The Manual of Musical Instrument

Conservation
This is the first book to combine museum-based conservation techniques with
practical instructions on the maintenance, repair, adjustment, and tuning of
virtually every type of historical musical instrument. As one of the world's
leading conservators of musical instruments, Stewart Pollens gives practical
advice on the handling, storage, display, and use of historic musical
instruments in museums and other settings, and provides technical
information on such wide-ranging subjects as acoustics, cleaning, climate
control, corrosion, disinfestation, conservation ethics, historic stringing
practice, measurement and historic metrology, retouching, tuning historic
temperaments, varnish, and writing reports. There are informative essays on
the conservation of each of the major musical instrument groups, the
treatment of paper, textiles, wood, and metal, as well as historic techniques of
wood- and metalworking as they apply to musical instrument making and
repair. This is a practical guide that includes equations, formulas, tables, and
step-by-step instructions.
STEWART POLLENS served between 1976 and 2006 as the conservator of
musical instruments at the Metropolitan Museum of Art, where he restored
and maintained a collection of over 5000 instruments. He is the author of
over eighty scholarly articles and six books, including Forgotten Instruments
(1980), The Violin Forms of Antonio Stradivari (1992), The Early Pianoforte
(1995; reprinted 2009), Giuseppe Guarneri del Gesù (1998), François-Xavier
Tourte: Bow Maker (2001), and Stradivari (2010), which won a 2011 Choice
award for “Outstanding Academic Title.” In 1997 he was the recipient of the
American Musical Instrument Society's Nicholas Bessaraboff Prize for The
Early Pianoforte, a study of the invention and early history of the pianoforte.
He is also a contributor to The New Grove Dictionary of Music and
Musicians and writes frequently for The Strad.
The Manual of Musical Instrument
Conservation

Stewart Pollens
University Printing House, Cambridge CB2 8BS, United Kingdom

Cambridge University Press is part of the University of Cambridge.

It furthers the University's mission by disseminating knowledge in the pursuit
of education, learning and research at the highest international levels of
excellence.

www.cambridge.org
Information on this title: www.cambridge.org/9781107077805
© Stewart Pollens 2015
This publication is in copyright. Subject to statutory exception and to the
provisions of relevant collective licensing agreements, no reproduction of any
part may take place without the written permission of Cambridge University
Press.

First published 2015

Printed in the United Kingdom by TJ International Ltd, Padstow Cornwall
A catalogue record for this publication is available from the British Library
Library of Congress Cataloguing in Publication data
Pollens, Stewart.
The manual of musical instrument conservation / Stewart Pollens.
pages cm
Includes bibliographical references and index.
ISBN 978-1-107-07780-5 (Hardback : alk. paper) 1. Musical instruments–
Conservation and restoration.
2. Musical instruments–Maintenance and repair. I. Title.
ML460.P7 2015
784.192′8–dc23 2015019710
ISBN 978-1-107-07780-5 Hardback
Cambridge University Press has no responsibility for the persistence or
accuracy of URLs for external or third-party internet websites referred to in
this publication, and does not guarantee that any content on such websites is,
or will remain, accurate or appropriate.

Every effort has been made in preparing this book to provide accurate and
up-to-date information which is in accord with accepted standards and
practice at the time of publication. However, the author and publisher
disclaim all liability for direct or consequential damages resulting from the
use of material contained in this book. Readers are strongly advised to pay
careful attention to information provided by the manufacturer of any
chemicals or equipment that they plan to use.
 Contents
List of figures

Introduction
Entries A–Z

Select bibliography
Index
List of figures
1 Composite trumpet with bell rim by Johann Georg Schmied, Pfaffendorf,
1736, shown prior to restoration. Ex-Rosenbaum collection;
Hamamatsu Museum of Musical Instruments
2a Anonymous brass bugle before treatment with trisodium EDTA.
Collection of the author
2b Bugle after treatment and light polishing with a slurry of precipitated
chalk
3 Dent removal tools, including stakes and hammers. Workshop of the
author
4 Bending a tube filled with lead or pitch at Boosey & Co. (from Algernon
Rose, Talks with Bandsmen, 1894)
5 Bronze disease on a brass bugle
6 Drafted clavichord tangent positions
7 X-ray of woodworm damage in a viola
8 Turning a graduated ivory cork adjuster for an English flute
9 Moth-eaten escapement cloths in a fortepiano by Conrad Graf, Vienna,
c.1839. The Metropolitan Museum of Art, 2001.272
10 Gilding tools. Workshop of the author
11a Clamping a loose bar in a flamenco guitar by Miguel Angel Senovilla,
Madrid, 1998
11b Clamping a separation between the back and sides with spool clamps.
Private collection
12 Copies of tuning-machine parts for a triple-necked guitar by André
Augustin Chevrier, Mirecourt, c.1830. The Metropolitan Museum of
Art, 1992.117.1
13 Exhibition of pianos at the Metropolitan Museum of Art in 1985
14 Harper's knot
15a Henry Greenway cross-strung harp, Brooklyn, after 1895. The
Metropolitan Museum of Art, 89.4.1235
15b Reconstruction of a damaged gilded-compo ornament on the Greenway
harp
16 Quilling a harpsichord jack
17 New parts fitted to a bentside spinet by Baker Harris, London, 1771.
Van Cortlandt House Museum, New York
18 Fading of varnish on a viola by Carleen Hutchins after one year's
exhibition at a luminance of 75 lux supplied by ceiling-mounted 60-
watt quartz-halogen lamp. The Metropolitan Museum of Art,
2002.571
19 Two methods of tying lute strings to the bridge. Lute by Stephen Barber
and Sandi Harris, collection of the author
20 Measuring instruments. Workshop of the author
21 Rubber mold with replicas of Broadwood square piano bolt covers
22 Cast copies of ormolu mounts from a Viennese fortepiano before cutting
off their sprues and electrogilding
23 Replacement mother-of-pearl parts ready to be glued into a nineteenth-
century qanun soundboard rosette. The Metropolitan Museum of Art,
89.4.330
24 Organ tuning and voicing tools. Workshop of the author
25 Organ windchest by William Crowell, New Hampshire, 1852. The
Metropolitan Museum of Art, 2001.272
26 Modern piano action. Line drawing from William Braid White's Piano
Tuning and Allied Arts, 1946
27 Standard set of piano action regulating tools. Workshop of the author
28 Piano tuning and voicing tools. Workshop of the author
29 Setting a sector with a pair of dividers. Ivory sector and drafting
instrument collection of the author
30 Spectrophotometry of Stradivari violin varnish (1694)
31 Front of Nicolò Paganini's Giuseppe Guarneri del Gesù violin showing
encrusted rosin. The city of Genoa
32a Interior of pianoforte by William Frecker, London, 1799. Private
collection
32b William Frecker pianoforte after restoration
33a Soundboard of a clavichord by J. C. Meerbach, Gotha, 1799, prior to
restoration. Moravian College, Bethlehem, PA
33b Soundboard after restoration
33c Underside of soundboard before restoration
33d Underside of soundboard after restoration
34a Loose bridge of a Conrad Graf fortepiano, Vienna, c.1838. The
Metropolitan Museum of Art, 2001.272
34b Clamping system used to reglue the bridge of the Conrad Graf
fortepiano using veneer press screws to apply pressure from above
and below the soundboard
35a Damper mechanism of a Conrad Graf fortepiano showing corroded
lead weight
35b Damper from a Conrad Graf fortepiano shown with replaced lead
weight
36a Padre Antonio Soler's monochord fretting template from his Theorica y
Practica del Temple para los Organos y Claves. With permission of
the Sociedad Española de Musicología
36b The author using a monochord fitted with Soler's fretting template to
assist in the tuning of a harpsichord
37 Werckmeister III tuning instructions from Andreas Werckmeister,
Musicalische Temperatur (Frankfurt and Leipzig, 1691)
38 Cross section of Stradivari varnish (1734) showing stratification
revealed by epi-fluorescence microscope
39 Gluing the lower bout of a violin made by Peter Guarneri, Venice, 1742.
Collection of Stephanie Chase
40 Gluing together a violin top
41 Method of attaching violin string to tuning peg
42 Stradivari tenor viola neck-foot template showing angled foot. Museo
Stradivariano
43 Original Stradivari bass-bars. Ex-Hill, Collection of James Warren
44 Overspun string made for a square piano by Alpheus Babcock, c.1820
45a Graves & Co. clarinet, Winchester, NH, c.1825, before restoration. Ex-
Lillian Caplin collection
45b Turning an ivory ferrule for the Graves & Co. clarinet
45c The Graves & Co. clarinet after restoration
46 Steps in making a key from the firm of H. Bettoney made for a didactic
display at the Metropolitan Museum of Art, c.1900
47 Woodworking tools. Workshop of the author
48 Preliminary fitting of replacement inlays in a Broadwood square piano,
London, 1801. Private collection
49 Turned bead-and-reel molding of rosewood for a Broadwood pianoforte
50 Scraped moldings
 Introduction

Much of the information included in this manual is derived from a shop

notebook that I compiled while serving as the musical instrument conservator
of the Metropolitan Museum of Art between 1976 and 2006, and it is a
reflection of the range of tasks that confronted me on a day-to-day basis. This
manual has not been conceived as a “how to” textbook on musical instrument
repair and conservation, but rather as a convenient alphabetically arranged
reference work organized for the quick retrieval of useful information. I have,
however, augmented the recipes, formulas, charts, and tables in that shop
notebook with essays on procedures and practices that I feel are relevant to
the care of musical instruments. I hope this manual will be of assistance to
those who pursue musical instrument repair and conservation as their
profession and to those who oversee such work.
During my thirty years as a museum-based conservator and as an outside
observer of museums since my formal retirement, I have witnessed a number
of significant changes within the profession. Around the time I entered the
field, major museums throughout the United States were in the midst of
altering their hiring policy: a foundation in practical craftsmanship was no
longer deemed sufficient or even relevant, and in an effort to “raise
standards” a master's degree in conservation from one of a coterie of
university-based programs became a virtual requirement for a full-time staff
position. Along with this alteration in hiring policy, the title “restorer” was
changed to “conservator” – a semantic refinement ostensibly intended to
signal a shift from a “glue-and-fix-it” approach to one based on scientific
training. The new breed of conservator may be guided by such laudable
precepts as “the preservation of cultural property,” an “informed respect for
cultural property,” “preventive conservation,” “documentation,” the
“stabilization of objects,” and the “reversibility” of treatments (to paraphrase
the American Institute for Conservation's 1994 Code of Ethics and
Guidelines for Practice and the International Council of Museums' 2013
Code of Ethics), though their classroom experience may not equip them to
tune a harpsichord or even to install a set of violin strings. In the specialty of
musical instrument conservation, academic credentials in conservation must
be supplemented by certain physical skills, such as those gained through a
formal apprenticeship with an established instrument maker, as well as a
modicum of musical training: in addition to being able to make an
instrument, a musical instrument conservator should be able to play one.
In recent years there has been a spate of articles and books decrying the
restoration of historic musical instruments (Barclay, 2005; Barnes, 1980;
O'Brien, 2008), and a few collections have declared moratoriums on restoring
their instruments because they believe that virtually any form of intervention
destroys historical evidence. This policy is a reaction to many restorations,
now considered ruinous, that were carried out in the course of the early music
revival. At the other end of the spectrum are musical instrument departments
that are called upon (or perceive it to be their responsibility) to serve as
entertainment or fundraising divisions of their museums. Some of the most
active departments do not even employ a staff conservator but instead rely
upon musicians, musical instrument makers, and outside repairmen to
maintain their instruments in playing condition. In such cases the standards of
practice as defined in the bylaws of the American Institute of Conservation,
for example, may be unknown to these individuals and therefore not
observed. Despite their training and special expertise, staff conservators often
exert very little authority within the museum hierarchy, and though they may
be consulted as a matter of procedure, it is the curator, and not the
conservator, who decides which objects are to be restored and how they will
be used.
It is assumed that there is rhyme and reason behind the formation of
collections, and that museums acquire objects on the basis of their relevance
to the collection, historical importance, quality, and condition. However,
most musical instrument collections gratefully accept virtually anything as a
gift, and often such instruments arrive, and remain, in disrepair. Anyone who
has ventured behind the scenes of the world's major collections has certainly
been appalled both by the condition of many thousands of instruments
crammed in their storerooms – dented and heavily corroded brass
instruments; tarnished silver flutes; keyboard instruments with torn-off lids
and missing everything from legs to ornamental hardware; stringed
instruments lacking pegs, bridges, and strings – and by, in general, evidence
of inadequate climate control and protection from dust, dirt, woodworm, and
other pests, as well as the aftereffects of floods and burst water pipes. Faced
with such a discouraging situation, it is surprising that some of the
profession's most respected figures are so vehemently opposed to any and all
forms of intervention. Certainly, much can be done to improve the
appearance of many of these neglected instruments, and perhaps make them
playable, with minimal or no compromise to their historical integrity. In
general, I neither advocate nor disapprove of the restoration of historic
musical instruments; rather, I believe that each instrument and proposed
treatment must be considered on a case-by-case basis with consideration
given to the requirements and needs of the collection or owner. The only
requirement that I would insist upon is that such work be carried out by
technically skilled and historically informed individuals who are cognizant of
their ethical responsibility to preserve original material and respectful of the
maker's concepts.
While this manual provides detailed instructions for employing the most
up-to-date techniques, it also advocates a number of traditional procedures
and materials that remain relevant and in some cases are preferable to their
modern counterparts. For example, many of the modern synthetic materials
that are widely employed in conservation work (such as the ubiquitous
acrylic adhesives and varnishes) are often inappropriate, ineffective, and
potentially harmful when used in musical instrument conservation, whereas a
traditional material such as animal-hide glue is not only eternally “reversible”
but ideally suited to certain tasks. Furthermore, some of the mainstays of
modern conservation practice, such as the consolidation, stabilization, and
impregnation of wood with synthetic resins and other substances, should be
generally avoided in musical instrument work because they alter an
instrument's acoustical and playing characteristics, and are, in fact,
irreversible.
Throughout this work, I have retained the units of weight and measurement
used in many of the original sources rather than converting everything into
metric, as the fundamental proportions are often more easily recognized in
their original units. For convenience, I have provided a table of equivalents
(see measurement system conversion). The references listed at the end of
each section include the sources of the information found in those sections as
well as recommendations for further reading.
Unless otherwise noted, all photographs and the conservation treatments
represented in them are the work of the author.

References
Atti della giornata di studi sul restauro liutario (Cremona, 1976).
Robert L. Barclay, The Preservation and Use of Historic Musical Instruments
(London, 2005).
John Barnes, “Does Restoration Destroy Evidence?” Early Music 8/2 (1980),
pp. 213–218.
Code of Ethics and Guidelines for Practice (American Institute for
Conservation, 1994).
ICOM Code of Ethics (International Council of Museums, 2013).
Grant O'Brien, “The Conservation of the 1690 Spinetta Ovale by Bartolomeo
Cristofori,” Restauro e Conservazione degli Strumenti Musicali Antichi:
La Spinetta Ovale di Bartolomeo Cristofori (Florence, 2008), pp. 137–150.
Stewart Pollens, “Curt Sachs and Musical Instrument Restoration,” The
Musical Times 130/1760 (October, 1989), pp. 589–594.
Stewart Pollens, “Flemish Harpsichords and Virginals in The Metropolitan
Museum of Art: An Analysis of Early Alterations and Restorations,”
Metropolitan Museum Journal 32 (New York, 1997), pp. 85–110.
Stewart Pollens, “Early Alterations Made to Ruckers, Couchet, and Grouwels
Harpsichords in the Collection of The Metropolitan Museum of Art,”
Kielinstrumente aus der Werkstatt Ruckers zu Konzeption, Bauweise und
Ravelement sowie Restaurierung und Konservierung (Halle an der Saale,
1998), pp. 136–170.
Standards in the Museum Care of Musical Instruments 1995 (Museums and
Galleries Commission, 1995).
 A
Acoustics

(see also Intervals, Mersenne's Law, Tuning and temperament)

Acoustics, or the science of sound, is a broad subject that includes the
study of vibration, noise abatement, room and architectural acoustics, the
physiology and perception of sound, speech and hearing, the processes of
recording and reproducing sound, sound reinforcement, hydrophonics, and
ultrasonics, as well as musical acoustics. So-called “acoustical” musical
instruments produce sound by causing bodies or columns of air, strings, rods,
bars, reeds, plates, and membranes to vibrate at certain frequencies.
The science of musical acoustics advanced significantly in the twentieth
century with the development of recording equipment, audio spectrum
analyzers, high-speed computers, hologram interferometry, and the
application of mathematical techniques such as Fourier transform, constant Q
transform, and finite element analysis. Though modern science has increased
our fundamental understanding of how musical instruments function, it has
unfortunately provided little practical benefit to musical instrument makers,
and the old, empirical “cut and try” approach still plays a large part in the
design, construction, and adjustment of musical instruments. As conservators,
our task is not to improve upon instruments but rather to preserve their
characteristics; nevertheless, many of the published studies and readily
available instrumentation, such as electronic tuning devices, frequency
counters, and audio spectrum analyzers, can be used to optimize their setup
and adjustment. For example, when replacing a missing fortepiano hammer,
one might employ a Lucchi meter (an electronic device that measures the
speed of sound in wood, which is proportional to its stiffness) in order to
select a piece of wood that matches the properties of the neighboring
hammers.

Acoustical knowledge up through the eighteenth century

 The science of acoustics may be thought to have originated with the
ancient Greek philosopher Pythagoras (c.570–497 BC), who is said to have
discovered that simple arithmetic ratios underlie musical intervals while
listening to the sounds produced by blacksmith's hammers striking an anvil
(that is, hammers whose weights were in the ratio of 2:1 produced sounds an
octave apart, and hammers whose weights were in the ratio of 2:3 produced
sounds a fifth apart). The Roman philosopher Boethius (480–524 CE)
maintained that Pythagoras invented the monochord for studying the pitch of
vibrating strings. Although this assertion has been disputed, by Boethius's
time the monochord was widely used to study musical intervals by physically
dividing the length of its string into various ratios and proportions (a ratio is
the relationship between two quantities; a proportion is the relationship
among three or more quantities, such as the equality of two ratios). Thus,
Pythagoras and his followers established a clear link between musical
intervals, mathematics, and geometry.
The Roman architect, engineer, and author of The Ten Books on
Architecture Marcus Vitruvius Pollio advised architects to study music
because of its underlying mathematical rigor. He also described the design
and construction of ancient theaters, which employed tuned niches and
resonant bronze vessels positioned around the circumference of the stage to
augment the voice of the speakers. These were “arranged with a view to
musical concords or harmony, and apportioned in the compass of the fourth,
the fifth, and the octave, and so on up to the double octave, in such a way that
when the voice of an actor falls in unison with any of them its power is
increased, and it reaches the ears of the audience with greater clearness and
sweetness.” The use of air resonance to accentuate certain frequencies
remains critical in the design of musical instruments. Optimally, the air
resonance of an instrument's body should lie at or below the lowest frequency
that can be produced by the instrument. For example, at Baroque A=415
pitch, a violin's G string vibrates at around 185 Hz, though a typical violin
body's air resonance is around 290 Hz, which is close to a Baroque d♯1. Thus,
the air resonance of a violin begins to support its D, or second string. The
earliest violins were made with only three strings; the G string was evidently
an afterthought added around the middle of the sixteenth century, though the
body of the instrument was never enlarged to accommodate it acoustically
(Pollens, 2011).
 In the sixteenth century the astronomer Galileo Galilei (1564–1642, the son
of the lutenist and music theorist Vincenzo Galilei, 1520–1591) demonstrated
that a sounding body produces ripples, or vibrations, when placed in water,
but the precise number of vibrations per second in musical pitches was not
then known. It was not until the 1660s that the French mathematician,
philosopher, and music theorist Marin Mersenne (1588–1648) conducted
experiments that enabled him to estimate the number of vibrations per second
produced by low-pitched organ pipes by comparing their pitches with
measurable vibrations in strings that were approximately 18 feet in length. He
made the first measurements of the speed of sound by timing the difference
between a gun's muzzle flash and its audible report at a measured distance.
Mersenne also codified the relationship between string length, string tension,
and pitch (see Mersenne's Law). In writing about the selection and testing of
strings in his Harmonie universelle (1636), Mersenne advocated the tuning of
soundboards and air resonances of instruments to the pitch of their strings:

It is enough to conclude that one will know which string sounds better
than all the others on a proposed instrument when one knows the pitch
of the table [soundboard] of the instrument, for that [string] will produce
the best sound which, having the requisite length and tension, will be in
unison with the said table, and if one finds many of the same thickness,
length, and tension, which are in unison, the one with more uniform
parts will sound the better; and if all the parts of some are as even as
those of the others, they will sound equally.
If those who make as great a thing of a good string as of the whole
instrument take the trouble of finding the tone of the soundboard, I feel
that they will have some satisfaction in comparing these two unisons and
they will admit that the unison is the most powerful and the most
excellent of all the consonances, as I have proved in the Books of
Theory, since the union made of the tone of the table with that of the
string produces an exquisite harmony, for it builds almost a single tone
from the two. Still, I do not wish to reject the other reasons that can be
given for the goodness of strings, for example, that the air enclosed in
the body of the instrument must be very well proportioned to the length
of the string, which ought to find not too great a quantity of air to stir up.
The simplest means of determining the fundamental pitch of a musical
instrument soundboard is to tap it and listen to the sound it produces. It
should be noted that the actual frequencies of musical tones were not known
until around 1700, when the French physicist Joseph Sauveur (1653–1716)
employed the beat rate (which represents the difference in pitch between two
notes) between two low-pitched organ pipes whose interval ratio was known
to calculate the actual number of vibrations produced by both pipes. (For
example, if we know that the ratio between two notes is 17:18, and that two
pipes tuned to those notes beat at 5 cycles per second, then the pitch of the
higher-pitched pipe can be obtained by multiplying the beat rate by 18, which
gives 90 cycles per second, while the lower-pitched pipe is calculated by
multiplying 90 by 17/18, or 85 cycles per second.) In Stradivari's day, violin
makers may have tap-tested the front and back plates of their instruments, but
it was not until 100 years after the death of Stradivari that the French
physicist Félix Savart (1791–1841) wrote that the disassembled front and
back plates of fine violins by Stradivari, Guarneri, and Vuillaume (the latter a
contemporary of Savart) vibrated at particular frequencies when lightly
tapped with a finger. Savart determined that the front plate of a Stradivari
violin vibrated at 512 Hz (the equivalent of c2 at A=430 Hz) and that the back
plate was tuned about a half or whole tone higher. However, violin makers
and theorists have not always agreed upon the optimal tap tones of violin
plates. For example, the Russian theorist Anatoli Leman (1853–1913) notes
in The Acoustics of the Violin (1903) that violin plates exhibit three tap tones
located at the upper, middle, and lower sections of the front and back plates,
and he proposed that the front plate's tap tones should be tuned to F, D, and
B, while the back plate's tap tones should be tuned to G, E, and C♯.
Mersenne's observations about the sympathetic vibration between tuned
strings, wood, and air volume were published just before Stradivari was born,
so it is possible that Stradivari used some method (perhaps tap tones) to
establish the resonant pitches of the front and back plates of his instruments.
It is important to consider that pitch was not standardized in Stradivari's
time, though in most locales it was generally somewhere between a whole
and a half step lower than it is today (see Pitch). With higher pitch, there is
more string tension as well as downward pressure exerted by the bridge on
the front plate, which has the effect of raising the front plate's resonant
frequency. However, most instruments made during Stradivari's time were
later “regraduated” (a euphemistic term for “thinning”). This was primarily a
timbre-altering procedure that was first carried out on a large scale in the
third quarter of the eighteenth century and continues to this day. Though the
thinning of front and back plates has the general effect of lowering their
resonant frequencies, it does not fully compensate for the moderate rise in
resonant frequency caused by the increased downward pressure on the top,
but compounds the discrepancy between the plates' resonant frequencies and
present pitch standards. Thus, if Stradivari somehow optimized the structure
and resonances of his instruments to conform to pitch standards then in
vogue, his instruments are not functioning optimally today (unless later
restorers somehow surmised Stradivari's plate tuning principle and retuned
the plates of his instruments to accommodate later pitch and string tension).
Mersenne also makes reference to the air resonance of an instrument's body
(which is derived from the body outline, rib height, and f-hole dimensions).
Though the rib structures of most of Stradivari's violins and violas have been
reduced by a millimeter or two as a result of shrinkage and repair, many of
his cellos have had their ribs deliberately reduced in height and their entire
bodies cut down in length and width, often by as much as several centimeters,
in an effort to render them more comfortable to play. With violins and violas,
the rise in air resonance resulting from the slight reduction in body volume is
not significant, but with cut-down cellos, the lowest notes are certainly
affected.
The German historian and music theorist Athanasius Kircher (1601–1680)
wrote extensively on the reflection, spreading, and compression of sound. An
engraving in his Phonurgia nova (Kempten, 1673) illustrates the
phenomenon of an ellipsoidal “whispering room.” As an ellipse has two foci,
if someone whispers at one focus, the sound will be clearly heard by someone
else standing at the other. Kircher's use of ray tracing to explain the
projection of sound may have stemmed from contemporary theories on the
nature of light and optics that used similar diagrams to explain how light rays
and images were brought into focus by lenses.
Francis Bacon also makes several references to room acoustics in his Sylva
Sylvarum (1627) and describes the acoustical properties of an architectural
beam that is comparable to a violin's soundpost:
 I remember in Trinity College in Cambridge, there was an upper
chamber, which being thought weak in the roof … was supported by a
pillar of iron of the bigness of one's arm in the midst of the chamber;
which if you had struck, it would make a little flat noise in the room
where it was struck, but it would make a great bomb [boom?] in the
chamber beneath.

Bacon's Sylva Sylvarum includes a number of general remarks on the

acoustical properties of musical instruments. For example, he describes the
interaction between the “knot” (soundboard rose), the “board” (soundboard),
and “concave underneath” (the volume of air enclosed by the instrument):

The strings of a lute, or viol, or virginals, do give a far greater sound, by

reason of the knot and board, and concave underneath, than if there were
nothing but only the flat of a board, without that hollow and knot, to let
in the upper air into the lower. The cause is the communication of the
upper air with the lower, and penning of both from expense of
dispersing.
An Irish harp hath open air on both sides of the strings: and it hath the
concave or belly not along the strings, but at the end of the strings. It
maketh a more resounding sound than a bandora, orpharion, or citter,
which have likewise wire strings. I judge the cause to be, for that open
air on both sides helpeth, so that there be a concave; which is therefore
best placed at the end.
In a virginal, when the lid is down, it maketh a more exile sound than
when the lid is open. The cause is, for that all shutting in of air, where
there is no competent vent, dampeth the sound: which maintaineth
likewise the former instance; for the belly of the lute or viol doth pen the
air somewhat.
The harp hath the concave not along the strings, but across the strings;
and no instrument hath the sound so melting and prolonged as the Irish
harp. So as I suppose, that if a virginal were made with a double
concave, the one all the length, as the virginal hath, the other at the end
of the strings, as the harp hath; it must needs make the sound perfecter,
and not so shallow and jarring. You may try it without any sound-board
 along, but only harp-wise at one end of the strings; or lastly, with a
double concave, at each end of the strings one.

Bacon also addresses the phenomenon of sympathetic vibration and the

phenomenon as it applies to the viola d'amore:

There is a common observation, that if a lute or viol be laid upon the

back, with a small straw upon one of the strings, and another lute or viol
be laid by it; and in the other lute or viol the unison to that string be
strucken, it will make the string move, which will appear both to the
eye, and by the straw's falling off. The like will be, if the diapason or
eighth to that string be stricken, either in the same lute or viol, or in
others lying by: but in none of these there is any report of sound that can
be discerned, but only motion.
It was devised, that a viol should have a lay of wire-strings below, as
close to the belly as a lute, and then the strings of guts mounted upon a
bridge as in ordinary viols: to the end, that by this means, the upper
strings stricken should make the lower resound by sympathy, and so
make the music the better; which if it be to purpose, then sympathy
worketh as well by report of sound as by motion.

In his Philosophical Essay of Music Directed to a Friend (London, 1677),

Francis North includes a section headed “How tones are produced, and of
assistances to the sound by instruments,” in which he comments on the
contribution to the production of sound made by the sides of an instrument,
the influence of mass on the bridge, and the function of the soundpost in the
violin:

In violins and harpsichords the tones are made wholly by the vibrating
strings, but the frame of the instrument adds much to the sound: for such
strings vibrating upon a flat rough board would yield but a faint and
pitifull sound.
The help that instruments give to the sound, is by reason that their sides
tremble and comply with any sound, and strike the air in the same
measure that the vibrations of the musick are, and so considerably
 increase the sound.
This trembling is chiefly occasioned by the continuity of the side of the
instrument with the vibrating string; therefore if the bridge of a violin be
loaded with lead, the sound will be damp; and if there be not a stick
called the soundpost to promote the continuity between the back and
belly of the instrument, the sound will not be brisk and sprightly.
Such a continuity to the nerve of hearing will cause a sense of sound to a
man that hath stopped his ears, if he will hold a stick that touches the
sounding instrument between his teeth.
The sound of itself without such continuity would occasion some
trembling, as may be seen by the moving the unison strings in the
instance before given; but this is not considerable in respect of the other,
though it be all the assistance that the structure of a chamber can give to
musick, except what is by way of eccho.
This tremble of the instrument changes with every new sound; the spring
of the sides of the instrument standing indifferent to take any measure,
receives a new impression: but a vibrating string can take no measure
but according to its tension.
Therefore instruments that have nothing to stop the sounding strings
make an intolerable jangle to one that stands near, as bells to one that is
in the steeple, and hears the continuing sound of dissonant tones; such is
the dulcimer; but the harpsichord that hath rags upon the jacks by which
the vibration of the string is staid, gives no disturbance by the
sonorousness of the instrument, for that continues not the sound after the
vibrations determined, and another tone struck, but changes and
complies with the new sound.

North also discusses the properties of strings and the phenomenon of

harmonics:

A perfect string produces a clear sound by entire and equall vibrations,

there being no inequality to hinder the motion from being uniform from
one end to the other, according to the laws of a pendulum: but if the
string hath any inequality toward one end, it will yield a jarring and
 distracted sound; for the resistances are not only at the ends of the string,
but there are cross tugges that alter the course of the vibrations; which is
evident in the manner whereby musicians try if their string be true: for if
the string be true the vibrations will appear as a clear filme; but they will
appear with cross threads if the string be false.
It is common experience, that a great string struck near the bridge with a
bow where the rosin takes but small hold, will whistle and break into
chords above; which it were struck by the thumb that removes it out of
its place, would give the true tone.
The trumpet marine that sounds wholly upon such breaks, is a large and
long monochord played on by a bow near the end, which causes the
string to break into shrill notes. The removing the thumb that stops upon
the string gives measure to these breaks, and consequently directs the
tone to be produced. The jar at the bridge takes the same measure and
makes the sound loud, in imitation of a trumpet, which otherwise would
be like a whistle or pipe.

In 1724 the French mathematician Pierre-Louis de Maupertuis presented a

paper to the French Royal Academy entitled “Sur la forme des instruments de
musique,” which explores how an instrument's shape affects the quality of
sound it produces. In this paper, he proposes that wood fibers act like
vibrating reeds, and that these reeds resonate at discrete frequencies
according to their length. Maupertuis suggests that the shape of the violin's
front plate and the position of the sound holes create a wide variety of reed
lengths, permitting the violin to resonate at many frequencies, and he
proposes that violins might benefit from being broken apart, thereby “freeing
up the reeds,” and then gluing them back together.
When an instrument maker continually experiments with new designs, it is
clear that he has not come upon the ideal proportions and shapes by scientific
means but is searching for a better means of achieving the tonal qualities he
desires. Stradivari's wood forms provide ample evidence of this empirical
approach. Four of his forms are graded in size and are inscribed, from largest
to smallest, P, S, T, and Q, which likely stand for prima (first), seconda
(second), terza (third), and quarta (fourth). Two of them, P and S, are
roughly duplicated (see below), and the S forms have a slightly longer
counterpart marked SL (probably an abbreviation for seconda lunga, or
second, long). A form marked MB (possibly modello buono, or good model)
is thought to be the earliest form and is said to be dimensionally similar in
outline to violins made by Nicolo Amati. The forms marked P, PG (prima
grande, or first, large), and G (grande, or large) are similar, though not
identical in shape, and were used to make his esteemed grand-pattern violins.
Two forms having slightly different dimensions are inscribed with the letter
B, and the longer of these appears to have been used to construct the so-
called “long-pattern” violins that were made primarily in the 1690s. This
form is similar to the SL form, which he made the previous year (see dates
below).
How important is the outline of a violin with regard to its tonal qualities?
Clearly, Stradivari thought it was a critical aspect of design because he
continually experimented and modified the size and shape of his violins,
violas, and cellos. With the cello, he gradually reduced its overall dimensions
during the course of his career, no doubt because of the gradual adoption of
metal overwound strings (invented around 1660), which enabled him to
shorten the string length and thus reduce the size of the instrument. After
Stradivari completed the set of five instruments for the Medici court in 1690,
he redesigned his contralto viola form by making it narrower, though
retaining the original length. In a sense, this slimmed-down viola was a better
visual and perhaps acoustical match for the elegant long-pattern violins that
came into production around that time. Though the long-pattern violin was all
but discontinued around 1700, Stradivari continued to use the slender 1690
contralto viola form for the rest of his life. With the demise of the long-
pattern violin, he reverted to his earlier P form and to the PG and G forms
derived from it that were used to make the “Alard” (1715), “Soil” (1714), and
“Sarasate” (1724) violins. This variety of shapes and sizes, in some cases
going in one direction (as in the cello and viola) or wavering between one
basic design and another (in the case of the violin) is indicative of Stradivari's
restless experimentation, an approach that typifies the work of many
important historic makers.
Stradivari also experimented with arching, which is perhaps more critical
to the tone of an instrument than the outline. His early instruments, in
keeping with the work of his Cremonese predecessors, are highly arched,
while his long-pattern and golden period instruments exhibit the flatter
arching that contributed to a more penetrating tone. Violins made in his later
years surprisingly exhibit a return to higher arching.
In the 1920s a cache of documents came to light in Italy that ostensibly
included the workshop notebook of Antonio Stradivari. In court proceedings,
these documents were declared fraudulent, and one of the perpetrators was
imprisoned. It is unclear whether a manuscript entitled Librem Segreti de
Buttegha is the workshop notebook that is mentioned in the inventory of
these documents (see Peluzzi, 1978; Dipper, 2013). This manuscript is said to
have passed into the hands of a book collector named Federico Patétta, who
allegedly donated it to the Vatican; however, it is not currently listed in the
Vatican's catalog and Vatican officials deny possession of it. The author's
impression of this manuscript, based upon the single page that has been
photographically reproduced by Peluzzi and Dipper, is that it is not in
Stradivari's hand. Furthermore, aspects would appear to be derived from
Antonio Bagatella's Regole per la costruzione de' violini, viole, violoncelli e
violoni: memoria presentata alla R. Accademia di Sienze Lettere et Arti di
Padova, which was first published in 1786 and went through innumerable
editions into the twentieth century. The manuscript ostensibly provides
geometric techniques for establishing the general dimensions, arching, and
thicknesses of the violin's plates; however, its instructions are ambiguous at
best. In all likelihood the Librem Segreti, if it exists or ever existed, would
appear to be a twentieth-century fabrication and unworthy of further
consideration.
One eighteenth-century document that does shed some light on the thought
processes of contemporary makers regarding violin acoustics and design is a
manuscript dated 1786 by the Bolognese violin maker Giovanni Antonio
Marchi (1727–1807), which is preserved in the Biblioteca Comunale in
Bologna (Ms. No. B3195). A transcription and English translation of this
important and insightful document was published by Roberto Regazzi (1986).
Another valuable work is the Abbé Sibire's La Chélonomie ou le parfait
luthier (published in Paris in 1806 and reissued in Brussels in 1823), which is
a treatise on violin making derived from the author's communication with the
violin maker Nicolas Lupot (1758–1824), who worked in Orléans between
1768 and 1794 and subsequently in Paris until his death.

Acoustical science in the nineteenth century

 Charles Hutton's Recreations in Mathematics and Natural Philosophy
(1803) presents a summary of acoustical knowledge at the beginning of the
nineteenth century. This work is a translation and enlargement of Jean-
Étienne Montucla's 1778 edition of Jacques Ozanam's 1694 work entitled
Recreations Mathématiques et Physiques. Montucla (1725–1799) and
Ozanam (1640–1718) were prominent French mathematicians, while Hutton
was an English mathematician and Fellow of the Royal Society. Hutton's
edition characterizes sound as “nothing else but the vibration of the particles
of the air, occasioned either by some sudden agitation of a certain mass of the
atmosphere violently compressed or expanded, or by the communication of
the vibration of the minute parts of a hard and elastic body,” and he correctly
notes that sound is not propagated in a vacuum. Regarding the velocity of
sound, he cites numerous experiments that demonstrated that sound travels at
a rate of 1172 Parisian feet per second, though he indicates that Gassendus
(Pierre Gassendi; French astronomer and physicist, 1592–1655) measured its
velocity at 1473 feet per second, while Mersenne found it to be 1474 feet per
second, Duhamel (Jean Baptiste Du Hamel; French astronomer and
philosopher, 1646–1706) measured it at 1338 feet per second, Isaac Newton
at 968 feet per second, while Derham (William Derham; English natural
philosopher, 1657–1735), Flamsteed (John Flamsteed; English astronomer,
1646–1719), and Halley (Edmond Halley; English astronomer, 1656–1742)
concurred that it was 1142 feet per second, which Hutton acknowledges to be
the generally adopted figure in England at the time of his writing (today, the
speed of sound in dry air at 20°C is reckoned to be 343.2 m/s or 1,126 feet
per second). Hutton correctly disputed Derham's experiment that ostensibly
demonstrated that the temperature of a body of air does not affect the speed
of sound traveling through it. There are also discussions regarding the general
propagation of sound and the phenomenon of echoes, and a short chapter
devoted to Ernst Chladni's discovery of vibration patterns formed by sand
sprinkled on plates of various shapes that were made to vibrate through the
action of a bow, which he published in 1787.
The nineteenth century witnessed important advances in theoretical
acoustics with the work of such physicists as Lord Rayleigh (John William
Strutt; 1842–1919), author of The Theory of Sound (vol. I, 1877; vol. II,
1878) and John Tyndall (1820–1893), author of Sound (1867). These treatises
introduced rigorous mathematical analyses in the study of the vibrations of
strings, membranes, and rigid plates of various shapes, as well as of gases
enclosed in open and closed tubes and other vessels, though it is unlikely that
instrument makers have ever been influenced by these works as they require
facility with algebra, trigonometry, and calculus.
The German physicist Hermann von Helmholtz (1821–1894) studied sound
in the laboratory. In his monumental publication Die Lehre von den
Tonempfindungen als physiologische Grundlage für die Theorie der Musik
(1863), he describes how small tuned vessels that resonate at specific
frequencies (termed “Helmholtz resonators”) can be used to analyze the
frequencies of overtones present in musical sounds, and his experiments with
a cleverly designed “vibration microscope” revealed the jagged waveform of
bowed strings caused by the grip and release of the string by the “sticky”
rosin applied to the bow's hair. He also explored the field of psychoacoustics,
or the perception of sound. August Kundt succeeded Helmholtz as the
director of the Berlin Physical Institute in 1888. Among his achievements
was the 1866 invention of the “Kundt's tube,” a glass tube containing
powdered cork dust or lycopodium (the spores of certain species of moss)
that settled at the nodal points of sound waves that were introduced into the
tube. Because the spaces between the nodes could be physically measured,
the actual wavelengths of sounds could be determined. If the frequency of a
sound wave was known, the speed of sound could then be calculated by
multiplying the wavelength by the frequency.
Though makers continued to develop and refine their instruments through
the nineteenth century, they tended to do so by empirical means rather than
by use of the laws of physics and mathematical calculation. For example,
though the position of tone holes in a woodwind instrument can be calculated
mathematically, these calculations are far more complex than those required
to position the frets of a lute or the tangents of a fretted clavichord, for
example, and were undoubtedly beyond the capacity of most makers. The
noted flute maker Theobald Boehm (1794–1881), who is credited with the
development of the cylindrical bore and “parabolic” head joint around 1847,
later described how he established the position of the flute's tone holes in Die
Flöte und das Flötenspiel (1871). While he states that “one must avail
himself of the help of theory,” he admits that:

I had a flute made with movable holes, and was thus enabled to adjust
 all the tones higher or lower at pleasure. In this way I could easily
determine the best positions of the upper three small holes, but it was
not possible to determine the tuning of the other tones as perfectly as I
desired; for in endeavoring to produce an entire true scale in one key,
the tones were always thrown out of the proportions of the equal
temperament, without which the best possible tuning of wind
instruments with tone-holes cannot be obtained.

Boehm continues:

I made a tube in which all the twelve tone sections could be taken off
and again put together, and which was provided with a sliding joint in
the upper part of the tube to correct for any defects in tuning. To
establish the optimal diameter of the tubing, I constructed many thin,
hard-drawn tubes of brass upon which the fundamental tone C, and also
higher notes, could be produced by a breath and be brought to any
desired strength without their rising in pitch. The fact that the hissing
noise heard in other flutes was not perceptible convinced me that the
correct dimensions of the tube, and its smooth inner surface permitted
the formation of the sound waves without noticeable friction. From this
as well as from the fine quality of tone of the harmonics or acoustical
over-tones, can be inferred the perfect fitness of my tube for the flute;
and with this I began the determination of the amount of shortening or
cutting of the air column, required for producing the intervals of the first
octave.

In the twentieth century, Arthur H. Benade, Donald E. Hall, and Neville H.

Fletcher and Thomas D. Rossing have published excellent general studies of
musical instrument acoustics, and researchers such as Lothar Cremer, Carleen
M. Hutchins, Frederick A. Saunders, and John C. Shelleng have conducted
fundamental research into the acoustics of the violin (see references below),
yet their combined labors have been of little assistance to most instrument
makers, who continue to rely upon the dimensional copying of earlier
designs.

Some basic principles of acoustics

 Acoustical, electrical, and mechanical systems are in many ways
analogous, and the electrical concepts of amperage, capacitance, impedance,
inductance, power, resistance, voltage, and wattage, as well as the mechanical
concepts of compliance, force, impedance, mass, power, resistance, and
velocity are sometimes used in describing and analyzing acoustical systems.
For example, a simple acoustical oscillator may be represented by an
electrical capacitor, inductor, and resistor wired together (the circuit of an
electrical oscillator). A Helmholtz resonator consists of a vessel with an open
neck; viewed as a mechanical system, the enclosed air in the vessel and the
air in the neck may be thought of as two masses suspended from springs,
each with their own natural frequencies of vibration.

Sound waves

A sound wave can be visualized as a sinusoidal curve generated through

one 360° cycle, or one crest and one trough. Because a stretched string is
motionless at its two fixed ends, the first, or fundamental mode of vibration,
consists of a crest, which is thus only half of a complete wavelength. The
second mode consists of a crest and a trough, the third mode a crest, a trough,
and a crest, and so on; thus, a stretched string exhibits both odd and even
multiples of the fundamental, otherwise known as harmonics. An open
cylindrical column of air (such as an “open flute” organ pipe) also exhibits
both odd and even harmonics. A cylindrical column of air closed at one end
(such as a “stopped flute” organ pipe) has a fundamental consisting of half of
a crest, followed by odd-number harmonics. Open and closed conical
columns of air behave like cylindrical ones in this respect.
The equation for the frequency of a harmonic oscillator is:

where

F is the frequency
k is the force constant
m is the mass.
Sound waves have several basic properties associated with them: speed,
frequency, wavelength, and amplitude. Pitch is a psychoacoustical
phenomenon associated with frequency, which is linked with wavelength and
the speed at which sound travels through a particular medium. The formula
that defines the relationship between the speed, frequency, and wavelength of
sound is:
c = fλ

where

c is the speed of sound in a particular medium

f is the frequency
λ is the wavelength.

The period, or duration, of one vibration is defined as:

P = 1/f

where

P is the period of vibration

f is the frequency.

The speed of sound varies in different media and obeys the Newton-Laplace
Law:

where

K is the coefficient of stiffness

ρ is the density of the medium.

For gases
where

c is the speed of sound

γ is the adiabatic index (approx. 1.4 for air)
P is the pressure
ρ is the density of the medium.

At 20°C, sound travels at the following nominal speeds through the following
media:

Air 343 m/s

Helium 927 m/s

Carbon dioxide 267 m/s

Water 1433 m/s

Aluminum 6420 m/s

Brass 3475 m/s

Copper 3901 m/s

Gold 3240 m/s

Iron 5130 m/s

Lead 1158 m/s

Steel 6100 m/s

Hardwood 3960 m/s

Softwood 3300–3600 m/s

Glass 3962 m/s

 Glass, Pyrex 5640 m/s

Plexiglas 2680 m/s

Brick 4176 m/s

Concrete 3200–3600 m/s

Cork 366–518 m/s

Diamond 12,000 m/s

Source: www.EngineeringToolBox.com

An interesting phenomenon that is not reflected in the above table (or the
relationship c = fλ; see above) is the effect of tension upon the speed of sound
in stretched strings (be they of metal, gut, or other material): as the tension is
increased on a string of a given length, the frequency increases, though the
wavelength remains constant because the endpoints of the string have not
been altered; thus, the speed of sound is not constant in a stretched string but
increases as the tension is raised.
The amplitude of a sinusoidal sound wave is the difference between the
peak and trough of one complete vibrational cycle. Sound intensity, or
volume, is the average rate at which sound energy is transmitted through a
unit area perpendicular to a specified direction.
Sound intensity level (SIL) is a logarithmic measure of sound intensity
compared to a reference level:
SIL = 10 log10 (I/Iref) dB

where

SIL is the sound intensity level in dB

I is the intensity in watts/m2
Iref is the reference sound intensity level of 10–12 watts/m2 (the threshold
of hearing).
Sound power is a measure of sound energy per unit time measured in watts:
P = IA

where

P is the sound power

I is the sound intensity
A is the area.

Sound power level (SWL) is the logarithmic measure of the sound power of a
source compared to a reference level:
SWL = 10 log10 W/Wref dB

where

SWL is the sound power level in dB

W is the sound power in watts
Wref is the reference sound power of 10–12 watt (0 dB level, or the
threshold of hearing).

A source having a sound power of 1 watt (such as a symphony orchestra) is

equivalent to 120 db.
Another logarithmic measure of sound level is sound pressure. Sound
pressure level (SPL) is the difference in pressure, relative to the ambient
atmospheric pressure, created by a sound wave. This can be expressed in
decibels (dB), and is defined as:
SPL = 20 log10 (p/pref) dB

where

SPL is the sound pressure level in dB

p is the effective sound pressure in microbars
pref is the reference pressure of 1 microbar (0.1 newton/m2).
Because the range of human hearing is so great (1:1012, or one to a trillion), a
logarithmic system, termed the decibel (dB; one-tenth of a bel), is used to
express differences in sound intensity or power. In the base-ten logarithmic
system a 3 dB difference between two sounds represents an approximate
doubling of the sound, while 10 dB represents a 10-fold difference, and 20
dB a 100-fold difference.
The phenomenon of beats is due to the interference of two frequencies that
are close enough to each other that they generate a sensible pulsation in
amplitude that can be counted or estimated. Acoustic beat frequencies range
from one beat over several seconds up to around 15 cycles per second, at
which point they are perceived as combination or difference tones, another
phenomenon first described by Giuseppe Tartini in his Trattato di musica
(Padua, 1754).
Resonance is a vibration that occurs when an object absorbs acoustical
energy at its natural frequency of vibration from another source. When a
string is plucked or bowed, it will vibrate at a frequency determined by its
length, mass, and tension (see Mersenne's Law). A body of air (as is enclosed
in an ocarina, organ pipe, or Helmholtz resonator) also has a natural
frequency of vibration. Many musical instruments have soundboxes with
openings in them (such as violins, viols, guitars, lutes, harps, harpsichords,
etc.) that exhibit a Helmholtz resonance. For a soundbox to be optimally
effective, it should have a resonance below that of its lowest pitched string.
The resonant frequency of a Helmholtz resonator can be calculated as
follows:

where

F is the frequency
c is the speed of sound in air
l is the length of the neck extending from the resonator (or thickness of
soundboard at opening)
v is the volume of the cavity or soundbox
a is the cross-sectional area of the neck or opening.
Sound is reflected by non-sound-absorbent surfaces and can be focused by
concave reflective surfaces (such as ellipsoidal “whispering rooms”); thus ray
tracing, which is employed in optics, has often been employed in
understanding room acoustics and in designing concert halls. Because sound
behaves like a wave, it exhibits diffraction, refraction, and interference
phenomena, such as bending around corners, spreading out after passing
through small openings, and confinement to narrow beams after passing
through openings that are large relative to their wavelength, alteration of
direction when traveling through media having different densities, as well as
the amplification and cancellation of multiple sources due to frequency and
phase relationships.
Sound waves are propagated in several ways. A longitudinal wave is
produced in a column of air inside an organ pipe when a stream of air is
blown across one end. This stream of air compresses the molecules of air in
the pipe standing just behind it, which then exert pressure on those beyond
them, and so on, until the pressure disturbance, or wave, travels down the
length of the tube. In the case of a stretched string, if it is plucked, a sideward
disturbance is propagated down the length of string, generating a traveling
wave. Two- and three-dimensional sound waves can be generated across
surfaces (such as a thin soundboard or stretched membrane) and through
solid, liquid, or gaseous bodies. Standing waves are created in stretched
strings when an impulse at one end rebounds at the other end and forms a
stationary pattern of nodes and antinodes. Standing waves formed in columns
of air, such as those in organ pipes, consist of masses of compressed and
rarefied air that are maintained by a steady stream of air supplied by the
pipe's mouth.

Vibrations in bars

A thin wood or metal bar that is free at both ends (such as those of a
xylophone) vibrates like an air column that is open at both ends. Its
fundamental frequency is approximately:

where
 F is the frequency
a is the thickness
L is the length
d is the density of the bar
y is Young's modulus (Hall, 1980).

A thin bar that is fixed at one end has a fundamental frequency of

approximately:

where

F is the frequency
a is the thickness
L is the length
d is the density
y is the Young’s modulus (Hall, 1980).

Vibrations in membranes and plates

Vibrational modes occur in stretched membranes (such as the skins of

drums) and rigid plates. In square- or rectangular-shaped membranes and
plates, the vibrational modes tend to align in square or rectangular segments,
while circular membranes and plates tend to form concentric bulges. More
complex shapes, such as the top and back plates of violins and keyboard
instrument soundboards, exhibit vibrational modes having complex shapes
(see works by Benade, Fletcher and Rossing, and Hutchins cited in the
References below).

Vibrations in wind instruments

Cylindrically and conically bored wind instruments that have terminal

diameters smaller than the wavelengths of the lowest pitches that emanate
from them (such as modern and Baroque flutes) are not efficient radiators of
sound. Flared bells (such as those found on clarinets and most brass
instruments) radiate lower-pitched sound more efficiently, especially of tones
whose wavelengths are shorter than the diameter of the bell. The bodies or
“walls” of woodwind and brass instruments are responsible for vibrational
losses due to their own resonances and surface roughness. Materials such as
wood, ivory, plastics, and various metals have different physical properties,
and though it is difficult to measure the effect of these properties upon the
tone of an instrument, makers and players are convinced that flutes made of
platinum, gold, silver, nickel silver, and various woods have different tonal
qualities.
The overtone series of “natural” brass wind instruments (i.e. those without
valves) begins with the fundamental, or so-called “pedal tone,” which is not
played as such. A Baroque trumpet pitched in C produces the following
harmonic series: c, g, c1, e1, g1, b♭1, c2, d2 e2, f♯2, g2, a2, b♭2 b2, c3, c♯3, d3,
e♭3, e3. Most of these harmonics are justly intoned, though the 7th, 11th,
13th, and 14th harmonics are out of tune (Smithers, 1988). To correct
intonation in modern “copies” of Baroque trumpets, finger or “node” holes
are added, though there are no original trumpets with such holes. Modern
trumpets (and other brass instruments) are fitted with extra lengths of tubing
operated by valves, which are used to correct intonation and add notes that
are absent from the harmonic series of the principal tube length.
There is a general misconception that acoustical musical instruments
“amplify” sound. In fact, most are extremely inefficient. For example, it has
been estimated that 150 milliwatts of power supplied by a violin bow results
in only 6 milliwatts of radiated acoustical power, indicating that the violin
has an efficiency of about 4%. What many instruments do is convert nearly
inaudible vibration into sound. For example, a thin vibrating string alone
projects very little sound, but if its energy is conveyed through a bridge to a
broad soundboard, its vibrations are projected into the surrounding air and are
more audible. The bridges of certain instruments have relatively high
impedances, which cause the strings' vibrations to be slowly transmitted to
the soundboard, thus producing a sustained sound. This is important in
instruments where the strings receive an impulsive stimulus, such as by
plucking or striking.

References
John Backus, The Acoustical Foundations of Music (New York and London,
1977).
Francis Bacon, Sylva Sylvarum: or a Natural History in Ten Centuries
(London, 1627).
Arthur H. Benade, Fundamentals of Musical Acoustics (New York, London,
and Toronto, 1976).
Leo L. Beranek, Acoustics (New York, Toronto, and London, 1954).
Theobald Boehm, Die Flöte und das Flötenspiel (Munich, 1871).
Theobald Boehm, The Flute and Flute Playing in Acoustical, Technical, and
Artistic Aspects, trans. Dayton Miller (New York, 1922).
Lothar Cremer, The Physics of the Violin (Cambridge, MA, and London,
1981).
Andrew Dipper, ‘Librem Segreti de Buttegha’: A Book of Workshop Secrets:
The Violin and Its Fabrication, in Italy, Circa 1725–1790 (Minneapolis,
2013).
Neville H. Fletcher and Thomas D. Rossing, The Physics of Musical
Instruments (New York, 2010).
Michael Forsyth, Buildings for Music: The Architect, the Musician, and the
Listener from the Seventeenth Century to the Present Day (Cambridge,
MA, n.d.).
Donald E. Hall, Musical Acoustics: An Introduction (Belmont, 1980).
Hermann L. F. Helmholtz, On the Sensations of Tone as a Physiological
Basis for the Theory of Music (New York, 1954); first published as Die
Lehre von den Tonempfindungen als physiologische Grundlage für die
Theorie der Musik (1863).
Frederick Vinton Hunt, Origins in Acoustics: The Science of Sound from
Antiquity to the Age of Newton (New Haven and London, 1978).
Carleen M. Hutchins, ed., Musical Acoustics, Part 1: Violin Family
Components (Stroudsburg, 1975).
Carleen M. Hutchins, ed., Musical Acoustics, Part 2: Violin Family
Functions (Stroudsburg, 1976).
Carleen M. Hutchins and Virginia Benade, eds., Research Papers in Violin
Acoustics 1975–1993, 2 vols. (Woodbury, 1997).
Charles Hutton, Recreations in Mathematics and Natural Philosophy
(London, 1803).
Athanasius Kircher, Phonurgia nova (Kempten, 1673).
Victor-Charles Mahillon, Elements d'Acoustique Musicale et Instrumentale
(Brussels, 1874; reprinted Brussels 1984).
Pierre Louis Maupertuis, “Sur la forme des instruments de musique,”
Histoire de l'Academié de Mathématique et de Physique (Paris, 1726), pp.
215–226.
Marin Mersenne, Harmonie universelle (Paris, 1636).
Philip M. Morse, Vibration and Sound (New York and London, 1936).
Cornelis J. Nederveen, Acoustical Aspects of Woodwind Instruments
(Amsterdam, 1969).
Francis North, Philosophical Essay of Music Directed to a Friend (London,
1677).
Euro Peluzzi, Tecnica Costruttive degli Antiche Liutai Italiani (Florence,
1978).
Stewart Pollens, “Some Thoughts on the Tuning of the Early Three-String
Violin,” The Galpin Society Journal 64 (March, 2011), pp. 61–66.
Stewart Pollens, Stradivari (Cambridge, 2010).
Lord Rayleigh, The Theory of Sound, 2 vols. (London: vol. I, 1877; vol. II,
1878).
Roberto Regazzi, Il Manoscritto Liutario di G. A. Marchi (Bologna, 1986).
Frederick A. Saunders, “The Mechanical Action of Violins,” Journal of the
Acoustical Society of America 9 (1937), pp. 81–98.
Frederick A. Saunders, “Recent Work on Violins,” Journal of the Acoustical
Society of America 25 (1953), pp. 491–498.
J. C. Shelleng, “The Violin as Circuit,” Journal of the Acoustical Society of
America 35 (1963), pp. 326–338.
J. C. Shelleng, “Acoustical Effects of Violin Varnish,” Journal of the
Acoustical Society of America 44 (1968), pp. 1175–1183.
J. C. Shelleng, “On Vibrational Patterns in Fiddle Plates,” Catgut Acoustical
Society Newsletter 9 (1968), pp. 4–10.
Abbé Sibire, La Chélonomie ou le parfait luthier (Paris, 1806; Brussels,
1823).
Don L. Smithers, The Music and History of the Baroque Trumpet before
1721 (Carbondale and Edwardsville, 1988).
Edward H. Tarr, “The ‘Bach Trumpet’ in the Nineteenth and Twentieth
Centuries,” in Michael Latcham, ed., Music of the Past – Instruments and
Imagination (Berne, 2006), pp. 17–48.
Giuseppe Tartini, Trattato di musica (Padua, 1754).
Vitruvius, The Ten Books on Architecture (New York, 1960).

Analysis of materials

Outlined below are the principal methods currently used by museums.

Gas chromatography/mass spectrometry (GC/MS) is one of the most
powerful analytical techniques used in the identification of organic materials.
The sample is either dissolved in a solvent or heated (the latter in pyrolysis-
GC/MS), which causes it to break down into chemical components. When the
sample is injected into the chromatograph, a carrier gas termed the “mobile
phase” forces the components through a helical column that is packed or
lined with an inert substance termed the “stationary phase.” The column is
heated in a temperature-controlled oven. The components of the sample have
various retention times, and as these elute from the column they are detected
by a flame-ion detector or mass spectrometer. The resulting spectrum is then
compared (by a computer) against a library of spectra of known materials.
This is one of the best methods of identifying the constituents of varnish, as
well as glues, oils, waxes, and other organic materials used in instrument
making.
Fourier-transform infrared spectroscopy (FTIR) employs a beam of
infrared light that either passes through or is reflected by the sample and is
then made to interfere with a reference beam. This produces an interference
pattern that is analyzed by a mathematical procedure termed Fourier
transformation, which yields an absorption spectrum of the sample. This
technique can be used to characterize the types of materials found in varnish;
however, many of the resins, oils, and other ingredients that are typically
used have similar infrared absorption spectra and may be difficult to
distinguish, especially if used in combination.
Raman microscopy directs a laser beam through a microscope objective
onto the sample. A phenomenon termed “Raman scattering” occurs, which
results in characteristic wavelength shifts that can be measured and analyzed.
Because the laser beam can be aimed precisely by the microscope, individual
particles (such as finely ground pigments suspended in a varnish or binding
medium) can be selected for study.
The author employed the above three methods in analyzing several samples
of Stradivari's varnish (see Pollens, 2010). Other researchers have used
similar instrumentation and produced comparable results. Stradivari used a
simple oil and pine resin varnish with relatively opaque inorganic pigments
for coloration (see Varnish).
High performance liquid chromatography (HPLC) works somewhat like
gas chromatography in that the unknown material is dissolved in a solvent,
which is injected into a column (generally much shorter than the coil used in
gas chromatography) packed with a granular sorbent. A high-pressure pump
is used to move the dissolved sample through the column. The dissolved
materials pass through at different rates, and as they emerge from the column,
various types of detectors can be used to identify the dissolved material.
HPLC has been used to detect organic dyes and lake pigments.
The advantage of a technique called thin-layer chromatography is that it
does not require expensive instrumentation. The sample is dissolved in a
solvent and then applied as a spot near the bottom of a sheet of plastic,
aluminum foil, or glass that is coated with a thin film of an inert, absorbent
material such as silica or alumina. Spots of known materials can be placed
alongside the sample for reference. The lower edge of the prepared sheet is
placed in a tray containing a solvent, which draws the individual components
of the spots upwards by capillary action. After drying, the sheet may be
sprayed with a reagent to make the resultant spots and streaks visible when
viewed under various sources of illumination, such as ultraviolet light. The
patterns are generally compared visually to identify the sample.
Unfortunately, many of the resins and other constituents used to make
varnishes produce similar patterns of streaks and spots, which may make
positive identification difficult.
X-ray fluorescence spectroscopy employs a beam of X-rays, which are
directed at the sample. This causes the re-emission of X-rays having the
characteristic energies of different chemical elements, which are plotted and
identified by computer. This is the best procedure for identifying inorganic
pigments and the constituents of metal alloys. It is important to note that X-
ray fluorescence spectroscopy only provides an elemental analysis and does
not identify compounds. Though it can detect carbon, it is not appropriate for
identifying the organic constituents of varnish. X-ray fluorescence
spectroscopy can be carried out with open-architecture equipment (which
does not require removal of a sample or placing the entire object in a
vacuum) or with a scanning electron microscope equipped for energy-
dispersive X-ray analysis (EDX). With the scanning electron microscope,
objects or samples must be small enough to fit inside the microscope's
vacuum chamber, but its main advantage is that it can be used to identify
individual particles on a microscopic scale. Parts like violin bow fittings will
fit inside the microscope's chamber, and the author's analysis of the silver
used in the fittings of a François-Xavier Tourte bow indicate that the alloy he
used to make the rings of one button consists of 88.7% silver and 11.3%
copper, while the reinforcement pins are of 96.9% silver and 3.1% copper.
The rose gold alloy he used to make the rings of another button consists of
74.3% gold, 5% silver, and 20.7% copper, while the solder used to fabricate
the ring is of a lower-melting-point alloy of 47.1% gold, 28.7% silver, and
24.2% copper. A Baroque trumpet attributed to Johann Georg Schmied of
Pfaffendorf was suspected of being a composite. X-ray fluorescence revealed
that the pommel and its underlying section of tubing, as well as other parts,
were made of different alloys of brass, thus confirming these suspicions
(Figure 1).
Figure 1 Composite trumpet with bell rim by Johann Georg Schmied,
Pfaffendorf, 1736, shown prior to restoration.

Another technique employing X-rays is X-ray diffraction, which can be

used to identify corrosion products on metal, as well as salts, minerals, and
other crystalline materials.
Visible light, fluorescence, and polarized light microscopy are valuable
techniques for studying the structure and layers of varnish. The venerable
optical microscope can be fitted with an epi-fluorescence illuminator and
associated filter sets, polarizing filters and rotating stage, heated stage, and
measuring reticules, which makes it a powerful tool for identifying various
substances. For example, the polarizing microscope can be used to identify
pigments by studying their crystalline structure. Epi-fluorescence
illumination of cross sections of varnishes can reveal the presence of layers
and the inclusion of pigments and other colorants (see Varnish: Figure 38).

References
N. S. Brommelle and Garry Thomson, eds., Science and Technology in the
Service of Conservation (London, 1982).
Michele R. Derrick, Dusan Stulik, and James M. Landry, Infrared
Spectroscopy in Conservation Science (Los Angeles, 1999).
Walter C. McCrone, Lucy B. McCrone, and John Gustav Delly, Polarized
Light Microscopy (Chicago, 2002).
John S. Mills and Raymond White, The Organic Chemistry of Museum
Objects (Oxford, 2003).
Stewart Pollens, “Recipe for Success,” The Strad 120/1429 (May, 2009), pp.
34–38.
Stewart Pollens, Stradivari (Cambridge, 2010).
Mary F. Striegel and Jo Hill, Thin-Layer Chromatography for Binding Media
Analysis (Los Angeles, 1996).
Barbara Stuart, Analytical Techniques in Materials Conservation (Chichester,
2007).
Authentication

(see also Analysis of materials, Dendrochronology and the dating of

wood, Inscriptions, faded, Labels, Paper, pencil, and ink)
Because of their special knowledge of materials and craft techniques,
conservators are often called upon to assist in the authentication of musical
instruments. Conventional curatorial expertise and connoisseurship can weed
out most fakes, but sometimes a technical approach is required. In the late
nineteenth century, when many of the world's musical instrument collections
were formed, unscrupulous dealers began to fabricate instruments of all kinds
and easily fooled naïve collectors and museum curators who were bent on
forming encyclopedic collections. The most infamous of these dealers was
Leopoldo Franciolini, who flourished in Florence from around 1879 until
1910, when he was charged with fraud and sentenced to four months'
imprisonment (Ripin, 1974). The Franciolini firm did sell some authentic
instruments, though the composites and altered examples that his workshop
turned out are still puzzling museum curators throughout the world.
Violins have never been of much interest to museums (in part because they
cannot afford the better ones), and the academic organological societies
rarely publish articles about them in their journals or review books written on
the subject. In this world apart, history writing and expertise reside in the
hands of a few dealers, who unfortunately have soaked out most of the
original violin labels and turned the attribution process into a lucrative
business. For example, a whole body of work is attributed to Michele
Deconet, though recent archival research has proven that this itinerant
Venetian street musician was never a member of the appropriate guild, nor is
there any record that he operated a workshop or commercial establishment in
Venice (Pio, 2002). Nevertheless, violins continue to be sold as authentic
Deconets.
Things to look for:

1 Was the instrument ever functional? An example would be a bassoon-

like object carved in the form of a sea serpent, fitted with six finger
holes, bocal, and double reed, yet that is utterly non-functional (it is
not a folded tube like a bassoon or dulcian, but is just a large, vented
 cavity). Nevertheless, such a preposterous nineteenth-century
fabrication (most likely hailing from the Franciolini shop) was
recently recatalogued by a major museum as a Renaissance
processional instrument.
2 Is the instrument consistent with others by the maker, school, region,
and period? An example would be a harpsichord superficially
resembling the work of Bartolomeo Cristofori but having a c′′ string
length of 255 mm. It is unlikely to have been made by him because
his other instruments have c′′ string lengths between 280 and 288
mm. The shorter scaling suggests that this harpsichord was designed
for a higher pitch and probably postdates Cristofori.
3 Are the expected “earmarks” present, such as special layout or tool
marks, brands, and inscriptions? For example, members of the
Serafin family typically branded their violins and carved a small
notch in the upper wall of the pegbox. If these are missing in a violin
ascribed to Santo Serafin, it is likely that the violin is misattributed.
4 Are the pitch, intonation, and compass consistent with the period? An
example would be a fretted clavichord that does not produce a
musical scale. Upon close examination, the soundboard and bridge
seem to be somewhat crude. They are probably replacements, with
the bridge improperly positioned.
5 Are the materials consistent with the supposed date and place of
manufacture? For example, a drum held together with machine-made
nails and bearing a coat of arms painted with synthetic ultramarine
pigment could not have been fabricated in the Renaissance.
Similarly, it is unlikely that Theobald Boehm would have marketed a
flute with a head joint of cocuswood and a body of blackwood. In all
likelihood such a flute is a composite.
6 Are the tool marks consistent with the period of manufacture? For
example, circular saw marks should not be present on original
internal structural parts of a seventeenth-century harpsichord.
Similarly, the bell of a sixteenth-century trumpet should show
parallel, overlapping marks made by a burnisher rather than evidence
of having been spun.
7 Are dates provided by such techniques as dendrochronology and
carbon-14 dating consistent with the supposed date of manufacture?
For example, a dendrochronological date of 1788 would not only call
 into question a label dated 1716 but also the attribution if the
purported maker died in 1737.
8 Examine inscriptions and labels carefully. If a clean, unblemished
label is glued over a dirty, oxidized, and scratched surface, it is not
original. If a knowledgeable observer specifically noted in 1902 that
a vital inscription was absent, but it currently exists, then in all
likelihood it is a forgery.

References
Helen Hayes, mod., “The Messiah: A Panel Discussion,” Journal of the
Violin Society of America 17/3 (2001), pp. 181–222. The author was a
member of the panel.
H. L.C. Jaffé, J. Storm van Leeuwen, and L. H. van der Tweel, eds.,
Authentication in the Visual Arts: A Multi-disciplinary Symposium
(Amsterdam, 1979).
Mark Jones, Fake? The Art of Deception (Berkeley and Los Angeles, 1990).
Mark Jones, ed., Why Fakes Matter: Essays on Problems of Authenticity
(London, 1992).
Walter McCrone, Judgement Day for the Turin Shroud (Chicago, 1996).
Karel Moens, “De Viool[bouw] in de 16de en 17de eeuw, I,” Musica Antiqua
2/1 (1985), pp. 24–26.
Karel Moens, “De Viool[bouw] in de 16de en 17de eeuw, II,” Musica
Antiqua 2/2 (1985), pp. 38–41.
Karel Moens, “De Viool[bouw] in de 16de en 17de eeuw, III,” Musica
Antiqua 2/3 (1985), pp. 85–90.
Karel Moens, “De Viool[bouw] in de 16de en 17de eeuw, IV,” Musica
Antiqua 2/4 (1985), pp. 123–127.
Karel Moens, “Authenticiteitsproblemen bij Oude Strijkinstrumenten,”
Musica Antiqua 3/3 (1986), pp. 80–87.
Karel Moens, “Authenticiteitsproblemen bij Oude Strijkinstrumenten,”
Musica Antiqua 3/4 (1986), pp. 105–111.
Karel Moens, “Authenticiteitsproblemen bij Oude Strijkinstrumenten,”
Musica Antiqua 4/1 (1987), pp. 3–11.
Karel Moens, “Renaissance Gambas in het Brussels Instrumentenmuseum:
Vragen rond Toeschrijvingen, Verbouwingen en Authenticiteit,” Bulletin
van de Koninklijke Musea voor Kunst en Geschiedenis Jubelpark Brussels
66 (1995), pp. 161–237.
Karel Moens, “Problems of Authenticity in Sixteenth-Century Italian Viols in
the Brussels Collection,” in The Italian Viola da Gamba: Proceedings of
the International Symposium on the Italian Viola da Gamba: Magnano,
Italy, 29 April–1 May 2000 (Solignac, 2002), pp. 97–114.
Karel Moens, “Les ‘Violons’ de Charles IX,” Musique, Image, Instruments 5
(2003), pp. 71–96.
Karel Moens and Klaus Martius, “Wie Authentisch ist ein Original:
Untersuchungen an zwei alten Streichinstrumenten des Germanischen
Nationalmuseums Nürnberg,” Concerto 6 (1988), pp. 15–21.
Stefano Pio, Liuteri & Sonadori Venezia 1750–1870 (Venice, 2002), pp. 44–
51.
Stewart Pollens, “Le Messie,” Journal of the Violin Society of America 16/1
(1999), pp. 77–101.
Stewart Pollens, “Messiah on Trial,” The Strad 112/1336 (August, 2001), pp.
54–58.
Stewart Pollens, “Messiah Redux,” Journal of the Violin Society of America
17/3 (2001), pp. 159–179.
Stewart Pollens, Stradivari (Cambridge, 2010).
Kenneth W. Rendell, Forging History: The Detection of Fake Letters and
Documents (Norman and London, 1994).
Edwin M. Ripin, The Instrument Catalogs of Leopoldo Franciolini
(Hackensack, 1974).
Ingrid D. Rowland, The Scarith of Scornello (Chicago and London, 2004).
 B
Benzotriazole

Benzotriazole (BTA) can be used to inhibit the corrosion of new or freshly

cleaned or polished objects made of copper and its alloys. To apply this form
of protection, one should first degrease the object with acetone or other
suitable solvent and then immerse it in a 1–3% aqueous solution of BTA at
room temperature for several minutes. Some conservators prefer to dissolve
BTA in alcohol, which has a lower surface tension than water, or to apply it
under a vacuum to suppress bubbles. BTA can also be swabbed or sprayed
on, but immersion provides more even application. Do not allow BTA
solutions to evaporate to dryness on the object being treated. After treatment,
the object must be washed with distilled or deionized water and then dried.
When copper or copper alloy objects are being chemically cleaned of
corrosion products using alkaline stripping agents (such as alkaline Rochelle
salts, alkaline glycerol, or Calgon), BTA can be added to the stripping agent
to reduce etching of non-corroded areas. If acidic stripping agents are used
(such as citric acid), the object can be pre-treated with BTA.
BTA can be used as a vapor phase inhibitor by immersing paper in a 3%
BTA solution and then allowing it to dry. The impregnated paper can then be
used to wrap or pack copper or copper alloy objects.
BTA is a component of the acrylic varnish Incralac (which consists of
Paraloid B-44 with BTA and a UV stabilizer in a solvent), which can be
sprayed or brushed onto the object.
BTA treatment slightly darkens freshly polished copper and copper alloy
surfaces, which may be undesirable in certain instances.
BTA is suspected of being a carcinogen, so protective gloves should be
worn when working with this chemical.

References
Linda Merk, “The Effectiveness of Benzotriazole in the Inhibition of the
Corrosive Behavior of Stripping Reagents on Bronzes,” Studies in
Conservation 26 (1981), pp. 73–76.
W. A. Oddy, “On the Toxicity of Benzotriazole,” Studies in Conservation 17
(1972), p. 135.
Catherine Sease, “Benzotriazole: A Review for Conservators,” Studies in
Conservation 23 (1978), pp. 76–85.

Bleach

(see also Paper, pencil, and ink, Stain removal, Textile cleaning, Wood)
Bleaching is best avoided as it can weaken the structure of paper, textiles,
and wood, and it does not actually remove the substances that cause stains
but merely reacts with chromophores (structures of certain molecules that
produce color) to de-colorize them. Paper or cloth that is primarily used for
physical reinforcement, such as bellows hinges, should not be bleached.
Bleaching can be accomplished through oxidation or reduction reactions.
Oxidizing bleaches include calcium hypochlorite and sodium hypochlorite (in
solution the latter is commonly known as eau de Javel, Javel water, and
ordinary household bleach), hydrogen peroxide, and sodium percarbonate
and sodium perborate (which are used in some household bleaches available
in powder form). Reducing bleaches include sodium dithionite and sodium
borohydride. Ultraviolet light, and to a lesser extent, visible light, also
produce a bleaching effect.
Bleaching should only be undertaken after testing to determine its effect on
the appearance and physical strength of the object being treated. For example,
paper and textiles made in early times rarely had the super-white appearance
of modern counterparts, and even mild bleaching may reduce or remove their
natural color and give them an unnatural appearance. Certain inks, dyes, and
pigments are adversely affected by bleaching. To test them, a cotton-tipped
applicator (or smaller applicator, if necessary) is moistened with a drop of the
bleaching solution and rolled against the dyed or pigmented area. If the color
is affected, then overall bleaching is unsafe.
 Sodium hypochlorite and other chlorine bleaches are difficult to wash out
of paper, textiles, and wood, and thus cannot be recommended. Furthermore,
they require an antichlor (see below) to halt their bleaching action, and these
agents must also be cleared with thorough washing. Insufficiently cleared
sulfur-containing antichlors (see below) may continue the bleaching process.
The advantage of using hydrogen peroxide is that the only residue left after
bleaching is water. Hydrogen peroxide solution may be used to bleach light
stains in paper, textiles, and wood (see Paper, pencil and ink, Textile
cleaning, Wood). Hydrogen peroxide is often stabilized by the addition of a
small amount of acid (the 3% hydrogen peroxide sold in pharmacies
generally has a pH of about 4.5), which must be neutralized to help release
the oxygen required for bleaching. If the pH is brought much beyond 9, the
energetic release of oxygen in the form of bubbles may be harmful to some
delicate objects. Household-strength ammonia can be used to alter the pH,
though some materials are adversely affected by it (oak, for example, is
darkened by ammonia). If this is the case, hydrogen peroxide can be activated
by other alkalis, such as sodium hydroxide. In order not to dilute the strength
of the hydrogen peroxide, alkalis should be added in relatively concentrated
form, while checking with a pH meter. The advantage of using ammonia is
that no chemical residue is left after the ammonia gas dissipates, while alkalis
such as sodium hydroxide must be cleared after bleaching is completed.
Hydrogen peroxide is sometimes sold in “10-volume” or “20-volume”
concentrations. This indicates that per volume of hydrogen peroxide solution,
10 and 20 times the volume of oxygen gas is released upon decomposition.
10-volume hydrogen peroxide is equivalent to a 3% solution; 20-volume to a
6% solution; 100-volume to a 30% solution. Hydrogen peroxide can be
purchased in the higher concentrations and diluted for use. Commercial two-
part wood bleaches sold in hardware stores start with a wash of sodium
hydroxide followed by a strong hydrogen peroxide solution. This is a very
powerful combination that will remove not only stains but also the natural
coloration of the wood. Protective gloves and goggles should be worn when
using high concentrations of hydrogen peroxide, especially in combination
with strong alkalis (see Safety equipment). When preparing bleaching
solutions, it is important to use distilled or deionized water, otherwise the
bleach may interact with iron or other materials found in tap water and cause
staining.
 Plenderleith and Werner (1979) provide instructions for preparing a non-
aqueous, ethereal solution of hydrogen peroxide by shaking together equal
parts of 20 vol. hydrogen peroxide and ether, and using the ethereal layer that
rises to the top. However, a mixture of hydrogen peroxide and ether is
potentially explosive, so this technique is not recommended.
Antichlors are chemical reducing agents that neutralize the action of
chlorine bleaches. These include sodium sulfite, sodium dithionite, sodium
thiosulfate, sodium borohydride, tetramethylammonium borohydride, and
tetraethylammonium borohydride. A 1–2% solution of sodium borohydride in
distilled or deionized water is an effective antichlor. After treatment with
chlorine bleaches and antichlors, paper and textiles must be thoroughly
washed in distilled or deionized water. Wood objects must be wiped down
repeatedly with distilled or deionized water to clear all of the above reagents.

References
Sheila Landi, The Textile Conservator's Manual (Oxford, 1992).
Guy Petherbridge, Conservation of Library and Archive Materials and the
Graphic Arts (London, 1987).
H. J. Plenderleith and A. E. A. Werner, The Conservation of Antiquities and
Works of Art (Oxford, 1979).

Bluing

Steel is blued by oxidation by subjecting it to a column of hot air from a

torch or in an oven at around 560–640°F (293–338°C).
A chemical bluing agent for steel:

2 parts ferric chloride

2 parts antimony chloride
1 part gallic acid
4–5 parts water

Brass can be blued by dipping into a solution of 5 drams ferric chloride and 1
ounce of sodium hyposulfite in 1 pint of water (Hasluck, Metalworking,
1904).
Another technique is to dissolve copper filings in nitric acid until it
becomes saturated, then dip the brass into this solution, dry it and heat it over
a moderate flame that is free from smoke.

Boxwood

To clean and polish unvarnished boxwood woodwind instruments and

boxwood key tops of harpsichords and other keyboard instruments, first mask
off areas that should be kept clean of oil, such as the surfaces of woodwind
instruments that underlie key pads and the sides of key levers. Wrap a soft
cloth around a fingertip and moisten it with a few drops of raw linseed oil.
Then dip the moistened cloth into rottenstone and rub the surface of the
boxwood to remove accumulated dirt, perspiration, etc. Rottenstone is an
abrasive that gently polishes the wood surface as it removes soil, while the
linseed oil revives the color of old boxwood. After using this mixture, wipe it
off with a soft cloth and let the work rest for a few hours or overnight. Any
oil that rises to the surface should be wiped off and the work polished with a
clean cloth. See Patination for methods of darkening and enhancing the figure
of boxwood.

Brass and bronze alloys

% % %
Common name Additives/properties
copper zinc tin

Gilding metal 95 5

Statuary metal 92 5 2 +1% lead

Medals 89 3 8

Tombac 88 6 6
Similor 85–90 10–
15

Red brass 85 15

Artificial gold 85 15 or 15% tin

Gun metal 85 10 5

Bell metal 75–85 15–

25

Pinchbeck 83 17

Low brass 80 20 very ductile

Dutch brass 80 20

Cymbals and 80 20
gongs

Prince Rupert's 75 25 imitation gold

metal

Rolled sheet 75 25
brass

Ordinary brass 71 29

Cartridge brass 70 30 worked cold

Nickel brass 70 24.5 + 5.5% nickel in

coinage

Admiralty brass 69 30 1 tin inhibits

dezincification

Yellow brass 67 33
British brass 66 34

Alpha brasses >65 <35 worked cold

High brass 65 35 high tensile strength

Common brass 63 37 worked cold

Aich's alloy 60.66 36.58 1.02 1.74% iron

Muntz metal 60 40 trace of iron

Naval brass 59 40 1 tin inhibits

dezincification

Alpha-beta brass 55–65 35– worked hot

45

German silver 50 25 +25% nickel

Speculum metal 50 50

White brass <50 >50 brittle; used as casting

alloy

Beta brasses 50–55 45– worked hot

50

German brass 49 51

German 33 67
watchmaker's

Brass 16–32 68– brittle

84

Britannia metal 2 92 +6% antimony

 Queen's metal 3.5 1 88.5 +7% antimony

Reference
Ernest Spon, Workshop Receipts for Manufacturers and Scientific Amateurs
(London, 1909).

Brass and nickel silver cleaning

(see also Benzotriazole, Brass instruments, Bronze disease,

Electroplating and electrocleaning)
The precise chemical nature of corrosion products can be determined by X-
ray diffraction and other analytical techniques, though they can often be
characterized by visual examination: acetates of copper are generally green;
carbonates of copper are blue-green; chlorides are green or brown; acetates
are green; carbonates are blue-green; chlorides are green or brown; formates
are blue; oxides are brown, reddish brown, or black; sulfates are blue; and
sulfides are gold, brown, and black. Reddish areas on brass may indicate
dezincification, which is often the result of contact with acids and handling
(thus it is not uncommon to find evidence of zinc depletion where cavalry
bugles were held while playing). Ammonia, which is often used to remove
corrosion, will also extract zinc from brass. Dezincification and the formation
of sulfides may result in “season cracking.” One cause of the formation of
sulfides on brass instruments, as well as “red rot” in leather, was the use in
the nineteenth century of gas lamps, which gave off sulfur dioxide. Gas
lamps were first used at the beginning of that century and their use continued
until the gradual switch to electric light bulbs that began in the 1880s was
complete. Sulfur is given off during the combustion of gasoline as well as by
many rubber products. Storage in commercial musical instrument carrying
cases may cause the formation of copper acetate and copper formate on the
surface due to the acidic vapors exuded by various woods and other materials
used in their construction. Wood carrying cases that have been fumigated by
the Thermo Lignum Warmair process (which uses heat to kill woodworm and
other wood-boring pests) can initiate the release of large quantities of acetic
and formic acids from the wood that will quickly react to form various
corrosion products on metal instruments stored in them. Moisture and the
presence of salt (for example, from a drop of sweat) can initiate so-called
“bronze disease” in brass, which appears as crusty or powdery pale green
spots (see Bronze disease).
To remove light corrosion products from copper, brass, bronze, gilt brass
or bronze (ormolu), nickel, and nickel silver (also known as German silver):

1 Disassemble the instrument (see Brass instruments).

2 Remove superficial coatings such as lacquer, varnish, wax, and oils
with acetone or, if very resistant, with methylene chloride (see
Solvents and solvent cleaning).
3 If tarnish is superficial, polish with a slurry of precipitated chalk in
ethanol or deionized water. If a high polish is required, use a slurry
of optical-quality rouge or gamma alumina (the latter is available in
grit sizes of 1.0, 0.3, and 0.05 µm). Deagglomerated forms of
abrasives are preferred; a nonionic surfactant such as Triton X-100
can be added as wetting agent.
4 To retard tarnishing of copper, brass, and bronze, immerse the object
in a 3% solution of BTA for several minutes at room temperature.
BTA may darken brass somewhat (see Benzotriazole). Do not let the
BTA solution dry on the surface, but rinse it off thoroughly after
treatment. Rinses should be carried out in tap water rather than
distilled or deionized water if the instrument has been assembled
with soft solder (see Lead and lead alloys.)
5 Light copper corrosion products can be removed from nickel silver or
nickel-plated instruments by swabbing with 5% formic acid,
followed by thorough rinsing in tap water.
6 Coat with Agateen no. 27 diluted 1:1 with lacquer thinner for spraying
or Paraloid B-48N, 10–12.5% solids in xylene for spraying and 15%
solids for brushing, or Incralac (contains Paraloid B-44,
benzotriazole, and a UV stabilizer) at similar concentrations. See
discussion of acrylic resins under Varnish.

More heavily corroded brass instruments can be immersed in a 5% solution

of trisodium ethylenediamine tetra-acetic acid (EDTA), again after removing
any coatings. This treatment may take several hours, during which time the
surface of the instrument should be periodically rubbed with a soft bristle
brush. As above, rinse thoroughly in tap water, lightly polish with a slurry of
precipitated chalk or other abrasives to even out the surface color, treat with
benzotriazole, and apply a protective coating if desired. Trisodium EDTA can
be prepared by mixing disodium and tetrasodium EDTA, using equal
proportions according to their formula weights (FWs). The FW of disodium
EDTA dihydrate is 372.24 while that of tetrasodium EDTA dihydrate is
416.20. (See Figures 2a and 2b.)

Figure 2a Anonymous brass bugle before treatment with trisodium EDTA.

Figure 2b Bugle after treatment and light polishing with a slurry of
precipitated chalk.

According to Spon’s Workshop Receipts (1896), the method used for

cleaning brass in the United States arsenals during the late nineteenth and
early twentieth centuries was as follows:
 Make a mixture of 1 part common nitric acid and ½ part sulfuric acid in
a stone jar, having also ready a pail of fresh water and a box of sawdust.
The articles to be treated are dipped into the acid, then removed into the
water, and finally rubbed with sawdust. This immediately changes them
to a brilliant color. If the brass has become greasy, it is first dipped in a
strong solution of potash and soda in warm water; this cuts the grease,
so that the acid has free power to act. Rub the surface of the metal with
rottenstone and sweet oil, then rub off with a piece of cotton flannel, and
polish with soft leather. A solution of oxalic acid rubbed over tarnished
brass soon removes the tarnish, rendering the metal bright. The acid
must be washed off with water, and the brass rubbed with whiting and
soft leather.
Regarding musical instruments made of brass, apply a mixture of 1 part
common sulfuric acid and 12 of water, mixed in an earthen vessel, and
afterwards polish with oil and rottenstone, well scouring with oil and
rottenstone, and using a piece of soft leather and a little dry rottenstone
to give a brilliant polish. In future cleaning, oil and rottenstone will be
found sufficient. Take a strip of coarse linen, saturate with oil and
powdered rottenstone, put round the tubing of instrument, and work
backwards and forwards; polish with dry rottenstone. Do not use acid of
any kind, as it is injurious to the joints. To hold the instrument, get a
piece of wood turned to insert in the bells; fix in a bench vise. The piece
of wood will also serve for taking out any dents you may get in the bells.
Oil and rottenstone for this purpose, are, though very efficacious,
objectionable on account of dirt, on account of the oil finding its way to
the pistons, and because the instrument cleaned in this manner so soon
tarnishes.

So-called “bright dips” employing strong acids (such as recommended in

Spon's Workshop Receipts) will attack the metal and leave the surface etched.
They are not used in modern conservation work, nor is rottenstone used as a
metal polish. However, highly refined and finely graded polishing abrasives
(such as laboratory grades of alumina) that are recommended in today's
conservation literature can produce a scratch-free, mirrorlike finish that
probably would not have been encountered on nineteenth-century military
marching-band instruments, either as they emerged newly made from the
workshop or after cleaning and polishing according to the procedure
described in Spon’s Workshop Receipts.
Household-strength ammonia is often advocated for removing greenish
corrosion products (often copper acetate or copper formate) from nickel
silver; however, ammonia can leach zinc from brass, which may lead to
embrittlement and stress-cracking. Therefore, objects made of thin sheet
brass (such as brass instruments) should not be treated with ammonia. More
massive objects (such as key work, valve caps, etc.) can be swabbed with
household-strength ammonia, but it should not be allowed to remain in
hollows or crevices.
Back in 1976, the author met the noted brass instrument historian Horace
Fitzpatrick, who related his recommendations for cleaning brass instruments:

1 Soak in Calgon (sodium hexametaphosphate).

2 Swab out and wipe surface while still wet.
3 Immerse in a solution of Rochelle salt and caustic soda.
4 Clean off with distilled water.
5 Polish with Solvol Autosol (commercially available from auto
accessory shops) or fine aluminum oxide in benzene.
6 Treat with weak solution of benzotriazole and distilled water.
7 Wipe off; then coat with light film of microcrystalline wax.
8 Polish with a soft cloth.

Note: Solvol Autosol and other paste polishes intended primarily for
polishing automobile chrome are aggressive and should be avoided in
conservation work. Soaking in sodium hexametaphosphate is a standard
conservation procedure for removing calcareous accretions on ancient
bronzes that have been excavated or dredged up from the ocean floor, but it is
rarely required of modern-era brass instruments that have not suffered such
indignities.

References
P. Fiorentino, M. Marabelli, M. Matteini, and A. Moles, “The Condition of
the ‘Door of Paradise’ by L. Ghiberti: Tests and Proposals for Cleaning,”
Studies in Conservation 27/4 (November, 1982), pp. 145–153.
Donny L. Hamilton, Methods for Conserving Archaeological Material from
Underwater Sites (www.nautarch.tamu.edu, 2010).
Colin Pearson, Conservation of Marine Archaeological Objects (London,
1987).
H. J. Plenderleith and A. E. A. Werner, The Conservation of Antiquities and
Works of Art (Oxford, 1971).
Ernest Spon, Workshop Receipts for the use of Manufacturers, Mechanics
and Scientific Amateurs, 5 vols. (London and New York, 1896). The
reference to cleaning brass instruments in US arsenals is on p. 111 of the
1896 edition. It does not appear in the later edition, Ernest Spon, Workshop
Receipts for Manufacturing and Scientific Amateurs, 4 vols. (London and
New York, 1909).
T. Stambolov, The Corrosion and Conservation of Metallic Antiquities and
Works of Art: A Preliminary Survey (Amsterdam, n.d.).
Glenn Wharton, Susan Lansing Maish, and William S. Ginell, “A
Comparative Study of Silver Cleaning Abrasives,” Journal of the
American Institute for Conservation 29/1 (Spring, 1990), pp. 13–31.

Brass instruments

(see-also Brass and nickel silver cleaning, Electroplating and

electrocleaning, Lubricants, Metallurgy, Metalworking, Silver cleaning)

Disassembly

Tuning slides, valves, and caps that were lubricated with commercial
greases and oils and then stored for many years are likely to become
immovable. Lightweight mineral oil or mineral spirits can be injected to try
to loosen these parts, but warming the area with a hot-air gun seems to be the
most efficient way of freeing up seized parts. Hot-air guns can generate
enough heat to melt soft-solder joints (see below and Soldering), so they
should be adjusted or held at a sufficient distance so that the joints are not
affected. A simple oven thermometer can be used to adjust the temperature.
Keep in mind that the lowest melting point for tin/lead alloy soft solders is
361°F (183°C), while “extra-easy” silver solder melts at 1145°F (618°C).
Sensitive areas can be protected with heat sinks or shaded with aluminum
foil.
A cloth or leather strap can be wrapped around the bouts of tuning slides
and used to pull them out, but it is important to distribute the applied force to
prevent kinking or distorting the metal. Leather work gloves are
recommended for getting a grip on recalcitrant valve caps. Moistening leather
often gives it a better grip on metal parts. Commercial mouthpiece-pullers
may be effective in removing stuck mouthpieces, but these should be lined
with cork or leather to prevent damage to the instrument and mouthpiece.
Bare metal-jawed pliers should never be used to loosen parts, as they will
mar the metal or distort threaded tubing, rendering disassembly even more
troublesome. Plastic-lined and curved-jawed pliers designed for the assembly
of optical equipment may be useful for grasping valve covers and other
threaded parts. Hammers (even those having rawhide or plastic heads) should
not be used for tapping on seized parts, as this technique is generally
ineffective and may damage the instrument.

Removing dents and creases

Metal that is dented has often been stretched by the impact. When
correcting dents it is important to avoid stretching the metal further, but
rather to encourage compression of the stretched areas by judicial use of
pressure and support. A good deal of the work involved in removing general
distortion and even pressing out large, shallow dents can often be
accomplished with hand pressure alone. Small dents and creases in bells and
bell stems can be repaired with burnishers and various metalworking stakes
anchored to a stable bench. A “spear” or “fish-shaped” burnisher can be made
by taking an old double-curved crossing file, grinding off its teeth on a belt
sander, and polishing it to a high luster using emery, crocus, and chromium
oxide. The combination of curves and the pointed end enables it to reach
most dents while accommodating the curves of bells and flared tubing. The
annealed end of a half-round burnisher can be hammered to form a hook,
which is useful for removing dents in cylindrical tubing. The burnisher is
given a back-and-forth motion while the instrument is supported by or rotated
against the curved section of a mandrel, bell iron, blow-horn anvil, or stake.
Good hand coordination is required, as the burnisher and stake must be in
alignment with the damaged area in order to smooth it out. Soapy water can
be used as a lubricant when burnishing, though to protect the patina when
rubbing out dents, a thin sheet of mylar can be interposed between the
burnisher and the surface of the instrument. Rubbing out dents may not
require annealing, though kinks and deep creases may require localized
annealing (see below) to soften the metal sufficiently to prevent cracking.
Some dents may be hammered out by tapping with lightweight
silversmith's hammers (those weighing 100 grams or less are recommended;
Figure 3). Mallets having heads of resin-impregnated rawhide, cow horn,
plastic, or an alloy of lead and tin (see “soft hammers” under Metalworking)
are useful when reshaping grossly deformed bells and flared tubing. When
using hammers in conjunction with stakes, it is important to avoid making
hard contact, as the metal may stretch. The silversmith's concepts of loose
and tight regions in distorted sheet metal may be helpful when attempting to
restore the shape of brass instruments. If a flat sheet of metal is hammered
centrally, the center will expand (or become loose), forming a bulge that can
be snapped back and forth like the bottom of an oil can. This internal tension
can be relieved by hammering around the perimeter of the bulge, thereby
freeing up the central area. Alternately, by hammering the perimeter of a flat
sheet of metal, the center becomes tight, which is manifested by a buckled
perimeter. In this case, hammering the center will reduce the buckling.
Annealing cancels out almost all stresses that develop in fabrication or repair,
but it will not undo bulges and buckling due to stretched or distorted surfaces.
Figure 3 Dent removal tools, including stakes and hammers.

In the case of straight cylindrical tubing, a mandrel can sometimes be

inserted and the dent burnished or tapped around its circumference from the
outside. This will tend to compress the dented metal and restore its original
shape. Dent balls (ovoid slugs of steel that are available in sets from brass-
instrument supply houses) can be inserted into curved tubing to press out
dents. One starts with a dent ball that is just large enough to clear the dented
area and progresses through a succession of larger balls until one reaches the
full internal diameter of the tubing. Dent balls can be drilled, tapped, and
mounted on a handle so that they can be inserted into the instrument. If the
dent is out of reach of a handle, dent balls may be used loose by use of
“drivers,” which are heavy slugs of metal of lesser diameter that are inserted
in front of and behind the dent ball. The instrument is then quickly turned up
and down so that the drivers knock against the dent ball and push it through
the damaged tubing. Because it is sometimes tricky to move and retrieve dent
balls in this way (there is always the fear that they may get stuck), they can
be linked by a braided metal cable to a handle and a flexible extension made
of tightly coiled spring-steel wire. The ball can then be pushed to the desired
spot by the handle and withdrawn by pulling on the cable. The Dentmaster is
a commercial version of this tool. Dent balls should be polished to a smooth
finish (rather than roughly turned or ground) to prevent scratching the bore. A
thin coating of light oil will assist in insertion and retrieval. When using dent
balls, it is important that they do not exceed the inside diameter of the tubing
at the point where the dent is located, otherwise the tubing will be stretched,
leaving a bulge or ridge. Dent balls can also be brought into position with
strong magnets (either permanent rare-earth magnets or electromagnets).
Magnets fitted with curved faces can be used in conjunction with special
spherical dent balls (which roll along with the magnet) to burnish out dents.
“Smooth-hammering” is a technique used to even out straight sections of
tubing after dents have been pushed out with dent balls or by other means. It
employs a fluted hardwood or metal block that is shifted back and forth over
the damaged area while being lightly tapped with a hammer.
After major dents have been removed from cylindrical tubing, a drawplate
or draw rings can be used to restore roundness. Bowed tubing can be
straightened by drawing it through a drawplate while applying slight pressure
to the outside curve, thereby compressing and straightening it. If necessary,
the graduated holes in drawplates and rings can be enlarged to the required
diameter with reamers.
The dent roller (available from brass-instrument supply houses), also
known to sheet metal fabricators as the “English wheel,” consists of a
rotating burnisher that is brought to bear against the damaged surface, which
is supported by an appropriately shaped shoe or mandrel. Because the
burnisher rotates, it is less likely to scratch the instrument. This tool is very
effective in removing dents and creases, especially those in the bells and
flared sections of very large brass instruments.
A snarling iron is sometimes helpful in reaching dents inside flared
sections. It is positioned so that its dome is directly below the dent. The shaft
of the iron is tapped with a hammer, which causes the dome to rebound and
impact the dent, thereby raising it.

Annealing

If annealing (softening) is required to soften the metal in order to remove

deep creases or repair heavily damaged areas, it is important to keep in mind
that soft solder, and even most hard solder joints, will come apart at
annealing temperatures, which are in the range of about 800–1300°F (427–
704°C) for brass, 1000–1400°F (538–760°C) for nickel silver, and 1200°F
(649°C) for silver. Brass is annealed by heating until it glows a light red
color, waiting for the red glow to subside, and then quickly dipping it in cool
water. Keep in mind that soft solder will pit or burn through brass and nickel
silver at around 500°F (260°C), which is considerably below the tempering
point. Heat sinks made of copper or aluminum should be used to isolate areas
that might be sensitive to high temperatures. The temperature required of
annealing will ruin silver plating, so the plating has to be removed
beforehand. This is generally carried out by dipping the instrument in a
stripping bath consisting of saltpeter (potassium nitrate) and 30% sulfuric
acid in a proportion of 1:32. This bath will quickly dissolve the silver plating,
though it will also attack the base metal if it is left in too long. After
annealing and repair, the instrument will then have to be replated (see
Electroplating and electrocleaning). Obviously, instruments of historic
importance should not be stripped and replated (see Silver cleaning for
methods of cleaning silver-plated brass). Annealing will also undo the work-
hardening of metal that occurred during manufacture as well as any
intentional heat treatment.

Repairing cracks

Soft-soldered patches are often used as an expedient in commercial repairs,

though they are unsightly and are generally avoided in museum restoration
work. However, if a decision is made to use a patch, it can be cut from 0.25
mm brass stock and rounded to fit by pulling the patch and a piece of tubing
(having the same diameter as the piece to be repaired) through a drawplate.
The patch is then soft-soldered in place, or it can be glued with Paraloid B-
48N (50% in acetone) or more securely with epoxy. Cracks are often mended
with soft solder, but this too is unsightly. Soldering cracks in brass with gold
solder or brazing will make a less noticeable repair, while silver solder
provides a reasonably good match with nickel silver (see below and
Soldering). However, before applying the high temperatures they require, it is
necessary to identify the types of soldering and brazing techniques that were
employed in the instrument's construction. For example, the seams of tubing
and bells may have been brazed or silver-soldered, while bouts and stays may
have been soft-soldered. Heat sinks or shielding may be required to protect
soft-soldered joints if silver solder or spelter is used in repair.
 Before soldering or brazing cracks, they should be leveled, and if necessary
burnished or lightly hammered to stretch the metal in order to eliminate any
gaps. The edges of the crack must be cleaned of oxidation and corrosion
(either mechanically or by swabbing with a pickling acid) otherwise the
solder will not bond with the metal. Hard solders and spelter can be filed into
granular form, made into a slurry with the appropriate flux (see below), and
applied to the crack with a brush. Use the minimum amount of solder or
spelter because it is difficult and potentially injurious to the instrument to
remove any excess. After cooling, carefully level the repaired area with a
sharp scraper and fine abrasive cloths.

Removing scratches

Scratches in brass can often be minimized by burnishing. If a burr was

created while the object was scratched, rubbing with a polished steel
burnisher will tend to push the raised burr back into the scratch. When
burnishing scratches, the burnisher should be directed along the scratch rather
than across it.
When removing scratches by abrasion, the patina and a significant amount
of metal will be lost, so scratches should only be removed by this technique if
they are especially disfiguring. If the decision is made to remove a scratch by
abrasion, it is pointless to rub away at a deep scratch with very fine abrasives.
Commence with a grit that is just coarse enough to quickly get through the
scratch, switching to progressively finer grits until you match the finish of the
untouched areas. For most scratch repairs, it is helpful to use cloth- or paper-
backed abrasives, as the backing tends to provide enough stiffness to help
level the area. If the scratch is on an uneven surface, balls of cotton wadding
or cheesecloth moistened with water and dabbed with abrasive powders will
conform to the topology of the surrounding area. Dry abrasives come in a
variety of grit sizes: silicon carbide is available in 40, 60, 80, 120, 220, 320,
400, and 600 grits; aluminum oxide in 25, 15, 12, 9, and 5 micron grit sizes
(see Grit size comparison chart); and rouge, cerium oxide, and gamma
alumina in grit sizes down to 0.3 micron and 0.05 micron in water
suspension. Most of the coarse-grit abrasives are also available in grease
suspensions. Abrasives should be applied across the scratch, crisscrossing
with successively finer grits until the surface is brought up to the appropriate
level of finish. Abrading in the direction of the scratch will deepen it. The
texture of cloths used in polishing affects the way abrasives work; for
example, cloths with a prominent weave may become clogged with abrasive
slurry, causing streaks or patterns to appear. Motorized buffing wheels should
not be used to remove scratches because they remove metal very quickly and
can erode engraving and other details of workmanship.

Soldering and brazing (see-also Soldering)

Again, it should be noted that any heating process will alter the physical
and acoustical characteristics of most metals. If heating and soldering are
inappropriate, synthetic adhesives (such as Paraloid B-48N, 50% in acetone,
or epoxy) can be used to reattach parts such as loose stays, or may even be
used in the reassembly of bouts and yards. Adhesive repairs, however, may
not be strong enough for instruments that will be frequently handled or
played, and when designing exhibition mounts, special precautions should be
taken to support the weight of parts that have been attached in this way.
Another approach to sealing small cracks, pinholes, or leaks is to fill them
with wax (such as beeswax or a microcrystalline wax, such as Cosmoloid
H80), a wax/resin mixture, or a hot-melt adhesive. An advantage of using
sealants and adhesives is their reversibility with solvents.
As indicated above, soft soldering, hard soldering, and brazing are three
distinct techniques used in the manufacture of brass instruments. Hard
soldering and brazing require high temperatures and are used to construct
tubing and to join up the seams of bells. Soft soldering is done at much lower
temperatures and is used to join lengths of tubing to bouts and to attach stays,
garlands, bosses, and other parts. Silver solder (which is classified as a hard
solder) is available in various alloys of silver, copper, and zinc that have
melting points ranging from “extra-easy” 1207°F (653°C) to “hard” 1450°F
(787°C). This range of melting points enables parts to be soldered in
sequence without loosening others that were previously soldered. Gold solder
is available in different alloys to match the colors of various karats of yellow
gold. These melt at somewhat higher temperatures than the silver solders,
1256–1512°F (680–822°C). Spelter is made by adding extra zinc to brass in
order to yield a melting point that is marginally below that of the typical
brass alloy used to make instruments, which is around 1710°F (930°C). For
example, one commercially available spelter has a composition of 59%
copper to 41% zinc and a flow point of 1650°F (899°C). Hard soldering and
brazing require heating the workpiece to a cherry-red color so that the solder
or spelter will flow. Soft soldering is done with tin/lead alloy solders that
have melting/flow points ranging from 361°F (183°C) to 594°F (312°C). A
60/40 tin-to-lead ratio solder is often used in the construction and repair of
brass instruments. This alloy has a melting point of 361°F (183°C) and a flow
point of 374°F (190°C). Between these temperatures, it is in a pasty state,
which is an advantage if workpieces have shifted in the course of heating and
need to be realigned before the solder has solidified.

Soldering technique

Pieces to be soldered must be free of oxidation and corrosion products

where they are to be joined, and it may be necessary to abrade or scrape the
area to expose a fresh surface. The application of the appropriate flux is also
important because it prevents the workpiece and solder from oxidizing during
heating and enables the solder to flow smoothly and bond with the metal
parts. Solder will bead up and fail to adhere to oxidized, tarnished, or poorly
fluxed work. Several fluxes can be used in soft soldering: sal ammoniac
(ammonium chloride), which is mixed with water to form a paste; so-called
“killed acid” flux (made by placing pieces of zinc in hydrochloric acid until
the evolution of hydrogen gas stops); and rosin (which can be crushed and
applied in powdered form or ground up with petroleum jelly to form a paste).
Rosin flux is less corrosive than acid flux. It can be ground with petroleum
jelly to form a convenient paste. The traditional flux for silver and gold
solders is made with borax (sodium tetraborate), which is mixed with water
to form a paste. A paste of cream of tartar (potassium tartrate or bitartrate) is
used for brazing. To protect old soft-soldered joints that might be affected by
the application of heat during subsequent soft soldering, a paste of
commercial whiting (calcium carbonate) and gum arabic can be painted on
the soldered joint and allowed to dry before applying heat. Soft and hard
solders can be prevented from flowing beyond the joint by painting the
surrounding area with a slurry of yellow ochre mixed with water or ethanol
and allowing it to dry before applying heat. Firescale can be prevented when
heating copper alloys (including sterling and coin silver) by dipping or
spraying the work with a solution of boric acid in methylated spirits (15 g
boric acid/50 ml methanol) or a combination of borax, trisodium phosphate,
and boric acid in water (80 g borax/80 g trisodium phosphate/120 g boric
acid/1 liter of water) and allowing it to dry before applying heat.
When soldering, use the minimum amount of solder, as any excess is
unsightly and difficult to remove. A soldering iron may be sufficient for soft-
soldering thin pieces of metal, but heavier work may require a propane torch.
For silver soldering, a jeweler's air/acetylene torch (such as a Prest-O-Lite
unit) is generally adequate, though a tip should be selected that provides a
wide enough flame. When adjusting an acetylene torch for soldering, use a
fuel-rich reducing flame rather than an oxidizing flame (an oxidizing flame
forms a sharp cone and makes a loud hissing sound). Solder tends to flow in
the direction of heat, so the soldering iron or torch should be positioned to
heat the workpiece rather than the solder. It is also important that the pieces
being joined are heated uniformly; for example, if a delicate part is being
soldered to a massive one, the heat should be applied to the massive part and
conducted to the delicate part. If feasible, heat sinks should be positioned to
prevent heat from spreading to previously soldered areas or parts that must be
kept cool. Hard soldering should not be attempted in the vicinity of soft-
soldered joints, as soft solder will burn through brass, silver, and gold at the
temperatures required for melting hard solder. If it becomes necessary to
make a repair with hard solder, any soft solder that cannot be shielded from
heat must be removed, either mechanically (by scraping or use of abrasive
cloths) or by cleaning with a mixture of glacial acetic acid and 6% (20 vol.)
hydrogen peroxide (3:1). In general, a minimal amount of solder should be
used. Soft solder in the form of ribbons or wire should not be applied at the
tip of the soldering iron or torch, but allowed to flow along the heated and
fluxed area being joined. Paillons (little squares approximately 1 mm square
and 0.25 mm thick) of silver and gold solders can be arranged around the
joint using a pair of tweezers or a soldering pick. If the solder does not flow,
the metal may not be clean or not enough heat is being applied; if the solder
balls up, there may be too much heat or not enough flux; if the solder flows
away from the joint, the source of heat may be closer to the solder than to the
joint or there may not be enough flux.
After soldering, it is necessary to remove the fused flux and oxidation
products in a pickling solution. For silver and brass instruments, a 30%
solution of sodium bisulfate or 10–15% sulfuric acid can be used, warmed to
about 360°F (182°C). Glass or ceramic tanks should be used to hold pickling
solutions. Be careful not introduce iron or steel objects (such as binding wire,
tongs or forceps) into a pickling bath as this may cause dissolved copper to
plate out on the instrument.

Repairing slides, valves, and keys

By far one of the greatest problems encountered with slides and valves is
disassembling them. Bare metal wrenches or pliers should not be used to
remove stuck parts, as they will mar or deform threaded parts – see
recommendations above.
During the nineteenth century, a great variety of valve systems was
developed, including the disc, box, Vienna (or double-piston), Berlin, Perinet,
and rotary valves. Though they may appear complex, their workings can
usually be understood upon examination. When disassembling valve parts,
sketches and photographs should be made if you are unfamiliar with the
system at hand, though it should not assumed that they were properly
reassembled in the past. Check the numbering and other markings (such as
scratch or prick marks) on valves, caps, etc. to determine if they are in the
proper sequence before removing them. Piston casings and covers were
generally numbered in order starting from the mouthpiece end of the
instrument; scratch marks are often found on rotary bearing plates and valve
casings to help realign them during reassembly. Closely examine the knurling
and decorative turning on valve covers and touches as these were often
replaced with ill-fitting parts cobbled from other instruments.
When removing screws and other parts, keep them in their proper sequence
by pressing them into a polyethylene foam block or labeling them. Dents in
valve casings can often be pushed out with dent balls, and those that are out-
of-round may be trued up with adjustable mandrels. One feature of some
early pistons is the use of cork plugs to seal the air passages; these must be
carefully pressed out before pistons are immersed in any cleaning solution.
Unfortunately, these cork plugs are often found in a shrunken or cracked state
and consequently do not seal properly, so it may be necessary to replace them
if the instrument is to be made playable. Cork plugs can be turned on a lathe
using a sharp knife supported on a raised tool rest. The edge of the knife
should be held nearly parallel to the cutting plane and light cuts made until
the proper diameter is reached. A sandpaper file can also be used to smooth
cork that is being turned on a lathe. A razor blade makes an excellent parting
tool when turning cork.
Pistons should not rotate in their casings; if they do, the valve key (or pin)
or key guide may have broken off or become worn. Some valve keys are
screwed on, while others are soldered in place. If soft-soldered, they can be
resoldered (though one must make certain that other parts will not be affected
by the heat); worn hard-soldered keys are best filed off and a new key fitted.
Hard-soldering a new key in place is dangerous as the heat may damage the
piston, so soft soldering is safer in this instance.
If old corks and cloth washers need to be replaced, use a valve gauge to
check whether the old ones bring the pistons into correct alignment with the
ports in the casings. The gauge is inserted into the valve casing and the depth
of the port is established. The setting on the gauge is then compared to the
corresponding point on the piston to establish the correct thickness of felt
washers and/or corks. If the old corks and felt washers are not the proper
height (very often they have become compressed through use) use an arch
punch to make new ones. When properly set up, the finger buttons should all
rise and fall to the same height. Spiral valve springs should be the proper
length so that the coils do not collapse upon themselves when the pistons are
fully depressed. If they are too long, the coils may collide and make a
clicking sound when the valve is depressed. The tops and bottoms of springs
should be normal to the axis of the spring and stand straight and centered so
that they do not scrape along the side of the valve casing. All of the valve
springs should have the same number of turns and provide equal resistance to
the touch.
Rotary valves are turned to a slight taper and generally rest on a removable
bearing. The lower screw cap is turned to adjust for play and wear. String
valves employ a gut cord that is about the diameter of a gut violin A string
(approximately 0.7–0.8 mm). If these must be replaced, note how the original
is dressed and copy this system. If the old cork bumpers must be replaced,
replacements should be carefully trimmed to the correct width to limit the
rotational motion of the valve so that the ports and conduits line up. Some
rotary valves use articulated crank actions with clock springs encased in
cylindrical spring boxes. If these springs are broken, replacements can be cut
from commercial clock mainsprings (which are available in a wide variety of
thicknesses and widths). A ratchet device is often fitted that allows the player
or restorer to “dial in” or adjust the speed and comfort of the valve action.
It may be unclear whether the scoring of pistons, rotaries, and casings, or
the lack of a good fit is an original manufacturing trait or due to wear. In
either case, pistons, rotaries, and other moving parts of historic brass wind
instruments should never be electroplated with copper or nickel in an effort to
“restore” their dimensions, nor should they be reground, lapped, or honed to
fit in accordance with modern standards.
If new cylindrical tubing must be fabricated, flat stock is cut out, shaped
around a mandrel, and both pulled together through a drawplate. The edges
that form the seam are filed to form a lap joint, and brazed or hard-soldered.
After brazing or hard soldering, the seam is cleaned up by scraping or filing.
Straight tubing may then be trued by again pulling it through a drawplate,
either manually or with the aid of a hydraulic draw bench. Slightly flared
tubing, such as trumpet lead-pipes, can be formed on a steel mandrel by using
a hydraulic draw bench or press. The mandrel is turned on a lathe to the exact
dimensions of the lead-pipe bore and then inserted within an oversize length
of brass tubing, and both are drawn, small end first, through a disposable
brass draw ring, which compresses the tubing against the mandrel.
To bend tubing, it must be temporarily filled to prevent it from collapsing
during the bending process. This can be done with soapy water (which is then
frozen), molten pitch (which is allowed to cool and harden), or molten lead
(which is allowed to cool and harden). If lead is used, the inside of the tubing
should be coated with oil so that the lead does not adhere and can be cleanly
melted out after bending. As a precaution, the tube must be free of water
droplets, as they may turn to steam when the molten lead is poured in and
cause the lead to erupt from the tubing. When pouring molten lead, wear
insulated gloves and eye protection or a face shield, and do not stand directly
over the open end of the tubing. Low-melting-temperature alloys, such as
Cerrobend, may also be used. Even though such alloys melt at relatively low
temperatures (some fuse below the boiling point of water), one must still
exercise care when handling them in a molten state. For example, do not pour
molten Cerrobend into a Pyrex glass tray or beaker, as it will shatter violently
due to the sudden and unequal application of heat. Instead, empty the molten
metal into a metal pot or can. Once the filler has frozen or hardened, the
tubing can be bent using a bending block or jig (Figure 4). If some flattening
occurs, the bent tubing can be rounded up by pushing dent balls through it or
by using a balling-out die.

Figure 4 Bending a tube filled with lead or pitch at Boosey & Co. (from
Algernon Rose, Talks with Bandsmen, 1894).

When joining sheet metal together to make highly flared tubing, such as
the bells of trumpets, the seam is generally “stitched” by cutting notches in
one of the edges, thus forming a series of tabs that are hammered down over
the adjoining edge. Sometimes a “gusset,” or an extra piece of roughly
triangular shape, is inserted when fabricating bells. After brazing or hard
soldering, the seam is flattened by hammering or rolling, or filing; the flared
tubing is then roughly hammered to shape on a blow-horn stake. The bell can
then be mounted on a metal spinning lathe, where it is directed against a
mandrel with long-handled burnishers. Widely flared bells, such as those
found on modern trombones and tubas, are often made by mounting discs of
brass on a metal spinning lathe. After the bells are spun, they are trimmed on
the lathe, and reinforcing bell wires are turned in with a notched burnisher or
roller. Soft or silver solder is then run into the rim to prevent the bell wire
from vibrating.
Unusually shaped parts (such as the bells of saxophones) that cannot be
formed by spinning are sometimes manufactured using a hydraulic blowout
press, which uses pressurized liquid to force the metal to expand into a die. In
the nineteenth century, firms such as Victor and Joseph Mahillon in Belgium
used pressurized steam to fabricate parts.

To polish or not to polish

Up until the end of my tenure at the Metropolitan Museum of Art (1976–

2006), the department did little conservation work on its brass instruments.
About a dozen or two were on display in the galleries, though scores
remained on open shelving in a climate-controlled storeroom. One of my
early tasks was to place these instruments in plastic bags to protect them from
dust. All of the brass instruments had a waxy feel, suggesting that they had
been given a coating of wax (very likely a commercial furniture paste wax),
and most of them had a uniform, light-brown patina. When these instruments
were moved or catalogued, they were generally handled with cotton gloves.
On occasion, a musician or instrument maker would make an appointment to
play, measure, or examine one of them. After these visits, the instruments
would be washed out and carefully dried (if they had been played) or wiped
down with a solvent to remove fingerprints before they were reinstalled in the
gallery or placed back in storage. I do not recall any specialist ever
complaining about the patina or suggesting that we polish these instruments.
However, in 2003, Associate Curator Herbert Heyde (an acknowledged
authority on historic brass instruments) formally requested that I polish a
group of brass instruments that he had selected for installation in the gallery.
I was opposed to this treatment, in part because the gallery wall cases were
constructed of plywood and had in them wool carpeting, both of which create
a corrosive environment. Although I informed him that freshly polished brass
and silver would quickly retarnish under those conditions, Heyde insisted
upon the treatment, and I had no choice but to comply. Briefly, the
instruments were dewaxed with solvents and immersed in tanks containing
5% trisodium EDTA (as outlined above). To even out the surface color after
the EDTA treatment, the instruments were lightly polished with a slurry of
precipitated chalk applied by hand with cotton cheesecloth pads. Chalk has a
Mohs hardness of 3, just below that of brass (3.5), so tool marks, engraving,
and scratches were minimally affected. Because of the myriad mechanical
parts (valves, keys, springs, corks, crooks, slides, tuning bits, etc.), I decided
not to lacquer the instruments, but instead treated them with benzotriazole
(see Benzotriazole) to retard retarnishing. Unfortunately, an ongoing regimen
of chemical treatment and polishing will now be required to maintain their
appearance, whereas the old patinated surfaces were stable and required
virtually no attention. One unfortunate aspect of the chemical cleaning of
these brass instruments is that their freshly polished surfaces now reveal
minor surface irregularities and scratches that were formerly consistent with
or disguised by the patina.

Protective coatings

Freshly cleaned or polished brass and other metals may be coated with
Agateen no. 27 (diluted 1:1 with lacquer thinner for spraying) or Paraloid B-
48N, (12.5% solids in xylene for spraying, 15% solids for brushing), or
Incralac (which contains Paraloid B-44, benzotriazole, and a UV stabilizer) in
concentrations similar to Paraloid B-48N. Microcrystalline wax may also be
used to retard corrosion.

Lubrication (see-also Lubricants)

Oils and greases are derived from plant, animal, and petroleum sources.
Lubricants derived from plant and animal sources tend to be acidic and will
attack brass, nickel silver, and other metals; some have a tendency to become
rancid or oxidize and become gummy or harden; others quickly evaporate.
For brass/wind instruments, lightweight (i.e. low-viscosity) lubricants are
used for valves and trombone slides, while more heavyweight or viscous
greases are used for tuning slides. For conservation purposes, it is best to
sacrifice performance characteristics and opt instead for lubricants that will
not attack the metal and are stable. Light- and heavyweight medicinal-grade
mineral oils, as well as petroleum jelly are relatively stable and non-acidic,
and can be used for general lubrication. Synthetic or partially synthetic clock
and watch oils and greases (such as those supplied by the Moebius, Nye and
Tillwich firms) are also recommended. Despite their heat resistance and
stability, silicone oils and greases are incompatible with mineral oils and
greases and their solvents, and are thus not recommended. Lubricants should
be cleaned off and renewed at regular intervals, regardless of whether
instruments are used or not.
Natural rubber parts and gaskets should not be lubricated with petroleum-
based products, and are best treated with a light dusting of talc.
References
Anthony Baines, Brass Instruments: Their History and Development
(London, 1976).
Murray Bovin, Silversmithing and Art Metal (Forest Hills and New York,
1995).
Erick D. Brand, Band Instrument Repairing Manual (Elkhart, IN, 1978).
Rupert Finegold and William Seitz, Silversmithing (Radnor, PA, 1983).
Herbert Maryon, Metalwork and Enamelling (New York, 1971).
Algernon Rose, Talks with Bandsmen (London, 1894).
Yamaha Band Instrument Repair Manual (Hamamatsu, 1988).

Bronze disease

Bronze disease occurs in the presence of dampness and salt. Though it

often develops in copper alloy objects that have been excavated or have been
immersed in sea water, it can also be initiated by salt deposited by
perspiration. In the presence of chloride ions, an ambient relative humidity of
about 63–65% is conducive to the development of this condition. Bronze
disease is recognized by the presence of pale green spots, often crusty or
powdery, which may spread and deepen over time and eventually eat through
the metal (Figure 5).
Figure 5 Bronze disease on a brass bugle.

To treat bronze disease, the superficial incrustation should first be removed

mechanically with scrapers, a knife blade, or other means. If the patina of the
instrument is to be retained, the instrument can be soaked in a 5% solution of
Calgon (sodium sesquicarbonate), which should be changed repeatedly until
it is free of chlorides. The solution can be tested for chlorides by neutralizing
10 ml with a few drops of nitric acid until effervescence stops, and then
adding a few drops of 2% silver nitrate; if the solution becomes cloudy,
chlorides are still present. If the patina is to be removed (for example, if the
instrument is to be polished), an alternative procedure involves soaking in a
5% solution of trisodium EDTA. After either of these treatments, the
instrument should be thoroughly washed in tap water (if the instrument is
soft-soldered), and dried. It can be treated with benzotriazole (see
Benzotriazole) to retard further corrosion.
To protect cleaned brass surfaces, they may be coated with nitrocellulose
lacquer (such as Agateen no. 27 diluted 1:1 with lacquer thinner for spraying)
or Paraloid B-48N (dissolved in xylene, 12.5% solids for spraying or 15%
solids for brushing). Another approach is to brush or spray on Incralac (an
acrylic varnish containing Paraloid B-44, benzotriazole, and a UV stabilizer)
that has been diluted to 12.5% solids for spraying or 15% solids for brushing.
Instruments that have been treated for bronze disease should be stored in
reduced humidity (less than 50%) to prevent its reappearance.

Reference
H. J. Plenderleith and A. E. A Berner, The Conservation of Antiquities and
Works of Art (Oxford, 1971).
 C
Calendar (historic)

Until 1750, many European countries began the New Year on the
Annunciation, March 25, rather than January 1 (the seventh, eighth, ninth,
and tenth months were thus named September, October, November, and
December). This should be taken into consideration when computing dates,
ages, and the sequences of events.
The French revolutionary (or Republican) calendar was used between 1792
and 1805. These years were renumbered I–XIV, and the months were
renamed. (See The Encyclopædia Britannica 11th edition, s.v. French
Revolution.) The months began on September 22, 23, or 24 of the old
calendar.

I Sept. 22, 1792–Sept. 21, 1793

II Sept. 22, 1793–Sept. 21, 1794

III Sept. 22, 1794–Sept. 22, 1795

IV Sept. 23, 1795–Sept 21, 1796

V Sept. 22, 1796–Sept. 21, 1797

VI Sept. 22, 1797–Sept. 21, 1798

VII Sept. 22, 1798–Sept. 22, 1799

VIII Sept. 23, 1799–Sept. 22, 1800

IX Sept. 23, 1800–Sept. 22, 1801

 X Sept. 23, 1801–Sept. 22, 1802

XI Sept. 23, 1802–Sept. 23, 1803

XII Sept. 24, 1803–Sept. 22, 1804

XIII Sept. 23, 1804–Sept. 22, 1805

XIV Sept. 23, 1805–Dec. 22, 1805

The French revolutionary calendar was officially discontinued on January

1, 1806 and the Gregorian calendar reinstated.

Cents conversion

(see-also Equal temperament frequency table, Intervals)

In equal temperament, each semitone is defined by the ratio of :1,
which is further divisible into 100 cents. There are 1200 cents in an octave.
Frequencies one cent apart are in the ratio of :1.
In many books and articles on tuning and temperament, intervals of non-
equally tempered scales are often expressed in cents or in cent deviations
from equal temperament. In one sense this is meaningless, because all of the
intervals (aside from the octave) in equal temperament are themselves
“tempered” and are thus neither just nor pure. For example an equally
tempered third consists of 400 cents, but a pure 5:4 major third is about 386
cents; an equally tempered fifth consists of 700 cents, while a perfect 3:2 fifth
is about 702 cents (see below).
Nonetheless, it is sometimes of interest to calculate the number of cents in
an interval. The following formula can be used:
C = 1200log2 (f2/f1)

where

C is cents
 f1 and f2 are the frequencies or frequency ratios
log2 is the base-2 logarithm (if base-10 logarithms are used, multiply the
right side of the equation by 3.322038403).

Some cents equivalents

Interval Cents

Unison 1:1 0

Minor second or diatonic semitone 16:15 112

Minor tone 10:9 182

Mean tone 193

Major tone 9:8 204

Pythagorean minor third 294

Meantone minor third 310

Minor third 6:5 316

Major third 5:4 386

Pythagorean third 408

Perfect fourth 4:3 498

Perfect fifth 3:2 702

Meantone fifth 697

Quarter comma wolf fifth 738

 Minor sixth 8:5 814

Major sixth 5:3 884

Minor seventh 9:5 1018

Major seventh 15:8 1088

Octave 1200

Reference
www.sengpielaudio.com

Clavichord maintenance

(see-also Stringed-keyboard restoration, Wire gauges for early

keyboard instruments)
The clavichord mechanism consists of a key lever with a metal tangent
projecting from the back of the key. This is the simplest action employed in
keyboard instruments and theoretically the easiest to maintain and adjust. The
key dip of a historic clavichord may be impaired by the deterioration of the
original key cloths or leathers, and it is important to evaluate evidence of the
original thicknesses of these materials in order to establish the distance
between the tangents and the strings when the keys are at rest, which in turn
affects the key dip. When adjusting or replacing missing or deteriorated key
cloths and leathers, a rule of thumb regarding key dip is that it should be
adjusted so that the front ends of the accidentals do not descend below the
upper surfaces of the natural key platings when the accidentals are played,
but settle a millimeter or two above the platings. The key dip of the natural
keys should be the same as that of the accidentals. If the accidentals have
ivory or bone slips, these slips should stand proud of the naturals when the
accidentals are fully depressed.
All adjustments regarding key dip and tangent height should be made with
the key cloths or leathers, and not by hammering down or prying up the
tangents. Because the tangents frequently become misaligned with the strings
due to wood shrinkage, it may be necessary to rock them from front to back
so that they engage their string pairs squarely; however, they should not be
bent from side to side unless it is clear that they have been inadvertently bent
or damaged. This is especially true of fretted clavichords, as their
temperament is determined by the precise position of the tangents of fretted
notes (see below). One of the skills of those who made fretted clavichords
was the layout of the bridge and key levers in accordance with the
mathematical principles of particular temperaments. Unfortunately, the
vicissitudes of woodworking and wood shrinkage may make it necessary for
the tuner or conservator to adjust the striking points of the tangents by
bending them from side to side in order to correct the temperament.
However, before doing so, one should make precise measurements of the
tangent positions and the string lengths they define, and subject these
measurements to mathematical scrutiny in order to determine or characterize
the instrument’s temperament (see below), because a few millimeters one
way or the other may distinguish one temperament from another.
The after-lengths of the clavichord’s strings (the sections of the strings to
the left of the tangent) must be damped so that they do not vibrate along with
the sounding lengths. The damping material also serves to stop the vibration
of the sounding lengths once the tangents descend from the strings. Most
clavichords employ what is termed “listing,” which is a thin strip of cloth that
is either woven over and under successive choirs of strings or simply tucked
between them. The disadvantage of weaving the listing is that it may make
the keys resistant to the touch, especially if the weaving is tight. The
disadvantage of tucking the cloth strip down between the keys is that the little
loops that have been pushed down between the strings have a tendency to pop
out during heavy playing. Some clavichords employ what is termed a
“damping board” that sits above the after-lengths of the strings. This board is
lined on its underside with cloth, felt, or leather, which makes contact with
the strings and dampens them. A disadvantage of the damper board is that it
may limit bebung (the vibrato or tremolo effect created by varying finger
pressure on the key).
The string gauges of many early clavichords are often inscribed on the key
levers or next to the tuning pins, but they can be difficult or impossible to see
if they have been partially or completely washed away during previous
cleanings. A careful investigation with a UV lamp may reveal them. In
restoring a clavichord, one should follow the original gauge markings (see
Wire gauges for early keyboard instruments).
If string tension is too great it may cause some notes to “bark.” If this
occurs, one has several options: lower the overall pitch, replace the wire of
the offending note with one of a thinner gauge, or weight the key at the
tangent end. Keys that were not originally weighted should not be modified,
but one should check to see if the original weights have fallen out or have
been pared back to eliminate corroded sections that rubbed against adjacent
keys. If this is the case, consider pressing out the pared-back weights and
replacing them with new ones. These can be fabricated by pouring molten
lead into a wooden form, and after the sheet of lead has cooled, an arch punch
can be used to stamp out discs or slugs of the appropriate diameter (see
Figures 35a and 35b in Stringed-keyboard restoration). It is generally
advisable to adhere to original string-gauge markings. To interpret these
markings see Wire gauges for early keyboard instruments.
The temperament of fretted clavichords can be determined from the
positions of tangents, or in certain instances from the slots in the key rack, in
the following ways:

1 To calculate Pythagorean scaling, start at C and obtain the next two

fifths by multiplying by 2/3, then doubling back to get D by
multiplying by 2, and continuing in this manner until all twelve notes
have been determined. This will give you the positions of all the
sharps. To obtain the flats, work downwards from C. When you have
finished these calculations, you will have all the possible fret
positions. Because it is unlikely that the bridge and all the tangents
were laid out with extreme precision from fretted-group to fretted-
group, it may make more sense to calculate the theoretical fret
positions individually, but one should be mindful of the fact that
there are two types of half steps in the Pythagorean scale: the minor
second, which has the ratio of 256:243 (a factor of 1.05349794), and
the augmented prime, which has the ratio of 2187:2048 (a factor of
1.06787109). The whole tone has a ratio of 9:8. These ratios should
be used to calculate the theoretical string lengths, which are then
 compared to the actual string lengths to determine if there is a match.
Angelo Mondino determined that the clavichord depicted in a late
fifteenth-century Italian intarsia has its tangents set for Pythagorean
scaling with all the fictae tuned as flats (Mondino, 1994).
2 To determine whether the 18:17 rule was used to approximate equal
temperament, multiply the longer string of the fretted group by 17/18
to compute the string length of the next fretted note. If there are three
fretted notes for the course, multiply this product again by 17/18 to
calculate the next fretted string length. Compare these products with
the actual string lengths. Equal temperament string lengths and fret
positions can also be ascertained by using the semitone ratio of the
twelfth root of two.
3 For quarter-comma meantone temperament, calculate the theoretical
string lengths in the following way: multiply C by 2/3, then “temper”
that fifth by lowering the pitch of the upper note by a quarter-
syntonic comma (i.e. increasing the string length) by multiplying by
the fourth root of the syntonic comma (81/80), or 1.0031. Continue
upwards in this way through C–G–D–A–E–B–F♯–C♯–G♯, and then
downwards C–F–B♭–E♭ by multiplying each length by 3/2 and
tempering the lower notes (i.e. raising their pitch by decreasing
string length) by dividing by the fourth root of the comma, and
doubling back to obtain all the notes in the octave. An alternate
method is to calculate the lengths of F–C–G–A–E as above and then
compute the lengths of the remaining notes by pure 5:4 thirds (the
sharps are computed upward from previously tuned notes, while the
flats are computed downward). Because the accidental keys may
have been tuned as sharps or flats (there are no enharmonics in
meantone temperament), it is necessary to calculate both lengths. To
calculate the frequency ratio of the meantone minor second, multiply
the ratio of the Pythagorean minor second (256/243) by the 5/4th
power of the syntonic comma (which yields 1.06998449); to find the
length of strings tuned to this interval (such as C–D♭), divide the
length of the strings of the lower note by this factor. To calculate the
frequency ratio of the meantone augmented prime, divide the
Pythagorean augmented prime (2187/2048) by the 7/4th power of the
syntonic comma (which yields 1.04490673); to find the length of
strings tuned to this interval (such as C–C♯), divide the length of the
 strings of the lower note by this factor. (Note that C♯ is lower in
pitch than D♭, which is reversed from Pythagorean intonation.) The
ratio for a meantone whole tone is 1.11803399 (the square root of
5/4). Because meantone is a “regular system,” all of the minor
seconds will have the same ratio, as will all of the augmented primes
(see Tuning and temperament).

Analysis of the fretting system of an anonymous German eighteenth-

century clavichord (The Metropolitan Museum of Art, 89. 4. 1215):

Of the keys making up the middle octave, c1–c♯2, the first two notes are
fretted together, while the others are fretted three to a string. The string
lengths tabulated below for the various temperaments have been
computed from the longest string of each fretted group, not from c1.
Measurements in millimeters

String Equal ¼ comma

Note Pythagorean
length temperament meantone

c1 410 – – –

c♯1 391 387 392 384

d1 382 – – –

e♭1 358 361 357 358

e1 342 340 342 340

f1 332 – – –

f♯1 318 313 318 311

g1 297 296 297 295

g♯1 285 – – –
 a1 266 269 266 271

b♭1 250 254 255 253

b1 240 – – –

c2 224 227 224 228

c♯2 214 214 215 213

The fretting of this clavichord thus more closely follows 1/4 comma
meantone than equal temperament (calculated using the ratio of the twelfth
root of two) and Pythagorean temperament.”
To take an investigation further, draft the positions of any temperaments
(such as 1/3, 1/5, 2/7 comma, etc.) on a line representing the length of the
temperament octave, and then use a pair of dividers to check the tangent
positions on the clavichord. This “rule” can be used to check the fretting of
other octaves (as well as clavichords having different scalings) by drawing a
line representing the alternate C string length at any acute angle from one
endpoint of the rule, connecting the opposing endpoints with another straight
line, and then marking off the new fret positions from the rule by replicating
the angle made by the line connecting the two endpoints. If drawn carefully
with a T-square and triangle or drafting machine, the fret positions on both
lines will have the same proportions (Figure 6).
Figure 6 Drafted clavichord tangent positions for an octave (see text).
Shown here are the 1/4 comma fret positions for C♯, F♯, G♯, B♭, and E♭ for
two different scalings.

Another technique for determining a clavichord’s temperament is to use an

electronic tuning device that is precise to a cent or better to determine the
number of cents between consecutive fretted notes, and to compare these
values to those listed for various temperaments in Murray Barbour’s Tuning
and Temperament (1951). When making relative pitch measurements, a
weight should be used to play each note, as slightly different forces exerted
by the fingertips will affect the pitch from key to key. A suitable weight can
be ascertained empirically by “playing” the clavichord with a gram-force
gauge. When a reasonable playing force is established by the gauge, a weight
can be made (for example, with a nut, a bolt, and a few washers) to provide a
constant force from key to key. In assessing a clavichord’s temperament, it is
not necessary to tune it to its intended pitch (though the strings should be
under sufficient tension to provide a reasonable touch weight), as the
proportional relationships established by the tangents of fretted notes are
independent of pitch.

References
Murray Barbour, Tuning and Temperament (East Lansing, 1951).
Peter Bavington, Clavichord Tuning and Maintenance (2007).
Angelo Mondino, “The Intarsia of Urbino,” De Clavicordio (Turin, 1994),
pp. 49–55.
Koen Vermeij, Tuning and Maintenance of the Clavichord (Amsterdam,
n.d.).

Cleaning: soaps and detergents

(see-also Bleach, Paper, Textile cleaning, Water, Wood)

Soaps are cleaning and emulsifying agents that are made by saponifying
fats or oils with an alkali. They are not generally recommended for
conservation work because they have a tendency to form an insoluble scum
or film with hard water and because of their alkalinity. They are, however,
extremely effective in removing oil and grease. Some of the early (circa
1940) cleaning done in the musical instrument department of the
Metropolitan Museum of Art was carried out with a commercial product
called Oakite, which was primarily trisodium phosphate with less than 10%
borax. Synthetic detergents are categorized as anionic, cationic, or nonionic:
anionic detergents (having a negative charge) are often used in household
cleaners; cationic detergents (having a positive charge) have germicidal
properties and are often used in institutional cleaners and shampoos; nonionic
detergents (having no charge) are more frequently used in conservation work
because they are safer to use when cleaning sensitive materials, such as dyed
textiles (see Textile cleaning).
The high alkalinity of soaps, certain detergents, and ammoniated cleaners
may alter the color of certain organic dyes found in various woods (such as
brazil wood) and resins (such as shellac). This is similar to the litmus
reaction; in some cases, the color change may be reversed by restoring the
initial pH.
When using soaps and detergents, it is important to clear them with several
rinses of water, as they and their byproducts may be harmful over time.
Conductivity and pH meters can be used for monitoring the wash water.
Distilled or deionized water should be used for preparing cleaning solutions
and for final rinsing (see Water).
Some cleaning agents used in conservation work:
Vulpex (potassium methyl cyclohexyloleate) is a liquid soap with a pH of
10.5–11.5 that can be used “wet” diluted in water (5% and higher) or used
“dry” with organic solvents, such as mineral spirits, Stoddard solvent, and
1,1,1-trichloroethane (1% solution). It is used to clean a wide variety of
materials, including stone, wood, precious metals, ivory, leather, and textiles.
Orvus WA paste (also available as a powder) is an anionic synthetic
detergent having a nearly neutral pH that is generally mixed with water (1%).
Because of its nearly neutral pH and the fact that it rinses easily, it is often
used in cleaning textiles and carpets.
Dehypon LS45 is a nonionic synthetic detergent that is used in general
cleaning, and is especially recommended for cleaning delicate textiles as well
as wool. A working-strength solution can be made by adding 0.3 grams to 1
liter of water.

Compo, or pastiglia

Prior to hardening, this kneadable combination of whiting, oil varnish, and

animal-hide glue can be modeled by hand or pressed into molds. One of the
properties of this form of plaster is that (until it hardens completely) it can be
stretched or compressed slightly so that ornaments and moldings fabricated
with it can be made to fit. Once hardened, it is considerably stronger than
ordinary plaster of Paris. In the past, compo was widely used in making
gilded picture frames and the ornamentation on harps. Though compo is not
the most precise mold-making compound, because it is generally on hand,
picture framers often use it to take impressions and make molds of carved
ornaments that need to be replicated. After a compo impression hardens, it is
lightly oiled and fresh compo is then pressed into it to make copies. (See
Mold making for instructions on making more accurate molds).
Most formulas for making compo involve combining one part hot oil
varnish with about two parts hot animal-hide glue, and then adding enough
bolted (sifted) whiting to give a doughlike consistency when the mixture is
kneaded. Ernest Spon’s Workshop Receipts (London, 1909) gives the
following recipe:

1 lb glue melted in water sufficient to make a thin glue.

Melt ½ lb pale resin in ½ pint raw linseed oil.
Pour the two together and boil for ½ hour, stirring and watching that
it does not boil over.
Now mix in sifted whiting and knead the mass to form a dough.
Keep damp, but if it hardens it can be made plastic by steaming it.

The Gilder’s Manual (New York, 1876) provides the following

instructions:

Boil 7 lbs of the best glue in 7 half-pints of water.

Melt 3 lbs of the best rosin in 3 pints of raw linseed oil.
 When the ingredients are well boiled, put in to a large vessel and simmer
for ½ hour, stir, and take care it does not boil over.
Pour in a large quantity of whiting, previously sifted and rolled very fine,
and mix to a consistency of dough.

A similar formula can be found in The Carver and Gilder (London, 1864),
though no quantities or proportions are given.
The Art and Science of Gilding: A Hand Book of Information for the
Picture Framer (Rochester, n.d.) provides the following instructions for
making compo:

It is necessary to have two pots. In one, put three pounds of glue and one
quart of water. After it is thoroughly soaked, place it on the stove to
melt. Into the other, put two pounds of rosin and one pint of rosin oil,
and place it on the stove to melt. After they are melted separately, allow
the rosin to cool for about fifteen minutes, then put it into the glue and
mix thoroughly. Into a box large enough for the purpose, put fifteen or
twenty pounds of bolted whiting. Bank this whiting around the sides to
prevent the mixture from sticking to the box. Then put the mixture into
it and stir with a stick until it becomes the consistency of dough, and
thick enough to handle. Sprinkle some whiting on a board or table, and
knead the compo until thoroughly mixed. In time the compo becomes
hard. It is softened by subjecting it to the action of live steam. In
factories and in places where it is in constant use, a steam box is used.
Where only used occasionally, a simple way is to make a stretcher and
cover with cheese cloth. Put the compo on this stretcher and place it
over a pan of boiling water, and over it put a box to keep the steam in.

Compo can be placed in plastic bags and stored in the refrigerator or

freezer. Warming it in a double boiler will restore its flexibility.

References
The Art and Science of Gilding: A Hand Book of Information for the Picture
Framer (Rochester, n.d.).
The Carver and Gilder (London, 1864).
The Gilder's Manual (New York, 1876).
Ernest Spon, Workshop Receipts for Manufacturers and Scientific Amateurs
(London, 1909).

Conservation reports

(see-also Ethics)
The late John Brealey, former head of the Paintings Conservation
Department at the Metropolitan Museum of Art, referred to the next
generation of conservators that he helped train as “the young
documentarians” in recognition of their propensity for writing lengthy
conservation reports. Indeed, report writing has become an art unto itself,
with many conservators spending more time and effort writing up elaborate
reports than in treating objects. Hans Rudolf Hösli, Marc Soubeyran, and
Tom Wilder (2010) have proposed a “detailed documentation form” (29
pages in length) and a “concise examination form” (13 pages in length) as
exemplars for violin restorers to follow when preparing conservation reports.
Both of these forms require making hundreds of measurements, providing a
physical description of the instrument, and documenting its provenance, legal
status, and sales records. In fact, much of this information is not required of a
conservation report but is rather the province of curatorial cataloging. The
primary function of a conservation report is to document the treatment;
however, if parts, structures, construction marks, and inscriptions become
accessible when an instrument is disassembled, it behooves the restorer to
note them in the report and make any necessary photographs, measurements,
or drawings before the instrument is reassembled. It is not necessary to make
scores of measurements of a violin’s scroll, as Hösli et al. suggest, if the
treatment simply involves regluing an open seam. Fervent report writers
might keep in mind that researchers typically have little regard for
measurements made by others and usually insist upon making their own.
Some museums (such as the British Museum) maintain an open policy
regarding conservation records and post them online, while others (such as
the Metropolitan Museum of Art) do not. Not all museums include
conservation reports with their official internal catalog but file them
independently. This is highly unfortunate, as those consulting curatorial
records miss out on the technical data that have been collected by
conservators over the years, as well as their recommendations and cautionary
advice regarding handling, tuning, and use.
A conservation report should include the following:

1 Documentation of the condition of the instrument prior to treatment.

This should include a summary or analysis of previous alterations
and repairs.
2 A statement regarding the purpose of the treatment (cleaning,
stabilization, repairing damage, to make playable, etc.), and who
ordered it.
3 Measurements and descriptions of relationships between parts that
might be disrupted or altered by the disassembly process or the
conservation procedure.
4 A list of new or replaced parts. If feasible, new parts should be
marked with the restorer’s name and date. Removed parts should be
saved.
5 Names of any chemical agents, adhesives, consolidants, fumigants,
coatings, etc. that were used, as well as the method of application.
6 If relevant, a record of the existing pitch and temperament, especially
if the instrument is tuned or its overall pitch is adjusted as part of the
treatment, or will be subsequent to it. Measurements should be made
with an accurate electronic tuning device.
7 Recommendations for care, exhibition, handling, and use.

Reference
Hans Rudolf Hösli, Marc Soubeyran, and Tom Wilder, “The Documentation
of Stringed Instruments,” in Tom Wilder, ed., The Conservation,
Restoration, and Repair of Stringed Instruments and Their Bows
(Montreal, 2010), pp. 290–363.
 D
Dendrochronology and the dating of wood

There are several scientific methods of dating wood; the most often cited
are radiocarbon dating (otherwise known as carbon-14 dating) and
dendrochronology.

Radiocarbon dating

Radiocarbon dating makes use of the rate of nuclear decay of carbon-14, a

heavy isotope of carbon formed in the upper atmosphere due to an interaction
with cosmic rays. This radioactive isotope combines with plants undergoing
photosynthesis, but after the plant dies, it no longer incorporates carbon-14.
Because carbon-14 undergoes radioactive decay and reverts to carbon-12 at a
constant rate of 1% every 83 years, it is possible to date old wood by
assessing the proportion of carbon-14 to carbon-12. Unfortunately, this
method of dating does not yield a precise date, but a range of possible dates.
Routine dating provides a range of ±80 years, though ±20 years is claimed in
some studies – but even these figures are statistically couched. While this
may be adequate for distinguishing a Gasparo da Salo violin from a
nineteenth-century copy, it is not precise enough for the type of dating
needed to authenticate most violins. Another problem with radiocarbon
dating is that a sample must be taken. While the required sample size has
been dramatically reduced in recent years (at some testing facilities, 0.1 gram
is sufficient), owners are often reluctant to have any amount of wood
removed from their instruments.

Dendrochronology

Dendrochronology is one of the techniques used to calibrate the

radiocarbon dating scale, which has been fraught with inaccuracies due to the
atmospheric testing of nuclear weapons and the spewing of carbon-bearing
gases into the atmosphere from the burning of petroleum products.
Dendrochronological dating relies upon the correlation of yearly weather
conditions with the growth rates of certain species of trees, such as spruce,
larch, and oak, which produce year-rings that vary in width in response to
average yearly weather conditions. Trees such as maple, mahogany, walnut,
and cypress unfortunately cannot be dated by this method because their year-
rings do not vary in strict accordance with weather conditions (they are said
to be “complacent”).
Trees that can be dated by dendrochronology exhibit growth patterns that
are unique to the regions from which they came, and numerous master
chronologies have been constructed from regions that are known to have been
the sources of wood used in making the tops or soundboards of violins, lutes,
guitars, harpsichords, pianos, and other instruments, as well as other objects,
such as the wood panels used in paintings. The dating of violins began in the
1950s with the assemblage of master chronologies drawn from spruce and
larch trees grown in the Alps and other European forests that were frequented
by the wood merchants who supplied the makers. Some master chronologies
extend from modern times back to the dawn of the violin (about 1500),
making them potentially useful for dating violins.
What does dendrochronological dating reveal? The date of the last year-
ring in a sequence merely indicates that the instrument could not have been
made before that date. Clearly, some year-rings may have been cut away
when the primary wedge used to make a violin top was re-split and planed for
joining. Furthermore, in certain types of wood, there is a distinction between
the inner heartwood and the outer sapwood, and some craftsmen choose to
discard the sapwood, which would result in the loss of an indeterminate
number of year-rings. Spruce, however, which is used to make the tops of
most stringed instruments and the soundboards of keyboard instruments,
typically does not display a clear distinction between heart- and sapwood, so
makers often use the wood right out to the last year-ring. Wood is generally
seasoned for at least a year before use, so this too must be considered when
arriving at a reasonable terminus post quem (the earliest date) for the
manufacture of the instrument. Clearly, an instrument maker may use an old
piece of wood or sections of wood taken close to the center of the trunk that
may be decades or even centuries removed from the year the tree was cut
down. Alpine spruce trees have been known to live for over 500 years, so it is
conceivable that an instrument made today could exhibit rings that are 400–
500 years old and have a dendrochronological date of 1600.
Unlike radiocarbon dating, the results obtained from dendrochronology are
potentially accurate to the year, but again, there is a statistical component to
this dating procedure: because there is never a 100% match between a master
chronology and a sample of wood, there is never absolute certainty as to the
results. The “Messiah” violin's dendrochronology is a case in point – one
dendrochronologist concluded that the most recent year-ring of this violin
dated to 1738, another asserted that 1682 is the correct date, and yet another
has arrived at a date of 1788. Why the different results? The reason is that
different statistical tests and mathematical manipulations of the raw data
often yield different dates.
The year-ring measurements of the object being dated, such as a violin top
or harpsichord soundboard, must be made along a radial axis of the log from
which it was cut, and this is easy to do when these parts are fabricated from
quarter-sawn or radial splits of segments of tree trunks, as the year-rings are
then in the proper orientation. Once the sequence of year-ring widths have
been measured (see below; measurements are generally made in hundredths
of a millimeter; 1.01 mm is recorded as 101), the sequence is formatted (so-
called “Tucson format” and simple columnar format are used) and can then
be imported into one of several dendrochronology programs (such as
ARSTAN, COFECHA, Synchro Search, and Tellervo) and compared against
master chronologies, such as those downloaded from the International Tree
Ring Database (ITRDB). When attempting to establish the date of a violin
top that is believed to be from northern Italy, for example, begin by using one
or more of the Alpine chronologies, such as the PCAB (Picea abies) Giertz
Obergurgl chronology, which provides a sequence dating from 1276 to 1974,
or the PCAB Schweingruber Obersaxen chronology, which dates from 1537
to 1995. Both of these are from the Austrian Alps. Another useful master
chronology is from the Fodara Vedla Alm, an Italian Alp sequence derived
from Pinus cembra and dating from 1474 through 1982. It is also important
to compare one's data against master chronologies from other regions,
including north of the Alps and even North America (several useful ones are
available on the ITRDB).
The programs listed above apply a variety of statistical operations, such as
de-trending and indexing, and perform tests, including Gleichläufigkeit,
Pearson's r, and Student's t. They typically uncover numerous matches at
varying levels of statistical significance. Closest scrutiny is given to matches
having the highest statistical scores. Some programs graph the year-ring
widths of the object being dated against the master chronology. An
examination of these graphs is essential, as visual similarities are critical in
deciding among the various statistical matches, and a compelling visual
match may trump the highest statistical score.
To measure the year-ring widths, the author uses a violin cradle mounted
on a rack-and-pinion device that is linked to a digital linear scale. This device
is positioned beneath a stereo microscope mounted on a boom stand. The
instrument is placed on the cradle and the year-rings are lined up in the cross-
hairs of one of the microscope's eyepieces using the rack-and-pinion device.
As the instrument is repositioned from year-ring to year-ring, the output from
the digital linear scale is sent to an Excel spreadsheet, which is programmed
to subtract the previous gross measurement from the most recent one, thereby
yielding a list of discrete year-ring widths. Some dendrochronology programs
can analyze data saved in columnar form as text files, while others require the
data to be organized in specialized formats, such as the “Tucson format.”
When measuring the year-rings, one begins with the oldest year-ring and
continues to the most recent. This direction can be established as it is the
same as early-growth (spring) to late-growth (summer) orientation. Growth in
spring is pale, gradually darkening going into summer, and then it abruptly
stops, as there is no growth in the fall and winter. With violin-family
instruments that have book-matched two-piece tops, the earliest year-rings
are generally located at the edges of the top, and the last year-ring runs along
the center line. It is critical to identify “wings” (narrow pieces of wood added
at the edges that are not a continuation of the principal chronology), one-
piece tops, and other anomalies. The soundboards of keyboard instruments,
lutes, and other instruments having broad soundboards or tops are generally
made of several planks that are joined edge to edge, so it is vital to determine
where each board begins and ends.

References
M. G. L. Baillie, Tree-Ring Dating and Archaeology (Chicago, 1982).
M. G. L. Baillie, A Slice Through Time: Dendrochronology and Precision
Dating (London, 1995).
Angelo Mondino and Matteo Avalle, Manual of Dendrochronology Applied
to the Dating of Musical Instruments (Cremona, 2005).
Angelo Mondino and Matteo Avalle, Nuova Procedura di Dendrodatazione
con Synchro Search Versione 2.2.1 (2010).

Detergents and soaps

See cleaning

Dictionary of common and obsolete chemical terms

Common
Chemical name
term

acetic ether ethyl acetate

alum potash alum or potassium alum is potassium

aluminum sulfate; ammonium alum is ammonium
double sulfate

ambergris a waxy substance secreted by the sperm whale

ammonia the gas NH3 or the gas dissolved in water

ammonia ammonium hydroxide solution

water

aniline dye synthetic dye made with aniline

antichlor sodium thiosulfate

antimony antimony trisulfide

black

antimony antimony trioxide

bloom

antimony antimony trisulfide, stibnite

glance

argol potassium bitartrate

aqua fortis nitric acid

aqua regia combination of nitric and hydrochloric acids used to

dissolve gold or platinum

azo dye a large group of synthetic dyes characterized by the

presence of an azo group in their chemical formulas

Bakelite trade name for a phenol-formaldehyde resin

baking soda sodium bicarbonate

barilla sodium carbonate

barita, barium oxide

calcined

barites natural mineral form of barium sulfate

battery acid 29–32% sulfuric acid

beaumontage filler for wood consisting of beeswax, rosin, and

shellac

benzene aromatic hydrocarbon

benzine volatile petroleum distillate, primarily aliphatic; not

to be confused with the highly toxic aromatic
 hydrocarbon benzene, also called petroleum naptha

bezoar concretion found in the alimentary tract of certain

ruminants; calcium phosphate

bicarbonate sodium bicarbonate

of soda

bitter salt Epsom salts

bitter spar see dolomite

black ash barium sulfate

black lead graphite

blanc d'argent white lead, lead carbonate

blanc-fixe barium sulfate

bleaching calcium chloride, calcium hypochlorite

powder

blue copperas blue vitriol, copper sulfate

blue lead lead sulfate

blue vitriol blue copperas, copper sulfate

bluestone cupric sulfate

bone ash tricalcium phosphate

borax sodium tetraborate

brimstone sulfur
brimstone sulfuric acid
acid

British gum dextrine; gummy polysaccharide derived from starch

burnt lime calcium oxide

butter of … chloride of a metal

calamine hydrous zinc silicate, zinc carbonate

calcareous containing calcium carbonate

calcimine mixture of glue and whiting or zinc white

calcine of … heated below fusing; often heated and pulverized; an

oxide of a metal

calcined tin tin oxide

Calgon trade name for sodium hexametaphosphate

caliche sodium nitrate

calomel mercurous oxide

calx calcine; roasted metal or mineral

calx of tin tin oxide

caoutchouc natural rubber

Cascamite trade name for urea-formaldehyde resin glue

Castile soap soap made by saponifying olive oil with sodium

hydroxide

caustic lime calcium hydroxide

caustic potash potassium hydroxide

caustic soda sodium hydroxide

celluloid cellulose nitrate often with camphor added as a

plasticizer

ceresin naturally occurring paraffin

cerussite lead carbonate

chalk calcium carbonate

chamber acid 62–70% sulfuric acid

chamber-pot stale urine, source of ammonia used in cleaning

lye

Chatterton's trade name for a cement made with 3 parts gutta

compound percha, 1 part rosin, and 1 part tar

Chile niter sodium nitrate

Chile sodium nitrate

saltpeter

Chinese zinc oxide

white

chloride of calcium hypochlorite

lime

chloride of sodium hypochlorite

soda

chrome alum potassium chromium sulfate

cinnabar mercuric sulfide

clay fine particles of hydrous aluminous minerals

Clorox trade name for bleaching solution of 6% sodium

hypochlorite

coal tar dye synthetic dye made from coal tar derivatives

coal tar benzene, xylene

naptha

coke residue of coal heated to 1200–1400°C.

collodion a solution of cellulose nitrate in alcohol and ether, or

other solvent

colophony rosin or pine resin

commercial finely ground calcium carbonate

whiting

compo a mixture of whiting, animal-hide glue, and oil

varnish

Condy's trade name for potassium permanganate

crystals

copperas, blue vitriol; copper sulfate

blue

copperas, green vitriol; ferrous sulfate

green

copperas, white vitriol; zinc sulfate

white

corrosive mercuric chloride

sublimate

corundum natural aluminum oxide abrasive

Courtrai, drier made with manganese and lead oxides

siccatif de

cream of lime calcium carbonate

cream of potassium bitartrate

tartar

creech calcium sulfate

cresol phenols derived from coal tar used as disinfectant

creosote oily substance obtained by the distillation of coal tar,

used as a preservative

crocus ferric oxide abrasive

crystal ammonium carbonate

ammonia

crystal sodium carbonate

carbonate

denatured ethanol to which methanol, 2-propanol, and other

alcohol poisonous substances (including aviation gasoline)
have been added

dextrine British gum; gummy polysaccharide derived from

starch

diamantine very fine aluminum oxide (alumina) abrasive

dolomite calcium magnesium carbonate

eau ecarlate solution of bichloride of tin

ebonite hard, vulcanized rubber

emery mixture of corundum and magnetite abrasive

epsom salts magnesium sulfate

ether diethyl ether, ethyl ether, or ethyl oxide

Everett's salt potassium ferricyanide

flowers of … the oxide of a metal

fuliggine soot

fuller's earth a soft clay primarily used as an absorbent in cleaning

textiles

galena lead sulfide

gallipot pine resin; also a small ceramic vessel

Genklene 1,1,1-trichloroethane

glacial acetic 99.5% acetic acid

acid

glair egg white

glance a dark colored mineral sulfide

Glauber's sodium sulfate

salts

glazing putty of linseed oil and whiting

compound
glover acid 78–80% sulfuric acid

Glyptal trade name for an alkyd resin

Graham's salt sodium hexametaphosphate

green green vitriol, ferrous sulfate

copperas

green verditer copper carbonate

green vitriol green copperas, ferrous sulfate

gumtion mixture of linseed oil and lead acetate

guncotton cellulose nitrate

gutta percha a rubber-like resin; can be vulcanized

gypsum calcium sulfate dihydrate

hartshorn ammonium carbonate; see salt of hartshorn

heavy spar barium sulfate

hepar calcis calcium sulfide

hepar sulfuris potassium sulfide

hypo sodium hyposulfite

industrial ethyl alcohol to which methyl alcohol has been

methylated added as a denaturant
spirits

iron alum ferric ammonium sulfate

isinglass water-soluble adhesive made from the swimming

 bladders of sturgeon

japonica gum catechu

Javelle water bleaching solution made with potassium chloride

(Eau de
Javel)

jeweler's ferric oxide

rouge

kalkspar calcium carbonate

Keene's a mixture of gypsum and alum yielding a durable,

cement hard-finish plaster

Kieselguhr infusorial earth, tripoli

lampblack carbon

lant stale urine rich in ammonia – used as a cleaning

agent

lead (pencil) combination of graphite and clay

lees (of wine) potassium bitartrate (active ingredient)

ligroine napthenic petroleum distillate; also benzine, heptane

lime calcium oxide

lime, calcium hypochlorite

chlorinated

lime, slaked calcium hydroxide

limewater alkaline solution of calcium hydroxide

linolein glycerol ester of linoleic acid

litharge lead monoxide

lithopone white pigment; double precipitate of barium sulfate

and zinc sulfide

liver of sulfur sulfurated potash (a mixture of potassium

polysulfide and potassium thiosulfate

lixivium lye

lunar caustic silver nitrate

lute general term for a cement or clay used in filling gaps

or binding parts together

lye potassium carbonate, sodium hydroxide, or

potassium hydroxide

malachite copper carbonate

marble calcium carbonate

Marseilles soap traditionally made by saponifying olive oil with

soap sodium hydroxide

massicot lead monoxide, lead ochre

megilp a gelatinous painting medium or vehicle made with

linseed oil and mastic varnish

methylated ethanol denatured (made unfit for drinking) by the

spirits addition of methanol

milk of lime calcium hydroxide suspension in water

milk of sulfur precipitated sulfur

minium red lead oxide

Mohr's salt ferrous ammonium sulfate

mucilage viscid solution of plant gum, protein, or sugar used

as an adhesive

muriate of … a substance dissolved in hydrochloric acid, such as

muriate of tin; a chloride

muriate of ammonium chloride

ammonia

muriatic acid commercial grade of hydrochloric acid

naptha volatile petroleum distillate primarily containing

aliphatic hydrocarbons

napthenic mineral oil distilled from napthenic crude oil

solvent primarily containing cycloalkanes

natal etching solution of nitric acid and ethanol, 1.5:100

natron hydrous sodium carbonate

neats-foot oil oil extracted from animal bones and feet used as a
leather dressing

niter (or potassium or sodium nitrate; sodium bisulfate

nitre)

Nitromors commercial name for paint remover containing

methylene chloride and methanol

Nordhausen sulfuric acid

acid
Oakite trade name for a household cleaning powder
containing trisodium phosphate and a small amount
of borax (circa 1940)

oil black lampblack

oil of vitriol concentrated sulfuric acid

oleum fuming sulfuric acid

orpiment arsenic trisulfide

ozokerite earth wax; naturally occurring paraffin

paraffin oil see paraffinic solvent

paraffinic linear-chain hydrocarbon distilled from parafinnic

solvent crude oil primarily consisting of n-alkanes

pearl ash potassium carbonate or potassium hydroxide

perhydrol hydrogen peroxide

permanganate potassium permanganate

potash

phenol carbolic acid

pitch residue of coal tar distillation

plaster of hemi-hydrate of calcium sulfate

Paris

Plasticine trade name for modeling compound containing

calcium carbonate, petroleum jelly, and stearic acid

plasteline modeling compound made with clay mixed with oil

 or wax

plate powder a fine abrasive used in polishing metal, such as

silver plate

plumbago graphite

potash potassium carbonate from leaching wood ash;

potassium hydroxide, potassium salts, potassium
oxide

potash lye potassium hydroxide

potassium potassium ferricyanide

prussiate

Prussian blue ferric ferrocyanide

prussiate salt of hydrocyanic acid

prussiate of potassium cyanide

potash

prussic acid hydrocyanic acid

putty a mixture of whiting and raw or boiled linseed oil;

lead white is sometimes added

pyroxylin cellulose nitrate

quicklime calcined (heated) limestone, calcium oxide

quicksilver mercury

realgar arsenic disulfide

rectified spirit alcohol concentrated through distillation

red lead lead oxide

red mercuric oxide

precipitate

red prussiate potassium ferricyanide

of potash

red prussiate sodium ferricyanide

of soda

regulus an impure mass of metal

regulus of impure mass of antimony

antimony

regulus of violet-colored alloy of copper and antimony

Venus

roche alum rock alum; ordinary alum

roche lime lime in lump form after it has been heated;

quicklime

Rochelle salts potassium sodium tartrate

rosin colophony or pine resin

rottenstone light gray, soft earthy stone used as an abrasive

rouge ferric oxide anhydrous

sal alembroth double chloride of ammonium and mercury

sal ammoniac ammonium chloride

salt cake sodium bisulfate

salt of ammonium chloride
hartshorn

salt of potassium sulfate

Lemery

salt of lemon potassium oxalate

salt of Saturn lead acetate

salt of Rochelle salt

Seignette

salt of soda sodium carbonate

salt of sorel potassium oxalate

salt of tartar potassium carbonate

salt of Venus copper sulfate

salt of vitriol zinc sulfate

saltpeter potassium nitrate

Schwerter's potassium dichromate

salts

Seccotine tradename for fish glue that is liquid at room

temperature

slaked lime hydrated lime, calcium hydroxide

soap bark bark of Quillaja saponaria, which yields a soapy

substance (saponin) used in cleaning

soda sodium carbonate, sodium bicarbonate, sodium

hydroxide, or sodium oxide
soda ash anhydrous sodium carbonate

soda crystals sodium carbonate

soda lime calcium hydroxide mixed with sodium hydroxide or

potassium hydroxide, or both

soda niter sodium nitrate

spirits of niter nitric acid

spirits of salt hydrochloric acid

spirits of sulfuric acid

vitriol

spirits of ethanol
wine

sugar of … acetate of a metal

sugar of lead lead acetate

sulfurated carbon disulfide

carbon

sulfurated mixture of potassium polysulfides and potassium

potash thiosulfate

talc hydrated magnesium silicate

tall oil resinous liquid byproduct of chemical wood pulp

manufacture

tallow rendered animal fat

tartaric acid sodium bitartrate

tincture a solution in alcohol

tower acid 78–80% sulfuric acid (see glover acid)

tripoli infusorial, diatomaceous earth; also a finely

granulated siliceous rock; used as an abrasive and
filler

trona natural form of sodium sesquicarbonate

turner's molten shellac used to attach objects to a cement

cement chuck

turpentine, a cheap grade of turpentine obtained through steam

steam distillation of wood
distilled

tutty powder pulverized zinc oxide

V.M. & P. varnish makers' and painters' naphtha or turpentine

naphtha substitute; petroleum distillate with boiling range of
100°–160°C

Vaseline trade name for petroleum jelly

venetian red ferric oxide

verdigris copper acetate

vermilion mercuric sulfide

vinegar acetic acid

Vienna chalk a high-magnesia lime prepared from calcined

dolomite used in polishing

vitriol, blue blue copperas; cupric sulfate

vitriol, green green copperas; ferrous sulfate

vitriol, oil of sulfuric acid

vitriol, white white copperas; zinc sulfate

vulcanite hard, vulcanized rubber

wall saltpeter calcium nitrate

washing soda sodium carbonate

waterglass sodium silicate

white acid mixture of hydrofluoric acid and ammonium

bifluoride

white caustic sodium hydroxide

white white vitriol, zinc sulfate

copperas

white lead lead carbonate

white spirits petroleum spirits often containing a mixture of

aliphatic and aromatic petroleum distillates

white vitriol white copperas, zinc sulfate

whiting finely ground calcium carbonate

wood spirits methanol

yellow potassium ferricyanide

prussiate of
potash

yellow sodium ferrocyanide

 prussiate of
soda

zieg urine

zinc vitriol white copperas, zinc sulfate

zinc white zinc oxide

Disinfection, disinfestation, and fumigation

Disinfection refers to the killing of disease-causing microorganisms.

Disinfestation refers to killing of insects, rodents, and other small animals.
Fumigation refers to the use of toxic fumes or smoke to dispose of all of the
above.
“Woodworm” is a nonspecific term for insect larvae that bore through
wood (Figure 7). There are many types of insects responsible for this type of
damage, including the furniture beetle (Anobium punctatum) and the powder-
post beetle (Lyctus brunneus). Until recently, infested objects were treated by
fumigation with highly toxic gases, such as sulfuryl fluoride and methyl
bromide. Unfortunately, some of these chemicals affect pigments, adhesives,
and other materials used in artwork, and the dangers of chemical leakage, as
well as the need to vent these gases after treatment, led to the development of
safer techniques.
Figure 7 X-ray of woodworm damage in a viola.

The best technique involves placing the object in an oxygen-free

environment until all of the larvae and eggs are killed. One method of
eliminating oxygen is to flush the treatment chamber or an impermeable
plastic bag of air and fill it with argon, nitrogen, or carbon dioxide – argon
being preferred because it is chemically inert. Packets of oxygen scavengers
(powdered iron moistened with saline solution, such as Ageless Z-2000
oxygen-absorbing packets) are inserted to remove any remaining oxygen.
One problem associated with using these gases is that they must be
humidified before releasing them into the disinfestation chamber or bag;
another is the high cost of the grades of these gases that are sufficiently
purified of oxygen. Another technique involves the use of oxygen scavenger
packets alone, which remove oxygen from the air that is sealed inside the
impermeable plastic bag along with the object. Battery-operated oxygen
monitors and chemical indicators (such as Ageless Eye) can be enclosed with
the object to determine the effectiveness of the oxygen scavenger as well as
any leakage of air into the plastic bag. A carbon dioxide sensor can also be
used to determine whether insects or their larvae are still alive. Due to the
absorption of oxygen by the oxygen scavenger, there may be an increase in
the relative humidity level in the enclosure, but this can be controlled by
inserting packets of preconditioned silica gel (such as Art-Sorb, which is
available in 40–70% humidity ranges). The absorption of oxygen by
scavengers is an exothermic process, so packets of these materials should not
be in direct contact with potentially sensitive objects but positioned a safe
distance from them.
Enclosures can be made of heat-sealed plastic film, but most types are far
too permeable to atmospheric oxygen to be of use. Several (such as
vinylidene dichloride and ethylene-vinyl alcohol copolymers) have very low
oxygen permeability, though they are relatively expensive. A much cheaper
type of plastic film is polypropylene laminated with aluminum foil
(Marvelseal is one commercial brand). This has very low oxygen
permeability and can be heat-sealed, though it is not transparent. To view
oxygen monitors or indicators, a window can be made by attaching a
Plexiglas or glass panel over a cutout with a silicone adhesive.
An Ageless Z-2000 packet is conservatively rated to remove 2000 ml (122
cubic inches) of oxygen from an airtight chamber or bag. One can calculate
the number of packets required by filling the chamber or bag with Styrofoam
peanuts and then pouring them into a graduated cylinder or pitcher to
determine the volume. Before heat-sealing the object in the plastic bag with
the Ageless Z-2000 packets, Art-Sorb, and an oxygen monitor or indicator,
keep in mind that about 20% of the volume of gas in the bag will be absorbed
by the oxygen scavenger, so a sufficient volume of air should be enclosed to
prevent the bag from being pulled tightly over the object. There should also
be sufficient space to position the Ageless Z-2000 packets at some distance
from the object to protect it from the heat that is released.
The amount of time it takes to kill insect larvae and eggs varies with the
species, the temperature and humidity, and the number of parts per million of
oxygen present, as well as with the massiveness of the object that is infested.
If carried out at about 20–25°C (68–75°F) and 40–55% humidity, and at an
oxygen concentration of less than 1000 p.p.m, commonly encountered insect
pests can be killed in stringed instruments and instrument cases in around ten
days. For massive objects such as a harpsichord or fortepiano, about fourteen
days is required. Before treating the object, it is recommended that the wood-
boring pest be identified, preferably by an entomologist, so that its life cycle
can be ascertained. In some cases, its eggs may survive in the absence of
oxygen, so a second round of disinfestation may be necessary to kill the next
generation.
Keep in mind that once a disinfested object is returned to the open
environment, it may be reinfested.
Disinfectants used in conservation include Dowicide 1 (soluble in mineral
spirits and oils, though only slightly soluble in water), Dowicide A (soluble
in water and other polar solvents), benzalkonium chloride, and Thymol. For
example, a small crystal of Thymol can be dissolved in a pint of hot animal-
hide glue to prevent bacterial growth.

References
Charles Selwitz and Shin Maekawa, Inert Gases in the Control of Museum
Insect Pests (Getty Conservation Institute, 2003).
Linda A. Zycherman and J. Richard Schrock, A Guide to Museum Pest
Control (Washington, 1988).

Drill bit sizes (fractional inch, number and letter, and metric)

Drill bit Inch Metric

no. 80 0.0135 0.3429

no. 79 0.0145 0.3680

1/64 in 0.0156 0.3969

0.4 mm 0.0158 0.4000

no. 78 0.0160 0.4064

no. 77 0.0180 0.4572

0.5 mm 0.0197 0.5000

no. 76 0.0200 0.5080

no. 75 0.0210 0.5334

no. 74 0.0225 0.5715

0.6 mm 0.0236 0.6000

no. 73 0.0240 0.6096

no. 72 0.0250 0.6350

no. 71 0.0260 0.6604

0.7 mm 0.0276 0.7000

no. 70 0.0280 0.7112

no. 69 0.0292 0.7417

no. 68 0.0310 0.7874

1/32 in 0.0313 0.7938

0.8 mm 0.0315 0.8000

no. 67 0.0320 0.8128

no. 66 0.0330 0.8382

no. 65 0.0350 0.8890

0.9 mm 0.0354 0.9000

no. 64 0.0360 0.9144

no. 63 0.0370 0.9398

no. 62 0.0380 0.9652

no. 61 0.0390 0.9906

1 mm 0.0394 1.0000

no. 60 0.0400 1.0160

no. 59 0.0410 1.0414

no. 58 0.0420 1.0668

no. 57 0.0430 1.0922

1.1 mm 0.0433 1.1000

no. 56 0.0465 1.1811

3/64 in 0.0469 1.1906

1.2 mm 0.0472 1.2000

1.3 mm 0.0512 1.3000

no. 55 0.0520 1.3208

no. 54 0.0550 1.3970

1.4 mm 0.0551 1.4000

1.5 mm 0.0591 1.5000

no. 53 0.0595 1.5113

1/16 in 0.0625 1.5875

1.6 mm 0.0630 1.6000

no. 52 0.0635 1.6129

1.7 mm 0.0669 1.7000

no. 51 0.0670 1.7018

no. 50 0.0700 1.7780

1.8 mm 0.0709 1.8000

no. 49 0.0730 1.8542

1.9 mm 0.0748 1.9000

no. 48 0.0760 1.9304

5/64 in 0.0781 1.9844

no. 47 0.0785 1.9939

2 mm 0.0787 2.0000

no. 46 0.0810 2.0574

no. 45 0.0820 2.0828

2.1 mm 0.0827 2.1000

no. 44 0.0860 2.1844

2.2 mm 0.0866 2.2000

no. 43 0.0890 2.2606

2.3 mm 0.0906 2.3000

no. 42 0.0935 2.3749

3/32 in 0.0938 2.3813

2.4 mm 0.0945 2.4000

no. 41 0.0960 2.4384

no. 40 0.0980 2.4892

2.5 mm 0.0984 2.5000

no. 39 0.0995 2.5273

no. 38 0.1015 2.5781

2.6 mm 0.1024 2.6000

no. 37 0.1040 2.6416

2.7 mm 0.1063 2.7000

no. 36 0.1065 2.7051

7/64 in 0.1094 2.7781

no. 35 0.1100 2.7940

2.8 mm 0.1102 2.8000

no. 34 0.1110 2.8194

no. 33 0.1130 2.8702

2.9 mm 0.1142 2.9000

no. 32 0.1160 2.9464

3 mm 0.1181 3.0000

no. 31 0.1200 3.0480

3.1 mm 0.1221 3.1000

1/8 in 0.1250 3.1750

3.2 mm 0.1260 3.2000

no. 30 0.1285 3.2639

3.3 mm 0.1299 3.3000

3.4 mm 0.1339 3.4000

no. 29 0.1360 3.4544

3.5 mm 0.1378 3.5000

no. 28 0.1405 3.5687

9/64 in 0.1406 3.5719

3.6 mm 0.1417 3.6000

no. 27 0.1440 3.6576

3.7 mm 0.1457 3.7000

no. 26 0.1470 3.7338

no. 25 0.1495 3.7973

3.8 mm 0.1496 3.8000

no. 24 0.1520 3.8608

3.9 mm 0.1535 3.9000

no. 23 0.1540 3.9116

5/32 in 0.1563 3.9688

no. 22 0.1570 3.9878

4 mm 0.1575 4.0000

no. 21 0.1590 4.0386

no. 20 0.1610 4.0894

4.1 mm 0.1614 4.1000

4.2 mm 0.1654 4.2000

no. 19 0.1660 4.2164

4.3 mm 0.1693 4.3000

no. 18 0.1695 4.3053

11/64 in 0.1719 4.3656

no. 17 0.1730 4.3942

4.4 mm 0.1732 4.4000

no. 16 0.1770 4.4958

4.5 mm 0.1772 4.5000

no. 15 0.1800 4.5720

4.6 mm 0.1811 4.6000

no. 14 0.1820 4.6228

no. 13 0.1850 4.6990

4.7 mm 0.1850 4.7000

3/16 in 0.1875 4.7625

4.8 mm 0.1890 4.8000

no. 12 0.1890 4.8006

no. 11 0.1910 4.8514

4.9 mm 0.1929 4.9000

no. 10 0.1935 4.9149

no. 9 0.1960 4.9784

5 mm 0.1969 5.0000

no. 8 0.1990 5.0546

5.1 mm 0.2008 5.1000

no. 7 0.2010 5.1054

13/64 in 0.2031 5.1594

no. 6 0.2040 5.1816

5.2 mm 0.2047 5.2000

no. 5 0.2055 5.2197

5.3 mm 0.2087 5.3000

no. 4 0.2090 5.3086

5.4 mm 0.2126 5.4000

no. 3 0.2130 5.4102

5.5 mm 0.2165 5.5000

7/32 in 0.2188 5.5563

5.6 mm 0.2205 5.6000

no. 2 0.2210 5.6134

5.7 mm 0.2244 5.7000

no. 1 0.2280 5.7912

5.8 mm 0.2284 5.8000

5.9 mm 0.2323 5.9000

A 0.2340 5.9436

15/64 in 0.2344 5.9531

6 mm 0.2362 6.0000

B 0.2380 6.0452
6.1 mm 0.2402 6.1000

C 0.2420 6.1468

6.2 mm 0.2441 6.2000

D 0.2460 6.2484

6.3 mm 0.2480 6.3000

1/4 in 0.2500 6.3500

E 0.2500 6.3500

6.4 mm 0.2520 6.4000

6.5 mm 0.2559 6.5000

F 0.2570 6.5278

6.6 mm 0.2598 6.6000

G 0.2610 6.6294

6.7 mm 0.2638 6.7000

17/64 in 0.2656 6.7469

H 0.2660 6.7564

6.8 mm 0.2677 6.8000

6.9 mm 0.2717 6.9000

I 0.2720 6.9088

7 mm 0.2756 7.0000
J 0.2770 7.0358

7.1 mm 0.2795 7.1000

K 0.2810 7.1374

9/32 in 0.2813 7.1438

7.2 mm 0.2835 7.2000

7.3 mm 0.2874 7.3000

L 0.2900 7.3660

7.4 mm 0.2913 7.4000

M 0.2950 7.4930

7.5 mm 0.2953 7.5000

19/64 in 0.2969 7.5406

7.6 mm 0.2992 7.6000

N 0.3020 7.6708

7.7 mm 0.3032 7.7000

7.8 mm 0.3071 7.8000

7.9 mm 0.3110 7.9000

5/16 in 0.3125 7.9375

8 mm 0.3150 8.0000

O 0.3160 8.0264
8.1 mm 0.3189 8.1000

8.2 mm 0.3228 8.2000

P 0.3230 8.2042

8.3 mm 0.3268 8.3000

21/64 in 0.3281 8.3344

8.4 mm 0.3307 8.4000

Q 0.3320 8.4328

8.5 mm 0.3347 8.5000

8.6 mm 0.3386 8.6000

R 0.3390 8.6106

8.7 mm 0.3425 8.7000

11/32 in 0.3438 8.7313

8.8 mm 0.3465 8.8000

S 0.3480 8.8392

8.9 mm 0.3504 8.9000

9 mm 0.3543 9.0000

T 0.3580 9.0932

9.1 mm 0.3583 9.1000

23/64 in 0.3594 9.1281

9.2 mm 0.3622 9.2000

9.3 mm 0.3661 9.3000

U 0.3680 9.3472

9.4 mm 0.3701 9.4000

9.5 mm 0.3740 9.5000

3/8 in 0.3750 9.5250

V 0.3770 9.5758

9.6 mm 0.3780 9.6000

9.7 mm 0.3819 9.7000

9.8 mm 0.3858 9.8000

W 0.3860 9.8044

9.9 mm 0.3898 9.9000

25/64 in 0.3906 9.9219

10 mm 0.3937 10.0000

X 0.3970 10.0838

Y 0.4040 10.2616

13/32 in 0.4063 10.3188

Z 0.4130 10.4902

10.5 mm 0.4134 10.5000

27/64 in 0.4219 10.7156

11 mm 0.4331 11.0000

7/16 in 0.4375 11.1125

11.5 mm 0.4528 11.5000

29/64 in 0.4531 11.5094

15/32 in 0.4688 11.9063

12 mm 0.4724 12.0000

31/64 in 0.4844 12.3031

12.5 mm 0.4921 12.5000

1/2 in 0.5000 12.7000

13 mm 0.5118 13.0000

33/64 in 0.5156 13.0969

17/32 in 0.5313 13.4938

13.5 mm 0.5315 13.5000

35/64 in 0.5469 13.8906

14 mm 0.5512 14.0000

9/16 in 0.5625 14.2875

14.5 mm 0.5709 14.5000

37/64 in 0.5781 14.6844

15 mm 0.5906 15.0000

19/32 in 0.5938 15.0813

39/64 in 0.6094 15.4781

15.5 mm 0.6102 15.5000

5/8 in 0.6250 15.8750

16 mm 0.6299 16.0000

41/64 in 0.6406 16.2719

16.5 mm 0.6496 16.5000

17 mm 0.6693 17.0000

43/64 in 0.6719 17.0656

11/16 in 0.6875 17.4625

17.5 mm 0.6890 17.5000

45/64 in 0.7031 17.8594

18 mm 0.7087 18.0000

23/32 in 0.7188 18.2563

18.5 mm 0.7284 18.5000

47/64 in 0.7344 18.6531

19 mm 0.7480 19.0000

3/4 in 0.7500 19.0500

49/64 in 0.7656 19.4469

19.5 mm 0.7677 19.5000

25/32 in 0.7813 19.8438

20 mm 0.7874 20.0000

51/64 in 0.7969 20.2406

20.5 mm 0.8071 20.5000

13/16 in 0.8125 20.6375

21 mm 0.8268 21.0000

53/64 in 0.8281 21.0344

27/32 in 0.8438 21.4313

21.5 mm 0.8465 21.5000

55/64 in 0.8594 21.8281

22 mm 0.8661 22.0000

7/8 in 0.8750 22.2250

22.5 mm 0.8858 22.5000

57/64 in 0.8906 22.6219

23 mm 0.9055 23.0000

29/32 in 0.9063 23.0188

21/23 in 0.9130 23.1913

59/64 in 0.9219 23.4156

23.5 mm 0.9252 23.5000

15/16 in 0.9375 23.8125

24 mm 0.9449 24.0000

61/64 in 0.9531 24.2094

24.5 mm 0.9646 24.5000

31/32 in 0.9688 24.6063

25 mm 0.9843 25.0000

63/64 in 0.9844 25.0031

1 in 1.0000 25.4000
 E
Ebony and ebonizing

The appearance of unvarnished ebony keys and fingerboards can be

improved by applying a drop of linseed oil to 0000 steel wool and rubbing
with the grain. After the oil has been absorbed, the surface should be wiped
down to remove steel wool fragments and excess oil, and then rubbed down
over the next day or two to remove any oil that might come to the surface.
Ebony is not always uniformly black in color and may be grayish or have
streaks of brown or even tan running through it. It can be stained by applying
India ink or a black pigment (such as black bone or ivory black) suspended in
linseed oil or French polish. Aniline dye may also be used.
Woods can be stained black either with pigments or chemically. A
common chemical stain involves applying a decoction of logwood chips, then
of nut galls, followed by a solution of iron sulfate.
Ebonizing (sometimes referred to as “black polishing”) is a French-
polishing technique (see French polishing) employing shellac colored with
drop black (i.e. bone black or so-called “ivory black” pigment) or black
aniline dye. This technique was often used in Europe for finishing piano
cases during the late nineteenth and early twentieth centuries, and in earlier
periods for creating black-accented case moldings, bridges, and nuts, as well
as for dying stringed instrument fingerboards, necks, and pegboxes. In the
early twentieth century, “black polishers” (as these specialized craftsmen
were called) often dissolved broken 78 r.p.m. records, which were made of
black-tinted shellac, to make their polish.
Early violin makers often blackened the outer two strips of their purfling
(Stradivari, for example, used poplar for all three strips) with a chemical
stain, which can be made by boiling or soaking wood in a decoction of
logwood and then applying a solution of copperas (iron sulfate), nut galls,
and vinegar.
References
The Cabinet-Maker's Guide (Concord, 1827).
Ernest Spon, Workshop Receipts (London, 1909).

Electroplating and electrocleaning

The plating of base metals with silver and gold dates back to antiquity. In
the West, this was first accomplished by the technique of foil gilding (i.e.
attaching silver or gold foil or leaf by the application of heat). The Romans,
for example, plated base-metal coins by melting grains of silver onto fluxed
base-metal coins. The Chinese perfected fire-gilding (mercury amalgam
gilding) in the fourth century BC, a technique that reached Rome about two
hundred years later. Mercury amalgam gilding is no longer widely practiced
because of the danger of inhaling poisonous mercury fumes, which are driven
off when the object is heated (see Gilding).
Electroplating dates back to 1772 when Giovanni Battista Beccaria (1716–
1781) deposited a layer of metal using the discharge from a Leyden jar. The
development of the voltaic pile, or battery, by Alessandro Volta (1745–1827)
in 1796 provided a stronger and more sustained source of direct electrical
current, though it was not until the development of the electromagnetic
dynamo by Michael Faraday (1791–1867) around 1830 that electroplating
became more than a laboratory curiosity; by 1860 electroplating had come
into wide use. Today, it is used to plate not only base metals with gold and
silver, but other metals as well, such as chromium, nickel, and copper. Plating
is sometimes done to disguise soldering or to improve appearance (such as
plating pure silver over sterling silver or rhodium over white gold).
Electroplating and electrocleaning are easily carried out using a direct
current (DC) power supply. These power supplies are available through
jeweler's suppliers or can be wired together using a schematic and readily
available transformers, rectifiers, potentiometers, voltmeters and ammeters,
fuses or circuit breakers, etc. Electroplating involves the use of an anode
(attached to the positive terminal of a DC power source) generally made of
the plating metal, an electrolyte containing salts of the metal being plated,
and the object, which is made the cathode (attached to the negative terminal).
To insure uniform thickness of the plating, the anode should have about the
same area and roughly conform to the shape of the object being plated. In
certain instances it is important to maintain a specific voltage and current
density, so the DC power supply should be equipped with voltage and
amperage meters. Highly recommended are models of DC power supplies
(such as those manufactured by Caswell, Inc.) that feature circuitry that
maintains constant current and voltage throughout the plating process.
Prepared copper, nickel, silver, and gold (in different karats and colors)
plating solutions are commercially available, but be aware that they generally
contain sodium or potassium cyanide, which are highly toxic. When
commercial plating solutions are used, follow the manufacturer’s suggestions
regarding voltage, plating bath temperature, and plating time. Certain metals
are not plated directly over others because they do not form a good bond. For
example, to plate chrome over brass, the brass is generally given a “copper
strike” (that is, a thin electroplating of copper), and then one of nickel. The
chrome is then plated over the nickel. When plating gold, silver, or rhodium
over brass, it is advisable to apply a thin, intermediate “flash” deposition of
copper over the brass object.
DC power supplies can also be used to strip plating. To do this, the object
is made the anode, and the electrolyte consists of commercially available
stripping salts (generally cyanide based). Stainless steel can be used as the
cathode. The voltage should be maintained at 6–12 volts.
There are two types of electrocleaning: cathodic, in which the object being
cleaned is made the cathode (negatively charged) and anodic, in which the
workpiece is made the anode and is positively charged. Cathodic
electrocleaning is most often employed in treating art objects. For robust
objects, current densities of 2–4 A/dm2 (ampere per square decimeter) can be
used for brass, 2–3 A/dm2 for nickel silver and silver, 2–5 A/dm2 for lead and
its alloys, and 5–10 A/dm2 for low-carbon steel. Considerably lower current
densities may be required in certain instances; for example a current as low as
0.01 A/dm2 might be employed when electrocleaning delicate objects,
especially to reduce the intensity of evolution of hydrogen bubbles at the
metal's surface. Though the turbulence created by hydrogen bubbles may
assist in the cleaning of robust objects, it might damage fragile ones. Certain
metals (such as high-strength steels) that are susceptible to a phenomenon
termed hydrogen embrittlement should not be cleaned cathodically. (Copper,
its alloys, and silver are not subject to hydrogen embrittlement.) A problem
that may occur when electrocleaning certain alloyed metals, such as silver
(which is often alloyed with copper), is the undesirable plating of one of the
constituent metals (or a metal inadvertently dissolved in the electrolyte) on
the surface of the workpiece. In some cases this can be prevented by making
certain that current densities do not fall below certain values. For example,
silver/copper alloys may develop a salmon-pink coating of copper if the
current density falls below 2 A/dm2. In other cases, it may be possible to
avoid the deposition of copper on silver by carefully controlling the voltage
so that it corresponds with the voltage required to reduce a particular
corrosion product, such as silver sulfide or silver chloride, in preference to
copper corrosion products that may be present. Voltage may be controlled by
use of specially designed constant-voltage-source power supplies or by use of
a reference electrode and a potentiostat controlled by a computer and special
software (Degrigny Jerôme, and Lacoudre, 1994; Espie et al., 2000). For
routine electrocleaning, a 5% sodium hydroxide solution can be used as an
electrolyte at a potential of 3–6 volts. For best results, the electrolyte should
be used close to the boiling point. Platinized titanium or no. 316 stainless
steel can be used as anodes for electrocleaning. To prevent dissolved metals
from plating on the surface of the workpiece, do not turn off the current
before it is withdrawn from the electrolyte. Furthermore, different types of
metal should not be electrocleaned in the same electrolyte bath. A 5 amp
power supply should be sufficient for cleaning musical instrument keys,
mouthpieces, and other small parts; a 50 amp power supply would be
required for cleaning moderate-size brass instruments.
It is important to remove lacquers, varnishes, waxes, oils, and greases from
the workpiece before electroplating, electrocleaning, and electrostripping.
Note that chemical and electrocleaning techniques may alter the surface
composition of alloys.

References
Murray Bovin, Jewelry Making for Schools, Tradesmen, Craftsmen (Forest
Hills, NY, 1971).
C. Degrigny, M. Jerôme and N. Lacoudre, “Surface Cleaning of Silvered
Brass Wind Instruments Belonging to the Sax Collection,” Corrosion
 Australasia 18, no. 2 (1994) pp. 16–17.
Terry Drayman-Weisser, ed., Gilded Metals: History, Technology and
Conservation (London, 2000).
L. Espie, N. Lacoudre, T. Beldjoudi and J. Dugot, “L'electrochimie au service
du patrimoine musical,” Techne (2000), pp. 19–27.
Susan La Niece and Paul Craddock, Metal Plating and Patination: Cultural,
Technical and Historical Developments (Oxford, 2003).

Epoxy removal

Solvents for epoxy glue include methylene chloride, dimethylformamide

(warmed), and 1-methyl-2-pyrrolidinone.
Epoxy softens between 100 and 160 degrees Celsius, so if the object can
bear the heat, warming the cured adhesive with a hot-air gun may weaken the
joint sufficiently.
Uncured components of epoxy glue can be removed with alcohol, xylene,
or acetone.

Equal temperament frequency table (see pages 73–103)

(see also Cents conversion)

Equal temperament frequency table

OCTAVE
C C♯ D D♯ E
0

−50 cents 31.7722 33.66147 35.66309 37.78373 40.03046

−49 cents 31.79056 33.68092 35.68369 37.80556 40.05359

−48 cents 31.80893 33.70038 35.70431 37.8274 40.07673

−47 cents 31.8273 33.71985 35.72494 37.84926 40.09989

−46 cents 31.84569 33.73934 35.74558 37.87113 40.12306

−45 cents 31.86409 33.75883 35.76624 37.89301 40.14624

−44 cents 31.8825 33.77834 35.7869 37.9149 40.16944

−43 cents 31.90093 33.79785 35.80758 37.93681 40.19265

−42 cents 31.91936 33.81738 35.82827 37.95873 40.21587

−41 cents 31.9378 33.83692 35.84897 37.98066 40.23911

−40 cents 31.95625 33.85647 35.86968 38.0026 40.26236

−39 cents 31.97472 33.87603 35.89041 38.02456 40.28562

−38 cents 31.99319 33.89561 35.91114 38.04653 40.3089

−37 cents 32.01168 33.91519 35.93189 38.06851 40.33219

−36 cents 32.03017 33.93479 35.95265 38.09051 40.35549

−35 cents 32.04868 33.95439 35.97343 38.11252 40.37881

−34 cents 32.0672 33.97401 35.99421 38.13454 40.40214

−33 cents 32.08573 33.99364 36.01501 38.15657 40.42548

−32 cents 32.10426 34.01328 36.03582 38.17862 40.44884

−31 cents 32.12281 34.03294 36.05664 38.20068 40.47221

−30 cents 32.14137 34.0526 36.07747 38.22275 40.49559

−29 cents 32.15995 34.07228 36.09832 38.24484 40.51899

−28 cents 32.17853 34.09196 36.11918 38.26693 40.5424

−27 cents 32.19712 34.11166 36.14004 38.28904 40.56583

−26 cents 32.21572 34.13137 36.16093 38.31117 40.58927

−25 cents 32.23434 34.15109 36.18182 38.3333 40.61272

−24 cents 32.25296 34.17082 36.20272 38.35545 40.63618

−23 cents 32.2716 34.19057 36.22364 38.37761 40.65966

−22 cents 32.29024 34.21032 36.24457 38.39979 40.68316

−21 cents 32.3089 34.23009 36.26551 38.42197 40.70666

−20 cents 32.32757 34.24986 36.28647 38.44417 40.73018

−19 cents 32.34625 34.26965 36.30743 38.46639 40.75372

−18 cents 32.36494 34.28945 36.32841 38.48861 40.77726

−17 cents 32.38364 34.30927 36.3494 38.51085 40.80082

−16 cents 32.40235 34.32909 36.3704 38.5331 40.8244

−15 cents 32.42107 34.34893 36.39142 38.55536 40.84799

−14 cents 32.4398 34.36877 36.41245 38.57764 40.87159

−13 cents 32.45854 34.38863 36.43348 38.59993 40.8952

−12 cents 32.4773 34.4085 36.45453 38.62223 40.91883

−11 cents 32.49606 34.42838 36.4756 38.64455 40.94247

−10 cents 32.51484 34.44827 36.49667 38.66688 40.96613

−9 cents 32.53363 34.46818 36.51776 38.68922 40.9898

−8 cents 32.55242 34.48809 36.53886 38.71157 41.01348

−7 cents 32.57123 34.50802 36.55997 38.73394 41.03718

−6 cents 32.59005 34.52796 36.5811 38.75632 41.06089

−5 cents 32.60888 34.54791 36.60223 38.77871 41.08462

−4 cents 32.62772 34.56787 36.62338 38.80112 41.10835

−3 cents 32.64657 34.58784 36.64454 38.82354 41.13211

−2 cents 32.66544 34.60783 36.66571 38.84597 41.15587

−1 cent 32.68431 34.62782 36.6869 38.86842 41.17965

0 cent 32.7032 34.64783 36.7081 38.89087 41.20344

+ 1 cent 32.72209 34.66785 36.72931 38.91334 41.22725

+ 2 cents 32.741 34.68788 36.75053 38.93583 41.25107

+ 3 cents 32.75992 34.70792 36.77176 38.95832 41.27491

+ 4 cents 32.77884 34.72797 36.79301 38.98083 41.29875

+ 5 cents 32.79778 34.74804 36.81427 39.00336 41.32262

+ 6 cents 32.81673 34.76812 36.83554 39.02589 41.34649

+ 7 cents 32.83569 34.78821 36.85682 39.04844 41.37038

+ 8 cents 32.85467 34.80831 36.87812 39.071 41.39429

+ 9 cents 32.87365 34.82842 36.89942 39.09358 41.4182

+ 10 cents 32.89264 34.84854 36.92074 39.11617 41.44213

+ 11 cents 32.91165 34.86868 36.94208 39.13877 41.46608

+ 12 cents 32.93066 34.88882 36.96342 39.16138 41.49004

+ 13 cents 32.94969 34.90898 36.98478 39.18401 41.51401

+ 14 cents 32.96873 34.92915 37.00615 39.20665 41.538

+ 15 cents 32.98778 34.94933 37.02753 39.2293 41.562

+ 16 cents 33.00684 34.96953 37.04892 39.25197 41.58601

+ 17 cents 33.02591 34.98973 37.07033 39.27465 41.61004

+ 18 cents 33.04499 35.00995 37.09175 39.29734 41.63408

+ 19 cents 33.06408 35.03018 37.11318 39.32004 41.65814

+ 20 cents 33.08319 35.05042 37.13462 39.34276 41.68221

+ 21 cents 33.1023 35.07067 37.15608 39.36549 41.70629

+ 22 cents 33.12143 35.09093 37.17755 39.38824 41.73039

+ 23 cents 33.14057 35.11121 37.19903 39.411 41.7545

+ 24 cents 33.15972 35.13149 37.22052 39.43377 41.77862

+ 25 cents 33.17887 35.15179 37.24203 39.45655 41.80276

+ 26 cents 33.19805 35.1721 37.26355 39.47935 41.82692

+ 27 cents 33.21723 35.19243 37.28508 39.50216 41.85108

+ 28 cents 33.23642 35.21276 37.30662 39.52499 41.87526

+ 29 cents 33.25562 35.2331 37.32817 39.54782 41.89946

+ 30 cents 33.27484 35.25346 37.34974 39.57067 41.92367

+ 31 cents 33.29406 35.27383 37.37132 39.59354 41.94789

+ 32 cents 33.3133 35.29421 37.39292 39.61641 41.97213

+ 33 cents 33.33255 35.3146 37.41452 39.6393 41.99638

+ 34 cents 33.35181 35.33501 37.43614 39.66221 42.02064

+ 35 cents 33.37108 35.35543 37.45777 39.68512 42.04492

+ 36 cents 33.39036 35.37585 37.47941 39.70805 42.06922

+ 37 cents 33.40965 35.39629 37.50107 39.731 42.09352

+ 38 cents 33.42896 35.41674 37.52273 39.75395 42.11784

+ 39 cents 33.44827 35.43721 37.54441 39.77692 42.14218

+ 40 cents 33.4676 35.45768 37.56611 39.7999 42.16653

+ 41 cents 33.48693 35.47817 37.58781 39.8229 42.19089

+ 42 cents 33.50628 35.49867 37.60953 39.84591 42.21527

+ 43 cents 33.52564 35.51918 37.63126 39.86893 42.23966

+ 44 cents 33.54501 35.5397 37.653 39.89197 42.26407

+ 45 cents 33.56439 35.56024 37.67476 39.91502 42.28849

+ 46 cents 33.58379 35.58078 37.69653 39.93808 42.31292

+ 47 cents 33.60319 35.60134 37.71831 39.96116 42.33737

+ 48 cents 33.62261 35.62191 37.7401 39.98424 42.36183

+ 49 cents 33.64203 35.64249 37.76191 40.00735 42.38631

+ 50 cents 33.66147 35.66309 37.78373 40.03046 42.4108

OCTAVE
C C♯ D D♯ E
1

−50 cents 63.5444 67.32294 71.32618 75.56745 80.06093

−49 cents 63.58111 67.36184 71.36739 75.61111 80.10718

−48 cents 63.61785 67.40076 71.40862 75.6548 80.15347

−47 cents 63.65461 67.43971 71.44988 75.69851 80.19978

−46 cents 63.69139 67.47867 71.49116 75.74225 80.24612

−45 cents 63.72819 67.51766 71.53247 75.78601 80.29248

−44 cents 63.76501 67.55667 71.5738 75.8298 80.33888

−43 cents 63.80185 67.59571 71.61516 75.87362 80.3853

−42 cents 63.83872 67.63476 71.65654 75.91745 80.43174

−41 cents 63.8756 67.67384 71.69794 75.96132 80.47821

−40 cents 63.91251 67.71294 71.73936 76.00521 80.52471

−39 cents 63.94944 67.75207 71.78081 76.04912 80.57124

−38 cents 63.98638 67.79121 71.82229 76.09306 80.61779

−37 cents 64.02336 67.83038 71.86379 76.13703 80.66437

−36 cents 64.06035 67.86957 71.90531 76.18102 80.71098

−35 cents 64.09736 67.90879 71.94685 76.22504 80.75761

−34 cents 64.1344 67.94803 71.98842 76.26908 80.80427

−33 cents 64.17145 67.98728 72.03002 76.31315 80.85096

−32 cents 64.20853 68.02657 72.07164 76.35724 80.89768

−31 cents 64.24563 68.06587 72.11328 76.40136 80.94442

−30 cents 64.28275 68.1052 72.15495 76.4455 80.99119

−29 cents 64.31989 68.14455 72.19664 76.48967 81.03798

−28 cents 64.35705 68.18392 72.23835 76.53387 81.08481

−27 cents 64.39424 68.22332 72.28009 76.57809 81.13166

−26 cents 64.43145 68.26274 72.32185 76.62233 81.17853

−25 cents 64.46867 68.30218 72.36364 76.6666 81.22544

−24 cents 64.50592 68.34164 72.40545 76.7109 81.27237

−23 cents 64.54319 68.38113 72.44728 76.75522 81.31933

−22 cents 64.58049 68.42064 72.48914 76.79957 81.36631

−21 cents 64.6178 68.46017 72.53103 76.84395 81.41333

−20 cents 64.65513 68.49973 72.57294 76.88835 81.46037

−19 cents 64.69249 68.53931 72.61487 76.93277 81.50743

−18 cents 64.72987 68.57891 72.65682 76.97722 81.55453

−17 cents 64.76727 68.61853 72.6988 77.0217 81.60165

−16 cents 64.80469 68.65818 72.74081 77.0662 81.6488

−15 cents 64.84214 68.69785 72.78284 77.11073 81.69597

−14 cents 64.8796 68.73754 72.82489 77.15528 81.74318

−13 cents 64.91709 68.77726 72.86697 77.19986 81.79041

−12 cents 64.9546 68.817 72.90907 77.24447 81.83766

−11 cents 64.99213 68.85676 72.9512 77.2891 81.88495

−10 cents 65.02968 68.89654 72.99335 77.33376 81.93226

−9 cents 65.06725 68.93635 73.03552 77.37844 81.9796

−8 cents 65.10485 68.97618 73.07772 77.42315 82.02697

−7 cents 65.14246 69.01604 73.11994 77.46788 82.07436

−6 cents 65.1801 69.05591 73.16219 77.51264 82.12178

−5 cents 65.21776 69.09581 73.20446 77.55743 82.16923

−4 cents 65.25544 69.13574 73.24676 77.60224 82.21671

−3 cents 65.29315 69.17568 73.28908 77.64708 82.26421

−2 cents 65.33087 69.21565 73.33143 77.69194 82.31174

−1 cent 65.36862 69.25564 73.3738 77.73683 82.3593

0 cent 65.40639 69.29566 73.41619 77.78175 82.40689

+ 1 cent 65.44418 69.3357 73.45861 77.82669 82.4545

+ 2 cents 65.482 69.37576 73.50105 77.87165 82.50214

+ 3 cents 65.51983 69.41584 73.54352 77.91665 82.54981

+ 4 cents 65.55769 69.45595 73.58602 77.96167 82.59751

+ 5 cents 65.59557 69.49608 73.62853 78.00671 82.64523

+ 6 cents 65.63347 69.53623 73.67107 78.05178 82.69299

+ 7 cents 65.67139 69.57641 73.71364 78.09688 82.74076

+ 8 cents 65.70933 69.61661 73.75623 78.14201 82.78857

+ 9 cents 65.7473 69.65684 73.79885 78.18716 82.83641

+ 10 cents 65.78529 69.69708 73.84149 78.23233 82.88427

+ 11 cents 65.8233 69.73735 73.88415 78.27753 82.93216

+ 12 cents 65.86133 69.77765 73.92684 78.32276 82.98007

+ 13 cents 65.89938 69.81796 73.96956 78.36801 83.02802

+ 14 cents 65.93746 69.8583 74.01229 78.4133 83.07599

+ 15 cents 65.97556 69.89867 74.05506 78.4586 83.12399

+ 16 cents 66.01368 69.93905 74.09785 78.50393 83.17202

+ 17 cents 66.05182 69.97946 74.14066 78.54929 83.22008

+ 18 cents 66.08998 70.0199 74.1835 78.59468 83.26816

+ 19 cents 66.12817 70.06035 74.22636 78.64009 83.31627

+ 20 cents 66.16638 70.10083 74.26925 78.68553 83.36441

+ 21 cents 66.20461 70.14134 74.31216 78.73099 83.41258

+ 22 cents 66.24286 70.18186 74.3551 78.77648 83.46077

+ 23 cents 66.28113 70.22242 74.39806 78.822 83.509

+ 24 cents 66.31943 70.26299 74.44104 78.86754 83.55725

+ 25 cents 66.35775 70.30359 74.48406 78.91311 83.60553

+ 26 cents 66.39609 70.34421 74.52709 78.9587 83.65383

+ 27 cents 66.43445 70.38485 74.57015 79.00432 83.70217

+ 28 cents 66.47284 70.42552 74.61324 79.04997 83.75053

+ 29 cents 66.51125 70.46621 74.65635 79.09565 83.79892

+ 30 cents 66.54967 70.50692 74.69948 79.14135 83.84734

+ 31 cents 66.58813 70.54766 74.74264 79.18707 83.89578

+ 32 cents 66.6266 70.58842 74.78583 79.23283 83.94426

+ 33 cents 66.6651 70.62921 74.82904 79.27861 83.99276

+ 34 cents 66.70362 70.67002 74.87228 79.32441 84.04129

+ 35 cents 66.74216 70.71085 74.91554 79.37025 84.08985

+ 36 cents 66.78072 70.75171 74.95882 79.41611 84.13843

+ 37 cents 66.8193 70.79259 75.00213 79.46199 84.18705

+ 38 cents 66.85791 70.83349 75.04547 79.5079 84.23569

+ 39 cents 66.89654 70.87442 75.08883 79.55384 84.28436

+ 40 cents 66.93519 70.91537 75.13221 79.59981 84.33306

+ 41 cents 66.97387 70.95634 75.17562 79.6458 84.38179

+ 42 cents 67.01256 70.99734 75.21906 79.69182 84.43054

+ 43 cents 67.05128 71.03836 75.26252 79.73786 84.47932

+ 44 cents 67.09003 71.07941 75.30601 79.78394 84.52813

+ 45 cents 67.12879 71.12047 75.34952 79.83003 84.57697

+ 46 cents 67.16758 71.16157 75.39305 79.87616 84.62584

+ 47 cents 67.20638 71.20268 75.43662 79.92231 84.67474

+ 48 cents 67.24522 71.24382 75.4802 79.96849 84.72366

+ 49 cents 67.28407 71.28499 75.52381 80.01469 84.77261

+ 50 cents 67.32294 71.32618 75.56745 80.06093 84.8216

OCTAVE
C C♯ D D♯ E
2

−50 cents 127.0888 134.6459 142.6524 151.1349 160.1219

−49 cents 127.1622 134.7237 142.7348 151.2222 160.2144

−48 cents 127.2357 134.8015 142.8172 151.3096 160.3069

−47 cents 127.3092 134.8794 142.8998 151.397 160.3996

−46 cents 127.3828 134.9573 142.9823 151.4845 160.4922

−45 cents 127.4564 135.0353 143.0649 151.572 160.585

−44 cents 127.53 135.1133 143.1476 151.6596 160.6778

−43 cents 127.6037 135.1914 143.2303 151.7472 160.7706

−42 cents 127.6774 135.2695 143.3131 151.8349 160.8635

−41 cents 127.7512 135.3477 143.3959 151.9226 160.9564

−40 cents 127.825 135.4259 143.4787 152.0104 161.0494

−39 cents 127.8989 135.5041 143.5616 152.0982 161.1425

−38 cents 127.9728 135.5824 143.6446 152.1861 161.2356

−37 cents 128.0467 135.6608 143.7276 152.2741 161.3287

−36 cents 128.1207 135.7391 143.8106 152.362 161.422

−35 cents 128.1947 135.8176 143.8937 152.4501 161.5152

−34 cents 128.2688 135.8961 143.9768 152.5382 161.6085

−33 cents 128.3429 135.9746 144.06 152.6263 161.7019

−32 cents 128.4171 136.0531 144.1433 152.7145 161.7954

−31 cents 128.4913 136.1317 144.2266 152.8027 161.8888

−30 cents 128.5655 136.2104 144.3099 152.891 161.9824

−29 cents 128.6398 136.2891 144.3933 152.9793 162.076

−28 cents 128.7141 136.3678 144.4767 153.0677 162.1696

−27 cents 128.7885 136.4466 144.5602 153.1562 162.2633

−26 cents 128.8629 136.5255 144.6437 153.2447 162.3571

−25 cents 128.9373 136.6044 144.7273 153.3332 162.4509

−24 cents 129.0118 136.6833 144.8109 153.4218 162.5447

−23 cents 129.0864 136.7623 144.8946 153.5104 162.6387

−22 cents 129.161 136.8413 144.9783 153.5991 162.7326

−21 cents 129.2356 136.9203 145.0621 153.6879 162.8267

−20 cents 129.3103 136.9995 145.1459 153.7767 162.9207

−19 cents 129.385 137.0786 145.2297 153.8655 163.0149

−18 cents 129.4597 137.1578 145.3136 153.9544 163.1091

−17 cents 129.5345 137.2371 145.3976 154.0434 163.2033

−16 cents 129.6094 137.3164 145.4816 154.1324 163.2976

−15 cents 129.6843 137.3957 145.5657 154.2215 163.3919

−14 cents 129.7592 137.4751 145.6498 154.3106 163.4864

−13 cents 129.8342 137.5545 145.7339 154.3997 163.5808

−12 cents 129.9092 137.634 145.8181 154.4889 163.6753

−11 cents 129.9843 137.7135 145.9024 154.5782 163.7699

−10 cents 130.0594 137.7931 145.9867 154.6675 163.8645

−9 cents 130.1345 137.8727 146.071 154.7569 163.9592

−8 cents 130.2097 137.9524 146.1554 154.8463 164.0539

−7 cents 130.2849 138.0321 146.2399 154.9358 164.1487

−6 cents 130.3602 138.1118 146.3244 155.0253 164.2436

−5 cents 130.4355 138.1916 146.4089 155.1149 164.3385

−4 cents 130.5109 138.2715 146.4935 155.2045 164.4334

−3 cents 130.5863 138.3514 146.5782 155.2942 164.5284

−2 cents 130.6617 138.4313 146.6629 155.3839 164.6235

−1 cent 130.7372 138.5113 146.7476 155.4737 164.7186

0 cent 130.8128 138.5913 146.8324 155.5635 164.8138

+ 1 cent 130.8884 138.6714 146.9172 155.6534 164.909

+ 2 cents 130.964 138.7515 147.0021 155.7433 165.0043

+ 3 cents 131.0397 138.8317 147.087 155.8333 165.0996

+ 4 cents 131.1154 138.9119 147.172 155.9233 165.195

+ 5 cents 131.1911 138.9922 147.2571 156.0134 165.2905

+ 6 cents 131.2669 139.0725 147.3421 156.1036 165.386

+ 7 cents 131.3428 139.1528 147.4273 156.1938 165.4815

+ 8 cents 131.4187 139.2332 147.5125 156.284 165.5771

+ 9 cents 131.4946 139.3137 147.5977 156.3743 165.6728

+ 10 cents 131.5706 139.3942 147.683 156.4647 165.7685

+ 11 cents 131.6466 139.4747 147.7683 156.5551 165.8643

+ 12 cents 131.7227 139.5553 147.8537 156.6455 165.9601

+ 13 cents 131.7988 139.6359 147.9391 156.736 166.056

+ 14 cents 131.8749 139.7166 148.0246 156.8266 166.152

+ 15 cents 131.9511 139.7973 148.1101 156.9172 166.248

+ 16 cents 132.0274 139.8781 148.1957 157.0079 166.344

+ 17 cents 132.1036 139.9589 148.2813 157.0986 166.4402

+ 18 cents 132.18 140.0398 148.367 157.1894 166.5363

+ 19 cents 132.2563 140.1207 148.4527 157.2802 166.6325

+ 20 cents 132.3328 140.2017 148.5385 157.3711 166.7288

+ 21 cents 132.4092 140.2827 148.6243 157.462 166.8252

+ 22 cents 132.4857 140.3637 148.7102 157.553 166.9215

+ 23 cents 132.5623 140.4448 148.7961 157.644 167.018

+ 24 cents 132.6389 140.526 148.8821 157.7351 167.1145

+ 25 cents 132.7155 140.6072 148.9681 157.8262 167.2111

+ 26 cents 132.7922 140.6884 149.0542 157.9174 167.3077

+ 27 cents 132.8689 140.7697 149.1403 158.0086 167.4043

+ 28 cents 132.9457 140.851 149.2265 158.0999 167.5011

+ 29 cents 133.0225 140.9324 149.3127 158.1913 167.5978

+ 30 cents 133.0993 141.0138 149.399 158.2827 167.6947

+ 31 cents 133.1763 141.0953 149.4853 158.3741 167.7916

+ 32 cents 133.2532 141.1768 149.5717 158.4657 167.8885

+ 33 cents 133.3302 141.2584 149.6581 158.5572 167.9855

+ 34 cents 133.4072 141.34 149.7446 158.6488 168.0826

+ 35 cents 133.4843 141.4217 149.8311 158.7405 168.1797

+ 36 cents 133.5614 141.5034 149.9176 158.8322 168.2769

+ 37 cents 133.6386 141.5852 150.0043 158.924 168.3741

+ 38 cents 133.7158 141.667 150.0909 159.0158 168.4714

+ 39 cents 133.7931 141.7488 150.1777 159.1077 168.5687

+ 40 cents 133.8704 141.8307 150.2644 159.1996 168.6661

+ 41 cents 133.9477 141.9127 150.3512 159.2916 168.7636

+ 42 cents 134.0251 141.9947 150.4381 159.3836 168.8611

+ 43 cents 134.1026 142.0767 150.525 159.4757 168.9586

+ 44 cents 134.1801 142.1588 150.612 159.5679 169.0563

+ 45 cents 134.2576 142.2409 150.699 159.6601 169.1539

+ 46 cents 134.3352 142.3231 150.7861 159.7523 169.2517

+ 47 cents 134.4128 142.4054 150.8732 159.8446 169.3495

+ 48 cents 134.4904 142.4876 150.9604 159.937 169.4473

+ 49 cents 134.5681 142.57 151.0476 160.0294 169.5452

+ 50 cents 134.6459 142.6524 151.1349 160.1219 169.6432

OCTAVE
C C♯ D D♯ E
3

−50 cents 254.1776 269.2918 285.3047 302.2698 320.2437

−49 cents 254.3245 269.4474 285.4695 302.4445 320.4287

−48 cents 254.4714 269.6031 285.6345 302.6192 320.6139

−47 cents 254.6184 269.7588 285.7995 302.7941 320.7991

−46 cents 254.7655 269.9147 285.9647 302.969 320.9845

−45 cents 254.9127 270.0706 286.1299 303.1441 321.1699

−44 cents 255.06 270.2267 286.2952 303.3192 321.3555

−43 cents 255.2074 270.3828 286.4606 303.4945 321.5412

−42 cents 255.3549 270.5391 286.6261 303.6698 321.727

−41 cents 255.5024 270.6954 286.7918 303.8453 321.9129

−40 cents 255.65 270.8518 286.9575 304.0208 322.0989

−39 cents 255.7977 271.0083 287.1233 304.1965 322.285

−38 cents 255.9455 271.1649 287.2892 304.3723 322.4712

−37 cents 256.0934 271.3215 287.4551 304.5481 322.6575

−36 cents 256.2414 271.4783 287.6212 304.7241 322.8439

−35 cents 256.3894 271.6352 287.7874 304.9001 323.0305

−34 cents 256.5376 271.7921 287.9537 305.0763 323.2171

−33 cents 256.6858 271.9491 288.1201 305.2526 323.4039

−32 cents 256.8341 272.1063 288.2865 305.429 323.5907

−31 cents 256.9825 272.2635 288.4531 305.6054 323.7777

−30 cents 257.131 272.4208 288.6198 305.782 323.9648

−29 cents 257.2796 272.5782 288.7865 305.9587 324.1519

−28 cents 257.4282 272.7357 288.9534 306.1355 324.3392

−27 cents 257.577 272.8933 289.1204 306.3123 324.5266

−26 cents 257.7258 273.051 289.2874 306.4893 324.7141

−25 cents 257.8747 273.2087 289.4546 306.6664 324.9018

−24 cents 258.0237 273.3666 289.6218 306.8436 325.0895

−23 cents 258.1728 273.5245 289.7891 307.0209 325.2773

−22 cents 258.3219 273.6826 289.9566 307.1983 325.4653

−21 cents 258.4712 273.8407 290.1241 307.3758 325.6533

−20 cents 258.6205 273.9989 290.2917 307.5534 325.8415

−19 cents 258.77 274.1572 290.4595 307.7311 326.0297

−18 cents 258.9195 274.3156 290.6273 307.9089 326.2181

−17 cents 259.0691 274.4741 290.7952 308.0868 326.4066

−16 cents 259.2188 274.6327 290.9632 308.2648 326.5952

−15 cents 259.3685 274.7914 291.1313 308.4429 326.7839

−14 cents 259.5184 274.9502 291.2996 308.6211 326.9727

−13 cents 259.6684 275.109 291.4679 308.7995 327.1616

−12 cents 259.8184 275.268 291.6363 308.9779 327.3507

−11 cents 259.9685 275.427 291.8048 309.1564 327.5398

−10 cents 260.1187 275.5862 291.9734 309.335 327.729

−9 cents 260.269 275.7454 292.1421 309.5138 327.9184

−8 cents 260.4194 275.9047 292.3109 309.6926 328.1079

−7 cents 260.5699 276.0641 292.4798 309.8715 328.2974

−6 cents 260.7204 276.2237 292.6488 310.0506 328.4871

−5 cents 260.8711 276.3833 292.8179 310.2297 328.6769

−4 cents 261.0218 276.5429 292.987 310.409 328.8668

−3 cents 261.1726 276.7027 293.1563 310.5883 329.0569

−2 cents 261.3235 276.8626 293.3257 310.7678 329.247

−1 cent 261.4745 277.0226 293.4952 310.9473 329.4372

0 cent 261.6256 277.1826 293.6648 311.127 329.6276

+ 1 cent 261.7767 277.3428 293.8344 311.3067 329.818

+ 2 cents 261.928 277.503 294.0042 311.4866 330.0086

+ 3 cents 262.0793 277.6634 294.1741 311.6666 330.1993

+ 4 cents 262.2307 277.8238 294.3441 311.8467 330.39

+ 5 cents 262.3823 277.9843 294.5141 312.0269 330.5809

+ 6 cents 262.5339 278.1449 294.6843 312.2071 330.7719

+ 7 cents 262.6856 278.3056 294.8546 312.3875 330.9631

+ 8 cents 262.8373 278.4665 295.0249 312.568 331.1543

+ 9 cents 262.9892 278.6273 295.1954 312.7486 331.3456

+ 10 cents 263.1411 278.7883 295.366 312.9293 331.5371

+ 11 cents 263.2932 278.9494 295.5366 313.1101 331.7286

+ 12 cents 263.4453 279.1106 295.7074 313.291 331.9203

+ 13 cents 263.5975 279.2719 295.8782 313.4721 332.1121

+ 14 cents 263.7498 279.4332 296.0492 313.6532 332.304

+ 15 cents 263.9022 279.5947 296.2202 313.8344 332.496

+ 16 cents 264.0547 279.7562 296.3914 314.0157 332.6881

+ 17 cents 264.2073 279.9179 296.5626 314.1972 332.8803

+ 18 cents 264.3599 280.0796 296.734 314.3787 333.0726

+ 19 cents 264.5127 280.2414 296.9054 314.5604 333.2651

+ 20 cents 264.6655 280.4033 297.077 314.7421 333.4576

+ 21 cents 264.8184 280.5654 297.2486 314.924 333.6503

+ 22 cents 264.9714 280.7275 297.4204 315.1059 333.8431

+ 23 cents 265.1245 280.8897 297.5922 315.288 334.036

+ 24 cents 265.2777 281.052 297.7642 315.4702 334.229

+ 25 cents 265.431 281.2143 297.9362 315.6524 334.4221

+ 26 cents 265.5844 281.3768 298.1084 315.8348 334.6153

+ 27 cents 265.7378 281.5394 298.2806 316.0173 334.8087

+ 28 cents 265.8914 281.7021 298.453 316.1999 335.0021

+ 29 cents 266.045 281.8648 298.6254 316.3826 335.1957

+ 30 cents 266.1987 282.0277 298.7979 316.5654 335.3893

+ 31 cents 266.3525 282.1907 298.9706 316.7483 335.5831

+ 32 cents 266.5064 282.3537 299.1433 316.9313 335.777

+ 33 cents 266.6604 282.5168 299.3162 317.1144 335.971

+ 34 cents 266.8145 282.6801 299.4891 317.2977 336.1652

+ 35 cents 266.9686 282.8434 299.6621 317.481 336.3594

+ 36 cents 267.1229 283.0068 299.8353 317.6644 336.5537

+ 37 cents 267.2772 283.1703 300.0085 317.848 336.7482

+ 38 cents 267.4316 283.334 300.1819 318.0316 336.9428

+ 39 cents 267.5862 283.4977 300.3553 318.2154 337.1374

+ 40 cents 267.7408 283.6615 300.5289 318.3992 337.3322

+ 41 cents 267.8955 283.8254 300.7025 318.5832 337.5271

+ 42 cents 268.0503 283.9894 300.8762 318.7673 337.7222

+ 43 cents 268.2051 284.1534 301.0501 318.9515 337.9173

+ 44 cents 268.3601 284.3176 301.224 319.1357 338.1125

+ 45 cents 268.5152 284.4819 301.3981 319.3201 338.3079

+ 46 cents 268.6703 284.6463 301.5722 319.5046 338.5034

+ 47 cents 268.8255 284.8107 301.7465 319.6892 338.699

+ 48 cents 268.9809 284.9753 301.9208 319.874 338.8946

+ 49 cents 269.1363 285.14 302.0953 320.0588 339.0905

+ 50 cents 269.2918 285.3047 302.2698 320.2437 339.2864

OCTAVE
C C♯ D D♯ E
4

−50 cents 508.3552 538.5836 570.6094 604.5396 640.4874

−49 cents 508.6489 538.8947 570.9391 604.8889 640.8575

−48 cents 508.9428 539.2061 571.269 605.2384 641.2277

−47 cents 509.2369 539.5177 571.5991 605.5881 641.5982

−46 cents 509.5311 539.8294 571.9293 605.938 641.969

−45 cents 509.8255 540.1413 572.2598 606.2881 642.3399

−44 cents 510.1201 540.4534 572.5904 606.6384 642.711

−43 cents 510.4148 540.7657 572.9213 606.9889 643.0824

−42 cents 510.7097 541.0781 573.2523 607.3396 643.4539

−41 cents 511.0048 541.3907 573.5835 607.6905 643.8257

−40 cents 511.3001 541.7035 573.9149 608.0417 644.1977

−39 cents 511.5955 542.0165 574.2465 608.393 644.5699

−38 cents 511.8911 542.3297 574.5783 608.7445 644.9423

−37 cents 512.1868 542.6431 574.9103 609.0962 645.315

−36 cents 512.4828 542.9566 575.2425 609.4482 645.6878

−35 cents 512.7789 543.2703 575.5748 609.8003 646.0609

−34 cents 513.0752 543.5842 575.9074 610.1526 646.4342

−33 cents 513.3716 543.8983 576.2402 610.5052 646.8077

−32 cents 513.6682 544.2125 576.5731 610.8579 647.1814

−31 cents 513.965 544.527 576.9062 611.2109 647.5554

−30 cents 514.262 544.8416 577.2396 611.564 647.9295

−29 cents 514.5591 545.1564 577.5731 611.9174 648.3039

−28 cents 514.8564 545.4714 577.9068 612.2709 648.6785

−27 cents 515.1539 545.7866 578.2407 612.6247 649.0533

−26 cents 515.4516 546.1019 578.5748 612.9787 649.4283

−25 cents 515.7494 546.4174 578.9091 613.3328 649.8035

−24 cents 516.0474 546.7332 579.2436 613.6872 650.179

−23 cents 516.3455 547.049 579.5783 614.0418 650.5546

−22 cents 516.6439 547.3651 579.9132 614.3966 650.9305

−21 cents 516.9424 547.6814 580.2482 614.7516 651.3066

−20 cents 517.2411 547.9978 580.5835 615.1068 651.6829

−19 cents 517.5399 548.3145 580.9189 615.4622 652.0595

−18 cents 517.839 548.6313 581.2546 615.8178 652.4362

−17 cents 518.1382 548.9483 581.5904 616.1736 652.8132

−16 cents 518.4375 549.2654 581.9265 616.5296 653.1904

−15 cents 518.7371 549.5828 582.2627 616.8858 653.5678

−14 cents 519.0368 549.9003 582.5991 617.2423 653.9454

−13 cents 519.3367 550.2181 582.9357 617.5989 654.3232

−12 cents 519.6368 550.536 583.2726 617.9557 654.7013

−11 cents 519.937 550.8541 583.6096 618.3128 655.0796

−10 cents 520.2374 551.1724 583.9468 618.67 655.4581

−9 cents 520.538 551.4908 584.2842 619.0275 655.8368

−8 cents 520.8388 551.8095 584.6218 619.3852 656.2157

−7 cents 521.1397 552.1283 584.9595 619.7431 656.5949

−6 cents 521.4408 552.4473 585.2975 620.1011 656.9743

−5 cents 521.7421 552.7665 585.6357 620.4594 657.3539

−4 cents 522.0436 553.0859 585.9741 620.8179 657.7337

−3 cents 522.3452 553.4055 586.3127 621.1766 658.1137

−2 cents 522.647 553.7252 586.6514 621.5355 658.494

−1 cent 522.949 554.0451 586.9904 621.8946 658.8744

0 cent 523.2511 554.3653 587.3295 622.254 659.2551

+ 1 cent 523.5535 554.6856 587.6689 622.6135 659.636

+ 2 cents 523.856 555.0061 588.0084 622.9732 660.0172

+ 3 cents 524.1586 555.3267 588.3482 623.3332 660.3985

+ 4 cents 524.4615 555.6476 588.6881 623.6933 660.7801

+ 5 cents 524.7645 555.9686 589.0283 624.0537 661.1619

+ 6 cents 525.0677 556.2899 589.3686 624.4143 661.5439

+ 7 cents 525.3711 556.6113 589.7091 624.7751 661.9261

+ 8 cents 525.6747 556.9329 590.0499 625.136 662.3086

+ 9 cents 525.9784 557.2547 590.3908 625.4972 662.6912

+ 10 cents 526.2823 557.5767 590.7319 625.8586 663.0741

+ 11 cents 526.5864 557.8988 591.0732 626.2203 663.4573

+ 12 cents 526.8906 558.2212 591.4147 626.5821 663.8406

+ 13 cents 527.1951 558.5437 591.7564 626.9441 664.2242

+ 14 cents 527.4997 558.8664 592.0984 627.3064 664.6079

+ 15 cents 527.8045 559.1893 592.4405 627.6688 664.9919

+ 16 cents 528.1094 559.5124 592.7828 628.0315 665.3762

+ 17 cents 528.4145 559.8357 593.1253 628.3943 665.7606

+ 18 cents 528.7199 560.1592 593.468 628.7574 666.1453

+ 19 cents 529.0254 560.4828 593.8109 629.1207 666.5302

+ 20 cents 529.331 560.8067 594.154 629.4842 666.9153

+ 21 cents 529.6369 561.1307 594.4973 629.8479 667.3006

+ 22 cents 529.9429 561.4549 594.8408 630.2118 667.6862

+ 23 cents 530.2491 561.7793 595.1845 630.576 668.072

+ 24 cents 530.5554 562.1039 595.5284 630.9403 668.458

+ 25 cents 530.862 562.4287 595.8724 631.3049 668.8442

+ 26 cents 531.1687 562.7537 596.2167 631.6696 669.2307

+ 27 cents 531.4756 563.0788 596.5612 632.0346 669.6173

+ 28 cents 531.7827 563.4041 596.9059 632.3998 670.0042

+ 29 cents 532.09 563.7297 597.2508 632.7652 670.3913

+ 30 cents 532.3974 564.0554 597.5959 633.1308 670.7787

+ 31 cents 532.705 564.3813 597.9412 633.4966 671.1663

+ 32 cents 533.0128 564.7074 598.2866 633.8626 671.5541

+ 33 cents 533.3208 565.0337 598.6323 634.2289 671.9421

+ 34 cents 533.6289 565.3601 598.9782 634.5953 672.3303

+ 35 cents 533.9372 565.6868 599.3243 634.962 672.7188

+ 36 cents 534.2457 566.0137 599.6706 635.3288 673.1075

+ 37 cents 534.5544 566.3407 600.0171 635.6959 673.4964

+ 38 cents 534.8633 566.6679 600.3637 636.0632 673.8855

+ 39 cents 535.1723 566.9953 600.7106 636.4307 674.2749

+ 40 cents 535.4815 567.3229 601.0577 636.7985 674.6645

+ 41 cents 535.7909 567.6507 601.405 637.1664 675.0543

+ 42 cents 536.1005 567.9787 601.7525 637.5345 675.4443

+ 43 cents 536.4103 568.3069 602.1002 637.9029 675.8346

+ 44 cents 536.7202 568.6352 602.4481 638.2715 676.2251

+ 45 cents 537.0303 568.9638 602.7961 638.6403 676.6158

+ 46 cents 537.3406 569.2925 603.1444 639.0093 677.0067

+ 47 cents 537.6511 569.6215 603.4929 639.3785 677.3979

+ 48 cents 537.9617 569.9506 603.8416 639.7479 677.7893

+ 49 cents 538.2726 570.2799 604.1905 640.1175 678.1809

+ 50 cents 538.5836 570.6094 604.5396 640.4874 678.5728

OCTAVE
C C♯ D D♯ E
5

−50 cents 1016.71 1077.167 1141.219 1209.079 1280.975

−49 cents 1017.298 1077.789 1141.878 1209.778 1281.715

−48 cents 1017.886 1078.412 1142.538 1210.477 1282.455

−47 cents 1018.474 1079.035 1143.198 1211.176 1283.196

−46 cents 1019.062 1079.659 1143.859 1211.876 1283.938

−45 cents 1019.651 1080.283 1144.52 1212.576 1284.68

−44 cents 1020.24 1080.907 1145.181 1213.277 1285.422

−43 cents 1020.83 1081.531 1145.843 1213.978 1286.165

−42 cents 1021.419 1082.156 1146.505 1214.679 1286.908

−41 cents 1022.01 1082.781 1147.167 1215.381 1287.651

−40 cents 1022.6 1083.407 1147.83 1216.083 1288.395

−39 cents 1023.191 1084.033 1148.493 1216.786 1289.14

−38 cents 1023.782 1084.659 1149.157 1217.489 1289.885

−37 cents 1024.374 1085.286 1149.821 1218.192 1290.63

−36 cents 1024.966 1085.913 1150.485 1218.896 1291.376

−35 cents 1025.558 1086.541 1151.15 1219.601 1292.122

−34 cents 1026.15 1087.168 1151.815 1220.305 1292.868

−33 cents 1026.743 1087.797 1152.48 1221.01 1293.615

−32 cents 1027.336 1088.425 1153.146 1221.716 1294.363

−31 cents 1027.93 1089.054 1153.812 1222.422 1295.111

−30 cents 1028.524 1089.683 1154.479 1223.128 1295.859

−29 cents 1029.118 1090.313 1155.146 1223.835 1296.608

−28 cents 1029.713 1090.943 1155.814 1224.542 1297.357

−27 cents 1030.308 1091.573 1156.481 1225.249 1298.107

−26 cents 1030.903 1092.204 1157.15 1225.957 1298.857

−25 cents 1031.499 1092.835 1157.818 1226.666 1299.607

−24 cents 1032.095 1093.466 1158.487 1227.374 1300.358

−23 cents 1032.691 1094.098 1159.157 1228.084 1301.109

−22 cents 1033.288 1094.73 1159.826 1228.793 1301.861

−21 cents 1033.885 1095.363 1160.496 1229.503 1302.613

−20 cents 1034.482 1095.996 1161.167 1230.214 1303.366

−19 cents 1035.08 1096.629 1161.838 1230.924 1304.119

−18 cents 1035.678 1097.263 1162.509 1231.636 1304.872

−17 cents 1036.276 1097.897 1163.181 1232.347 1305.626

−16 cents 1036.875 1098.531 1163.853 1233.059 1306.381

−15 cents 1037.474 1099.166 1164.525 1233.772 1307.136

−14 cents 1038.074 1099.801 1165.198 1234.485 1307.891

−13 cents 1038.673 1100.436 1165.871 1235.198 1308.646

−12 cents 1039.274 1101.072 1166.545 1235.911 1309.403

−11 cents 1039.874 1101.708 1167.219 1236.626 1310.159

−10 cents 1040.475 1102.345 1167.894 1237.34 1310.916

−9 cents 1041.076 1102.982 1168.568 1238.055 1311.674

−8 cents 1041.678 1103.619 1169.244 1238.77 1312.431

−7 cents 1042.279 1104.257 1169.919 1239.486 1313.19

−6 cents 1042.882 1104.895 1170.595 1240.202 1313.949

−5 cents 1043.484 1105.533 1171.271 1240.919 1314.708

−4 cents 1044.087 1106.172 1171.948 1241.636 1315.467

−3 cents 1044.69 1106.811 1172.625 1242.353 1316.227

−2 cents 1045.294 1107.45 1173.303 1243.071 1316.988

−1 cent 1045.898 1108.09 1173.981 1243.789 1317.749

0 cent 1046.502 1108.731 1174.659 1244.508 1318.51

+ 1 cent 1047.107 1109.371 1175.338 1245.227 1319.272

+ 2 cents 1047.712 1110.012 1176.017 1245.946 1320.034

+ 3 cents 1048.317 1110.653 1176.696 1246.666 1320.797

+ 4 cents 1048.923 1111.295 1177.376 1247.387 1321.56

+ 5 cents 1049.529 1111.937 1178.057 1248.107 1322.324

+ 6 cents 1050.135 1112.58 1178.737 1248.829 1323.088

+ 7 cents 1050.742 1113.223 1179.418 1249.55 1323.852

+ 8 cents 1051.349 1113.866 1180.1 1250.272 1324.617

+ 9 cents 1051.957 1114.509 1180.782 1250.994 1325.382

+ 10 cents 1052.565 1115.153 1181.464 1251.717 1326.148

+ 11 cents 1053.173 1115.798 1182.146 1252.441 1326.915

+ 12 cents 1053.781 1116.442 1182.829 1253.164 1327.681

+ 13 cents 1054.39 1117.087 1183.513 1253.888 1328.448

+ 14 cents 1054.999 1117.733 1184.197 1254.613 1329.216

+ 15 cents 1055.609 1118.379 1184.881 1255.338 1329.984

+ 16 cents 1056.219 1119.025 1185.566 1256.063 1330.752

+ 17 cents 1056.829 1119.671 1186.251 1256.789 1331.521

+ 18 cents 1057.44 1120.318 1186.936 1257.515 1332.291

+ 19 cents 1058.051 1120.966 1187.622 1258.241 1333.06

+ 20 cents 1058.662 1121.613 1188.308 1258.968 1333.831

+ 21 cents 1059.274 1122.261 1188.995 1259.696 1334.601

+ 22 cents 1059.886 1122.91 1189.682 1260.424 1335.372

+ 23 cents 1060.498 1123.559 1190.369 1261.152 1336.144

+ 24 cents 1061.111 1124.208 1191.057 1261.881 1336.916

+ 25 cents 1061.724 1124.857 1191.745 1262.61 1337.688

+ 26 cents 1062.337 1125.507 1192.433 1263.339 1338.461

+ 27 cents 1062.951 1126.158 1193.122 1264.069 1339.235

+ 28 cents 1063.565 1126.808 1193.812 1264.8 1340.008

+ 29 cents 1064.18 1127.459 1194.502 1265.53 1340.783

+ 30 cents 1064.795 1128.111 1195.192 1266.262 1341.557

+ 31 cents 1065.41 1128.763 1195.882 1266.993 1342.333

+ 32 cents 1066.026 1129.415 1196.573 1267.725 1343.108

+ 33 cents 1066.642 1130.067 1197.265 1268.458 1343.884

+ 34 cents 1067.258 1130.72 1197.956 1269.191 1344.661

+ 35 cents 1067.874 1131.374 1198.649 1269.924 1345.438

+ 36 cents 1068.491 1132.027 1199.341 1270.658 1346.215

+ 37 cents 1069.109 1132.681 1200.034 1271.392 1346.993

+ 38 cents 1069.727 1133.336 1200.727 1272.126 1347.771

+ 39 cents 1070.345 1133.991 1201.421 1272.861 1348.55

+ 40 cents 1070.963 1134.646 1202.115 1273.597 1349.329

+ 41 cents 1071.582 1135.301 1202.81 1274.333 1350.109

+ 42 cents 1072.201 1135.957 1203.505 1275.069 1350.889

+ 43 cents 1072.821 1136.614 1204.2 1275.806 1351.669

+ 44 cents 1073.44 1137.27 1204.896 1276.543 1352.45

+ 45 cents 1074.061 1137.928 1205.592 1277.281 1353.232

+ 46 cents 1074.681 1138.585 1206.289 1278.019 1354.013

+ 47 cents 1075.302 1139.243 1206.986 1278.757 1354.796

+ 48 cents 1075.923 1139.901 1207.683 1279.496 1355.579

+ 49 cents 1076.545 1140.56 1208.381 1280.235 1356.362

+ 50 cents 1077.167 1141.219 1209.079 1280.975 1357.146

OCTAVE
C C♯ D D♯ E
6

−50 cents 2033.421 2154.334 2282.438 2418.158 2561.95

−49 cents 2034.596 2155.579 2283.756 2419.556 2563.43

−48 cents 2035.771 2156.824 2285.076 2420.954 2564.911

−47 cents 2036.947 2158.071 2286.396 2422.352 2566.393

−46 cents 2038.124 2159.318 2287.717 2423.752 2567.876

−45 cents 2039.302 2160.565 2289.039 2425.152 2569.359

−44 cents 2040.48 2161.814 2290.362 2426.554 2570.844

−43 cents 2041.659 2163.063 2291.685 2427.956 2572.329

−42 cents 2042.839 2164.312 2293.009 2429.359 2573.816

−41 cents 2044.019 2165.563 2294.334 2430.762 2575.303

−40 cents 2045.2 2166.814 2295.66 2432.167 2576.791

−39 cents 2046.382 2168.066 2296.986 2433.572 2578.28

−38 cents 2047.564 2169.319 2298.313 2434.978 2579.769

−37 cents 2048.747 2170.572 2299.641 2436.385 2581.26

−36 cents 2049.931 2171.826 2300.97 2437.793 2582.751

−35 cents 2051.116 2173.081 2302.299 2439.201 2584.244

−34 cents 2052.301 2174.337 2303.63 2440.611 2585.737

−33 cents 2053.486 2175.593 2304.961 2442.021 2587.231

−32 cents 2054.673 2176.85 2306.292 2443.432 2588.726

−31 cents 2055.86 2178.108 2307.625 2444.843 2590.221

−30 cents 2057.048 2179.366 2308.958 2446.256 2591.718

−29 cents 2058.236 2180.626 2310.292 2447.669 2593.215

−28 cents 2059.426 2181.886 2311.627 2449.084 2594.714

−27 cents 2060.616 2183.146 2312.963 2450.499 2596.213

−26 cents 2061.806 2184.408 2314.299 2451.915 2597.713

−25 cents 2062.998 2185.67 2315.636 2453.331 2599.214

−24 cents 2064.19 2186.933 2316.974 2454.749 2600.716

−23 cents 2065.382 2188.196 2318.313 2456.167 2602.218

−22 cents 2066.576 2189.461 2319.653 2457.586 2603.722

−21 cents 2067.77 2190.726 2320.993 2459.006 2605.226

−20 cents 2068.964 2191.991 2322.334 2460.427 2606.732

−19 cents 2070.16 2193.258 2323.676 2461.849 2608.238

−18 cents 2071.356 2194.525 2325.018 2463.271 2609.745

−17 cents 2072.553 2195.793 2326.362 2464.694 2611.253

−16 cents 2073.75 2197.062 2327.706 2466.118 2612.761

−15 cents 2074.948 2198.331 2329.051 2467.543 2614.271

−14 cents 2076.147 2199.601 2330.396 2468.969 2615.782

−13 cents 2077.347 2200.872 2331.743 2470.396 2617.293

−12 cents 2078.547 2202.144 2333.09 2471.823 2618.805

−11 cents 2079.748 2203.416 2334.438 2473.251 2620.318

−10 cents 2080.95 2204.689 2335.787 2474.68 2621.832

−9 cents 2082.152 2205.963 2337.137 2476.11 2623.347

−8 cents 2083.355 2207.238 2338.487 2477.541 2624.863

−7 cents 2084.559 2208.513 2339.838 2478.972 2626.38

−6 cents 2085.763 2209.789 2341.19 2480.405 2627.897

−5 cents 2086.968 2211.066 2342.543 2481.838 2629.415

−4 cents 2088.174 2212.344 2343.896 2483.272 2630.935

−3 cents 2089.381 2213.622 2345.251 2484.706 2632.455

−2 cents 2090.588 2214.901 2346.606 2486.142 2633.976

−1 cent 2091.796 2216.181 2347.962 2487.579 2635.498

0 cent 2093.005 2217.461 2349.318 2489.016 2637.02

+ 1 cent 2094.214 2218.742 2350.676 2490.454 2638.544

+ 2 cents 2095.424 2220.024 2352.034 2491.893 2640.069

+ 3 cents 2096.635 2221.307 2353.393 2493.333 2641.594

+ 4 cents 2097.846 2222.59 2354.752 2494.773 2643.12

+ 5 cents 2099.058 2223.875 2356.113 2496.215 2644.647

+ 6 cents 2100.271 2225.16 2357.474 2497.657 2646.176

+ 7 cents 2101.484 2226.445 2358.837 2499.1 2647.704

+ 8 cents 2102.699 2227.732 2360.199 2500.544 2649.234

+ 9 cents 2103.914 2229.019 2361.563 2501.989 2650.765

+ 10 cents 2105.129 2230.307 2362.928 2503.435 2652.297

+ 11 cents 2106.345 2231.595 2364.293 2504.881 2653.829

+ 12 cents 2107.563 2232.885 2365.659 2506.328 2655.362

+ 13 cents 2108.78 2234.175 2367.026 2507.776 2656.897

+ 14 cents 2109.999 2235.466 2368.393 2509.225 2658.432

+ 15 cents 2111.218 2236.757 2369.762 2510.675 2659.968

+ 16 cents 2112.438 2238.05 2371.131 2512.126 2661.505

+ 17 cents 2113.658 2239.343 2372.501 2513.577 2663.042

+ 18 cents 2114.879 2240.637 2373.872 2515.03 2664.581

+ 19 cents 2116.101 2241.931 2375.244 2516.483 2666.121

+ 20 cents 2117.324 2243.227 2376.616 2517.937 2667.661

+ 21 cents 2118.547 2244.523 2377.989 2519.392 2669.203

+ 22 cents 2119.772 2245.82 2379.363 2520.847 2670.745

+ 23 cents 2120.996 2247.117 2380.738 2522.304 2672.288

+ 24 cents 2122.222 2248.416 2382.113 2523.761 2673.832

+ 25 cents 2123.448 2249.715 2383.49 2525.219 2675.377

+ 26 cents 2124.675 2251.015 2384.867 2526.678 2676.923

+ 27 cents 2125.902 2252.315 2386.245 2528.138 2678.469

+ 28 cents 2127.131 2253.617 2387.624 2529.599 2680.017

+ 29 cents 2128.36 2254.919 2389.003 2531.061 2681.565

+ 30 cents 2129.59 2256.222 2390.384 2532.523 2683.115

+ 31 cents 2130.82 2257.525 2391.765 2533.986 2684.665

+ 32 cents 2132.051 2258.83 2393.147 2535.45 2686.216

+ 33 cents 2133.283 2260.135 2394.529 2536.915 2687.768

+ 34 cents 2134.516 2261.441 2395.913 2538.381 2689.321

+ 35 cents 2135.749 2262.747 2397.297 2539.848 2690.875

+ 36 cents 2136.983 2264.055 2398.682 2541.315 2692.43

+ 37 cents 2138.218 2265.363 2400.068 2542.784 2693.986

+ 38 cents 2139.453 2266.672 2401.455 2544.253 2695.542

+ 39 cents 2140.689 2267.981 2402.843 2545.723 2697.1

+ 40 cents 2141.926 2269.292 2404.231 2547.194 2698.658

+ 41 cents 2143.164 2270.603 2405.62 2548.666 2700.217

+ 42 cents 2144.402 2271.915 2407.01 2550.138 2701.777

+ 43 cents 2145.641 2273.228 2408.401 2551.612 2703.338

+ 44 cents 2146.881 2274.541 2409.792 2553.086 2704.9

+ 45 cents 2148.121 2275.855 2411.185 2554.561 2706.463

+ 46 cents 2149.362 2277.17 2412.578 2556.037 2708.027

+ 47 cents 2150.604 2278.486 2413.972 2557.514 2709.592

+ 48 cents 2151.847 2279.802 2415.366 2558.992 2711.157

+ 49 cents 2153.09 2281.12 2416.762 2560.47 2712.724

+ 50 cents 2154.334 2282.438 2418.158 2561.95 2714.291

OCTAVE
C C♯ D D♯ E
7

−50 cents 4066.841 4308.668 4564.875 4836.317 5123.899

−49 cents 4069.191 4311.158 4567.513 4839.111 5126.86

−48 cents 4071.542 4313.649 4570.152 4841.907 5129.822

−47 cents 4073.895 4316.141 4572.792 4844.705 5132.786

−46 cents 4076.249 4318.635 4575.435 4847.504 5135.752

−45 cents 4078.604 4321.13 4578.078 4850.305 5138.719

−44 cents 4080.961 4323.627 4580.723 4853.107 5141.688

−43 cents 4083.318 4326.125 4583.37 4855.911 5144.659

−42 cents 4085.678 4328.625 4586.018 4858.717 5147.631

−41 cents 4088.038 4331.126 4588.668 4861.524 5150.606

−40 cents 4090.4 4333.628 4591.319 4864.333 5153.582

−39 cents 4092.764 4336.132 4593.972 4867.144 5156.559

−38 cents 4095.129 4338.638 4596.626 4869.956 5159.539

−37 cents 4097.495 4341.144 4599.282 4872.77 5162.52

−36 cents 4099.862 4343.653 4601.94 4875.585 5165.503

−35 cents 4102.231 4346.162 4604.599 4878.402 5168.487

−34 cents 4104.601 4348.674 4607.259 4881.221 5171.474

−33 cents 4106.973 4351.186 4609.921 4884.041 5174.462

−32 cents 4109.346 4353.7 4612.585 4886.863 5177.451

−31 cents 4111.72 4356.216 4615.25 4889.687 5180.443

−30 cents 4114.096 4358.733 4617.917 4892.512 5183.436

−29 cents 4116.473 4361.251 4620.585 4895.339 5186.431

−28 cents 4118.851 4363.771 4623.254 4898.167 5189.428

−27 cents 4121.231 4366.292 4625.926 4900.998 5192.426

−26 cents 4123.612 4368.815 4628.599 4903.829 5195.426

−25 cents 4125.995 4371.339 4631.273 4906.663 5198.428

−24 cents 4128.379 4373.865 4633.949 4909.498 5201.432

−23 cents 4130.764 4376.392 4636.626 4912.334 5204.437

−22 cents 4133.151 4378.921 4639.305 4915.173 5207.444

−21 cents 4135.539 4381.451 4641.986 4918.013 5210.453

−20 cents 4137.929 4383.983 4644.668 4920.854 5213.463

−19 cents 4140.319 4386.516 4647.351 4923.697 5216.476

−18 cents 4142.712 4389.05 4650.037 4926.542 5219.49

−17 cents 4145.105 4391.586 4652.723 4929.389 5222.505

−16 cents 4147.5 4394.124 4655.412 4932.237 5225.523

−15 cents 4149.897 4396.662 4658.102 4935.087 5228.542

−14 cents 4152.294 4399.203 4660.793 4937.938 5231.563

−13 cents 4154.694 4401.745 4663.486 4940.791 5234.586

−12 cents 4157.094 4404.288 4666.18 4943.646 5237.61

−11 cents 4159.496 4406.833 4668.877 4946.502 5240.637

−10 cents 4161.899 4409.379 4671.574 4949.36 5243.665

−9 cents 4164.304 4411.927 4674.273 4952.22 5246.694

−8 cents 4166.71 4414.476 4676.974 4955.081 5249.726

−7 cents 4169.118 4417.026 4679.676 4957.944 5252.759

−6 cents 4171.527 4419.578 4682.38 4960.809 5255.794

−5 cents 4173.937 4422.132 4685.086 4963.675 5258.831

−4 cents 4176.348 4424.687 4687.793 4966.543 5261.869

−3 cents 4178.762 4427.244 4690.501 4969.413 5264.91

−2 cents 4181.176 4429.802 4693.211 4972.284 5267.952

−1 cent 4183.592 4432.361 4695.923 4975.157 5270.995

0 cent 4186.009 4434.922 4698.636 4978.032 5274.041

+ 1 cent 4188.428 4437.485 4701.351 4980.908 5277.088

+ 2 cents 4190.848 4440.048 4704.068 4983.786 5280.137

+ 3 cents 4193.269 4442.614 4706.785 4986.665 5283.188

+ 4 cents 4195.692 4445.181 4709.505 4989.547 5286.241

+ 5 cents 4198.116 4447.749 4712.226 4992.43 5289.295

+ 6 cents 4200.542 4450.319 4714.949 4995.314 5292.351

+ 7 cents 4202.969 4452.89 4717.673 4998.2 5295.409

+ 8 cents 4205.397 4455.463 4720.399 5001.088 5298.469

+ 9 cents 4207.827 4458.038 4723.126 5003.978 5301.53

+ 10 cents 4210.258 4460.613 4725.855 5006.869 5304.593

+ 11 cents 4212.691 4463.191 4728.586 5009.762 5307.658

+ 12 cents 4215.125 4465.769 4731.318 5012.657 5310.725

+ 13 cents 4217.56 4468.35 4734.052 5015.553 5313.793

+ 14 cents 4219.997 4470.931 4736.787 5018.451 5316.864

+ 15 cents 4222.436 4473.515 4739.524 5021.35 5319.936

+ 16 cents 4224.875 4476.099 4742.262 5024.252 5323.009

+ 17 cents 4227.316 4478.686 4745.002 5027.155 5326.085

+ 18 cents 4229.759 4481.273 4747.744 5030.059 5329.162

+ 19 cents 4232.203 4483.863 4750.487 5032.966 5332.241

+ 20 cents 4234.648 4486.453 4753.232 5035.874 5335.322

+ 21 cents 4237.095 4489.046 4755.978 5038.783 5338.405

+ 22 cents 4239.543 4491.639 4758.726 5041.695 5341.489

+ 23 cents 4241.993 4494.235 4761.476 5044.608 5344.576

+ 24 cents 4244.444 4496.831 4764.227 5047.522 5347.664

+ 25 cents 4246.896 4499.43 4766.98 5050.439 5350.754

+ 26 cents 4249.35 4502.029 4769.734 5053.357 5353.845

+ 27 cents 4251.805 4504.63 4772.49 5056.277 5356.939

+ 28 cents 4254.262 4507.233 4775.247 5059.198 5360.034

+ 29 cents 4256.72 4509.837 4778.006 5062.121 5363.131

+ 30 cents 4259.179 4512.443 4780.767 5065.046 5366.23

+ 31 cents 4261.64 4515.05 4783.529 5067.973 5369.33

+ 32 cents 4264.102 4517.659 4786.293 5070.901 5372.432

+ 33 cents 4266.566 4520.269 4789.059 5073.831 5375.537

+ 34 cents 4269.031 4522.881 4791.826 5076.762 5378.642

+ 35 cents 4271.498 4525.494 4794.594 5079.696 5381.75

+ 36 cents 4273.966 4528.109 4797.365 5082.631 5384.86

+ 37 cents 4276.435 4530.726 4800.136 5085.567 5387.971

+ 38 cents 4278.906 4533.343 4802.91 5088.506 5391.084

+ 39 cents 4281.379 4535.963 4805.685 5091.446 5394.199

+ 40 cents 4283.852 4538.583 4808.462 5094.388 5397.316

+ 41 cents 4286.328 4541.206 4811.24 5097.331 5400.434

+ 42 cents 4288.804 4543.83 4814.02 5100.276 5403.555

+ 43 cents 4291.282 4546.455 4816.801 5103.223 5406.677

 + 44 cents 4293.762 4549.082 4819.584 5106.172 5409.801

+ 45 cents 4296.243 4551.71 4822.369 5109.122 5412.926

+ 46 cents 4298.725 4554.34 4825.155 5112.074 5416.054

+ 47 cents 4301.209 4556.972 4827.943 5115.028 5419.183

+ 48 cents 4303.694 4559.605 4830.733 5117.983 5422.314

+ 49 cents 4306.18 4562.239 4833.524 5120.94 5425.447

+ 50 cents 4308.668 4564.875 4836.317 5123.899 5428.582

Ethics

A number of museum and conservation groups (such as the International

Council of Museums and the American Institute for Conservation) as well as
individual conservators have drafted and published rules of ethical conduct
(see References below). These are certainly worth consulting and should be
given due consideration before undertaking any project. Below are some
general principles that should guide conservators in their work:

1 It should be recognized that instruments are affected by time,

environment, use, and neglect. Though it may be technically possible
to compensate for some of the ill effects caused by wood shrinkage,
warpage, structural distortion, damage by insects and other pests, as
well as general wear, certain corrective measures may impair or
destroy important aspects of the original that should be preserved,
and this should be given careful consideration when deciding on a
course of treatment.
2 All restoration work should be guided by knowledge of the maker's
work, then-current manufacturing techniques and materials, and
relevant musical repertoire and performance practice.
3 Treatments should not alter the intended capabilities and
 characteristics of the instrument or attempt to improve upon them.
4 Old or original materials that must be removed in the course of
restoration should be documented with regard to function, precise
position, etc., and preserved if they are not reinstalled.
5 New parts should be marked with the restorer's name and date if
feasible or otherwise documented.
6 Reversible procedures should be employed whenever feasible.
7 Treatments should be selected that will not interfere with future use,
material identification, dating, and research.
8 Conservation or treatment reports (see Conservation reports) should
include suggestions regarding use, general care and handling, and
safe storage. Reports should be readily available to those who
handle, display, and play the instruments, as well as those
contemplating future treatments.

Additional concerns

Many types of early musical instruments had limited lifespans or were not
generally altered to suit changing musical styles or tastes. For example,
instruments such as eighteenth-century one-keyed flutes and three-keyed
oboes could have been “modernized” with the addition of keys, though this
was infrequently done, often because the prevailing pitch had changed or
because such instruments could be replaced with more up-to-date ones at
comparatively little cost. Violins, on the other hand, have exhibited a
remarkable ability to adapt to different musical styles, and with a few minor
alterations and new fittings a violin made by Andrea Amati in the mid to late
sixteenth century can be used in a modern concert hall to perform twentieth-
century repertoire. Instruments such as the harpsichord were frequently
rebuilt to enlarge their keyboard range, to increase the number of registers, or
to alter their pitch. When such complex mechanical changes were made,
harpsichords were often repainted and placed on newer stands to better suit
contemporary décor. The restoration of an instrument that has had a complex
history of alteration poses a dilemma for the conservator, namely, to what
state should it be restored? Grant O'Brien, the noted musicologist and former
curator of the Raymond Russell Collection at the University of Edinburgh,
has addressed this issue at many conferences and most recently on his
website (www.claviantica.com):
 One of the basic tenets of the ethical restoration of musical instruments
is that one should aim to restore the instrument to its last state of
historical use. This, in general, means two things. (1) Any accretions
added to an instrument after the historical period, such as lead weights in
the key levers, sound bars under the soundboard, or the use of any
modern material such as plastic, piano felts, leather plectra, etc. that
would not or could not have been used in the historical period should be
removed. The same would normally apply to any later decorations or
changes to the appearance of the instrument. (2) Although many
instruments have been altered from their original state during the
historical period, no attempt should be made to return the instrument to
its original state. It is the last historical state that should be considered in
the restoration. This means, for example, that a Ruckers harpsichord that
has undergone a petit ravalement or a grand ravalement should not be
returned to its original width, compass, disposition, and decoration. To
do so means that the history of the instrument, which is so important to
our understanding of musical style and performance practice in the
intervening historical period, is destroyed and a great deal of
information is lost in the process.

Certainly, there are many instruments that have been altered to suit changing
musical styles that one might consider restoring to their original
configuration – for example, an original arpeggione made by Johann Georg
Stauffer or a seven-string bass viola da gamba by Nicolas Bertrand that were
crudely converted into cellos, a guitar by Stradivari that was converted from
five double courses to six strings, or an eighteenth-century Parisian
harpsichord by Jean Goermans that was clumsily converted into a piano in
the nineteenth century. In their altered states, these instruments functioned in
a limited capacity, and though there may be sociological and musicological
interest in the historic use of converted instruments, with respect to these
examples compelling arguments could be made for undoing the alterations
and restoring them to their original states.

An earlier perspective

In 1938, the collection of musical instruments that had been assembled at

the Metropolitan Museum of Art by Mrs. John Crosby Brown some fifty
years earlier was languishing without a full-time curator or conservator.
Because of the deterioration of the instruments then on exhibit, the museum
consulted with Dr. Curt Sachs, a noted musicologist and former director of
the musical instrument collection at Berlin's Staatliche Instrumenten
Sammlung, who had fled Nazi Germany and settled in New York in 1937.
Sachs selected several instrument makers to restore a group of the museum's
instruments and oversaw their work. In a letter dated July 6, 1939 addressed
to executives of the Metropolitan Museum of Art, Sachs provided a summary
of the ongoing work:

To the President and the Acting Director

Metropolitan Museum of Art
New York City
Gentlemen:
The decision of the Metropolitan Museum a year ago to begin
restoration of the Crosby Brown Collection of Musical Instruments
causes me to submit a report of the first twelve months at this time.
Three workmen have been employed eighty hours a week, one Mr.
Schlesinger [previously a piano technician employed by the Bechstein
firm], 40 hours regularly, the others Messrs. Staub [who had served a
brief apprenticeship with the Mendler-Schramm harpsichord-making
firm] and Markert [a violin maker], dividing forty hours between them.
Certain operations connected with harpsichord repairing demanded the
collaboration of at least two keyboard workers at one time. This scheme
was found most practical. All three men turned out to be highly
satisfactory.
My first task was to save those instruments which were in the worst
condition. Though playing was not required by the authorities of the
Museum, it was necessary to make the restored instruments ready for
performance, for that is, of course, the criterion of true reconstruction.
Musical instruments can resist temperature conditions only if the
permanent natural tension on which their structure is based, is regained.
Consequently all instruments that have left the workshop have gone out
in playing condition.
The first principle in restoring an instrument is to preserve as many parts
of the original as possible. No pains have been spared in mending,
gluing, ribbing, and studding the innumerable parts worn by age,
circumstances, or clumsy repair. Thus it has been possible to preserve all
soundboards except that of the Cristofori piano [see Pollens, 1989,
1995].
Only those parts which were too worm-eaten to resist restoring have
been replaced. In such cases not only the measurements but also the
various kinds of woods have been reproduced as closely as possible and
specimens of the original parts kept to facilitate research work.
I shall not burden you with further details. It will suffice to innumerate
the items restored within the last twelve months.

I. Keyboard instruments:
1. The Clavichord no. 1215
2. The Kirkman harpsichord no. 1678
3. The Cristofori piano no. 1219
4. The Grovvelius double spinet no. 1196
5. The Zenti harpsichord no. 1220
6. The German claviorganum no. 2741 (not yet completed).
II. Stringed instruments without keyboards:
21 items (viols, lutes, guitars, etc.).
III. Oriental instruments:
1. One of our workmen, Markert, is mending the broken
skins of Oriental drums.
2. The painters' shop of the Museum is restoring those
Oriental instruments of high value on which the lacquer
scales off as the wood shrinks.

The cooperation received from the other Museum employees has been
most helpful.
 Respectfully yours,
Curt Sachs

On November 8 of that year, Sachs addressed a letter to Dr. Brown

(presumably the Reverend Williams Adams Brown, son of Mrs. John Crosby
Brown) in response to questions regarding the future preservation and
utilization of the instruments in the Metropolitan Museum of Art's collection.
Sachs wrote:

The restoration carried out during the last year is a definite responsibility
because if the instruments are not looked after and kept in repair they
will deteriorate and the work will have to be done over again at a later
date. Musical instruments are like flowers. They demand a certain
amount of constant attention or they wilt. It seems a pity to think that the
money spent this year should be practically wasted for want of future
care … It is a pity to see visitors lingering about the room without
knowing how to approach this dead world of muted instruments.
Occasionally, when for a few minutes notes are heard in the gallery,
people float in from the adjoining rooms to listen to the sweet sounds of
the music of yesterday. This is perfectly natural. There is no doubt about
it that the instruments should be played at least once a day as they are in
Leipzig, Munich, Berlin and the other centers of musical instruments
abroad. The late Oscar Sonneck was keenly aware of the indifferent use
made of our collections of musical instruments in this country and he
pointed out long ago that old musical instruments did not have to be
silent continuously – indeed, their entire educational significance
depended on their being played. He might have added that music is a
wonderful relaxation in a museum primarily devoted to the eyes. The
Mannes concerts are of course a strong proof of this fact. Lectures on
the musical instruments could be given at regular intervals by expert
persons in the present gallery and musical illustrations should
accompany them. If something more formal could be arranged it could
perhaps take place in the auditorium with good results, as the concert put
on by the American Musicological Society so happily proved.
As for the replacement of the Cristofori piano's soundboard, despite several
consultations by Sachs and his workmen the wrong wood was selected for the
new soundboard (Atlantic white cedar was used rather than the original
Mediterranean cypress), and only around a square foot of the original was
preserved, which is insufficient to reconstruct the original scaling prior to a
later compass shift, or to determine whether in fact the case has been cut
down, according to a recently proposed theory (Schwarz, 2001).
Another ethical issue concerns the use of certain materials derived from
endangered species in conservation work. At present it is still legal in most
parts of the world to own musical instruments made of these materials,
though with the passage of resolutions by the Convention on International
Trade in Endangered Species of Wild Fauna and Flora (CITES) and the US
Department of the Interior Director's Order no. 210, the international and US
interstate transport, sale, and transfer of such musical instruments – of any
date of manufacture – that were purchased after 1976 are now restricted.
While the author condemns the killing of elephants, whales, and turtles, as
well as the destruction of tropical forests for any purpose, and promotes the
use of synthetic materials in the reconstruction of missing or damaged parts,
he fears that fine instruments will be stripped of parts or fittings made of
ivory, tortoiseshell, whalebone, rosewood, etc., in order to avoid legal action.
Clearly, nothing is gained and no species is spared by prying the ivory facing
off a 200-year-old bow, or even one made a year ago. Furthermore, musicians
must be able to travel freely in pursuing their international careers without
fear of having their instruments confiscated, as well as buy and sell them.

References
Instruments pour demain: Conservation et restauration des instruments de
musique (SFIIC: Champs-sur-Marne, 2000).
Keilinstrumente aus der Werkstatt Ruckers: Schriften des Händel-Hauses in
Halle 14 (Halle an der Saale, 1998).
Stewart Pollens, “Curt Sachs and Musical Instrument Restoration,” The
Musical Times 130/1760 (October, 1989), pp. 589–594.
Stewart Pollens, The Early Pianoforte (Cambridge, 1995), pp. 90–92.
Gabriele Rossi-Rognoni, ed., Restauro e conservazione degli strumenti
musicali antichi: la spinetta ovale di Bartolomeo Cristofori (Florence,
2008).
Kerstin Schwarz, “Bartolomeo Cristofori: Hammerflügel und Cembali im
Vergleich,” Scripta Artium 2 (Autumn, 2001), pp. 23–67.
 F
Flute

For general recommendations and reference materials for maintaining and

restoring flutes, particularly early ones made of wood or ivory, see
Woodwind Instruments.

Figure 8 Turning a graduated ivory cork adjuster for an English flute.

Modern metal flutes have bodies made of plated brass, nickel silver, silver,
gold or platinum. To repair dents, deformations, and areas requiring soldering
see Brass Instruments and Soldering. Special mandrels are available for
reshaping dented or misshapen head and body joints of certain models or
flutes; mandrels can also be fabricated to fit other models (see
Metalworking). When flute tenons wear from repeated assembly and
disassembly, the head and foot joints may be at risk of sliding off. It is
dangerous to rely upon heavy greases to stabilize worn parts; special tenon
expanding tools and swedging dies are available that can be used to achieve a
proper fit. The sections on Electrocleaning and Electroplating and Silver
Cleaning describe techniques for refreshing the surfaces of metal parts.
See Vulcanite and ebonite for recommendations for cleaning head joints
and other parts made of these materials. Additional information can be found
under Boxwood and Ebony and ebonizing.

Fortepiano maintenance

(see also Piano action regulation and voicing (modern grand),

Stringed-keyboard restoration, Tuning and temperament, Wire gauges
for early keyboard instruments)
Conservators often encounter fortepianos in a deplorable state of
preservation. More often than not, they have been improperly restrung with
overly heavy gauges of modern piano wire that have placed undue strain on
the case structure and soundboard. Setting up the action of a 200-year-old
fortepiano is a complex problem, especially if the original key cloths and
action leathers have disintegrated, become badly compressed and hardened –
or, worse, been replaced with inappropriate modern materials. The original
thicknesses of these key cloths and leathers were critical in establishing the
key dip, hammer let-off, check height, and other critical relationships.
Fortunately, a number of fortepiano maintenance manuals were published in
the early nineteenth century that shed light on stringing, action regulation,
and tuning. These manuals include (in order of writing):Johann Heinrich
Ernst Nachersberg, Stimmbuch oder vielmehr: Anweisung wie jeder liebhaber
sein Clavierinstrument selbst repariren und also auch stimmen könne
(Breslau and Leipzig, 1801).Andreas Streicher, Kurze Bemerkungen über das
Spielen, Stimmen und Erhalten der Forte-piano, welche von Stein in Wien
verfertiget werden (Vienna, 1802).Karl Lemme, Anweisung und Regeln zu
einer zweckmässigen Behandlung englischer und teutscher Pianoforte's und
Klaviere/nebst/einem Verzeichnisse der bei dem Verfasser verfertigen Sorten
von Pianoforte's und Klavieren, von Karl Lemme, musikalischen
Instrumenten macher und Organisten in Braunschweig (Braunschweig,
1802).Franz Josef Gall, Clavier-Stimmbuch oder Deutliche Anweisung wie
jeder Musikfreund sein Clavier-flügel, Forte-piano und Flügel-fortepiano
selbst stimmen, repairiren, und bestmöglichst gut erhalten könne. Nebst einer
Nachricht von einigen neuerfundenen musikalischen Instrumenten des herrn
Joseph Wachtl (Vienna, 1805).Christian Friedrich Gottlieb Thon, Ueber
Klavierinstrumente, deren Ankauf, Behandlung und Stimmung. Ein
nothwendiges handbuch für jeden Besitzer dieser Art Metallsaiteninstrumente
(Sonderhausen, 1817).Carl Dieudonné and Johann Lorenz Schiedmayer,
Kurze Anleitung zu einer richtigen Kenntniss und Behandlung der Forte-
pianos in Beziehung und Erhalten derselben, besonders derer, welche in der
Werkstatte von Dieudonné und Schiedmayer in Stuttgart verfertigt werden
(Stuttgart, 1824).

General comments on construction

Of the works cited above, Christian Friedrich Gottfried Thon's handbook

deals at greatest length with the general principles of piano design. He
indicates that the case should be made of solid, well-aged, dry woods such as
oak, ash, maple, walnut, plum, cherry, or mahogany, and that they may be
covered with an expensive veneer, such as mahogany. He notes that the wood
should be clean, knot-free, without warpage, and well joined and glued. The
bottom of the instrument should be quite heavy, and he states that it is a great
mistake to make it too thin, as the case might twist under string tension. He
recommends that the soundboard should be of old, well-dried fir
(Tannenholz) that is smooth, flat, and carefully prepared by the maker. Franz
Josef Gall's Clavier-Stimmbuch indicates that soundboard wood was obtained
from Bohemia or from the Black Forest in boards six to seven Fuss (foot; the
old Viennese foot equals 316.08 mm) long, half a Fuss wide, and one-quarter
Zoll (inch; the old Viennese inch equals 26.34 mm) thick. Concerning the
bridge, Thon advocates that it should be “neither too high nor too low,”
though it should be positioned to produce proper string length and should be
well glued to the soundboard. The principles of establishing string lengths are
not discussed in any of the works cited above, although Thon states that
“herein lies a great part of the true art of an instrument maker.” Concerning
the keyboard, he notes that the keys should be level, the key dip not
excessive, and that the keys themselves “should be neither too wide nor too
small.” Gall states that “the thickness of the key levers is optional” but that
“the under [natural] keys should be three-quarters Zoll in width; the above
[accidental] keys only half that width.”
While these manuals present very little information about the construction
of the fortepiano, they are a bit more helpful with regard to maintenance,
adjustment, and tuning.

Stringing

The Dieudonné and Schiedmayer handbook states that a fortepiano should

be restrung after a period of ten years to give it a more powerful and rounder
tone. If past owners heeded their advice, there may be very little original wire
remaining in early fortepianos, and the vestiges of old wire that we find today
may in fact be from early restringings rather than that installed by the maker.
During the decades of fortepiano making between 1770 and 1830, the
stringing and case structures of pianos became increasingly heavier. One can
well imagine that an instrument made in 1780 might have been “modernized”
with heavier strings even as early as its first restringing, and it is conceivable
that such restringing could have been carried out by the original maker using
his signature hitchpin loops and tuning pin coils. The practice of increasing
the thickness of a fortepiano's strings is not suggested in any of these
manuals, and Thon does in fact discuss the problem involved in matching
string gauges when replacing strings. He notes that some makers mark the
gauge numbers on their instruments, while others do not, and that the string
makers' gauges are not always identical. Additionally, he advises that it is
unwise to restring by matching the gauge markings found on instruments
with those found on the spools of wire, and that gauging the diameters by eye
is a more reliable method. Gall warns against using “blue tempered steel … it
has been made weaker through the action of heat and is not as good as other
steel strings.” He provides eight tables of stringing schedules for a variety of
keyboard instruments, including four types of fortepiano, two harpsichords,
and a clavichord. Identical tables appear in the slightly earlier printing of
Nachersberg's Stimmbuch cited above (see Wire gauges for early keyboard
instruments).
Though Gall's Clavier-Stimmbuch does not propose stringing schedules
that require overspun strings (as were used in English, French, and American
square pianos of the period), he does describe a spinning machine for making
such strings. It consists of
 a long iron rod that has a gear mounted on either end. The rod is turned
by a crank connected to a wheel. Two iron supports hinged to a board
carry this rod. Each stand has under its gear a small gearbox, in which
runs a small spinner of iron mounted in a bearing. These spinners are
mounted close to the wheel, so that when the crank is turned, the gear
teeth mesh and turn on their axles. In addition, through each gearbox
extends a hook which is attached to each spinner, and thus they turn
together. Upon these hooks the wire to be overspun is stretched. With
the right hand, the crank is turned, and with the left the finer wire is
directed such that the string is tightly overspun … Furthermore, the
strings may be overspun with imitation silver.

This description was used by the author to construct a motorized version.

Gall provides practical instructions for replacing strings. He first warns that
“one should be careful when removing wire from the roll or else it might
spring out, and it may become so entangled that one will have to throw the
better part of it out the window.” Once the process of unwinding the wire has
been mastered, he makes a one-Zoll hitchpin loop by doubling over two
“inches” of wire, holding the doubled-over wire together with the thumb and
forefinger, and then after placing the loop over the hook on the tuning
hammer, twisting the hammer eighteen times. When making the loop, he
recommends holding the wire loop together with one's foot, using a doorstep,
sill, or other elevated support. He indicates that six or seven turns of wire
should be made around the tuning pin. Thon also recommends a one-inch
loop, but states that ten to twelve turns are sufficient for heavy wire, while
sixteen to eighteen turns are needed for thinner gauges. He also suggests
running the wire six Zoll beyond the tuning pin to provide a sufficient amount
of wire to make the coil. Gall mentions that “between the nut and the tuning
pin the string must fall slightly from parallel,” thereby providing some down-
bearing. “If the nut is high [with respect to the tuning pin] then the last turn of
the string around the tuning pin must lie above the other turns.” An
illustration of this method is found in the text. In the engraving, the coil is
simply wound upside down, rather than winding up and over a downwardly
wrapped coil.
Thon describes two techniques of cleaning rust from the strings. One
method is to polish with pumice, the second is to rub the string with a small,
round piece of lead. Concerning loose tuning pins, which may arise when
“pins are often removed,” he recommends that new and heavier pins be
installed. Gall suggests that tuning pins be tightened either by driving them in
further with a hammer, or by rubbing them with rosin, or by putting a small
piece of paper in the tuning pin hole.

Action regulation

Even in today's pianos, there is a lack of standardization of key height (the

distance between the key at rest and the bottom of the key frame), hammer-
to-string distance, let-off (the distance between the hammer and the string just
before escapement), key dip (how far the key travels before it is stopped
either by a cloth punching under the key front or an over-rail at the back end
of the key lever), and check height (the height of the hammer after being
caught on the rebound by the back-check; see Piano action regulation and
voicing [modern grand]). The same lack of standardization characterizes the
fortepiano. Unfortunately, most of these critical relationships are not
mentioned in the works cited above. Key dip, for example, is never specified,
nor is a rule of thumb given to establish it. About the only adjustment
described in these manuals is let-off. Dieudonné and Schiedmayer give this in
one instance as 1⅛ Linien (one Linie equals Zoll, or 2.195 mm) and in
another as Zoll. They suggest a method of effecting a general correction in
let-off (as might be required after an instrument has settled in for some length
of time, if key or action cloths have deteriorated (Figure 9), or there has been
a change in humidity): drive a wooden wedge between the tenon of the rail
upon which the escapement levers rest and the front or back walls of the
mortises in the side pieces that support the rail. A hammer blow against the
wedge would then drive the rail forward or backward to alter the engagement
between all of the hammer beaks and their respective escapement levers.
They state that “as a rule one-quarter the thickness of a knife blade is
sufficient” in shifting this rail.
Figure 9 Moth-eaten escapement cloths in a fortepiano by Conrad Graf,
Vienna, c.1839.

With regard to English-style grand and square actions, certain critical

adjustments (such as hammer let-off) can be made with the action installed in
the instrument's case, which is a great advantage. One of the unfortunate
aspects of Viennese action design is that the entire keyboard and action must
be removed from the instrument in order to adjust the hammer let-off. If the
keyboard and action are placed on a flat workbench while this adjustment is
made, the action may not function properly when it is reinstalled in an
instrument that has twisted as a result of string tension. Thus, it may be
necessary to simulate these conditions on the workbench, for example by
wedging up the front right-hand corner of the key frame. An alternative
solution may be to install shims under the key frame when it is replaced.
Another problem resulting from a twisted case may be a distorted (rather than
flat) string band, and it may be necessary to simulate a sinuous strike line so
that hammer let-off and check height can be accurately adjusted at the bench.
This can be done by fitting (with a block plane) a thin slat of wood to the
strings along the strike line and then mounting it at the appropriate height
above the hammers while the action is on the workbench. When establishing
key dip, lost motion, and the point of damper engagement, original key and
action cloths and leathers should provide guidance.
Voicing and tonal characteristics

The construction and leathering of fortepiano hammers have an important

effect upon the tone of an instrument. In these early nineteenth-century
manuals it is surprising that nothing is said about the care and releathering of
hammers or about methods of manipulating the hammer coverings to alter the
tonal characteristics of the instrument. A later source, H. Welcker von
Gontershausen's Der Flügel (Frankfurt am Main, 1856), mentions that the
sheepskin from the firm of Kaindel in Linz and the deerskin sold by
Trumpfer in Vienna are the best for leathering fortepiano hammers. A patent
tanning process devised in 1837 by Johann Gottlieb Steininger of Ortenberg
involved dehairing skins with lime and ashes, tanning with fir bark, treatment
with olive oil and potash, washing with milk, as well as mechanical beating
and scraping with dogfish skin (Harding, 1933). Not only were specially
prepared leathers used for hammer coverings (see Leather), but there were
various means of attaching the leather to the hammer head. Evidently some
makers stretched the leather layers while others apparently did not. Some tied
the outer layers in place, presumably to prevent them from becoming
impregnated with glue and hardening. In early Broadwood pianos employing
leather-covered hammers, a brown-colored, vegetable-tanned (chestnut bark)
sheepskin was often used. The Viennese fortepiano maker Johann Schantz
used hammer coverings of a golden- to reddish-colored oil-tanned deerskin.
Other early Viennese makers used sheepskin, while some later makers of that
school used deerskin of a fibrous, feltlike quality. Dampers were often
covered with tawed sheepskin or calfskin, as leather cured this way is not
only soft but relatively noncorrosive to the iron and brass wire. Thick
hammer-beak leather was sometimes made from elk hide. Hardened
fortepiano hammer coverings and dampers can sometimes be revived by
brushing with a stiff natural bristle or a jeweler's soft wire brush.
Fortunately, the effects of proper leathering are described in general
aesthetic terms in the works cited above. The Andreas Streicher maintenance
manual states that a piano should have a tone “that resembles that of the best
wind instruments” and that “it should not have the so-called Silberton (silver
tone) which becomes the Eisenton (iron tone) with hard playing, and borders
on the dry, thin, and meager.” Dieudonné and Schiedmayer also mention
wind instruments and specify the clarinet and horn as exemplars of good tone
– “The tone should sustain … not get lost in the higher octaves” – and that “a
singing tone is the most important thing … a keyboard instrument should be
able to produce softness or brilliance.” They too warn against the
“Silberton/Eisenton” phenomenon and conclude that “instruments … through
roundness, softness, and pliancy fill the ear and affect our emotions.”

Miscellaneous repairs

Gall provides instructions for repairing soundboard cracks. He advises

shimming and leveling the fill with a knife, or as an alternative simply gluing
a strip of paper over the crack. If the crack needs leveling when shimming, he
suggests cutting a hole in the bottom of the instrument and installing a prop
to level both sides of the crack during the gluing process.
Dieudonné and Schiedmayer recommend that when gluing together a
broken hammer, it is best to add either vinegar or alcohol to the glue. While
searching for the causes of hissing and buzzing noises, they suggest that
damper wires may be the cause, particularly in square pianos. Other
possibilities are the after-lengths of strings, which they suggest can be
muffled with wool cloth.
When replacing the Fagotzug (bassoon stop), Thon recommends using a
roll of either thin parchment or paper. Gall recommends only paper.
Dieudonné and Schiedmayer recommend lubricating the key pins every
two years with goose grease, bone marrow, or olive oil. They also suggest
yearly oiling of the hammer kapsels if they are made of brass; if made of
wood, the kapsels should have their felt bushings enlarged if the hammers
stick. They add that a little oil may be used to lubricate felt bushings: even
though “it is against the rules … do whatever works.” If a hammer drags
against the escapement lever, they recommend burnishing the front surface of
the escapement lever with pencil lead until it shines, or bending the spring
behind the escapement lever to weaken it. Lemme provides an interesting
remedy for tight balance-pin mortises – he recommends rubbing the pin with
Wasserblei (molybdenum disulfide, a dry lubricant similar to graphite).
To clean and polish the cases of fortepianos, Thon suggests using an oil-
impregnated wool cloth, but cautions against using warm water. For
polishing ivory, ebony, mother-of-pearl, and bronze mounts, he recommends
whiting or tripoli, but never with oil. Gall, on the other hand, suggests that
olive oil and tripoli can be used to polish ebony keys, and that pumice be
used for cleaning ivory. Several authors recognize the problems created by
heat, cold drafts, and changing humidity and suggest that instruments be
protected with covers. Gall recommends that cracks in the lid and hinges be
sealed by gluing strips of parchment over them to prevent dust from entering
the workings of the instrument.
Lengthy, step-by-step troubleshooting procedures are given in Gall and
Thon for the elimination of problems encountered with hammer actions.
These involve checking to see whether the hammers are striking the right
strings, if hammer heads have come unglued, or if key cloths have become
displaced or worn out. Typical problems such as stuck keys and ripped
parchment hinges are dealt with in detail.

Broken hammer shanks

If a broken shank and a new piece that is to be grafted onto it are held
together and planed at the same angle, they will line up perfectly when the
graft is turned around for gluing. If a block plane is clamped in a workbench
vise with the sole facing up, the planing operation is easier. If the break is
short (for example, straight across at the hammer head), the shank and new
extension can be planed at an acute angle without appreciably shortening the
original section, and thereby providing an extended gluing surface. Wrapping
with thread is a convenient method of holding the two pieces together, and
one can shove the action back into the piano while the glue dries. However,
the thread should be removed after the glue has set. The author suspects that
one often finds thread wrapped around old repaired hammer shanks because
tuners did not trouble themselves to come back the next day to remove it.

Making hitchpin loops

The loop-making tools sold by piano-technician suppliers work very well

for iron and steel wire (they make neat, consistent loops, and one can adjust
the length of the loop), but brass wire has a tendency to crack when these
devices are used. This is due to the wire becoming overstressed as it is
twisted during the loop-making process. To avoid this problem, it is
necessary to allow both the loose end and the section of wire coming off the
spool to slip and untwist as the loop is being made. By lightly clamping the
end of the wire coming off the spool against a tabletop with a cork-faced cam
clamp and placing a piece of felt under the wire, sufficient pressure can be
applied to make a tight loop while allowing the end of the wire coming off of
the spool to unwind as the loop is being formed. The free end (which can be
held in a pair of pliers or between fingertips) can be released every few turns
to relieve stress at that end. To make the loop, the hook at the end of a T-
handle tuning wrench (or one fashioned from a headless nail chucked to a
stout pin vise) must be held in such a way that it bisects the angle (around
60°) created by the wire coming off the spool and the free end. If the angle is
not bisected, the two ends of the wire will not twist around each other evenly.
Hitchpin loops made in brass wire sometimes unravel as the wire work-
hardens over time (this may happen years after the string was installed);
however, if the wire is given a sufficient number of turns (around 12 per inch
over an inch and a half), the hitchpin loop will hold for the life of the string.
Some technicians and makers simply spool out a long section of wire
(especially when making loops in brass wire) and clamp it firmly (rather than
allowing it to slip, as suggested above). If this procedure is followed, the wire
will develop a twist between the loop and the clamp, which may cause the
string to sound false if the twisted section extends past the bridge pin and
forms part of the speaking length of the string.

References
Claire Chevallier and Jos van Immerseel, eds., Matière et Musique: The
Cluny Encounter (Antwerp, 2000).
Peter Donhauser, ed., Restaurieren, renovieren, rekonstruieren: Methoden
für Hammerklaviere (Vienna, 1997).
Rosamond Harding, The Piano-Forte: Its History Traced to the Great
Exhibition of 1851 (Cambridge, 1933).
Stewart Pollens, “Early Nineteenth-Century German Language Works on
Piano Maintenance,” Early Keyboard Journal 8 (1990), pp. 91–109.
Thomas Steiner, ed., Instruments à Claviers – Expressivité et Flexibilité
Sonore (Bern, 2004).
H. Welcker von Gontershausen, Der Flügel (Frankfurt am Main, 1856).

French polishing

Though reviled by many modern conservators, French polishing is a

legitimate and highly refined wood-finishing technique that was apparently
developed in the early nineteenth century. Despite admonitions against its
use, it remains an important method of reviving both spirit and oil varnishes,
as well as French-polished objects, which often include the casework of
pianos as well as the bodies of violins, guitars, and other stringed
instruments.
The earliest known published reference to the technique of French
polishing is found in the anonymous The Cabinet-Maker's Guide; or Rules
and Instructions in the Art of Varnishing, Dying, Staining, Japanning,
Polishing, Lackering, and Beautifying Wood, Ivory, Tortoise-shell, and Metal
(Concord, 1827). The title page indicates that this American printing was
“drawn from the latest London edition” published in 1825. The earlier 1809
and 1819 editions of this work do not mention the term “French polishing” or
describe the process, and the author of the 1825 edition indicates that this
technique “is of comparatively modern date.” Why it was termed “French
polishing” is not known, as this method of applying spirit varnish is not
described in earlier French sources, such as Watin's L'Art du peintre, doreur,
vernisseur (Paris, 1772) or Roubo's L'Art du menuisier (Paris, 1769–1774),
though it is found in later French writings, such as Nosban and Maigne's
Nouveau Manuel Complet de l'Ébéniste (Paris, 1887). Curiously, spirit
varnishes similar in formulation to that used in French polishing were
referred to as vernis anglais (English varnish) in some French sources (such
as A. Romain's Nouveau Manuel complet de fabricant de vernis de toute
espèce, published in Paris in 1888).
In the American edition of The Cabinet-Maker's Guide, “friction
varnishing, or French polishing” is described as a “method of varnishing
furniture, by means of rubbing it on the surface of the wood.” Its author states
that “a coat of clear size” must first be applied to the wood, which prevents
the polish from “being absorbed into the wood.” The polishing pad used to
apply the varnish is described as “a wad … of coarse flannel or drugget
[made by] rolling it round and round, over which, on the side meant to polish
with, put a very fine linen rag several times doubled, to be as soft as
possible.” This polishing pad is then held up against the mouth of a bottle
containing the polish, which is then shaken in order to dampen the pad. The
work is then rubbed “in a circular direction, observing not to do more than
about a square foot at a time, [rubbing] lightly till the whole surface is
covered, repeat this three or four times, according to the texture of the wood,
each coat to be rubbed until the rag appears dry, and be careful not to put too
much on the rag at a time, and you will have a very beautiful and lasting
polish.”
The Cabinet-Maker's Guide neglects to mention one very important
procedure that is essential to the technique, especially when French-polishing
small objects such as violins and guitars: that a small amount of oil must be
deposited on the surface of the pad to prevent it from sticking as one circles
over freshly applied polish. This author suggests holding a palm over a
narrow-necked bottle of raw cold-pressed linseed oil and inverting it, thereby
picking up a small quantity of oil on one's hand, and then pressing the
charged polishing pad into it. In this way, a light film of oil can be spread
across the pad's surface. Another technique is to use an atomizer to apply a
mist of oil on the pad. In any case, very little oil should be used – just enough
to prevent the pad from sticking. After the object has been sufficiently
polished, another pad lightly moistened with ethanol is lightly rubbed over
the work to remove the oil. This is known as “oiling out.” The author likes to
finish with strokes running with the grain rather than with a circular motion.
The advantages of French polishing are that one can build up a highly
polished finish in very little time and that the finish is virtually dry to the
touch immediately after the oiling-out process (though it is advisable to let
the instrument rest for a while before handling or using it). Because of the
solvent power of the ethanol and the fluidity of the shellac, French polish
tends to amalgamate underlying varnishes, thereby reducing cloudiness and
opacity created by crazing, as well as the visibility of scratches.
To get the knack of French polishing, one can practice on cheap pieces of
furniture and small wooden objects. Commonly made mistakes include
applying too much shellac to the pad and using insufficient or too much oil.
Some French polishers use mineral or baby oil instead of linseed oil. This is
not good practice because the oiling-out process never completely removes
all of the oil, and non-drying oils that are trapped between layers of shellac
have a tendency to break through the polished surface. Crosssections of
French-polished wood stained with Sudan Black (a dye that stains
triglycerides and lipids) and viewed under an epi-fluorescence microscope
often reveal remnants of oil between the layers of shellac.
Early references to French polishing recommend a variety of combinations
of resins dissolved in alcohol, including shellac, mastic, sandarac, Venice
turpentine, and colophony (pine resin). Today, orange or bleached, dewaxed
shellac is generally used alone, though French-trained luthiers often add some
gum benzoin to the shellac to provide a somewhat softer finish (as well as a
lovely scent). A tincture of benzoin has sometimes been used as a final glaze,
though this cannot be recommended because it is a bit sticky to the touch. A
“1½ pound cut” (i.e. 1½ pounds of shellac dissolved in 1 gallon of ethanol) is
a typical proportion. The author prefers bleached and dewaxed shellac
dissolved in 95% ethanol (such as 190-proof Everclear brand grain alcohol).
Over time, orange or bleached and dewaxed shellac flakes have a tendency to
become insoluble in alcohol. This occurs even if the flakes are stored in dark,
well-sealed bottles, so it is best to use fresh supplies when mixing batches of
polish. If the shellac does not dissolve at room temperature overnight, then
discard it and use fresh flakes. Once dissolved in ethanol, however, shellac
has a fairly long shelf life, even in partially full bottles.
Early sources often give elaborate instructions for folding pads of linen
cloth for the application of the shellac; however, a neatly folded piece of
ordinary cheesecloth covered with an outer layer of the same material will
suffice (a lumpy pad may interfere with an even application of polish and
produce streaks). Whatever material one chooses to make one's polishing
pad, the fabric should not contain starch and should be lint-free.
If French polishing is used to revive an old finish, such as the case of an
English square piano, the underlying varnish, or French polish, should first be
cleaned as best as possible. One way to accomplish this is to use a polishing
pad lightly charged with ethanol (95%). When cleaning the lid, case sides,
and stand, wipe the surfaces with a light, quick motion, following the
direction of the grain (rather than using a circular pattern). Keep replacing the
outer surface of the pad with fresh cloth until the wiping action ceases to
bring up grayish matter (which is dirt and/or varnish or French polish that has
become embedded with dirt) and begins to bring up clean, yellowish varnish
or French polish. At that point, one can begin to polish with a fresh pad
charged with shellac. In building up the surface, apply a light film that is just
sufficient to even out and renew the surface of the original varnish, if that can
be ascertained. The bad reputation of French polish is primarily due to the
tendency of polishers to apply too much of it and thereby obliterate the
texture of the underlying varnish with a heavy, glassy overcoat.
Some early instructions for the French polishing of piano cases involve
what was termed an “acid finish,” which was achieved by sprinkling the
freshly French-polished surface with Vienna chalk and then going over this
with dilute sulfuric acid (made by gradually adding the acid to water to
produce a 1:8–1:10 solution). The acid solution was applied over the chalk in
circular or straight motions with the palm of one's hand. This was said to
create a creamy paste with the chalk, and one continued polishing until the
chalk dried and could be removed as a fine powder. The purpose of this
technique was to harden the surface of the shellac. Clarified ox-gall was
sometimes used after the acid process or after spiriting-off with alcohol. The
author does not recommend either of these techniques, though wiping down
freshly applied French polish with a pad lightly moistened with hexanes may
help remove residual oil that remains after spiriting-off.

A final warning on French polishing

Among antiquarians, dealers, restorers, and perhaps a few museum

curators, there is a growing awareness of the value of preserving original
finishes. Unfortunately, there are very few musical instruments (be they
violins, guitars, lutes, harpsichords, or pianos) that have not been adulterated
with household polishes, over-varnishing, French polishing, or in some cases
by complete over-varnishing or revarnishing. Furthermore, any instrument
that has evaded such interventions is likely to be in an otherwise horrendous
state of preservation, which may include loss of parts and veneer, severe
scratches and bruises, water damage, accumulated filth, etc. If one is
fortunate enough to encounter a well-preserved specimen with its original
varnish, one should certainly attempt to convince the owner to preserve the
finish. If they are not amenable, one can always find an area (such as the
underside of the lid, the back surfaces of the legs or stand, etc.) that can be
left untouched. Another approach would be to clean the surface as best as
possible, apply a thin barrier coat of Regalrez 1078 or 1094, and then French
polish over it. Regalrez is insoluble in alcohol and can later be removed with
mineral spirits, along with the overlay of French polish.

References
The Art and Science of Gilding: A Handbook of Information for the Picture
Framer (Rochester, NY, 1909).
The Cabinet-Maker's Guide, or Rules and Instructions in the Art of
Varnishing, Dying, Staining, Japanning, Polishing, Lackering, and
Beautifying Wood, Ivory, Tortoise-shell, and Metal (Concord, NH, 1827).
The French Polisher's Handbook, with a Section on Gilding and Bronzing by
“A Practical Man” (London, n.d.).
Paul N. Hasluck, Woodfinishing, Comprising Staining, Varnishing, and
Polishing (London, Paris, New York, and Melbourne, 1906).
R. Moore, The Artizans' Guide and Everybody's Assistant (Montreal, 1874).
N. Nosban and W. Maigne, Nouveau Manuel complet de l'ébéniste (Paris,
1887).
A. Romain, Nouveau Manuel complet de fabricant de vernis de toute espèce
(Paris, 1888).
André-Jacques Roubo, L'art du menuisier (Paris, 1769–1774),
Ernest Spon, Workshop Receipts for Manufacturers and Scientific Amateurs
(London and New York, 1909).
J. Stokes, The Cabinet-Maker and Upholsterer's Companion (Philadelphia,
1889).
Jean-Felix Watin, L'Art du peintre, doreur, vernisseur (Paris, 1772).
The Woodworker Series: Staining and Polishing (Philadelphia and London,
n.d.).

Frets, tied-gut and fixed (for guitars, lutes, and viols)

 Frets for guitars, lutes, and viols are generally positioned according the
18:17 rule (i.e. the speaking length of the string is divided into eighteen parts
and the first fret is positioned on the 17th division from the bridge; to
position subsequent frets, the speaking length is taken from the previously
positioned fret, divided into eighteen parts, and the next fret placed on the
17th division from the bridge, and so on). The 18:17 rule, which was
advocated by Vincenzo Galilei (1581), Mersenne (1637), and others,
provides a good approximation of equal temperament. Other methods include
dividing the overall string length and successive speaking lengths by the
twelfth root of 2, or 1.059463, to get the distance between the bridge and
each fret, or dividing the string length and successive speaking lengths by
17.817, which gives the distance between the nut and the first fret, as well as
distances between successive frets. Historically, Pythagorean and variations
of meantone fretting systems have also been advocated, the former being of
extremely limited utility, though meantone fretting still has its advocates. The
movable frets used in lutes, early guitars, and viols enable players to
experiment with and employ tunings other than equal temperament.
There are several methods of tying on frets. One way of tying a single fret
is to make a loose overhand knot in one end; pass the free end around the
fingerboard and through the knot, and then pull both ends tight. The ends are
then trimmed close to the knot and singed with a soldering iron. The
mushroomed ends will prevent the knot from loosening. A more secure knot
can be made by simply tying the free ends in a square or reef knot and
trimming closely, singeing being optional in this case. Double frets can be
made by doubling a length of gut to create a bight, passing both of the free
ends around the neck, directing one end through the bight and the other
around it, pulling both tight, tying the two ends together with a square knot,
and trimming the ends closely; again, singeing with a soldering iron is
optional. Frets should be tied further up the neck where it is narrower so that
they will be become tighter when they are shifted into position.
When fretting instruments with shallow actions, such as lutes, it is
generally necessary to use progressively thinner gut as one proceeds from the
nut down to the end of the fingerboard. In his Varietie of Lute-Lessons
(1610), Robert Dowland advises that when tying frets on lutes and viols, the
first two frets nearest the “head” (pegbox) should be made of gut having the
diameter of the “countertenor” strings (his fourth course), that the third and
fourth frets should be made of gut having the diameter of the “great meanes”
(third course), that the fifth and sixth frets should have the thickness of the
“small meanes” (second course), and that the rest should be as thick as the
“trebles” (first course). Robert Lundberg, author of Historical Lute
Construction (2002), advocates fret gradations of 1.0 mm or 1.1 mm ranging
down to 0.7 mm or 0.66 mm.

References
Robert Dowland, Varietie of Lute-Lessons (London, 1610).
Charles Ford, ed., Making Musical Instruments: Strings and Keyboard (New
York, 1979).
Vincenzo Galilei, Dialogo della musica antica et della moderna (Florence,
1581).
Paul N. Hasluck, Handbook of Knotting and Splicing (London, 1904).
Mark Lindley, Lutes, Viols and Temperaments (Cambridge, 1984).
Robert Lundberg, Historical Lute Construction (Tacoma, 2002).
Marin Mersenne, Harmonie universelle (Paris, 1636–1637).
Des Pawson, The Handbook of Knots (New York, 1998).
 G
Gilding

Several types of gilding have been used in musical instrument making:

water gilding, oil gilding, mercury amalgam gilding, electrochemical
deposition, and electrogilding. When restoring gilded surfaces, one must
identify the type of gilding being repaired; the surfaces, mordants, sizes, etc.
should be built up in precisely the same manner as the undamaged or
surrounding areas.

Water gilding

This type of gilding is often found on seventeenth- and eighteenth-century

harpsichord stands and harps. In this process, ornamental wood carving was
covered with layers of gesso grosso (coarse gesso), then gesso sottile (fine
gesso), and finally bole prior to the application of gold leaf. The function of
the gesso is to conceal the grain of the wood and provide a smooth, receptive
surface that can be shaped and ornamented with recutting tools, stamps,
burnishers, and other techniques. To make gesso, Italian artisans generally
used plaster of Paris (calcium sulfate) whereas French gilders tended to use
whiting (calcium carbonate).
The wood surface is first sized with hot rabbitskin glue (1:10 with water)
and allowed to dry.
To make gesso grosso, prepare rabbitskin glue with water in the same
proportion as above. Carefully sift plaster of Paris or gilder's whiting into a
quantity of this glue until it makes a stiff paste (adding water to dry plaster
tends to form bubbles that may remain in the set plaster). The gesso is applied
with a bristle brush, and several coats may be used. After the gesso is
thoroughly dry, it was traditionally smoothed down with cabinet scrapers and
abrasive materials, such as dogfish skin, “horsetail” (Equisetum, scouring
rush), and tamiso (a rough cloth made of woven horsehair). With the advent
of sandpaper in the early nineteenth century, this material was also used.
After applying the gesso, details in decorative moldings and underlying
carving that may have gotten clogged can be sharpened with recutting tools
(Figure 10). Other ornamental details can be added at this point as well.

Figure 10 Gilding tools including mops, gilder's tip, agate burnishers,

cushion, knife, books of gold leaf, and recutting tools.

Gesso sottile is then applied over the gesso grosso. Gesso sottile is made
with slaked plaster of Paris or gilder's whiting. Once slaked, the plaster of
Paris or gilder's whiting will no longer set or harden when remixed with
water. To slake plaster of Paris or whiting, it is sifted into a large quantity of
water (for example, a pound of plaster can be added to a gallon of water), and
is then left to stand for at least a few hours (some early recipes, such as that
found in Cennini's fifteenth-century Il Libro dell’ Arte, suggest letting the
plaster soak for a month, but this is not necessary). The water is then poured
off, and the slaked plaster is placed in a piece of cloth, which is wrung out to
remove most of the water. The plaster is then formed into little cakes and left
to dry. When needed, the cakes are soaked in water, thoroughly pulverized
with a glass muller, and then left to dry once more. This finely powdered
material is then sifted into a double boiler containing warm glue size until it
reaches the consistency of a light batter. It is important that this mixture not
be brought to a boil. The gesso sottile is then lightly brushed over the gesso
grosso. Several coats may be applied. One of the properties of gesso sottile is
that it can be water-polished, that is, rubbed over with a lightly moistened
cloth or pad to bring up a very smooth surface.
Bole is a soft clay, generally colored, though sometimes white, that is
emulsified with size until it reaches a creamy consistency. This is applied on
top of the gesso sottile and creates a surface that can be burnished after the
gold leaf has been applied to it. Bole is commercially available in several
forms: in dried cakes and in paste form. The paste form is easier to work with
as it eliminates the step of mulling the dried cakes with water. If the paste
form is used, take 4 parts of the bole paste and combine it with 3 parts warm
water until it reaches the consistency of light cream. If dried cakes are used,
add warm water until this creamy consistency is reached, though make sure
that the clay is well mulled on a glass slab to eliminate any coagulated
masses of clay. Then warm 1:10 rabbitskin glue (4 parts) and gradually stir it
into the above creamy mixture until it begins to stiffen and stand in a peak
(like mayonnaise). After it has reached that consistency, slowly add more
glue until it reaches the consistency of light cream. Several coats of bole may
be applied.
When repairing damaged surfaces, it is important to match the color of the
bole with the surrounding area, as this affects the hue of the gold that is
ultimately laid on top. Bole is available in several basic colors: black (or
gray), yellow (ochre-colored), red (burnt-sienna-colored), and white, but
these colors can be adjusted by mixing them together or by adding
watercolors from tubes. After the bole has dried, the gold (or silver) leaf is
laid down. Today, gold leaf comes in a variety of karats (24, 23¾, 23½, etc.),
which should be matched for color to the surrounding areas. When selecting
the proper karat leaf, keep in mind that old surfaces may be soiled or covered
with varnish, which affects their hue. It is helpful to have sample pieces of
veneer or cardboard gilded with different karats of leaf, sections of which can
be varnished and “patinated” to ascertain which karat of leaf to use when
reconstructing lost or damaged areas.
Double-weight gold is somewhat thicker than standard-weight leaves, and
thus it is easier to handle. So-called “patent” or “transfer” leaf comes lightly
adhered to a paper backing and is often used when gilding outdoors or in
drafty areas that might blow away regular leaf, which is generally interleaved
between sheets of thin paper dusted with rouge. Gold leaf is sold in “books”
of 25 leaves 3¼ inches square; silver leaves are generally 3¾ inches square.
Patent leaf can also be bought in rolls of various widths for applying
decorative banding, such as is found on harpsichord cases. When using loose
leaves of gold, they are generally spread out on a gilder's cushion, which can
be made with a piece of ¾ inch thick wood or plywood cut to about 6 inches
by 12 inches (150 mm × 300 mm) (Figure 10). Lay several sheets of felt or
cotton batting over the top side and cover with white (alum-cured) sheepskin,
flesh side up, which is stretched taut and tacked along the edges. A folding
draft-screen made of parchment or heavy paper can be fitted to the back end
of the cushion to prevent leaves from blowing off, which is useful in drafty
environments. Individual leaves can then be picked up with a gilder's knife
and carefully deposited on the leather surface of the front section of the
cushion. You then gently blow straight down on the leaf to flatten it out
against the cushion. The gilder's knife, which has a straight blade about 6
inches (150 mm) or 8 inches (200 mm) in length and a polished, nick-free
cutting edge, is used to cut pieces from the leaf to fit areas to be gilded. The
gilder's tip, a special brush made with long squirrel hairs projecting from a
folded card, is wiped along your cheek to pick up a little facial oil, which
helps the gold leaf adhere to it when it is transferred from the gilder's cushion
to the object being gilded. If your skin is insufficiently oily, rub some skin
cream or a little petroleum jelly on the back of your hand to rub the gilder's
tip across. Just before the leaf is laid down, the bole is brushed with water to
which a little ethanol has been added to reduce surface tension; once this is
done, the gold leaf is applied with a gilder's tip. As soon as the gold leaf
touches the wet surface of the bole, it immediately adheres. If there are any
“holidays” (i.e. areas not covered by leaf), the bole can be remoistened and
“faulted” with “skewings” (small fragments of gold). These skewings are best
handled with gilder's squirrelhair mops, which come in a variety of sizes.
Mops are also useful in tamping down the gold leaf against the wet bole.
Because they have a tendency to pick up moisture from the wet bole, it is
useful to have a number of dry ones on hand. Wads of cotton wool can also
be used to tamp down the gold leaf.
After the gilded areas have dried, areas of gold leaf can be burnished with
agate burnishers, which produce a lustrous surface that resembles polished
metal. Burnishing is sometimes restricted to highlights that contrast with
nonburnished areas. Water-gilded surfaces are fragile and can be damaged by
moisture, so after burnishing they should be given a thin coat of shellac or
synthetic varnish such as Soluvar or Paraloid B-72. Crevices, hollows, and
embossed areas that tend to collect grime can be subtly “antiqued” with
pigments mixed with thinned Soluvar, Paraloid B-72 varnish, or shellac. Be
aware that watercolor, egg tempera, and acrylic emulsion paints (which
contain water) may disturb water gilding. To further integrate newly gilded
areas, places that might have been subjected to gradual wear can be lightly
rubbed with a piece of coarse wool cloth to wear through the gilding and
reveal a bit of the underlying bole.

Oil gilding

This method, also called mordant gilding, is often used when gold leaf is
applied directly over painted surfaces, such as the gold banding on painted
harpsichord cases. It is also used directly on wood, metal, and glass. Briefly,
an oil varnish, or size, is applied in areas to be gilded, and when the surface
of the varnish becomes tacky, the gold leaf is applied to it. Japan gold size
dries very quickly and is convenient for gilding small areas. Commercially
gilding varnishes are available that have specified drying rates (such as 3 and
12 hours), which are useful when large areas of gold must be applied. Oil
gilding cannot be burnished.

Mercury gilding

Also known as fire-gilding, mercury gilding was often used in gilding brass
or bronze mounts, such as those found on early nineteenth-century Viennese
fortepianos, and sixteenth- through eighteenth-century silver-gilt trumpets
and sackbutts. This process, known since ancient times, involves combining
mercury and gold to form an amalgam. To form this amalgam, the gold does
not have to be fused, but only raised to red heat, though the mercury must be
close to its boiling point, which is about 600°F (316°C). At this and lesser
temperatures, mercury has an affinity for gold and will readily absorb it.
Traditionally, a crucible was lined with a slurry of yellow ochre, allowed to
dry, and then heated over glowing embers (without the use of bellows,
according to the sixteenth-century sculptor and goldsmith Benvenuto Cellini).
Thin sheets of gold and mercury in a 1:6 to 1:8 proportion were stirred with
an iron rod until the gold and mercury amalgamated. After cooling, the
amalgam was wrapped in a piece of leather and squeezed to remove excess
mercury. The object to be gilded had to be absolutely clean of oxidation and
oil or grease, and could be scratch-brushed or etched with nitric acid just
prior to gilding, not only to expose a fresh surface but also to provide tooth
for the amalgam, which was applied with a stiff brush. A ground made by
dissolving 1 g of mercury in 3.5 ml nitric acid was sometimes painted on just
prior to gilding. After the amalgam was brushed on, the object was placed in
glowing embers to drive off the mercury, while the gold adhered to the brass
or bronze object. The amalgam could be reapplied and refired to achieve a
thicker and more durable layer of gold. Another technique involved painting
the surface of the object with mercury (called “quicking”), laying gold leaf on
top, and then subjecting the object to heat. After firing, mercury-gilded
surfaces were matte, but they could be scratch-brushed or burnished with
polished agate to create a brilliant effect. As with water gilding, burnishing
was often used to create highlights that contrasted with matte areas. Mercury-
gilded surfaces could also be patinated, or colored, with solutions of
ammonium chloride, potassium nitrate, and cupric acetate.
Due to the obvious dangers of working with mercury and its vapors,
mercury gilding has been largely supplanted by electroplating, a process that
came into wide use around 1860 (see Electroplating and electrocleaning). It
may be difficult to discern whether an object has been mercury-gilded or
electroplated; however, unlike most electroplated objects, the undersides of
mercury-gilded objects were often left ungilded, and one may see brush
marks or drips where the mercury amalgam inadvertently ran. If the leafing
technique was used, one may see overlapping lines left by the square leaves
of gold. Electroplating is sometimes preceded by the application of mercury,
so its discovery by X-ray fluorescence, for example, does not necessarily
indicate that the object was mercury-gilded.

Gilding on leather

Gilded designs and lettering on leather are done with embossing tools,
patterned rollers, and metal type. Glair (slightly whisked egg white) is first
applied to the leather and allowed to dry thoroughly. The leaf is then laid
down on the leather, and the embossing tools, rollers, or type that have been
heated up to about 240°F (115°C) are pressed against the leaf for a few
moments, which softens the glair sufficiently for the leaf to adhere. After
cooling, the excess leaf is wiped off with a soft cloth. The gilded areas can be
given a thin coat of shellac or synthetic varnish (such as Soluvar retouching
varnish or Paraloid B-72, 10–15% in xylene) to protect it. If special
ornaments need to be matched, embossing tools can be fabricated out of
square or round brass stock using gravers and punches. Bookbinding and
leather-working suppliers sell a wide variety of shapes that may suffice.

Gilding on paper and parchment

The areas to be gilded are painted with bole that has been tempered with
rabbitskin glue (see above). After drying, breathe on the bole to activate the
glue and lay down the gold leaf. After it has set, it can be burnished with an
agate burnisher. Glair, a weak solution of gum arabic, or garlic juice can also
be used as mordants to attach gold leaf directly to paper or parchment. These
liquids are applied with a quill pen or brush, and when almost dry, the gold
leaf is laid down and rubbed with a burnisher until it adheres. Another
technique is to write or paint directly with shell gold, which is pulverized
gold leaf mixed with gum arabic. Shell gold (as well as shell silver) can be
made by grinding gold or silver leaf together with gum arabic using a pestle
and mortar. In earlier times, the resulting paste was deposited in a seashell
(hence the name “shell gold”), where it could be picked up with a quill or
brush. Shell gold and silver can be purchased today in little discs or tablets
glued onto a small plastic dish. Shell gold and silver cannot be burnished.

Cleaning and retouching gilding

Do not use water to clean water-gilded areas, as this will remove the
gilding. Mineral spirits, or a mixture of mineral spirits with increasing
proportions of ethanol, acetone, or ethyl acetate, may be more effective, but
any combination of solvents must be tested in an inconspicuous area.
Small unburnished areas can be retouched with shell gold or gold powder
applied with Soluvar retouching varnish or Paraloid B-72 (10–15% in
xylene). If gesso and bole have been lost, they must be built up as described
above. Patination can be applied with watercolors, artist's quality acrylic
paints, or pigments applied in egg tempera, Soluvar retrouching varnish, or
Paraloid B-72.
On old, gilded work, some show-through of bole and gesso is acceptable
and even aesthetically desirable, particularly in areas that might have been
subjected to wear in the course of use or handling. However, fresh chips and
areas that look overtly damaged should be reconstructed, gilded, and
patinated to blend in with undamaged areas. (See sections on Mold making
and Compo, or pastiglia for techniques of reconstructing carved and gesso
ornaments.)

References
The Art and Science of Gilding: A Handbook of Information for the Picture
Framer (Rochester, 1909).
The Carver and Gilder (London, 1864).
Benvenuto Cellini, The Treatises of Benvenuto Cellini on Goldsmithing and
Sculpture, trans. C. R. Ashbee (New York, 1967).
Cennino D’Andrea Cennini, The Craftsman's Handbook: The Italian “Il
Libro Dell’ Arte”, trans. Daniel V. Thompson Jr. (New York, 1954).
The Gilder's Manual (New York, 1876).
Peter and Ann Mactaggart, Practical Gilding (Welwyn, 1984).

Glues, pastes, and other adhesives

In most traditional musical instrument making and repair, fish, animal-

hide, and bone glues have been the principal adhesives for use with wood,
leather, paper, and textiles. They not only have sufficient strength for general
woodwork but also suffice for the most demanding applications, such as the
gluing of lute and guitar bridges and keyboard wrestplanks, which require
great strength in order to bear the constant pull of the strings. Many modern
synthetic adhesives, such as those made of polyvinyl acetate (“white glue”),
aliphatic resin (yellow carpenter's glue), and acrylic resin (Paraloid), tend to
remain flexible after setting, which permits parts to creep under tension. Fish
and animal-hide glues, on the other hand, hold fast. By diluting fish and
animal-hide glues with water, one can easily adjust their strength to suit
different requirements. For example, violin bellies can be glued onto the ribs
with diluted hide glue so that this joint will release during extreme changes in
humidity, thus preventing the formation of cracks. The diluted glue also
makes it easy to pry bellies off when internal repairs must be attended to.
Fish and animal-hide glues are also eternally reversible with water, which
allows instruments to be easily disassembled and their joints to be cleaned
prior to reassembly. (See below for mixing instructions.)

Animal-hide glue

The hides, horns, and bones of cattle, as well as the skins and parchment
derived from the skins of sheep, goats, and pigs, rabbits, and the swim-
bladders, or “sounds,” of certain fish are used to make animal glue. The
process of manufacturing glue from animal hides begins very much like the
process of making leather, in that the fresh hides are first “limed,” or steeped
in solutions of lime (calcium oxide) or caustic soda to remove the hair. In
making hide glue, the liming is continued somewhat further in order to
remove other unwanted material. After liming, the hides are neutralized with
acid or simply washed. Further treatment of the hides with alkali and acid
removes some of the grease, though boiling drives off the rest, which is
skimmed off. The resulting liquor may be bleached and preservatives added
before it is poured into pans or molds and cooled. The gelled glue is often cut
into sheets, thoroughly dried on metal netting, and then broken into flakes or
ground up.
Animal glue consists primarily of gelatin, though another constituent,
chondrin, has greater adhesive power. In the manufacture of gelatin, chondrin
and other materials are removed, yielding a pale jelly. While gelatin is
sometimes considered a “purer” substance than glue, it lacks adhesive
strength and is often used as a size or as an adhesive when great strength is
not required or desirable.
Bone glue is prepared in a similar fashion to hide glue, though the
considerable amount of grease contained in bones may require steaming or
treatment with petroleum solvents to remove it. The resulting liquor is
generally more dilute than that derived from hides and may not gel upon
cooling. Continued boiling to concentrate the stock would weaken the
resultant jelly, so excess water is extracted in a vacuum evaporator. Glue
made from bones is typically less flexible than that made from hides; for this
reason, it is sometimes preferred for gluing mortises and tenons, where
rigidity is desirable. The two kinds can be mixed to produce a glue having
intermediate properties. French woodworkers employ three types of wood
glue: colles de peaux (hide glues, considered the strongest and most
expensive), colles de nerfs (nerve glues, considered of good quality), and
colles d'os (bone glues, generally thought to be of lesser quality but
recognized for their rigidity). For veneering, hide and bone glues are
sometimes used in a proportion of 1:4; for general joinery, a proportion of 1:2
is advocated.
Commercial animal-hide and fish glues that are liquid at room temperature
have been chemically treated (often by the addition of urea) to prevent them
from gelling. They are generally not as strong as the glues that must be
heated, and in some cases may revert to a gummy state in high humidity.
Glues are graded according to their viscosity under standardized test
conditions, and various grading and numbering systems are employed (for
example, those of Peter Cooper and the National Association of Glue
Manufacturers). The Peter Cooper grades (which run from A extra, 1 Extra,
No. 1, 1X, 1¼, 1⅜, etc. down to No. 2) are determined by dissolving 25
grams of glue in 100 ml of water, heating to 180°F and allowing the liquid to
pass through a viscosimeter that has been adjusted so that 50 ml of water
passes through in 15 seconds. The various Peter Cooper grades pass through
the viscosimeter in the following times (in seconds):

A Extra 45

1 Extra 40

No. 1 35

1X opaque 32

1X clear 29

1¼ 27
 1⅜ 25

1½ 23

1⅝ 21

1¾ 19½

1⅞ 18

No. 2 16½

Another system for grading animal-hide glue is gram-strength. 150–200

gram-strength glue is comparatively weak, but appropriate for veneering. It
has the longest setting time. 250 gram-strength glue is stronger and adequate
for general cabinetmaking. 300–350 gram-strength glue is for high-stress
applications (such as gluing wrestplanks), and has the shortest setting time.
As is evident in the above table, the better grades of glue are more viscous
than the lower grades; therefore, they can tolerate greater dilution than the
lesser grades to achieve a given viscosity. “A Extra” grade glue at a dilution
of 1:2¾ has a gram strength of 379, whereas 1½ grade glue at a dilution of
1:1¼ has a gram strength of 135. When purchasing glue from suppliers, the
grade is rarely specified, so it is impossible to recommend standard dilution
for hide glues.
In preparing dry animal-hide, bone, or fish glue for use, it must first be
soaked and then heated. If the glue is finely granulated, the soaking process
can take less than an hour; if the glue is in sheet or block form, it must be
broken into small pieces or crushed, but if the pieces are relatively thick (such
as the leathery sheets of isinglass that are currently on the market) they must
be soaked overnight or until they have completely gelled before they are
heated. Heating should be carried out in a temperature-controlled glue pot or
hotplate with the temperature adjusted to about 140°F (60°C). Glue should
not be allowed to boil as this reduces its strength, as will prolonged heating at
the recommended temperature. Animal-hide, bone, and fish glue are
preferably prepared just prior to use. A general rule to follow when mixing
up a small quantity of glue (an ounce or two) for a restoration project is to
place the granules of glue in a small glass jar and add enough water to
moisten it. After the glue has swelled, add water to a depth of about half an
inch above the surface of the glue, and then heat in a glue pot. For most
gluing operations, adjust the viscosity of the glue so that when a glue brush is
dipped into it and then lifted about a foot above the glue pot, the glue forms
an unbroken stream. If the stream breaks, add a little more water. If the glue
seems too watery, continue heating the glue to drive off some of the water.
Once the correct viscosity has been achieved, keep in mind that water will
evaporate rather quickly from glue while it is being heated, thereby
increasing its viscosity, so it is advisable to keep the jar covered with a lid or
perhaps with a piece of aluminum foil that has a hole poked in it for the glue
brush. Some believe that it is best to let the glue gel after its first heating and
then to reheat it for the first and subsequent gluing jobs. Glue can be kept in a
refrigerator for months, but it should well capped to prevent the evaporation
of water. If glue is kept at room temperature for an extended period (several
days) during the summer months, a crystal of Thymol can be added to the
glue to prevent bacterial decomposition.
Isinglass is a strong glue made from the swim-bladders of sturgeon. The
best grades tend to be pale, which makes them useful for gluing ivory and
other light-colored materials. Unfortunately, the tough, leathery pieces that
are available from art material suppliers are very time-consuming to soak,
sometimes requiring more than an overnight soaking (while soaking,
isinglass should be kept in a refrigerator to prevent it from rotting during this
extended process). To render this form of isinglass more convenient to use,
after this preliminary soaking and heating it can be poured out onto a clean
Teflon-coated baking pan. When this thin sheet has hardened, it can be
crushed into fine flakes that will quickly reliquify with the addition of a small
amount of water and the application of heat. Isinglass is also available in
shredded form from firms that supply materials for beer and wine making (it
is used in the clarification process).
Parchment glue is made by soaking parchment scraps in water and then
heating. This makes a light-colored glue that is often used as a size and in
gilding. Parchment scraps can be purchased from gilding suppliers such as
Sepp Leaf Products.
As indicated above with regard to animal-hide and bone glues, different
types of glue can be mixed in order to adjust or combine their properties, and
some glues that have been traditionally used for certain types of work have
been found to have properties that make them unexpectedly suited for other
types of gluing operations. For example, rabbitskin glue, which has
traditionally been used in diluted form as a size and in the preparation of bole
for water gilding, and thus does not have a reputation for being particularly
strong, has been found to be a very strong glue at higher viscosities.
Furthermore, it is extremely flexible, which makes it excellent for certain
applications.
Some years ago the author tested about a dozen fish, animal-hide, and bone
glues that he had on hand. Salianski isinglass from Kremer Pigmente was the
strongest, followed closely by fish glue acquired in sheet form from Zecchi in
Florence. In third place was a rabbitskin glue from Kremer, and tied for
fourth place were a clear, light-colored, filtered animal-hide glue purchased
from a violin restorer in New York and a French bone glue. The weakest
glues were a rabbitskin glue from another source and a French hide-and-bone
glue mixture. In conducting this test, the glues were diluted with water in a
1:3 proportion, weight to volume, and strips of maple simulating lute bridges
were glued to spruce soundboard stock. A shearing force was gradually
applied and the breaking tension was measured with a spring scale.
When using glues that must be heated, it is important to keep in mind that
such glues chill very quickly upon application to wood surfaces that are at
room temperature. Working time is generally around 1–2 minutes, so it is
necessary to have one's clamps or go-bars (see Woodworking) ready to go
before the glue is applied. Heat lamps may be of some benefit when working
with large surfaces or on cooler days. In early times, some Italian harpsichord
makers set fire to a small pile of wood shavings inside the cases of their
instruments just before gluing in their soundboards. They may have done this
to warm the soundboard so that the glue would not chill as the soundboard
was set in place. Work glued with animal-hide, bone, and fish glues should
generally be kept under clamping pressure overnight.

Wheat paste

Stir 1 part wheat starch (such as Aytex-P) into 4 parts deionized or distilled
water.
 Let stand for at least 1 hour.
Cook over a medium-high heat until thick and translucent for 15–25
minutes.
Remove from the heat and allow to cool.
Before use, strain well (in Japan, this was traditionally done with a special
horsehair paste strainer) and dilute with water until a creamy consistency is
reached.
Undiluted, cooked paste may be stored in the refrigerator for several
weeks.

Rice paste

Stir 1 part rice starch into 6 parts of deionized or distilled water.

Let stand for at least 1 hour.
Cook over a medium-high heat until thick and translucent (20–40 minutes).
Remove from the heat and allow to cool.
Before use, strain well and dilute with water until a creamy consistency is
reached.
Undiluted, cooked paste may be stored in the refrigerator for several
weeks.

Casein glue

Casein is a protein that is extracted from milk, either through fermentation

or the action of acids and alkalis. It must be combined with an alkali (such as
lime or borax) to render it soluble. Some brands of commercial dry casein
contain an alkali so that the product can be readily mixed with water; others
require the addition of an alkali.
Other adhesives used in fabrication and conservation work include:
Acryloid–see Paraloid below.
Araldite 2020, Hxtal NYL1, West System, and Milliput epoxy resins are
two-part adhesives available in liquids of various viscosities as well as in
putty form. Epoxies are not easily reversed, though acetone and ethanol can
be used to remove the uncured resin and hardener before they cure.
AYAA, AYAB, AYAC, AYAF, and AYAT are polyvinyl acetate resins
available in a range of molecular weights, viscosities, and glass transition
temperatures.
BEVA 371 is an aqueous dispersion of acrylic and ethyl vinyl acetate
resins, often used in relining paintings.
Cellofas B3500, Hercules 7LC (sodium carboxymethylcellulose) is water
soluble and often used for gluing paper.
DAP Weldwood resorcinol glue is highly waterproof and used in boat
building.
Duco Cement, HMG is cellulose nitrate dissolved in various solvents. It
yellows and has been largely supplanted by B-72 and other synthetic resin
glues.
Elmer's Glue, Elvace 45675, Jade 403 and 711 are aqueous dispersions of
polyvinyl acetate that are not reversible with water. These are widely used for
gluing paper and general woodwork. Some types, such as Lineco, are
remoistenable and thus reversible.
Ethulose, or ethyl hydroxyethyl cellulose, is soluble in water or alcohol.
4% in water produces a thick solution, while a 8–10% solution has a gel-like
consistency.
Evasol is an ethylene vinyl acetate co-polymer having a neutral pH that is
widely used to bond paper and textiles.
Gorilla Glue is made with polyurethane resin. It expands to fill voids while
curing.
Hot-melt glues (polyamide and hydrocarbon resins) set upon cooling but
generally have poor bonding strength.
Klucel G (hydroxypropylcellulose) is nonionic and soluble in alcohols,
cold water, and aromatic hydrocarbons. It is widely used as a leather
consolidant.
Lascaux 360 HV is a butyl-methacrylate copolymer in an aqueous
dispersion that is thickened with acrylic butyl ester. It may be thinned with
water but is insoluble in water when dry. This is a tacky contact adhesive
used to bond paper, textiles, and wood.
Methocel A4C (methyl cellulose) is water soluble and widely used for
gluing paper.
Mowilith 20, 50 (polyvinyl acetate) is used as a consolidant and retouching
medium.
Mowilith DMC2 is a copolymer of vinyl acetate and maleic acid di-n-butyl
ester. It is available in an aqueous dispersion as a binder for paints and
adhesives for paper and textiles.
Paraloid B-72 (ethyl methacrylate copolymer), B-67 (isobutyl methacrylate
polymer), B-48N (methyl methacrylate copolymer), and B-44 (methyl
methacrylate copolymer). These resins are soluble in xylene, toluene, and
high-aromatic solvents. They are widely used in protective coatings, though
they can also be used as adhesives.
Plextol D360, D498, B500 are butyl acrylate and methacrylate copolymers
used as adhesives in relining paintings on canvas. They are available in
aqueous dispersions.
Pyroxylin glue (nitrocellulose) is widely used for household mending.
Regalrez is a hydrocarbon resin that can be used as a hot-melt adhesive or
dissolved in mineral spirits and used as a coating; it is said to be permanently
reversible in mineral spirits.
Rhoplex N-580 and MC-76 are acrylic resins formulated as aqueous
emulsions, but they are not reversible with water after setting. Used to bond
paper and fabric and as a contact adhesive.
Seccotine is a refined fish glue that is liquid at room temperature.
Commonly used to glue paper and cardboard.
Starch pastes (see preparation instructions above) are used for gluing
paper, paper to matte board, or wood.
Super Glue is a cyanoacrylate resin that bonds well to skin.
Titebond, or yellow carpenter's glue, is an aliphatic resin in an aqueous
emulsion. It is not reversible with water after setting.
 Tylose MH300 (methyl 2-hydroxyethyl cellulose) is water soluble and
widely used for gluing paper.
Urea resin glue (DAP Plastic Resin Glue). Resistant to many solvents.

References
R. Livingston Fernbach, Glues and Gelatin, a Practical Treatise on the
Methods of Testing and Use (New York, 1907).
C. V. Horie, Materials for Conservation (London, 1987).
Allen Rogers, Industrial Chemistry: A Manual for the Student and
Manufacturer (New York, 1931).
Ernest Spon, Workshop Receipts for Manufacturers and Scientific Amateurs
(London, 1909).

Gold and gold-plated brass and bronze

Corrosion products on gold-plated brass or bronze can be removed by

immersion in a 5–10% solution of trisodium EDTA or in a solution of 150 g
Rochelle salt (sodium potassium tartrate) and 50 g sodium hydroxide in 1
liter of distilled or deionized water.

Reference
P. Fiorentino, M. Marabelli, M. Matteini, and A. Moles, “The Condition of
the ‘Door of Paradise’ by L. Ghiberti: Tests and Proposals for Cleaning,”
Studies in Conservation 27/4 (November, 1982), pp. 145–153.

Grain painting and marbling

Faux finishes have been used to decorate the cases of organs, harpsichords,
and other keyboard instruments since the sixteenth century (notably those
made by Flemish and “Antwerp” school makers). The elaborately carved legs
of many of the rosewood-veneered square and grand pianos made in the mid
to late nineteenth century were often not made of solid rosewood but were
grain-painted over cheap, light-colored woods such as poplar. Even smaller
instruments, such as lyre guitars and harps, often had grain-painted bodies.
In the nineteenth century, grain painting was a specialized craft that was
extensively employed in decorating interior woodwork, such as doors,
columns, wainscoting, and moldings, as well as furniture. Today it is largely
the domain of hobbyists, and perhaps because of its popularity, the
specialized tools that were once the province of this professional craft are
again widely available. These tools include steel and rubber graining combs,
graining rollers, and special brushes, such as hog-bristle mottlers and badger
blenders.
In the nineteenth century, professional grainers generally did their interior
wood-grain painting in distemper, or watercolor medium. The commonly
used pigments, such as raw and burnt umber, raw and burnt sienna, the
ochres, Venetian red, Vandyke brown, drop black (i.e. bone or “ivory”
black), and lead white, were available pre-mixed in a watercolor medium,
generally a solution of gum arabic. Some grainers ground dry pigments in a
mixture of water and ale or beer. Generally, a ground color was laid down
over a well-smoothed surface that was sometimes sealed with a thin coat or
two of shellac to prevent the resins in knots from breaking through the
painted surface. Graining was done over the ground tone, and the best
grainers were extremely deft at imitating the year-rings, rays, birds’ eyes, and
other characteristics of virtually every type of cabinet-grade wood. Marbling
was often done in oil colors, which was thought to provide a greater sense of
depth and transparency required of marble and other stones, such as porphyry
and granite.
Restorers must often retouch these faux finishes, or employ these
techniques in general retouching. For example, it is often necessary to
integrate new edgework in violins or to integrate fills in wood that have been
made with putties or other materials. For grain painting to be convincing, the
transitions from early to late year-ring growth, as well as from sapwood and
heartwood, must be imitated. Some of the techniques used in marbleizing
paper are also valuable in general restoration work.
For imitating mahogany, Ernest Spon, Workshop Receipts for
Manufacturers and Scientific Amateurs (London, 1909) provides the
following instructions:
 Vandyke brown and a little crimson lake ground in ale laid on, allowed
to dry and then smoothed, form the ground. Then lay on a second thicker
coat, soften with a badger-hair brush, take out the lights while it is wet,
and imitate the feathery appearance of mahogany heart. Soften, and top
grain with Vandyke brown laid on with an overgraining brush of flat
hog-hair combed into detached tufts. In softening, be careful not to
disturb the under color.

Ashmun Kelly's The Standard Grainer Stainer and Marbler (Philadelphia,

1923) gives instructions for imitating oak. After setting in the ground using
umber (either raw or burnt), he advises:

Lay the color off in the direction of the grain. To remove any
objectionable brush marks use the badger blender. Stipple and blend
both with this brush. It will be observed that a panel of oak will have
coarse lines on one side, and that these gradually become finer as they
approach the opposite side. To make these growths use a 3-inch leather
coarse comb to begin with, and follow with steel combs of requisite
fineness to about the middle of the panel, or a little beyond, according to
the character of the wood sample you are imitating. The side opposite to
the coarse grain may be done with the rubbing-in brush, in some cases,
by drawing the side of the brush downward. Over these lines or grains
run a steel comb in a wavy manner, to produce the pores; use combs of
sizes corresponding to the grain … On top of this combing the figures
are made. With a dry rag held tight over the right thumb-nail and with
the loose end thereof held in the left hand proceed to wipe out the heart
growth; some call it the sap … This graining rag is used in several folds,
and sometimes, in place of the thumb, a piece of bone is used, it being
flat and about the width of the thumb-nail, which it is to replace, the left
hand keeping the loose end of the rag away from the work … Note,
when using the graining combs frequently wipe them off with a dry
clean rag. Another effect often made with the combs is had by placing a
piece of cheese-cloth over them, changing the cloth by drawing it with
the left hand as you work; this gives a softer grain effect than where the
plain combs are used. It is particularly good next to a wiped-out part,
where the clear-cut comb marks would be too harsh. It produces a more
 woody appearance … After doing the wiping-out work it only remains
to do the overgraining. Overshading is also a feature of this part of the
work. The purpose of both is to furnish a natural variety of shade and
light, and should not be overdone. The color for the purpose must be
mixed specially, using distemper color, with burnt umber as the
pigment. Lines are put in with the bristle liner and then blended with the
badger brush. This work requires care, as it is liable to soften up the
undercoating of stippling, which is usually done. I have usually laid a
glaze of the graining color over the work when done and dry, for this
takes off the raw mechanical look of the job, and also makes it more
uniform of texture and color. In running the overgrainer, the hairs will
naturally separate, and, if [they do] not, then separate them with a coarse
comb, or with the finger of the hand that handles the brush. Fill the
brush with a weak color and draw it the way of the natural grain of the
wood. The overgraining should be lighter where the figures have been
made. Then, before the work is dry, draw the overgrainer lightly and
with a wavy motion across the grain … Vandyke brown is a useful color
for glazing with, owing to its richness of tone and its transparency, but
for a warm effect it is toned with burnt sienna, while for a cool effect a
little blue-black is added.

This gives some sense of the complexity of the wood-graining process. Kelly
also gives instructions for marbling. Here are his recommendations for
imitating porphyry:

Make the ground with purple, brown, and rose pink. The graining colors
are vermilion and white lead, ground separately in turpentine, with a
little gold size as binder; more turpentine must be added before the color
is applied. When the ground is dry, fill a large brush with the vermilion,
and remove nearly all of it by scraping it off with a palette knife over the
edge of a paint pot. Then, holding a short iron rod, or a piece of broom-
handle firmly in the left hand, strike the brush smartly upon it, which
will cause a shower of particles of color. These spots must appear very
fine on the surface of the work. Now repeat the operation with the red,
then lighten the color a little with white lead, and sprinkle again. Finally,
give it a shower of white lead spots. When the work is dry, place a few
 white veins across it. Some put in the fine spots with a graining disc
wheel, and such work has greater regularity of form than that done by
sprinkling. Any parts that are not to be spotted will have to be protected
by paper. This marble may also be done in water color, with a coat of
varnish for protection; it is very good for interior work. Some varieties
of this marble have a narrow opaque white vein running among the
spots. It cannot be put in until the spots are dry. These veins are made
with a sable pencil [fine brush], while the threads are drawn out
afterward with the feather.

For imitating walnut, Kelly suggests that the heartwood can be formed with
crayons, which are also recommended for imitating certain features in oak
and other woods as well. These crayons were not the wax-based type in
general use today but were made in the following manner:

To make graining and marbling crayons take pipe clay and the color and
mix both together; then have ready some strong hot soapsuds, and mix
with the clay and color until like putty in consistency; roll it out on a
board to a thickness of about ¼ inch; with a broad-bladed knife cut it
into strips of about ¼ inch, making the crayon a little less than ¼ inch
square when dry.

To prevent distemper (water-based) media from drying out too quickly, and
to make the graining thicker so that it stands in relief, Kelly recommends use
of a preparation termed megilp. He states that megilp for water media can be
made by taking equal parts of soap and wax, the later being melted into the
hot soapy water, and then adding this liquid to the graining color in very
small amounts. Megilp for oil media was made in other ways, though very
often by mixing equal quantities of a drying oil (such as linseed oil to which
sugar of lead, or lead acetate, was added as a drier) with mastic varnish
(generally mastic resin dissolved in turpentine). Other proportions of drying
oil to mastic varnish were also used, but the net result was a medium having
the consistency of jelly or butter.
While professional grainers of the nineteenth and early twentieth centuries
were adept at imitating woods and stone with paint, the early grain painting
and marbling found on musical instruments (in particular the outer case
painting of Ruckers harpsichords, which were often painted in imitation of
porphyry) is sometimes extremely crude – almost a caricature of this art.
Though the Ruckers’ case painters typically got the basic ground color right,
the veining and other inclusions were often perfunctorily dashed in with a
few brush strokes. In later years, most of these original painted surfaces were
deemed expendable and were overpainted.

References
Phileas Boeck, Die Marmorirkunst (Vienna, Pest, and Leipzig, 1896).
Leslie Carlyle, The Artist's Assistant (London, 2001).
Ashmun Kelly, The Standard Grainer Stainer and Marbler (Philidelphia,
1923).
Spon's Workshop Reciepts for Manufacturers and Scientific Amateurs
(London, 1909).
J. Stokes, The Cabinet-Maker and Upholsterer's Companion (Philadelphia,
1889).

Grit size comparison chart

US graded (CAMI) European graded (FEPA)

Average Old Average

Microns diameter Mesh sandpaper Mesh diameter Microns
(inches) designation (inches)

0.3 0.0000118

0.5 0.0000197

1.0 0.0000394

2.0 0.0000787
3.0 0.000118

4.0 0.000158

5.0 0.000197

6.0 0.000236

6.5 0.00026 1200

9.0 0.00035

9.2 0.00036 1000

12.0 0.00047

12.2 0.00048 800

15.0 0.00059

16.0 0.00062 600 P1200 0.00060 15.3

19.7 0.00077 500 P1000 0.00071 18.3

20.0 0.00079

23.6 0.00092 400 10/0 P800 0.00085 21.8

Extra fine

25.0 0.00098

28.8 0.00112 360 P600 0.00100 25.8

30.0 0.00118

P500 0.00118 30.2

36.0 0.00140 320 9/0 P400 0.00137 35.0

 Extra fine

40.0 0.001575

P360 0.00158 40.5

44.0 0.00172 280 8/0

Very fine

45.0 0.00177 P320 0.00180 46.2

50.0 0.00197

53.5 0.00209 240 7/0 P280 0.00204 52.5

Very fine

55.0 0.00217

66.0 0.00257 220 6/0 P220 0.00254 65.0

Very fine

78.0 0.00304 180 5/0 P180 0.00304 78.0

Fine

93.0 0.00363 150 4/0 P150 0.00378 97.0

Fine

116.0 0.00452 120 3/0 P120 0.00495 127.0

Fine

141.0 0.0055 100 2/0 P100 0.00608 156.0

Medium

192.0 0.00749 80 1/0 P80 0.00768 197.0

Medium

268.0 0.01045 60 1/2 P60 0.01014 260.0

Medium
 351.0 0.0139 50 1 P50 0.01271 326.0
Coarse

428.0 0.0169 40 1½ P40 0.01601 412.0

Coarse

535.0 0.02087 36 2 P36 0.02044 524.0

Coarse

638.0 0.02488 30 2½ Coarse P30 0.02426 622.0

715.0 0.02789 24 3 P24 0.02886 740.0

Very coarse

905.0 0.03535 20 3½ P20 0.03838 984.0

Very coarse

1320.0 0.05148 16 4 P16 0.05164 1324.0

Very coarse

1842.0 0.07174 12 4½ P12 0.06880 1764.0

Very coarse

Emery paper is mentioned in the Philosophical Transactions LXII

(London, 1772). The Cabinet Maker’s Guide (anon: Concord, NH, 1827)
provides instructions for making glass-paper that involve pounding glass in a
mortar, passing the ground glass through sieves, and then gluing the glass
particles to heavy paper with animal-hide glue.

Guitar

(see also Gut strings, Mersenne's Law)

Stringing
 Below are Antonio Stradivari's string recommendations written on his
guitar fingerboard pattern, Museo Stradivariano no. 375:

1st course, 2 cantini of the guitar [approx. 0.48 mm]

2nd course, 2 sotanelle of the guitar [approx. 0.58 mm]
3rd course, 2 cantini of the violin [E string, approx. 0.67 mm]
4th course, 1 canto of the violin [A string, approx. 0.90 mm]
4th course octave, 1 sotanella of the guitar [approx. 0.58 mm]
5th course, 1 canto of the violin, but a little heavier [A string, approx.
1.10 mm]
5th course octave, 1 cantino of the violin [E string; approx. 0.67 mm]

Classical-period guitar strings can be strung at around 7 kg tension. Evidence

for this is provided by old or original strings preserved with a guitar given by
Percy Bysshe Shelley to Jane Williams in 1822 (guitar and strings are in the
collection of the Bodleian Library, Oxford) as well as guitar strings ordered
from the Schott firm by Niccòlo Paganini (see Segerman, 1992, 2014).
Northern Renaissance Instruments recommends the following gut string
dimensions based upon this evidence:

String Pitch Diameter in mm

1st e1 0.64

2nd b 0.81

3rd g 1.02

4th d 1.36

5th A 1.82

6th E 2.42
 Below are classical guitar gut string gauges recommended by the guitarist,
composer, and musicologist Emilio Pujol in his Escuela razonada de la
guitarra (1934). Pujol is known to have played a guitar made by Antonio de
Torres.

String Gauge no.** Diameter in mm

1st 12.5–13.5 0.63–0.68

2nd 16–17.5 0.80–0.88

3rd 20–21.5 1.00–1.08

4th* 15–16 0.75–0.80

5th* 18.5–19.5 0.93–0.98

6th* 23–24 1.15–1.20

* Overspun, probably copper over silk

** Probably similar to the Pirastro gauge system, where each gauge number equals 0.05 mm.

Figures 11a, 11b, and 12 illustrate a few guitar repairs made by the author.
Figure 11a Clamping a loose bar in a flamenco guitar by Miguel Angel
Senovilla, Madrid, 1998.

Figure 11b Clamping a separation between the back and sides with spool
clamps.
Figure 12 Copies of tuning-machine parts for a triple-necked guitar by
André Augustin Chevrier, Mirecourt, c.1830. Bronze gears and tuners were
cast from silastic rubber molds. The steel screws were turned on a
watchmaker's lathe.

References
Stefano Grondona and Luca Waldner, La Chitarra di Liuteria (Sondrio,
2001), pp. 168–176.
Emilio Pujol, Escuela razonada de la guitarra (Buenos Aires, 1934).
Ephraim Segerman, “Shelley's Guitar and 19th Century Stringing Practices,”
FoMRHI Quarterly 67 (April, 1992), pp. 41–42.
Ephraim Segerman, Northern Renaissance Instruments website, February 16,
2014.

Gut strings
 (see also Guitar, Harp, Lute stringing, Mersenne's Law, Viola da
gamba, Violin setup and stringing)
Gut strings for musical instruments were, and continue to be made, by
twisting prepared small intestines of sheep (though cow and hog gut are also
used). The intestines are first emptied of their contents by soaking them in
water and running them between the fingers, though today pressurized water
is used to clean them out. The guts are then soaked in brine for eight to ten
days and afterwards in fresh water for three or four days. They are then
placed on a bench and scraped with either a wooden blade or the back of a
knife until the softer interior membranes become detached and are pressed
out. The gut is soaked again for seven or eight days in a weak alkaline
solution made with 2 ounces of pearl ash (potassium carbonate or potassium
hydroxide) per gallon of water, and then lightly spun using an ordinary
spinning wheel. The number of strands of gut that are twisted together to
form musical instrument strings varies with the intended diameter; three or
four strands are commonly used for the thinner violin strings. After wiping
them down, the strings are passed through progressively smaller holes in a
brass plate, which makes them almost perfectly round, reduces them to the
proper diameters, and also polishes them. Gut strings are sometimes bleached
with sulfur fumes and oiled or varnished to preserve and strengthen them.
According to Robert Dowland's Varietie of Lute Lessons (1610), quality lute
strings were round, smooth, and clear when held up to the light, with the
thinnest strings having a whitish-gray or ashen color.
Gut strings are twisted to balance strength with flexibility. A low twist
provides the greatest strength because the individual fibers are more closely
aligned with the direction of force applied to the string; high twist provides
greater flexibility but the tensile strength is reduced because the fibers are at a
greater angle to the direction of force. When making roped strings, several
individual strings are twisted together. If the individual strings have been
twisted in a clockwise fashion and are then twisted together in a
counterclockwise manner, the fibers will realign along the length of the string
and thus will provide optimal strength along with flexibility. Unfortunately,
such roped strings do not have a smooth surface, as Dowland indicates.
In his Musick's Monument (1676), Thomas Mace describes four different
types of strings used in stringing lutes (see Lute stringing): Minikins, Venice-
Catlins and Lyons (for basses), and Pistoy Basses (which he suspected were
nothing more than thick Venice-Catlins dyed dark red). For the first three
choirs of the lute, as well as the octave strings (especially the sixth-course
octave string), he recommends Minikins. For the fourth, fifth, and most of the
other octave strings he recommends Venice-Catlins. For the great bass
strings, he recommended Pistoys or Lyons. Regarding the thick Venice-
Catlins and Pistoys used for bass strings, he describes them as being well-
twisted and smooth, which would seem to rule out the roped construction
described above. As it is improbable that roped strings would have been used
for the fourth, fifth, and octave courses, it is therefore unlikely that Venice-
Catlins were of roped construction, as is currently thought.
The tensile strength of gut strings varies considerably, and their breaking
frequency is not independent of diameter (as it is in metal strings) because of
the different twist rates used in various gauges (see Scaling); however, the
following formula is often used to calculate the safe pitch limit for a gut
string of a given length:
F = 240,000/L

where F is the frequency in Hz, and L is the string length in millimeters.

Calculating gut string diameters for stringed instruments

If we turn to museums for examples of early or original strings that may

still be mounted on sixteenth- through eighteenth-century instruments, we
find that very few have been preserved. Mimmo Peruffo's excellent study
(1994) of the diameters of string holes in historic lute bridges provides some
indication of the largest diameters of strings that might have been fitted to
those instruments, though unfortunately not the actual gauges, which were
presumably somewhat narrower than the bored holes.
Though there is scant physical evidence of historical stringing practice,
there is documentary evidence found in such sources as Mersenne's
Harmonie universelle (1636) and Leopold Mozart's Versuch einer
gründlichen Violinschule (1756). Mozart states that the strings of the violin
should be at equal tension, though he does not recommend string gauges.
Mersenne indicates that the diameters of strings should be in the same
proportion as the ratios of their pitches, which is a prescription for equal
tension (see below). Furthermore, he provides the following examples: if the
largest, or the eleventh string of a theorbo or lute is one ligne (2.25 mm) in
diameter, then the seventh string, which is a 5th above, should then be 2/3
ligne (1.5 mm) in diameter. He then gives diameters for the fourth string,
which is a 12th higher (an octave and a 5th), 1/3 ligne (0.75 mm); and for the
2nd string, which is tuned a 17th higher (two octaves and a major 3rd), 1/5th
ligne (0.45 mm). To calculate the diameter of the top string, which is a 4th
above the second string, multiply its diameter by the ratio of a 4th, or 3:4
(0.34 mm). Though Peruffo lists a Matteo Sellas lute dated 1640 that has
bridge holes 2.20 mm in diameter, bridge hole diameters of 1.50–1.80 mm
are more common, and 0.43 mm was about the thinnest gut string available
through early Roman string makers, according to Peruffo. Thus, there may
have been some deviation from the calculated diameters.
What kind of tensions are provided when string length is kept constant and
the gauges increase according to the ratio of pitches? Though the answer may
be intuitively obvious to some readers, we can use Mersenne's Law to
calculate the relationship:
T = F2L2D2ρπ

where

T is the tension in newtons (the newton is a unit of force in the metric

system, whereas the gram and kilogram are units of mass; the unit
kilogram-force, or kgf, is equal to 9.81 newtons)
F is the frequency in Hz
L is the length of the string in meters
D is the diameter in meters
ρ is the density of gut, approximately 1,300 kg/m3.

Using this formula, if C is 110 Hz, L is 0.6 m, D is 0.00225 m, and ρ is 1,300

kg/m3, then T is 90.06 newtons, or 9.18 kgf. If e′ (a 17th above C) is 550 Hz,
L is 0.6, D is 0.00045 (1/5 of 2.25 mm), and ρ is 1300 kg/m3, then T is also
90.06 newtons or 9.18 kgf. Thus, Mersenne's instruction for increasing or
decreasing the diameter in proportion to the ratio of their pitches provides
equal tension among the strings, just as Leopold Mozart recommends.
Pirastro gut string gauge system

Since the nineteenth century, a gut string gauge system has been used to
denote different diameters. The Pirastro gauge numbers used today follow
this system. Each gauge number is equal to 1/20 mm, so to convert a gut
string gauge number to millimeters, multiply by 0.05. The manufacturer's
gauging of strings is often not precise, so it is good practice to verify the
actual string diameters with a micrometer.

References
Robert Dowland, Varietie of Lute Lessons (London, 1610).
Thomas Mace, Musick's Monument (London, 1676).
Marin Mersenne, Harmonie universelle (Paris, 1636).
Leopold Mozart, Versuch einer gründlichen Violinschule (Augsburg, 1756).
Mimmo Peruffo, “The Mystery of Gut Bass Strings in the 16th and 17th
Centuries: The Role of Loaded Weighted Gut,” Lute Society of America
Quarterly vol. 29, no. 2 (May, 1994), pp. 5–14.
Ernest Spon, Workshop Receipts for Manufacturers and Scientific Amateurs
(London, 1909).
 H
Handling, storing, and transporting of musical instruments: general
guidelines

Environmental considerations

Instruments that are made of wood, animal skins, and other organic
materials should be protected from extremes and fluctuations of temperature
and humidity: 70°F (20°C) and 50% humidity are considered optimal for
most instruments. Wood shrinks when the humidity decreases and swells
when it increases, but the rates of shrinking and swelling vary with grain
direction. Longitudinal dimensional changes are negligible, but the radial and
tangential changes are significantly greater (see Wood). Because instruments
such as violins, viols, lutes, guitars, harpsichords, and pianos have
soundboards and case structures that are constrained or reinforced by ribs and
bars that are glued across the grain, changes in humidity may cause these
parts to crack, warp, or separate. Changing humidity causes gut strings to
expand and contract and thus go out of tune, and temperature change will
have a similar effect upon metal strings. Because brass and steel strings
expand and contract at different rates, they will not rise and fall in pitch
together, thereby compounding the tuning problems of keyboard instruments.
Metal and wood organ pipes are affected differently by temperature and
humidity variations, causing them to go out of tune with each other.
Mechanical parts made of wood and other organic materials tend to go out of
adjustment or may cease to operate as a result of changes in humidity and
temperature.
Sustained humidity levels above 65% encourage the growth of mold and
mildew, which are especially destructive to key and action cloths, felt,
leather, and wood. It is also advisable to keep metal instruments at reduced
humidity levels, for at about 65% or higher, so-called “bronze disease” may
erupt in brass instruments, and steel and iron parts are prone to rust. In certain
instances, ethnographic instruments made in tropical environments may be
more safely stored and exhibited at higher humidity levels, but generally not
above 65%. Asian lacquered objects should be stored at 60% to prevent the
lacquer from buckling, cracking, and detaching from the underlying structure,
which in the case of musical instruments is generally of wood.
When instruments are made of both organic and inorganic materials – for
example, a wood flute fitted with silver keys – the environmental levels
suitable for the most sensitive component (in this case the wood) should take
priority.
Illumination levels should be kept as low as possible to prevent the fading
of varnishes, which are often tinted with fugitive organic dyes (see
Illumination levels: Figure 18). Cabinetry made of woods such as walnut,
mahogany, and rosewood is readily bleached by sustained exposure to light.
Ivory will turn yellow if stored in total darkness, though moderate levels of
natural or artificial illumination will keep it white. For this reason, ivory
instruments should be stored and exhibited so that a moderate amount of light
reaches all surfaces, otherwise masked or shaded areas will gradually turn
yellow. It is advisable to store pianos and other keyboard instruments with
their fall-boards in a raised position so that their ivory keys receive some
ambient light. Clear acrylic covers can be fitted to keep the keyboards clean
and prevent unauthorized handling.
Instruments and storage cases should be examined for woodworm, mites,
silverfish, and other pests. Chemical fumigants, desiccation, and freezing are
not appropriate methods of ridding musical instruments of these pests; anoxic
treatment is the best method, providing that the inert gas used to kill the pests
is humidified (see Disinfection, disinfestation, and fumigation).

Handling and playing instruments

Whether or not historic musical instruments should be played is a

philosophical issue that is debated endlessly by curators, conservators, and
the public. Some museums have declared a moratorium on restoring and
playing their instruments, while others continue to use them on a regular
basis. Curators and collection administrators should be aware that repetitive
tuning and adjustment, prolonged rehearsal and recording sessions, and full-
length recital and concert programs take a toll on old instruments. Concerts
and recordings invariably involve moving instruments, which can be
damaging to large and delicate ones such as harpsichords, fortepianos, and
chamber organs. For instruments that are played, the same handling and
environmental safeguards should be followed as for all other instruments in
the collection: performance areas should be maintained at the same
temperature and relative humidity levels as display galleries and storage
rooms, and strong stage lights should be avoided because they generate heat
that can cause instruments to go out of tune. Instruments must never be tuned
above the pitch level for which they were originally designed.
All instruments should be carefully examined before handling, noting
existing damage and points of weakness. With regard to brass and woodwind
instruments, determine how the separate components are assembled in order
to prevent them from loosening or falling off when they are examined or
moved. More than one person may be required to move large instruments
(see below). Plan the course of action in advance. Be sure that work surfaces
are clean and that any jewelry, watches, keys, dangling ID cards, and the like
are removed or secured before examining or moving an instrument to prevent
inadvertent scratches or other damage. Each type of instrument has special
handling requirements that are briefly described below.

Keyboard instruments

Keyboard instruments should be moved as little as possible, and ideally

they should not be repositioned after they have been given their final tuning
prior to a performance. Lifting or rolling early keyboard instruments even
over short distances or uneven flooring not only will disturb their tuning but
also may weaken legs and stands. Never touch the metal strings of a
keyboard instrument with bare hands as skin oils and perspiration are highly
corrosive; avoid touching varnished, painted, and gilded surfaces as well.
After playing or handling keyboard instruments, fingerprints and perspiration
should be removed from keyboards, stop knobs, etc. with a soft cloth.
Harpsichords with pedal-operated stop actions should have their pedals
shifted into their “off” position to relieve tension on springs.

Organs
 After playing, stop knobs should be pushed back and any couplers or
combination mechanisms that work against springs should be released or
disengaged. The dump-valve should be activated to empty the reservoir.

Stringed instruments

Stringed instrument cases are designed to safely hold the instrument and
several bows; extraneous objects (such as music, concert programs, rosin,
metronomes, and tuning devices) should be stored in outer pockets or
separately. When removed from their cases, violins, viols, lutes, and guitars
are safely carried by the neck with the other hand supporting the bottom of
the instrument. It is not necessary to wear protective gloves when handling
stringed instruments unless the neck itself is of special historic importance or
unusually fragile (nothing is more ludicrous to violinists than having an
instrument handed over to them before a performance by a conservator
wearing gloves). However, one should avoid touching varnished surfaces,
and any fingerprints, perspiration, or saliva that happen to land on the
instrument during handling or playing should be wiped off with a soft cloth.
After playing violins and viols, rosin dust (which is abrasive to varnish)
should be gently brushed or wiped off varnished surfaces as well as from the
strings and fingerboards. Musicians and researchers who visit collections
should not be allowed to use solvents, commercial cleaners, polishes, or
specially treated cloths (such as those impregnated with oil or silicone) to
remove rosin or to shine up instruments. Old, caked-on rosin should only be
removed by a conservator.
When handling bows, do not touch the hair as skin oil will prevent it from
drawing sound from the strings. Bows are safely held by the frog, leather
grip, or wrapping, and carried in an upright position. When examining the
head, one may support the tip of the head in the palm of the opposing hand,
but again, one should take care not to touch the hair. Bow hair should be
loosened after playing by giving the button a few twists or by unclipping the
frog. If left under playing tension, a bow may warp or lose its camber. Hair
that is reasonably loose under humid conditions during the summer months
may tighten up if the humidity drops in the winter.

Wind instruments
 Before hand-carrying a woodwind or brass instrument, check to see that the
joints, tuning slides, bits, and crooks are sufficiently secure. Mouthpieces and
clarinet bells, for example, are relatively heavy and can fall off the instrument
if they are not properly fitted. Special oils and greases are employed in
lubricating tenons, valves, tuning slides, and crooks (see Brass instruments).
Over time, these lubricants may evaporate or harden, causing parts to seize. If
the various parts cannot be disassembled with a light touch, do not apply
force (again, see Woodwind instruments and Brass instruments). If a
woodwind instrument is new or has not been played regularly, it is advisable
to “break it in” by gradually increasing the playing time, beginning with
perhaps ten minutes the first day, and gradually increasing playing time to
about an hour over a period of a week or more (see Woodwind instruments).
This allows the moisture level in the wood to build up gradually. When a
woodwind instrument is played, condensed moisture drips down the bore,
where it is adsorbed, while the external surface remains comparatively dry,
thereby setting up tension in the wood that may result in cracking or warping.
After playing, an appropriate swab should be inserted and pulled through the
bore to absorb condensed moisture. A traditional method was to use a turkey
feather, which was inserted and twisted to spread the moisture evenly around
the bore to facilitate even evaporation. The bores of woodwind instruments
that are in regular use are swabbed with oil to help repel moisture. The types
of oil include almond oil, though it is apt to become rancid; linseed oil, which
hardens to form a varnishlike film; and mineral oil, which avoids the
problems of rancidity and hardening, but was not traditionally used. Oiling is
done periodically, not necessarily before each use. Historic woodwind
instruments that are not played and are kept in storage or on display should
not be oiled. Brass instruments, even if lacquered, should be handled with
cotton or nitrile gloves. Latex gloves should not be used because they are
vulcanized (processed with sulfur) and are thus potentially corrosive. For this
reason latex gloves must never be left inside carrying or storage cases. Before
playing brass instruments, pistons and slides should be cleaned of old
lubricants, and fresh oils and greases should be applied (see Brass
instruments). After playing, they should be rinsed with running water and
rotated to eject as much water as possible; bells and other accessible areas of
the bore should be wiped dry with a soft cloth. If metal instruments have been
played without gloves, they should be wiped down with a quick-drying,
grease-cutting solvent (such as hexanes) before putting them back on display
or in storage. Brass wind and woodwind instruments are easily damaged by
extraneous objects (such as music, concert programs, swabs, metronomes,
and tuning devices) placed in their fitted cases.

Percussion instruments

Percussion instruments that have old, embrittled parchment heads should

not be played as they may rip. If there is some mechanism for adjusting the
tension on drum heads, reduce it to a minimal level. This should be done
gradually and evenly around the perimeter to maintain even tension. Unequal
tension may cause the body or shell of the drum to become distorted, or the
head to buckle or rip. Raising the tension and tuning a drum head for
performance is a delicate operation that is best left to a professional
percussionist.

Display and storage

Storage, display, and carrying cases made of wood, plywood,

particleboard, masonite, cardboard, fabrics, paints, and adhesives may exude
corrosive fumes that can tarnish instruments made of brass, nickel silver, and
silver. Manufacturer-supplied carrying cases are generally not recommended
for long-term storage of brass and silver instruments and should be kept in a
separately vented room. Display and storage cases made of glass or stainless
steel are safe for long-term storage or display; baked enamel or powder-
coated steel storeroom shelving or enclosed cases are also safe for long-term
storage. Polyethylene sheeting, bags, and foam are safe for wrapping
instruments or making supports for storage. Bags should not be hermetically
sealed, as condensation may form inside the bag if climate-control equipment
fails, and there should be a free flow of air to prevent the buildup of corrosive
gases that may be exuded by storage cabinets and shelving or by the
instruments themselves. Rubber and elastic bands should not be left in cases
or used to bind parts together because they contain sulfur that will quickly
tarnish brass and silver instruments. Latex gloves should not be left in
carrying cases or in proximity to instruments made of brass, nickel silver, or
silver.
When mounting instruments for display, refrain from using wood,
plywood, or dowels for supports. Mounts and armatures are preferably made
of silver-soldered brass. Stands can be fabricated out of Plexiglas
(polymethyl methacrylate, or PMMA) or other stable materials. Display
mounts should be designed to facilitate easy removal of instruments that are
frequently played or examined and should adequately support detachable
parts that might fall off, such as clarinet bells, mouthpieces, slides, tuning
crooks, and bits. Vinyl, plasticized tubing, and wool felt or cloth should not
be used as padding between metal mounts and instruments because they are
corrosive. Acrylic or polyester felt is safer to use and may be attached with
Paraloid B-72 adhesive (50% solids in acetone). Polyolefin heat-shrink
tubing can also be used as a protective barrier on metal mounts. The Oddy
test should be employed to determine the safety of any materials used in
exhibition vitrines and storage cases.

Keyboard instruments

Keyboard instruments that are no longer played should have their strings
tuned down. This should be done gradually (perhaps over a period of several
weeks) and evenly throughout the entire keyboard range. Some tension
should be retained to prevent tuning pin coils from springing loose and to
keep the strings reasonably straight. Keyboard instruments that are in regular
or frequent use should be maintained at their optimal pitch. Continually
raising and lowering the pitch disturbs the delicate equilibrium between
string tension and case structure that is required for tuning stability.

Stringed instruments

Stringed instruments that are no longer played or that are infrequently used
are best tuned down to reduce tension on the soundboard and body of the
instrument. De-tuning should be done gradually and evenly throughout the
entire range of the instrument, leaving sufficient tension to prevent bridges
and soundposts from falling and string coils from springing loose from their
tuning pegs. A pad of soft cloth or cotton batting should be fitted under the
tailpieces of violins and viols to prevent damage to the instrument's top
should the tailpiece come loose (this is especially important if metal fine-
tuners are installed on the tailpiece). Stringed instruments should be checked
from time to time to determine if their strings have broken; if so, they should
be replaced to rebalance the tension on the bridge, tailpiece, and other
fittings. If it is unlikely that a stringed instrument will ever be used, consider
removing all the strings and fittings and storing them securely. Before putting
bows away, make sure that the hair is well loosened; if the humidity drops,
moderately slackened bow hair may shrink sufficiently to damage the stick.
When placing harps in storage, lower their pitch to reduce string tension and
hitch the pedals in the highest position to reduce tension on springs. Unless
they are on display, the pedals should be folded up to avoid damage.

Wind instruments

Wind instruments made of metal or having metal key work and fittings
should not be stored in their carrying cases, as most commercial cases are
made of materials that are potentially corrosive. In general, it is advisable to
store wind instruments in a disassembled state so that joints, slides, and
tuning bits do not become cemented together by aging grease and other
lubricants. To prevent tarnishing, silver instruments may be wrapped or
stored in sacks made of cloth impregnated with colloidal silver (such as
Pacific Silver Cloth) or placed in special polyethylene bags impregnated with
copper. Glass trays filled with activated charcoal (which absorbs airborne
sulfur compounds and other corrosive gases) can also be placed in drawers
and cabinets. These specially treated cloth and plastic bags, as well as the
activated charcoal, must be replaced periodically. If metal instruments are
stored in cabinets or drawers, avoid using foam rubber or wool carpeting as
padding, as they generally contains sulfur that will tarnish metal.
Polyethylene foam sheets and blocks are safe for padding cabinets and
drawers. Woodwind instruments that have bells and other protuberances that
extend beyond the body of the instrument should not be stored or displayed in
a horizontal position unless some provision is made to support the body of
the instrument to prevent weight-induced distortion. Corked and thread-
wrapped tenons of assembled woodwind instruments should be checked and
lubricated periodically to prevent sockets and tenons from seizing. Brass
wind instruments displayed in an assembled state should have their sliding
parts and valves cleaned of old grease and oil and relubricated with stable
lubricants on a regular basis (see Brass instruments). Do not attach tags or
labels with rubber bands, as they will tarnish metal.
Music boxes and automatic instruments

Those with wind-up mechanisms should be stored with their springs let
down. The safest way of doing this is to allow the music box to run down on
its own, as the quick release of the mainspring may damage the gear train.
Never touch reeds, combs, or other metal parts with bare fingers. Paper
treated with a vapor phase corrosion inhibitor (VPI or VCI) can be cut to
shape and placed over exposed mechanical parts, such as steel combs,
cylinders, and discs, to prevent rust.

Ethnographic instruments

It is imperative that special storage and exhibition methods be devised to

protect these often delicate objects from airborne dust, because many of the
materials from which they are made, such as feathers and minimally cured
animal skins, are virtually impossible to clean. Ethnographic instruments
should be stored in acid-free boxes or polyethylene bags. Again, bags and
boxes should not be tightly sealed to prevent condensation should climate-
control equipment fail, and they should permit the free flow of air.
Instruments from tropical regions may require special micro-climates for
storage and display. Asian lacquered objects should be maintained at 60%
humidity.

Transport

Keyboard instruments

Again, early keyboard instruments should be moved as little as possible.

Even unrestored examples with moderately slackened strings may still be
under hundreds or even thousands of pounds of string tension. Old wood-
frame instruments, and those with veneered, painted, and gilded surfaces, are
especially delicate, and moving them often causes damage.
Before rolling a modern piano across the floor, check to see that the legs
are in good condition, properly tightened or toggled in place, and that the
casters are firmly attached and roll freely. Never give a keyboard instrument
a shove to clear a threshold. When rolling older pianos, lift a bit while rolling
to take some weight off the legs. Most harpsichords have separate stands that
are not attached to the instrument itself. When moving these instruments, lift
from the stand or remove the instrument from the stand and have extra
movers on hand to reposition the stand under the instrument. Always close
the lid before moving a piano or harpsichord. When moving a keyboard
instrument, it is advisable to have a spotter to oversee the moving operation
and check for steps and other obstacles.
Before packing a piano or harpsichord for transport, the casework should
be checked carefully for loose veneer, moldings, and ornamental mounts. A
polyethylene foam block can be cut to shape and positioned over the hammer
shanks to immobilize them during the move. Before wrapping with moving
blankets, partially detached veneer or moldings can be temporarily secured
with drafting tape. Prop sticks (if not hinged), candle holders, music desks,
benches, tuning wrenches, and other loose parts should be labeled,
inventoried, and packed separately.
Turned legs on pianos are usually threaded and are removed by twisting;
carved or square legs may be screwed on, though many modern pianos are
fitted with cast-iron plates that dovetail into metal mounts screwed to the
undercarriage of the piano. Toggles prevent the legs from sliding off these
fittings, and they must be rotated away from the legs before they can be
removed. It may be necessary to use a rubber mallet to loosen dovetailed
plates or toggles. Pedal lyres may be screwed to the case or held in place with
toggles. When removing pedal lyres, number the pedal rods or secure them in
place so that they do not get mixed up. Some manufacturers (such as
Steinway) employ a short dowel between the pedal linkage and the internal
mechanism that lifts the dampers; this dowel (termed the Pitman linkage)
may fall out of the instrument when the pedal lyre is removed; if so, make
sure it does not get lost, and reinsert it when the piano is reassembled.
Exposed pedal trap-work must be protected if piano cases are placed on carts
or dollies, or are shipped flat. This can be done by placing boards of
appropriate thickness under the instrument's bottom.
To move a modern grand piano, remove any loose parts. The lid can either
be removed or secured with canvas straps; the fall-board should also be
secured, either by locking it or with tape. If the lid is removed, the music
desk should be as well. After disassembling the pedal lyre, the left front leg is
removed (the casters of the other legs having first been rotated away from the
spine, or back), and the front left corner is gently lowered onto a folded
moving blanket. If a piano board (sometimes termed a “skid-board”) is used,
position the piano's spine so that the lid overhangs the padded board. The
other legs can then be removed. The case may then be wrapped with moving
blankets and secured to the piano board with canvas straps. To place a dolly
under the piano board, lift the lighter back end of the board and shove the
dolly below it. The dolly should then be repositioned so that the case of the
piano is balanced. To set up the piano, reverse the process by installing and
securing the back and right front leg first, turning the casters away from the
spine, and then carefully lifting the spine of the case (movers should position
themselves to prevent the piano from rolling as it is lifted). The left front leg
is then attached, followed by the pedal lyre. At least three people should be
on hand to move a piano.
Painted lids and cases should be treated like panel paintings. A painting
conservator should examine all painted surfaces prior to packing. Special
handling or crating may be required.
Figure 13 is a view of the 1985 Keynotes exhibition at the Metropolitan
Museum of Art – a product of safe handling, storage, and transport.

Figure 13 Exhibition of pianos at the Metropolitan Museum of Art in 1985.

The author worked on the cabinetry of most of the thirty-two pianos in this
exhibition and restored to playability a newly acquired fortepiano by
Ferdinand Hofmann, Vienna, c.1790 (MMA, 1984.34). Ten of these pianos
were then playable, and the author prepared several for use in concerts and
gallery demonstrations during the exhibition.

Stringed instruments

Violins can be safely transported in their fitted cases. The best carrying
cases employ a suspension system that isolates the instrument from shock.
Dampits or other means of maintaining humidity in the case should be
employed, especially when transporting instruments in airplanes or to dry
climates. Never leave an instrument in a sealed car or trunk because of heat
buildup under the sun. Do not place heavy objects on top of an instrument
case. Music, programs, metronomes, tuners, and other extraneous materials
that might cause damage to the instrument should not be placed inside the
case but in outside pockets.
If a fitted carrying case is not available for transport, the instrument will
have to be packed in such a manner that minimal pressure is exerted on the
body of the instrument. The bridge and strings of violin- and viol-family
instruments must be completely isolated from pressure by the case or packing
material, as an impulse delivered to the bridge or strings may cause great
damage to the body of the instrument. It is advisable to de-tune stringed
instruments during shipment, but leave enough tension so that the bridge and
soundpost will not fall.
Whenever transporting instruments of the violin or viol families, it is
advisable to place a folded cloth under the tailpiece to prevent it from
damaging the instrument should the bridge fall in transit. Violins with neck
stability problems are sometimes stored or shipped with a thick strip of
cardboard gently wedged between the fingerboard and the top of the
instrument. Fitting and installing this strip should be done by a specialized
conservator.

Wind instruments

Most woodwind and brass instruments can be safely transported in their

fitted carrying cases. If fitted carrying cases are not available, brass and
woodwind instruments should be disassembled and the individual joints and
other loose parts inventoried and packed separately. Loose or extra crooks
and mouthpieces of brass wind instruments should also be inventoried and
carefully packed. Instruments can be supported by shaped blocks of
polyethylene foam, and acid-free tissue paper or polyethylene bags or
sheeting used for wrapping. Avoid using foam rubber, rubber bands, latex
gloves, or any material that might contain sulfur.

Percussion instruments

If drums are being moved to a drier climate, it is advisable to loosen the

tension on the heads, if possible. Provide extra padding or a fitted board for
drum heads, as they are delicate and easily perforated. Marimbas and
xylophones have bars that are generally strung together to keep them in order
and attached to their stands. If the bars are loose, some system should be
devised to keep them attached to the stand, or they may be individually
wrapped, numbered sequentially, and packed separately.
Beaters and drumsticks should be labeled and inventoried to prevent them
from getting lost or switched with other instruments.

Ethnographic instruments

These may be problematic with regard to packing and shipment because

they are often made of extremely fragile materials, such as gourds, palm
leaves, reed and plant fibers, minimally cured animal skins, feathers, etc.
Painted and lacquered instruments should be examined by a specialized
conservator prior to packing and shipping. Again, lacquered objects should
be maintained at 60% relative humidity.

References
Robert Barclay, Anatomy of an Exhibition: The Look of Music (Ottawa,
1983).
Robert Barclay, Florence Gétreau, Friedemann Hellwig, Cary Karp, Jeannine
 Lambrechts-Douillez, and Frances Palmer, CICIM Recommendations for
Regulating the Access to Musical Instruments in Public Collections (1985).
Mario Igrec, Pianos Inside Out (Mandeville, LA, 2013).
Simon Knell, ed., Care of Collections (London and New York, 1994).
Nathan Stolow, Conservation and Exhibitions (London, 1987).
John M. A. Thompson, ed., Manual of Curatorship: A Guide to Museum
Practice (Oxford, 1994).
Garry Thomson, The Museum Environment (London, 1978).

Harp

Modern concert harps generally have 46–47 strings with a 61/2-octave

range. The large concert harp of 47 strings has a range of CC to g4; harps of
46 string have ranges of either DD–g4 or CC–f4. So-called “baby grand”
harps and Irish harps often have a compass of 31 notes, with a range of E–g3.
Single- and early full-size double-action harps had ranges of 36 to 43 strings,
the latter with a range of E1 to e4. The system for designating the octaves of
the harp is different from that used for most other instruments (that is: CCC,
CC, C, c, c1, c2, c3, c4, with c1 representing middle C on the piano keyboard)
in that the numbering of the harp's octaves begins on F rather than C, with the
top f4 and g4 designated as being in octave 0, the first full octave downward
being octave 1, middle C (on the modern piano) being in octave 4, and the
lowest octave termed octave 6. This is important to remember when ordering
or replacing strings.
Modern concert harps are strung primarily in gut or nylon, with steel-core
strings overwound with copper wire (harp string suppliers often refer to these
as “wires”) generally used for the lowest octave. Early single- and double-
action harps (up to around 1835) often used wire-wrapped silk for the lowest
octave. To orient the harpist, C strings are dyed red and F strings are dyed
black, blue, or sometimes purple, while the other strings are left in their
natural color. In another coloring system, the C strings are dyed green, the F
strings dyed purple, and the remaining strings dyed red. For the overspun
strings, the G and F strings are left their natural copper color, while the other
notes are plated with silver or nickel (white). In restringing a harp or
replacing missing strings, existing dyed strings can be used as guides for
determining the pitch of the string that needs to be replaced, and hence the
proper gauge. If the colored strings are missing and the harp has either a
single or double pedal action, the middle pedal for the left foot operates the
sharps and flats for C, so pressing the pedal will identify that note. The pedal
sequence for modern harps – as well as those from the late eighteenth century
– runs from left to right as follows: D, C, B (left foot); E, F, G, A (right foot).
With the single-action harp, when the pedal is depressed and locked into
position, the string is “stopped,” or shortened and raised in pitch by a
semitone. In the double-action harp, the string can be raised two semitones,
so the upper position is the flattened note, the middle position is the natural
note, and the lower position is the sharpened note. It may be difficult to
determine the range of early or folk harps that have no pedals, hooks, or note
designations near their tuning pins or on the soundboard. Absent any physical
indication or other evidence (such as the maker's name, model number, etc.),
consider the traditional length of the violin's E string, approximately 330 mm
(this optimal length for gut strings was established in the mid sixteenth
century and is still in use), as a starting point, keeping in mind that the lowest
notes for most harps tend to be C, F, or G; the highest notes of very early
harps run the gamut.
At the soundboard end, the gut strings generally pass through eyelets or
holes in a central wooden batten. In some harps, the strings are secured in
these holes with tapered wood or ivory board pins. These pins have grooves,
which should be aligned with the strings to hold them in place. The ends of
the strings are tied with a so-called “harper's knot.” This is essentially a
slipped overhand knot, which is a simple overhand knot with the free end
directed back through the crossing turn and tightened to secure the loop. This
forms a stopping knot that prevents the string from slipping through the
eyelet or past the board pin. A short length of heavy gut string (from about
the fifth octave upwards) can also be secured by the knot to prevent the
smaller knots of thinner strings from pulling through the holes (Figure 14).
Figure 14 Harper's knot.

Modern professional-grade harps (like many other types of modern

instruments, such as concert grand pianos, orchestral woodwind and brass
instruments, and “classical” guitars) tend to develop structural problems that
are difficult and thus expensive to fix; when this occurs, professional harpists
often replace them with new instruments. Frequently, the older harps are
given to museums, where they languish, or are sold to students and amateurs
who either tolerate their defects or have them rebuilt with new parts, such as
soundboards, necks, and shoulders. They may also fall into the hands of
interior decorators, who typically go over them with bronze-powder paint to
disguise the often chipped or broken-off gilded gesso ornaments. Decorators
and their restorers are generally at a loss to replace strings, which is when the
musical instrument conservator gets called in.
Harps support tremendous string tension, which can exceed 700 kg (1500
lb) in the case of modern concert harps. Unlike the strings of the piano, which
are supported by a cast-iron plate, and only press downward on the
soundboard at a fraction of their total tension, the strings of the harp are
attached to the soundboard at full tension and pull up on it. Over time this
stress causes harp soundboards to bulge upwards, their necks to distort and be
pulled towards the soundboard, and the hollow fore-pillars to warp. This
structural distortion throws the pedal action out of regulation due to “over-
motion” of the pedal rods; fortunately there is generally some means of
adjusting their effective length in the pedal box. Over-motion can sometimes
be corrected by simply replacing compressed pedal felts. New felt is wrapped
around the pedal and glued or sewn in place. In a well-regulated harp, the
strings are centered relative to the pins projecting from the natural and sharp
discs. If the harp has adjustable nuts, the strings can be centered by loosening
the nuts' screws and shifting the nuts from side to side to center the strings.
Distortion may also create intonation problems due to the effective
shortening of the strings. If this has occurred, movable nuts can be
repositioned to adjust the distance between the string's endpoint and the
natural disc pin or pins. Unfortunately, the distance between the natural and
the sharp discs cannot be altered, but the degree of sharpening can be
adjusted somewhat by regulating the amount of string deflection caused by
the sharp disc as it rotates. This is done by raising the pedal to the flat
position and loosening the retaining screw that holds the disc in place. Once
loosened, the disc can be rotated a few degrees so that it will provide more or
less tension after the retaining screw is retightened.
Buzzing may occur if semitone levers, forks, or discs are not providing
sufficient pressure on the strings. As described above, the discs on modern
harps can be loosened with a screwdriver and reoriented to increase pressure.
In certain folk harps and earlier-style mechanisms, it may be necessary to
bend the hooks, the so-called “butter-knife blades,” or the semitone posts to
provide better contact with the string, thereby eliminating the buzz.
Compressed felt coverings on pedals may also affect pedal action, requiring
adjustment to the effective lengths of the pedal rods. Re-covering the pedals
with special harp felt or leather may restore proper regulation.
Harp soundboards often have their wood grain running transversely. This is
done to help resist the pull of the strings, but it also increases the tendency for
the soundboard to develop shrinkage cracks. While these can be shimmed if
the soundboard is not painted or decorated, a better approach is to remove the
soundboard and join up all the cracks. A go-bar deck is very useful for
leveling the cracks in this clamping and gluing operation (see Woodworking
section). After compressing all of the spaces, it may be necessary to add
sections of soundboard at the top and bottom to restore its original
dimensions.
 The carved or compo ornamentation on harps often gets damaged,
especially at the top and bottom of the fore-pillar and the pedal box. The
designs generally repeat, so it is often possible to make impressions of intact
ornaments and replicate them in gesso or compo (Figures 15a and 15b; see
Mold making and Compo, or pastiglia). Silicone dental impression materials
that have a puttylike consistency (such as Optisil or Cuttersil) are convenient
for taking impressions. Missing carved wood ornaments, such as feet, should
be replicated in wood, as cast gesso or compo will not withstand the weight
of the instrument or the general abuse that these parts receive in the course of
playing, handling, and transport. Harps are often beautifully gilded with
contrasting matte and burnished areas (see Gilding).

Concert harp gut string gauges

(in inches), suitable for harps made from c.1860 to present.

0 octave

G 0.022

F 0.022

1st octave

E 0.022

D 0.023

C 0.024

B 0.024

A 0.025

G 0.026
 F 0.026

2nd octave

E 0.028

D 0.029

C 0.030

B 0.031

A 0.033

G 0.035

F 0.036

3rd octave

E 0.037

D 0.039

C 0.041

B 0.043

A 0.046

G 0.047

F 0.049

4th octave

E 0.051
 D 0.053

C 0.057

B 0.061

A 0.063

G 0.067

F 0.071

5th octave

E 0.073

D 0.077

C 0.082

B 0.086

A 0.090

Concert “wires”

(steel core with copper or copper-plated windings)

5th octave

G 0.062

F 0.066

6th octave
 E 0.071

D 0.075

C 0.078

B 0.082

A 0.084

G 0.089

F 0.091

7th octave

E 0.098

D 1.002

C 1.011

Clark Irish harp string gauges (in inches), 31 notes

1st octave

G 0.026 Red

F 0.028 Purple

2nd octave

E 0.0275 Red

D 0.0285 Red
 C 0.032 Green

B 0.033 Red

A 0.034 Red

G 0.035 Red

F 0.036 Purple

3rd octave

E 0.037 Red

D 0.038 Red

C 0.043 Green

B 0.048 Red

A 0.048 Red

G 0.048 Red

F 0.048 Purple

4th octave

E 0.049 Red

D 0.055 Red

C 0.058 Green

B 0.060 Red

A 0.063 Red
 G 0.067 Red

F 0.068 Purple

5th and 6th octave

Metal-wound strings

Silk harp strings made by Baud of Versailles, 1798, GG–b3, (in

millimeters)*

No. Diameter Color (red used for C, blue used for F)

1 0.40

2 0.48

3 0.42 Blue

4 0.50

5 0.56

6 0.60 Red

7 0.60

8 0.60–0.66

9 0.70

10 0.78 Blue

11 0.80
12 0.80

13 0.80 Red

14 0.86

15 0.90

16 0.96

17 1.0 Blue

18 1.0

19 1.2

20 1.2 Red

21 1.2

22 1.3

23 1.4

24 1.4

25 1.4

26 1.6

27 1.6

28 1.20 Wire wound (silver-plated copper, 0.120)

29 1.28 Wire wound (silver-plated copper, 0.114)

30 1.38 Wire wound (silver-plated copper, 0.142)

 31 1.44 Wire wound (silver-plated copper, 0.193)

32 1.52 Wire wound (silver-plated copper, 0.240)

33 1.74 Wire wound (silver-plated copper, 0.221)

34 1.75 Wire wound (silver-plated copper, 0.218)

35 1.74 Wire wound (silver-plated copper, 0.237)

36 1.98 Wire wound (silver-plated copper, 0.329)

37 2.05 Wire wound (silver-plated copper, 0.326)

* The specific gravities of silk and gut are similar, so these diameters may serve as a guide when
stringing in gut.
Figure 15a Henry Greenway cross-strung harp, Brooklyn, after 1895
(Metropolitan Museum of Art, 89.4.1235). This harp required extensive
restoration, including removal and repair of both soundboards, cleaning of
gilded and varnished surfaces, removal of corrosion from metal parts,
rebuilding lost gilded-compo ornaments, and restringing.
Figure 15b Reconstruction of a damaged gilded-compo ornament on the
Greenway harp (see Figure 15a) using casts of grape leaves made from other
areas of the harp and hand-fabricated stems and clusters of grapes. This
photograph was taken prior to completion of the reconstruction, the use of
recutting tools to sharpen details (see text) and gilding.

References
Aspects of the Historical Harp: Proceedings of the International Historical
Harp Symposium, Utrecht,1992, ed. Martin van Schaik (Utrecht, 1994).
Albert Cohen, “A Cache of 18th-Century Strings,” The Galpin Society
Journal 36 (March, 1983), pp. 37–48.
Rajka Dobronić-Mazzoni, The Harp (Zagreb, 1989).
Happiness is a Contented Harp: A Manual on the Care and Regulation of the
Harp (Chicago, 1970).
Roslyn Rensch, The Harp, Its History, Technique and Repertoire (New York,
1971).
Zur Baugeschichte der Harfe vom Mittelalter bis zum 19. Jahrhundert; 13.
Musikinstrumentenbau-symposium in Michaelstein am 6. und 7.
November, 1992, ed. Monika Lustig (Michaelstein, 1995).

Harpsichord maintenance

(see-also Stringed-keyboard restoration, Tuning and temperament, and

Wire gauges for early keyboard instruments)

Quilling

It is preferable to quill antique harpsichords that have original (or old)

wooden jacks and tongues with feather (and leather when appropriate) rather
than with Delrin, Celcon, and other synthetics because these natural materials
are less likely to damage the mortises in the tongues. Plastics are very tough
substances, and they can easily split old wood tongues or “straighten out”
curved or crescent-shaped mortises in them. The traditional materials for
quilling harpsichords were crow and raven quill (the latter being for all
intents and purposes unobtainable today). Some modern makers report
having had success with goose and turkey feathers, which are readily
available, and even those from swans and vultures. Regarding crow, only the
principal flying feathers from the wings are suitable for use as plectra. The
jacks and tongues of most modern harpsichords are designed for plastic
plectra and often do not work properly with natural quill (see below).
Before quilling an entire register, it is important to adjust the “on” and
“off” positions of a register so that all of the plectra will be able to project
past the strings by about half their diameter in the “on” position, and close to
the strings, but not brushing against them, when the register is in the “off”
position. The back-and-forth motion of the registers can be adjusted by
gluing in shims of veneer or cardboard, though many modern harpsichords
employ capstan screws for regulating the motion of registers.
Bird feathers consist of the quill, shaft, and barbs. The end of the quill is
generally not used to make plectra, but only portions of the shaft. The shaft
tapers from the quill to the tip of the feather. When working with crow quill,
the barbs are first removed by slicing or scraping with the knife blade held
roughly perpendicular to the shaft. A well-sharpened violinmaker's knife is
preferable to scalpels or hobby knives with disposable blades because it is
firmer and offers more control. To use the feathers efficiently, it is important
to employ as much of the feather's shaft as possible, so when quilling an
entire register, the narrow end of the shaft is used in quilling the 4-foot
register and the treble of the 8-foot, while the thick end is reserved for the
bass notes. Start with about half a dozen feathers so that one may quill a
series of notes in a row. It is best to start cutting from the thin end of the shaft
rather than from the quill end in order to take advantage of its natural taper.
To fit the shaft to the tongue, use an angled cut as though you were creating a
pen point and trim the sides so that it fits the mortise. The foamy white core
that is encountered when cutting through the shaft should be trimmed back in
shaping the plectrum, but it is helpful to retain a bit where the quill is held in
the mortise of the tongue, as it helps hold the plectrum in place. Glue should
never be used to secure the plectrum to the tongue, as this interferes with later
requilling. With the tongue supported at the front, push the quill in firmly
from the back with the fingertips (Figure 16) or with a pair of needle-nose
pliers. Then trim the back of the quill with a pair of nippers, but leave a short
length protruding from the back of the tongue so that the plectrum can be
pushed forward if necessary. After the plectrum is installed, trim it to length
(again, it should project past the string by about half the diameter of the
string) by inverting it against a voicing block (a piece of end-grain mahogany
or other dense wood) and cutting downwards. Do not cut through the
plectrum from above as this might create a slight burr that could interfere
with the return of the plectrum.
Figure 16 Quilling a harpsichord jack.

The instrument must be up to pitch and well-tuned before final voicing

commences. False strings (that is, strings that “beat” when played alone or
produce an uncharacteristic sound) should be replaced as they interfere with
voicing. Voicing involves the careful trimming of the plectra so that they
produce an even volume, repeat well, and have similar touch (the actual
volume, repetition speed, and touch weight actually change gradually from
treble to bass, but from note to note the transition should be undetectable).
Plectra taper towards the string, and are generally stouter in the bass and
narrower in the treble. Delrin and Celcon plectra often have the undersides of
their tips cut at an angle to provide a backward vector as the jack falls, thus
facilitating the tongue's retreat and enabling the plectrum to clear the string
upon its return. Because of the curved structure of crow quill, beveling is not
feasible, which is why the mortises of tongues designed for quill plectra were
sometimes broached at an angle. If there is sufficient clearance, the tongues
of modern jacks can be made to lean backwards by turning in their
adjustment screws, thereby providing an upward tilt to the plectra and
preventing them from “hanging.” Some makers apply a light coating of olive
oil or graphite to the underside of the tip to assist the plectra in clearing the
string. It is also believed that oil toughens the quill and makes it more
durable. If oil is used, it should be applied sparingly and not allowed to
penetrate the wood tongues, jacks, or other mechanical parts. Synthetic
materials, such as Delrin and Celcon, have a grain direction, and to make the
strongest and most durable plectra they should be cut with the grain running
lengthwise. The grain direction can be ascertained by cutting a perfectly
square piece and bending it back and forth along both axes to determine their
relative strengths. These synthetic materials also have a “skin,” so cutting or
scraping through it will diminish their strength (this is also true of nylon
filament, which is often used in place of natural boar bristle for the springs in
jacks). Voicing is accomplished by adjusting the width and thickness of the
plectra: a broad plectrum produces a full, round sound while a narrow one
produces a somewhat nasal sound; strength can be adjusted either by using
plectra of greater or lesser thickness or by shaving down the under surface.
Regardless of the type of timbre one wishes to create, the widths and
strengths of the plectra should gradually decrease as one moves from the
heavier to the lighter strings.
Many modern harpsichords are fitted with cast Delrin jacks and tongues.
Very often the springs are cast extensions of the tongues, and these have a
tendency to break off over time. The usual recourse is to replace the entire
tongue, though some harpsichord owners are choosing to replace all of the
jacks with wooden facsimiles, which feature traditional replaceable boar
bristle or brass springs. Delrin is a difficult plastic to repair with glue, but
broken-off springs or replacements made by cutting new ones from strips or
sheets of Delrin can be easily “welded” in place with a soldering iron. The
best type of soldering iron to use is a low-wattage pencil-type fitted with a
pointed tip used to solder printed circuit boards. A quick touch will fuse the
spring in place.

Regulation

When regulating an old instrument, one should first determine the original
plucking order of the registers. In the absence of evidence to the contrary, one
may adhere to the modern convention for single-manual harpsichords that
calls for the 4-foot to pluck first, followed by the close-plucking 8-foot, and
then the back 8-foot. If there is a 16-foot that should pluck last. To stagger
the plucking, the back 8-foot plectra should sit about 1/16 inch below the
strings and the front 8-foot a bit closer. The 4-foot plectra, which are
generally quite a bit weaker than the 8-foot plectra, tend to pluck somewhat
quicker, so it may not be necessary to position them closer to the strings than
the front 8-foot rank. The 16-foot plectra should be positioned below the back
8-foot. In double-manual harpsichords, the order of the 8-foot registers is
generally reversed, with the lower-manual 8-foot plucking before the upper
manual 8-foot. If the plucking order is not staggered, the key action will
become progressively heavier as one engages registers. With wood jacks, an
overall adjustment to the heights of the plectra relative to the strings can be
made by placing or removing shims under the key frame, or to individual
registers by placing slips of paper or cardboard under the cloth or leather
padding that their jacks rest on. If just a few jacks need adjustment, it may be
easier to extend their lengths by gluing shims at the bottom end of each jack,
though it is inadvisable to trim old jacks, as trimming is irreversible. The
author worked on the 1744 Jacob Kirckman harpsichord whose jacks had
been lengthened by the application of sealing wax at the bottom. The
solidified sealing wax (see Sealing wax) had been carefully filed to the
precise width and breadth of the jacks, and then filed to provide the required
added length. If all the jacks sit too high, check to see if shims have been
placed under the key frame. If so, removing these shims may restore the
proper position of the jacks. If this is not the case, one may have to replace
the cloth or leather strips at the ends of the keys with thinner material. As a
last resort, the jacks themselves may need to be shortened, but this should be
avoided if possible. If original key cloths and leather are replaced, the old
material should be preserved for future reference.
In most early harpsichords, key dip is established by the jack rail, which is
fitted with layers of cloth or leather padding. Again, a careful examination of
all the original action cloths and leathers, even if greatly deteriorated, may
provide sufficient guidance for establishing (or reestablishing) the original
key dip, as well as the fore- and after-touch. If they are not serviceable and
must be replaced, it is important to save the old material and document how it
was positioned and secured (for example, by gluing, nailing, or sewing).
Problems caused by changes in humidity

Changes in humidity may cause all of the jacks to sit too high or too low
relative to the strings due to the swelling or shrinkage of the case sides. If all
of the jacks sit too low or hang from their dampers, the easiest solution is to
place shims under the key frame in order to raise the keyboard. These shims
can be removed when weather conditions change or the instrument is moved
to a more favorable climate. If the jacks sit too high and the plectra do not
clear the strings, check to see if shims are present under the key bed before
making changes to the key cloths, leathers, or jack length. Modern
instruments fitted with plastic jacks often have individual adjustment screws
that can be turned in or out to lower or raise the jacks, though this is a chore
if seasonal adjustments must be made on a regular basis to 180 or so jacks.
Some makers install screws in the bottom of the instrument that can be used
to raise or lower the key frame in order to adjust the height of the keyboard
and jacks simultaneously.
Changing humidity may also cause entire registers of jacks to over- or
under-pluck. In early instruments, this can be fixed by installing or removing
shims in order to limit or increase the register's travel. Many modern
instruments use capstan screws located at the ends of the registers to make
these adjustments. In spinets and virginals that have fixed registers with jacks
plucking alternately to the left and to the right, variations in humidity or
general dimensional changes occurring over time may cause the left- or right-
facing jacks to over- or under-pluck. If the register's proper orientation
relative to the strings cannot be restored, the only recourse (other than
repinning or replacing the nut) is to adjust the lengths of the plectra, which
unfortunately may make the touch uneven from note to note.

Stringing

See also Wire gauges for early keyboard instruments for information on
stringing.
Figure 17 is a detail of the jacks of a restored bentside spinet made by
Baker Harris of London in 1771, which were originally quilled in leather.
The photograph shows an intermingling of old and new replacement parts.
The new leather plectra were made from hard sole leather purchased from a
shoemaker and split by him to a workable thickness. Vestiges of the old
plectra, which had decomposed due to red rot (see Leather), served as models
for the replacements, which were shaped with a sharp knife. The new strings
seen here were made with Malcolm Rose type “A” iron wire.

Figure 17 New parts fitted to a bentside spinet by Baker Harris, London,

1771.

References
Frank Hubbard, Harpsichord Regulating and Repairing (Boston, 1963).
Edward Kottick, The Harpsichord Owner's Guide (Chapel Hill, 1987).

Helmholtz resonator

(see-also Acoustics)
Musical instruments having soundboxes with openings (such as
harpsichords, guitars, lutes, violins, viols, etc.) behave like Helmholtz
resonators (Helmholtz, 1885) in that the enclosed air resonates at a particular
frequency. Optimally, the air resonance should be at or below the lowest
frequency produced by the instrument's strings.
The formula for the resonant frequency of a Helmholtz resonator is:
where F is the resonant frequency

c is the speed of sound in air

A is the cross-sectional area of the neck or opening
V is the volume of air in the principal cavity or soundbox
L is the length of the neck.

Reference
Hermann Helmholtz, On the Sensations of Tone (London, 1885).

Historical metrology

(see-also Measurement, , Proportion

Historical metrology is the study of the units of measurement that were
used in various locales prior to and including the introduction of the metric
system, which was developed in France around 1795 and officially enacted
into law on December 10, 1799. Before the international adoption of the
metric system, the basic units of measurement employed throughout much of
Europe had been loosely derived from those used in ancient Rome.
Charlemagne attempted to maintain measurement standards throughout his
empire, though after his death in 814 CE, various locales adopted their own
measurements for the various units of length, weight, and volume, resulting
in thousands of different “standards.” The metric system was one of the
products of the French Enlightenment. It was devised to provide a universal
system of measurement and eliminate the need to convert from one local
measurement system to another, though it had a difficult start in its country of
origin. Although it was immediately implemented in Paris and was to become
mandatory throughout France in 1800, the transition to a decimal system
proved difficult for merchants and their clients, who were used to buying and
selling goods using simple fractional divisions of the old pieds (feet), livres
(pounds), and chopines (pints). In response, a system of measuring length,
weight, and volume termed mesures usuelles was introduced in 1812 under
the direction of Napoleon I as a temporary bridge between the older units and
the new metric system. This system adjusted the size of the older basic units
to conform with the new metric standards, but retained the old fractional
divisions. For example, the toise (fathom), a unit of length, was redefined as
having a length of two meters, but this unit was divided into 6 pieds (feet)
and 72 pouces (inches) rather than decimally. The metric system was taught
in French schools between 1812 and 1840, the year it was reinstated. Other
parts of Europe gradually adopted the metric system; for example, Portugal
began using it in 1852, while most of Germany did not convert to it until
1871–1872.
In recent years, organologists have explored the use of pre-metric units of
measurement in their analyses of the design and construction of early musical
instruments. In some instances they have even attempted to deduce where an
instrument was made based upon the premise that certain gross dimensions of
instruments (such as their overall length and width) follow whole or major
fractional divisions of local units of measure. The fundamental assumption of
these researchers is that makers relied upon the use of local units of measure
in the design of their instruments. While it is an obvious assumption that
instrument makers had locally made rulers at their disposal, they may have
been used primarily for mundane tasks (such as spacing the pegs on a violin
pegbox or establishing the lengths of tailpieces); mitigating factors, such as
the prevailing pitch, the breaking stress of stringing materials, and
proportional considerations would have played a more critical role in
determining the actual dimensions of instruments.
A major problem with this analytical technique is the lack of consistency
among the various published tables of equivalents that are being used by
today's researchers to determine the actual dimensions of these old local units
of measure. Though some of their findings appear to present compelling
revelations about musical instrument design, a closer look at the data and
statistical workups may raise doubts about their conclusions.

Nominal and actual dimensions

One point to consider is that the dimensions of instruments given in early

treatises may be nominal rather than actual. For example, back in 1699, Claas
Douwes noted in his Grondig ondersoek van de tonen der musijk that “six-
foot harpsichords are not fully six feet in length but about one-third of a foot
shorter. Likewise, those of five, four, and three feet too have not quite the full
length but are usually somewhat shorter.” Just as in the seventeenth century,
one cannot today use so-called 7-foot and 9-foot Steinway pianos as
measuring sticks, as they are invariably “off” by a few inches. Another
example is organ pipe length, which is traditionally defined by the “foot.”
One might conclude that local units of measurement played a role in
establishing the lengths of organ pipes, and hence local pitch. However, if
known pitches are compared with old local units of measurement, one finds
that some cities that had comparatively long units of measure did not
necessarily have low pitch. For example, in the seventeenth and eighteenth
centuries, Rome is known to have had low pitch, while Venice had
comparatively high pitch, yet according to one source the Roman foot was
about 294 mm in length while the Venetian foot was about 353 mm. Thus if
organ pipe lengths had followed the local units of measure in these two cities,
their relative pitches would have been reversed. In the USA, an estimated
half-trillion running-feet of so-called “2 × 4” lumber have been used in the
construction of suburban houses; however, the actual dimensions of 2 × 4s
after kiln-drying and dressing is about 1½ inches by 3½ inches. If historians
of the future were to evaluate the USA’s measuring system on the basis of the
overwhelming evidence provided by the actual dimensions of lumber and
sales records, rather than rulers and tape measures, they would arrive at a
distorted notion of its units of length.

Did makers use local units of measure?

Some researchers assert that a local unit of measure was used by

Cremonese violin makers in the design of their instruments. However, there
are scant references to the Cremonese braccio on Antonio Stradivari's forms
and patterns preserved in the Museo Stradivariano in Cremona: on a crudely
traced pattern for a harp that he restored he noted that its overall length was
two braccia, and on two tracings of lute bodies that he rebuilt he indicated
their depths in oncia. However, if we consider Stradivari's violin forms and
patterns there is conflicting evidence for his direct use of the Cremonese
braccio, which is inscribed on the south outer wall of the thirteenth-century
Torrazzo in Cremona (the author measured it with a steel rule and found it to
be 484 mm). If we measure the principal widths and lengths of Stradivari's
violin forms and finished instruments with a reconstruction of a Cremonese
ruler, we find that their dimensions rarely fall on whole, half-, third-, or
quarter-oncia (“inch”, or braccio, approximately 40.33 mm), as can be
seen in the table below, which gives measurements of Stradivari's violin and
viola forms rounded to the nearest punto (“point,” or twelfth division of the
oncia, approximately 3.36 mm).

Measurements of Stradivari's violin and viola forms in the Cremonese

oncia

The italicized form letters in the first column are Stradivari's own
designations written on the forms. In each case the MS no. given is the
Museo Stradivariano (Cremona) inventory number.

Inventory
Form Length Width Width Width
no.

upper center lower

bout bout bout

MB [MS no. 1] 8 3 2 4

S [MS no. 2] 8 3 2 4

P (B) [MS no. 6] 8 4 2 4

T [MS no. 8 3 2 4
11]

Q [MS no. 8 3 2 4
16]

PG [MS no. 8 4 2 5
21]

SL [MS no. 8 3 2 4
28]
 B [MS no. 8 3 2 4
33]

B [MS no. 8 3 2 4
38]

S [MS no. 8 3 2 4
39]

P [MS no. 8 4 2 5
44]

G [MS no. 8 4 2 5
49]

V. [MS no. 6 2 1 3
Piccolo 54]

Viola [MS no. 9 4 3 6

55]

CV [MS no. 9 4 2 5
205]

TV [MS no. 11 5 3 6
229]

Of the 64 principal length and width measurements tabulated above, only 8

round off to whole oncia; 3 round off to three-quarters oncia (9/12), 13 to
half oncia (6/12), 3 to one-third oncia (4/12), and none to one-quarter oncia
(3/12). Thus, it does not appear that Stradivari scrupulously relied upon the
principal divisions of the local unit of measure in designing his forms.
However, if we narrow the field to the forms used to make his “grand-
pattern” violins (the P, PG, and G forms), their overall lengths are 8 1/2 once,
while the upper widths are 4 once, the center-bout widths are 2½ once, and
the lower-bout widths are 5 once. When set up with Baroque-style fittings,
the string lengths of Stradivari's full-size violins made on these forms are a
bit longer than 8 once, which is an inconvenient figure because the string
length ultimately must be divided into a 2:3 proportion, with two parts
representing the neck's length (as measured from the nut to the neck's
juncture with the body) and three parts representing the distance from the top
of the body to the bridge. If Stradivari relied solely upon a Cremonese ruler
when designing his instruments, he would have had difficulty dividing up this
length in order to draft neck and f-hole positioning patterns. Certainly, this
task would have been made easier if he had arbitrarily made the string length
of his full-size violins an even 5 or 10 once, as these lengths can be easily
divided into two parts having a 2:3 ratio. However, 5 once would not have
worked because the strings would have been too slack at Baroque pitch, and
at 10 once they would have been much too long to be pulled up to pitch. 8
once turned out to be just right from a mechanical standpoint as the strings
were tense enough to produce a good sound and had sufficient reserve
strength to resist the extra stresses imposed by tuning, and the pressures
exerted by the fingers and bow during playing (though a substantial drop in
humidity causes gut violin E strings to break). Stradivari could have
established a workable string length with the assistance of a monochord,
though by his day the optimal string length had already been established.
Once the string length was set, if the dimensions of the body of the
instrument were to bear any proportional relationship to it (as many
organologists assume), they could have been laid out using a proportional
divider or sector. Thus, the true “unit” underlying the design of any stringed
instrument, including keyboards, was determined by the tensile strength of
the stringing material and the concomitant relationship between pitch and
string length (see Scaling). Furthermore, the ubiquitous copying of violins,
which is documented from the sixteenth century onward, weakens the case
for attempting to discover a violin's country or city of origin by comparing its
basic dimensions with local units of measurement. With the introduction of
the metric system in Europe, some makers redesigned their violins around a
body length of 360 mm, a nice round figure that is amenable to proportional
subdivision by factors of 2, 3, 4, 5, 6, 8, 9, etc.
The ratios and proportions of musical intervals played an important part in
the design of musical instruments. For example, Stradivari left numerous
patterns for mandolins of various pitches, whose neck patterns indicate that
several were designed to carry an extra 2, 5, 10, and 11 frets in accordance
with the 18:17 rule. The fretting sequences, which are marked out on his
paper neck patterns, were probably laid out geometrically with a pair of
dividers, rather than by calculating the fret spacing in oncia and laying them
out with an ordinary ruler, which would have been extremely awkward.

Measurements and statistics

In recent years a number of studies have used historical metrology in the

analyses of musical instrument design. One such study by Grant O’Brien
(2002) contends that Bartolomeo Cristofori scaled a spinet he made in1690
using a local Florentine unit of length, the soldo (then about 27.56 mm);
however, a closer reading of the actual measurements indicates that the 10
soldi figure he posits for the c2 string length is actually the rounded-off
average length (actual average 10.1 soldi) of the shorter and longer strings
that make up the c2 pair, which have individual lengths of 9.8 soldi (271 mm)
and 10.4 soldi (287 mm). O’Brien also concludes that the spacing of
successive alternate jack slots of the spinet is 1 soldo, though this figure was
not arrived at by direct measurement of the spaces between the individual
slots but was calculated by dividing the distance between the first and last
slots by the number of slots. He also points out that the three-octave span of
Cristofori's keyboard averages out to 495.7 mm, or exactly 18 soldi,
indicating an octave span of 6 soldi. This is a compelling observation, though
this metric length also happens to be the three-octave span used by the
Steinway firm in New York, which is equivalent to 19½ inches and
corresponds to a single octave span of 6½ inches.
While O’Brien may be entirely correct in his analysis, when evaluating
conclusions drawn from measurements that only approximate the purported
unit or that are derived from averages of measurements or the division of
overall lengths, it is important to keep in mind that these are not actual
measurements. If a maker used a local ruler to establish various dimensions
or to mark out a sequence of points, then there should be incontrovertible
physical evidence of it, such as transfer marks made by a scriber or divider.
While seventeenth- and eighteenth-century instrument makers such as
Cristofori may not have achieved (or aspired to) micrometer accuracy in their
woodwork, it is unlikely that they would have wandered off the mark by
several divisions of a soldo when using rulers, dividers, compasses, straight
edges, and scribers. When unusual fractional units emerge during an analysis,
the possibility that other factors were employed in the design of the
instrument must be considered.

Reference works

Before the adoption of the metric system, merchants who bought and sold
commodities in other cities, as well as in foreign countries, had to be versed
not only in currency conversion but also in the various weights and measures
used by their clients in other towns and countries. Numerous handbooks of
conversion tables were published to assist local merchants in their
commercial ventures, and similar tables were included in early encyclopedias
and other reference works. Some of the encyclopedic reference works that
predate the adoption of the metric system convert various local units into the
old French or English foot, which were de facto standards before the
introduction of the metric system, whereas most reference works that
postdate the metric system convert those units directly into metric
measurements. Unfortunately, many of the reference works that are currently
being used by organologists to study the dimensions of musical instruments
rarely indicate the sources of their data or reveal how their tables of
equivalents were compiled, though most appear to have acquired their
information from earlier secondary sources rather than by measuring old
rulers or consulting measurement standards preserved by municipalities,
bureaus of standards, or museums. Abraham Rees's Cyclopaedia is one of the
few that gives sources, though they are all of a secondary nature (see below).
Another point to consider is the fact that many local units of length changed
over time – for example in 1782 the Florentine braccio was increased by
decree from about 551.2 mm to 583.6 mm, while in 1765 the Spanish
pulgada was increased from around 23.2–25.6 mm to around 27 mm in
accordance with the length of the old French pouce (though in modern-day
Spain, rulers and tape measures are manufactured with a pulgada that has
reverted back to around 23 mm). Unfortunately, most of the secondary
reference sources do not indicate periods of use for the various units found in
their tables. To further complicate matters, though municipalities may have
adopted new measurement standards, this did not mean that instrument
makers had to change the dimensions of their instruments. As a case in point,
while the Viennese pianoforte maker Conrad Graf appears to have adopted
metric units when laying out his keyboards (he used an octave span of 160
mm), his contemporary, Nannette Streicher & Sohn, retained a marginally
smaller pre-metric octave span of 6 Viennese inches.
In reviewing a number of recently published organological studies, the
author noted a number of discrepancies in the metric equivalents of old local
units of measurement. In Ruckers: A Harpsichord and Virginal Building
Tradition (1990), author Grant O’Brien acknowledges several possible metric
equivalents for the Antwerp duim (thumb, or the eleventh division of the
voet, or foot): 25.7 mm, 26.6 mm, 25.8, mm, 26.07 mm, 25.879 mm, 25.881
mm, and 25.477 mm; Herbert Heyde's Musikinstrumentenbau 15.–19.
Jahrhundert Kunst-Handwerk Entwurf (1986) provides a few more: 25.83
mm, 25.73 mm, 25.96 mm, and 25.84 mm. Both these authors cite only
secondary sources for these equivalents. Though a few of the minor
discrepancies are due to rounding off of the figures, the major differences are
significant when one is dealing with large instruments such as a harpsichord.
For example, a harpsichord that measures 2300 mm in length would be 86.47
duimen in length when the overall length is divided by 26.6 mm/duim and
90.28 duimen in length when divided by 25.477 mm/duim. In examining
several other eighteenth- and nineteenth-century encyclopedias and
specialized treatises that have not been previously cited, the author
discovered scores of discrepancies in local unit equivalents that call into
question the reliability of these secondary reference works as well as
conclusions that have been drawn from them.
Despite these caveats, historical metrology now plays an important role in
the analysis of early musical instrument design and construction. Listed
below are local unit equivalents adapted from Nicolas Bion's The
Construction and Principal Uses of Mathematical Instruments (1758), Denis
Diderot and Jean le Rond D’Alembert's Encyclopédie ou dictionnaire
raisonné des sciences, des arts et des métiers (1780), Charles Hutton's
Recreations in Mathematics and Natural Philosophy (1840), John Henry
Alexander's Universal Dictionary of Weights and Measures, Ancient and
Modern: Reduced to the Standards of the United States of America (1850),
and Abraham Rees's The Cyclopaedia, or, Universal Dictionary of Arts
(Philadelphia, no date). The author makes no claim that the equivalents given
in these sources are any more accurate than those that have been cited in
recently published studies. While researchers may find the following tables
useful, they should note any discrepancies among them and exercise caution
when drawing conclusions from the data they contain.

Source: Nicolas Bion, The Construction and Principal Uses of Mathematical Instruments (1758), pp. 87–88.

Nicolas Bion (1652–1733) was an engineer and maker of mathematical

instruments to the French court. In this English edition of a pre-metric French
work originally published in 1709, the Royal foot of Paris is used as the
standard. The local units in the original table are expressed in terms of the
Paris foot, which was divided into 12 inches, which in turn were divided into
12 lines. Bion's equivalents have been converted into metric units
(millimeters) below using 324.84 mm as the French foot equivalent.

Besançon foot 309.0

Dijon foot 313.6

Grenoble foot 340.6

Lyons foot 340.6

Macon foot 333.9

Paris foot 324.8

Rouen foot 324.8

Sedan foot 331.6

Lorraine foot 291.0

Brussels foot 291.0

Amsterdam foot 282.0

Rhine foot 313.6

London foot 304.5

Danzig foot 286.5

Sweden foot 327.1

Denmark foot 291.0

Rome foot 293.3

Bologna foot 381.2

Venice foot 322.6

Milan foot, great 595.5

Milan foot, small 397.0

Turin foot 512.1

Savoy foot 270.7

Genoa foot 487.2

Vienna foot 315.8

Constantinople foot 708.3

Roman palme 221.1

Genoa palme 245.9

Naples palme 263.9

Portugal palme 221.1

Italy pan 216.6

Italy pan 243.6

Paris ell 1191.1

Avignon ell 397.0

Montpelier ell 397.0

Provence ell 397.0

Flanders ell 694.8

German ell 694.8

Milan fathom, mercers 523.3

Milan fathom, linen drapers 947.4

Florence fathom 582.0

Lucca ras 595.5

Piedmont ras 595.5

Seville yard 836.9

Madrid varre 994.8

Portugal varre 994.8

Spain varre, general 1773.1

Toulouse cane 1773.1

Rome cane 2262.6

Naples cane 2224.2

 Constantinople pic 708.3

India geuse 945.2

Persia geuse 945.2

Source: Denis Diderot and Jean le Rond D’Alembert, Encyclopédie ou dictionnaire raisonné des sciences, des arts et des
métiers vol. 21 (1780), pp. 637–638.

The Encyclopédie of Diderot and D’Alembert, originally published in Paris

in 28 volumes between 1751 and 1772, was one of the most important
literary accomplishments of the French Enlightenment. It went through a
number of editions, including the Swiss one cited here. Its articles on musical
instrument making and various associated crafts are of great interest to those
studying eighteenth-century construction techniques. The table of length
measurements in this French, pre-metric source surprisingly compares local
units of length with the “London foot,” which is divided into 1000 parts for
ease of comparison. For example, the Paris foot is listed as having a length of
1068 parts, as well as its equivalent in English units: 1 foot, 0 inches, and
8/12 of an inch.
To convert the various local units into the metric system (millimeters), the
French foot was here taken as 324.84 mm, and the other units were calculated
in the following manner: 324.84 mm/1068 parts = Local mm/Local parts;
thus, the millimeter equivalent of the London foot was calculated as follows:
(324.84/1068) × 1000 = 304.15.

London foot 304.2

Paris foot 324.8

Amsterdam foot 286.5

la Brille foot 335.5

Antwerp foot 287.7

Dortmund foot 329.7

Rhine foot 314.2

Leyden foot 314.2

Lorraine foot 291.4

Maline foot 279.5

Middlebourg foot 301.4

Strasbourg foot 279.8

Bremen foot 293.2

Cologne foot 290.2

Frankfurt-am-Main foot 288.3

Spanish foot 304.5

Toledo foot 273.4

Roman foot 294.1

Ancient Roman foot 295.6

Bologna foot 366.2

Mantua foot 477.2

Venice foot 353.4

Danzig foot 287.1

Copenhagen foot 293.5

Prague foot 312.1

Riga foot 556.9

Turin foot 323.0

Greek foot 306.3

Paris foot, M. Bernard 324.2

Universal foot 331.2

Ancient Roman foot 295.0

Bologna foot, M. Auzout 346.7

Lyon ell 1209.3

Bologna ell 625.3

Amsterdam ell 690.1

Antwerp ell 691.3

Rhine ell 687.4

Leyden ell 687.4

Frankfurt ell 555.4

Hamburg ell 579.4

Leipzig ell 687.4

Lübeck ell 580.3

Nuremberg ell 677.4

Bavaria ell 290.2

Vienna ell 320.3

Bologna ell 653.0

Danzig ell 578.8

Florence braccio 581.9

Spanish palm 228.4

Castille palm 228.4

Spanish vare or verge 912.8

Lisbon vare 836.4

Gibraltar vare 839.5

Toledo vare 807.5

Naples palm 109.8

Naples braccio 608.3

Naples canne 2092.6

Genoa palm 115.6

Milan calamus (canne) 1990.4

Parma coudée 567.6

China coudée 309.0

Cairo coudée 554.8

 Turkey pique 669.1

Persia arish 972.4

Source: Charles Hutton, Recreations in Mathematics and Natural Philosophy (1840), pp. 188–190.

Charles Hutton (1737–1823) was a distinguished English mathematician

noted for his calculation of the density of the earth. In this source, the
“English foot” consists of 12 inches, which is divided into 12 lines, which
Hutton further divides into 10 parts. He multiplies these together and then
assigns a length of 1440 parts for the English foot for purposes of
comparison. Because this source postdates the introduction of the metric
system, its author gives a comparative meter length of 4731 parts. Hutton's
table of local units of length measure lists the number of parts as well as the
equivalent in English feet, inches, lines, and tenths of lines.
To convert the local units into the metric system, the number of parts given
in the original table were divided by 4731; thus, the millimeter equivalent of
the English foot was calculated as follows: (1440/4731) × 1000 = 304.37
mm.

Ancient Roman foot 294.2

Greek and Ptolemaic foot 307.1

Greek Phyleterian foot 355.3

Archimedes foot 222.2

Drusian foot 331.9

Macedonian foot 353.0

Egyptian foot 432.5

Hebrew foot 368.8

Natural foot 247.7

Arabian foot 333.3

Babylonian foot 348.3

Babylonian foot 345.6

English foot 304.4

Altdorf foot 235.9

Amsterdam foot 282.2

Ancona foot 390.2

Ecclesiastical States foot 390.2

Antwerp foot 286.0

Aquileia foot 343.3

Arles foot 270.3

Augsburg foot 295.7

Avignon foot 270.3

Barcelona foot 301.8

Basle foot 287.5

Bergamo foot 435.4

Berlin foot 301.8

Besançon foot 309.0

Bologna foot 378.8

Bourg-en Bresse and Bogey foot 313.5

Bremen foot 290.6

Brescia foot 475.0

Breslau foot 342.4

Bruges foot 228.1

Brussels foot 274.6

Chambery and Savoy foot 336.9

China Tribunal of Mathematics 343.1

China Imperial foot 319.8

Cologne foot 274.8

Constantinople foot 668.1

Constantinople foot 354.7

Copenhagen foot 319.4

Danzig foot 280.9

Delft foot 166.3

Denmark foot 318.7

Dijon foot 313.5

Dordrecht foot 234.6

Ferrara foot 400.8

Florence foot 302.9

Franche-Compté foot 356.6

Frankfurt am Main foot 283.9

Genoa palm 247.3

Geneva foot 584.0

Grenoble and Dauphigny foot 340.5

Harlem foot 285.4

Halle in Saxony foot 297.4

Hamburg foot 283.9

Heidelberg (Palatinate) foot 274.8

Innsbruck foot 335.2

Krakow foot 356.0

Leipzig foot 314.7

Leyden foot 311.4

Liege foot 287.5

Lisbon foot 289.8

Lombardy, foot of Luitprand or Aliprand 433.9

Lorraine foot 291.1

Lübeck foot 283.9

Lucca foot 589.1

Lyons, and the Lyonnese, Fores, Baujalois foot 340.5

Madrid foot 278.6

Maestricht foot 278.8

Malta palm 278.6

Mantua brasso 462.9

Marseilles foot 247.7

Mechlin foot 229.1

Mentz foot 300.8

Milan Decimal foot 260.2

Milan Aliprand foot 433.9

Modena foot 633.5

Monaco foot 234.6

Montpellier pan 236.5

Moscow foot 282.6

Munich foot 288.3

Naples palm 262.1

Netherlands foot 278.8

Nuremberg town foot 303.1

Nuremberg country foot 276.1

Padua foot 427.8

Palermo foot 227.4

Paris foot 324.5

Paris meter 1000.0

Parma foot 569.0

Pavia foot 468.6

Prague foot 301.0

Provence foot 247.7

Rhinelandish foot 311.4

Riga foot 283.9

Rome palm 223.0

Rouen foot 324.5

Savoy foot 336.9

Seville foot 301.8

Sienna foot 377.1

Stettin (Pomerania) foot 372.6

Stockholm foot 326.6

 Strasburgh town foot 291.1

Strasburgh country foot 294.9

Toledo foot 278.6

Turin foot 510.3

Trent foot 365.5

Valladolid foot 276.3

Venice foot 346.2

Verona foot 340.1

Vicenza foot 345.8

Vienna foot 315.4

Vienne in Dauphigny foot 322.1

Ulm foot 251.5

Urbino foot 353.6

Utrecht foot 225.5

Warsaw foot 356.0

Wesel foot 234.6

Zurich foot 298.0

Source: John Henry Alexander, Universal Dictionary of Weights and Measures, Ancient and Modern: Reduced to the
Standards of the United States of America (1850).
 In 1845 John Henry Alexander (1812–1867) prepared a report on standard
weights and measures for the State of Maryland, and in 1857 he published An
Inquiry into the English System of Weights and Measures. His Universal
Dictionary of Weights and Measures was dedicated to Alexander Dallas
Bache, then the Superintendent of Weights and Measures of the United
States. In this work, local units are given in US yard equivalents, the US yard
itself being equivalent to 0.9144 meter. To obtain the millimeter equivalents
listed below, the measurements tabulated in his work were multiplied by
0.9144 and then by 1000.

Ancona braccio 643.5

Basel braccio 544.1

Bergamo braccio 656.2

Bologna braccio 645.2

Brescia braccio, woolens 674.0

Brescia braccio, silk 640.0

Cremona braccio 594.9

Ferrara braccio, silk 634.4

Ferrara braccio, woolens 673.6

Florence braccio 583.0

Genoa braccio 581.2

Ionian Isles braccio, silk 644.5

Ionian Isles braccio, woolens 690.5

Leghorn braccio 583.6

Lodi braccio 456.1

Lucca braccio 595.1

Milan braccio, old measure 586.5

Milan braccio, since 1803 1000.0

Modena braccio 577.6

Naples braccio 698.7

Parma braccio, silk 588.0

Parma braccio, woolens 640.0

Parma braccio, wood, surveying 542.1

Pisa braccio, di panno, woolens 583.6

Placentia braccio 675.0

Reggio braccio 529.8

Rome braccio, merchants’ 848.0

Rome braccio, weavers’ 636.0

Sienna braccio, linens 600.3

Sienna braccio, woolens 377.6

Treviso braccio, silk 634.0

Treviso braccio, woolens 676.0

Venice braccio, silk 638.7

Venice braccio, woolens 683.4

Aix-en-Provence pied, old measure 248.6

Alsace pied, old measure 295.3

Avignon pied, old measure 247.9

Bordeaux pied, old measure 356.7

Brussels pied, old measure 275.7

Burgundy pied 331.2

Daughiny pied, old measure 341.0

France, Pied du Roi until 1812 324.8

France pied metrique, 1812 to 1840 333.3

Franche-Compté 357.5

Geneva pied 487.9

Guienne, old measure 346.5

Hainault, old measure 293.4

Liège, builders 295.6

Liège, land 292.7

Lorraine, old measure 286.3

Lyons, old measure 342.5

Normandy, old measure 297.8

Rouen, old measure 268.4

Aarau Fuss 293.3

Aix-la-Chapelle Fuss, builders’ 288.7

Aix-la-Chapelle Fuss, surveyors’ 282.0

Altona Fuss 286.5

Antwerp Fuss 285.6

Anhalt Fuss, “Rhine foot” 313.9

Appenzell Fuss 314.7

Augsburg Fuss 296.2

Baden Fuss 300.0

Bamberg Fuss, old measure 303.7

Basel Fuss 304.6

Bavaria Fuss, since 1809 291.9

Rhenish Bavaria Fuss 333.4

Berlin Fuss, since 1816 313.9

Berlin Fuss, surveyors’ 376.6

Berne Fuss 293.3

Berne Fuss, quarry 317.7

Berne Fuss, of Schweitz 300.0

Bohemia Fuss, old measure 296.4

Bohemia Fuss, imperial 316.1

Bremen Fuss 289.2

Brunswick Fuss 285.4

Carlsruhe Fuss, old measure 333.3

Cleves Fuss 295.5

Coblentz Fuss 290.6

Cologne Fuss 287.6

Courland Fuss, “Rhine foot” 313.9

Danzig Fuss, old measure 286.9

Dresden Fuss 283.3

Frankfurt Fuss 284.6

Frankfort Fuss, surveyors’ 355.9

Freiburg Fuss 293.3

Glaris Fuss 300.0

Göttingen Fuss 291.0

Gotha Fuss 287.6

Hamburg Fuss 286.5

Hamburg Fuss, “Rhine foot”, surveyors’ 313.9

Hanover Fuss 292.0

Heidelberg Fuss, old measure 279.3

Hesse Cassel Fuss 287.6

Hesse Cassel Fuss, surveyors’ 284.9

Hesse Darmstadt Fuss, old measure 287.6

Hesse Darmstadt Fuss, since 1818 250.0

Holstein Fuss 298.5

Hungary Fuss 316.1

Leipzig Fuss, common 282.2

Leipzig Fuss, builders’ 282.7

Lippe Fuss 289.5

Lübeck Fuss 287.9

Luneburg Fuss, old measure 291.0

Mannheim Fuss, old measure 290.3

Mecklenberg Fuss 291.0

Moravia Fuss 296.0

Munich Fuss 291.9

Neufchâtel Fuss 293.3

Neufchâtel Fuss, surveyors’ 287.1

Nürnberg Fuss, old measure 303.8

Oldenberg Fuss 295.9

Osnabrück Fuss 279.3

Rostock Fuss 287.9

Schweiz Fuss 300.0

Solothurn Fuss 293.3

Thurgau Fuss, “Rhine foot” 313.9

Trent Fuss 365.9

Vaud Fuss 300.0

Vienna Fuss, imperial 316.1

Weimar Fuss 282.0

Weimar Fuss, surveyors’ 451.2

Wesel Fuss, old measure 235.1

Würtemberg Fuss 286.5

Zurich Fuss 300.0

Zurich Fuss, builders’, since 1820 301.4

Aragon pié 257.0

Canary Islands pié 282.6

Castille pié 278.3

Curaçao pié 282.6

Havanna pié 282.7

Mexico pié 282.7

Valencia pié 302.4

Bassano pié 357.4

Bologna pié 380.1

Brescia pié 471.0

Carrara pié, surveyors’ 293.3

Casal pié 484.0

Cremona pié 480.3

Ferrara pié 403.9

Florence pié, architects’ 548.2

Genoa pié, liprando 513.8

Genoa pié 296.8

Ionian Isles pié 332.5

Lodi pié 455.3

Lucca pié 589.9

Malti pié 283.6

Mantua pié 466.9

 Milan pié 435.2

Milan pié, architects’ 396.5

Modena pié 523.1

Padua pié 354.2

Parma pié, surveyors’ 544.7

Placentia pié 469.9

Ravenna pié 584.6

Reggio pié 530.9

Rome pié 297.9

Sardinia pié 478.7

Turin pié 342.5

Turin pié, liprando 513.8

Urbino pié 409.6

Venice pié 347.8

Venit. Lombardy pié, mean 347.4

Verona pié 342.9

Vicenza pié 357.4

Source: Abraham Rees, The Cyclopaedia, or, Universal Dictionary of Arts, vol. 24 (undated Philadelphia edition of the
London edition of vol. 22, 1812).
 Rees credits Dr. Young (Thomas Young, see below) for gathering the
equivalents from various sources, which are identified by the following
initials and names that are listed in column one of the table below: Bernard is
Edward Bernard (1638–1697) a British astronomer who published a work on
ancient measurement systems, De mensuris et ponderibus antiquis (1688);
Celsius refers to Anders Celsius (1701–1744), the Swedish physicist and
astronomer for whom the temperature scale is named; Eytelwein is the
German engineer Johann Albert Eytelwein (1764–1848) who served on a
commission that established weight and measurement standards for Prussia. C
refers to E. Cavalli, the author of a subsequently published work entitled
Tables de comparison des mesures (Marseille, 1869); F refers to Folkes,
likely Martin Folkes (1690–1754), who published Dissertations on the
Weights and Values of Ancient Coins in 1733; H refers to Charles Hutto (see
above); Howard is John Howard (1753–1799), the British mathematician
noted for his work on spherical geometry; V refers to Vega, the Slovenian
mathematician Georg von Vega (1754–1802), author of Natürliches Mass-,
Gewichts- und Münz-system, Vienna, 1803; Thomas Young (1773–1829), the
distinguished English physicist and polymath who was one of a group of
scientists involved in calculating the length of the meter (which was intended
to be one ten-millionth of the distance measured along the earth's meridian
from the equator to the north pole), though he was opposed to the meter's
decimal division. He also devised a system for tuning keyboard instruments
that is today called “Young's temperament” (see Tuning and temperament);
Wolfe is likely Nathanael Matthaeus von Wolf (1724–1784), a Polish
physician and astronomer who published an article that makes passing
reference to the Dresden system of measurement in the Philosophical
Transactions of the Royal Society (London) in 1769; Ph. M. is an
abbreviation for the noted British scientific journal Philosophical Magazine
founded in 1798.
The figures given below are from Rees's Table XXVI entitled “Modern
Measures of various Countries compared with those of England.” In this
table, the English foot is given a value of 1, and all other local units are
expressed in terms of a decimal factor; for example, the Altdorf foot is given
as 0.775 while the Amsterdam ell is given as 2.233. In his Table XXV, the
meter is defined as equaling 3.281 feet. The English foot is equivalent to
304.785 mm, which was used here to calculate the millimeter equivalents of
the factors given in Table XXVI.

Altdorf foot, H 236.2

Amsterdam foot, H 282.5

Amsterdam foot, C 283.5

Amsterdam foot, Howard 283.8

Amsterdam ell, C 680.6

Ancona foot, H 390.7

Antwerp foot, H 286.5

Aquileia foot, H 343.8

Arles, Avignon foot, H 270.6

Augsburg foot, H 296.3

Barcelona foot, H 302.3

Basle foot, H 287.7

Bavarian foot, Bernard 295.0

Bergamo foot, H 436.1

Berlin foot, H 302.3

Bern foot, Howard 293.2

Besançon foot, H 309.4

Bologna foot, H 379.2

Bologna foot, C 381.0

Bourg en Bresse foot, H 313.9

Brabant ell, German, V 691.3

Bremen foot, H 291.1

Brescia foot, H 475.5

Brescia braccio, C 637.6

Breslau foot, H 342.9

Bruges foot, H 228.3

Brussels foot, H 274.9

Brussels foot, V 290.8

Brussels greater ell, V 694.3

Brussels lesser ell, V 684.2

Castilian vara 836.9

Chambery foot, H 337.4

China mathematical foot, H 343.5

China imperial foot, H 320.3

China imperial foot, C 320.0

Chinese li, C 184699.7

Cologne foot, H 275.2

Constantinople foot, H 669.0

Constantinople foot, H 355.1

Copenhagen foot, H 319.7

Danzig foot, H 281.3

Dauphiné foot, H 341.1

Delft foot, H 166.7

Denmark foot, H 319.1

Dijon foot, H 313.9

Dordrecht foot, H 235.0

Dresden foot, Wolfe, V 283.1

Dresden ell, V 566.0

Ferrara foot, H 401.4

Florence foot, H 303.3

Florence braccio, C 579.1

Florence braccio, C 582.1

Franche Comté, H 357.2

Frankfurt, H 284.4

Genoa palm, H 247.5

Genoa palm, C 243.8

Genoa palm, C 249.0

Genoa canna, C 2224.9

Genoa foot, H 584.9

Grenoble, H 341.1

Haarlem foot, H 285.6

Halle foot, H 297.8

Hamburg foot, H 284.4

Heidelberg foot, H 275.2

Krakow foot, H, V 356.3

Krakow greater ell, V 616.9

Krakow smaller ell, V 565.4

Leipzig ell, H 558.7

Leyden foot, H 311.8

Liege foot, H 287.7

Lisbon foot, H 290.2

Lucca braccio, C 596.8

Lyons, H 341.1

Madrid foot, H 278.9

Madrid foot, Howard 279.8

Madrid vara, C 994.5

Maestricht foot, H 279.2

Malta palm, H 278.9

Mantua brasso, H 463.6

Mantua braccio, C 637.6

Marseilles foot, H 248.1

Mechlin foot, H 229.5

Mentz foot, H 301.1

Milan decimal foot, H 260.6

Milan aliprand foot, H 434.6

Milan braccio, C 525.8

Modena foot, H 634.3

Monaco foot, H 235.0

Montpelier pan, H 236.8

Moravian foot, V 295.9

Moravian ell, V 790.6

Moscow foot, H 282.8

Munich foot, H 288.6

Naples palm, H 262.4

Naples palm, C 261.8

Naples canna, C 2105.5

Nuremberg town foot, H 303.6

Nuremberg town foot, V 303.9

Nuremberg country foot, H 276.4

Nuremberg artillery foot, V 292.9

Nuremburg ell, V 660.2

Padua foot, H 428.5

Palermo foot, H 227.7

Paris foot, H 324.9

Paris meter, Y 1000.0

Parma foot, H 569.6

Parmesan braccio, C 683.3

Pavia foot, H 469.4

Placentia braccio, C 683.3

Prague foot, H 300.8

Prague foot, V 296.3

Prague ell, V 593.7

Provence, H 248.1
Rhineland foot, H 311.8

Rhineland foot, V, Eytelwein 313.9

Riga foot, H 284.4

Rome palm, H 223.4

Rome foot, F 294.4

Rome palmo, F 76.7

Rome palmo di architettura, F 223.3

Rome canna di architettura, F 2232.6

Rome staiolo, F 1283.8

Rome canna dei mercanti, F 1992.2

Rome braccio dei mercanti, F 849.6

Rome braccio dei mercanti, C 870.5

Rome braccio di tessitor di tela, F 636.0

Rome braccio di architettura, C 780.6

Rouen, H 324.9

Russia archine, C 720.1

Russia archine, Ph. M. XIX 711.2

Savoy, H 337.4

Saville, H 302.3
Saville vara, C 841.2

Sienna foot, H 377.6

Stettin foot, H 373.1

Stockholm foot, H 327.0

Stockholm foot, Celsius, Ph. M. 296.9

Strasburg town foot, H 291.4

Strasburg country foot, H 295.3

Toledo, H 278.9

Trent foot, H 366.0

Trieste ell for woolens, H 676.6

Trieste ell for silk, H 642.2

Turin foot, H 510.8

Turin foot, C 512.3

Turin ras, C 596.8

Turin trabuco, C 3073.8

Tyrol foot, V 334.0

Tyrol ell, V 804.3

Valladolid foot, H 276.7

Venice foot, H 346.5

 Venice foot, Bernard, Howard, V 347.5

Venice foot, C 355.7

Venice braccio for silk, C 642.5

Venice ell, V 636.7

Venice braccio for cloth, C 685.8

Verona foot, H 340.4

Vicenza foot, H 346.2

Vienna foot, H 315.8

Vienna foot, Howard, C, V 316.1

Vienna ell, V 779.3

Vienne in Dauphiné foot, H 322.5

Ulm foot, H 251.8

Urbino foot, H 354.2

Utrecht foot, H 225.8

Warsaw foot, H 356.3

Wesel foot, H 235.0

Zurich foot, H 298.4

Zurich foot, Ph. M. VIII 299.9

References
John Henry Alexander, Universal Dictionary of Weights and Measures,
Ancient and Modern: Reduced to the Standards of the United States of
America (Baltimore, 1850).
Nicolas Bion, The Construction and Principal Uses of Mathematical
Instruments; Translated from the French of M. Bion, Chief Instrument
Maker to the French King by Edmund Stone (London, 1758).
E. Cavalli, Tables de comparaison des mesures, poids et monnaies anciens et
modernes (Marseilles, 1869).
F. W. Clarke, Weights, Measures, and Money of all Nations (New York,
1876).
Denis Diderot and Jean le Rond D’Alembert, Encyclopédie ou dictionnaire
raisonné des sciences, des arts et des métiers vol. 21 (Lausanne, Berne,
1780).
Herbert Heyde, Musikinstrumentenbau 15.–19. Jahrhundert Kunst-Handwerk
Entwurf (Wiesbaden, 1986), pp. 68–85.
Charles Hutton, Recreations in Mathematics and Natural Philosophy
(London, 1840).
Grant O'Brien, Ruckers: A Harpsichord and Virginal Building Tradition
(1990)
Grant O’Brien, “The Development of an Idea: From the Design to the
Instrument,” in Gabriele Rossi-Rognoni, ed., Bartolomeo Cristofori: La
Spinetta Ovale del 1690 (Florence, 2002), pp. 62–79.
Abraham Rees, The Cyclopaedia, or, Universal Dictionary of Arts, vol. 24
(Philadelphia, n.d.).
Georg von Vega, Natürliches, aus der wirklichen Grösse unserer Erdkugel
abgeleitetes … Maß, Gewichts- und Münz-System (Vienna, 1803).
Ronald Edward Zupko, French Weights and Measures before the Revolution
(Bloomington and London, 1978).
Ronald Edward Zupko, A Dictionary of Weights and Measures for the British
Isles: The Middle Ages to the Twentieth Century (Philadelphia, 1985).
Ronald Edward Zupko, Revolution in Measurement: Western European
Weights and Measures since the Age of Science (Philadelphia, 1990).

Humidity control

(see-also Relative humidity)

Ideally, metal objects should be maintained at a relative humidity of less
than 40%; paper, parchment, and wood (including most woodwind, stringed,
keyboard, and percussion instruments) at around 50%; oil paintings at around
50%; Asian lacquer at around 60%; and textiles at around 40–50%.
Storerooms and galleries should be fitted with hydrostatically regulated
humidifiers and dehumidifiers to maintain the relative humidity within these
recommended ranges. In dry climates or in buildings with central heating, it
may be very difficult to maintain 50% humidity, especially in the winter, and
45% may represent a more realistic level. Stability at a compromised level is
preferable to continual fluctuation.
Another method of maintaining constant relative humidity is to store or
display objects in sealed cases along with cloth sacks or glass trays
containing preconditioned silica gel. Silica gel is often used in a completely
dry state as a desiccant. However, in addition to absorbing moisture, it can
also desorb moisture and thus serve as a humidity buffer. Silica gel can be
preconditioned to control relative humidity in an enclosed space by first
placing it in a chamber or room that is adjusted to the required relative
humidity. After the silica gel achieves equilibrium with this environment, it
will serve as a buffer by either absorbing or desorbing moisture from the
surrounding air in sealed exhibition or storage cases. Equilibrium is
determined by weighing the silica gel: when its weight stabilizes at the
desired relative humidity, it has either absorbed or desorbed sufficient water,
a process that may take several weeks. Preconditioned silica gel is
commercially available in a range of relative humidity levels (such as the
products Art Sorb and Sorb-It). One pound of preconditioned silica gel beads
is sufficient to buffer 16 cubic feet. Because most exhibition and storage
cases are not completely airtight, the relative humidity inside the case should
be monitored. When it begins to deviate from the required level, the silica gel
must be replaced or reconditioned. One problem with using preconditioned
silica gel to regulate relative humidity is that opening an exhibition or storage
case in order to remove an object for use or study will upset the equilibrium,
which must then be reestablished.

Hurdy gurdy stringing

For hurdy gurdies tuned in the Auvergne and Limousin modes:

Chanterelles Violin D or G 3rd (uncovered) of guitar

Trompette Violin D or G 3rd (uncovered) of guitar

Mouche Violin D or G 3rd (uncovered) of guitar

Petit Bourdon G of cello

Gros Bourdon C of cello (overwound)

For hurdy gurdies tuned in the Bourbonnais mode:

1st Chanterelle Violin A or E 1st of guitar

2nd Chanterelle Violin D or G 3rd of guitar (uncovered)

Trompette Violin D or G 3rd of guitar (uncovered)

Mouche E 6th of the guitar

Petit Bourdon Cello G

Gros Bourdon Cello C

Guitar E 1st or B 2nd steel strings can be used as sympathetics.

A small piece of cotton wool is often applied under the strings where they
touch the wheel in order to smooth the bowing and tone. The wheel, like
violin bow hair, should never be touched (otherwise skin oil will prevent the
wheel from bowing the strings). A small amount of powdered rosin may be
applied to the wheel.

Reference
Susan Palmer and Samuel Palmer, The Hurdy Gurdy (London, 1980).
 I
Illumination levels

Lux and foot candles are measures of illuminance. 1 lux equals 1 lumen per
square meter, while one foot candle equals 1 lumen per square foot. To
convert from foot candle to lux, multiply by 10.764.
The consensus on a safe illumination level for the display of easel
paintings, undyed leather, wood, bone, and ivory is 200 lux, while
watercolors, drawings, textiles, dyed leather, and natural history exhibits
should be subjected to no more than 50 lux (Thompson, 1994). Special light
meters for the direct measurement of lux are available from suppliers of
conservation equipment, though a conventional camera or light meter can
also be used to determine illuminance levels. To do this, set the camera or
light meter to ASA 50 (18 DIN) and point it at an 18% gray card, or use the
incident light diffuser on the light meter. An exposure value (EV) of 3 (which
is equivalent to an f-stop/shutter-speed combination of 8 seconds at f/8)
indicates 44 lux, and an EV of 4 (equivalent to 4 seconds at f/8) indicates 88
lux, while an EV of 5 (equivalent to 2 seconds at f/8) indicates 175 lux. The
Gossen Luna-Pro SBC light meter features a lux table. Another popular light
meter, the Sekonic Studio Meter L-398M, reads directly in foot candles and
is calibrated down to 2.5 fc.
In the author's experience, a level of illumination of 75 lux over a one-year
period does not insure that fading will not occur in violin varnish (Figure 18).
On the positive side, even this low level of illumination is sufficient to
prevent ivory from yellowing, and over time it will even bleach an existing
yellow cast. Ivory objects should not be exhibited or stored in such a way that
one side is shielded from light, as that side will gradually turn yellow while
the illuminated side will become white.
Figure 18 Fading of varnish on a viola by Carleen Hutchins after one year's
exhibition at a luminance of 75 lux supplied by a ceiling-mounted 60-watt
quartz-halogen lamp. Note the “shadows” of the bridge and tailpiece.

References
Barbara Appelbaum, Guide to Environmental Protection of Collections
(Madison, 1991).
Knell, Simon, ed., Care of Collections (London and New York, 1994).
Nathan Stolow, Conservation and Exhibitions (London, 1987).
John M. A. Thompson, ed., Manual of Curatorship: A Guide to Museum
Practice (Oxford, 1994).
Garry Thomson, The Museum Environment (London, 1978).

Inscriptions, faded

Old, faded, or even scraped-off iron-gall ink inscriptions can often be made
legible under long-wavelength (approximately 365 nanometers) ultraviolet
(UV) light. UV lamps emit invisible ultraviolet rays that cause certain
substances to emit (fluoresce) visible wavelengths, while other substances
absorb these rays and appear dark. This fluorescence is generally weak, so it
is necessary to work in a completely darkened room to prevent ambient light
from overwhelming the fluorescent inscription. The effect of this ultraviolet
fluorescence may be enhanced through photography, using either film or
digital cameras.
Ultraviolet lamps consist of fluorescent tubes or mercury vapor lamps that
have been fitted with an exciter filter, such as a Wratten 18A filter, which
absorbs most of the visible light and transmits the ultraviolet rays. To capture
the visible light fluorescence on film, a UV-blocking filter (such as a Wratten
2A or 2E) is placed over a camera's lens. If a hand-held exposure meter is
used, position the UV filter over the sensor when taking a light reading.
Exposure times are generally fairly long, ranging from several seconds to
minutes. Unfortunately, many amateur-grade digital cameras will not record
images at this low level of illumination, and there is a further problem caused
by infrared (IR) radiation transmitted by the 18A filter (as well as various
coatings found on many UV lamps and tubes) that may penetrate the so-
called “hot mirrors” and infrared filters that are installed in most digital
cameras and mask the weak UV fluorescence. Conventional black-and-white
film, however, is relatively insensitive to IR, so this problem does not occur
with the 18A filter.
Infrared (IR) imaging is another technique for enhancing faded
inscriptions. Special devices are available for IR reflectography, but they are
rather costly. Unfortunately, the “hot mirrors” and infrared filters fitted over
the sensors of most digital cameras render them somewhat insensitive to
infrared and thus unusable for IR photography – though for a modest sum,
the IR filters can sometimes be removed by camera repairmen. (A few digital
cameras, such as the Leica M8, which lacks an IR filter, and Sony models
equipped for night viewing, can be used directly for IR photography.)
Conventional tungsten or quartz-halogen lamps are good sources of IR. An
infrared-passing filter (such as a Wratten 87 or 87C) must be placed over the
lens, but because this type of filter passes virtually no visible light, the
camera must be aimed and focused before attaching the filter. Most lenses
exhibit a focus shift in the infrared region, so some manufacturers engrave a
red line on the lens barrel that indicates how much to offset the focusing ring
after focusing in visible light. Refocusing is not necessary when
apochromatic lenses are used. Infrared film must be used to record IR
images. Unfortunately, it is very grainy and has low acutance (sharpness),
which is a disadvantage when using 35 mm film; however, 4in. × 5in.
infrared sheet film (for use in view cameras) provides better results. When
using IR film, a Wratten 87 or 87C filter must be placed over the lens before
making the exposure.
Attempting to revive faded or scraped-off iron-gall ink inscriptions by
chemical means is risky because the inscription may run or wash away, or the
surrounding area may become discolored. The application of a 1–2% solution
of gallic acid or potassium ferricyanide may “develop” the remains of iron
left from a faded ink inscription. An old recipe (Spon’s Workshop Receipts,
1909) suggests brushing a decoction of oak galls in wine on the faded
inscription to make it visible. Ammonium sulfide vapors have been used to
revive faded iron-gall ink inscriptions.

References
Norman Brommelle and Perry Smith, eds., Conservation and Restoration of
Pictorial Art (London and Boston, 1978).
Ernest Spon, Workshop Receipts for Manufacturers and Scientific Amateurs
(London, 1909).

Intervals

Interval Frequency ratio

Unison 1:1
Schisma 32805:32768 (1.001)

Lesser diesis 2048:2025 (1.0113)

Syntonic comma 81:80 (1.0125)

Pythagorean comma 531441:524288 (1.014)

Greater diesis 253:243 (1.041)

Chromatic semitone 25: 24 (1.042)

Pythagorean minor second 256:243 (1.05)

Diatonic semitone 16:15 (1.067)

Pythagorean augmented prime 2187:2048 (1.068)

Minor tone 10:9 (1.11)

Meantone :1 (1.12)

Major tone 9:8 (1.125)

Pythagorean minor third 32:27 (1.18)

Minor third 6:5 (1.2)

Major third 5:4 (1.25)

Pythagorean major third 81:64 (1.26)

Fourth 4:3 (1.33)

Augmented fourth 45:32 (1.41)

Diminished fifth 64:45 (1.42)

 Meantone fifth :1 (1.49)

Fifth 3:2 (1.5)

Augmented fifth 25:16 (1.56)

Pythagorean minor sixth 128:81 (1.58)

Minor sixth 8:5 (1.6)

Major sixth 5:3 (1.67)

Pythagorean major sixth 27:16 (1.69)

Harmonic or “Blues” seventh 7:4 (1.75)

“Grave” minor seventh 16:9 (1.78)

Minor seventh 9:5 (1.8)

Major seventh 15:8 (1.88)

Pythagorean major seventh 243:128 (1.89)

Octave 2:1 (2.0)

To determine the frequency ratio when intervals are added, convert the
proportions to fractional form and multiply; to subtract intervals, cross-
multiply or divide the fractions. For example, to add a fourth to a fifth,
, an octave. To subtract a fifth from an octave,
, a fourth. To determine how many cents there are in an interval see Cents
conversion.

Iron and steel

 Iron is a chemical element, though it is rarely found in nature in its metallic
state. Its ore consists primarily of iron oxide along with impurities, such as
carbon, silicon, and phosphorus. To convert the ore into metallic iron, it is
heated to drive off the oxygen. This produces an iron that contains over 2.2%
carbon (generally 3–4%), which is suitable for casting (hence it is referred to
as “cast iron”) but that is neither malleable nor ductile and thus cannot be
hammered into shape or drawn into wire. Remelting cast iron in a blast
furnace or by the Bessemer process (1855) exposes it to oxygen, which burns
out most of the carbon and converts it into wrought iron, a form that can be
shaped by both hot and cold working methods. An early technique for
converting iron ore directly into low-carbon iron employed a bloomery,
which was a low-temperature charcoal-burning furnace fitted with inlets for
air to burn off most of the carbon. The intentional incorporation of small
amounts of carbon (0.30–2.2%) to iron yields steel, which has the important
properties of achieving great hardness upon quenching as well as the ability
to be tempered (a form of controlled softening) by gentle reheating. These
characteristics enable steel to be made into fine tools, especially those used in
cutting. Both iron and steel may be drawn into wire, though pure iron cannot
be used to make wire for musical instruments because it continues to stretch
under tension and thus does not provide stable tuning. The inclusion or
retention of small amounts of carbon or phosphorus renders wire that can be
tuned. Modern music wire (such as used to make piano strings and violin E
strings) has a high carbon content and considerably greater tensile strength
than so-called “early iron wire” that was used in stringing clavichords,
harpsichords, and early pianos.
An early description of the wire drawing process can be found in
Vannoccio Biringuccio's De La Pirotechnia (1540); however, Joseph
Moxon's Mechanick Exercises (1677) provides insight into the production of
a special type of iron that was used in making wire:

There is another sort of Iron used for making Wyer, which of all Sorts is
the softest and toughest: But this Sort is not peculiar to any Country, but
is indifferently made where any Iron is made, though of the worst sort;
for it is the first iron that runs from the Stone when it is melting, and is
only preserved for the making of Wyer.
According to Goodway and Odell (1987), the Westphalian process of fining
iron burnt off the residual carbon and removed inclusions, but left
phosphorus, thus producing a type of iron that was uniquely suited for
drawing musical instrument wire. High-tensile-strength steel music wire was
developed in the mid nineteenth century as a result of the invention of the
Bessemer steelmaking process.
The study of the composition and properties of early iron wire continues,
though the limited quantity required by the world's instrument makers and
restorers may make production of such wire unfeasible.

Removing rust and corrosion

To removal rust mechanically from iron and steel objects, flat or relatively
smooth areas can be scraped with a razor blade, jeweler's wire brush, or glass
bristle brush. Graded abrasive cloths or Micromesh can also be used. Crudely
made iron objects such as early tuning pins and wood screws can be
immersed for several hours or overnight in mineral spirits and then wire-
brushed to remove rust. Keep in mind that the use of abrasives, as well as of
wire and glass bristle brushes, will affect the finish and possibly the size of an
object (an important consideration when cleaning tuning pins). Tumbling is
an effective and efficient method of removing rust and corrosion when
dealing with hundreds of small objects. There are two types of tumbler, one
that has a chamber (containing the parts to be cleaned and the cleaning
medium) that rotates, and another that vibrates. Tumbling media include
ground walnut shells and corn cobs; more aggressive media, such as shaped
ceramics and plastics, will erode the object. To remove light deposits of rust
from delicate parts (such as harmonica reeds), dental-grade sodium
bicarbonate can be sprayed on with an air brush.

Chemical cleaning

Rusted iron objects can be immersed in a 5% solution of tetrasodium

EDTA (ethylene diaminetetra-acetic acid). Disodium EDTA works faster
than tetrasodium EDTA, and is thus often recommended for working with
iron, but its acidity (pH of around 5) may cause pitting. A 5% solution of
tetrasodium EDTA has a pH of about 10. The acidity of working solutions of
disodium EDTA may also be adjusted with sodium hydroxide or with
buffers.
Rust can also be removed by soaking in a solution of 100 g sodium
gluconate and 100 g sodium hydroxide in 1 liter of distilled or deionized
water. This works best at about 90°C (194°F).
After chemical cleaning, objects should be thoroughly washed in tap water
and transferred to acetone before drying to prevent the formation of fresh
rust.

Electrolytic reduction

Iron or steel objects to be cleaned are made the cathode (attached to the
negative terminal of a rectified power supply), and mild steel or No. 316
stainless steel is used as the anode (attached to the positive terminal). If large
or unusually shaped objects are being cleaned, the anode can be made of
flexible mesh or screening and shaped to enclose the object, but it should not
touch the object, otherwise the power supply will short out. A glass tank can
be used to hold the electrolyte, such as a 5% solution of sodium carbonate or
2–5% sodium hydroxide. If the current of the power supply is adjusted to
provide a density of about 0.1 amp/cm2, there will be vigorous evolution of
hydrogen gas at the surface of the object. This activity assists in cleaning, but
it may be injurious to delicate objects, in which case the current density can
be reduced to 0.001 or even to 0.0001 amp/cm2, though cleaning will proceed
at a much slower rate.
Tannic and phosphoric acids have been recommended for treating iron
objects to prevent corrosion because they produce a protective surface
coating of ferrous tannate or ferrous phosphate. However, these acids tend to
etch the surface and produce a bluish or grayish surface coloration that this
author finds unacceptable.
Derusted iron objects can be coated with microcrystalline wax, or
varnished with nitrocellulose lacquer such as Agateen no. 27 or an acrylic
varnish such as Paraloid B-48N (12.5% dissolved solids for spraying, 15%
dissolved solids for brushing). Tuning pins, however, should not be treated
with wax (which will cause them to slip) or varnish (which may cause them
to stick).

References
Georgius Agricola, De Re Metallica, trans. Herbert Clark Hoover and Lou
Henry Hoover (New York, 1950).
Vannoccio Biringuccio, De La Pirotechnia (1540), trans. Cyril Stanley Smith
and Martha Teach Gnudi (New York, 1959).
Martha Goodway and Jay Scott Odell, The Metallurgy of 17th- and 18th-
Century Music Wire, The Historical Harpsichord no. 2 (Stuyvesant, 1987).
Donny L. Hamilton, Methods of Conserving Archaeological Material from
Underwater Sites (www.nautarch.tamu.edu, 1999).
Geoffrey Michael Lemmer, “The Cleaning and Protective Coating of Ferrous
Metals,” Bulletin of the American Group – The International Institute for
Conservation of Historic and Artistic Works, 12/2 (1972), pp. 97–108.
Joseph Moxon, Mechanick Exercises (London, 1677).
Colin Pearson, Conservation of Marine Archaeological Objects (London,
1987).
H. J. Plenderleith and A. E. A. Werner, The Conservation of Antiquities and
Works of Art (Oxford, 1979).
T. Stambolov, The Corrosion and Conservation of Metallic Antiquities and
Works of Art: a Preliminary Survey (Amsterdam, n.d.).

Ivory

Cleaning

Antique ivory objects should not be immersed in water, as they may crack,
warp, or become distorted; however, they may be gently swabbed with
distilled or deionized water or with a 1:1 mixture of distilled or deionized
water and ethanol. A nonionic surfactant may be added to assist cleaning, but
afterwards it must be cleared with fresh distilled or deionized water. Blot dry
with a soft cloth after swabbing with water.
Ivory may be polished with a soft cloth moistened with ethanol.
The Victoria and Albert Museum (1971) recommends a teaspoon of
ammonia in a cupful of acetone to clean ivory, but ammonia may dissolve
proteinaceous material from ivory and possibly etch its surface.
In general, bleaching should be avoided, though, if necessary, 3%
hydrogen peroxide can be mixed with whiting to form a paste and applied
locally to bleach organic stains. Iron stains can be removed with 5–10%
oxalic acid, 5% ammonium citrate, or 5% disodium EDTA. After treatment,
these chemicals must be cleared with deionized water.
Paraloid B-72 in acetone can be used to glue ivory. When gluing on piano
ivories, whiting is often added to animal-hide glue in order to mask the color
of the wood key levers below.
Ivory can be made flexible by placing it in warm (140°F) water.
Modest levels of illumination, even from incandescent lamps, will prevent
ivory from yellowing, and it will gradually whiten ivory that has turned
yellow (see Illumination levels).
Mammoth ivory can be distinguished from elephant ivory by observing the
Schreger lines (cross-hatching that resembles engine turning) that are visible
in cross sections. If the Schreger lines intersect at less than 90°, it is
mammoth ivory; if they intersect at more than 90°, it is elephant ivory.
Unfortunately, many of the ivory parts used in musical instruments (such as
the facings of violin bows, the edgework of guitars, and piano key platings)
are too thin to identify by this method.
Ivory from African and Asian elephants can often be distinguished
visually: African ivory is somewhat translucent and often “ivory”-colored;
Asian ivory is more opaque and whiter. African ivory tends to be harder than
the Indian variety.

Ivory care at the Kunsthistorisches Museum in Vienna

The author received the following letter, from Dr. Sabine Haag, Curator, in
response to his inquiry regarding the cleaning of ivory at the
Kunsthistorisches Museum in Vienna:

Dear Mr. Pollens,

Thank you so much for your friendly words about our collection. Our
conservation department uses only simple methods for cleaning the
ivories: a cotton swab soaked in deionized water and ethanol (1:1) is
used first. If still needed, they [the conservators] use spit (enzymatic
cleansing). In this case thorough cleansing is required in order to avoid
later microbe attacks. Nonionic tensids are seldom used because they
hardly show equally good results as spit. To avoid a change in color, we
keep our ivories in a light spot, bleaching is not used any more for some
years. It was however done decades before. Due to the lack of
documents in our files we assume that in former times the conservators
(who had no academic training of course) used ammonia and hydrogen
peroxide. To make the surface more shining, our conservators polish the
object with brushes soaked with ethanol, continuing the process until the
ethanol has evaporated.
[October 27, 2004]

References
The Care of Ivory, Technical Notes on the Care of Art Objects no. 6 (Victoria
and Albert Museum, London; 1971).
H. J. Plenderleith and A. E. A. Werner, The Conservation of Antiquities and
Works of Art (London, 1979).
 J
Japanning

Japanning may be defined as the application of specially formulated

colored varnishes to metal and wood, often with the application of heat. The
varnish formulas can be as simple as shellac mixed with a pigment such as
ivory black, though most early formulations included asphaltum (used in
black japanning), resins (such as amber, colophony, copal, copaiba balsam), a
drying oil, and oil of turpentine. Ernest Spon, Workshop Receipts (1909)
gives the following formula for a “superior black Japan ground”:

amber 12 oz

asphaltum 2 oz

boiled oil ½ pint

resin 12 oz

oil of turpentine 16 oz

The amber, asphaltum, and resin were fused, the hot oil was added, and as
it cooled, the oil of turpentine was added.
After application of the varnish, the object was placed in an oven heated to
200–300°F (93–149°C) for several hours (termed “stoving”), which drives
off the solvent and fuses the resinous components so that they flow and
produce a glossy surface that hardens upon cooling.
Japanning was sometimes used to decorate the cast-iron plates and other
metal parts of nineteenth-century pianos. One decorative technique involved
embedding small pieces of mother-of-pearl (see Mother-of-pearl and abalone)
into the Japan varnish, either by cementing the pieces down to the underlying
surface prior to brushing on the varnish or by pressing the pieces down into
the wet or molten varnish. After the varnish cooled, it was ground down with
pumice and finally rottenstone to expose the mother-of-pearl pieces and to
level the surface. Gilding and painting were also used to further embellish
japanned surfaces.

Reference
Ernest Spon, Workshop Receipts for Manufacturers and Scientific Amateurs
(London, 1909).
 K
Keyboard instruments

See Clavichord maintenanceFortepiano maintenanceHarpsichord

maintenanceOrgan restorationPiano action regulation and voicing
(modern grand)Stringed-keyboard restorationTuning and
temperamentWire gauges for early keyboard instruments
 L
Labels

(see also see also paper, pencil, and ink)

One of the roles of a musical instrument conservator is to determine the
authenticity of manuscript and printed inscriptions in musical instruments. In
Europe, printing by movable metal type dates from around 1440 (before the
earliest known violins). It is easy to distinguish such labels from those
reproduced by lithography (invented in 1796) and the gravure process
(developed around 1850) because movable type generally leaves an
impression in the paper, whereas the later processes deposit the ink on the
surface of the paper without creating an indentation. Under magnification,
labels printed using halftone screening (a technique developed around 1830)
reveal a dotlike appearance under magnification, as is seen in newspapers and
magazines when viewed under a hand lens. Offset lithography (invented in
1875), which produces especially sharp and clean text, became the
predominant form of printing in the 1950s, though manual typesetting is still
used to a limited extent in fine printing.
Labels printed in the sixteenth through the twentieth century by movable
type were generally arranged in columns and rows on sheets of paper, so the
individual labels, which were cut apart after printing, sometimes exhibit
slightly different wordings and line breaks, damaged or inverted letters, etc.
This is especially true of the early labels printed in the sixteenth and
seventeenth centuries, whereas later printers seem to have worked with
greater care. Therefore, slight discrepancies between two labels of a maker do
not indicate that one of them is a fake.
In the eighteenth and nineteenth centuries, some labels were printed by the
engraving process. In this and other intaglio processes (such as
photogravure), the printing ink filled engraved or acid-etched lettering or
images on a flat plate, and the paper was pressed against the plate with
sufficient force to draw out the ink. As a result, the ink stands a bit above the
surface of the paper, and one can often feel it by running one's finger across
the printed surface (as on a freshly minted dollar bill).
Many printed labels have handwritten dates and other marks and
inscriptions added to them, such as the last one or two figures of the dates on
Stradivari's labels and the double score marks that Giuseppe Guarneri del
Gesù often added after the handwritten figures of the date. The earliest iron-
gall ink inscriptions were made using quill pens, while the steel-nibbed pen
came into wide use towards the end of the eighteenth century. Because of the
comparative flexibility of quill, downward strokes tend to broaden, while
steel nibs tend to produce a line of even width that is independent of stroke
direction – though special “calligraphic” steel nibs have been made that
imitate the variations in line width that are characteristic of quill. Steel points
sometimes produce furrows on either side of the ink line, whereas this is
generally not the case with quill. The ballpoint pen came into wide use
towards the end of World War Two. The felt-tipped pen was invented around
1964.
Typed labels are occasionally found in musical instruments. The typewriter
was developed in the early nineteenth century but did not come into wide use
until the latter part of that century. Most typewriters use inked cloth ribbons,
whereas later IBM Selectrics (the original model was introduced in 1961), as
well as some other models and brands, use plastic film ribbons coated with
fine carbon particles. Impressions made with cloth ribbons often show the
weave of the cloth, whereas those made with plastic film ribbons produce a
very clear impression free of the weave pattern. A new version of the IBM
Executive typewriter introduced around 1944 introduced proportional spacing
and had provision for justifying the right margin, whereas most typewriters
leave a ragged right margin.
Fake labels have been made with rubber stamps and by electroformed
metal plates, the latter creating depressions in the paper that are similar to
labels printed with metal type. In making a rubber or electroformed stamp,
the original is photographed, and as a result there may be a slight size
discrepancy between the original and the replica, which may be augmented or
offset by the vulcanization and electroforming processes. Size discrepancy
can be used to determine whether labels are authentic, but one needs an
authentic label for comparison.
 Labels have been reproduced with photocopy machines and by computer
scanning in conjunction with dot matrix, daisywheel, laser, and ink-jet
printers. They may even be printed by these means on old paper and thus be
difficult to detect (see Paper, pencil, and ink). However, each of these
printing techniques produces an idiosyncratic effect that can generally be
identified under a low-power stereo microscope or even a hand-held
magnifier. Under magnification, photocopied labels and those printed by
laser printers reveal melted particles caused by a heating element in the
copier or printer that fuses the particles to the paper. Under magnification,
characters formed by ink-jet printers can be seen to have been formed by tiny
dots of ink, and there is often a halo of dots around the perimeter of each
letter. Some laser printers print a barely visible array of tiny light-yellow dots
across the paper, a procedure which was instigated by the US government to
help distinguish counterfeit money.
One should become familiar with the various historical fonts and typefaces
used in label production. Labels used in nineteenth- and twentieth-century
commercial copies of Stradivari and other violins are often not facsimiles as
such and employ typefaces that can be easily recognized as anachronistic.
Many original labels are reproduced in Vannes (1951), Lütgendorff (1913),
and de Wit (1910).
To determine if a label has been in position for some length of time or is
new, place a drop of distilled or deionized water on a corner of the label, and
after the glue has softened, lift the corner of the label with a pair of tweezers.
If the label is original, the wood underneath should be cleaner and less
oxidized than the area around the perimeter of the label; in fact, there should
be a crisp demarcation where the underlying wood has been protected by the
label. In the early 1990s the author determined that the label on the
soundboard of a square piano attributed to Johann Socher (Germanisches
Nationalmuseum, Nuremberg) was glued in place long after the instrument
was made because scratches, dirt, and disfigurements on the soundboard were
covered by the undamaged label. This piano had long been thought to be the
earliest known square piano because of the date on the label, which is now
considered spurious (Pollens, 1995).

References
Willibald Leo Freiherrn von Lütgendorff, Die Geigen- und Lautenmacher
vom Mittelalter bis zur Gegenwart (Frankfurt, 1913).
Stewart Pollens, The Early Pianoforte (Cambridge, 1995).
Kenneth W. Rendell, Forging History: The Detection of Fake Letters and
Documents (Norman and London, 1994).
René Vannes, Dictionnaire Universel des Luthiers (Brussels, 1951).
Paul de Wit, Geigenzettel alter Meister vom 16. zur Mitte des 19.
Jahrhunderts (Leipzig, 1910).

Lacquer

Urushi is the Japanese term for the sap of the Rhus vernicifera (or
verniciflua) tree that grows in Japan, China, and Korea. Like the sap of
poison ivy (Rhus toxicodendron), it is a skin irritant in its uncured state.
Enzymes and carbohydrates in the sap, in conjunction with heat and
humidity, catalyze the cross-linking of urushiol, the main component of
urushi, resulting in an extremely hard and durable film that is highly resistant
to water, moisture, and many solvents. Nitric acid is one of the few agents
that can attack this remarkable material. Up through the eighteenth century,
European artisans were mystified by Asian lacquer and unsuccessfully
attempted to replicate it with conventional resins.
Lacquering is a complex art, and there are many ways to apply it. When it
is used to decorate wooden objects, the ground layer may consist of a
preliminary coating of lacquer followed by layers of silk or linen cloth
impregnated with a mixture of urushi and rice flour, which are applied as
reinforcement. After smoothing with whetstones, multiple layers of lacquer
are applied, including mixtures of urushi and coarsely pulverized
earthenware called jinoko, followed by applications of increasingly finer
grades of pulverized clay called tonoko mixed with urushi. Each application
of urushi must be allowed to dry before the next layer is applied. Drying
takes about three or four days, but this can be hastened if the object is placed
in a box or chamber that is kept warm, about 100°F (40° C) and at 70–80%
relative humidity. After drying, the layers are gone over with whetstones or
abrasives until a perfectly smooth surface has been created. Layers of colored
urushi are then applied, and after drying, the surface is polished with water
and magnolia wood charcoal, followed by finer grades of camellia wood
charcoal. Another layer of clear urushi is then applied with cotton wool, and
after thorough drying, it is polished with tonoko and finally with oil and
calcined deer's horn (called tsunoko). The final layer consists of diluted
urushi rubbed into the surface and wiped off with matted rice paper. This in
turn may also be polished with oil and calcined deer's horn. Again, it should
be noted that there are innumerable variations to the basic technique of
applying lacquer, and many methods of ornamentation, including the
application of gold (and other metals) in the form of leaf and powders,
inlaying shell and other decorative objects, and carving.
As lacquer ages, it becomes somewhat transparent and there may be subtle
color shifts. It can be irreversibly bleached (blacks turn brown, for example)
by exposure to strong light. Ultraviolet is especially destructive and causes
not only bleaching but loss of gloss and more serious surface deterioration.
Lacquered objects should not be subjected to UV or continuous illumination
above 200 lux (plus filtration to remove UV). Lacquer may become detached
from a wooden substrate if subjected to low humidity. To protect lacquered
objects, they should be stored at 60% relative humidity. Other forms of
deterioration include corrosion of inlaid metals, such as silver and copper.
Lacquered objects should preferably be handled with gloves; any fingerprints
must be wiped clean before placing the objects in storage or on display.
When cleaning lacquered wooden objects, solvents such as distilled or
deionized water, water with nonionic detergents, and mineral spirits are
relatively safe to use, though they should be tested first, especially on very
old lacquered pieces whose surfaces have lost their gloss and become
absorbent. Loose or detached lacquer can be reattached using natural urushi,
which is sometimes mixed with wheat or rice paste and fillers. Fine cracks
may be filled with multiple applications of urushi diluted with mineral spirits.
Surface gloss can be revived by polishing with camellia wood charcoal or
calcined deer horn and water and by rubbing with raw urushi.
If the conservator is not familiar with the technique of applying natural
lacquer, animal-hide glue or synthetic resin adhesives (such as Paraloid B-72)
may be used. In the past, molten wax–resin mixtures have been used to
consolidate flaking lacquer. This mixture was allowed to run under the loose
lacquer, which was pressed down with a heated spatula. One disadvantage to
using wax–resin mixtures to consolidate lacquer is that it interferes with the
curing of urushi, and thus it must be completely removed if urushi is used at
a later date for consolidating or rebuilding lost areas. Losses and cracks can
be built up using the natural lacquering technique, though pigmented wax or
wax–resin combinations can also be used to fill voids. One advantage to
using wax and wax–resin fills is that they can be leveled and polished using
solvents that will not affect the surrounding lacquer. Watercolor, acrylic
paint, or pigments mixed with synthetic resin varnish can be used for
retouching, though these invariably dry to a different surface luster and thus
must be overcoated with varnish, such as Paraloid B-72, to provide an even
gloss. In the past, surface gloss was sometimes revived by French-polishing
with shellac, but this too is not recommended. The pigments used in making
traditional lacquered objects include a black made by immersing finely
powdered iron in rice vinegar, as well as carbon black; reds include cinnabar
(mercuric sulfide) and iron oxide; orpiment (arsenic trisulphide) is used for
yellow.
While the restoration of classically lacquered objects with Western
materials and techniques corrupts their structure and alters their appearance,
it is difficult to achieve effective matches when filling and retouching with
urushi because it undergoes color changes as it cures and ages.

Western lacquer

Modern synthetic lacquer consisting of nitrocellulose resin dissolved in

solvents (such as a combination of acetone and toluene) was developed in the
1920s. Nitrocellulose lacquer is generally sprayed on in very thin coats,
though it can also be brushed on or applied by dipping. Because of the use of
highly volatile solvents, lacquer dries very quickly. Modern lacquers are also
formulated with acrylic resins.

References
N. S. Brommelle and Perry Smith, eds., Urushi (Marina del Rey, 1988).
John S. Mills, Perry Smith, and Kazuo Yamasaki, eds., The Conservation of
Far Eastern Art (London, 1988).
Ernest Spon, Workshop Receipts for Manufacturers and Scientific Amateurs
(London, 1909).

Lake pigments

Lake pigments are made by extracting colorants from various organic

substances (such as madder root, buckthorn berries, kermes, and cochineal)
by soaking, heating, or boiling in water, combining with a mordant (such as
alum), and precipitating the pigment from solution. Lakes typically have a
lower refractive index than earth pigments and are therefore somewhat
transparent in most varnishes. Because of the high cost of manufacturing true
lakes, most artists' suppliers have stopped producing them and have
substituted synthetic pigments that have similar color but lack the
transparency. One such lake, Roberson's orange, was much prized by violin
restorers for retouching varnish, but unfortunately it is no longer
commercially available.

To make an orange lake

Add 1 pound of crushed Persian berries (ripe buckthorn berries) to 1 gallon

of distilled or deionized water, boil to extract the colorant, filter out the
berries, and add 4 ounces of tin chloride (“muriate of tin,” which was
prepared by dissolving tin in muriatic acid, was once used). Then slowly add
sodium carbonate to precipitate the lake pigment. Filter the precipitated
pigment through fine cloth or filter paper, wash with distilled or deionized
water, and then dry.

To make carmine lake

Boil 1 pound of cochineal in 1 gallon of water, add 4 ounces of alum

(potassium aluminum sulfate), decant the clear liquid, then slowly add cream
of tartar (potassium tartrate) or potassium carbonate to precipitate the lake
pigment. Filter the precipitated pigment, wash with distilled or deionized
water, and dry.

To make madder lake

 Add 1 pound of ground madder root to 1 gallon of distilled or deionized
water. This can be left to stand for a day or so to set up a fermentation
process which helps release the colorant. The resultant liquor can be warmed,
but should not be boiled. Filter out the madder root, and add 4 ounces of
alum. Then slowly add sodium carbonate to precipitate the lake pigment.
Filter the precipitated pigment, wash with distilled or deionized water, and
dry. If tin chloride is used instead of alum, the madder lake will have a warm
hue instead of a purplish color.

References
George Hurst, Painters' Colours, Oils, and Varnishes: A Practical Manual
(London, 1901).
A. P. Laurie, The Materials of the Painter's Craft in Europe and Egypt from
Earliest Times to the End of the XVIIth Century (London, 1910).
Jean René Denis Riffault, Armand Denis Vergnaud, and Claude Jacques
Toussaint, A Practical Treatise on the Manufacture of Colors for Painting
(Philadelphia, 1874).
Pierre François Tingry, The Painter and Varnisher's Guide (London, 1804).

Lead and lead alloys

(see also Organ restoration, Leather)

Gray corrosion products on lead and lead/tin alloys (such as organ pipes)
are generally oxides and carbonates, which form a protective layer that
retards further oxidation. Lead is attacked by organic acids, such as acetic,
formic, and tannic acids, as well as by strong alkalis. For this reason, lead
objects should not be stored in wooden cabinets or come in contact with
tanned leather (this is why organ bellows were traditionally made with tawed
rather than tanned skins). Oxides, carbonates, and sulfides can be removed by
a quick immersion in 10% hydrochloric acid. Another method is to immerse
the object in a 5% solution of disodium EDTA for up to several hours. Either
treatment must be followed by a thorough rinsing with tap water. Distilled
and deionized water are not recommended for the prolonged rinsing or
washing of lead or lead-alloy objects (including those that are soldered with
tin/lead solder) because water of such purity readily absorbs carbon dioxide
from the air and becomes acidic; the resulting acid can attack lead and soft
solders.
Electrolytic reduction is not recommended because the alkaline electrolytes
that are generally employed are difficult to remove from lead with a simple
water wash and must be neutralized with an acid.

References
Donny L. Hamilton, Methods of Conserving Archaeological Material from
Underwater Sites (www.nautarch.tamu.edu, 1999).
Colin Pearson, Conservation of Marine Archaeological Objects (London,
1987).
H. J. Plenderleith and A. E. A. Werner, The Conservation of Antiquities and
Works of Art (Oxford, 1979).
T. Stambolov, The Corrosion and Conservation of Metallic Antiquities and
Works of Art: A Preliminary Survey (Amsterdam, n.d.).

Leather

(see also Parchment, vellum, and slunk)

Before modern procedures were developed, the skins of small or young
animals, birds, and reptiles, and the hides of larger animals were converted
into leather either by vegetable tanning, oil curing (chamoising), or tawing (a
curing technique using alum). Modern mineral tanning includes the use of
titanium, zirconium, and chromium compounds. These materials and
processing techniques have largely replaced earlier methods because of their
speed – skins and hides can now be converted into leather in a matter of
hours rather than days or weeks. Unfortunately the finished products rarely
duplicate the qualities of softness, resilience, and flexibility that were
imparted by traditional methods. The difficulty in locating sources of
traditionally tanned and cured leathers presents problems for musical
instrument restorers, because the tonal and playing qualities of certain
instruments are affected by the quality of various leathers used in their
construction (for example the leathers used in the hammer and damper
mechanisms of late eighteenth- and early nineteenth-century fortepianos, the
peau de buffle plectra of eighteenth-century French harpsichords, and the
purse pads of early nineteenth-century woodwind instruments).
In traditional tanning, fresh skins and hides were first made pliable by
soaking for several hours in cold water that was often “softened” by the
addition of soap and borax (bran and salt were sometimes added). Sulfuric
acid was added to this soaking liquid if the hair was to be retained; if it was to
be removed, the skins were soaked for several days in a liquor made with
cold water, slaked lime, and hardwood ashes (lye). The hair was then
removed by placing the skins on a fleshing beam and scraping with a fleshing
knife. The skins could then be treated by a variety of methods: vegetable
tanning involved soaking in a preparation of oak and chestnut bark or the
leaves and panicles of sumac (which contain tannins); oil curing involved the
application of plant or animal oils (such as cod oil) or brains; and tawing
employed alum or salt. After conversion into leather, skins and hides could
be stretched, beaten, rolled, or boarded to adjust their flexibility, as well as
“curried” with cod liver oil or tallow, or “stuffed” with egg yolk, flour, oil, or
grease to impart strength, water resistance, pliancy, or body. So-called
“rawhide” is not “tanned” as such, but is simply soaked in lime and stretched
while drying.
The species of animal from which leathers derive can now be identified by
antibody reaction and DNA analysis, while the tanning processes can be
determined by various forms of chemical analysis. However, most of the
commonly used leathers can be roughly characterized by collecting samples
from leather merchants and learning to recognize their grain and follicle
patterns (see references below).

Causes of deterioration

Atmospheric sulfur dioxide caused by the burning of oil and gas lamps and
industrial pollution have been the major causes of “red rot” (a powdery
condition) in vegetable-tanned leather. Airborne sulfur dioxide should not
exceed 0.06 parts per billion in museum galleries and storage areas. Oxygen,
ozone, and nitrogen dioxide are also responsible for the deterioration of
leather. Excessive heat and low humidity (below 40%) dry out leather and
render it inflexible, and excessive levels of illumination and ultraviolet (see
Illumination levels) cause dyes and natural coloration to fade. Leather should
be kept at 70°F (18°C) and 50% humidity.

Cleaning leather

Vegetable-tanned leather, such as sole leather and split skins, may be dry-
cleaned with hexanes, 1,1,1-trichloroethane, or 2% potassium methyl
cyclohexyloleate soap (such as Vulpex) in mineral spirits. If potassium oleate
soap is used, it must be cleared afterwards with multiple applications of the
solvent. Saddle soap can also be used to clean tanned leather that has a
smooth surface. It consists of a mixture of conventional soap and emulsified
oils that is worked up into a lather and applied with a cloth pad or sponge;
after the excess is removed with a moist cloth, the leather surface is buffed
with a soft dry cloth.
Tawed (alum- and salt-cured) leathers, which are generally white in color
and very supple, should never be cleaned with or immersed in water because
it will dissolve the alum or salts. To remove superficial dirt, tawed skins may
be lightly sponged with hexanes, 1,1,1-trichloroethane, or 2% potassium
oleate soap in mineral spirits followed by clearing with mineral spirits.
Chromium-tanned skins and “white leather” made by the
zirconium/aldehyde process can be lightly sponged with soap and water, and
then cleared with fresh water, or 2% potassium oleate soap in mineral spirits
cleared with fresh mineral spirits.
If “red rot” has not begun, the leather may be protected by treating with a
solution of potassium lactate. Because the working solution is unstable, it
should be diluted (1:9) from a stock solution of 50% potassium lactate just
prior to use. Distilled or deionized water should be used to prepare both the
stock and the working solutions. The working solution is lightly sponged on.
Before cleaning gilded or painted leather, first test any gilding, painting,
and coatings for sensitivity to solvents.

Dressings for tanned leather

 The traditional leather dressings listed below have long been used to
restore the flexibility of old leather as well as to improve its appearance.
Unfortunately, they tend to darken leather, and if applied too liberally or too
often, their oily components can become sticky and attract dust. There is
growing controversy about the use of leather dressings, and it may be
preferable to leave the leather parts of early musical instruments untreated. In
some cases, the stiffness that leather acquires through age renders it tougher
and thus safer to handle and mount for exhibition purposes. The hammer
coverings and dampers of fortepianos should never be oiled or treated with
the dressings listed below; light brushing with a bristle or soft wire brush
(available from jewelers' suppliers) will loosen encrusted dust and dirt and
may revive its texture.

British Museum leather dressing

200 g lanolin (anhydrous)

30 ml cedarwood oil
15 g beeswax
330 ml hexanes

Apply sparingly with soft cloth. Buff with a soft cloth or brush after a few
days.
Substitute ceresin for beeswax if used in a warm climate.

Library of Congress leather dressing

40% lanolin (warmed)

60% neatsfoot oil

Apply sparingly with a soft cloth. Buff with a soft cloth or brush after a few
days.

Smithsonian glycerin treatment

 59% glycerin
39% water
1% formaldehyde or Dowicide 1

or

25% glycerin
75% ethyl alcohol

Note: glycerin restores flexibility to brittle leather but supports mold growth
(hence the addition of the disinfectant Dowicide 1) and attracts dirt.

Bavon ASAK ABP

This is a leather lubricant; not sticky like the British Museum or Library of
Congress dressings.
Use 10% in either mineral spirits or 1,1,1-trichloroethane. Up to ten coats
can be applied with a brush to leather at fifteen-minute intervals.

Traditional paste wash for leather book bindings

Lightly pad wheat or rice paste (see Glues, pastes, and other adhesives)
over the surface and allow to dry. This removes superficial dirt and adds
gloss.

Microcrystalline wax

Microcrystalline wax (such as Cosmoloid H80, Victory wax, or

Renaissance wax) may be lightly applied to leather to improve its appearance.

To prevent red rot

Prepare a 50% stock solution of potassium lactate; dilute the stock solution
1:9 with deionized water to produce a working solution, and sponge this on
both sides of the leather. If red rot has begun, this treatment is ineffective.
Leather that has badly deteriorated due to this condition can be consolidated
by applying several coats of hydroxylpropylcellulose (such as Klucel or
Cellugel), 10% in ethyl alcohol.

Adhesives for leather

Animal-hide glue was traditionally used in attaching leather to wood (such

as hinges, gussets of bellows, fortepiano hammer leathers and dampers, etc.).
To prevent glue from being absorbed by leather, adjust its viscosity either by
using less water when making up the glue or by heating the glue to drive off
water.

Consolidant for powdery leather

A 2% solution of Klucel or Cellugel in ethanol or 2-propanol can be

applied to the surface of leather to consolidate its surface. Klucel G is a
medium-viscosity consolidant; Klucel E has lower viscosity and penetrates
deeper into the leather.

References
N. R. Briggs, The American Tanner (New York, 1802).
C. N. Calnan, Fungicides Used on Leather (Northampton, 1985).
James Jackman, ed., Leather Conservation: A Current Survey (London,
1982).
Marion Kite and Roy Thomson, Conservation of Leather and Related
Materials (Oxford, 2007).
Ernest Spon, Workshop Receipts for the use of Manufacturers, Mechanics,
and Scientific Amateurs (London, 1909).
D. H. Tuck, Oils and Lubricants Used on Leather (Northhampton, 1983).
John Waterer, John Waterer's Guide to Leather Conservation and
 Restoration (London, 1972).

Lubricants

Lubricants create a slippery barrier between mechanical parts that enable

them to work smoothly and with less wear. There are three basic types of
traditional oils: mineral oils derived from the fractional distillation of
petroleum, vegetable oils (often from seeds and nuts), and animal oils
(produced by rendering animal fat or extracted from the head cavities of
certain whales, dolphins, and porpoises). So-called “neutral oils” are light-
grade mineral oils derived from the distillation process without cracking; they
do not emulsify when they come in contact with water. Mineral oils come in
several basic viscosities: light, medium, and heavy. The grades available in
pharmacies are generally highly refined, stable, and adequate for lubricating
the valves and slides of brass instruments (see Brass instruments). Vegetable
oils are greasier and oilier than mineral oils, but they oxidize and are acidic.
Animal oils are also greasier than mineral oils, acidic, and prone to rancidity.
Acidic oils will slowly attack metals, such as brass. Lubricating greases are
often made by compounding mineral oil with soda or lime soaps. Animal fat
and lard, sometimes thickened with rosin, wax, or talc, have also been used as
greases. Petroleum jelly (also called petrolatum; Vaseline is the trade name
for a highly refined petroleum jelly used medicinally as an ointment), which
is derived from the fractional distillation of petroleum, is a stable material
that can be used as a grease.
Synthetic oils are more stable and reduce friction better than the natural
oils. Though expensive, watch- and clockmakers' synthetic and partially
synthetic oils made by the Moebius, Nye, and Tillwich companies are
recommended.
Silicone oils and greases are noted for their stability at extremely low or
high temperatures (−50° to 500°F; −46° to 260° C), which is generally not
required of musical instruments. However, they are incompatible with
mineral oils and greases and are insoluble in petroleum solvents, so they
cannot be recommended in conservation work.
Oils and greases used to lubricate mechanical parts of musical instruments
should be periodically cleaned off with the appropriate solvent and replaced
with fresh lubricant. Volatile oils such as kerosene (which are sometimes
used to lubricate trombone slides) quickly evaporate, thereby leaving metal
parts without lubrication and likely to seize.

Reference
George S. Brady, Henry R. Clauser, and John A. Vaccari, Materials
Handbook (New York, 1997).

Lute stringing

(see also Gut strings, Mersenne's Law)

Thomas Mace (1676) presents a fascinating discussion of lute strings in his
Musick's Monument published in 1676:

The first thing you are to consider, is the Size of your Lute; 2ly. The
Substance and Strength of it.
And as to the Size, if it be a Large Lute, it must have the Rounder
Strings; and a Small Lute, the Smaller.
Then again (as to the Substance) if it be a Strong firm-made Lute, it may
bear the Thicker Strings; but if Weak and Crazy, then the Smaller
Strings.
Yet I rather advise to String it, according to the Size, than the Strength,
&c.
First, Because a Lute has more Natural Right done it, and will return
you, more Acceptable Content, in token of Its Gratefulness.
2ly. Because a Lute that is Crazy and Weak, may have Ease done it, in
setting it at a Lower Pitch, (if you see cause) sometimes.
But if you be to use your Lute in Consort, then you must String it, with
such siz'd Strings, so as it may be Plump, and Full Sounded, that it may
bear up, and be heard, equal with the other Instruments, or else you do
little to the purpose …
The first and Chief Thing is, to be careful to get Good Strings, which
would be of three sorts, viz. Minikins, Venice-Catlins, and Lyons (for
Basses). There is another sort of Strings, which they call Pistoy Basses,
which I conceive are none other than Thick Venice-Catlins, which are
commonly Dyed, with a deep dark red colour. They are indeed the very
Best, for the Basses, being smooth and well-twisted Strings, but are hard
to come by; however, out of a Good parcel of Lyon Strings, you may
(with care) pick those which will serve very well.
And out of these three sorts, First chuse for your Trebles, 2nds, 3rds,
and some of your small Octaves (especially the sixth) out of your
Minikins.
Then out of your Venice-Catlins, for your 4ths, 5ths, and most of your
other Octaves.
Your Pistoys, or Lyons, only for the Great Basses.
There is a small sort of Lyons, which many use, for the Octaves; but I
care not for Them, they being constantly Rotten, and good for little, but
to make Frets of.
Now that you may know, all these strings, and also how to know Good,
from Bad, take these following Observations.
First know, that Minikins are made up always, in long-thin-small Knots,
and 60 are to be in a Bundle.
Venice-Catlins are made up, in short double Knots, and 30 doubles in a
Bundle.
Both which, are (generally) at the same Price, and the signs of
Goodness, both the same; which are, first the Clearness of the String to
the Eye, the Smoothness, and Stiffness to the Finger, and if they have
Those two qualities, dispute their Goodness no further.
The Lyon String, is made up in a double Knot; but as Long as the
Minikin.
They are sold (commonly) by the Dozens, and not made up in Bundles.
Their Goodness may be perceiv'd, as were the other; But they are much
more Inferiour Strings, than the other.
I have sometimes seen Strings of a Yellowish Colour, very Good; yet,
 but seldom; for that Colour is a general sign of Rottenness, or of the
decay of the String.
There are several Sorts of Coloured Strings, very Good; But the Best (to
my observation) was always the clear Blue; the Red, commonly Rotten;
sometimes Green, very good.
As concerning the keeping of your Strings, you must know, there ought
to be a Choice Care taken; for they may be very Good when you buy
them, but spoiled in a quarter of an hours time, if they take any wet, or
moist Air. Therefore your best way is, to wrap them up close, either in
an Oyl'd Paper, a Bladder, or a piece of Sear-cloath, such as often
comes over with Them, which you may (haply) procure, of them who
sell your Strings: Yet they are not very willing to part with it, except
they sell a Good quantity of Strings together.

See Gut strings for a discussion of the distinctions between Minikins, Venice-
Catlins, Lyons, and Pistoy Basses.
To compute the maximum safe pitch (about a semitone below the breaking
point) to which a gut-strung instrument can be tuned, divide 240,000 by the
string length in millimeters. Thus, the top string of a lute having a string
length of 580 mm can be safely brought up to about 413.8 Hz, which would
allow the top G string of a Renaissance-style lute to be safely tuned to 392
Hz, the pitch of g1 at A=440 Hz.
As with metal strings, the mass and strength of gut strings are both
proportional to the cross-sectional area of the string, so simply increasing the
diameter of a gut string does not increase the breaking pitch (see Scaling).
However, unlike plain metal wire strings, gut strings are made of multiple
filaments that are twisted together, and the degree of twist affects both the
strength and the flexibility of the string: low-twist gut strings have the
greatest strength (tensile strength around 392 MPa) though the least
flexibility, while high-twist strings have decreased strength (tensile strength
around 294 MPa) but increased flexibility. Thus, the breaking pitch varies
with the twist rate. With instruments such as the lute, mandolin, guitar, viol,
and violin, only the top string is tuned close to its breaking point, and as one
descends to the bass, the strings are tuned further away from their breaking
point, though they are made thicker to provide similar tension from string to
string. Thick, slack strings have a poor tone quality and tend to play out of
tune when fretted in higher positions. A higher twist rate, which renders them
more flexible, alleviates these defects.
Marin Mersenne (1636) advocated increasing or decreasing the diameter of
lute strings in proportion to the pitch; that is, a string a fourth lower should be
4/3 times thicker (see Gut strings for calculations of string diameters), which
yields equal tension among the strings. Some modern-day lutenists feel that
this results in bass strings that are too thick (if one calculates backwards from
the top string), so they use thinner strings than those considered ideal by
Mersenne (see table below). This results in a gradual reduction in tension as
one proceeds from treble to bass.

Mersenne's system A typical setup used today*

g′ 0.40 mm 0.40 mm

d′ 0.53 mm 0.48 mm

a 0.71 mm 0.60 mm

f 0.89 mm 0.74 mm

c/c′ 1.19 mm/0.60 mm 0.98 mm/0.50 mm

G/g 1.58 mm/0.79 mm 1.28 mm/0.66 mm

F/f 1.77 mm/0.89 mm 1.36 mm/0.74 mm

D/d 2.12 mm/1.06 mm 1.50 mm/0.88 mm

* As recommended by Olav Chris Henriksen of Boston Catlines for the author's eight-course
lute having a 630 mm string length.

For an eleven-course Baroque lute having a 700 mm string length, the

following string diameters are recommended:

f1 0.42 mm

d1 0.50 mm

a 0.58 mm

f 0.72 mm

d 0.86 mm

A/a 1.20 mm/0.56 mm

G/g 1.32 mm/0.64 mm

F/f 1.46 mm/0.64 mm

E/e 1.6 mm/0.76 mm

D/d 1.8 mm/0.86 mm

C/c 1.96 mm/0.96 mm

Methods of tying gut strings to a lute bridge are illustrated in Figure 19.
Figure 19 Two methods of tying lute strings to the bridge.

References
Thomas Mace, Musick's Monument; or Remembrancer of the Best Practical
Musick, both Divine, and Civil, That Has Ever Been Known, to Have Been
in the World (London, 1676).
Marin Mersenne, Harmonie universelle (Paris, 1636).
 M
Measurement

(see-also Metalworking, Woodworking, Proportion, Historic

metrology)
When studying, cataloging, restoring, and copying musical instruments, it
is often necessary to make precise and accurate measurements. Precision
refers to the degree of exactness or refinement of a measurement, while
accuracy refers to how well the measurement conforms to an accepted
standard. Precision is a relative term: for example, measuring the length of a
grand piano to the nearest millimeter would be considered a precise
measurement, but measuring the diameter of one of its strings to the nearest
millimeter would be considered imprecise. Accuracy depends upon the
quality of the measuring device and whether it is appropriate to the task: for
example, a micrometer or caliper might come with a manufacturer's
certificate indicating that its accuracy has been checked against standards that
are traceable to the National Bureau (or Institute) of Standards, or a set of
gauge blocks might be certified from ASME Grade 2 (workshop tolerance)
up to Grade 00 (calibration tolerance). A tape measure might be accurate, but
it is inappropriate for measuring the diameter of a drill bit, for example. Other
considerations are repeatability (whether a measuring instrument provides the
same results if the same object is measured over and over) and discrimination
(which is the finest unit that a measuring tool can measure).
While the measuring of musical instruments contributes to our
understanding of them, we should consider how early examples were made
and what measuring equipment their makers had at their disposal. For
example, while today's violin makers typically adjust the thicknesses of their
top and back plates to 0.1 mm with a deep-engagement dial caliper, it is
unlikely that early makers exercised such a fine control.
A number of museums prohibit the use of metal measuring tools when
examining musical instruments in their collections. Plastic rulers and calipers
are generally sufficiently accurate for most work, though metal measuring
instruments can often be made less dangerous by gently rounding off sharp
corners and edges that might come in contact with instruments. For example,
a strip of crocus cloth can be pulled across the corners and edges of the jaws
of a vernier caliper without impairing the accuracy of its measuring surfaces.
Similarly, the edges of a micrometer's anvil and spindle can be beveled with a
sharpening stone. Electronic calipers and micrometers that can be readily
“zeroed” with the push of a button can have their jaws, anvils, and spindles
lined with plastic tape.

Traditional measuring tools

Many of the traditional measuring tools used in musical instrument work

are shown in Figure 20.

Figure 20 Measuring instruments, including calipers, dial indicator,

Hacklinger gauge, gram-force gauge, set of small-hole gauges, vernier
protractor, universal bevel, screw pitch gauge, wire and string gauges,
micrometer, and dial caliper.

Dress maker's cloth measuring tape

While not the last word in accuracy, cloth measuring tapes have been used
by generations of violin dealers for measuring body lengths and widths over
the arching. Such measurements are slightly inflated when compared to those
made around the arching with a caliper, which should be taken into
consideration when consulting measurements found in old certificates.

Steel tape measure

This ubiquitous measuring tool is available in different lengths and in both

the inch and metric system. Steel tape measures generally have a hook that
can be pushed in to make inside measurements or pulled out to make outside
measurements. Tape measures are rarely accurate to their finest gradation or
over their entire length and should be checked against more accurate
measuring devices, such as a steel rule (see below). The hook can be adjusted
to provide a more accurate end point by filing it down or attaching shims to
it.

Steel rule

These are generally made of tempered stainless steel and have accurately
engraved scales. They come in various lengths and widths and in rigid and
flexible types. The smaller sizes are sometimes fitted with a hook, which is
useful when making outside measurements of objects, but it may interfere
with measuring the length of a string from the middle of a bridge or nut pin.
The finest graduations of steel rules are generally 1/64 in. and 0.5 mm. Every
restorer should have a basic set consisting of 6 in./150 mm, 12 in./300 mm,
18 in./450 mm, 24 in./600 mm, 36 in./900 mm, and 48 in./1200 mm.

Dividers and calipers

These are not measuring instruments as such, but are used to gather
dimensions from objects and are then held up against a steel rule or other
graduated device to ascertain the measurement. Dividers generally have
sharp, straight points, while most calipers have rounded, curved ends that are
used to take dimensions from the inside or outside of an object. The best type
is the so-called “spring” divider or caliper, which has legs that are held in
position by a strong bow spring and are adjusted by turning a knurled nut. So-
called “firm-joint” dividers and calipers have legs that pivot from a friction
joint and are simply collapsed or expanded to fit against or within an object.
Hermaphrodite calipers have one pointed foot and one curved foot (the latter
can be rotated to take inside or outside dimensions) and are used to measure
or scribe lines from an edge. These are usually available with firm-joint
construction. Spring-bow dividers and calipers come in a wide variety of
sizes up to 12 inches, while firm-joint calipers are made up to 36 inches. A
beam compass or a pair of trammel heads with interchangeable points fitted
to a straightedge can be used to take dimensions of larger work. One of the
most useful calipers is the “transfer” caliper, which is equipped with a short
locking arm that rotates with one of the legs. After the legs have been
positioned around an object and tightened, the leg attached to the locking arm
is temporarily released so that the caliper can be removed or extracted from
the object; the leg is then returned to the position it was in when the
dimension was taken and is locked in place.

Measuring caliper

In the past, most were equipped with vernier or dial gauges, but now many
are available with electronic digital readouts, which are much easier to read
and provide measurements in both inches and metric units. The 6 in./150 mm
size is the most common and is generally equipped with long jaws for
measuring outside dimensions, as well as a shorter pair of blade-shaped jaws
for making inside measurements and a rod or beam that extends for use as a
depth gauge. For making accurate measurements of the lengths of violins and
violas, an 18 in./450 mm or 24 in./600 mm size is useful. Larger-size calipers
are available, but in musical instrument work, it is generally unnecessary to
measure very large objects with the precision that a mechanical or electronic
caliper provides (0.001 in. or 0.02 mm).

Micrometers

Most mechanical micrometers are equipped with divisions engraved on

both the fixed sleeve and rotating thimble for measuring 0.001 in. or 0.01
mm. Some micrometers are also equipped with vernier scales on their
sleeves, which provide measurements to 0.0001 in. or 0.002 mm. Electronic
digital micrometers are easier to read, though they are generally much bulkier
than the mechanical type. Micrometers have a more limited range than
calipers but are better suited to making accurate thickness measurements of
wire and gut strings. They come in a wide range of sizes, the most common
being 0–1 in. (0–25 mm), 1–2 in. (25–50 mm), and 2–3 in. (50–75 mm).
Larger ones are available, but in those sizes the precision they afford is
rarely, if ever, required in musical instrument work. The standard micrometer
has flat measuring surfaces (termed the anvil and spindle), though other types
are available, including micrometers with rounded anvils and/or spindles (for
measuring curved objects), those shaped like blades (for measuring slots and
keyways), discs (for measuring narrow grooves or recesses), rods (for
measuring the thickness of tubing and cylinder walls), and points (for small
or inaccessible areas), as well as with V-shaped anvils (for measuring the out-
of-roundness of cylindrical rods and tubes, or the diameters of cylindrical
objects with interrupted surfaces, such as three-flute cutting tools). Multi-
anvil micrometers have interchangeable anvils for accommodating different
surfaces. Screw-thread micrometers with interchangeable anvils are used to
determine the pitch diameter of threads. A standard micrometer can also be
used to measure pitch diameter by employing the three-wire thread measuring
system (see below). Micrometer heads (consisting of sleeve, thimble, and
spindle, but lacking a frame and anvil) can be adapted to various devices
(such as microscope stages) for precise movement and positioning, as well as
for measurement.
The accuracy of micrometer readings can be affected by the force exerted
on the thimble by the person making the measurement. To make micrometers
less subject to inconsistencies due to handling, most are fitted with either
ratchet or friction thimbles, which control to some extent the degree to which
the thimble can be tightened against the object being measured. The friction
thimble is generally considered a more sensitive device and is always fitted to
micrometers equipped with verniers measuring to 0.0001 in. and 0.001 mm.
Even with ratchet or friction thimbles, the speed at which the micrometer is
tightened will still affect the measurement. To develop a reliable technique,
novices should practice measuring gauge blocks or objects of known
dimension to develop a “feel” for their micrometer.
One advantage of electronic micrometers and calipers is their ability to
transmit measurements directly to computers for storage or arithmetical
manipulation in spreadsheets, CAD (computer-aided design), and other
programs. Electronic digital scales, which are essentially electronic digital
calipers without jaws, can be attached to various devices (such as X-Y tables
and microscope stages) for positioning and measurement.

Depth gauges

These are available with micrometer and vernier mechanisms, as well as

with dial and digital readout. The micrometer type is generally equipped with
interchangeable steel rods for extending the measuring depth to 12 in./300
mm and greater. The vernier type generally measures directly to 12 in./300
mm.

Height gauges

These are available in vernier, dial, and digital type. 12 in./300 mm and 24
in./600 mm are the most useful sizes.

Deep-engagement dial calipers

These are very useful for measuring violin plate and soundboard thickness.
They are made in various sizes with throats up to around 400 mm for
measuring the plate thicknesses of double basses. One type is constructed
with a very deep frame that enables the caliper to reach over and around cello
ribs.

Reinert caliper

This device is shaped in such a way that one of its legs can pass through
violin f-holes to take thickness measurements of violin tops. It has a limited
reach and precision and has been supplanted by the Hacklinger thickness
gauge.

Hacklinger thickness gauge

This pen-shaped measuring device employs a small but powerful free

magnet that is dropped through a violin's f-hole or into the body of an
instrument through an opening. Another magnet mounted inside the gauge is
used to drag the free magnet around the interior surface of a violin's belly or
instrument soundboard. A spring-loaded marker is lifted against a calibrated
scale, and a clicking sound occurs when the marker reaches a point on the
scale that coincides with the distance between the free magnet and the cloth-
lined contact surface of the gauge positioned on the outer surface of the
instrument. This device measures to 0.1 mm. Several versions are available
that offer measuring ranges up to 12 mm in thickness. New electronic
thickness gauges have been recently introduced; though they are a
convenience and can be used in conjunction with thickness mapping
software, their readings may be affected by temperature fluctuations, so they
must be carefully calibrated before use.

Small-hole gauges and telescoping gauges

Like dividers and calipers, these do not measure directly but are used to
take dimensions of the insides of circular recesses (such as the bores of
woodwind and brass instruments) and then transfer them to micrometers or
vernier calipers for measurement. Small-hole gauges are available singly or
in sets, which range from 0.125–0.500 in. (3.2–12.7 mm); sets of telescoping
gauges generally range from about 0.5–6 in. (12.7–152 mm). Extension
handles can be fitted that allow dimensions to be taken deep within the bores
of wind instruments. The measuring surfaces of small-hole gauges are
adjusted with a knurled knob, while telescoping gauges have rods with
rounded ends that extend under spring pressure and are then locked in place
with a knurled knob.

Dial test indicators

These are extremely precise gauges (sometimes measuring to 0.00005 in.

or 0.001 mm), generally with a limited measuring range, that are fitted to
adjustable bases for use as comparators, for checking lathe alignment, and
out-of-roundness, as well as the dimensions of objects that are made on
lathes, milling machines, and other machinery.
Taper and feeler gauges

These consist of either a tapered feeler with graduations on one surface or

sets of strips of steel of different thicknesses that are used to measure
clearances or spaces between closely fitted objects.

Screw gauges

Screws have four basic measurements: the major diameter, which is the
outer diameter of the thread; the minor diameter, which is measured at the
roots of the thread; the pitch diameter, which is an imaginary intermediate
diameter located at a point where the thread and the space between the
threads are the same width; and the pitch, which is the distance between
points of the threads measured parallel to the longitudinal axis of the screw.
The major diameter must be determined in order to select cylindrical stock
for threading or when drilling holes for the free passage of threaded parts; the
minor diameter must be determined in order to drill holes that will be tapped;
the pitch is required to select the correct tap or die or to set up the change
gears of a machine lathe if one is turning the threads. There are several
methods of determining pitch and diameters: screw pitch gauges are used to
determine the number of threads per inch or the spacing of threads in
millimeters of both inside and outside threads; screw-checker gauges are used
to determine the pitch as well as the size of outside threads; plug gauges are
used to measure the pitch, size, and degree of fit of inside threads; while
screw-thread micrometers with interchangeable anvils are used to measure
the pitch diameter of threads (this diameter can also be measured by the
three-wire system, which uses sets of measuring wires in conjunction with an
ordinary outside micrometer).

Protractor and vernier protractor

A protractor is a device for measuring angles. The protractor that comes

with a combination-square set is very useful for measuring and transferring
angles. Vernier protractors with engraved scales and vernier (or with dial or
electronic digital readouts) are used for more accurate measurements. Vernier
protractors are generally graduated in degrees (°) or 5 minutes of arc.
Electronic protractors resolve to 0.01°. Some vernier and electronic
protractors come with an acute-angle measuring attachment, which is very
useful for measuring the angle between nonparallel surfaces and the taper of
shafts of tuning pegs.

Bevel and universal bevel

The bevel is a simple device for transferring angles to a measuring device

such as the protractor. The universal protractor has an offset slot that permits
transfer of acute angles of tapered surfaces and shafts.

Machinist's squares

These are made of precision-ground hardened steel and are generally more
accurate than woodworker's squares. They are invaluable when squaring-up
work and setting up machinery. Every shop should have several sizes ranging
up to 6 in. (150 mm).

Measuring magnifiers and optical comparators

These are useful when objects cannot be measured directly with a

micrometer, caliper, or protractor. Small optical comparators with
interchangeable reticles are available for both inch and metric measurement.

Force gauges

Small, mechanical force gauges (sometimes called tension or compression

gauges) are useful for recording or adjusting key pressure on keyboard
instruments. These are calibrated in gram-force or centi-newtons and come in
different measuring ranges. Those that measure up to 150 grams-force are
useful for early keyboard instruments such as clavichords, lightly voiced
harpsichords, and early fortepianos; modern pianos, some double-manual
harpsichords, and tracker-action organs require scales measuring up to 250
grams-force or even greater. Electronic force gauges are also available, but
they are rather expensive.
Lucchi meter

This electronic device measures the speed of sound in wood, which is

proportional to its stiffness. It is used by violin and bow makers for selecting
wood, as it is widely believed that stiffer wood produces better-sounding
stringed instruments and better-playing bows.

Acoustical measurement (see also Acoustics)

An accurate and stable electronic tuning device should be on hand for

measuring and setting pitch. The digital frequency counter is another device
that can be used to determine pitch, though it is only effective with tones
having a steady pitch because it must accumulate vibrations over a period of
one or more seconds. A variable audio oscillator used in conjunction with a
frequency counter can be used to generate tones of any desired pitch, which
is useful for identifying resonant frequencies and wolf tones. Audio spectrum
analyzers can be used to graph the frequency response of soundboards or
violin plates. Software is available that enables laptop computers and other
digital devices to function as audio oscillators and spectrum analyzers; cell
phones and other hand-held devices can now be equipped to work as audio
oscillators and electronic tuners.

References
Edward G. Hoffman, Student's Shop Reference Handbook (New York, 1986).
John E. Traister, Machinists’ Ready Reference Manual (New York, 1988).

Measurement system conversion

Length

1 inch (in. or ″) = 25.40 millimeters (mm)

1 foot (ft or ′ = 12 inches) = 304.80 millimeters
1 yard (yd = 3 feet) = 0.914 meters
1 millimeter (mm) = 0.03937 inches
1 centimeter (1 cm = 10 millimeters) = 0.3937 inches
 1 meter (1 m = 100 centimeters) = 3.281 feet

Weight (dry; avoirdupois)

1 grain (gr) = 0.0648 grams

1 dram (1 dr = 27.343 grains) = 1.772 grams
1 ounce (1 oz = 16 drams) = 28.349 grams
1 pound (1 lb = 16 ounces) = 453.59 grams
1 gram = 0.035 ounces
1 kilogram = 2.205 pounds

Weight (dry, troy; used to weigh precious metals)

1 grain (gr) = 0.0648 grams

1 pennyweight (1 dwt or pwt = 24 grains) =1.555 grams
1 ounce (1 oz t = 20 pennyweight) = 31.103 grams
1 pound (1 lb t = 12 ounces) = 373.24 grams
1 gram = 0.03215 ounces
1 kilogram = 2.679 pounds

Weight (apothecary)

1 grain (gr) = 0.0648 grams

1 scruple (1 s ap = 20 grains) = 1.296 grams
1 dram (1 dr ap = 3 scruples) = 3.887 grams
1 ounce (1 oz ap = 8 drams) = 31.103 grams
1 pound (1 lb ap = 12 ounces) = 373.24 grams
1 gram = 0.03215 ounces
1 kilogram = 2.679 pounds

Capacity (liquid; U.S.)

1 minim (min) = 0.0616 milliliters

1 fluidram (1 fl dr = 60 minims) = 3.697 milliliters
1 fluidounce (1 fl oz = 8 fluidrams) = 29.574 milliliters
1 gill (1 gi = 4 fluidounces) = 118.291 milliliters
1 pint (1 pt = 4 gills) = 0.473 liter
 1 fifth (one-fifth US gallon) = 0.757 liter
1 quart (1 qt = 2 pints) = 0.946 liter
1 gallon (1 gal = 4 quarts) = 3.785 liter
1 teaspoon = 1⅓ fluidrams = 4.929 milliliters
1 tablespoon = 4 fluidrams = 14.787 milliliters
Imperial measure
1 fluid ounce (fl oz) = 28.413 milliliters
1 pint (1pt = 20 oz) = 0.568 liter
1 quart (1qt = 40 oz) = 1.137 liters
1 gallon (1 gal = 160 oz) = 4.546 liters
1 milliliter (ml) = 1 cubic centimeter (cc) = 0.0338 ounce
1 liter (l = 1000 ml or 1000 cc) = 1.0567 quarts or 33.814 ounces

Force

1 pound-force = 0.453592 kilograms-force

1 newton (N) = 0.10197 kilogram-force
1 kilogram-force = 9.80665 newtons

Mersenne's Law

Marin Mersenne (1588–1648) published a number of observations

regarding the interrelationships of string length, string mass, tension, and
pitch in his Harmonie universelle (Paris, 1636), which can be summarized as
follows:

1 Pitch is inversely proportional to string length.

2 Pitch is proportional to [the square root of] string tension.
3 Pitch is inversely proportional to [the square root of] the mass of the
string.

What is often referred to today as “Mersenne's Law” was not originally

expressed as a mathematical equation by Mersenne himself; this was done by
the English mathematician Brook Taylor (1685–1731), who reduced the
above relationships to an equation and published it in his Methodus
Incrementorum Directa et Inversa (London, 1715).
 Mersenne's Law is as follows:

or as

where

F is the frequency in Hz
L is the length of the string in meters
T is the tension in newtons
M is the mass per unit length in kilograms per meter
D is the diameter of the string in meters
ρ is the density of the string (the density of gut is around 1300 kg/m3, a
sample of old iron wire is around 7690 kg/m3; modern steel piano wire is
around 7830 kg/m3; a sample of old yellow brass wire is around 8240
kg/m3, a sample of old red brass is around 8680 kg/m3; copper is
8940kg.m3 (Goodway and Odell, 1987).

(The newton is a unit of force in the metric system. Though the kilogram is
more properly a unit of mass rather than of force, another metric unit, termed
the kilogram-force, or kgf, is often used for convenience when calculating
string tensions. 1 kgf is equal to 9.80665 newtons.)

Reference
Martha Goodway and Scot Odell, The Metallurgy of 17th- and 18th-Century
Music Wire (Stuyvesant, 1987).

Metallurgy
 (see-also Brass and brass alloys, Iron and steel, Lead and lead alloys,
Nickel silver, Gilding)

Properties of metals

Alloys often exhibit markedly different properties than their constituents;

for example, the malleability of gold will be reduced by the addition of
copper, which itself is a highly malleable metal. Silver can be hardened by
the addition of a small amount of copper, which by itself is extremely soft.
Certain metals, such as nickel and platinum, are very difficult to fuse in their
pure state; however, the addition of copper will render nickel more easily
fusible. Arsenic, tin, and zinc will have a similar effect upon platinum.
The melting point of an alloy is determined by the proportions of its
constituents. Lead, for example, melts at 626°F (330°C) while tin melts at
455°F (235°C); however, an alloy of 5 parts tin to 1 part lead melts at 381°F
(194°C). The eutectic is defined as the specific proportion of an alloy's
constituents that exhibits the lowest melting point, which is called the eutectic
point. For tin/lead alloys, the 361°F (183°C) eutectic point occurs at a
proportion of 61.9:38.1; for silver/copper alloys, the eutectic point of 1434°F
(779°C) occurs at a proportion of 71.9:28.1. Alloying is sometimes used to
increase resistance to corrosion and oxidation. Tin, for example, is added to
copper (yielding bronze) to make it more resistant to corrosion. Machinability
may also be improved by alloying: for instance, the addition of a few percent
of iridium reduces platinum's extreme malleability and ductility and renders it
more easily drilled and turned. Copper is difficult to cast in its pure form, but
the addition of small amounts of zinc or tin makes it easier to cast.
Brittleness is the tendency of metal to crack when struck.
Ductility is a measure of plastic deformation that can be sustained prior to
fracture. It enables a metal to be drawn into wire or extruded into a variety of
shapes without breaking. Though ductility is similar to malleability in that it
involves the ability to tolerate deformation without fracture, a metal such as
lead exhibits poor ductility even though it is soft and highly malleable.
Metals in descending order of ductility include gold, silver, iron, copper,
nickel, zinc, tin, and lead.
Elasticity is the ability of a metal to resume its original shape after being
subjected to force.
Fatigue is the weakening or failure of metal caused by cycles of stress and
strain. It typically occurs at stress levels that are appreciably below the tensile
or yield strength of the material under static conditions.
Fusibility refers to the ability of a metal to melt with the application of
heat. Metals such as lead and tin melt at low temperatures and are considered
highly fusible; metals such as platinum and titanium are less fusible and
require great temperatures to melt them. Alloys often have lower melting
points than the individual metals that make them up. See discussion of the
eutectic and eutectic point above. Fluxes affect the ability of a metal to flow
when it reaches its melting point.
Hardness is the ability of a metal to resist scratching or penetration.
Degrees of hardness are measured by the Vickers, Knoop, Brinell, Mohs, and
Rockwell systems, which use diamond points or hardened balls to make an
impression in the metal's surface under controlled force or impact. The width
or diameter of the impression is used to determine the hardness.
Machinability refers to a metal or alloy's ability to yield to the cutting
action of tools. The factors affecting machinability involve hardness,
toughness, ductility, and the nature of chip formation. Metals such as
properly annealed low-carbon steels have high machinability ratings, whereas
metals such as copper or hot-rolled low-carbon steels have low machinability
ratings. By correct choice of cutting speed, cutting tool geometry, and cutting
fluid, most materials can be machined. Unlike wood, which cleaves ahead of
the cutting tool, metals tend to shear at the point of contact with the cutting
tool's point or edge. Alloys of brass that contain only copper and zinc are
difficult (and potentially dangerous) to machine because cutting tools, such as
drills, tend to grab the metal; however, if alloyed with 1–2% lead, they
become “free-cutting brass,” which machines more easily.
Malleability refers to the ability of a metal to be beaten or rolled out
without cracking. Metals in descending order of malleability include gold,
silver, copper, tin, lead, zinc, iron, and nickel. Certain alloys, such as alpha
brass, have poor hot-working properties and are best worked cold, whereas
other alloys, such as beta brass, can only be worked hot. A small amount of
copper added to silver will improve its malleability, though the addition of
small amounts of lead, tin, or zinc will sharply reduce it.
 Resilience is the ability of a material to absorb energy during deformation
and to release this energy upon removing the source of the deformation.
Malleability, ductility, toughness, hardness, brittleness, elasticity, and
resilience can be affected both by cold-working the metal and by heat
treatment, and thus are not immutable properties.
Tensile strength is the maximum amount of stress that can be sustained
prior to fracture. For metals, the range varies considerably (from about 50 to
3,000 megapascals) according to metal type, alloy, degree of work-hardening,
and heat treatment.
Toughness refers to the ability of a metal to withstand constant load or
impact without fracture. In descending order of toughness are iron, nickel,
copper, silver, gold, zinc, tin, and lead.

Heat treatment

The properties of metals are affected by the initial method of manufacture,

and subsequent heat treatment, working, and reheating. Most metals become
harder if drawn through dies or hammered against a hard surface (see section
on Work-hardening, below). If one continues to work the metal, it can
become so brittle that it will crack. When some metals are heated to certain
temperatures, they regain their softness, ductility, and malleability.
Nonferrous metals such as silver, copper, and certain alloys of brass achieve
their maximum state of softness by heating them red-hot and then quenching
them in cold water; slow cooling in air produces less extreme softness. Brass
should be allowed to lose its red glow before quenching, otherwise it may
crack upon quick immersion. After annealing, certain alloys of silver and
copper can be rehardened by the process of precipitation heat treatment,
which involves exposure to controlled amounts of heat over periods of time
up to several hours. Steel, in contrast to silver, copper, and certain alloys of
brass, is brought to maximum hardness by heating it until it becomes bright
red and then quickly quenching it in oil, water, or brine. The quenching
medium depends upon the alloy and the relative size of the piece. Some steels
are air-hardened rather than quenched, while others develop a hardened
“skin” from air cooling while the interior remains relatively soft. Steel can be
tempered (softened) by controlled reheating after quenching – for example,
by holding the steel in a draft of hot air rising from coals or by heating it with
a torch. Oxidation colors associated with various reheating temperatures can
be used to judge the resulting hardness (see Tempering steel). These
oxidation colors are best observed on a surface that has been brightly
polished after initial quenching. In tool making, steels are deliberately
brought to different degrees of hardness by tempering. Though hardness is a
desirable property for a cutting tool, extreme hardness renders the cutting
edge brittle and liable to chip; thus the hardness must be adjusted to a level
appropriate to the work.
A number of modern brass-instrument makers have experimented with the
cryogenic treatment of brass, in which it is gradually brought down to −325°F
(−198° C). This process is said to relieve internal stress and to yield
acoustical advantages. Some modern horn makers carefully control the
temper of the metal used in making the bells of their instruments and may
even offer the musician a choice of tempers, which are said to affect the tonal
characteristics of the instrument.

Work-hardening

When metals such as steel, brass, nickel silver, and silver are bent,
hammered, or undergo forming operations, they are gradually hardened.
Hammering or rolling causes a reduction in thickness, and the reduction of
thickness is proportional to an increase in hardness. For example, a 10.9%
reduction in the thickness of soft, fully annealed brass results in what is
termed 1/4 hard brass; a 20.7% reduction in thickness results in 1/2 hard
brass; a 29.4% reduction in thickness results in 3/4 hard brass, and a 37.1%
reduction in thickness results in what is termed hard brass. High-grade steel
used to make piano wire can sustain up to forty draws through successively
smaller dies without fracture (this however, is not done in practice – see Wire
gauges for early keyboard instruments). The alloys of brass that are
customarily used to make wire (i.e. the so-called alpha brasses, which are at
least 65% copper and can be cold-worked) are generally sufficiently work-
hardened after seven or eight passes through dies that annealing is required to
continue drawing them. Similarly, repeated annealing is required to safely
hammer brass to the desired shape without cracking. The strengthening of
iron and brass as a result of successive passes through wire drawing dies is
termed tensile pickup.
Stress and strain

As stress is applied to a metal object, it will undergo strain (dimensional

change). This dimensional deformation may be either elastic (temporary) or
plastic (permanent), depending upon whether the stress exceeds the material's
elastic limit. When stress levels are below the material's elastic limit, stress
and strain are related to each other according to Hooke's law:
σ = Eε

where

σ is stress
E is the modulus of elasticity
ε is strain.

At the point where this simple proportional relationship fails, deformation

becomes plastic, or permanent. That point is termed the yield strength.

Metallurgical analysis

Qualitative and quantitative analysis can be conducted without removing

samples with open-architecture X-ray fluorescence equipment; corrosion
products can be identified by X-ray diffraction. The examination of specially
prepared samples under a metallurgical microscope may reveal whether the
metal has been cast, rolled, hammered, quenched, annealed, and reworked
(and often the sequence of these operations), thus providing valuable
information about the steps of manufacture and possibly the original temper
or hardness of the object. Unfortunately, to perform this type of analysis, it is
necessary to remove a sample, which must then be mounted, polished, and
etched for examination. The determination of a metal's hardness may be
misleading with regard to a musical instrument's original state, as later
manipulations, such as those employed in making repairs, may have altered
the original characteristics. Furthermore, as certain metals age (including
some copper alloys), they may become harder due to natural changes in their
crystalline structure. Areas of stress that were the result of bending and
soldering may become the loci of embrittlement and cracking, particularly
when exposed to corrosive substances such as sweat, saliva, and polishing
compounds containing ammonia (which cause dezincification). Oxidation, as
well as airborne pollutants from combustion (such as sulfur and nitrogen
oxides, along with their resulting acids), can also cause embrittlement. Thus,
the hardness of a metal after two or three hundred years may differ from
when it left the instrument maker's hands, even if it has not been subjected to
subsequent hammering, burnishing, and heat treatment.
Evidence of previous repairs may indicate that the metal has already
suffered irreversible alteration of its structure or characteristics, but this does
not necessarily open the door to further treatment. One should keep in mind
that brass wind instruments are made primarily of thin metal: the sheet brass
used to make a Baroque trumpet may range in thickness from 0.2 mm to 0.3
mm; in some cases plating may be as thin as 0.1µm. When working with such
thin material, it is clear that whatever affects the surface will have a profound
effect upon the entire structure of the instrument. For example, a severely
dented or creased bell might require annealing to remove these
disfigurements; however, annealing would leave the bell very soft (likely
softer than it was when it left the maker's hands as a finished instrument); the
removal of dents and creases through burnishing and hammering would
work-harden the metal, though this process might not necessarily reestablish
the original hardness. Furthermore, hammering and burnishing might stretch
the metal, reduce its thickness, and also impart subtle changes to the
instrument's overall shape. Soldering causes oxidation and firescale, which
are traditionally removed by immersion in acid pickles – a process that may
leave brass a pinkish color due to the depletion of zinc or the deposition of
copper on the surface. To reestablish an instrument's original color after
pickling, it would be necessary to polish the surface with abrasives. Clearly,
this sequence of interventions would result in the alteration of shape,
thickness, hardness, surface composition, and finish of an instrument.

Metal identification

X-ray fluorescence (either with an open-architecture device or a scanning

electron microscope equipped with EDX) is an excellent technique for
identifying metals and determining the relative composition of alloys.
However, the composition of lead/tin alloy organ pipes cannot be reliably
determined by X-ray fluorescence due to their lack of homogeneity
(“spotted” metal is a perfect example). While working at the Metropolitan
Museum of Art, a curator and conservation scientist attempted to use X-ray
fluorescence to determine the alloy of a pipe so that several others missing
from its rank could be fabricated. The alloy's composition, as determined by
peak-height ratios, turned out to be grossly inaccurate, and the new pipes
fabricated with an overabundance of lead soon collapsed under their own
weight. To determine the alloy of organ pipe metal by its specific gravity, see
Organ restoration.
For a simple method of distinguishing silver from nickel silver, prepare the
following testing solution:

1 gram potassium dichromate

22 ml nitric acid, technical grade (60–65%)
8 ml distilled or deionized water.

When preparing this solution (sometimes termed Schwerter's solution),

remember the safety precaution of adding the acid to the water, and then add
the potassium dichromate.
Because this solution will etch the metal, apply a small drop to the
underside or to an inconspicuous area using a thin glass rod or toothpick.
Fine silver will turn the solution bright or blood red; 0.925 silver will turn the
solution dark red; nickel silver will turn the solution dark blue. Blot off the
solution and wash with distilled or deionized water after testing.
Gold is best assayed by a testing laboratory (X-ray fluorescence alone may
not give an accurate indication of karat), though simple testing kits are
available from jewelers’ suppliers. These kits each consist of a slate
touchstone, a set of needles tipped with different karats of gold, and several
vials of acids of different strength. To determine the karat, the object is
rubbed against the slate touchstone to leave a mark. Marks are then made
alongside with the tipped needles, and drops of testing acid are then placed
on them. The reaction is then observed and the streaks are compared to
ascertain the karat.
To determine whether an object is gold or gold-plated, make a scratch in an
inconspicuous spot. Use a toothpick or glass needle to deposit a small drop of
nitric acid in and around the scratch. If the scratch and the surrounding area
turn green, the object is not made of gold; if the acid turns green in the
scratch but not in the surrounding area, the object is made of a base metal
plated with gold; if the acid in the scratch turns a milky white, the object is
gold-plated silver. If there is no reaction, either in the scratch or in the
surrounding area, the object is entirely gold.
The karat (not to be confused with the carat weight of precious stones; 1
carat = 200 milligrams) refers to the percentage of pure, or “fine” gold in an
alloy. 24K is 100% gold; 18K is 18/24 or 75% gold; 14K is 58.3% gold; 12K
is 50% gold, etc.
To distinguish brass from bronze, dissolve a small sample of the metal in a
drop of concentrated nitric acid (or place a small drop of the acid on the
underside or an unexposed side of the metal object for a few moments). After
the metal has dissolved in the acid and turned it a greenish color, pick up the
acid with a glass capillary tube or eyedropper and add it gradually to an equal
volume of distilled water. If the sample is bronze, a precipitate will form; if it
remains clear it is brass.
High-carbon steels can be distinguished from iron and mild steel by
observing the spark pattern during high-speed grinding: iron produces
straight sparks, while mild steel produces a few tiny forks at the ends of the
sparks, and high-carbon steel exhibits a multitude of forks.
A magnet can be used to distinguish iron and steel from most white metals,
such as silver, lead, tin, and zinc. Nickel is slightly magnetic, though nickel
silver barely responds to a small hand-held magnet. Copper, brass, and
bronze are not magnetic.
Test strips (available from companies such as Merck and Gallard-
Schlesinger) can be used to identify metals such as aluminum, copper, iron,
lead, manganese, molybdenum, tin, and zinc. These test strips (which
generally must be moistened with a drop of distilled or deionized water or
with a reagent) are pressed against the object and change color in the
presence of the metal they are designed to identify.

Reference
William D. Callister Jr., Material Science and Engineering (New York,
 1997).

Metalworking

Alloying is the combining of different metals to provide a metal of new or

improved properties. For making replacement parts, solders, and spelter,
small ingots of an alloy can be prepared by combining the component metals
together with a small amount of crystalline boric acid and heating them
directly with an acetylene torch in a fused silica or graphite crucible, or by
placing the crucible in a small electric furnace (available from jewelers’
supply houses). Carbon or quartz rods are used to stir molten metal.
Bending jigs are useful for accurate bending of rods and strips of metal.
They are especially good at making tight bends in heavy stock. Long sharp
bends in sheet metal are easily achieved on a bending brake.
Burnishing is a technique of forming and smoothing metal by rubbing with
a hard, highly polished material, typically steel, agate, and stellite (an alloy of
cobalt and chromium). Thin sheet metal can be shaped against mandrels by
stroking the metal against the mandrel with a burnisher. Gold and silver can
be burnished to a luster that is difficult to achieve with abrasives, and without
the concomitant loss of metal. When used in the fabrication of wind-
instrument keys, burnishers serve to sharpen edges and corners (whereas
abrasive polishing tends to round them off), thereby creating a stronger visual
effect. Steel burnishers were once used to polish the surface of brass
instruments. After several parallel, overlapping strokes, the burnishers were
repolished by running them through grooves in heavy strips of walrus hide
charged with emery and crocus. The burnisher was first rubbed back and
forth in a groove charged with emery and then in a groove charged with
crocus. After careful cleaning off of the abrasives, the burnisher was dipped
into soapy water, which served as a lubricant, and the process continued. A
motorized buffing wheel charged with chromium oxide compound can also
be used to polish burnishers.
Chasing tools are used with special lightweight chasing hammers to
emboss designs and textures on metal surfaces. In a technique termed
repoussé, similar tools are used to impress designs from the reverse side.
Dapping is a technique of stretching metal into a shape by hammering it
into a concave die. The cups of salt-spoon keys can be formed by using a
dapping block and punches.
Drilling can be done by hand using bow, pump (Archimedean spiral), and
“egg-beater”-style drills, which is often more convenient and affords better
control than electric drills. If a pump or bow drill is used, it is advantageous
to use spade bits because they can be sharpened to drill on both clockwise
and counterclockwise strokes.
Before drilling, one should locate the center of the hole with crossed scribe
lines followed by a prick punch and then a center punch. If regularly spaced
holes must be drilled, dividers can be used to make short arcs across a center
line. For accuracy, it is a good idea to align the prick punch with the scribe
marks under magnification. A machinist's hammer fitted with a magnifying
lens is helpful in accurately positioning prick and center punches. A center
drill bit, which is primarily designed to make countersunk holes in
workpieces that are to be mounted on lathe centers, can also be used to
position drill bits. Because they are short and rigid, center drill bits are less
likely to wander than narrow-diameter twist-drill bits, which are commonly
used to drill pilot holes. Large-diameter holes should be drilled by degree,
starting with a small-diameter bit followed by bits of increasing diameter
until the final size is reached. This reduces the amount of torque required to
drive large bits and results in cleaner holes. When drilling deep holes, the bit
should be withdrawn periodically to eject chips, which might otherwise
interfere with drilling and mar the surface of the hole. When drilling by hand
in thin sheet metal, no lubricant is generally required, though one may
moisten the tip of the bit with saliva.
Drill presses in which the workpiece or bit is brought into contact by direct
hand control are referred to as “sensitive drill presses.” Such machines enable
the operator to sense resistance to the drill bit and variations in material
hardness. Small precision drill presses used in watchmaking allow the
operator to raise the work up to the spinning bit, which supposedly offers
greater control and less breakage of delicate wire bits. On some drill presses,
the head can be swiveled for angular drilling, but for most angular drilling the
work is held in a drill press vise. When positioning the drill bit, the
workpiece can be allowed to “float” on the drill press table until the bit
catches a mark made with a center punch or center bit; the workpiece is then
firmly clamped to the table to complete the drilling operation. Some vertical
milling machines are fitted with a traveling quill, which permits them to be
used like a drill press. Their X-Y tables can be used to accurately position the
work under the drill bit. A machine termed the “jig borer” is essentially a
massive drill press that is primarily designed to drill accurately positioned
holes in steel drilling jigs. Lathes are used for concentric drilling or boring of
turned work. When drilling with the lathe, either the work is rotated and a
nonrotating bit is directed against the work from the tailstock, or the bit is
chucked to the headstock and the work is supported by a tailstock chuck or
mounted on the cross slide by means of a vise or by clamping to an angle
plate. A crotch center mounted on the tailstock is very helpful for drilling
perpendicularly into rods or tubing. Boring is generally done with a special
lathe cutter called a boring bar, which is mounted on a tool rest supported by
the cross slide.
The advantage of motorized over hand-operated drills is that the drill bit
can be set in motion before it makes contact with the work and operates at
higher speeds. This can be of great assistance in the accurate positioning of
drilled holes, especially when they are drilled at an angle to the work surface.
Special cutting fluids are recommended when drilling various types of metals
with drill presses and electric hand drills (see Oberg et al., Machinery's
Handbook, 2012).
Note that when drilling brass with an electric hand drill or drill press,
unless one is confident that one is drilling into so-called “free-cutting” brass
(an alloy that contains 1–2% lead), it is vital to clamp the workpiece firmly to
the workbench or drill press table, because ordinary brass has a tendency to
climb drill bits and grab onto them as they pass through. If the workpiece is
hand-held, it will likely be torn free, which can be dangerous.
Draw rings and drawplate. Misshapen tubing can be rounded-up by
passing it through successively smaller draw rings or holes in a drawplate.
Bowed tubing (such as a trumpet yard or trombone slide) can be straightened
by pulling it through a slightly larger hole in a draw plate while applying
moderate pressure to the outside curve, thereby compressing it and
straightening the tubing. A hydraulic draw press is used to compress tubing
around tapered mandrels by pulling the tubing and mandrel through an
expandable brass draw ring.
Files are versatile and important tools for refining metal surfaces. They are
available in a wide variety of shapes, including barrette, cant, checkering,
crochet, crossing, equaling, half-round, hand, joint edge, knife, lozenge, oval,
pillar, pippin, round, slitting, screw head, square, three-square (or triangular),
and warding. Files come in a variety of cuts: American-pattern grades include
coarse, bastard, second, smooth, and dead smooth; the Swiss-pattern grades
run from 00 (coarse) to 6 (extremely fine). The actual coarseness of a file
depends upon two factors: the cut and the length. For example, a 10-inch
“smooth” file will have a coarser cut than a 6-inch “smooth” file. Single-cut
files have a single set of teeth, while double-cut files have a second set cut at
about a right angle to the first. Single-cut files are used in lathe work and are
better for draw filing (see below). Some flat files are made without teeth on
one edge; this edge is referred to as the “safe” edge and is useful when filing
up to an edge or surface without risk of cutting into it. Floats are files with
curved teeth specially designed for work with soft materials such as
aluminum. Files come in a wide variety of sizes, typically from about 300
mm in length down to needle files of about 100 mm in length. The smallest
needle files are termed “escapement” files because of their use in making
clock and watch parts.
Separate sets of files should be kept for brass and ferrous metals. In
general, files for brass must be extremely sharp, while those for iron or steel
sometimes work better if they are a bit dull. When files used for brass have
become worn, they can be cleaned with a file card and used for iron or steel.
If files are used on lead or lead alloys, they must not be used for silver or
brass, because lead filings left on these metals will cause pits or holes if the
workpiece is heated for hard soldering.
Files can be used on the lathe to remove the coarse marks left by cutting
tools. In lathe filing, the file is hand-held over the work (the tang or handle is
supported by the dominant hand, while the thumb and forefinger of the other
hand support the far end of the file). A light touch is used to simultaneously
push the file across and along the rotating work. For best work at the lathe,
files with a single-cut, long-angle tooth pattern are employed. Jeweler's lathes
can be fitted with a file rest; when used in conjunction with an indexing head,
facets can be filed on cylindrical work.
There are two main techniques of hand filing: flat filing and draw filing.
Hand filing is generally done while standing at the bench, with the work
supported at elbow height in a vise. Flat filing of larger pieces generally calls
for the work to be firmly mounted in a vise, while the file is supported by
both hands to prevent it from moving in an arc. Sheet aluminum or other soft
metal can be cut to fit the jaws of vises to prevent marring the work. Wood
“chops” (consisting of two pieces of hardwood hinged together) can also be
employed for holding thin or especially delicate pieces in a vise. A hand-held
ring clamp is a convenient tool for holding small objects for filing. When
filing into corners, select a file that has a “safe” edge (i.e. no teeth) to prevent
removal of material from a previously dimensioned or finished surface.
Triangular files, otherwise known as “three-square files,” (as opposed to truly
square files) are used for enlarging and refining square holes because they
come in contact with only one surface at a time (square files, even if tapered,
are generally not used for enlarging square or rectangular holes because it is
difficult to bring one face of the file into a corner without another face
coming in contact with an adjoining surface). Needle files can be mounted on
reciprocating scroll saws for refining the profiles or edges of a workpiece.
Specialized filing machines are also available for this type of work. Draw
filing is a technique of filing that provides a finished surface. In draw filing,
the workpiece is supported in a vise and the file is held by both hands at right
angles to the surface and pushed across it. This technique is especially useful
when refining the edges of metal objects. Single-cut files give the best finish
with this technique.
When fitting together irregularly shaped metal pieces with a file, one can
coat the finished part with soot from an oil lamp, and then position it against
a part that must be fitted to it. Soot will be transferred to “high” spots, which
can then be filed or scraped (see below) away. A perfect fit is achieved when
soot is transferred evenly from one part to the other. This is similar in
principle to “chalk-fitting” used in woodwork.
File cards should be used to clean filings from files, rifflers, and rasps.
Files may also be rubbed with chalk to prevent “pinning,” or the collection of
filings between their teeth. Trapped filings can score the work and interfere
with efficient filing.
Forging is a technique of metal forming employing continuous pressure or
hammer blows, either with or without the use of dies or heat. It can be used to
produce complex shapes and increase strength in critical areas by localized
hammering.
 Forming (see also Raising below …) by use of stakes, mandrels, hammers,
and burnishers was the traditional method of making the flared sections of
brass wind instruments, and it is also an effective technique for removing
dents, creases, and distortions from damaged areas. The manufacture of
highly flared sections (such as bells) primarily involves stretching the metal
to its final configuration. Stretching is accomplished by hammering the metal
against a stake or by supporting it on a stake and hammering just beyond the
point of contact. Metal can also be shaped by contraction. Certain conical
sections of brass instruments (such as trumpet lead-pipes) are fabricated by
use of a draw bench or hydraulic press, which forces cylindrical tubing and a
shaped mandrel through an expandable orifice (such as a brass ring), thereby
compressing the tubing against the mandrel. Certain metals, such as iron,
steel and the beta brasses, are formed red-hot, while silver and alpha brasses
are worked cold.
Gravers are made of high-grade steel and are principally used for
engraving designs and lettering. They come in a variety of shapes (onglette,
bevel, chisel, flat, knife, lozenge, square, oval, and round) and in graduated
widths to produce a variety of cuts. When engraving small pieces and thin
stock (such as sheet metal), the work is generally supported by an engraver's
block or pitch bowl (the pitch used to fill the bowl is tempered with plaster of
Paris, rosin, and tallow). These blocks and bowls are mounted on felt or
leather rings so that the workpiece can be comfortably positioned and steadily
moved or rotated into the graver. The engraver's block is actually a miniature
vise fitted to a rotating turntable that permits the work to be directed with
great ease and fluidity. Gravers are also used as cutting tools in conjunction
with a tool rest to turn delicate metal objects on the lathe.
Grinding is used to true the surfaces of cast iron or hardened materials. In
surface grinding, the workpiece is generally clamped to a movable table and
fed into a rotating grinding wheel. In a Blanchard-type grinder, the workpiece
is rotated against a counter-rotating grinding wheel. Cylindrical grinding can
be done on a machine lathe by mounting a motorized grinder on the cross
slide.
Hammering is used to shape metal parts in conjunction with stakes, anvils,
and snarling irons. Silversmiths’ hammers and stakes come in a great variety
of shapes for raising, forming, and planishing. These tools can be used in
fabricating new parts (such as woodwind instrument keys) or removing dents,
creases, and other damage to sheet metal parts, such as the bells and tubing of
brass instruments (see Brass instruments). Wood and rawhide mallets,
plastic-faced hammers, and “soft hammers,” which have heads of cast lead or
Babbitt metal, can be used when it is important to protect the surface of metal
during shaping or re-forming.
Knurls (either mounted in a tool holder and supported by hand on a lathe's
tool rest or mounted on the tool post of a lathe's cross-slide) can be used to
emboss decorative elements on mounts and rings.
Lapping and honing are used to closely fit pieces together through abrasive
action. Lapping bored holes involves the use of expanding split sleeves fitted
with abrasive stones or laps charged with abrasives that are rotated and
moved back and forth within the holes in order to true the surface and
provide a close fit between running parts, such as the rotary valves and
pistons of brass wind instruments. In the final fitting of brass-wind
instrument pistons to their respective cylinders, the piston is charged with a
very fine abrasive compound and worked back and forth within the cylinder,
a technique called “honing.” Flat surfaces can be closely fitted or refined by
lapping and honing with abrasive stones, diamond-dust-encrusted metal
plates, or glass plates charged with abrasive slurries.
Mandrels and expanding mandrels can be placed within tubing to round up
or enlarge tubing to achieve a better fit.
Marking out is generally done with a pointed scriber or knife. Metal
straightedges and templates are used as guides, while dividers are used to
transfer dimensions from rules or other measuring instruments to the
workpiece. Marking gauges and hermaphrodite calipers are convenient for
scribing and laying out lines parallel to edges.
Milling typically refers to the machining of surfaces with cutting tools.
There are several types of milling machines, though most have rotating
cutters that cut either on their ends or their periphery, with the workpiece
being fed into the cutter along any of three axes that are perpendicular or
parallel to the cutter's axis of rotation. Vertical milling machines are useful in
instrument work for cutting slots and grooves.
Pliers and cutters permit the manipulation of small metal objects. Pliers are
made with flat, round, half-round, and chain nose jaws, which are useful for
opening and closing metal bows, bending rings, coiling springs, etc. Those
used by jewelers, dentists, and opticians have specially shaped jaws that can
be useful in musical instrument repair work. Pliers are available with smooth
and serrated jaws; for instrument work smooth jaws are generally preferred as
they are less apt to mar the work. They are available in a variety of jaw sizes;
both fine (jewelers’) and medium (electricians’ and mechanics’) sizes are
useful. Large duck-billed pliers are useful for bending and straightening the
keys of woodwind and brass instruments, and special swedging pliers are
used to compress, re-form, and elongate metal tubing used in woodwind
instrument key work. The working edges of cutters are ground for regular,
semi-flush, or flush cutting. Flush-cutting nippers have their cutting bevels on
the inside surfaces of the jaws, which permit projecting pins or sprues to be
cut off close to the surfaces from which they project. Special cutters are
available with hardened or carbide jaws for cutting music wire (these often
have compound action for greater leverage). To make perfectly straight cuts
in wire or rods, it is best to use a shearing action. To make a shear for wire
and small-diameter rods, close the jaws of a stout pair of lineman's pliers and
drill holes in proximity to the joint that permit insertion of standard sizes of
heavy wire or rods. After the material to be cut is inserted, the handles are
pulled apart, and the wire or rod will be sheared straight across.
Polishing with motorized buffing wheels has been an essential part of brass
instrument manufacture for well over a century, though it is rarely employed
in conservation work because it is extremely aggressive and can quickly
erode the metal's original surface texture, engraving, and construction marks.
In fabrication work, tool marks and heavy scratches are removed with emery,
tripoli, and lime applied with muslin buffing wheels. Stitched buffs, which
are firmer than unstitched ones, are used to reach into recesses, while ganged
buffs are used to polish broad surfaces. Finer grades of polishing compounds,
such as red rouge, are then applied with soft flannel buffs. Goblet,
cylindrical, and small-diameter buffs can be used for polishing the interiors
of bells and around valve casings. Buffs used for one compound must not be
used with another and should be marked and stored separately to avoid
contamination. Special rakes are used to true and break up glazed surfaces on
buffs. For small-scale work (such as polishing newly fabricated exhibition
mounts or replacement parts), ½-horsepower electric motors turning at 1725
r.p.m. are adequate for buffs up to about 6 inches in diameter. Larger buffs
should turn at slower speeds, which can be adjusted with combinations of
pulley wheels and belts. The tapered arbors used to mount buffing wheels
should project well past the motor housing to provide adequate clearance for
the workpiece. A length of heavy rubber or flexible plastic hosing should be
installed over arbors and projecting motor spindles to prevent damage to the
workpiece should it be drawn into contact with them. Fume hoods and
exhaust systems should be of adequate strength to clear airborne buffing
compounds away from the operator and the rest of the workroom. A face
mask should be worn to prevent inhalation of the abrasives. In most
instrument-making shops, the buffing room is isolated from other work areas.
Raising, sinking, bouging, and planishing (see also Forming above) are
techniques of forming and smoothing sheet metal by hammering it against
stakes or dies. Raising is a technique of transforming a flat sheet of metal into
a three-dimensional object by hammering on the outer surface while it is
supported internally by a stake. In sinking, the metal is hammered into a
form. Both raising and sinking stretch the metal. Bouging is a corrective
procedure employing heavy, nonoverlapping hammer blows; it is used to
remove rough spots, to even out metal thickness, and to equalize tension in a
piece. It is often done after raising and sinking. Planishing is a smoothing and
hardening technique employing lightweight, polished, convex-headed
hammers. Light, overlapping blows are delivered to the metal, which is
supported by a stake.
Reamers are used to refine drilled holes or to achieve diameters that are not
available with standard-size drill bits. Reamers can be chucked in lathes and
drill presses or fitted to tap handles or pin vises and used manually. They are
generally tapered, though they are also available with nontapered shafts.
Expanding reamers can be adjusted until the desired diameter is achieved.
Reamers are made with single and multiple flutes. Those having single flutes
are generally turned by hand, whereas straight and spiral multi-fluted reamers
can be used either on a lathe, or a drill press, or by hand. Watchmakers’
pentagonal reamers are held in a pin vise and turned by hand; they are used to
fit center pins in modern piano actions.
Rifflers are files that have short cutting surfaces and are often curved. They
are useful for getting into areas that straight files cannot reach. Die-sinkers'
rifflers come in a wide variety of shapes and cuts, including very fine cuts
(No. 6) for detail work.
 Rolling mills (either hand- or motor-operated) are useful for the reduction
of commercial sheet metal or ingots to required thickness. After several
passes through the mill, brass and silver must be re-annealed to prevent
cracking and buckling. Silver can be reduced to about a third of its original
thickness before re-annealing is required, but brass is not as malleable and
must be annealed repeatedly to reduce it by this amount.
Sawing is done by hand with conventional metal-cutting hacksaw blades.
Parts fabricated from thin stock can be cut out with jeweler's saws fitted with
fine blades. Two types of jeweler's saw blades are available: piercing blades,
which have rounded back corners and permit the blade to saw curves, and flat
blades (generally made in larger sizes), which lack the rounded corners and
are primarily used for straight cuts. If a blade has trouble turning corners and
cutting tight curves, the back corners can be relieved by rubbing them against
a sharpening stone after the blade is mounted on a saw frame (see below).
When sawing sheet metal, it is necessary to have at least two saw teeth in
contact with the metal at all times; if sufficiently fine blades are unavailable,
tilting the saw blade relative to the work may engage a sufficient number of
teeth to facilitate cutting. Saw frames come in a variety of depths; however, it
is best to use the smallest frame required for the work as the deeper the
frame, the more difficult it is to maintain sufficient tension on the blade. To
install a blade, it is first secured in the clamp at the upper end of the frame
with its teeth pointing downwards. The frame is adjusted so that its length is
somewhat greater than that of the blade, and then flexed by pressing the top
of the frame against the edge of a workbench to bring the lower part of the
blade between the jaws of the clamp located at the handle, which is then
tightened with a wing nut. If the blade is under sufficient tension, it should
produce a bright, ringing sound when plucked with the finger. Some saw
frames have rotating blade clamps, which allow the blade to be mounted at
any angle relative to the frame. In this way, a narrow strip can be sawed from
a sheet of any length, even though the frame is of limited depth. If the frame
does not have rotating blade supports, the blade can be twisted at the top and
bottom with a pair of pliers to achieve the proper angle. When sawing metal,
the blade can be periodically drawn through a block of beeswax or paraffin to
lessen the tendency of the blade to catch. Wintergreen oil is sometimes used
to lubricate very fine jeweler's saw blades. When sawing out small parts, a
bench pin with a V-shaped cutout can be used to support the workpiece,
while the saw blade is positioned at the vertex of the V and drawn up and
down as the workpiece is directed against it. The bench pin can be clamped to
the edge of the workbench, though a carriage bolt and wing nut make a
perfect means of temporarily attaching it through a hole drilled in the
workbench or through a bench-dog mortise. Motorized scroll saws can also
be used for freehand sawing of sheet metal; however, they must be supplied
with effective hold-downs to prevent the workpiece from jumping up and
down uncontrollably with the saw blade. Metal-cutting blades can be
installed in band saws that are primarily intended for wood, though the blade
speed should be reduced when sawing metal. Power hacksaws are used to cut
heavy bar and rod stock, and special metal-cutting band saws are designed for
cutting curved shapes in sheet metal and heavy metal plates. When sawing
most metals on a motorized scroll saw or band saw, the operator can apply a
block or stick of beeswax to the blade where it makes contact with the metal;
the beeswax will melt on contact and act as a cutting lubricant.
Scrapers can be used to fit metal parts together. To determine where to
scrape, soot deposited by an oil lamp to the finished piece can be transferred
to the part that requires refinement; flat or curved scrapers (or files) can then
be used to remove “high” spots that have picked up soot. Hand-scraped lathe
beds and milling-machine ways were once the mark of fine, handcrafted
machine tools.
Shears or snips are used to cut out straight, curved, or irregular patterns.
These are available with jaws designed for straight, as well as left- and right-
curved, cuts. Bench shears are stationary or table-mounted and are
convenient for cutting strips or curved shapes. Nibblers (both handheld and
bench-mounted) are used to cut irregular shapes in sheet metal.
Soldering (see also Soldering).
Spinning is done on special lathes designed to turn at high speeds and bear
heavy loads on their headstock and tool rest. The tool rest consists of a
horizontal support with strong vertical posts upon which handheld spinning
tools bear. Spinning tools are essentially stout, specially shaped burnishers
that direct the sheet metal disc against a form attached to the headstock. The
disc is pressed against the center of the form by a wooden cone projecting
from the tailstock center. Rosin can be used to prevent the metal disc from
slipping while spinning.
 Swaging is a technique of altering the shape of metal by compressing it
between two dies or the jaws of swaging pliers. A piece of tubing can be
elongated or have its diameter reduced (to effect a better sliding fit, for
example) by placing it in a swaging die. The small tubes used in woodwind
instrument key work can be adjusted (either to reduce play or to increase
length) with special swaging pliers (see Woodwind instruments).
Threading of new or replacement parts can be done directly on the lathe,
either by hand by the technique of thread-chasing or automatically on a
machine lathe by advancing the cutter against the work by means of a set of
change gears. Up through the early nineteenth century, special screw-cutting
lathes had headstocks fitted with sets of threaded guides. When one of these
guides was engaged, the chucked workpiece was advanced against a fixed
cutter, which turned the thread. Some early woodwind instruments, such as
the French musette de cour, often have threaded sections that were formed on
such thread-box lathes. Thread-chasing was another method of threading that
made use of simple hand-held cutting tools (termed combs) that had several
teeth spaced to match the pitch of the intended thread. The tool was
supported by a tool rest and directed in an arc against the leading edge of the
rotating workpiece. Gradually, and with repeated passes, the angle between
the tool and the workpiece was reduced and the tool would begin to cut the
thread. The back teeth of the comb would follow (or “chase”) the groove,
while the leading tooth would begin to cut the thread. By these methods, both
inside and outside threads of any diameter could be formed.
To cut a thread using a modern machine lathe, it is necessary to arrange the
feed gears that coordinate the rotational motion of the headstock with the lead
screw, which provides lateral motion to the carriage. The simplest screw-
cutting lathe has a gear mounted on the lead screw, and another gear mounted
on the back end of the headstock spindle. To determine the sizes of gears to
use, multiply the number of threads per inch or millimeter that one needs to
cut by a small number, such as 3 or 4, which will give the number of teeth for
the gear mounted on the lead screw. Then, by this same small number,
multiply the number of threads per inch or millimeter of the lead screw, and
that will give the number of teeth for the gear mounted on the headstock
spindle. Machine lathes that are fitted with more complex gear trains
generally come with a diagram and chart to assist in selecting gears for
turning various threads.
 To establish the size of a screw or screw hole, one must determine the
major (or nominal) diameter and the pitch of metric threads (which is defined
as the distance between adjacent threads) or the number of threads per inch.
Metric thread sizes are designated by a series of letters and numbers: M10 ×
1.5. 2. This code indicates a metric thread of 10 mm nominal diameter with a
pitch of 1.5 mm, having a class of fit of 2. In the English, or American,
system 1/4–20 UNF 2 indicates a Unified Fine thread (the designation UNC
stands for Unified Coarse; UNEF stands for Unified Extra Fine) of ¼-inch
nominal diameter, with twenty threads per inch, having a class of fit of 2 (fit
sizes range from 1 to 3, with 1 being a loose fit and 3 being a tight fit).
Screws smaller than ¼-inch are graded in a series of sizes from 0000 to 12.
To calculate the major diameter of the numbered sizes from 1 to 12, multiply
the size number by 13 and add 60. This will give the diameter in thousandths
of an inch. By using a “screw checker” or screw plate, the pitch and nominal
size of a screw may be ascertained by turning the screw into the screw
checker or screw plate. If the screw can be turned in at least 1½ turns without
binding, the pitch and size is established. Another method of determining the
size of a screw is to hold a screw pitch gauge against the screw to determine
the metric pitch or number of threads per inch; a caliper is then used to
measure the major diameter. To determine the size of a threaded hole, screws
of known size or plug gauges can be screwed into the threaded hole.
Taps and dies can be used for making small threaded parts (such as the
screws or pivot rods used in modern woodwind instruments). Taps are used
to make female threads; dies to make male threads. If a small screw is
missing from an instrument, with any luck a similar one can be found
elsewhere on the instrument to gauge the size, or a series of stock screws can
be tested in the hole to determine or estimate the correct size. Taps are
available with tapered threads for easy starting. Plug and bottoming taps
allow threading to the bottom of a flat-bottom hole and are generally used
after the standard tapered tap has been used to start the hole. If a plug or
bottoming tap is not available, a regular tap can be converted into one by
grinding off the tapered end (though the tap then becomes useless for starting
new threads). Gun taps are designed to throw chips ahead of the tap. These
should only be employed when the hole to be threaded passes completely
through the work. Dies generally flare out at the starting face to ease the
beginning of the threading process. To thread completely up to the face of
screw head, for example, one can reverse the screw through the back of the
die to carry the threading right up to the face of the head. If the die has a
chamfered back face, it can be ground back to facilitate threading up to the
face of the head. Oil can be used as a lubricant when threading iron and steel;
brass is generally tapped dry, while beeswax can be used in aluminum (see
also Tap drill sizes).
To make a small screw, select or turn a length of mild steel, iron, or brass
rod to the diameter of the screw head and mount it in a collet. Then turn
down the section to be threaded to the major diameter. A screw plate or die is
then held perpendicularly to the end to be threaded and the lathe head is
rotated by hand until the screw shaft engages the screw plate and the thread is
cut to the required depth. After the thread has been formed, the screw can
then be cut off the spinning lathe with a parting tool. The slot can be sawn
with a jeweler's saw or filed with specially tapered screw-slot files. To blue
small steel screws, they should first be polished, degreased, and then placed
on a brass plate and held over an alcohol lamp or torch until the desired color
is achieved.
If the screw is not a standard size, it may be possible to make a die that will
“roll” the desired thread. A rolling die can be made by screwing an exemplar
into a root-diameter hole that has been drilled in a fully annealed steel plate.
After removing the exemplar, the steel plate is heated red-hot, quenched, and
tempered to a light straw color. Screw blanks (turned on a lathe) can then be
screwed into this die to form the thread. A cutting die can also be made from
an existing screw: first make a rolling die as described above, but before
annealing it, drill two smaller holes on either side of the threaded hole and
connect them to the threaded hole with cuts made with a jeweler's saw. The
die is then heated until red-hot, quenched, and tempered to a light straw
color. The sawn openings will serve as cutters while the two smaller holes
provide clearance for the chips. Light machine oil can be used as a cutting
lubricant for threading iron or mild steel; beeswax (melted into a bit of
turpentine to soften it somewhat) can be used as a lubricant for threading
aluminum. Brass is generally threaded dry.
Turning of metal and other materials was first accomplished on a simple
instrument known as the turns, which employed fixed centers and a tool rest.
Today, turning is done on a lathe. With the turns, motion was supplied
directly to the workpiece by use of a hand-operated bow or by a cord that was
wrapped around it and drawn back and forth by a foot pedal, with tension
supplied by an overhead spring (often a tree branch or flexible plank of
wood). Because the cord was flexible, it could even rotate an irregularly
shaped or square billet, though it would have to be repositioned to gain
access to the entire piece being turned. While this use of the turns sounds
primitive, its fixed centers provided perfect centering of the object being
turned, and the workpiece could be turned end to end or removed and
remounted without impairing accuracy. This was generally not the case with
early lathes, unless they were fitted with precision bearings and the lathe bed
was true. Even after the invention of the lathe, many watchmakers continued
to rely upon the turns for work requiring the greatest accuracy, such as the
turning of balance shafts. When using a lathe, the workpiece is often
supported between centers, but it is driven by one of them. As with the turns,
the use of centers is the best means of assuring concentricity of the object,
and it is generally vital if the work must be turned end to end. Soft materials,
such as wood and ivory, can be turned on center by using a spur driver in the
headstock. With metal objects, the work can be turned on centers using a
drive plate and lathe dog. One cannot turn between centers if one needs
access to the end of the workpiece or if drilling and boring must be done. If
this is the case, the work must be chucked, and if long or spindly, it should be
supported by a steady or traveling rest. Collets provide a high degree of
accuracy in holding and driving work. These are available with circular and
square apertures (which are also useful for holding square or octagonal
stock). They generally require a collet closer or draw bar to tighten them
around the workpiece. The holding range of a collet is very limited, and so
they must be acquired in sets. Specialized collets are available that expand to
grip tubular or bored workpieces from the inside.
Another means of supporting the workpiece between centers is by use of a
mandrel or arbor. Mandrels and arbors may be tapered or cylindrical.
Expanding mandrels maintain their cylindrical shape over their range;
expanding stub mandrels are useful for gripping short objects, such as rings
and ferrules (Figure 45b).
A variety of chucks are available for holding work, including self-centering
three- and six-jaw chucks (the latter chuck is better for holding objects that
might tend to become deformed by pressure exerted by the jaws), and four
jaw chucks with independent movement of the jaws. Bezel chucks have
multiple, curved jaws that hold thin, circular pieces or plates without marring
them. Objects can also be clamped to a faceplate for turning or boring.
Cement chucks are designed to hold small work for turning on a jeweler's
lathe. To attach the workpiece to the cement chuck, stick shellac or turners’
cement is melted and a drop is placed on the face of the chuck. As the
workpiece spins, parts can be centered by lightly pressing them against the
molten cement with a pointed scriber until the part ceases to wobble and the
cement hardens. Cement chucks are often made of brass to conduct heat
when melting the shellac or turners’ cement. In a similar fashion, large work
can be temporarily cemented or glued to the faceplate of a conventional lathe.
The relatively small metal, ivory, and wood pieces required of musical
instruments can generally be turned with hand tools supported by a tool rest.
When turning soft materials (such as wood and ivory) with hand tools, it is
important to be able to raise and lower the tool rest so that the cutting edge
can be positioned to make a slicing cut, though a great deal of turning is also
done with scraping tools that are level with the axis or rotation. When turning
metal, cutting tools are generally oriented so that their top edge is level with
the axis of rotation.
Metalworking lathes are generally fitted with a cross slide, which is
essential for accurate machining of cylindrical parts. The compound slide
allows the cutter to be directed at an angle, and this is very useful for taper
turning. On some lathes, tapers can also be turned by offsetting the tailstock
or headstock (Figure 8, p. 107). For thread cutting, metalworking lathes are
generally equipped with back gears and a lead screw to guide the cross slide
and its cutter at the appropriate rate as the workpiece rotates.
Trumming is a method of polishing with abrasives that are applied with
strips of cloth that are coiled and pulled around sections of tubing and joints.

References
Erick Brand, Band Instrument Repairing Manual (Elkhart, 1978).
Donald de Carle, Practical Clock Repairing (London, 1994).
Donald de Carle, Practical Watch Repairing (London, 1995).
Rupert Finegold and William Seitz, Silversmithing (Radnor, 1983).
Henry B. Fried, Bench Practices for Watch and Clockmakers (Fairfax, 1997).
Ward Goodrich, The Watchmakers’ Lathe (Fairfax, 1999).
Paul N. Hasluck, Metalworking Tools, Materials and Processes for the
Handyman (London, 1904).
Edward G. Hoffman, Student's Shop Reference Handbook (New York, 1985).
Richard R. Kibbe, John E. Neely, Roland O. Meyer, and Warren T. White,
Machine Tool Practice (Englewood Cliffs, 1995).
Herbert Maryon, Metalwork and Enamelling (New York, 1971).
T. McCreight, The Complete Metalsmith (Worcester, 1991).
Erik Oberg, Franklin D. Jones, Holbrook L. Horton, and Henry H. Ryffel,
Machinery's Handbook 29th Edition (New York, 2012).
Harold O’Conner, The Jeweler's Bench Reference (Taos, 1988).
Spon's Workshop Receipts for Manufacturers and Scientific Amateurs
(London, 1909).
John E. Traister, Machinists' Ready Reference Manual (New York, 1987).

Metronome

The oscillation frequency of a simple pendulum is defined as

where

f is the frequency
g is the acceleration due to gravity (9.81 m/sec2)
l is the length of the pendulum.

The oscillation frequency of the type of double pendulum used in the

Winckel or Maelel metronome is defined as follows:
where

f is the frequency
g is the acceleration due to gravity
M is the mass of the heavier weight mounted below the pivot
R is the distance of the heavier weight below the pivot
m is the mass of the upper movable weight
r is the distance of the upper movable weight above the pivot.

The number of “beats” per minute can be calculated from the frequency as
follows:
beats = 60f/π

References
Sture Forsén, Harry B. Gray, L. K. Olof Lindgren, and Shirley B. Gray, “Was
Something Wrong with Beethoven's Metronome?”, Notices of the
American Mathematical Society 60/9 (October, 2013), pp. 1146–1153.
Rosamond E. M. Harding, The Metronome and Its Precursors (Henley-on-
Thames, 1983).

Mold making

Musical instruments are often missing small metal parts (bolt covers,
casters, decorative mounts, hinges, keys, etc.), and if similar parts are
available, molds can be made from the originals and replacements cast using
the lost-wax process (Figure 21).
Figure 21 Rubber mold with replicas of Broadwood square piano bolt
covers, six of which have been patinated.

The mold-making materials in wide use today include vulcanized rubber

(often used in the jewelry industry), silicone rubbers (both condensation- and
addition-cured rubbers), and polyether rubber (the latter two often used in
dentistry). Vulcanized rubber molds have the disadvantage of releasing sulfur
during the vulcanization process, which involves heating and compressing
sheets of rubber around the object being cast. The heat and sulfur will cause
original parts of brass, bronze, and silver to tarnish badly. This is not the case
with room-temperature-curing silicone rubbers, which are safe to use with
these metals. However, so-called “condensation-cured” silicone rubber
exhibits relatively high shrinkage, low tear strength, and poor dimensional
stability, and it gives off alcohol, acetic acid, and other byproducts during the
curing process, which may be detrimental to varnishes, lacquers, waxes,
paints, and other coatings on the original part. So-called “addition-cured”
silicone rubber, particularly the type employing chloroplatinic acid as a
catalyst, does not produce alcohol as a byproduct, though hydrogen gas
bubbles are released from formulations that lack a hydrogen scavenger. It has
relatively high tear strength and produces very accurate and dimensionally
stable molds, though curing may be inhibited by substances containing tin,
sulfur, and amines.
Silicone casting rubbers are available in different consistencies, from putty
to liquid, and with setting times varying from a few minutes to up to twenty-
four hours. Curing time may be adjusted by modifying the temperature; for
example, the curing of liquid casting rubbers such as Dow corning 3110 RTV
or E RTV can be lengthened by placing them in the refrigerator, thereby
providing an opportunity for air bubbles to rise to the surface. The putties can
be molded by hand around objects or the objects can be pressed into them. If
the thick, syrupy types are used, the object must be set in a leakproof
container, such as a casting frame, flexible plastic box, or container
constructed of or lined with wax. Those having a liquid consistency can be
used in a technique known as slip casting (in which the object is dipped into
the mold-making liquid), or it can be brushed on. Overly flexible or slip-cast
molds must be set into a “mother mold” to provide dimensional stability.
Some casting materials require the object to be given a thin coating of a
release agent, such as mineral oil or petroleum jelly; other casting materials
(such as some of the silicone rubbers) have oils suspended in them that serve
the same purpose. Casting materials that are intermixed with oily release
agents are not appropriate for making casts of wooden objects or other
materials that may absorb or be damaged by these oils. After mixing two-part
systems or adding catalysts, syrupy and liquid casting materials can be placed
in a vacuum chamber to eliminate bubbles. To further suppress bubbles, after
the casting material has been poured into the molding frame or container, it
can be placed in a pressure chamber to force the bubbles into solution.
Molds can also be made using plaster of Paris, compo (see compo), or
Hydrocal (a gypsum-based casting plaster), though these rigid-impression
materials cannot be used if there is undercutting or the object is sensitive to
water. Alginate (a casting material used in dentistry) is dimensionally
unstable, produces poor detail, and tears easily; furthermore, it is mixed with
water, so it is again unsuitable for casting objects that might be affected by
water.
One-part molds can be used when casting flat-backed objects such as piano
lid hooks and bolt covers; two- or multi-part molds, or those having a
removable core, may be required for more complex shapes. For most small
pieces, the object is suspended within a molding frame by a sprue-form and
the molding material is poured around it. Liquid molding materials can be
brushed around complex shapes to assure that there are no voids and that all
the detail will be captured. The cured material is then cut apart using
specially shaped mold-cutting knives. Locks (truncated pyramidal shapes) are
first cut into the corners to help realign the two halves of the mold after they
have been separated and the original part has been removed.
After the original has been removed from the mold, specially formulated
casting waxes are poured or injected into the reassembled mold. The Kerr
company produces waxes in a range of flexibilities and hardnesses. The
molten waxes are injected by means of a hand-pumped or compressed-air
wax injector or by a centrifugal device (both of which can be obtained from
jewelers’ supply firms). After the wax impression cools, it should be closely
inspected for defects. Voids can be filled with molten wax, which is then
carved or filed to shape; excess wax along parting lines can be cut away with
a sharp knife. Special forming tools and heated spatulas can be used to refine
the wax impression.
After the wax impression is corrected and refined, wax sprues are attached
using a heated spatula. These provide passages for the removal of the wax
during the burnout process as well as for the entry of molten metal. Once the
sprues are attached, the wax is suspended in a metal flask and a heat-resistant
plasterlike material termed the investment is poured around the wax. The
investment is made of gypsum, silica, cristobalite, and reducing agents,
which can withstand the heat required to melt out the wax as well as that of
the molten metal that will be poured into it. While the investment plaster is
still wet, it should be placed in a vacuum chamber to remove bubbles.
Burnout is done in a special temperature-controlled furnace. After the wax
has been burnt out, the investment flask is placed in a centrifugal casting
machine (either spring- or motor- driven) and the nozzle of a graphite or
silicon-carbide crucible is inserted in the sprue opening. Molten metal is
poured into the crucible or melted directly in the crucible with a torch, and
the casting machine then spins the investment flask and crucible so that the
molten metal fills the cavity in the investment.
After the metal has cooled, the investment is broken away. The sprues
(Figure 22) are then sawn off the casting, and the surface is refined with files
and rifflers (see Metalworking).
Figure 22 Cast copies of ormolu mounts from a Viennese fortepiano before
cutting off their sprues and electrogilding.

The equipment described above (vacuum chamber, centrifugal casting

machine, etc.) is not terribly expensive and can be purchased from jewelers’
supply firms. Musical instrument conservators may not wish to get involved
in the entire casting process as it is fairly complex and requires experience to
master its intricacies and refinements. In urban centers that have a jewelry-
making industry (such as New York City), there are specialized craftsmen
who make molds and do the casting (as well as others who do polishing,
patinating, enameling, etc.). While working at the Metropolitan Museum of
Art, the author learned the mold-making process from craftsmen working in
the museum's reproduction studio. Making molds and waxes in-house
reduced the possibility of damage and loss of original parts, and relieved the
author of the administrative details of removing artwork from the museum.

References
Murray Bovin, Centrifugal or Lost Wax Jewelry Casting for Schools,
Tradesmen, Craftsmen (Forest Hills, 1995).
Benvenuto Cellini, The Treatises of Benvenuto Cellini on Goldsmithing and
Sculpture (New York, 1967).
Cennino d’Andrea Cennini, The Craftsman's Handbook “Il Libro dell’ Arte”
(New York, 1960).
Herbert Maryon, Metalworking and Enamelling: A Practical Treatise on
Gold and Silversmiths’ Work and Their Allied Crafts (New York, 1971).

Mother-of-pearl and abalone

Mother-of-pearl and abalone are hardened nacreous secretions found inside

the shells of various mollusks. These substances consist of thin layers
(approx. 0.0002 mm thick) of calcium carbonate crystals, termed lamellae,
that are cemented together with a protein called conchiolin. Because calcium
carbonate is dissolved by acids, mother-of-pearl and abalone gradually erode
from continued contact with perspiration, which is mildly acidic. Mother-of-
pearl may be bent (for example, to make a curved facing for the head of a
violin bow) or flattened by soaking it in a mild alkaline solution for several
weeks and then clamping it against a form. Mother-of-pearl and abalone shell
decompose into lime (calcium oxide) under the heat of an open flame, such
as that of an alcohol lamp.
A jeweler's saw can be used to cut out various designs for inlays (Figure
23) and turned on a lathe to make circular eyes for bow frogs and other parts.
Figure 23 Replacement mother-of-pearl parts ready to be glued into a
nineteenth-century qanun soundboard rosette.

References
Hermann Kühn, Conservation and Restoration of Works of Art and
Antiquities vol. I (London, 1986).
Tom Wilder, ed., The Conservation, Restoration, and Repair of Stringed
Instruments and Their Bows (Montreal, 2010).
 N
Nickel silver

Nickel silver (also known as German silver) is an alloy that generally

contains copper, zinc, and nickel. The proportions vary, but typically consist
of about 3:1:1 of these metals. If there is no zinc in the alloy, it is reasonably
safe to use ammonia (household strength or further diluted) to remove copper
corrosion products from nickel silver. If there is a high percentage of zinc,
which is fairly typical of the alloys used in making musical-instrument keys,
treatment with ammonia may leach out the zinc, resulting in embrittlement
and possibly leading to stress-cracking. EDTA or alkaline Rochelle salt
solutions are safer to use (see Brass and nickel silver cleaning).
Acids are not generally recommended for cleaning copper corrosion
products that have formed on nickel silver, as they may react with cuprite to
form a pink film of copper on the object.
 O
Oddy test

The Oddy test is used to determine whether materials are safe for use in
storing or displaying artwork. It involves enclosing a sample of the material
from which the artwork is made (for example, a polished coupon of silver or
brass) along with the material to be tested (for example, a piece of fabric used
to upholster an exhibition case). A moisture buffer, such as a wad of cotton
moistened with distilled or deionized water, is placed in the container, which
is then sealed and placed in an incubator at 60°C for 28 days, after which the
coupon is examined for traces of tarnish or corrosion.

Organ restoration

The organs most likely encountered in private collections or museums are

the smaller types: portatives (portable), positives (those that can be moved,
though generally with some difficulty, and “deposited” where required),
cabinet organs, regals (small organs with beating reads), harmoniums, and
melodeons (the last two employing free reeds). Less typical of instrument
collections are the larger “installed” organs found in churches, concert halls,
theaters, and townhouses, though these grander instruments may share the
same basic elements as the smaller ones: feeder-and-reservoir wind supply,
pallet-and-slider windchest, mechanical key and stop actions, and
conventional pipework. Much larger organs, especially those equipped with
free-standing consoles and remote divisions made in the mid to late
nineteenth century or the twentieth century, may have stop-channel, tubular-
pneumatic, Barker-lever, or pouch-and-membrane chests, as well as
electropneumatic or electrical key actions, computerized stop actions, and
other electro/mechanical systems – however, those contrivances will not be
covered here. The author has unfortunately encountered numerous cabinet
organs in museums that are compilations of parts and pipes cobbled together
from various sources; in many cases these organs never functioned, and it
would be extremely frustrating to attempt to “restore” them.
Tuning and voicing of pipework (see below) should never be undertaken
before the intended wind pressure has been established. Pressure is easily
determined by use of a manometer. A manometer can be made by bending a
piece of soda-lime glass tubing into a U-shape, using an alcohol lamp or gas
flame to soften the glass. The manometer is held or mounted in an upright
position and partially filled with water. A pipe is removed from the chest and
the manometer is fitted in its place either directly or through a length of
flexible tubing fitted with a coupler shaped like a pipe foot. When the
bellows is pumped, the air pressure is indicated by the difference in height of
the two columns of water in the U-shaped tube. Slant-tube manometers
(designed to measure pressure in forced-air ventilating systems) provide a
more accurate reading due to their extended scales.
Though it is a relatively easy matter to determine pressure, establishing
whether it is correct or not is more difficult and relies upon a historical
knowledge of organ building, and, better still, upon measurements of
unaltered instruments by the maker. The wind pressure of organs varies
considerably: many sixteenth- through eighteenth-century chamber organs,
for example, range between 30–75 mm, while the reeds of the Atlantic City
Convention Hall organ completed in 1932 were voiced at over 2500 mm
pressure! An instrument originally designed for 60 mm wind pressure would
have its pitch and timbre substantially altered if the pressure were increased
or decreased by as little as 10 percent. With time, pressure gradually tends to
decrease due to the formation of leaks, though it is often the case that it has
been deliberately increased to raise the organ's pitch or to increase the
volume of sound. Wind pressure and pipe mouth cut-up are closely
interrelated, so if wind pressure is increased without raising the heights of the
mouths, the pipes may fly off their speech. Consequently, if the pressure has
been raised, the pipe mouths may have been cut up to permit them to be
blown more forcibly. One must also establish whether the organ's pitch has
been transposed. For example a Renaissance cabinet organ may have been
adapted to lower Baroque pitch by shifting all of its pipes up a semitone;
similarly, a Baroque organ may have been brought up to modern pitch by
shifting its pipes down or reducing their lengths. If pipes have been shifted, a
new pipe may have been fitted for the top or bottom note. When examining
an organ, check to see if the note names scratched on the pipes agree with the
notes on the keyboard or if they are off by a note or more. Often the openings
in the pipe feet have been changed in the process of revoicing: knocking-up
cups or tapered mandrels may have been used to close or open the toes, or the
feet of wood pipes may have been drilled out or partially occluded with
slivers of wood. A pipe originally designed for open-toe voicing would
sometimes be modified for closed-toe voicing if new stops requiring greater
wind pressure (such as reeds) were added to its division. Often the alterations
of mouth height and pipe length are subtle or difficult to discover or date.
A visual assessment of the instrument may indicate whether the bellows
weighting is original. Bellows weights may be of cast lead or pipe metal, iron
plates, bricks, slabs of marble or limestone, or even field stones. If the
reservoir is fitted with neatly cast slabs of common pipe metal held in place
with wood cleats whose age seems consistent with the top board of the
reservoir, the presence of a couple of stones thrown on top of the cast lead or
the presence of helper springs pressing down on the top board of the reservoir
are indications that the weighting was later increased, either to compensate
for leaks, to raise the pitch, or to increase volume. As the tuning and voicing
of pipes are affected by wind pressure, the restorer must rely upon his or her
knowledge of historical parameters of pressure and pitch as well as a
developed aesthetic sense that the pipes are working at their original
specifications. If the original pressure has been compromised due to leaks,
organ tuners sometimes resort to the expedient of cutting down pipes to
restore their pitch. By altering the pipes in this way, the scaling, voicing, and
original tonal characteristics of the pipes are changed. It is therefore
important to attend to leaks and repair the wind supply before attempting to
tune or regulate pipe speech.

Wind supply

Assuming for a moment that the organ's wind is supplied by hand or foot
power and not by an electric blower, one of the first problems that one
encounters may be leaky feeders. Feeders may be operated by the organist
using a foot pedal or by an assistant operating a pump handle behind or at the
side of the instrument. In some early designs, such as regals and some chest
or cabinet organs, the feeders are themselves the reservoirs. These are
generally operated in pairs; as one falls, the other is lifted. In this way, a
continuous flow of air is maintained. Most early organs employ a two-part
wind supply, consisting of one or more cuneiform feeders and a parallel
reservoir. As the feeders are pumped, the folds of the reservoir rise. The folds
are often controlled by a pair of regulators, which consist of pivoted strips of
wood mounted on either side of a central, floating frame; these cause the
reservoir to rise evenly. Weights of cast lead or stone slabs placed on top of
the reservoir regulate the wind pressure.
One of the feeder's valves permits the intake of air while another allows air
to be delivered from the feeder to the reservoir. “Puff” or “flap” valves
consist of either a hinged flap of leather or a piece of leather loosely tacked
around a series of holes. Air pressure within the feeder forces the feeder valve
to seal when it is delivering air and to open when air is drawn in; air pressure
from the feeder causes the reservoir valve to open, and internal pressure
within the reservoir causes it to shut when the feeder is drawn open for air
intake. Another valve, a safety valve, is generally located on top of the
reservoir to prevent it from bursting if overinflated. This valve may be held
shut with a moderately strong spring and is either pulled open (as the
reservoir reaches full extension) by a cord fastened within the reservoir or
pushed open by a block of wood located above the reservoir. Valves are often
sources of leaks and should be checked to see if they are sealing properly.
Unfortunately, it is often necessary to cut open the feeders or reservoir to
gain access to the valves for repair.
Many early organs originally operated by hand or foot feeders were later
refitted with electric blowers. The simplest way of attaching an electric
blower is to make another pallet-box bung board that is equipped with a hose
coupling. In this way, the hand- or foot-operated feeder can be left intact and
may always be employed if desired. Air supplied by manual or foot-operated
feeders is often less turbulent and of a more stable temperature than that
supplied by an electric blower, and is thus preferable for concerts and
recordings, while the blower can be used for practice or normal day-to-day
use. Very often, however, the feeders have been discarded and an electric
blower has been installed in their place. If this is the case, one must decide
whether to reconstruct the missing feeders or retain the blower. In museums,
the general public gets a thrill out of seeing an organ being hand-pumped; if
an organ is situated where the pumping can be observed, it is advisable to use
or restore the original mechanism. In some organs of recent manufacture, no
reservoir is used. Instead, the blower supplies air to a rigid wind chamber
fitted with Schwimmers (a shallow bellows fitted with springs), which
maintain steady pressure as the wind is depleted.
Even a small rent in the leather hinges or gussets may render the bellows
totally ineffective. Defective valves may also make a bellows inoperative.
Sometimes the leather couplings of wind conduits or trunks may be ripped
(this often occurs as wood parts shrink or parts of the case shift or settle),
resulting in loss of wind pressure. For centuries, tawed (alum-cured)
sheepskin has been used to make organ bellows (see Leather). Vegetable-
tanned skins are avoided because their acidic tannins are corrosive to
pipework, and tanned skins are subject to “red rot.” Nevertheless, modern
pneumatic pouch leather is generally prepared from vegetable- or chromium-
tanned sheep or goat skins. Its ready availability and the fact that it comes
skived to a remarkable thinness have made it a choice for patching bellows;
however, because this leather is acidic (and unsightly), it should not be used
for this purpose; nor should white-colored leather tanned with zirconium salts
and formaldehyde, which is sometimes fraudulently offered as “alum leather”
by firms that sell organ supplies. Its strong chemical smell distinguishes it
from tawed skins, and it generally lacks suppleness; furthermore, it is highly
corrosive to pipe metal. Therefore, it is best to use alum-tawed skins for
repairs and for reconstructing hinges and gussets.
If bellows leather appears to be original and essentially sound, localized
splits and holes can be patched rather than resorting to complete releathering.
Patches should have their edges carefully skived (i.e., beveled on the flesh, or
gluing side). This is done not only for the sake of neatness, but also to reduce
the tendency of edges to curl away from the wood as time goes by. When
skiving, a razor-sharp knife is used to pare away the flesh side of the skin
while the hair side is supported by a hard, smooth surface. Small skiving
devices employing replaceable razor blades (such as the Scharf-Fix) can be
used to bevel the edges of strips and patches. When gluing on a patch, hot
animal-hide glue is applied to the flesh side and a cloth lightly dampened
with hot water is used to press it into place and to wipe away any drops of
glue. A polished bone or ivory paper folder (of the type used in bookbinding)
can be used to press down the leather and squeeze out excess glue. In places
where the thickness of a leather patch may interfere with some operation of
the mechanism, goldbeater's skin may be glued over a split. Rubberized cloth
(also known as tosh or mackintosh), which was historically used in reed-
organ bellows, should not be used in reconstructing or repairing the wind
supplies of pipe organs.
If all of the leather is suspect, then it must be replaced. Before releathering,
it is helpful to make paper templates of gussets and corner pieces before the
pieces are removed (or afterwards, if they can be removed intact or their
shapes can be reconstructed). Carefully observe and copy the system of cloth
hinges or tapes (generally made of artist's-grade linen canvas) that are used
on the inside of the ribs, as well as hemp rope, gut, or sinew hinges that are
sometimes employed at the vertex of diagonal feeders.
The internal surfaces of bellows ribs were sometimes lined with parchment
to make them airtight. If the parchment is well adhered, it should be left in
place; if not, it can be soaked off by applying cool damp cloths, and the ribs
can then be carefully dried under pressure so that they remain flat. The
parchment must then be cleaned of old glue and reglued (or replaced, as
required) using animal-hide glue. When gluing parchment, it should first be
moistened until it becomes supple. Fresh parchment may have to be
roughened a bit (either by scraping or sandpapering) before glue is applied, or
else it may not adhere well. If grease is preventing good adhesion, rubbing
the parchment with ethanol or acetone is generally effective in removing the
grease. As the glue and parchment dry, the parchment shrinks and tends to
curl at the edges and corners. These curls can be pressed down with a
dampened cloth, and, if necessary, fresh glue can be applied where the
parchment has failed to adhere. A bone or ivory paper folder can be used to
rub down the edges until the glue sets. Some makers sized the inside surfaces
of ribs with hot animal-hide glue to make it more airtight. If so, a fresh
application of thinned glue may be brushed on.
When releathering bellows, skins should be selected that are similar in
thickness and quality to the originals. Hinges and gusset leather must be
strong and especially pliant. A dial gauge with wide contacts (such as a
violinmaker's thickness gauge) can be used to determine the thickness. If
necessary, skins can be reduced in thickness with a skiving knife or skiving
machine. Again, and especially if one is not overly familiar with this type of
work, one should make detail photographs and sketches of the hinging,
taping, and leathering before and during the various stages of disassembly in
order to facilitate reconstruction of the original system. Sharp knives and
heavy, steel straightedges can be used to cut the strips used for hinges. To
prevent straightedges from slipping, their bottom surfaces can be lined with
thin cork or rubber; this will also keep the leather from being pulled towards
the knife blade during cutting. Work should be neatly executed, with edges
carefully beveled where necessary. Hot animal-hide glue should be used to
attach the leather. Again, a damp cloth and polished ivory or bone paper
folder can be used to press down the leather. All drips and visible traces of
glue should be wiped off with a damp cloth or sponge while the glue is still
wet. In some organs, the bellows ribs are lined on the outside with decorative
paper, so care must be taken that it does not get damaged by water and glue
during releathering operations. Keep in mind that alum-cured leather should
not be allowed to get too wet. Neatness and good organization are important
when leathering bellows, reservoirs, and conduits. When releathering valves
and pallets, glue should not be allowed to drip on the face that serves as the
seal, as it will stiffen the leather and prevent it from making good contact.
Very often the top and bottom boards of the feeders and reservoir have
shrinkage cracks. If these cracks are relatively small and do not impair
physical stability (which might be the case if the boards are bread-boarded)
but are merely leaking air, it may suffice to seal the cracks with strips of
parchment or leather. If the boards are unstable, the cracks should be closed
with animal-hide glue and bar clamps, though this may require complete
disassembly of the reservoir. If the shrinkage cracks are very wide,
compressing them will alter the widths of the boards, and the ribs would then
have to be trimmed. If this is the case, the cracks should be shimmed to
preserve the original dimensions of the boards.
If the reservoir and bellows are disassembled for releathering, this is a
perfect opportunity to check the leather valves to determine whether or not
they are sealing properly. If they have curled or become stiff with age, they
should be replaced. Intake valves are sometimes made of two pieces of
leather glued hair-side to hair-side to lessen their tendency to curl, though
single layers are often employed.
If the organ is very old and the leathering is original or of some antiquity, it
may be preferable to make a replica of the existing bellows, preserving the
original as found. This is particularly true if maker's marks are located inside,
or if interesting parchment or paper manuscripts have been used in their
construction. In any case, all interior inscriptions should be transcribed and/or
photographed before repairs are made and the bellows are closed up.

The windchest

Historically, several types of windchests have been used. The spring-chest

found in some sixteenth- to eighteenth-century Netherlandish, Italian,
Spanish, Swiss, and south German organs uses two sets of pallets: one set is
operated by the keys and the other by the stop action. The stop action of the
spring-chest employs a pallet valve below each pipe. To activate a stop of
pipes, a stout batten of wood is brought to bear against a set of stickers that
opens all of the stop-action pallets simultaneously. The key-lever pallet
admits air to the key channel, though only those pipes whose stop pallets
have been opened receive air. The term “spring-chest” derives from the
myriad springs employed.
Another type of windchest, known as the cone-valve or stop-channel chest,
gained popularity in the nineteenth century. In this type of chest, the pipes are
arranged in stop channels (rather than key channels), and each key operates a
set of valves, one for each stop. Both the spring-chest and cone-valve chest
dispense with the sliders used in the pallet-and-slider chest, which have a
propensity to stick or leak due to changing humidity.
The more common pallet-and-slider windchest consists of two principal
sections, a pallet box and a channeled section, the latter generally formed by
a series of bars rabbeted into the walls of the chest. In small organs, the
channeled section may consist of a solid plank of wood that has been bored
or routed out. The pallet box receives air from the bellows or reservoir, and
pallet valves operated by the keys allow air to rush into their respective
channels. The channels are generally graduated in width from bass to treble
to accommodate the varying demands for air, and they are long enough to
accommodate one pipe from each stop (or a group of pipes if the stop is a
mixture). Mounted above the channels is the table (sometimes misleadingly
termed the “soundboard”), which is pierced with holes corresponding to the
position of each pipe foot. The pipes are not planted in these holes, but two
“layers” above in the toe boards (sometimes called the “top boards”). The
intervening “layer” consists of the sliders (strips of wood drilled to allow
wind to pass from the table to the toe boards) and strips between the sliders
termed “bearers.” The bearers, which are glued to the table, support the toe
boards and pipes and provide just enough clearance for the sliders to move
back and forth, but they must not be too loose, otherwise wind will escape.
The sliders are often “black-leaded,” that is, brushed over with finely
powdered graphite and then burnished, or lubricated with finely pulverized
soapstone so that they run freely. Sliders sometimes run on strips of tawed
leather, which provides a seal while permitting them to move freely. The
sliders are operated by the stop action in such a way that when the stop is
engaged, the holes in its slider align with those of the table and toe boards,
thus activating an entire rank of pipes. When a stop is off, no pipes in that
rank will sound when the keys are played. Sliders are sometimes divided so
that the bass and treble sections work independently. In some English organs
there may be two sliders per rank, one located above the other; the upper
slider is often operated by a foot pedal (generally in combination with
another rank) to switch certain stops on and off together. Thus, in the slider-
chest, each pipe is in effect controlled by two devices: the pallet valve, which
opens up an entire groove in the windchest that coincides with its respective
key or pedal, and the slider, which opens up a hole in the channel that
coincides with the pipe in the particular rank, or “stop.”
The most common problem involving the slider-chest is “ciphering,” or the
unwanted sounding of a pipe. Before blaming the chest, one should first
check the key action of the instrument to see if keys, stickers, trackers, or
rollers are out of adjustment or are warped or rubbing against one another,
causing a pallet valve to remain open (see discussion of key action below). If
this is not the case, one cause of ciphering may be the pallets themselves: dirt
may have become lodged between the pallet valve and its seating, preventing
a good seal. The remedy is to remove the bung board to expose the pallets,
and brushing any dirt off the face of the pallet. To prevent ciphering from
occurring, it is important to prevent debris from falling into the holes in the
toe board or into the chest if toe boards, sliders, or pipes are removed for
repair. Careful brushing and vacuuming of the chest will help prevent the
incursion of dirt and grit.
A weak pallet spring may also be the cause of a cipher. This is remedied by
removing the spring (a special pair of pliers designed for compressing and
withdrawing the spring may be of assistance; Figure 24) and flexing the
spring to give it more strength. The spring should not be made so strong that
the pallet will be more difficult to open than the adjacent ones, otherwise the
key action will seem uneven. Sometimes a spring may be broken, misshapen,
or missing. To fabricate a new spring, select the proper gauge of wire
(measure the damaged spring or an adjacent one with a micrometer), draw it
through a wire-straightening jig, and then shape it to match a neighboring
spring. The proper shape may be formed with a pair of pliers, though it is
helpful to make a simple spring-forming jig to shape the coils neatly. When
fabricating a new spring, a spring-operated gram gauge can be used to match
the tension of other springs.

Figure 24 Organ tuning and voicing tools.

It is not good practice to add a helper-spring to aid in seating a pallet. If

one finds that considerably greater spring pressure is required, then the
problem may be caused by faults in the key action (a corroded pull-down
wire, for example) or, more seriously, by a warped pallet or unevenness in
the pallet leather. In general, pallets are made of clear, quarter-sawn wood
and should not warp. If they have warped, they should be removed and trued
with a plane (if the warp is slight) or replaced with a copy. Should the leather
be at fault, the pallet will have to be releathered. In releathering, one should
copy the system used by the maker, observing the thickness of the leather or
use of multiple layers, if appropriate. Sometimes a pallet will stick if it opens
too wide. It may get caught on a guide pin, or the pull-down may get
snagged. The solution here is to adjust the key action in order to limit the
pallet's excursion, perhaps by adjusting the threaded leather nut at the end of
a tracker that operates the pallet pull-down. The pallet should open wide
enough so that the wind enters promptly, however. Voicing and tuning
problems may result from the uneven opening of pallet valves. Very often
pull-down wires become bent, impeding the motion of the pallet. The obvious
solution is to straighten or replace the pull-down wire.
When a pipe unexpectedly sounds that is adjacent to one that is deliberately
played, there is said to be a “run” or “cipher.” Sometimes this occurs because
a toe board is loose. As these boards are generally screwed down, tightening
the screws may eliminate the run. If, when two non-consecutive notes are
played and the note between them sounds, the running of air may be caused
by the slider being too loose (it may be pushed up by the wind pressure,
allowing air to escape). This may again be remedied by tightening the toe
board screws. If the sliders are leathered, replacing the leather with new or
thicker strips may solve the problem of running. It may be necessary to plane
the bearers down to reduce the gap. Sticking sliders, caused by slight
shrinkage of the bearers, may be eased by gluing paper shims on top of the
bearers. Chalk can be used to determine where contact is being made.
Another cause of ciphers may be a crack in the table or its becoming
unglued from the channel bars. When the table develops cracks or comes
unglued from the channel bars, it is most often the result of excessive heat or
dryness. Smaller cracks that do not pass completely through the table can
sometimes be filled with epoxy, taking care to carefully level the hardened
epoxy fill with a sharp chisel or cabinet scraper after it has set (Figure 25).
Runs are sometimes caused by cracks in the channel bars or if the channel
bars come unglued. This can be ascertained by opening pallet valves and
inserting a flashlight and dental mirror, or by inserting a borescope through
the holes in the toe board. If the bar is simply loose, it may be possible to run
hide glue into the joint, though it may be necessary to invert the chest to get
the glue to run in the right direction. If the problem is serious, it may be
necessary to disassemble the chest and reglue or replace the loose parts. The
channel bars and frame might need to be touched up with a hand plane to
restore flatness and assure good glue contact. Because these types of repairs
are complex and time-consuming, windchests are often replaced rather than
repaired. In the restoration of an organ, every effort should be made to retain
and repair the original windchest, as it is a crucial part of the original design.

Figure 25 Organ windchest by William Crowell, New Hampshire, 1852,

showing original irrigation channels and later cracks that were causing
ciphers. These cracks were filled with epoxy resin.

Ciphers are often concealed by devious means, such as by drilling small

holes in the pipe feet to bleed off just enough air to prevent the pipe from
speaking from the leakage. This is obviously not an acceptable practice; such
holes should be filled and the problem dealt with in the appropriate manner.
Occasionally, a pipe may speak into another that sits close by, causing it to
whistle or hum, thereby simulating a run. Rotating either pipe slightly will
generally solve that problem. It is inappropriate to alleviate runs by
“irrigating” a chest (i.e. making gouge- or V-shaped cuts in the toe boards or
slider bed to allow air to escape), though many makers cut this channeling as
a precaution, and so its presence may not indicate an old problem with
leakage (see Figure 25).

Pipework

Organ pipes are readily damaged by careless tuning and rough handling,
especially when organs are disassembled or moved. Before working on pipes,
one should examine them closely to determine if they were made by the same
hand and whether inscribed note designations correspond to the notes the
pipes actually play. It is important to make this assessment, as pipes are often
cobbled from other instruments or transposed upwards or downwards to alter
the pitch. Also check for alterations in the pipe mouths, as they may have
been cut up or the pipe may have been disassembled and recut so as to lower
the mouth.
Pipes are made of various alloys of tin and lead, though bismuth or
antimony are sometimes added for stiffness. Because of its low cost, sheet
zinc has sometimes been used to make larger pipes. Sheets of tin/lead alloy
pipe metal are formed by pouring the molten metal into a trough that is drawn
along a casting table. An adjustable slot in the trough allows the metal to pour
out in a uniform thickness. In some schools of organ building, the casting
table was covered with cloth, and one can see the cloth pattern on the inner
surfaces of the pipes. Because the melting point of pipe metal is lower than
the burning point of cloth, the cloth did not scorch. After the metal cooled, it
was peeled off the casting table. Pipe metal was sometimes hammered to
increase its density and stiffness and to reduce thickness. Today, drum
planers are often used to reduce metal to the required thickness (the parallel
marks made by their cutters are clearly recognizable). While drum-planing
reduces thickness, it does not impart the same physical changes that
hammering does. Pipe metal is sometimes stored at room temperature to
“season” it, which causes it to become stiffer. Special temperature-controlled
ovens are sometimes used to accelerate this seasoning through a process
termed precipitation heat treatment. To construct pipes, tables or graphs of
dimensions are consulted when marking out the various components (such as
the resonator, foot, languid, etc.) on the sheets of cast metal, which are then
cut out and soldered together (see discussion of pipe soldering below). Wood
pipes are joined up with hide glue and clamps in the usual manner.
When handling pipes, gloves should be worn to protect the metal from
perspiration, which has a corrosive action due to the presence of oils, lactic
acid, salt, and moisture. Pipe metal is quite soft and must be treated with
great care. Typical damage includes general shape distortion, dents, open
seams, tears due to excessive or improper tuning, misshapen mouths and
displaced languids (either from manhandling or abortive attempts at voicing),
as well as dirt and corrosion.
If pipes are removed so that work can commence on other parts of an
organ, or when an organ is moved, pipe trays should be made to store the
pipes safely. Cleats should be fastened to the sides of the trays to prevent the
trays from falling into one another when they are stacked. In storing mixture
ranks, the pipes associated with each note can be gently bundled together
with flat, braided cotton tape or cord.
Check to see if pipes are clean before voicing or tuning them. Over time,
dust and debris may settle in their mouths and windways and affect their
speech. Sometimes brushing with a soft camel-hair brush is sufficient
(gilding mops are the softest brushes, but stiffer brushes may be required to
loosen dirt). A vacuum cleaner can be used, with soft, supplementary brushes
used to push the dirt in the direction of the vacuum nozzle. Care should be
taken that pipes are not marked or dented in cleaning. If the dirt is resistant to
simple brushing and vacuuming, washing may be required. As pipes may be
quite large, washing can present a problem. Ideally, a wet room with long
troughs and sinks should be available. Softened, distilled, or deionized water
should not be used for soaking or protracted washing of metal pipes as lead is
attacked by absorbed carbon dioxide. Soaps and detergents should also be
avoided. Pipes can be lightly brushed with soft bristle brushes to remove
caked-on dust while soaking, and then given a final wash in tap water. They
should be inverted for drying and blotted with a soft, clean cloth to prevent
spotting. Display pipes, if of tin or high-tin alloy, may be polished with a
slurry of precipitated chalk and water applied with a soft cloth, but be aware
that the surfaces of burnished display pipes will be altered by any type of
abrasive action. Especially after washing and polishing, pipes should not be
touched with bare hands; cotton gloves should be worn when handling and
reinstalling the pipes.
Bent or dented pipes can be straightened on mandrels. If one does not have
access to the facilities of a pipe-making shop, wood or metal mandrels can be
fabricated on a lathe. When turning mandrels, all cutting tool marks must be
removed, otherwise they will be impressed into the surface of the pipe during
reshaping. For small pipes, ground drill rod works well as it comes in a wide
variety of sizes (both English and metric) and is usually very well finished.
The ends of these rods should be beveled or rounded off at the end to prevent
damage while inserting them into the pipe. Dent balls and cylindrical slugs
used in brass instrument repair (see Brass instruments) can also be used to
push out dents in organ pipes, but they must be mounted on wood or metal
rods to facilitate insertion and retrieval. It may be possible to push out dents
in pipe feet through the bottom opening; however, if a pipe foot is badly
deformed or crushed, it may be necessary to cut it off with a sharp knife, re-
form it on a stake or conical mandrel, and then resolder it to the body of the
pipe.
Languids and pipe lips can be straightened, smoothed, raised, lowered, and
otherwise adjusted with steel blades, chisels, and probes specially made or
adapted for this function. Standard voicing tools (see discussion below) may
also come in handy when re-forming damaged mouths and languids.
Neighboring pipes should be closely studied to serve as guides in re-forming
damaged ones. The rudiments of pipe voicing are discussed below.
Rips, holes, and open seams of pipes require soldering. Sometimes the tops
or feet of pipes have become so distorted and torn that it is necessary to trim
them down and solder on new pieces. Pipe extensions can be trimmed using a
rotary hand trimmer available from organ supply houses. A chisel can also be
used to “plane” a pipe to length or to straighten a ragged top.
Soldering organ pipes (see Soldering) is a tricky operation owing to the
fact that the melting point of soft solders is precariously close to that of most
pipe metals, and the heat of the soldering iron will be considerably higher
than both the solder and the pipe. In the days of Dom Bedos (1706–1799),
soldering irons (which were generally made of copper rather than iron) were
heated in a coal fire and used in rotation (as one cooled in use, the other was
being heated). Today, temperature-controlled electric soldering irons are
available. These differ from conventional soldering irons used in electrical
soldering in having massive copper tips that retain a great deal of heat and
permit long seams to be smoothly soldered. Despite the element of control
provided by their variable transformers, electrically heated irons used in
soldering organ pipes must still be hotter than the melting point of the solder
and pipe.
To solder a seam or rip, the pipe must first be sized. The sizing prevents
molten solder from spreading beyond the seam or rip and also insulates the
pipe from heat. The size is made with whiting and hot animal-hide glue or
gum arabic, which is prepared by placing one part pulverized glue or gum
arabic in two parts of water, allowing it to swell overnight, heating to around
140°F (60°C), and straining before use. Commercial whiting is sifted into the
hot glue or gum arabic until it reaches the consistency of heavy cream. The
size is then painted on the pipe metal along the seam, making a band
sufficiently wide to protect the pipe from the tip of the hot iron. After drying,
the first coat is rubbed until it turns gray and a second coat is added. Spotted
metal should be sized on both sides; common pipe metal may be sized on one
side. When the size is dry, if one presses a thumbnail against the size, it
should not flake off. One should also be able to cut the sized metal with a
knife or file across it without the size chipping. If it chips, a greater
proportion of glue or gum arabic should be used. When the sizing is
thoroughly dry, the edges of the metal to be soldered are beveled at a 45-
degree angle with a sharp block plane (if the seam is straight) or filed, so that
when the edges are brought together they form a V-shaped trough. Before
soldering commences, a file or abrasive cloth should be used to remove
oxidation from the tip of the iron. The iron can also be rubbed against a block
of sal ammoniac (ammonium chloride) to clean its tip. After cleaning, the
iron is heated, fluxed, and tinned (i.e. coated with a thin layer of molten
solder). Tinning the iron is vital as molten solder will not adhere to it if it is
not prepared in this way. Puddles of solder deposited on a stone or brick slab
are picked up as required by the hot iron. Low-temperature eutectic solder
(63/37 tin-to-lead) may be used; however this alloy passes from solid to
liquid without exhibiting a pasty stage. While this is generally acceptable, a
pasty stage may be advantageous in certain soldering procedures. In his L’Art
du Facteur d’Orgues (1766–1778) Dom Bédos in fact provides instructions
for preparing four different types of solder for a range of tasks: he begins by
making solder having a 6:1 tin-to-lead ratio, which he advocates for soldering
common pipe metal; to this alloy more lead is added to produce solder for
soldering pure tin pipes; additional lead is added to make “turning” solder,
which is suitable for joining pipe feet to bodies, and finally more lead is
added to make solder for repairing reed blocks. Dom Bédos produces these
solders in sequence, pouring some out to form little ingots before adding lead
to make the next batch.
To solder the seam or rip, the V-shaped trough is fluxed with tallow or
drippings from a stearin candle and lightly tacked with bits of solder to hold
it together. Solder picked up from the stone or brick is then drawn along the
seam and drawn along the trough. The tip of the iron should be held at an
angle to the pipe body – the degree of inclination depending upon the heat of
the iron (a very hot iron is held at a steeper angle). Finally, the seam is
refluxed, and the iron is drawn in one smooth motion to unite the solder in a
continuous strip. When soldering around the circumference of the pipe, it is
helpful to have an assistant rotate the pipe, using a cradle to avoid jerky
movement. After soldering, the sizing is washed off with water and a soft
brush, leaving the neatly soldered seam.
If new sections of metal must be added or new pipes made, it may be
necessary to cast sheets of the proper alloy. The composition of pipe metal
may be roughly ascertained from its appearance (high tin and spotted metal
instantly reveal themselves) or more accurately by determining the specific
gravity of a sample. This can be calculated by weighing a sample or pipe in
grams and then dividing its weight by the volume in milliliters, which can be
ascertained by submerging the sample or pipe in a narrow graduated cylinder,
taking care to dislodge trapped air bubbles, and noting the amount of water
that is displaced. Another method is to suspend a sample or pipe from a
balance using a fine thread. The sample or pipe is weighed in air and then
submerged in water and weighed again. The difference is the weight of the
water (one milliliter of water weighs one gram) displaced by the submerged
pipe. To calculate the specific gravity, the weight difference is divided into
the weight of the pipe in air – the quotient is the specific gravity. Töpfer's
four-volume Lehrbuch der Orgelbaukunst (1955–1960 edition prepared by
Paul Smets) provides the following table of specific gravities and melting
points for various proportions of tin to lead:

Ratio of tin to Specific Melting point in degrees

lead gravity Celsius

10:3 7.927 –

3:1 7.994 188

7:3 8.150 –

2:1 8.267 188

5:3 8.408 –
 3:2 8.498 190

1:1 8.864 197

2:3 9.265 215

1:2 9.553 238

2:5 9.770 240

1:3 9.939 252

1:4 10.184 274

1:7 10.594 290

The alloy of pipe metal can also be determined by casting a slug with a pair
of casting pliers and comparing its weight against cast slugs of metals of
known composition (a set of casting pliers and reference slugs can be
obtained from organ suppliers). Because pipe metal is not uniform in
composition (spotted metal is an obvious case in point), analytical techniques
that measure small areas, such as X-ray fluorescence spectroscopy, may yield
wildly inaccurate results.
Wood pipes are generally more rugged than metal pipes, and though they
may get knocked around, scratched, and dented, these marks generally do not
impair the tone of the pipe. Open seams, loose caps, and ill-fitting stoppers
are the problems most encountered. Wood parts should be repaired with hot
animal-hide glue, though to make a good airtight joint, all of the old hide glue
should first be washed off and the joint cleaned. If stoppers must be
releathered, choose skins of the proper thickness or skive them down to
achieve a fit that will allow the stopper to seal well and move smoothly. The
front cap of a wood pipe forms the lower lip and one face of the windway,
and its position is vitally important to the working of the pipe. The grain
direction of most caps runs vertically; in others the grain is oriented
horizontally. In the latter, the caps often shrink or curl away from the body of
the pipe. If this happens, the cap should be unglued and shifted upward so
that it is in proper alignment with the top of the lower block, and then
clamped and glued in place.
Reed pipes are composed of the boot, reed block, shallot, tongue, wedge,
tuning wire, and resonator. The boot and block are usually made of pipe
metal, though they are sometime made of wood. The shallot is generally a
tapered tube of brass, capped at the bottom and having the front cut away to
provide a seat for the tongue. The tongue, generally made of brass, is held in
place over the front of the shallot by a wedge of wood. The tuning wire
passes through a hole in the block and is bent in such a way that it bears
against the tongue. A common problem with reed pipes is instability of the
parts assembled in the reed block. If the wedge does not fit snugly, the tongue
may shift around during tuning, or the tuning will be unstable, requiring
frequent attention. It is important to make sure that the tongues are well
secured by properly fitted wedges.
Reeds that do not speak may be fouled with dirt. Sometimes raising and
lowering the tuning wire will allow dirt particles to fall away. A soft brush
(camel hair or sable) can be used or a thin sheet of paper may be drawn
between the tongue and the shallot face to clear debris. If necessary, the
entire reed assembly may be disassembled and blown or brushed out. If the
tongue does not seat properly because of localized corrosion, the corroded
area can be smoothed with fine emery or crocus cloth. If this occurs on the
shallot's face, take care to preserve its flatness or curvature. Corroded brass
can be treated chemically by immersion in a solution of the trisodium salt of
EDTA (see Brass and nickel silver cleaning).
If tongues are misshapen, they can sometimes be gently bent back into
shape by hand, using neighboring tongues as guides. More serious bends or
creases may need to be peened or burnished on a reed block (Figure 24),
though this may harden the metal and alter its acoustical properties. Severely
damaged tongues are generally replaced. Stock for making reeds is available
from organ suppliers, though if the correct thickness is not available, the
closest available gauge can be drawfiled or rubbed against a sanding block to
bring it to the proper dimension. A rolling mill should not be used to alter the
thickness of a reed blank because each pass will harden the metal somewhat.
When cutting out tongues from sheet stock, use a straightedge and a stout
knife to score the brass. The sheet can then be clamped in a smooth-jawed
vise or between wood chops with the score mark on the edge of the vise jaw
or chop, and the tongue broken away. A special reed guillotine can be used,
though a heavy-duty guillotine-type paper cutter may be used for lighter
tongues. Final fitting is accomplished by holding the tongue up to the shallot
face and marking it with a scriber, then clamping it between wood chops and
planing or filing it to shape. The surface that beats against the shallot must be
perfectly smooth and finely finished.
The curvature of a new tongue should be modeled closely on its neighbors.
The curve is formed by drawing a steel burnisher down the length of the
tongue while it is supported against a flat or curved steel reed block. The
degree of curvature varies with the type of stop and voicing; soft-toned reeds
such as the oboe generally begin with a gentle curve that increases at the tip
while trumpet reeds typically begin more sharply. As a rule, the stronger the
curve, the louder the tone and the slower the speech; the weaker the curve,
the softer the sound and the quicker the speech. However, if curved too
strongly, the tongue may not beat against the shallot and there will be no
sound; if too weak, buzzing and rattling may occur or the tongue may remain
fixed against the shallot, again resulting in no sound. It may be necessary to
alter the burnishing pressure as one moves down the tongue to achieve the
correct curve. Whatever the ultimate curve, the mounted tongue should make
continuous contact with the shallot face when it is pressed down with a
fingertip at its extreme end. This can be checked against a strong light: if
there is any discontinuity, curve reversal, or lifting away from the shallot, the
light will shine through and reveal the flaw, which can generally be corrected
by burnishing or direct manipulation by hand.
Bass reeds are sometimes weighted to lower their resonant frequency. Bits
of lead may be soldered to the end of the tongue, though around 1900,
“Chatterton compound” was sometimes used to attach felted lead weights.
This compound was formerly used by electricians and consisted of 3 parts
gutta percha, 1 part rosin, and 1 part Stockholm tar. If encountered, it can be
remelted through the gentle application of heat to reseat loose weights.
The free reeds of harmoniums are shaped differently than beating reeds.
They are sometimes given an S shape along their length and may have a
slight twist at the free end to make the tone more mellow. Free reeds are
typically riveted to their frame. Loose rivets will cause the reed to chatter; to
alleviate this problem the rivets may be tightened by hammering them with a
steel punch. The chattering also may be caused by the reed touching the
reed's frame. If the reed has shifted, it may be straightened relative to the
frame by holding a flat tool against the riveted end and giving it a light tap. If
the reed is bent, it should be carefully straightened, taking care not to alter the
intended curve. Neighboring reeds should always be examined and used as
reference when adjusting the curvature of a damaged or replacement reed.
If pipes or reeds are missing, it is necessary to make measurements of
metal or wood thickness, length, width, circumference, mouth width, mouth
height, and all other vital dimensions of pipes in the rank, and then plot these
dimensions on graph paper. In this way, the dimensions of the missing pipes
or reeds can be interpolated. If facilities are not on hand for casting pipe
metal and fabricating the pipes, the specifications can be sent to a pipe maker.
Pipes can be made slightly over length and with the upper lip left uncut so
that the restorer can tune and voice the new pipe to match its neighbors. A
proportional divider is used to establish the cut-up of the upper lip, which is
generally trimmed with a sharp knife.

Keywork

Mechanical-action organs may have balanced or suspended keyboards (the

latter widely used in French Classical organs). In the balanced action, the key
levers are pivoted centrally, as in most keyboard instruments, with the
trackers hinged to the back of the key lever. In the suspended action, the keys
pivot from the back end, and the trackers are hinged centrally.
Even in small cabinet organs, key levers may be of sufficient length to be
adversely affected by warping. If keys rub against each other, they may be
straightened by the application of heat, care being taken that the keys are not
scorched, and flexed back into shape. A hot-air gun, bending iron, or the
radiant heat from a hotplate or alcohol lamp can be used to heat the key lever.
The wood will need to be brought to a temperature that is too hot to hold, so
insulated gloves should be worn to protect the hands. As a last resort, it may
be necessary to plane or trim keys down to eliminate rubbing, though one
should be aware that removing wood reduces key strength and alters weight,
balance, and flexibility.
 The balance or fulcrum pins should operate smoothly within their key
mortises. If the front of the key is grasped between the thumb and forefinger
and pulled back and forth, there should be no detectable motion. If the key is
wiggled from side to side, a slight motion is acceptable, but the key should
not clatter or touch a neighboring key. Twisting should also yield minimal
movement. If the mortises are tight, a sharp, pointed knife can be used to
shave away the excess wood in order to ease the key. If the balance mortises
are badly worn, a traditional method involved making incisions at either side
of the mortise with a thin knife and then gluing narrow wedges of wood into
the cuts to restore tightness. If the wear is too extreme for this technique, thin
slips of wood can be glued against the mortise walls, though the walls may
have to be trued with a knife or chisel to provide good gluing surfaces for the
shims. The shims are then glued in, using wood wedges to temporarily hold
them in place until the glue dries. It is best to fit shims that are thicker than
necessary, and then trim them with a knife or chisel until a perfect fit with the
pin is achieved. If keys are bushed with cloth or leather, it may be necessary
to rebush them with similar material to restore the correct fit and feel (see
Piano action regulation and voicing [modern grand]).
Unlike harpsichord and piano keyboards, organ keyboards typically do not
have a back rail. Assuming the balance rail is true and all balance-rail
punchings are the correct height, the keys can often be leveled by turning
leather nuts threaded on the tapped wires of trackers or stickers located at the
back ends of the key levers. Some organs are fitted with a thumper rail,
which consists of a weighted bar that is laid on top of the key levers in front
of the balance pins. This device is also used to prevent keys from bouncing.
To increase or decrease the key dip of the entire keyboard, first prop up or
remove the thumper rail if one is present, adjust the height of the top and
bottom natural and accidental keys using the leather nuts attached to trackers
or pull-down wires at the ends of the levers, and then use a straightedge to
adjust the remaining keys. After the thumper rail is put back in position,
check that none of the pallets is being pulled open when its key is at rest. If
so, adjust the appropriate tracker or sticker (generally by turning a leather
nut) until the pallet is firmly seated. If the thumper rail is lightly pressed, no
notes should sound.
The system of linkages in an organ is often complex, but it is made up of a
series of mechanical devices that are quite simple in their design and
function:
Trackers are generally thin strips of wood that work by pulling. Tapped
wires (often of brass) are usually fitted at both ends of trackers to connect
them to other parts of the action. One end of each wire is bent at a right angle
for insertion in a hole drilled in the tracker. After the wire is inserted, linen
twine is whipped (i.e. tightly wrapped) around the end of the tracker, and
painted with shellac to bind it together and prevent it from unravelling. The
free end of the wire should be threaded with a die, preferably before it is
bound to the tracker. Engagement nuts can be made from firm sole leather
using an arch punch. These should be drilled slightly undersize before
attempting to thread them onto the wires, but they should not be so tight that
undue stress is placed upon the tapped wires, as these are apt to break off.
Cloth washers may be used between the leather nut and the linkage to prevent
noise.
Stickers are dowels or square sections of wood that work by pushing. Pins
are set into the ends of the stickers, which allow them to link with other parts
of the action. Stickers often run in drilled or mortised registers.
Backfalls are arrays of wood or metal slats that pivot centrally from a rail
or frame. They reverse the motion of whatever mechanical device (such as a
tracker or sticker) pulls or pushes them. They typically direct the motion
some distance and may be fanned out. Double backfalls are frequently used
in octave couplers. Backfalls may pivot from axles or vertical pins like those
used in aligning key levers.
Squares are made of single pieces of wood shaped like isosceles right
triangles and pivoted at the corner, or constructed by mitering two pieces of
wood to form an L-shape with an axle inserted at the joint. Squares are used
to change the direction of movement, such as converting up-and-down to
side-to-side or front-to-back motion.
Rollers are wood or metal rods with pivot pins driven into each end. The
pivot pins are supported by brackets or spade-headed screws. Arms attached
to both ends of the roller allow for motion to be offset laterally. Rollers are
often used to convert from key-scale to wider chest-scale. A roller board is a
device (generally a plank of wood) for mounting a series of rollers.
Trundles are rollers mounted vertically. They are often employed in stop
actions. Parts may be of wood, though metal is frequently employed for
strength.
Keyboard and pedalboard couplers allow a key or pedal of one division to
operate the corresponding key or pedal of another. These allow stops of
different divisions to be combined. The simplest system involves fixed or
sprung coupler dogs that engage and disengage when an upper keyboard is
slid back and forth. With fixed keyboards, stickers, or tumblers, between the
manuals may be brought into position by shifting a register that maintains
them in alignment. Tumblers or drum-stick couplers, which are similar in
action to stickers, are used to couple manuals that are in close proximity to
one another. When regulating couplers, some degree of lost motion should
remain, otherwise problems may arise (such as the binding of parts or
ciphering) when the couplers are engaged or disengaged.
Nearly every organ has its own unique arrangement of mechanical parts,
but most depend upon combinations of the above devices for conveying
motion from key to pallet and stop knob to slider. Regulation is often carried
out by adjusting the positions of leather nuts, which provide a means for
changing the effective length of various linkages. The disintegration of
leather nuts will result in their sliding out of position or even falling off the
tapped wires. When this happens, notes will not sound. If this is the case,
replacing the nut will solve the problem. Only in the very simplest of organs,
such as regals and melodeons (where hinged keys sit directly over the pallet
valves and motion is conveyed by short stickers made of wood or metal),
may there be no ready provision for adjusting the interaction between keys
and pallets. If so, it may be necessary to glue bits of cloth or leather to the
underside of the key where it makes contact with a sticker.
The squeaking of mechanical parts is often a vexing problem. Tallow was
traditionally used to alleviate squeaks or other noise caused by metal or wood
parts bearing against each other, though clock-spring grease is a good
alternative.

Voicing

The voicing of pipes is a complex art. Once a pipe has been straightened
and the dents have been removed, then the careful adjustment of the upper
and lower lip and of the position of the languid may commence. As indicated
above, many organs contain pipes that were drawn from different instruments
and may even be a collection of entire ranks that were originally voiced at
different wind pressures or designed to speak at different pitches. In general,
the cut-ups of flue pipes of the same pitch in different ranks of the same
division should be at approximately the same height. If this is not the case, it
may indicate that ranks have been assembled from different divisions or
sources, and it may be impossible to get them to work together in an artful
manner. It is often the case that a conservator is given the hopeless task of
“restoring” a composite instrument that never worked well in the first place.
In the discussion that follows, it is assumed that the instrument is not a
composite and that the pipework is original to the instrument.
Flue pipes work by directing a stream of air at the edge of the upper lip of
the pipe's mouth. In metal pipes this sheet of air is created by the languid and
the lower lip, which are in close proximity and form a narrow slit. In wood
pipes the sheet of air is created by a gap between the lower block and its cap.
The pitch of a flue pipe is determined by the resonant frequency of the air in
the resonator (i.e. the pipe body), which interacts with the thin stream of air
generated at the pipe mouth and causes it to vibrate back and forth and create
a sound. The resonant frequency of a closed, or stopped, pipe is
approximately half that of an open pipe of the same length.
The timbre of a pipe is determined by a number of factors. Open,
cylindrical flue pipes produce even- and odd-order harmonics, while closed
cylindrical tubes produce the odd-order harmonics. The “scaling,” or ratio
between the length and the diameter or width of a pipe, is an important factor:
wide-scaled pipes produce a mellow, flute-like sound due to the lack of
development of higher harmonics, while narrow-scaled pipes produce a
string-like sound due to the presence of these harmonics. The science (or,
more properly, the “art”) of establishing the rate at which pipe diameters
increase or decrease as one moves down or up the rank is documented in an
extensive literature extending back to the Middle Ages (see Scaling). Pipes
are made in a wide variety of shapes in order to produce different timbres,
often in imitation of various types of wind instruments. Other factors
generally affecting timbre include:

1 The ratio of the width of the mouth to the circumference of the pipe (a
 wide mouth tends to produce greater volume and enhances the
fundamental, while a narrow mouth produces a softer sound with
greater development of harmonics).
2 The height of the mouth (a high cut-up tends to cause the tone to
become flutelike while a low cut-up generally yields a brighter
string-like tone).
3 The shape and position of the languid (a strongly beveled front edge
produces a brighter sound), while its height affects the quickness of
speech (raising it slows the speech while lowering it quickens the
speech).
4 Nicking of the languid (no nicking or light nicking produces a rougher
sound that is slow in speech and tends to begin with a “chiff,” while
heavy nicking smooths the tone, though if excessive may dull it).
5 The shape of the lower lip (a vertical lower lip coupled with a squarish
languid tends to produce a soft, flutelike timbre, while an incurved
lower lip and a sharply beveled languid produce a more strident
tone).
6 The shape of the upper lip (a sharply cut and burnished upper lip
encourages overtone production while rounding and leathering it
reduce overtones).
7 The use of “ears,” “beards,” “roller beards,” and “harmonic bridges”
(these help maintain the direction of the airflow and tend to stabilize
pitch and prevent overblowing).
8 The thickness of pipe walls and the tin/lead content (thin metal pipes
cannot be blown as hard and are typically used for softly voiced
stops; high-tin alloy and hammered pipe metal tend to reinforce
higher harmonics).
9 The shape and dimensions of the pipe foot (an open toe admits more
wind and produces a louder and more strident tone; a closed-in toe
produces a softer tone).

When conserving an organ one should not revoice ranks or attempt to change
an instrument's sound. For example, in the late nineteenth century, “chiff,”
which was so much treasured in the Baroque period for the articulation it
provided, was considered a defect. Consequently, pipes exhibiting
pronounced “chiff” were nicked to soften their attack. Certainly, it is
inappropriate to “nick” pipes in the course of conservation work; however,
burnishing or attempting to fill in nicks should be given careful consideration
because it may further damage the pipe and may even be inconsequential if
the pipe has been altered in other ways. It is sometimes necessary to accept
changes that have been made to an instrument as part of its history (see
Ethics).
The following faults in flue pipes may be correctable by simple voicing
techniques: slowness of speech, overblowing the octave, excessive loudness
or softness, windiness, hollowness, excessive reediness, excessive flutiness.
Pipes may be slow of speech or fail to speak if there is gross misalignment of
the upper and lower lips. This may be due to the pipe having been twisted as
a result of improper cone tuning technique (tuning cones should be pressed or
lightly dropped into the pipe, rather than twisted). This problem may be
alleviated by manually untwisting the pipe. Mouth alignment should in any
case be checked by sighting down the pipe. In general, the upper and lower
lips of the mouth should be aligned with each other, and the languid should
be flat and in line with the lower lip. Sometimes the lower lip may become
slightly convex and needs to be pushed in to straighten it. Convexity (or
concavity) may be due to the pipe being out-of-round. If this is slight, hand or
finger pressure may be sufficient to round up the pipe. If extreme, it may be
necessary to use mandrels to round up the pipe.
Slowness of speech can often be corrected by directing the upper lip
outwards or by lowering the languid. If the lower lip is pressed in too far, no
sound may develop, and this can be corrected by pulling out the lip with a
lancette or lip tool. A flat chisel or lip tool is useful for straightening or
redirecting the upper or lower lips; rods inserted either through the foot or
resonator may be used to raise or lower the languid.
If the pipe sounds an overtone before settling into its fundamental, or if the
fundamental does not develop, the pipe is said to overblow. Pushing the
upper lip inward or raising the front edge of the languid will generally correct
this. The problem may also be caused by too much wind reaching the pipe,
which may be resolved by closing up the foot with a toe cone or knocking-up
cup. If a pipe is too soft, the foot can be opened with a tapered mandrel. With
wood pipes, strips of wood may be wedged into the foot or removed to
regulate volume.
Windiness is sometimes caused by air rushing past rough edges or
obstructions. If the roughening has been caused by damage, edges may be
smoothed over with a burnisher. Dust or obstructions in the mouth or
windway may also be the cause. Windiness can sometimes be reduced by
pushing in the lower lip. Very light nicking, or “feathering” will often reduce
windiness, but this should be avoided in conservation work as it alters the
original tonal characteristics of the instrument.
Reeds may be softened or strengthened by adjusting the curve of the
tongue. Slowness in speech is often due to excessive curvature, which may be
reduced by gently burnishing the striking side of the reed. Overblowing is
often due to insufficient curvature, which again can be corrected by
burnishing the outer surface. A replacement tongue that is too thick may
produce a dull sound and the attack may be slow. A tongue that has been
made too thin may produce a harsh sound or rattle. Adjusting the length of
the resonator (by use of tuning flaps, slots, or rolls) is part and parcel of reed
voicing. Though resonator length is generally not changed to alter the tuning
of a reed pipe (this is accomplished by adjusting the length of the reed tongue
by manipulating the tuning wire; see below), the resonator must be in
reasonable tune with the reed. Subtle changes in the resonator's effective
length can be made to adjust the pipe's volume, though a resonator that is too
long or too short will cause the pitch of the reed to be unstable. If a resonator
is shaded at the top and the pipe does not return to its proper pitch when the
obstruction is removed, the resonator's tuning flap or roll should be adjusted
to bring the resonator into better tune with the reed.

Tuning (see also Tuning and temperament)

Before attempting to tune an old organ it is important to assess its pitch and
temperament. If the organ is in a reasonably good state of preservation, it
may be possible to recognize the temperament (see Tuning and
temperament). For purposes of documentation, it is advisable to measure and
record the pitch of each note of the principal and other major ranks with an
accurate electronic tuner, particularly if the pitch and temperament cannot be
firmly established prior to restoration and tuning. Before measuring the pitch
or temperament, check that all of the pipes are properly racked, well seated,
and standing vertically. The mouths of nondisplay pipes generally face
outward, though due to crowding, it may be necessary to reorient some to
prevent them from being shaded or from blowing into the mouths of their
neighbors. If the wind supply is not operational, the pipes can be mounted on
a voicing jack to document pitch. A voicing jack can be improvised by
attaching a hose from the exhaust of a vacuum cleaner to an airtight box that
is fitted with a manometer, an adjustable valve (such as a hole with a sliding
cover) to regulate the air pressure in the box, and a countersunk opening to
plant the feet of the pipes.
When tuning an organ, the principal, prestant, or Octave 4 foot of the
Great or principal manual division should be the first rank checked, for it is
from this rank that all of the other ranks are tuned. It is of the utmost
importance that the original temperament be retained, and for that reason one
hopes that the principal has not been damaged or strayed significantly from
its original temperament. Before any pipe is adjusted in this rank, be certain
that it is the one that is out of tune. To determine this, begin by making all of
the interval tests that are appropriate for the most likely temperaments in
order to establish any wayward notes. After carefully adjusting the
temperament octave, tune upwards and downwards by octaves until the entire
rank is tuned. Carefully recheck the temperament octave and the entire rank
before continuing, otherwise any errors will be duplicated in the other stops
and divisions. After tuning the principal of the Great, those of other divisions
should next be brought into tune with it, and then the other ranks in each
division with the principals of their respective division. It may not be
possible to tune soft-voiced ranks directly from the principal, as they may be
overwhelmed or drawn into tune by this loud stop. In such cases a less
brilliant rank may be used to tune them. Soft-voiced ranks should be gone
over separately to check that they are in tune with themselves. When tuning
mixture ranks, the pipes that are not being tuned should be silenced with a
mute (i.e. a fringe of cloth mounted on a handle). Reeds are generally tuned
last. In general, one should tune one's way out of an organ to avoid reaching
over previously tuned ranks.

Tuning technique

The tuner requires an assistant to press the keys and operate the stops, as
well as a pumper to supply the air if there is no electric blower. Pipework is
susceptible to the heat of the tuner's hands and body (holding one's hand
close to the resonator of a sounding pipe will cause it to go flat due to
expansion of the pipe metal due to the radiant heat), so it is important to keep
clear of it as much as possible. Cloth or leather gloves should be worn as
insulation if pipes must be handled. Open flue pipes are tuned by several
means. Most small-diameter pipes are generally tuned with tuning cones
(Figure 24). To sharpen a pipe, the pointed end of the cone is carefully held
directly above and pressed or dropped into the pipe, thereby causing the pipe
to flare out and rise in pitch. To flatten the pitch, the hollow end of the cone
is used to constrict the pipe, thereby lowering its pitch. Tuning cones should
not be twisted as this may cause the pipe mouth to become distorted. Cones
may be lightly lubricated with petroleum jelly prior to use. Frequent cone
tuning takes its toll on pipes, and excessive pressure can distort a pipe or even
rip open a seam. One solution to the wear and tear of cone tuning is the
installation of tuning sleeves. Larger open pipes that cannot be cone-tuned
are fitted with sleeves, flaps, rolls, or slots. These should be adjusted with the
appropriate tools (Figure 24) rather than by hand. Stopped pipes are tuned by
gripping the handle of the stopper and raising or lowering it as required.
Reeds are adjusted with a special reed tuning knife, which is used to tap the
tuning wire up or down. To ascertain whether an open flue pipe is sharp or
flat, the top of the pipe can be shaded. This will flatten the pipe and enable
one to determine where it stands. A stopped pipe can be checked by shading
its mouth. As the ambient temperature increases, the pitch of labial pipes rises
because the air becomes less dense, while the pitch of lingual pipes falls due
to the expansion of their reeds; thus, the labial and lingual ranks will go out
of tune with one another. Though the lingual pipes are generally less affected
by temperature change, it is the reeds that are often tuned to the flue pipes
because of the relative ease of manipulating their tuning wires. To further
complicate matters, humidity fluctuations cause wood pipework to go out of
tune relative to metal pipes. Thus, it is unwise to tune an organ at times of the
year when the temperature and humidity are in the process of changing.
Furthermore, organs should never be tuned when sunlight is striking the case
or pipes. Swell shutters should be opened the day before tuning to allow the
temperature of the enclosed pipes to reach that of other divisions.

References
J. Adlung, Musica Mechanica Organoedi (Berlin, 1768).
Paul-Gerhard Andersen, Organ Building and Design (New York, 1969).
George Ashdown Audsley, The Art of Organ Building (New York, 1905).
François Bédos de Celles, L’Art du Facteur d’Orgues (Paris, 1766–1778).
Thomas Elliston, Organs and Tuning: A Practical Handbook for Organists
(Sudbury, Suffolk; 1898).
A. Hemstock, On Tuning the Organ (London, 1924).
Walter and Thomas Lewis, Modern Organ Building (London, 1939).
L. G. Monette, The Art of Organ Voicing (Kalamazoo, 1992).
Henry Nicholson, The Organ Manual for the Use of Amateurs and Church
Committees (Boston, 1879).
Organ Voicing and Tuning (Cincinnati, 1881).
Johann Gottlob Töpfer, Lehrbuch der Orgelbaukunst, 4 vols., ed. Paul Smets
(Mainz, 1955–1960).
Andreas Werckmeister, Orgelprobe (Quedlingburg, 1698).
Reginald Whitworth, The Electric Organ (London, 1930).

Overspun strings

The formula (Rose, 1995) for the string tension of overspun strings is:

where

T is the tension in newtons

ρc is the density of the core in kg/m3
ρw is the density of the winding in kg/m3
dc is the diameter of core in millimeters
 dw is the diameter of winding in millimeters
W is the winding pitch in millimeters (for close-wound strings, W = dw)
L is the string length in millimeters
F is the frequency in Hz.

Making overspun strings

Overspun strings are used in pianos and in some clavichords and

harpsichords. So-called “open-wound” strings, which have a helical
wrapping, are commonly found in late eighteenth- and early nineteenth-
century English square pianos; “close-wound” strings, which have a tight coil
of wire covering the core, are found in virtually all modern pianos, though
they were also used in some early nineteenth-century fortepianos. To make
both types of overspun strings, the core wire must be rotated at both ends to
prevent it from twisting while the covering wire is wrapped over it.
The author's string-winding machine is loosely based upon a description
provided in Gall's Clavier-Stimmbuch (Vienna, 1805), which uses gears
attached to the ends of a long rod (mine is half-inch diameter steel) which
mesh with gears at either end that are fitted with hooks (mine are formed
from half-inch steel rod, turned down, annealed, and bent with a pair of pliers
to form hooks). All four of the spur gears have the same pitch diameter (3½
in.) and number of teeth (84). One of the gears is mounted alongside a pulley
that is run by another pulley chucked to an electric hand drill. The pulleys are
connected with 3 mm diameter jeweler's lathe belting (this is cut to length
and joined by heating with an alcohol lamp or cigarette lighter). Most electric
hand drills have enough speed and torque for this application. The gears run
on pillow block ball-bearings, which are screwed to the ends of a 2in. × 8in.
× 100in. plank (the gears overhang the ends of the board). A foot-operated
speed control from a Foredom flexible shaft tool enables me to start up
slowly, work up to speed, and slow down to a controlled stop at the end. The
electric drill I use has a reversing switch, which is handy and allows me to
back up if a winding problem develops. When attaching the core wire, the
hitchpin loop is fixed to the hook at one end; the string is drawn around the
hook at the other end using a spring gauge to adjust the tension, and is then
tied off. I generally use somewhat less tension than the final tuning tension so
that the final stretch will tend to tighten the winding. Before applying the
covering, I lightly run a fine half-round file the length of the core to provide
it with a little tooth, which helps grip the covering. A uniform open winding
can be achieved by “chasing” the winding with the fingers of one hand while
using the thumb of the same hand to direct the covering wire at a constant
angle. In this way one can maintain a uniform winding pitch despite the
various amounts of deflection of the core during spinning. In early English
square pianos, a brass core was often used with an overwinding of copper.
One sometimes encounters tinned copper overwinding. Copper wire can be
tinned by boiling it with tin filings in a solution of cream of tartar. If one
employs a winding of brass, it should be soft, or annealed brass. Other
combinations of metals have been used, though for best results, the
overwinding material should be annealed (see Metalworking).
When I was apprenticed to the harpsichord maker John Challis in the early
1970s, he used a much simpler device for making the overwound strings. His
winding machine employed a motorized hook and a separate hook running on
a ball-bearing that was simply clamped at the other end of the workbench.
There was no connection between the hooks other than the core wire. This
system seemed to work for the extremely delicate overspun strings that he
used in his clavichords, many of his harpsichords, and even some of the
strings of the 2-foot stop found in one of his pedal harpsichords (which were
overspun with 0.0065-inch annealed brass). However, such a spinning
machine does not work with the heavier strings that one encounters in square
pianos.
In Diderot's engraving of a luthier's shop, there is a depiction of a string
winding machine that resembles a spinning wheel. The hand-operated large
wheel turns a hook mounted on a smaller wheel, which greatly increases the
spin rate. Though this device was probably used to make covered gut strings
for bowed string instruments, like Challis's spinning machine, the far end of
the string was not geared to turn at the same rate as the near end; instead, the
far end of the string was weighted, draped over a fixed hook, and allowed to
spin at its own rate. Modern bowed string instrument strings use a variety of
covering materials (including silver, aluminum, and tungsten) over gut or
synthetic cores.
In instances where the coils in old piano strings buzz or the strings have
gone dead, restorers generally remove and replace them, but I have found that
they can be sometimes be salvaged with a light application of thinned
nitrocellulose lacquer applied with a small watercolor brush. This binds the
winding to the core. The lacquer dries almost immediately and is invisible.
Incralac is a brand of lacquer that contains benzotriazole, which protects
copper and brass from corrosion.
Twisted wire strings may have preceded the development of overspun
strings. I discovered evidence of the use of such strings in the 1581 Ruckers
double virginal (muselar) in the Metropolitan Museum of Art (Pollens, 1997,
1998). Unfortunately, the two surviving twisted strings, which had been used
for short-octave notes, were discarded by someone working in the museum
before I was employed there. However, I was able to reconstruct the
diameters and twist rates using a microscope and the original 8in. × 10in.
negatives that were made shortly after the instrument was acquired in 1929.
The Ruckers double virginal was discovered in a chapel in Peru, where it sat
undisturbed for hundreds of years, and there is documentary evidence that it
arrived there from Antwerp in 1582. Some New World woods were used to
make some early repairs, and thus there has been some speculation that the
twisted strings were installed in the New World, but I believe that the idea of
twisted strings, several of which were recently discovered on a Mexican
clavichord of undetermined date, originated with early Old World
instruments that were imported to South and Central America. It is widely
believed that overspun strings were invented around 1660, through Michael
Praetorius (1619) mentions gedrehete (spun or twisted) brass strings used as
sympathetic strings in a particular type of bowed stringed instrument made in
England. Twisted strings are easy to make by doubling a length of wire
around a hook chucked to a hand drill; an elaborate string winding machine is
not required. I saw such strings in a “baby grand” fortepiano by Carl Lemme.
These were replacements made by the instrument's restorer, Horst Rase.
Twisted strings, perhaps original, have been reported in other fortepianos.
The twisted strings on the 1581 Ruckers instrument were probably made of
red brass. The C/E string had a total diameter of 1.12 mm, with component
strands approximately 0.56 mm, and a pitch of 4.2 mm, while the E/G♯ string
had a total diameter of 0.95 mm with component strands of around 0.47 mm
and a pitch of 7.4 mm.

References
Stewart Pollens, “Flemish Harpsichords and Virginals in the Metropolitan
Museum of Art: An Analysis of Early Alterations and Restorations,”
Metropolitan Museum Journal 32 (1997), pp. 85–110.
Stewart Pollens, “Early Alterations Made to Ruckers, Couchet and Grouwels
Harpsichords in the Collection of the Metropolitan Museum of Art,”
Kielinstrumente aus der Werkstatt Ruckers (Haale an der Saale, 1998), pp.
136–170.
Michael Praetorius, Syntagma Musicum 2: De Organographia (Wolfenbüttel,
1619), p. 47.
Malcolm Rose and David Law, A Handbook of Historical Stringing Practice
for Keyboard Instruments (Lewes, 1995).
 P
Paper, pencil, and ink

Paper

Paper was invented in China around 100 CE, or possibly earlier, and was
introduced into Europe in the twelfth century. Prior to the invention of the
paper-making machine in France in 1799 (Henry Fourdrinier's English patent
dates from 1806), paper was made by casting macerated rags in a rectangular
mold, termed a deckle, fitted with a brass screen that consisted of closely
spaced wires crossed by widely spaced ones. This produced so-called “laid
paper,” which when held up to the light shows closely spaced “laid lines” and
widely spaced “chain lines,” which are oriented at 90 degrees to the laid
lines. Laid paper has been made in Europe since the late fourteenth century.
So-called “wove” paper (without the conspicuous laid and chain lines)
became popular in Europe around 1750 and in America around 1800. The
casting molds employed in making wove paper use a screen of fine brass wire
that is woven like cloth. Because of its smoother surface, this kind of paper
was favored by John Baskerville (1706–1775) and other printers because it
enabled metal type to produce a cleaner impression. Watermarks are
produced by stitching designs made of wire onto the casting screen. A
“deckle edge” is an untrimmed edge of a sheet of cast paper. Today, the
deckle edge is often produced artificially on machine-made paper by sawing
(rather than guillotining) through a stack of paper or by moistening individual
sheets and tearing them against a ruler. The cotton gin was invented in 1793,
which led to the gradual replacement of linen with cotton fiber in the making
of rag paper. The paper-making machine enabled long sheets to be made by
linking together a number of casting molds. Most paper is now manufactured
in continuous rolls, which are cut into large sheets of various standard sizes
depending upon the type of paper. Paper made in rolls has a distinct grain
direction and does not fold or tear evenly against the grain. To produce
smaller sheets for folding (such as leaflets or personal stationery), the grain
direction must be properly oriented when cutting from large stock. Alum-
rosin size was first used in 1807, while clay and barite fillers were introduced
in 1810; these coatings help control the absorption of ink and improve
printing and writing properties. Paper made with ground wood pulp appeared
around 1840; it was refined by the soda process around 1850, the sulfite
process in 1857, and the sulfate process in 1884 (the latter processes greatly
increased the longevity of wood pulp paper). Foxing is a brownish
discoloration of paper due to fungal action. While it can develop in books
stored under damp conditions, the root cause may be the deposition of mold
spores during the manufacturing process when wet sheets were turned out
onto felt pads for drying.
Paper documents should be stored in a relative humidity of between 50 and
65 percent and at a temperature of around 20°C (68°–70°F).

Bleaching paper (see Bleach)

Iron-gall ink inscriptions, dyes, and certain pigments may be adversely

affected by bleaching, and wood pulp paper should not be subjected to
chlorine-based bleaches, including Chloramine-T (see below). Foxing and
many types of stains in both wood pulp and rag paper may be bleached with a
weak hydrogen peroxide solution (3%). Hydrogen peroxide that is purchased
from chemical supply houses and pharmacies is generally stabilized by the
addition of an acid. To activate the release of oxygen, the acidity must be
neutralized and the hydrogen peroxide made slightly alkaline (to a pH of
about 9), which can be accomplished by adding household-strength ammonia.
The ammonia should be added gradually with a pipette and the pH checked
with a pH meter. Prior to immersion in the hydrogen peroxide solution, it is
recommended that the paper be washed with distilled or deionized water and
de-acidified by immersion in a saturated solution of calcium hydroxide if it
contains no susceptible dyes, pigments, or inks. After bleaching and thorough
washing, the paper can be re-immersed in a weaker solution of calcium
hydroxide (about 1 gram per liter of distilled or deionized water; adjust to
pH=9) and then air-dried. This will hinder acidification because the carbon
dioxide in the air will react with the calcium hydroxide to form a calcium
carbonate buffer in the paper. If stains are localized and it is not necessary to
immerse the entire paper object, hydrogen peroxide may be applied by lightly
rolling with a moistened cotton-tipped applicator. Oil stains can sometimes
be removed with mineral spirits or ethanol, again gently rolled on with a
cotton-tipped applicator. With any mechanical action, care must be taken not
to abrade the paper's surface.
The following procedure can be used for bleaching rag paper with
Chloramine-T:

1 De-acidify the paper by immersion in a saturated solution of calcium

hydroxide (approx. 2 grams per liter of distilled or deionized water).
2 Then immerse in 2–5% Chloramine-T. The solution should be
prepared immediately before use and used fresh. Immerse for no
longer than ten minutes.
3 Immerse in an antichlor, such as 1% sodium borohydride or 2–5%
sodium thiosulfate, sodium bisulfite or metabisulfite solution.
4 Wash thoroughly after treatment in distilled or deionized water, using
a pH meter to check for acidity in the wash water. If necessary, de-
acidify and wash again.

Pencil

From the fifteenth through the nineteenth century the word “pencil” was
often used to denote an artist's brush. One of the earliest references to
graphite pencils is found in Johann Mathesius's Bergpostille published in
Nuremberg in 1564, though sticks of graphite may have been used for
drawing as early as 1500. Prior to the invention of the lead pencil (which in
fact is a misnomer as it does not contain lead), drawings were done with
wood or bone styli tipped with lead metal, which rubbed off and left a dark
mark similar to that of graphite. Silver point was also used from the fifteenth
century for drawing, though the line it created was initially a very light shade
of gray that darkened with time as the silver turned to silver sulfide. It was
not until around 1790 that graphite pencils came into wide use as a result of a
manufacturing process developed by Nicolas Jacques Conté in France. Pierre
François Tingry (1803) mentions the lead pencil and the use of caoutchouc
(natural rubber) erasers. Today, graphite pencils are made by mixing graphite
and clay with water to form a paste that is extruded into rods and then fired in
a kiln; the hardened rods are then sandwiched between pieces of wood or
used in mechanical holders. Pencil “leads” are graded by hardness, 9H being
the hardest and 9B the softest. The hardness is increased by increasing the
proportion of clay to graphite. The best method of removing pencil marks on
paper and wood remains the rubber or plastic eraser. Graphite is difficult to
remove from porous materials such as stone because it is a lubricant, and
rubbing or brushing tends to drive it into the surface. Alcohol is sometimes
useful in loosening binders in pencil lead and thus removing pencil marks
from nonporous surfaces, such as metal.

Ink

Carbon-particle ink (made with lampblack or finely ground carbon or

charcoal) has been used since ancient times. India ink contains fine carbon
particles suspended in water. Traditional printer's ink consists of carbon
particles suspended in oil. Abraded, charred, or otherwise difficult-to-
decipher inscriptions written with inks composed of carbon particles (such as
lampblack, bone black, ivory black, and vine black) can sometimes be read
using special equipment designed for infrared reflectography or conventional
film and digital cameras (see Inscriptions, faded).
Iron-gall ink has been used for writing since the Middle Ages. Because
freshly made iron-gall ink often lacked deep color when it was first prepared
(its color tends to deepen over time), dyes such as indigo were sometimes
added to it. Iron-gall inks are acidic due to residual tannic acid from the galls
used in making the ink. The acid is highly corrosive to paper and often eats
through it under inscriptions. Fine sand or pulverized sandarac resin was
often sprinkled on ink immediately after writing to blot up any excess and
speed drying. These fine particles sometimes adhered to the ink and may give
it a glittery appearance. Faded iron-gall ink inscriptions fluoresce under long-
wavelength ultraviolet illumination and can often be read and photographed
(see Inscriptions, faded). Inscriptions may sometimes be revived by boiling
gall nuts in ethanol and lightly spraying the inscription with this decoction,
but this procedure is extremely risky as wetting an inscription may wash it
away or cause it to bleed. Ammonium sulfide vapors have also been used to
revive iron-gall ink inscriptions.
Writing inks have also been also made from cuttlefish (sepia) ink, aqueous
suspensions of beechwood soot (bistre), and plant extracts (such as crushed
pokeweed berries).
Aniline inks were invented in 1856. These are generally water soluble.
Stains made by iron-gall, aniline, and unfixed plant-extract inks may be
bleached with 2–5% Chloramine-T, followed by antichlor (see below and
Bleach), and thorough washing. 5% oxalic acid or 10% citric acid can also be
used to bleach iron-gall ink, again followed by thorough washing.

Some historic formulas for inks

Robert Dossie's The Handmaid to the Arts (London, 1764) gives a formula
for preparing common black iron-gall writing ink.

1 gallon of boiling soft water poured over 1 pound of powdered galls.

Let stand for 2 or 3 days.
Then add half a pound of powdered green vitriol and let stand for
another 2–3 days, stirring occasionally.
Then dissolve 5 ounces of gum arabic in a quart of boiling water and
add to above.
Finally, add 2 ounces of alum.
Strain before use.

Dossie gives the following formula for black printing ink formula. Boil nut
oil in an iron pot, and then allow it to catch fire. Allow it to burn for half an
hour with stirring, and then extinguish the fire. This is termed “weak oil,” and
is the principal oil used in making printing ink.
To make so-called “strong oil,” prepare as above but permit the oil to
remain on fire until it becomes “very thick and glutinous, and when cool is
extremely adhesive and ropy, so as to be drawn out in long threads.”
To prepare the ink, take one-half pound of “Frankfort black” (a charcoal
black possibly made by burning lees of wine from which the tartar has been
removed, or simply wood charcoal) or any other good black and grind it on a
stone with about half that weight of the weak oil. To prepare for use, grind in
“a quantity” of the strong oil.

References
Robert Dossie, The Handmaid to the Arts (London, 1764).
Dard Hunter, Papermaking through Eighteen Centuries (New York, 1930).
Hermann Kühn, Conservation and Restoration of Works of Art and
Antiquities vol. 1 (London, 1986).
Guy Petherbridge, Conservation of Library and Archive Materials and the
Graphic Arts (London, 1987).
H. J. Plenderleith and A. E. A. Werner, The Conservation of Antiquities and
Works of Art (London, 1979).
Kenneth W. Rendell, Forging History: The Detection of Fake Letters and
Documents (Norman and London, 1994).
Pierre François Tingry, Traité théorique et pratique sur l'art de faire et
d'appliquer les vernis (Geneva, 1803).

Parchment, vellum, and slunk

(see also Leather)

The distinction between parchment, vellum, and slunk is based upon the
age of the animal from which the skin came: parchment is made from the
skins of adult sheep while vellum is made from the skins of calves and kids.
Slunk is made from the skins of unborn or prematurely born calves. The skins
of other animals are also used to make these materials.
Upon arrival from the slaughterhouse, the skins are washed and then
soaked in a lime (calcium oxide) bath for several days and are then taken
from the lime bath and dehaired and defleshed. This was traditionally carried
out by scraping with moon-shaped blades, which also shaved down the skins
to the desired thickness. Sheepskins are often split after being in the lime
liquor for several days; the flesh side, which is stronger, is used to make
parchment or chamois leather, while the hair side is used to make skivers.
Today, machinery is used to squeeze and cut the fat and other extraneous
matter from the skins. The skins are then returned to a lime bath for further
whitening, if necessary, stretched on frames where the lime liquor is
squeezed out with blunt moon-shaped knives, and finally dried. After drying,
the skins were traditionally shaved down to thickness with a very sharp
moon-shaped knife, though machinery and abrasives are often used today.
Parchment and vellum are very stable materials, and if preserved under the
right conditions they can last almost indefinitely. However, if they become
wet, they will disintegrate; if they dry out due to low humidity or heat, they
tend to shrink and become brittle. Parchment and vellum should be stored at a
relative humidity of 55–65% and at a temperature between 68° and 72°F
(around 20°C).
In musical instrument construction, parchment and vellum are used to
make the heads of percussion and stringed instruments (such as banjos) as
well as flexible hinges in piano action and organ mechanisms. Thin strips of
vellum are sometimes used to reinforce stringed-instrument rib structures as
well as cracks in soundboards. Over time, and with use, parchment and
vellum parts may wear out, tear, or come unglued. Dryness causes skins to
contract and stiffen, which may cause drum and stringed-instrument heads to
rip. If the instruments are not being used, the tension on their heads should be
reduced if possible (such as those of timpani, rope-tensioned drums, and
banjos) to prevent tearing.

Conservation of parchment and vellum

Ripped hinges used in piano escapement mechanisms and organ bellows,

for example, generally must be replaced with new material (see Stringed-
keyboard restoration, Organ restoration). Torn drum and stringed-instrument
heads (such as those used on banjos and Chinese erh-hus) can be
cosmetically repaired, though repaired tears generally will not withstand
renewed tension. A skin that has dried out, shrunk, and pulled loose from its
framework can sometimes be relaxed, stretched, and remounted by placing it
in a workroom or chamber where the relative humidity has been raised to 65–
70%. If the skin cannot be removed from the instrument, humidified air can
be directed at it from a hose and nozzle connected to an ultrasonic humidifier
or by a Preservation Pencil (with the heating element turned off). Water
vapor from these devices should not be allowed to condense or bead up on
the skin during rehydration, especially if there is painted decoration on it.
Once softened, the skin may be supple enough to be retucked or refastened
onto the instrument.
Tears are often caused by shrinkage. By humidifying the edges of a tear,
they will swell, and it may be possible to bring them back together for repair.
Strips of goldbeater's skin can then be glued over the tear to hold it together.
If the underside of the skin is accessible, the tear should be similarly
reinforced from below. Parchment glue is an excellent adhesive for joining
goldbeater's skin to parchment and vellum. To make parchment glue, place
parchment trimmings in cold water and allow them to swell, then heat them
with a little water to about 140°F (60°C) as one would with regular animal-
hide glue. (Firms that sell gilding supplies, such has Sepp Leaf Products, Inc.,
New York, sometimes sell parchment scraps at reasonable prices.) Gelatin,
isinglass, or light-colored animal-hide glue can also be used to repair
parchment and vellum. When using goldbeater's skin to reinforce a rip, it
should be stretched to remove wrinkles and held in place while gluing with a
sheet of weighted glass or Plexiglas. A repair made with goldbeater's skin
will be nearly invisible due to its extreme thinness and transparency, though
it can be retouched with dyes or watercolors to match the surrounding area if
necessary. Goldbeater's skin is quite strong, and though repairs made with it
may withstand a modicum of tension (perhaps enough to remove buckling or
surface unevenness), the head will lack the strength required to restore full
playing tension. If a decision is made to replace the original head of a drum,
banjo, or other instrument in order to restore playability, the conservator
should consult with a specialist or maker (particularly in the case of
ethnographic instruments) so that the appropriate materials and mounting
procedure will be followed.

References
Marion Kite and Roy Thomson, Conservation of Leather and Related
Materials (Oxford, 2007).
Guy Petherbridge, Conservation of Library and Archive Materials and the
Graphic Arts (London, 1987).
Patination

Metal (see also Bluing)

Late eighteenth- and early nineteenth-century English square pianos are

often missing brass hardware, such as bolt covers, hooks, eyes, casters,
hinges, and lock fittings. Replicas can be cast from original parts (see Mold
making), but they generally must be patinated to match. For brass hardware
of this period, a 1–5% solution of cupric nitrate can be used to impart a slight
greenish cast, followed by a 1–5% solution of liver of sulfur (a mixture of
potassium polysulfide and potassium thiosulfate) to add brown. These
solutions can be used cold, though their action is quicker if they are warmed.
The degree of patination can be better controlled if the chemicals are applied
in low concentration, allowed to air-dry, and reapplied until the desired effect
is achieved. The best method is to immerse workpieces in a succession of
trays or tanks containing diluted patination solutions. This method often will
produce a more even color than applying the chemicals in more concentrated
form with a brush.
After chemically treating newly cast brass mounts, it is sometimes
effective to polish them in such a way that the recesses remain dark. This will
allow new parts to blend with originals if the latter have been superficially
polished and retain some of their natural patina in hollow, incised, and
engraved areas. Polishing can be done by hand or with a motorized muslin or
cotton buffing wheel. Once the desired effect has been achieved, the color
can be stabilized either by lacquering with a nitrocellulose lacquer such as
Agateen no. 27 diluted 1:1 with lacquer thinner, or by varnishing with
Incralac or Paraloid B48N (12.5% solids for spraying or 15% solids for
brushing).

Wood (see also Wood)

No effective method has been developed for patinating wood, which is the
bane of violin makers attempting to make copies as well as those involved in
the conservation of wooden objects. As far back as the sixteenth century,
Alessio Piemontese (1555) advocated the use of horse manure to darken
wood, a material that does not lend itself to modern conservation work.
 New parts that have been fabricated of light-colored woods (such as
spruce, fir, pine, and maple) are especially difficult to “age,” though they can
be darkened somewhat by prolonged exposure to sunlight or by placing them
in “sunning boxes” equipped with long-wavelength ultraviolet lamps. The
nineteenth-century violin maker J. B. Vuillaume is said to have baked his
wood in an oven to darken it, though this method is only effective at
temperatures that cause resins in the wood to exude and form spots on the
surface. Baking also produces a somewhat lifeless appearance, as well as
causing excessive dehydration and shrinkage of the wood. Some modern
violin makers have experimented with ozone chambers to darken wood. They
place their instruments (prior to varnishing) in a sealed chamber with a high-
voltage arc generator. The arc produces ozone, nitrogen dioxide, and nitric
acid fumes in the presence of moisture. Unfortunately, this procedure is
damaging to wood and glue joints, as well as to human mucous membranes
and respiratory organs. Caustic chemicals such as strong alkalis, ammonia
fumes, a solution of potassium dichromate, and nitric acid have been used,
but again, these tend to have a deleterious effect upon the physical structure
of wood and glue, and it is difficult to predict the final outcome. Other
popular chemical treatments include painting the wood with a solution of
ferric acetate (made by placing iron filings in acetic acid or vinegar) followed
by an application of a solution of tannic acid (derived from steeped tea
leaves), the use of saltpeter (potassium or sodium nitrate), potassium
permanganate, and other oxidizers. Colorants such as coffee, tea, chicory,
walnut hulls, fuliggine (fireplace soot), and tars extracted from old cigarette
butts have also been used to color wood. Synthetic and natural dyes are
sometimes effective, but in softwood they tend to be absorbed to a greater
degree by the spring growth than by the summer growth and thus produce an
unnatural visual effect of “grain reversal,” in which the late growth appears
lighter (or may even disappear entirely) against the more heavily stained
early growth. Painting the surface with artist's pigments (such as burnt umber,
ochre, etc.) suspended in water (watercolor) or organic solvents tends to
produce a masked effect.
The flute maker Rod Cameron uses the following method to darken
boxwood: he places iron nails in warm nitric acid (this evolves nitrous oxide,
which is poisonous, so this should be done outdoors or in a fume hood;
furthermore, the evolution of this gas may also cause containers to explode if
they are tightly sealed). The resulting solution of ferric nitrate is applied to
the boxwood, and after the desired color is achieved the acid is “neutralized”
by applying warm linseed oil. He creates artificial figuration in boxwood by
playing the surface with a torch.

Ivory (see also Ivory)

The violin bow maker and restorer Henryk Kaston developed an effective
method for patinating newly made ivory fittings for bows: he moistened
wood shavings with alcohol and set them aflame. He then sprinkled them
with water to extinguish the flame and generate smoke. The fully polished
ivory part was then exposed to this smoke. If necessary the process was
repeated until the desired yellowish appearance was achieved.

References
William T. Brannt, The Metal Worker's Handy Book of Recipes and
Processes (London, 1896).
J.-P. Coutrait, Nouveaux Trucs et procédés du bois (Paris, 1991).
George Frank, Adventures in Wood Finishing (Newtown, 1981).
Richard Hughes and Michael Rowe, The Colouring, Bronzing and Patination
of Metals (London, 1982).
Susan La Niece and Paul Craddock, Metal Plating and Patination (Oxford,
1993).
Herbert Maryon, Metalwork and Enamelling (New York, 1971).
Alessio Piemontese, De' Secreti del Reverendo donno Alessio Piemontese
(Venice, 1555).
Ronald D. Young, Contemporary Patination (San Rafael, 1988).

Pegs

Peg shapers and reamers are available with a 1/30 taper (for modern violin
and viola pegs), 1/25 taper (for cello pegs), and 1/17 taper (for cello endpins).
3° and 5° reamers are available for fitting guitar bridge pins and harp board
pins (see Tapered reamers). Some early violin pegs have a steeper taper of
around 1/20, and when working with early stringed instruments having
pegboxes that have not been reamed out with modern tools, it may be
necessary to make pegs to fit those nonstandard tapers. To set the angle of the
compound slide of a machine lathe to turn the proper taper on the shaft of a
new peg, use a caliper to measure the major and minor diameters of the
tuning peg holes at the outer walls of the pegbox, as well as the distance
between these walls. Calculate half the angle of the taper (which is the angle
to which the compound slide should be offset) by subtracting the minor
diameter from the major diameter and dividing by twice the distance between
them: this is the tangent of half the taper angle (the angle itself can be
determined with an electronic calculator, slide rule, or table). If an old peg
exists, its taper angle can be measured directly with a vernier protractor fitted
with an acute-angle attachment. The measured angle should be halved to set
the compound slide. If no lathe is available, an alternative is to make a peg
shaper that matches the taper. This can be made out of hardwood and fitted
with a cutter from a standard peg shaper or the blade from a plane; the
tapered hole in the shaper can be made by drilling a hole having the minor
diameter of the peg and then enlarging it with a rat-tail file. One can also
make a peg reamer by shaping the appropriate taper in a length of hardwood
with a spokeshave, plane, or file, then sawing a groove in it and inserting a
piece of tempered steel (such as a fretsaw blade) that has been beveled like a
cabinet scraper (see Woodworking and Sharpening tools). The cutting edge
of the blade should project just beyond the surface of the reamer, and some
wood should be trimmed away in front of the cutter to provide clearance for
the shavings. This reamer can be used to refresh the holes in the pegbox as
well as to adjust the taper in the improvised peg shaper (described above) to
produce a perfectly matched set of cutting tools.
When using peg shapers to fit violin pegs, use a sharp knife to score around
the shaft of the peg just beyond the decorative turnings. This will prevent
them from chipping or breaking off when the shaft is turned into the shaper.
A shaper blade works best when it is well sharpened and projects beyond the
mouth of the shaper by about 0.25–0.75 mm. It should also project past the
front of the shaper's opening by about 0.5–1.0 mm. Do not apply excessive
inward pressure while rotating the peg into the shaper as this may cause the
wood to tear or crumble.
 When using a reamer (the new, multi-flute helical reamers work well),
make very light cuts and continually check the fit of the shaped peg to avoid
reaming out too much wood. If the pegbox holes have been widened to the
extent that standard-size pegs are too small, they will have to be plugged,
drilled, and reamed – a complex process.

Peg dope

When the air gets dry, pegs tend to loosen because they shrink and the peg
holes widen (as a result of the pegbox walls shrinking). In humid weather
pegs tend to stick because they swell and the peg holes narrow due to
swelling of the pegbox walls. Pegs will slip if their shafts become burnished
from continued turning. Baillot (1835) recommends rubbing the pegs with
soap if they stick and applying lead white or chalk if they slip. To lubricate
pegs, use a dry bar of “toilet soap,” but avoid “floating soap,” as this type
contains about 30% water. To provide friction, use commercial whiting and
avoid chalk that contains oil (such as blackboard chalk). Some of the
commercial peg dopes (such as Hills') are too slippery.

Reference
Pierre Marie François de Sales Baillot, L'art du violon (Paris, 1835).

Piano action regulation and voicing (modern grand)

Preliminaries (Figure 26)

1 Prior to regulating the action, make sure that the key frame makes
good contact with the key bed. If not, adjust the frame studs until
they make firm contact with the key bed (it may be necessary to
retract all of the studs and then systematically retighten them). If
there are no studs, paper or cardboard shims can be placed under the
frame so that it makes contact with the key bed. When pressing down
on various parts of the framework from bass to treble, there should
be no movement of the key frame and no knocking sound made by it
 striking the key bed.
2 Each key lever should be checked for smooth action. Sometimes
overall tightness of the key levers and action parts is due to high
humidity. If so, the action can be dried out in a box or cabinet
containing a tray of silica gel or warmed by a source of modest heat,
such as an incandescent light bulb. There should be a slight amount
of side-to-side play at the front- and balance-rail pins, but no front-
to-back motion when the keys are grasped at the front and pushed
and pulled back and forth. If so, the hole at the bottom of the key at
the balance rail may have become enlarged. If so, plug the hole with
a basswood dowel and redrill the hole. If the hole is only slightly
enlarged, it may respond to sizing with dilute hot animal-hide glue
(allow the glue to dry completely before reinstalling the key lever).
Excessive side-to-side motion or rattling may be fixed by rotating the
front guide pins, which are oval; by turning them a few degrees with
the appropriate piano technician's tool (Figure 27), the play can be
reduced. If this adjustment is insufficient to remove excessive
motion, the front pin mortise can be rebushed with new felt. Bent
front or balance pins may cause key levers to stick; these pins can be
straightened with standard piano technician's tools. Key levers can be
spaced and any tilt removed using these same tools. The heights of
the key levers relative to the key bed are adjusted (see table below)
by increasing or decreasing the thickness of the back rail key cloth
and/or by adding or removing paper and cardboard punchings under
the balance-rail felts. The keys should be about 1/32 in. (about 1
mm) higher at middle C than at the top and bottom keys (a batten of
wood planed to this subtle curve can be used as a guide). With this
slight convexity in mind, adjust the key dip by adding or removing
paper and/or cardboard punchings under the front rail felts (see table
below). The accidental key levers should also stand about 1/32 in.
higher than the natural key levers. When fully depressed, the fronts
of the sharps should not descend below the natural keys, but come to
rest about 1/32 in. above the tops of the naturals. The key-stop rail
should be adjusted so that it is 1/16 in. above the top of the
accidental key levers.
3 Other points to check before regulating the action include: the
condition of the leather coverings of the hammer rollers (also called
the “knuckles”), which if hardened can be softened by needling with
a hammer voicing tool; the spacing of the jacks (also known as
“flies”) within their repetition-lever mortises; the adjustment of the
back edges of the jacks, which should align with the centers of the
wooden cores of the rollers (this is done by turning the jack's
regulating screw); and the height of the repetition levers, which
should be adjusted with the regulating screw so that the tops of the
these levers are just in contact with the rollers. (This last adjustment
can be checked by depressing the jack's arm [or “tender”] and noting
if the hammer moves slightly.) Check the pivots of the hammers,
jacks, wippen flanges, and repetition levers. The hammer and wippen
flanges should be free-acting, though firm enough that the flange
drops slowly by the weight of its screw. The jack should move freely
on its pivot but not rattle or exhibit excessive side-to-side movement.
Pivot pins that are slightly tight may respond to the application of a
drop of ethanol diluted with water in a 3:1 ratio (or 1:1 in severe
cases). If the pins are still too tight, the cloth bushing can be reamed
or burnished, or a smaller pin can be installed. If pins are loose, a
larger pin can be installed. Center pins come in graded sizes, and
specially designed sets of tapered, hexagonal reamers and burnishers
are available for this work. If the felt bushings of the hammer and
wippen flanges are worn, the bushings should be drilled out and the
holes rebushed with new felt. Finally, check and correct the string
spacing with the appropriate tool, and center the hammers relative to
the strings by loosening the hammer flange screw. When this general
maintenance has been completed, regulation can begin.
Figure 26 Modern piano action. Line drawing from William Braid White's
Piano Tuning and Allied Arts, 1946.

Figure 27 Standard set of piano action regulating tools.

Regulation
 Before removing the action from the piano:

1 Adjust the distance between the hammers at rest and the strings (see
table below) by turning the capstan screws that are located directly
beneath the wippens. This is an extremely important first step (that is
generally not mentioned in maintenance manuals) because the action
cloths of older pianos may be greatly compressed, causing the
hammers to sit very low relative to the strings. If the hammers are
not raised to the recommended height, it may be impossible to adjust
the let-off, for example. The capstan screws below the wippens are
also used to level the hammers. When set at the proper distance from
the strings, the hammer shanks should rest about 3/16 in. above the
hammer cushions mounted on the wippens.
2 Adjust the hammer let-off (see table below) by turning the regulation
screws or buttons.

Remove the action from the piano:

1 Adjust the hammer drop to 1/8 in. by turning the drop screws.
2 Adjust the back checks so that the hammers check at the proper height
(see table below).
3 Check the strength of the repetition springs by hitting each key solidly
so that the hammer is caught by the back check. Then slowly release
the key and note whether the hammer rises smoothly without
jumping. The strength of repetition springs can be adjusted by
turning small adjustment screws located on the repetition levers (in
actions lacking these screws, each spring can be adjusted by flexing
it with the appropriate piano technician's tool).
4 Adjust damper under levers so that they engage the key levers when
the hammer is about halfway between its rest position and the
strings. This is done by loosening the under lever top flange screw
and adjusting the length of the damper wire.

Grand piano action regulation

Key Hammer- Let- Key Check

 Maker height to-string off dip height

Steinway 2⅝ in. 1¾ in. in. 0.390 ⅝ in.

in.

H. Steinway 65 47 mm 1 mm 10 15 mm
mm mm

G. Steinweg 69 45 mm 2 mm 9.66 15 mm
mm mm

Baldwin 2 1⅞ in. in. ⅜ in. ⅝ in.

in.

Bechstein 63 46 mm 1.5 9.5–10 –

mm mm mm

Blüthner 71 45 mm 1.5 9 mm 24 mm
mm mm

Bösendorfer – 46 mm 1.5 9.5– 15 mm

mm 9.8
mm

Chickering 2½ in. 1¾ in. ⅛ in. in. ¾ in.

Yamaha 2 1 in. ⅛ in. in. ⅝ in.

in.

Kawai 2¾ in. 1¾ in. –⅛ in. in.

in.

Voicing

Voicing refers to the adjustment of a piano's timbre by altering the

hardness of the hammers' felt. Some pianists prefer a mellow sound, which is
produced by a softer hammer, while others like a more percussive sound,
which is produced by a harder hammer. A piano must be regulated and in
perfect tune before attempting to voice its hammers. Before beginning, check
to see if all of the strings are well spaced and making contact with their
respective hammers (they can be spaced and leveled with a stringing hook).
Any strings that are not making firm contact with the bridge can be pressed
down with a piece of brass. Spacing and leveling may throw the strings out of
tune, so they should be gone over carefully before voicing. The felt hammer
heads of modern pianos can be softened with a special needling tool or
hardened with a curved iron (Figure 28). Never needle the hammer's striking
surface directly, but work in front and behind it, jabbing the needles
downward in a radial fashion to loosen up the supporting felt; a core of
unneedled felt should remain directly below the striking surface of the
hammer. If the tops of the hammers have deep indentations from the strings,
the outer surface of the felt can be removed with a sandpaper file. Do not
begin by filing through the indentations at the top of the hammer, but file
towards the top from the front and the back in order to preserve its shape. A
60 grit sandpaper file can be used for rough reshaping, followed by a 100 grit
file. To finish, file off the “top knot” and go over the outer surface of the
hammer to produce a smooth contour, again maintaining the proper shape.
The tops of the hammers should not be flat, but nicely rounded in the bass
and somewhat pointed in the treble. Some pianists prefer very hard hammers
and have their technicians “dope” the felt with dilute lacquer. Obviously, the
original leather and felt hammer coverings of historic fortepianos should not
be impregnated with any material that might alter their acoustic quality.
Figure 28 Piano tuning and voicing tools, including various sizes of tuning
wrenches, tuning forks, an electronic tuning device, iron and needling tool for
voicing hammers, and dampers used in tuning.

References
Franz Rudolf Dietz, Das Intonieren von Flügeln (Frankfurt, 1968).
Alfred H. Howe, Scientific Piano Tuning and Servicing (Clifton, 1976).
Walter Pfeiffer, The Piano Key and Whippen (Frankfurt, 1967).
Piano Action Handbook (Seattle, 1971).
Steinway Grand Key and Action Regulation (New York, 1949).
William Braid White, Piano Tuning and Allied Arts (Boston, 1946).

Pitch

For thorough treatments of this subject see Ellis and Mendel (1968) and
Haynes (2002).
In the seventeenth and eighteenth centuries the terms low and high
chamber pitch (Cammer-Ton), choir pitch (Chor-Ton), and cornett pitch
(Cornett-Ton) referred to distinct relative pitches, each separated by about a
semitone, with low chamber pitch being the lowest and cornett pitch the
highest. Regional pitch differences may have overridden these distinctions
(Praetorius, 1619).
Today, for practical purposes, when tuning early keyboard instruments and
stringed instruments, especially for use with modern copies of early wind
instruments, Renaissance pitch is often set at A=440 or A=466, low French
Baroque pitch is often set at A=392 Hz, late Baroque pitch is set at A=415
Hz, and Classical and early Romantic pitch is set at A=430 Hz. Modern pitch
is supposedly standardized at A=440 Hz, though many orchestras and concert
halls (such as the New York Philharmonic and the house piano at Carnegie
Hall in New York) are at A=442 Hz. The pitches 392, 415, 440, and 466 are
approximately equal semitones apart. Some modern harpsichords have
provision for transposing between two or three pitches that are semitones
apart.

Nominal pitch

The nominal pitch, or “key,” of brass instruments is generally the

fundamental pitch without the valves being pressed or the slides being
extended. In woodwind instruments (other than the bassoon), the nominal
pitch is generally taken as one tone below the “six-finger” note in the lowest
register for instruments overblowing the octave (such as the oboe) and second
register for instruments overblowing the twelfth (such as clarinets). (Myers
1990)

Perfect pitch

Some musicians can detect whether an instrument is tuned a few cycles per
second above or below standard pitch and may be completely thrown off by
the “transposition” when they sit down at or play along with a keyboard
instrument that is tuned a half or full tone below modern pitch. There are
those who argue that perfect pitch is an innate ability, while others believe
that it can be learned. For those of us who must work with instruments from
various periods and locales, perfect pitch is a hindrance, so to whatever
extent it can be learned, it is best not to cultivate this skill. Musical
instruments should not be retuned to pitches they were not designed for in
order to accommodate musicians encumbered with perfect pitch.

References
Alexander J. Ellis and Arthur Mendel, Studies in the History of Musical Pitch
(Amsterdam, 1968).
Bruce Haynes, A History of Performing Pitch: The Story of “A” (Lanham
and Oxford, 2002).
Hermann Helmholtz, On the Sensations of Tone, translated by Alexander J.
Ellis (New York, 1954). (This includes Ellis's table of historical pitches.)
Arnold Myers, Historic Musical Instruments in the Edinburgh University
Collection (Edinburgh, 1990), p. 14.
Michael Praetorius, Syntagma musicum (Wolfenbüttel, 1619).

Proportion

See Historical metrology

Proportion is defined as a constant ratio between three or more numbers, or
the equality of two ratios. There are three different types of proportions:
arithmetic (in which the second number is related to the first by the same
quantity as the third is to the second, as in 3:4:5); geometric (in which the
second number is the same multiple of the first as the third is to the second,
as in 2:4:8); and the harmonic (in which the first number is to the third as the
difference between the first and second is to the difference between the
second and third, as in 3:4:6). These proportions and their means can be
expressed algebraically as follows:

Arithmetic proportion b – a = c – b
Arithmetic mean b = (a + c)/2
Geometric proportion a/b = b/c
Geometric mean
Harmonic proportion a/c = (b – a)/(c – b)
 Harmonic mean b = 2ac/(a + c)

The idea that ratios and proportions are the basis of musical intervals (see
Intervals) can be traced back to a discovery said to have been made by
Pythagoras in the sixth century BC while he listened to the sound of
blacksmith's hammers striking an anvil: two hammers having weights in the
ratio of 1:2 produced a clanging sound an octave apart, those having weights
in the ratio of 2:3 produced sounds a fifth apart, while those having weights
in the ratio of 3:4 produced sounds a fourth apart. This was a brilliant
discovery, because up until that point, the concept of consonance had been a
purely aesthetic matter with no mathematical or physical basis. An example
of consonance occurs as a violinist tunes one string to another: the interval of
a perfect fifth stands out in an arresting manner amidst the discordance on
either side of that very distinct pitch. The music theorist Boethius (480–524
CE) credited Pythagoras with having invented the monochord, which is a
simple device consisting of a single string stretched across a soundbox (hence
the name “monochord”). By sliding the monochord's bridges back and forth,
one can hear the sounds produced by various proportional string lengths.
Many late fifteenth-, sixteenth-, and seventeenth-century treatises devoted
to music theory (such as those written by Ludovico Fogliani, Franchino
Gafurio, Athanasius Kircher, Marin Mersenne, Nicola Vicentino, and
Gioseffo Zarlino) include dissertations on the mathematics of musical
intervals, the construction of musical scales, and geometric methods for
dividing the monochord. For example, Zarlino's Le istitutioni harmoniche
(Venice, 1558) provides instructions for adding and subtracting musical
intervals by multiplying and cross-multiplying the numerators and
denominators of their ratios: adding a fifth (3/2) to a fourth (4/3) by
multiplying the numerators and denominators produces the ratio of the octave
(12/6, or 2:1); subtracting a fourth from a fifth by cross-multiplying produces
a ratio of 8/9, or a whole tone. Zarlino also describes the mesolabio (an
ancient geometric calculating device consisting of a set of rectangular frames
sliding between parallel tracks) that could be used, for example, to divide the
Pythagorean whole tone into two equal parts.

Proportion in musical instrument design

 Because ratios and proportions are the basis of musical intervals, it stands
to reason that they have also played an important role in the design of
musical instruments. An ordinance passed in Toledo in 1617 provides insight
into the formal technique of musical instrument design then in use. It states
that anyone applying to become a master vihuela maker must not only submit
a finished instrument but also demonstrate knowledge of the instrument's
proper “proportions, rules, and [musical] range.” Candidates had to design an
instrument from scratch, and part of the master's examination involved

[The making of a] paper template, which must be done in the presence

of the overseers and examiners so that [they] can see it being made, and
it must be made using only a knife, a pair of compasses, a ruler, and a
set square, without recourse to any other pattern, but drawing only on
experience and an understanding of the said craft.

This reliance upon a pair of compasses, straightedge, and set square is telling,
for these drafting instruments are directly related to Euclid's five fundamental
postulates (which involve straight lines, circles, and right angles) and they are
the only drafting tools used in the geometric proofs of his theorems.
The basic ratios and proportions that underlie the design of the violin, viol,
lute, mandolin, guitar, and other instruments are associated with fundamental
musical intervals, such as the ratios of the octave (2:1), fifth (3:2), fourth
(4:3), and major third (5:4). Early keyboard instruments exhibit what is
termed “Pythagorean” scaling (see Scaling), meaning that through most of
the keyboard range the string lengths double or halve on the octave. Though
the gross case dimensions of keyboard instrument cases rarely follow
proportional rules, the bodies of stringed instruments such as the violin and
guitar often do.
The classical violin form developed in Cremona exhibits the following
basic proportions: upper bout width to lower bout width to overall length
have the ratio 4:5:9; the neck stop to body stop forms a ratio of 2:3; the body
is divided by the bridge in a ratio of 5:4; f-hole length to body length is 1:5;
and the bridge divides the length of the f-hole in half. From the lute,
mandolin, and viola da gamba neck and fingerboard patterns preserved in the
Museo Stradivariano in Cremona, it is clear that Antonio Stradivari was
familiar with laying out tempered fret positions and designing families of
instruments that were proportionately scaled to different pitch relationships
(Pollens, 2010). Several of his patterns have simple arithmetic calculations
scribbled on them, though none of these calculations appears to relate to
musical instrument construction.
Using simple arithmetic and the basic proportional relationships, it would
have been possible for an instrument maker to calculate the dimensions of an
instrument of any size, but this task was greatly simplified with a device
termed the sector (also known as a pantometer), a type of proportional divider
that was often included with sets of drafting instruments made in the
sixteenth through nineteenth centuries. There is documentary evidence that
Stradivari owned a sector, though his heirs evidently lent it to the violin
maker Carlo Bergonzi after Antonio's death and unfortunately failed to
recover it when most of the other shop materials were gathered up and sold to
Count Cozio di Salabue in 1775–1776 (Pollens, 1992, 2010).
To use a sector, the legs are opened to the distance one needs to divide, and
then an ordinary pair of dividers is used to take the actual length of whatever
subdivision one requires from the engraved scales (Figure 29). Most sectors
are engraved with scales for subdividing lines, surfaces, circles (in order to
draw polygons), and solids, and some have scales for determining
trigonometric functions and for comparing the weights of objects made of
different metals. A few have scales designed for marking off musical
intervals. In Athanasius Kircher's Musurgia universalis (Rome, 1650), there
is a description and illustration of a sector having a specially engraved scale
that was designed to subdivide any length into the intervals of diatonic,
chromatic, and enharmonic genera; however, even a conventional sector
lacking this special scale could have been used to mark out the basic
proportions of musical intervals.
Figure 29 Setting a sector with a pair of dividers.

If one measures Stradivari's instruments, forms, and patterns, it is evident

that he did not strictly adhere to the proportional scheme established by his
predecessors; nor did he follow a single proportional system throughout his
career. For example, in 1690 he developed a narrower contralto viola form
expressly for the Medici commission, though the violins that were part of that
commission were of the older, broader style. The development of the “long-
pattern” violin (which was about seven millimeters longer and five
millimeters narrower across the upper and lower bouts than typical full-size
violins made up until that time) appears to have been concurrent with the
design of the Medici violas. Around 1700, Stradivari abandoned the “long-
pattern” violin form and reverted to the broader model that he had previously
employed, though he continued to make use of the narrower contralto viola
form throughout the rest of his career.
Was Stradivari a nonconformist or a participant in a general movement
away from the strict use of proportions that had been established a century
and a half earlier? The noted art historian Rudolf Wittkower wrote that at the
end of the seventeenth century there was a decided breaking-away from the
laws of harmonic proportion in the field of architecture, which was a kindred
discipline of instrument making (Wittkower, 1971). Among other
documentary evidence, Wittkower quotes the seventeenth-century architect
Guarino Guarini, who wrote: “to please the eye one must take away from, or
add to, the proportions, because one object is placed under eye level, another
at great height, another in an enclosed space, and yet another in the open air.”
Wittkower notes that by subordinating the rules of architecture to those of
perspective, Guarini “does not even discuss the possibility of the objective
truth on which Renaissance aesthetics was founded.” Guarini, however, was
not truly at odds with Renaissance theorists, as Wittkower suggests, nor even
with those from the ancient world, for in the first century BC, the Roman
architect and engineer Marcus Vitruvius Pollio offered similar advice in The
Ten Books on Architecture:

There is nothing to which an architect should devote more thought than

to the exact proportions of his building with reference to a certain part
selected as the standard. After the standard of symmetry has been
determined, and the proportionate dimensions adjusted by calculations,
it is next the part of wisdom to consider the nature of the site, or
questions of use or beauty, and modify the plan by diminutions or
additions in such a manner that these diminutions or additions in the
symmetrical relations may be seen to be made on correct principles, and
without detracting at all from the effect. The look of a building when
seen close at hand is one thing, on a height it is another, not the same in
an enclosed place, still different in the open, and in all these cases it
takes much judgment to decide what is to be done. The fact is that the
eye does not always give a true impression, but very often leads the
mind to form a false judgment.

In keeping with Vitruvius, two of the Renaissance's greatest architects,

Alberti and Palladio, also advocated the “swelling” and “diminution” of
structures as well as the subtle arching of horizontals and columns in order to
overcome various visual illusions and thus to render architecture more
pleasing to the eye. Thus, Stradivari's deviation from the established
proportions that underlie violin design was not truly at odds with Renaissance
or Baroque aesthetics. Other Cremonese makers from Stradivari's time were
also highly innovative. Giuseppe Guarneri del Gesù, for example, appears to
have used a form similar to one designed by his father until around 1731, at
which point he developed his own model, and finally a broader model around
1740.
References
Antonio Bagatella, Regole per la costruzione de' violini, viole, violoncelli, e
violoni (Padua, 1782).
Cristina Bordas and Luis Robledo, “José Zaragozá's Box: Science and Music
in Charles II's Spain,” Early Music (August, 1998).
Kevin Coates, Geometry, Proportion and the Art of Lutherie (Oxford, 1985).
Euclid, The Thirteen Books of The Elements, trans. Thomas L. Heath, vol. I
(New York, 1956).
Roger Hargrave, “The Working Methods of Guarneri del Gesù and Their
Influence upon His Stylistic Development,” in Giuseppe Guarneri del
Gesù, II (London, 1998).
Bruce Haynes, A History of Performing Pitch (Lanham and Oxford, 2002).
Athanasius Kircher, Musurgia universalis (Rome, 1650).
Mark Lindley, Lutes, Viols and Temperaments (Cambridge, 1984).
Grant O'Brien, Ruckers: A Harpsichord and Virginal Building Tradition
(Cambridge, 1990).
Stewart Pollens, The Violin Forms of Antonio Stradivari (London, 1992).
Stewart Pollens, Stradivari (Cambridge, 2010).
Roberto Regazzi, Il Manoscritto liutario di G. A. Marchi (Bologna, 1986).
José Romanillos, Vega and Marian Harris Winspear, The Vihuela de Mano
and the Spanish Guitar (Guijosa, 2002).
Marcus Vitruvius Pollio, The Ten Books on Architecture, trans. Morris Hicky
Morgan (New York, 1960).
Rudolf Wittkower, Architectural Principles in the Age of Humanism (New
York, 1971).
Gioseffo Zarlino, Le istitutioni harmoniche (Venice, 1558).
Ronald Edward Zupko, Italian Weights and Measures from the Middle Ages
to the Nineteenth Century (Philadelphia, 1981).
 R
Recording

Recording the sound of instruments that one has restored can be part of the
documentation process. The moderately priced digital recording equipment
that is available today can provide audio quality that is equal to or better than
the most sophisticated analog equipment that was used in professional
recording studios just a few decades ago. For recording individual
instruments or small ensembles, all one needs are a pair of high-quality
microphones (or a stereo microphone) and an analog-to-digital converter
hooked up to a laptop computer or a dedicated digital recorder.
Room acoustics are the most critical factor with regard to producing a good
sounding recording. For example, a harpsichord, a fortepiano, or a small
ensemble of stringed or wind instruments placed in a furnished period room
will tend to produce a sense of ambience that is appropriate for such
instrumentation. However, more often than not, one is confronted with an
environment that is not ideal: there may be too much reverberation, or not
enough. The latter can be remedied electronically, but an overly reverberant
room can be difficult to record in. It may be necessary to put down carpeting
or to position acoustical damping materials around the room to limit
reverberation. If recordings are made of live concerts, one should be aware
that the acoustics of auditoriums tend to become drier (less reverberant) once
an audience enters the hall.

Positioning of microphones

The purpose of stereophonic recording is to capture sound the way our ears
do and thus create a sense of space in which we can locate and distinguish
discrete sources of sound. This sense of space is created by the time delay or
sound-pressure differences captured by the spaced microphones. A simple
pair of microphones is thus a more accurate means of creating a sense of
auditory space than the technique of multiple miking that is typically used by
most recording studios, in which sections of an orchestra or ensemble are
recorded on separate tracks and later mixed to create an artificial balance.
Because the sounds of the instruments are picked up at slightly different
times by arrays of widely spaced microphones, the composite mix lacks
detail and the illusion of acoustical space. This is why virtually all
commercial recordings, especially those of large ensembles such as
orchestras, sound terrible.
When using a pair of microphones, several methods are employed: the very
popular “X-Y” technique uses two directional microphones (often cardioid
pattern) with the head of one placed directly above the other with both heads
facing at a 45° angle away from the sound source (thus, they are at a 90°
angle with respect to each other); if the microphones have a figure-eight
pickup pattern and have their heads crossed as in the X-Y technique, they are
termed a “Blumlein pair.” The so-called “A-B” technique employs two
omnidirectional microphones set a distance apart (often several feet) and
parallel to each other. The near-coincident technique (which the author
favors) uses a pair of cardioid microphones placed more closely than in the
A-B method. In one popular near-coincident technique, termed ORTF
(abbreviation for Office de Radiodiffusion-Télévision Français), two cardioid
microphone heads are placed 17 cm apart and held at an angle of 110°,
covering an angle of about 180°. Another technique used by the Dutch
Broadcasting System (NOS) also uses two cardioid microphones but spaces
them 30 cm apart with the heads mounted at an angle of 90°, which covers an
angle of about 160°. Because of the spacing and angling of the microphone
heads, these systems are good for miking larger ensembles, such as orchestras
and choral groups. Another interesting technique that mimics the way we
hear employs two omnidirectional microphones spaced about 16.5 cm apart
and separated by a sound-absorbent disc (termed a Jecklin disc).
In the author's experience of recording early keyboard instruments, the
near-coincident technique, with closely spaced microphones positioned about
four or five feet from the open lid and facing in the general direction of the
soundboard, produces a pleasing effect, particularly if the room acoustics are
less than favorable. To create a sense of ambience, the microphones can be
moved back. When recording a small ensemble of string instruments (such as
a string trio or quartet), the closely spaced pair of microphones should be
placed fairly high (consider the direction of the tops of violins and cellos,
which face slightly upward) and not too far away from the musicians. Again,
as the microphones are moved further from the musicians, the sense of room
ambience increases. Cardioid-pattern microphones, which have a heart-
shaped sensitivity that has negligible sensitivity at the audience side of the
microphone, may be placed somewhat further from musicians than
omnidirectional microphones, which tend to pick up more room sound and
background noise. Omnidirectional microphones are generally considered the
most accurate transducers and the best choice unless an audience is present.
There are several types of microphones. Dynamic microphones (these are
the cheap, rugged microphones used in public address systems and by rock-
and-roll singers on-stage) are not appropriate for recording classical music.
Condenser microphones tend to have a much flatter and extended frequency
response and lower distortion, but require an external source of power
(termed phantom power); these come in small- and large-diaphragm types,
the best of the large-diaphragm type tending to be more highly regarded,
though they can be extremely expensive. Ribbon microphones are said to
produce a warmer sound than condenser microphones, although they tend to
be less sensitive and may require a supplemental microphone preamplifier.
Stereo microphones (which consist of two microphone cartridges mounted in
a single housing) are generally set up in X-Y configuration; the best of these
feature a pair of condenser cartridges or ribbon units.

Reference
Leo L. Beranek, Acoustics (New York, 1954).

Relative humidity

A sling psychrometer can be used to determine relative humidity and is

useful for checking or calibrating mechanical and electronic humidity
measuring equipment. When using a sling psychrometer, check to see that
both thermometers read the same temperature when the psychrometer is at
rest and before the wick on the wet-bulb thermometer has been moistened.
Moisten the wick on the wet-bulb thermometer with distilled water at room
temperature, whirl the psychrometer overhead for about a minute, and then
quickly read the wet-bulb thermometer, noting the difference between the
two thermometers. An alternative is to use one thermometer and take two
readings, the first without the wick and the second with a moistened wick.
Use the chart below to determine the relative humidity.

% Relative humidity

Difference between dry Dry bulb temperature (Celsius)

bulb and wet bulb
temperatures (degrees
15 18 20 22 25 27 30 33
Celsius)

1 90 91 91 92 92 92 93 93

2 80 82 83 84 85 85 86 87

3 71 73 75 76 77 78 79 80

4 62 65 67 68 70 71 73 74

5 53 57 59 61 64 65 67 69

6 44 49 52 54 57 59 61 63

7 36 42 45 47 51 53 55 58

8 28 34 38 41 45 47 50 53

9 21 27 31 34 39 41 45 48

10 13 20 25 28 33 36 40 43

Retouching

Retouching light scratches in varnished wood

 Light surface scratches in varnished wood can often be made to disappear
without the application of a colorant. For example, rubbing with a soft
polishing pad moistened with ethanol (as one would do in the final step of
French polishing) may soften and amalgamate the fractured surface of the
scratch, rendering it invisible. If this is ineffective, padding the surface with a
very light application of dilute natural-resin varnish, synthetic-resin varnish,
or French polish may penetrate the scratch sufficiently to disguise it.
Some commercial scratch removers and polishes employ low-viscosity
oils, silicones, and waxes to wet down scratches. Oils and silicones can
penetrate the varnish and become absorbed by the wood, where they may
eventually darken and disfigure the surface. Silicones may also interfere with
subsequent gluing, retouching, and varnishing. Waxes (including
microcrystalline wax) hinder retouching and are difficult to remove, often
requiring the use of warm aromatic solvents.

Repairing deep scratches and bruises

If the original varnish and ground can withstand the application of steam,
dents in the wood can often be raised by placing a piece of cotton wool or
cloth dampened with distilled or deionized water over the dent and warming
with a heated spatula or tacking iron. Deep scratches in varnish are best filled
with varnish rather than with an opaque filler. Violin restorers have
traditionally used fillers consisting of combinations of sandarac, shellac, and
other resins dissolved in alcohol. Today, some restorers use synthetic-resin
floor varnishes (such as polyurethane) or even clear nail polish to fill cracks,
but these materials should not be used because they require extremely strong
solvents to remove them, and thus are essentially irreversible. Synthetic resin
varnishes such as Soluvar picture varnish and Regalrez 1078 and 1094 can be
used to fill scratches and dents in natural resin finishes. One advantage of the
Regalrez resins is that they dissolve in aliphatic solvents and after drying
remain soluble in them indefinitely (see Solvents and solvent cleaning).
While natural resin fills must be leveled mechanically or by abrasion,
Regalrez and Soluvar can be leveled and polished by wiping with a wad of
lint-free cloth or a piece of paper moistened with mineral spirits. This
procedure is much less disruptive to the surrounding varnish than mechanical
or abrasive leveling. Because Regalrez resins are insensitive to alcohol, one
may French-polish over them. By the use of Regalrez as a barrier, French
polish that has been applied over it can be later removed along with the
Regalrez. When filling scratches with any type of varnish, the scratch should
first be brushed with thinned varnish in order to saturate minute fractures,
followed by more viscous preparations. If the fill must be tinted to match the
surrounding varnish, color is best built up with successive layers (see below).
Whether filling with natural or synthetic resin varnishes, the refractive
index (RI) of the filler should match the surrounding varnish. If the RI of a
varnish fill is within 0.01 of that of the surrounding area, the repair will be
nearly invisible. A discrepancy of 0.04 or greater will create an optical
disturbance (visible when the instrument is tilted or rotated) that draws
attention to the fill. The RI of the surrounding varnish can be determined
under a microscope by placing a sample of the dried varnish in fluids of
known RI (such as those made by Cargille, Locke, and Dow Corning) and
employing the Becke line test to determine whether the sample lies above or
below the RI of the medium. (The Becke line is a halo surrounding a
transparent particle that moves towards a region of higher RI as the
microscope is focused upwards and towards a region of lower RI as the
microscope is focused downwards.) An inexpensive gemological
refractometer can also be used to determine RI. Once the RI of the damaged
varnish has been determined, the RI of the crack filler can be adjusted by
combining resins of different refractive indices. Use a weighted average of
the resins’ refractive indices and check a thoroughly dried sample by the
above methods. Below are the refractive indices of some commonly used
varnish-making materials:

Resin Refractive index

Amber 1.546

Cellulose nitrate 1.51

Dammar 1.539

Escorez 5300 1.546

 Laropal K80 1.529

Linseed oil, dried 1.52

Mastic 1.536

Manilla copal 1.544

MS2A 1.518

Polymethacrylate 1.48–1.52

Regalrez 1078 1.510

Regalrez 1094 1.519

Rosin 1.525

Sandarac 1.545

Shellac 1.516

Retouching with color

Violin varnish color extends from yellows, oranges, and reds to broken
hues of light straw to deep brown. This coloration is the net result of the
colored resins, plant extracts, dyes, and pigments used in the original
formulation, as well as the effects of age. A 200–300-year-old varnish will
not appear the same today as it did when it was first applied because the
original colorants may have faded or shifted in hue or the resins themselves
may have deepened in color. Furthermore, the refractive indices of certain
components of varnish (such as the drying oils) tend to increase, which may
render embedded pigments less opaque. Therefore, even if the original
varnish formula can be re-created, it might not be completely effective for
retouching. Nevertheless, we do have at our disposal many old violin varnish
recipes (see Varnish), and most of the natural materials used to make them
are still available. With our knowledge of these formulas and materials, it is
often possible to use or modify these old recipes to provide a reasonably good
retouching varnish. One such formulation, termed “1704” (which was
claimed by Giacomo Stradivari [1822–1901] to have been written alongside
that date in the Stradivari family bible) can be made by combining 45g of
seedlac, 9 ml of oil of spike, 7.5g of gum elemi, and 180 ml of ethanol.
Though this formula is probably spurious from a historical standpoint, it is
used in some major violin shops as a basic retouching varnish for fine
Cremonese violins.
When retouching, it is often more effective to use an almost dry brush and
apply color with a dotting or dabbing motion rather than with long, wet brush
strokes. Color is best built up gradually rather than by attempting to achieve
the final shade and concentration in a single application. The traditional
colorants used in violin varnish include insect and plant extracts (such as
kermes, cochineal, alkanet root, and buckthorn berries), lake pigments (such
as madder and alizarin; see Lake pigments), and earth pigments (such as
burnt umber, red ochre, and cinnabar). The unfixed plant extracts, and to
some extent the lake pigments, tend to be fugitive, though they are sometimes
useful in retouching. Effective retouching of scratched or damaged wood and
varnished surfaces can be carried out with watercolors or with dry pigments
that are mulled on a glass plate with virtually any solvent, medium, or
varnish. If mulled and applied with ethanol, pigments will adhere to the
underlying varnish by virtue of the ethanol's solvent power. For retouching
varnishes that are tinted with old aniline dyes (developed in the 1850s,
followed by the azo dyes in the 1880s), or when great transparency is
required, the newer, relatively light-fast Orasol dyes may be used. These can
be dissolved in ethanol and allowed to dry on a glass plate or a porcelain
mixing pan; for use, the colors are picked up with a retouching brush that has
been lightly moistened with ethanol. It is best to touch the wet brush to a
piece of absorbent paper to draw off any excess liquid before beginning to
retouch with these dyes.

Color theory

The color circle

 First devised around 1660 by Isaac Newton, the color circle or wheel
consists of an arrangement of colors in the order they appear in the spectrum
(as produced when light is split up by a prism or in a rainbow). Newton
divided his circle into seven principal colors: red, orange, yellow, green, blue,
indigo, and violet. Around 1730, a theorist named J. C. Le Blon discovered
that the colors red, yellow, and blue could be intermixed to produce any
color, which is the basis of the subtractive color system. In 1766, Moses
Harris published a color circle divided into eighteen segments, with red,
yellow, and blue as the primary colors, which he called “primitives,” with
three “compounds” of orange, green, and purple, along with twelve
intermediary hues. Johann Wolfgang von Goethe, Ignaz Schiffermuller, M.
E. Chevreul, Charles Blanc, and Charles Hayter also published works on
color theory that include color circles based upon the primary colors red,
yellow, and blue, though Thomas Young proposed red, green, and violet as
primaries and Ogden Rood, A. H. Church, and R. A. Houston proposed
subtractive systems that employed color wheels having red, green, and blue
primaries, which when combined produced black. Other subtractive color
systems have been designed. One, termed the process wheel, is used in color
printing, photography, and ink manufacture; it employs yellow, cyan, and
magenta as primaries, with green, violet, and orange as secondaries. The
Munsell system employs five principal colors: yellow, red, green, blue, and
purple. The concept of additive color was demonstrated by spinning color
discs and tops, which produced white from primary colors. James Clerk
Maxwell devised a color box that used prisms to produce spectral colors,
from which primaries could be isolated and superimposed to produce white.
Another example of additive color is theatrical lighting, which uses stage
lights fitted with gels of various colors that are aimed together to produce
white or other colors. In addition to circles or wheels, colors have been
arranged in a sphere (Philip Runge), a bi-conical solid (Wilhelm Ostwald), an
oblated sphere (Albert Munsell), a triangle (J. C. Maxwell, Josef Albers), and
a shoehorn-shaped form (Commission Internationale d’Éclairage, or CIÉ).
Color matching systems, such as those devised by Munsell, CIE, and Pantone
(the latter's used extensively in the printing industry), and information found
in the Kodak Color Dataguide (which includes a filter set used in
determining and correcting subtle color casts) are highly instructive for
developing skill in analyzing and mixing colors.
Secondary, tertiary, quaternary, and quinary colors

Secondary colors are obtained by mixing two primary colors. In the

subtractive system using red, yellow, and blue as primaries, the secondary
colors are orange (composed of red and yellow), green (composed of yellow
and blue), and violet (composed of red and blue). Tertiary colors are made by
mixing a primary with a secondary color that is adjacent to it on the color
wheel. A mixture of primary and tertiary colors yields quaternary colors, and
a mixture of a secondary and a tertiary color yields a quinary color. In the
additive system using red, green, and blue primaries, the secondary colors are
yellow (composed of red and green), cyan (composed of green and blue), and
magenta (composed of red and blue).

Broken hues

Combinations of unequal proportions of all three primary colors yield what

are referred to as broken hues. Examples of these include brown and olive
green.

Complementary colors

Complementary colors are diametrically opposed to each other on a color

wheel and thus consist of either a primary against a secondary color, or a
tertiary against another tertiary. In the additive color system, complementary
colors combine to form white. In the subtractive system, the intensity of a
color is reduced by adding its complement.

Shade, tint, and tone

Shades of a color are obtained by adding black. Tints are obtained by

adding white. Different tones (the relative lightness or darkness of a color)
are obtained by adding black, white, or gray.

Chroma
 Chroma refers to the saturation of a color. Pure colors are fully saturated.
Saturation is diminished by the addition of white, black, gray, or the
complementary color.

Value

The term value refers to relative lightness or darkness. Pure yellow, for
example, has a higher value than its complement, violet. Mixed colors tend to
have reduced saturation and are darker than pure colors, which is a point to
consider when retouching.

Tinctorial strength and covering power

Tinctorial strength refers to the ability of a particular pigment to tint white.

Covering power refers to the capacity of a pigment to hide colors beneath it.
The tinctorial strength and covering power of a pigment are related to the
opacity, refractive index, and size of the pigment particles.

Dichroism

This term has several meanings: it is used in reference to materials that

reflect and transmit different colors; to those that split light into various
colors; and to those that selectively absorb light having different
polarizations. It also refers to solutions or transparent materials that exhibit
various colors in different concentrations or thicknesses. Violin varnishes are
sometimes thought of as being dichroic because thin and thick layers exhibit
different colors (for example, as varnish coats are applied, the resultant color
may change from yellow to red).

Selective or metallic reflection

Certain pigments and dyes reflect a different color than the one they
transmit. The aniline dye eosin, for example, reflects a metallic yellow-green
color, though it transmits red. Gold leaf reflects the color yellow but
transmits green.
Transparency, translucency, and opacity

Transparency is the ability to pass light without scattering (such as a pane

of glass), translucency the ability to pass diffused light (such as frosted
glass), and opacity the ability to block light.

Cool and warm shades

Colors that tend towards red, orange, and yellow are termed warm; those
that tend towards blue and violet are said to be cool. Certain reds, such as
alizarin and carmine, are considered cool because they have a purplish cast.
Vermilion, which has an orange cast, is considered a “warm” red.

Physiological and psychological aspects of color

There are two types of light receptors in the retina of the eye: rods and
cones. Rods do not convey color information but are extremely sensitive to
light (a single photon of light may be enough to activate them) and respond in
low-light vision. Cones have chromophores that exhibit different spectral
sensitivities; some cones peak in the red region, some in the blue, and some
in the green. (This is very much like the dye-coupled silver grains used in
photographic color film, though the spectral response of color film does not
match that of the human eye). A particular wavelength or combination of
wavelengths of light will stimulate individual cones to different degrees. The
brain compares the strengths of the impulses produced by these receptors,
and color is perceived. It is well known that red and yellow combine to form
orange, but in fact, when red and yellow pigments are mixed together, their
wavelengths are not actually transformed into the wavelength of orange (a
spectrophotometer would show that the two distinct wavelengths are still
present). Thus, the color orange may be perceived directly by the wavelength
of spectral orange or by the reception of pairs of wavelengths whose average
equals that of orange. This is unlike the auditory sense, in which sounds of
different frequencies are heard as distinct tones (such as the notes of a chord)
rather than as averages of the individual frequencies.
Phenomena such as complementary-color afterimages and the Purkinje
effect (in which color shifts are perceived as the eye adapts to different
intensities of illumination) are physiological in nature. These may be factors
to consider in retouching; for example, neutral retouches may appear to take
on a hue that is complementary to the surrounding area. Concepts such as
color harmony and the association of colors with moods and musical pitches
are psychological or cultural in nature.

Mixing colors

Most pigments and dyes do not behave like pure spectral colors, so
mixtures of them tend to produce somewhat dull secondary and tertiary
colors, and sometimes even unexpected effects. An example of this can be
observed when certain black pigments are mixed with ochre, which results in
a dull green. This occurs because pigments such as bone black and lampblack
have a slight bluish cast that mingles with ochre to form green. For a darker
shade of yellow, a more neutral black (such as true ivory black), or the
addition of a warmer color such as burnt umber, is needed. Generally
speaking, it is difficult to create a vibrant secondary or tertiary color by
mixing pigments; this is why attempting to replicate a particular hue through
the use of “primary” pigments and dyes invariably fails to produce a true
spectrographic match and may result in metameric defects (i.e. colors that
appear to change hue when viewed in light sources having different spectra).
Unfortunately, there is often no recourse but to work with a primary palette
or to combine colors.

Color changes due to aging

When an old violin varnish is analyzed by a spectrophotometer, the

resultant color curves are often fairly smooth and lacking in strong spikes
(Figure 30). Over time, unfixed organic dyes and colorants tend to fade,
causing their spectrophotometric curves to smooth out. Cooked-oil varnishes,
made without added colorants, also tend to exhibit smooth
spectrophotometric curves. It should be noted that such curves cannot be used
to identify or characterize the formulas of varnishes as they are merely
indicators of color characteristics.
Figure 30 Spectrophotometry of Stradivari violin varnish (1694). The x axis
represents the various wavelengths in nanometers; the y axis represents
percent reflectance.

Retouching under artificial light

Retouching is often done in north light or under artificial light similarly

balanced. Though incandescent lamps yield a continuous spectrum, their
color temperatures are considerably lower than daylight, which ranges from
around 5000°K to 6500°K. Some quartz-halogen bulbs used in photographic
lighting are balanced for 3200° K, which is a bit higher than average
incandescent bulbs used in domestic lighting, which are around 2700°K to
3000°K. So-called “daylight” fluorescent bulbs have a higher color
temperature (around 5000°K) but have discontinuous spectra, which can
create color matching problems due to metamerism. Tri-phosphor fluorescent
bulbs are the best of this type, and those having a color rendering index (CRI)
above 90 are often used for retouching. LED bulbs also have discontinuous
spectra, though some are claimed to have CRIs as high as 98.

References
Philip Ball, Bright Earth: Art and the Invention of Color (New York, 2002).
Faber Birren, Principles of Color (Atglen, 1987).
M. E. Chevreul, The Principles of Harmony and Contrast of Colors and
Their Application to the Arts (London, 1859).
Arthur H. Church, The Chemistry of Paints and Painting (London, 1915).
Edith Anderson Feisner, Color Studies (New York, 2001).
George Field, Chromatics or, an Essay on the Analogy and Harmony of
Colours (London, 1817).
George Field, Field's Chromatography, or, Treatise on Colours and
Pigments as Used by Artists (London, c.1869).
Trevor Lamb and Janine Bourriau, eds., Color, Art, and Science (Cambridge,
1995).
A. P. Laurie, The Painter's Methods and Materials (London, 1960).
Ian Sidaway, Color Mixing Bible (New York, 2002).

Rosin

One of the problems encountered in cleaning violins and bows is the

removal of caked-on rosin. Players should be advised to blow or brush off
freshly deposited rosin dust immediately after playing. Rosin is abrasive, so if
it must be wiped off, this should be done with a soft cloth and a light touch.
Some musicians do not attend to this, and the rosin piles up like snow. The
longer it remains on the instrument, the harder it is to remove. When left
unattended, rosin ultimately bonds with the underlying varnish, absorbs dirt,
and turns black; at that stage it becomes a major conservation problem. The
blackened varnish of Niccolò Paganini's Guarneri violin is a striking example
of what can happen if deposited rosin is neglected for many years (Figure
31).

Figure 31 Front of Niccolò Paganini's Giuseppe Guarneri del Gesù violin

showing encrusted rosin.
 The natural solvent (see Solvents and solvent cleaning) for fresh rosin is oil
of turpentine, but conservators generally do not use it today because sticky
residues may be left behind. These residues are either gums that remain after
incomplete distillation or are partially polymerized turpentine oil (a reaction
caused by exposure to oxygen and light). Recently solidified deposits of rosin
can often be removed with mineral spirits containing a moderate proportion
of aromatics, such as Stoddard solvent (which contains 7–22% aromatics
depending upon the source), Shellsol 7EC (8% aromatics), or Shellsol 15
(22% aromatics). Some violin restorers resort to xylene, but many varnishes
cleaned with straight xylene become blanched (a whitish, cloudy effect). This
phenomenon is probably caused by remains of rosin that have not been
sufficiently cleared from the surface, though it may be due to the leaching of
some of the resinous components of the underlying varnish (which may
include rosin) that are redeposited as a discontinuous film. A safer approach
is to dilute xylene with an aliphatic solvent, which will restrain its action and
permit the rosin to be removed in a more gradual fashion with possibly less
effect upon underlying varnish. It may be advantageous to use toluene rather
than xylene because it evaporates more quickly. Very old and highly oxidized
rosin may require a stronger solvent, such as ethanol. In this case, start with
10% ethanol in an aliphatic solvent, and increase the percentage if necessary,
but keep a careful eye on the underlying varnish, which may be sensitive to
this combination of solvents. Vulpex soap (the concentrate can be mixed with
water or mineral spirits in a 1:6–7 proportion) has been found to be effective
in removing recently hardened rosin; one advantage of using this soap over
xylene and toluene is its lower toxicity, though it must be thoroughly cleared
with water or mineral spirits.

References
Robert L. Feller, Nathan Stolow, and Elizabeth H. Jones, On Picture
Varnishes and Their Solvents (Cleveland, 1971).
Knut Nicolaus, The Restoration of Paintings (Cologne, 1998).
Helmut Ruhemann, The Cleaning of Paintings: Problems and Potentialities
(New York, 1982).
Richard Wolbers, Cleaning Painted Surfaces (London, 2000).
 S
Safety equipment

When using volatile, flammable, and otherwise hazardous liquids (such as

most hydrocarbon solvents, acids, alkalis, etc.) and gases (such as those
released by aqueous ammonia and chlorine bleaches), conservators should
work under a fume hood or with a fume extractor equipped with an
explosion-proof fan positioned over the work. If this equipment is not
available, use a respirator or face piece equipped with the appropriate filter
cartridge (various filters are available for protection against organic vapors,
ammonia, acids, etc.) and arrange for the free flow of fresh air. Safety glasses
or face shields should be used to protect against chips thrown back by
machinery, as well as splashes of caustic or poisonous chemicals. Emergency
eye and skin wash stations should be set up in proximity to areas where
chemicals are used. Specially formulated sorbents should be on hand to
neutralize and pick up chemical spills; these are available in granulated form,
which can be shaken or scooped onto a spill, or in the form of blankets or
pads, which can be used to sponge up hazardous liquids. After spills are
cleaned up, the sorbents are swept or picked up and then disposed of in
accordance with local, state, and federal regulations.
Dangerous chemicals should be clearly labeled and stored in specially
vented storerooms or cabinets. Venting should accommodate gaseous
substances that are both heavier and lighter than air (i.e. vents should be
placed at both the floor and ceiling levels of storerooms or at both the top and
the bottom of cabinets). Flammable solvents should be stored in insulated
cabinets that are designed to provide protection in the event of fire. Chemical
cabinet shelves should be equipped with solvent- and corrosion-resistant spill
trays. Safety cans fitted with wire-mesh flame arresters are recommended for
storing large quantities of flammable liquids. Smaller quantities of solvents
and other liquids should be stored or decanted into unbreakable dispensing
bottles that are made of materials that will not be degraded by the chemicals
stored within them. Borosilicate glass is resistant to most chemicals (except
certain acids), though it is breakable. Low- and high-density polyethylene
bottles are adequate for many chemicals but are inappropriate for solvents
such as acetone and the halogenated hydrocarbons, which will dissolve this
plastic. Bottles made of fluorinated ethylene propylene or
polytetrafluoroethylene (Teflon) are unbreakable and can withstand most
solvents, as well as some acids and strong oxidizing agents. Though rather
expensive, containers made of these fluorinated plastics are recommended for
dispensing most of the chemicals used in conservation work.
Storage cabinets, safety cans, and bottles should be clearly marked with the
appropriate National Fire Protection Association (NFPA) hazard codes for
health hazard, flammability, reactivity (i.e. chemicals that may become
explosive or that are violently reactive); with the Globally Harmonized
System of Classification and Labelling of Chemicals pictograms (GHS); or
with European Union classifications. One should be mindful of potential
dangers that might result from spills, leakage, or evaporation. For example,
potassium and sodium cyanides are often sold in metal cans, which must
never be stored in cabinets that contain acids, as leakage or slow evaporation
of acid fumes from stoppered bottles will corrode the metal cans and react
with their contents to produce deadly hydrogen cyanide gas. Acids and their
fumes can also react with sulfide compounds to yield poisonous hydrogen
sulfide gas. Acids and bases are best stored in separate cabinets and
preferably on shelves below eye level.
While ubiquitous latex gloves can be used to protect the hands during
routine cleaning with water, soaps, detergents, alcohols, and certain mild or
diluted acids and alkalis, they are permeable to many of the solvents used in
conservation work, such as mineral spirits, Stoddard solvent, hexanes,
xylene, toluene, and many chlorinated solvents. Nitrile gloves are also
resistant to soaps, detergents, mild and diluted acids, alkalis, alcohols, oily
substances, as well as aliphatic hydrocarbons, though they are not safe to use
with acetone, trichloroethylene, and methylene chloride, for example. Gloves
coated with polyvinyl alcohol can be used with xylene and toluene, as well as
with chlorinated solvents, such as methylene chloride, carbon tetrachloride,
and methyt ethyl ketone, which will attack most other types of gloves.
Polyvinyl-alcohol-coated gloves, however, must not be exposed to water, as
it will dissolve the coating. Laminate-film gloves (such as the Silver Shield
and Barrier brands) are resistant to virtually all of the organic solvents, but
they tend to be bulky and uncomfortable. See Permeation/Degradation
Resistance Guide for Ansell Chemical Resistant Gloves (Ansell, 2003), which
provides recommendations for handling over 160 chemicals.
One should observe common safety rules when mixing chemicals, such as
adding acid to water (rather than adding water to acid, which may result in
the violent release of heat and dangerous splattering), never combining
chlorine bleach with ammonia or vinegar (which results in the release of
poisonous and potentially explosive substances such as chlorine gas, nitrogen
trichloride, and hydrazine), keeping sodium or potassium cyanide away from
acids, and never sniffing unmarked bottles to determine their contents (for
example, a whiff of household-strength ammonia is highly irritating). One
should be aware of problems arising from certain types of chemical reactions,
including those that are exothermic (heat-releasing), which might cause glass
containers to shatter or plastic ones to melt. Electrolytic cleaning results in
the release of hydrogen gas, which is explosive, so there should be a free
flow of fresh air if large objects are being treated. Never mix hydrogen
peroxide and acetone or other flammable liquids as these combinations are
explosive. Fumigants must be properly vented after use, and even inert,
nonpoisonous gases such as argon and nitrogen used in anoxic disinfestation
can be dangerous in high concentrations. When heating flammable liquids,
one should be aware of their flashpoints and take precautions to prevent and
deal with fire and explosion. Many chemicals should not be washed down the
sink but require certain procedures for disposal. Be aware of local ordinances.
Read and heed all safety warnings that come with chemicals, and download
product and safety information from the manufacturer or supplier. Exercise
appropriate care when using chemicals that are poisonous, or irritating, or
whose vapors have narcotic effects.
When using noisy power machinery, it is advisable to wear ear plugs or
acoustic ear muffs. When using machinery that throws off chips, particles, or
sparks (such as motorized saws, drill presses, lathes, sanders, grit-blasters,
and grinders), it is imperative to wear goggles, safety glasses, or face shields.
When brazing and soldering, wear goggles or face shields and protective
gloves to protect against molten metal and flux. When welding, wear
protective gloves and special welder's goggles to protect eyes from the bright
light and ultraviolet rays emitted by the torch's flame. Welding aprons and
gloves made of leather are recommended because they are less flammable
than cloth or plastic and provide better insulation from heat. In many locales,
special training and a license are required to use welding equipment.
All electrical equipment should be properly grounded (earthed). Be
conscious of the current (amperage) drawn by machinery so that wiring is not
overloaded. To determine the amount of current drawn by a piece of
machinery under heavy load (or other electrical equipment, including
photographic lighting), divide the wattage rating by the operating voltage: the
quotient will be the amperage. Thus, if you attempt to use four 500-watt
tungsten bulbs to photograph a musical instrument, you will blow a fuse or
trip a circuit breaker set for 15 amps (4 × 500 watts/120 volts = 16.6 amps).
There are five classes of fires: Class A (common combustibles, such as
paper, wood, textiles); Class B (flammable gases and liquids, such as
solvents); Class C (live electrical equipment); Class D (combustible metals,
such as magnesium, lithium, sodium); Class K (cooking oils and fats). Fire
extinguishers should be selected according to the type of work being done
(use of flammable solvents such as alcohols and mineral spirits would require
a fire extinguisher capable of putting out a Class A fire, while making oil
varnish might cause a Class K fire). So-called “chemical extinguishers” use
chemicals such as Halon, monoammonium phosphate, or sodium bicarbonate
to put out types A, B, and C fires. The Halon extinguishers are extremely
effective, but they leave a powdery substance that must be cleaned up and
may damage certain types of equipment. Carbon dioxide fire extinguishers
are used on Class B and C fires and leave no residue, though they temporarily
deposit a film of dry ice (frozen carbon dioxide) that may damage musical
instruments that are temperature sensitive. Water fire extinguishers should
not be used in electrical fires (as water may conduct electricity) or when oily
materials are on fire (oil floats on water and will spread the flame);
furthermore, water may damage some types of musical instruments). Fire
extinguishers should be clearly marked according to the type of fire they are
designed to extinguish; they should be mounted so that they can be easily
reached; and they should be located near areas that are likely to catch fire
(such as soldering booths, welding stations, or areas where hotplates and glue
pots are used); furthermore, they should be positioned to facilitate both egress
should a fire get out of control, and entry for those attempting to put out the
fire (i.e. extinguishers should be located at the far end of a room as well as at
points of entry).
 A first-aid kit, including a fire blanket, should also be on hand.
The regulatory and safety agencies in the US are as follows:

American Society for Testing and Materials (ASTM)

Food and Drug Administration (FDA)
National Fire Protection Association (NFPA)
Occupational Safety and Health Administration (OSHA)
Underwriters Laboratories (UL).

Scaling

The term scaling is used in several ways. With regard to stringed-keyboard

instruments (such as harpsichords, clavichords, and pianofortes), scaling
refers to the progression of string lengths throughout the compass of the
instrument, while the scaling of organ pipes refers not to their speaking
lengths but rather to their widths in proportion to their lengths (a “wide-
scaled” pipe, such as an open flute, would have a greater diameter than a
similarly pitched string-toned pipe, such as a viola da gamba).
Many early stringed-keyboard instruments exhibit what is termed
“Pythagorean scaling,” which has come to mean that their strings double or
halve in length as one proceeds from octave to octave. However, just as
Pythagorean temperament is constructed with the ratios of the octave and
fifth (see Tuning and temperament and Intervals), pure 3:2 ratio fifths were
evidently also employed in establishing the scaling of keyboard instruments,
for in many harpsichords one finds pin or scribe marks on the soundboard
and wrestplank that indicate where both the C and F pins of the bridges and
nuts are positioned. Early makers typically adhered to this scaling system
through the middle range of an instrument's compass, but they were
compelled to deviate from the Pythagorean ideal as they approached the bass,
otherwise a four-octave keyboard instrument having a c2 string length of 300
mm (around a foot) would have a C string length of 2400 mm or around 8
feet. (Were a seven-octave instrument, such as a modern piano, to employ
Pythagorean scaling, a c2 string length of 364 mm [14.33 in.] would require a
C1 string length of 5824 mm, or over 19 feet.) In the treble of most early
keyboard instruments, the string lengths also tend to deviate from the
Pythagorean ideal because makers took advantage of a phenomenon known
as “tensile pickup,” in which wire made from iron and steel, and to a lesser
extent brass, becomes progressively stronger as it is drawn down in size
through successive dies. Because of this extra tensile strength, the treble
strings could be made longer, which enabled makers to position the treble
ends of their bridges further back from the acoustically restrained edge of the
soundboard. However, with the development of the modern piano, makers
abandoned Pythagorean scaling entirely and instead they employ a constant
scaling factor throughout much of the range of their instruments. For
example, in A Treatise on the Art of Pianoforte Construction (1916), author
Samuel Wolfenden advocates calculating string lengths from the top note c5
down to c by starting with a top note string length of 54 mm and obtaining
each successive string length by multiplying by 39/37 (as opposed to 35/33,
which is his approximation of the equal-temperament semitone interval of
). Thus, his octaves progress by a factor of 1.88 rather than 2.
For convenience, the length of c2 (C above middle C) is often given as the
“scaling” of keyboard instruments. Because yellow brass and iron wire have
different strengths and densities, some keyboard historians (Thomas and
Rhodes, 1967; O'Brien, 1981, 1990) have hypothesized that instruments were
designed, or “scaled,” for yellow-brass wire if their c2 strings range in length
up to about 290 mm and were scaled for iron wire if their c2 strings were over
290 mm. Barnes (1965) concluded that scaling differences were simply
reflections of pitch rather than stringing material, while Shortridge (1960)
maintained that short- and long-scaled instruments were designed for regular
and transposed tuning. It is also possible that some harpsichord makers
preferred or were indifferent to the sound of short-scaled instruments strung
in iron wire. The distinction between brass and iron scaling breaks down with
the pianoforte, as most were strung in iron regardless of their scaling.

Calculating the breaking frequency of strings

It is generally acknowledged that early instrument makers were aware of

the tensile strength of wire and gut. String lengths were probably determined
empirically by use of a monochord or simply by copying successful designs.
In addition, makers had to provide a safety factor to accommodate the raising
and lowering of pitch during tuning, additional stresses imposed by changes
in temperature and humidity, as well as by playing. Attempts have been made
to determine the pitch of keyboard instruments from their string lengths, but
there are too many unknowns, such as variability in the tensile strength of
wire and makers' predilections or tolerances for the sound of strings tuned
below their breaking points (Wraight 1997). Nevertheless, it is valuable to
know the absolute breaking frequencies of strings of any length in order to
assess safety margins at a given pitch. Mersenne's Law can be adapted in the
following way to compute an approximation of these frequencies:

where

F is the frequency in Hz
L is the length of the string in meters
T is the tensile strength in pascals (Pa; substituted for “tension”)
ρ is the density in kg/m3 (substituted for “mass per unit length”).

Using data reported by Mersenne in 1635 for a particular sample of iron wire
having a diameter equivalent to 0.38 mm and a breaking load of 19 livres, the
tensile strength of this wire would have been 821 MPa (equivalent to
821,000,000 Pa; Goodway and Odell 1987), while the density of an unrelated
sample of old iron wire has been measured at 7690 kg/m3 (Goodway and
Odell, 1987). Using the above formula and these data, the breaking frequency
for Mersenne's wire at a c2 length of 315 mm (0.315 m) would have been 519
Hz. This would have provided a safety margin of about 84 cents at a pitch of
A=415. If an instrument with a c2 string length of 290 mm were strung in
brass having a tensile strength of 634 MPa (as derived from data reported by
Coulomb in 1784) and a density of 8240 kg/m3 (Goodway and Odell, 1987),
the breaking frequency of c2 would have been around 478 Hz, or about 44
cents above c2 at a pitch of A=392 (French Baroque pitch). These meager
safety margins may be reflections of uncharacteristically weak samples of
wire or inaccurate measurements of tensile strength.

Gut strings (see also Gut strings)

 While the density of gut is around 1300 kg/m3, its tensile strength varies
according to the rate of twist: strings used for the high-pitched strings of
violins, viols, lutes, guitars, mandolins, etc. are traditionally made with a low
twist rate, which has greater tensile strength (around 392 MPa) than the high-
twist gut used for the lower-pitched strings (around 294 MPa). Necked
stringed instruments (such as those of the violin and viola da gamba families,
lutes, guitars, and mandolins) may also be thought to exhibit “scaling,” but
only the highest-pitched string is tuned close to its breaking point. If the
remainder of the strings are mounted across the same fingerboard they are in
a sense progressively foreshortened, though this problem is mitigated to some
extent in lutes and theorbos that have secondary pegboxes and extended
necks for the lowest-pitched strings. Generally only the top few strings are
strung in the high-strength, low twist gut, while the lower-pitched strings are
made of gut having a greater twist rate. The high twist rate imparts greater
flexibility, which enables the thicker, slack strings to produce overtones that
are in better tune with their fundamentals. If we consider a typical violin
string length of 327 mm, using the formula above, a low-twist gut E string
would break at a frequency of about 840 Hz, or about a major third higher
than an E tuned to a pitch of A=440. A rule of thumb used by some of today's
string makers for determining the safe pitch for gut and nylon strings is:
F = 240,000/L

where F is the frequency, and L is the string length in millimeters.

For an E string length of 327 mm, this formula yields a pitch of 734 Hz,
which provides a safety margin of 186 cents, or slightly less than a whole
tone at a pitch of A=440. Though the violin can be safely tuned to modern
pitch with gut, it could not have been tuned (even to Baroque pitch) with the
low-tensile-strength iron wire that was used up through the early nineteenth
century for stringing harpsichords and early fortepianos – its E string would
have broken at around 500 Hz. Strangely, Michael Praetorius (1619) states
that violins will produce a much more gentle and pleasant sound when strung
with brass and steel strings than with gut. The high-tensile-strength steel
music wire that was developed in the mid nineteenth century is strong enough
to be tuned to A=440 and higher, though steel violin E strings were not
widely employed until the early twentieth century.
Determining overall string length from fret positions

In the author's study of Stradivari's mandolin neck patterns (which have

equal-temperament fret positions marked on them; Pollens, 2010), the
following formula was used to compute the overall string lengths associated
with each pattern:
L = N/1 − (17/18)F

where

L is the overall string length

N is the distance between the nut and the fret
F is the fret number.

Scaling of organ pipes

While the lengths of the pipes cannot be foreshortened, if the pipes' widths
doubled on the octave they would require enormous quantities of wind for
them to speak, and windchests would have to be much larger to accommodate
their girth. Rather than doubling in diameter from octave to octave, pipe
diameters are scaled differently. For example, in his Lehrbuch der
Orgelbaukunst (Weimar, 1855), Johann Gottlob Töpfer suggests that pipes
double in diameter on the 16th, 17th, or 18th note rather than on the octave; a
1: ratio was preferred by many early German and French organ builders.

References
John Barnes, “Pitch Variations in Italian Keyboard Instruments,” Galpin
Society Journal 18 (1965), pp. 110–116.
John Barnes, “The Specious Uniformity of Italian Harpsichords,” in Edwin
M. Ripin, ed., Keyboard Instruments: Studies in Organology (Edinburgh,
1971).
Martha Goodway and Jay Scott Odell, The Metallurgy of 17th- and 18th-
Century Music Wire (Stuyvesant, 1987).
John Koster, Keyboard Instruments in the Museum of Fine Arts, Boston
(Boston, 1994).
Grant O'Brien, “Some Principles of Eighteenth-Century Harpsichords and
Their Application,” The Organ Yearbook 12 (1981), pp. 160–176.
Grant O'Brien, Ruckers: A Harpsichord and Virginal Building Tradition
(Cambridge, 1990), pp. 55–66.
Stewart Pollens, Stradivari (Cambridge, 2010).
Michael Praetorius, Syntagma Musicum: De Organographia (Wolfenbüttel,
1619).
John D. Shortridge, “Italian Harpsichord Building in the 16th and 17th
Centuries,” Contributions from the Museum of History and Technology,
United States National Museum Bulletin 225, Paper 15 (Washington, DC,
1960), pp. 93–107.
William R. Thomas and John J. K. Rhodes, “The String Scales of Italian
Keyboard Instruments,” Galpin Society Journal 20 (1967), p. 48.
Johann Gottlob Töpfer, Lehrbuch der Orgelbaukunst, 4 vols., ed. Paul Smets
(Mainz, 1955–1960).
Samuel Wolfenden, A Treatise on the Art of Pianoforte Construction (Old
Woking, 1975).
Denzil Wraight, “The Stringing of Italian Keyboard Instruments c.1500–
c.1650,” PhD thesis (Queen's University Belfast, 1997).

Sealing wax

Wax seals are occasionally found affixed to the makers' labels of early
keyboard instruments as well as to the backs or pegboxes of violins and viols
as marks of ownership. Sealing wax has also been used to attach woodwind
instrument pads to keys, to lengthen harpsichord jacks, and as a quick-drying
sealant and cement.
According to Robert Dossie's The Handmaid of the Arts (London, 1764),
there were two types of sealing wax: soft and hard. Soft sealing wax was used
for charters, patents, and other documents, while the hard sealing wax was
intended for closing up letters (in Dossie's time, posted letters were typically
folded up to enclose the writing and to form an integral “envelope,” which
was sealed with wax).
Dossie's soft sealing wax was made by heating the following together:

16 ounces beeswax
3 ounces turpentine
1 ounce olive oil
1 ounce of coloring matter: vermilion (red), red lead (coarser red), ivory
black (black), verdigris (green), smalt (blue), verditer (light blue),
massicot (yellow), or a mixture of vermilion and smalt (purple).

The hard sealing wax was made by heating the following together:

2 parts shellac or seedlac

1 part resin (likely pine colophony or the resin derived from boiling off
the spirit oil in Venice turpentine)
1 part vermilion (or the other pigments listed above).

Conservation issues

Seals made of shellac, seedlac, and rosin are sensitive to alcohols, acetone,
cellosolve, chlorinated solvents, and many other solvents, while those
containing beeswax may also be sensitive to aliphatic and aromatic
hydrocarbon solvents. If casts must be made of seals, plaster of Paris or
Hydrocal are safer than silicone rubber impression materials, which may
contain oily release agents or produce alcohol during curing.

Reference
Robert Dossie, The Handmaid of the Arts (London, 1764).

Sharpening tools

General
 New cutting tools, such as chisels, carving gouges, and plane blades,
generally come from the factory with their primary bevels ground at too great
an angle, and some manufacturers unfortunately grind a small bevel on the
underside of the blade that interferes with sharpening and must be ground off.
Motorized bench grinders with 6-inch and 8-inch abrasive wheels can be
used to regrind primary bevels. If there is a bevel on the underside of the tool
or blade, it is important to grind this off while regrinding the primary bevel of
the top surface. Bench grinders equipped with half-horsepower 60-cycle AC
motors generally turn at very fast speeds (often 1725 r.p.m.), which can
quickly heat the blade and draw its temper (the oxidation color at the cutting
edge should never extend beyond a straw color; see section on Tempering
steel). By working slowly, with light pressure, and with frequent immersion
of the tool in cool water, these relatively inexpensive and ubiquitous
motorized grinders will suffice. The gray-colored aluminum oxide abrasive
wheels that normally come with these grinders are the toughest, but the white
or pink aluminum oxide abrasive wheels are preferable because the abrasive
particles are not bound together as tightly and tend to wear off, providing a
fresh grinding surface rather than becoming clogged with metal debris. They
are also less likely to heat up the tool and draw its temper. The white-colored
wheels are the most friable, with pink classified as semi-friable. If grinding
wheels become scored or glazed, the surface can be renewed with a dressing
stick or diamond-pointed dressing tool. Gray, white, and pink grinding
wheels should not be used with nonferrous metals. If a nonferrous metal is
inadvertently ground on a wheel, it will have to be resurfaced with a dressing
stick. To grind nonferrous metals such as aluminum and brass, special black
silicon carbide wheels are available. Ruby-colored grinding wheels are used
to shape hard tool steels. Green silicon carbide wheels are used for carbide or
carbide-tipped cutters. Because of the relatively small diameter of 6–8-inch
grinding wheels, they produce a distinct hollow-ground bevel, which is an
advantage when creating the secondary bevel (see below). A 60 grit abrasive
wheel is useful for removing metal quickly; a 120 grit wheel is more suitable
for regrinding primary bevels.
Grinding wheels are often marked with series of code letters and numbers
that indicate the abrasive type (such as aluminum oxide and silicon carbide),
grain size (60, 80, 150, etc.; see grit size comparison chart), grade (soft to
hard), structure (density of abrasive particles), and bond type (such as
vitrified, shellac, or resin). See Oberg, et al., Machinery's Handbook (2012)
for details.
To maintain the proper angle when grinding, motorized grinding wheels
have adjustable tool rests that can be set at the proper angle with a protractor.
On the better grinders, these tool rests are made of cast iron that has been
ground flat; less expensive models have tool rests made of bent sheet metal
that are often not perfectly shaped. The tool rest must have a true surface so
that the blade can be smoothly shifted from side to side during grinding,
which is necessary for achieving a perfectly straight bevel. If the blade is
wider than the grinding wheel, this back-and-forth motion will have a planing
effect and maintain a perfectly straight bevel. If one does not have facilities
for regrinding or milling the tool rest, it can be trued up by the technique of
flat-filing (see Metalworking).
After the primary bevel has been ground on a grinding wheel, the
secondary bevel is ground and refined by hand on a series of flat bench
stones. Because of the hollow ground primary bevel produced by a 6–8-inch
grinding wheel, one can use the back edge of the bevel to maintain a constant
angle on the bench stone while forming and refining the secondary bevel,
which is the cutting edge of the tool or blade. Do not use a trailing motion
when working with a bench stone; instead, push the cutting edge against the
stone or use an elliptical or circular motion. Make a few passes on the bevel
side and then turn the blade over and do the same to the back, keeping the
back of the blade perfectly flat against the sharpening stone. A few strokes on
a 1000 grit waterstone should produce a relatively smooth secondary bevel
that can be further refined with 4000 and 8000 grit waterstones. The
secondary bevel should ideally be as narrow as possible (around 1–2 mm) to
reduce cutting friction. With repeated sharpening, the secondary bevel will
gradually widen, ultimately requiring regrinding of the primary bevel.
Cutting edges can be further refined by honing them against a plank of soft
and evenly textured wood (such as lime or basswood) that has been smeared
with chromium oxide polishing compound or the pastes used on strops for
sharpening old-fashioned cut-throat razors. Here, a trailing motion should be
employed, with the blade being flipped from one side to the other after each
stroke. Blades honed by this method should be sharp enough to shave the hair
off one's arm. Do not hone blades on muslin or flannel buffing wheels, as this
will destroy their flatness and effectively increase the cutting angle.
Chisels

Cabinet and paring chisels should have their primary bevels reground at an
angle of 25°. For fine work on soft woods (such as spruce and other
soundboard woods), the primary bevel can be reduced to about 20°, while
mortise and firmer chisels, which are intended to be struck with a mallet, may
be ground at 30°. With the exception of skew chisels, the primary bevel is
generally ground at 90° relative to the length of the blade. The bottom faces
of woodworking chisels should be perfectly flat, with no bevel.

Plane blades

For block and smooth planes used for planing broad surfaces, such as the
soundboards of lutes, guitars, and keyboard instruments, the blade should be
ground so that it is perfectly straight through most of its width, though at the
corners the cutting edge should be relieved or curved back a bit so that sharp
corners do not project past the plane's mouth; projecting corners would create
disfiguring ridges and marks in the planed surface. The blades of jointing
planes, which are primarily used for planing the edges of boards, and those of
rabbet planes, should have their blades ground perfectly straight through their
corners. The secondary bevel and honing procedures are the same as for
chisels.

Carving gouges

Grinding the primary bevel on out-cannel carving gouges is problematic

because the curvatures of these tools are often roughly shaped by forging or
hand-grinding, and the undersides often have a different radius of curvature
than the upper surfaces, which makes it difficult to produce a properly shaped
and even bevel if one uses the grinding wheel's tool rest to support the chisel.
To grind a perfect bevel, make a jig to support the gouge's handle at such a
distance from the grinding wheel that the cutting edge is brought up to the
wheel at the proper angle. Commercially made jigs are available for this
purpose, but because they generally support the tool by cradling the end of
the handle, they do not work well with gouges that have faceted handles. To
make a jig for sharpening gouges fitted with faceted handles, drive a
sharpened metal spike at an angle through a length of 1-inch by 1-inch wood.
This strip of wood is then clamped to the tabletop just below the grinding
wheel so that when the sharpened end of the spike is inserted in a depression
in the gouge's handle (which can be made with an awl or center punch), the
bevel of the gouge comes to rest at the correct angle against the grinding
wheel. By rotating the tool against the grinding wheel (using the sharpened
end of the spike as a pivot), a perfectly smooth and uniform bevel can be
ground.
The secondary bevel on carving gouges is hand-ground, using a back-and-
forth rocking motion on flat waterstones or carver's slips, while the inside
surface is refined with curved stones. The cutting edge can be honed on a 6–
8-inch diameter hard felt wheel charged with chromium oxide polishing
compound. The cutting edge should be positioned so that the motorized
wheel turns away from the cutting edge (if the wheel turns towards the
cutting edge, it may catch and tear the gouge out of one's hands). A
sandpaper file can be used to shape a felt wheel so that it can reach into the
concave surface of the gouge. Another technique for honing the cutting edge
is to use the gouge to trim off a corner and carve a channel in a soft and even
textured block of wood. The rounded corner and channel will then have the
same radii as the inside and outside cutting edges of the gouge and can be
charged with chromium oxide compound to bring the tool to razor sharpness.

Knives

Knife grinding generally requires some modification of the grinding

wheel's tool rest so that the knife blank can be brought against the grinding
wheel at a sharp angle (about 12.5° for knives that are beveled on both sides).
Once this angle is set, even curved knife blades (such as those used in violin
making) can be easily ground. Some knives are beveled on one side with the
opposing side left flat; in this case, the grinding wheel's tool rest should be
adjusted to about 25°. Secondary bevels are made on flat grinding stones and
can be further refined on a piece of soft, evenly textured wood charged with
chromium oxide compound, as described above. Again, knives should be
honed to razor sharpness.

Hand saw blades

 Hand saws and saw blades designed for cutting wood come with teeth
designed for either cross-grain or rip sawing. Before sharpening, the teeth of
saws should be “set” with a special tool (called a “saw-set”) that alternately
bends the teeth from one side to another. This provides clearance for the back
of the saw so that it does not bind in the cut. The degree of set can be
adjusted by a set screw. Cross-cutting teeth are sharpened to a point and do
their work by grinding the wood; ripping teeth are sharpened straight across
to form little chisels that produce a less ragged edge when cutting with the
grain. After using the saw-set, the teeth are sharpened with a single-cut,
three-square (triangular) file (see Files under Metalworking) in such a manner
as to preserve their proper profile.

Scrapers

The cutting edges of heavy-duty cabinet scrapers used in finishing

hardwood (such as oak floors) are often ground at 90° and have cutting burrs
raised from that angle. For fine work (such as scraping softwood) it is
advisable to grind the cutting edge to around a 45° angle, and then refine the
edge with sharpening stones. The burr is then gradually raised with a few
light strokes of a burnisher, starting by orienting the burnisher at the bevel
angle and gradually increasing the angle until the burr is turned. The larger
the burr, the heavier the cut it will make. Scrapers are made of soft-tempered
steel and can be ground to any shape on a grinding wheel. The fractured edge
of a sheet of plate glass makes an excellent scraper, especially for softwood.
Curves can be laid out with a diamond scriber and then broken apart; the
cleanly broken edge can be used for scraping, though the edge on the scribed
side is roughened by the scriber and will not scrape cleanly.

Drill bits

The geometry of twist drill bits is complex, and special grinding jigs or
grinding machines are required to sharpen them. The cost of these grinding
machines is now quite reasonable, though modern high-speed steel bits hold
their cutting edge rather well (especially in wood), and it may be more
economical to replace bits rather than resharpen them when they become dull.
A note on sharpening stones

Up until the early 1970s, extremely fine Belgian or Ardennes yellow

coticule honing stones and hard natural Arkansas stones were available, but
those being quarried today are of comparatively poor quality despite the high
prices being charged for them. The author recently purchased an Ardennes
yellow coticule stone to sharpen a straight razor – but it was completely
useless due to a slight curvature of its grinding surface (flattening such a
stone is a laborious process). Some excellent natural waterstones are still
coming out of Japan, but they are rather expensive and difficult to find. Japan
used to be the primary source for man-made waterstones, but the US firm
Norton now has an excellent selection of them.
It is important to keep sharpening stones flat and their surfaces unclogged.
Man-made waterstones tend to wear down quickly and develop hollows, so
after use they should be rubbed while wet against a flat stone of coarser grit
in order to true up their surfaces and clear away metal debris. Rubbing stones
against each other prior to use will also produce an abrasive slurry that assists
in sharpening. It is unnecessary to keep waterstones in a bucket of water
between sharpening sessions; running them under the tap for a minute or two
just prior to use is sufficient.
If waterstones are unavailable, oilstones can also be used. These include
coarse silicon carbide, medium-grit India stones (man-made aluminum
oxide), and fine Arkansas stones (natural soft, black, and hard translucent
natural novaculite). The quality of natural Arkansas stones has declined in
recent years.
Other types of sharpening stones include die-makers' slips, engravers'
points, silversmiths' stones, toolroom finishing sticks, and garnet and
diamond files. These come in a wide range of shapes and grits and are useful
in forming and refining the shapes of complex molding cutters and scrapers,
deburring, and other surfacing operations. Flat metal plates charged with
emery, carborundum, and diamond compounds can be used to true up the
soles of planes and other flat surfaces.

References
Paul N. Hasluck, Manual of Traditional Wood Carving (London, 1911).
Albert Jackson, David Day, and Simon Jennings, The Complete Manual of
Woodworking (New York, 1994).
Erik Oberg, Franklin D. Jones, Holbrook L. Horton, and Henry H. Ryffel,
Machinery's Handbook 29th Edition (New York, 2012).
Harry Walton, Home and Workshop Guide to Sharpening (New York, 1967).

Silver cleaning

Silver is readily tarnished by exposure to sulfur dioxide in the air and

sulfurous fumes emitted by wood, cloth (especially wool), rubberized carpet
matting, rubber bands, some varnishes and paints, and certain foods (such as
egg). Silver sulfide is yellow to light brown, sometimes verging on iridescent
purple. Silver chloride, generally black in color, is formed by contact with
chlorinated tap water and salt from perspiration, food, and ocean water.
The silver used to make musical instruments is rarely pure and is generally
alloyed with copper. X-ray fluorescence (using peak-height ratios) has
determined that a button and ferrule made by François-Xavier Tourte consists
of 88.7% silver and 11.3% copper, which is nearly identical to a particular
écu coin minted in France in 1781 in the possession of the author, which is
88.4% silver and 11.5% copper. Some silver alloys include American coin
silver (90%), sterling silver (92.5%), “French standard” silver (95%), and
Britannia silver (95.8%). Both coin and sterling silver have been widely used
in making flutes. Flutes made by Louis Lot in Paris around 1880–1890 have
body tubes made of an alloy of about 95% silver and 5% copper; the
American maker Verne Q. Powell used sterling silver, while the Haynes
company has used 90% silver.
Lightly tarnished silver may be cleaned as follows:

1 Remove protective coatings with acetone or gelled methylene

chloride.
2 Wash in a nonionic detergent (such as 1% Dehypon LS45).
3 Polish with a slurry of precipitated chalk, 0.1 micron gamma alumina
or optical-quality rouge (obtainable from suppliers of materials for
grinding telescope mirrors). If these abrasives are unavailable, a
 good grade of silver polish containing jeweler's rouge, such as Mish's
Silver Polish, can be used. A soft, open-weave cloth, such as
cheesecloth, may be used to apply the abrasive slurry.
4 Thoroughly remove polishing abrasives with running water, followed
by rinses with distilled or deionized water.
5 Dry with a soft cloth.

Chemical cleaning

Moderately tarnished silver can be immersed in a 5–10% solution of

disodium EDTA. Treatment may extend for hours. Subsequent treatment with
5–10% formic acid may be required to remove residual tarnish.
Immersion in 15% ammonium thiosulfate or 30% sodium thiosulfate may
be effective in removing both silver sulfides and chlorides.

Electrochemical and electrocleaning

The author has never had much success with the technique of using
aluminum foil and a baking soda (or sodium hydroxide) electrolyte to remove
tarnish from silver.
The electrocleaning of silver and silver-plated instruments is risky because
of the possibility that dissolved copper (from either the silver alloy or the
base metal of plated instruments) will be deposited on the silver surface
unless the voltage and current density are closely controlled and the anode is
carefully shaped to follow the contours of the instrument. Degrigny et al.
(1994) and Espie et al. (2000) describe the successful electrolytic cleaning of
tarnished silver-plated instruments in the collection of the Musée de la
Musique in Paris. This work was not carried out in the museum but by a
private firm that is specially equipped for electrocleaning and plating.
Perhaps the most famous electrolytic treatment of a musical instrument was
the reduction of wholly mineralized silver parts of an ancient lyre from Ur,
which was carried out in 1962 by R. M. Organ of the British Museum. The
silver parts were made the cathodes (attached to the negative terminal of a
partially rectified AC power supply) and immersed in an electrolyte
consisting of 3% sodium hydroxide (which was replaced periodically with
fresh solutions). A carbon rod was used as the anode, and the current density
was set at 10 mA/dm2. After three to four weeks, the corroded fragments
were converted into metallic silver, which was then polished with a glass-
bristle brush (Organ, 1965).

Silvered objects

Silvered base-metal instruments that are slightly discolored may be

refreshed by lightly brushing with a soft bristle brush moistened with a slurry
of cream of tartar (potassium bitartrate), followed by thorough rinsing. If the
instruments have been soft-soldered, rinsing should be done with tap water
rather than distilled or deionized water.
To remove silver plating immerse the object in a solution of 3% potassium
nitrate dissolved in 30% sulfuric acid.

Spot plating with silver

Small areas where silver plating has worn off underlying brass or copper
can be replated with the following:

3 parts silver chloride

20 parts potassium bitartrate (cream of tartar)
15 parts sodium chloride (common salt)

Mix these together and moisten with water. The resulting paste is then picked
up with a piece of soft cloth and rubbed over the worn area, which should be
clean and free of corrosion. The freshly applied silver can then be polished
with a soft cloth and precipitated chalk or a slurry of optical-quality rouge.
Heavier deposits of silver can be applied electrolytically (see
Electroplating and electrocleaning).

Precautions

During the manufacturing process, the soldering and annealing of objects

made of sterling silver and other silver alloys containing copper often causes
firescale (reddish-brown cuprous oxide and black cupric oxide) to form on
the surface. Silversmiths may disguise this disfigurement by quenching the
workpiece in sulfuric acid, which plates the surface with a thin layer of pure
silver. Subsequent abrasive polishing may cause the underlying firescale to
break through the surface.
Commercial “silver dips” are very effective in removing tarnish. They
generally consist of thiourea and an unspecified acid (often phosphoric or
sulfuric acid), which may etch silver and leave a reactive surface that tends to
retarnish fairly quickly, particularly if it is not removed after treatment.
Alkaline washes may stain silver.
Ammonia should not be used to clean silver that is alloyed with copper.
Tarnish inhibitors, such as benzotriazole (BTA) and 1-phenyl-5-
mercaptotetrazole (PMTA), are not effective with silver that is alloyed with
copper.
Deionized or distilled water absorbs carbon dioxide from the atmosphere
and becomes acidic, thus becoming corrosive to lead-based solders that are
often used in the construction of instruments made of silver. Soft-soldered
silver instruments should thus be soaked or washed in tap water, as long as it
is reasonably free of impurities.
Pacific Silver Cloth contains colloidal silver, which is effective in
absorbing airborne sulfur compounds, thus protecting silver objects that are
wrapped in it.
See Oddy test for determining whether materials are safe for storing or
displaying silver (and other metal) objects.

Protective coatings

Nitrocellulose lacquer (such as Agateen no. 27 dissolved 1:1 with lacquer

thinner for spraying) or acrylic varnish (such as Paraloid B-48N dissolved in
xylene, 12.5% solids for spraying, 15% solids for brushing) can be applied to
delay the formation of tarnish.

References
C. Degrigny, M. Jerôme, and N. Lacoudre, “Surface Cleaning of Silvered
Brass Wind Instruments Belonging to the Sax Collection,” Corrosion
Australasia 18, no. 2. 1994, pp. 16–17.
L. Espie, N. Lacoudre, T. Beldjoudi, and J. Dugot, “L'electrochimie au
service du patrimoine musical,” Techne (2000), pp. 19–27.
Rupert Finegold and William Seitz, Silversmithing (Radnor, 1983).
P. Fiorentino, M. Marabelli, and A. Moles, “The Condition of the ‘Door of
Paradise’ by L. Ghiberti: Tests and Proposals for Cleaning,” Studies in
Conservation 27/4 (November, 1982), pp. 145–153.
Donny L. Hamilton, Methods of Conserving Archaeological Materials from
Underwater Sites (www.nautarch.tamu.edu, 2010).
Tim McCreight, The Complete Metalsmith (Worcester, 1991).
Harold O'Conner, The Jeweler's Bench Reference (Taos, 1988).
R. M. Organ, “The Reclamation of the Wholly-Mineralized Silver in the Ur
Lyre,” Application of Science in Examination of Works of Art (Boston,
1965), pp. 126–144.
Colin Pearson, Conservation of Marine Archaeological Objects (London,
1987).
H. J. Plenderleith and A. E. A. Werner, The Conservation of Antiquities and
Works of Art (Oxford, 1971).
Jiří Šrámek, Toe. B. Jakobsen, and Jiří B. Pelikán, “Corrosion and
Conservation of a Silver Visceral Vessel from the Beginning of the
Seventeenth Century,” Studies in Conservation 23/3 (August, 1978), pp.
114–117.
T. Stambolov, The Corrosion and Conservation of Metallic Antiquities and
Works of Art (Amsterdam, no date).
Aive Villus and Mart Viljus, “The Conservation of Early Post-Medieval
Period Coins Found in Estonia,” Journal of Conservation and Museum
Studies 10/2 (2013), pp. 30–44.
Soldering

Soft soldering

This type of soldering is often used in joining the bouts, yards, and stays of
brass instruments. Because soft soldering is done at around 400°F (200°C),
areas that have been previously subject to silver, or “hard,” soldering (which
is carried out at much higher temperatures) will not be affected.
For soldering brass instruments, so-called “60/40” solder (60% tin, 40%
lead) is often used. A soldering iron may be sufficient for small joints or
repairs, but a propane or air/acetylene torch (such as the Prest-O-Lite torch)
may be required to heat large areas evenly. Prior to fluxing, the pieces to be
soldered must be completely free of oxidation, oil, grease, or wax. Liquid or
paste flux should only be applied where the solder should flow. “Killed” acid
flux (zinc chloride) may be used and can either be purchased or made by
placing pieces of zinc in hydrochloric acid until the ebullition of hydrogen
ceases. Rosin may also be used as a flux, and is less corrosive. Crushed rosin
can be mixed with petroleum jelly to make a paste.
Parts can be held in position with a soldering jig or with annealed iron
binding wire. If holding the work together with clamps, be careful that the
clamps do not insulate or draw heat from the work, as this may interfere with
the flow of solder, and do not clamp the pieces together too firmly, as there
will be no space for the solder to flow. If a soldering iron is used, the heated
tip must first be cleaned by rubbing against a block of sal ammoniac
(ammonium chloride), then “tinned” with solder, and the excess wiped off
with a rag. Apply solder only when the work is sufficiently heated, and use it
sparingly. Never apply solder at the source of heat (such as the tip of the
soldering iron or torch) but rather at the metal parts being joined. Molten
solder tends to flow towards the source of heat, so if the solder is applied at
the point most distant from the iron or torch tip and melts freely, one can be
reasonably certain that the solder has reached the entire joint. Excess soft
solder can be removed by reheating and wiping with a rag, wicking it off with
braided copper tape, or by using a suction bulb. Scrapers are used to clean off
solder after it has solidified, but care should be taken that the workpiece is
not scratched or damaged in the process.
 In addition to the 60/40 solder commonly used in brass instrument work,
there are other types of soft solders, including common plumber's solder,
which has a 50/50 tin-to-lead proportion. This solder passes through a pasty
stage between its melting point (361°F; 183°C) and flow point (421°F;
216°C), which is useful when parts must be repositioned before the solder has
completely solidified. The eutectic invariant point (the lowest melting point)
for a tin/lead alloy (61.9% tin, 38.1% lead) is 361°F (183°C), which is very
close to 60/40 solder. This alloy goes directly from solid to liquid at this
temperature and thus has no pasty stage. Some tables of solder melting points
refer to the solidus and liquidus temperatures for various proportions of lead
and tin. The solidus temperature is the point at which the alloy leaves its solid
state (its melting point), while the liquidus temperature is the point at which
the solder is in a liquid state (its flow point).

Approximate melting and flow points of soft solders

% tin % lead Melting point °F/°C Flow point °F/°C

95 5 361/183 460/238

90 10 361/183 415/213

70 30 361/183 379/193

63 37 361/183 361/183

60 40 361/183 370/188

50 50 361/183 421/216

40 60 361/183 477/247

30 70 361/183 495/257

20 80 361/183 536/280
 10 90 527/275 576/302

Hard soldering

Hard soldering, which includes silver and gold soldering, employs solders
consisting of various alloys of silver, gold, copper, and zinc. These alloys
melt at much higher temperatures than soft solders, and the metal being
soldered must be brought to red heat. For this, an air/acetylene torch (such as
a Presto-O-Lite unit) is generally sufficient for musical instrument work. For
hard soldering, adjust the torch to produce a neutral flame (blue with a slight
amount of yellow), rather than a reducing (yellow) or oxidizing flame
(intense blue accompanied by a rushing noise), which tends to leave oxides
and firescale on the surface of the workpiece. As with soft soldering, the
work must be clean of oil, grease, wax, and oxidation. A paste of borax
(sodium tetraborate) is generally used as a flux. Jewelers traditionally use
cone borax (i.e. borax cast in the shape of a cone), which is rubbed with
water on a hollowed slate to form a paste, and then brushed onto the parts
that are to be joined.
Solder will only flow where there is flux. A paste of yellow-ochre pigment
and water can be painted on areas where one does not want solder to flow.
Use as little solder as possible when soldering; a well-made joint should
show only the smallest fillet of solder where parts have been joined. Silver
solder can be purchased in low-, medium-, and high-melting-point alloys
(termed “extra-easy,” “easy,” “medium,” and “hard”; see table below). The
higher the melting point, the stronger the solder joint, but the true utility of
these different grades is that they permit pieces to be soldered in sequence
without loosening previously soldered work. Silver solder is sold in wire
form and sheets. Jeweler's suppliers also sell silver solder in small squares
termed paillons that are about 1/16 in. (1.5 mm) square. These small squares
of solder are positioned around the area to be joined and are held in place by
the flux as it is heated and the water is driven off. Small bits of solder can
also be melted on a block of charcoal and allowed to harden; they can then be
picked up on a steel point (termed a “solder pick”) that has been heated red-
hot, and the solder deposited on the heated work. If wire solder is used, do
not apply the torch directly to the solder, but rather touch the solder to the
heated work. As with soft soldering, hard solder tends to flow toward the
source of heat or towards the hottest area; therefore, if a small piece is to be
joined to a larger one, it is best to apply heat to the larger part. If too much
heat is applied to the work, the solder may fail to flow and form into little
spheres.
Brazing is done at a higher temperature than most hard soldering and
employs spelter rather than solder. Spelter is an alloy of brass that contains
about 15% extra zinc to reduce its melting point below that of ordinary brass.
Brazing has been used in the manufacture of brass instrument tubing and
bells since the late nineteenth century. It can be employed in repair work, but
its melting point is very close to that of brass so great care must be exercised
to prevent the workpiece from melting.
When heating copper alloy metals (which include brass, and bronze, as
well as sterling silver and coin silver) to red heat for hard soldering or
brazing, the formation of firescale can be prevented by dipping or spraying
the workpiece with a solution of boric acid in methanol (300 g boric acid / 1
liter methanol) or a combination of borax, trisodium phosphate, and boric
acid in water (80 g borax / 80 g trisodium phosphate / 120 g boric acid /1 liter
of water) and allowing it to dry before applying heat.

Approximate melting points (MP) and flow points (FP) of hard solders
and spelter

Silver solders MP °F/°C FP °F/°C

Extra-easy 1145/618 1207/653

Easy 1240/671 1325/718

Medium 1275/691 1360/738

Hard 1365/741 1450/788

 Gold solders MP °F/°C

8 kt. yellow, easy 1275/691

10 kt. yellow, easy 1390/754

10 kt. yellow, hard 1415/768

12 kt. yellow, hard 1485/807

14 kt. yellow, easy 1390/754

10 kt. white, easy 1350/732

12 kt. white, hard 1440/782

14 kt. white, easy 1375/746

Brazing spelters MP °F/°C Strength

Copper Zinc

50 50 1607/875 for brass, strong

33 66 1508/820 for brass, weaker

66 33 1706/930 for brazing iron

Pickling

After hard soldering and brazing, the work may be covered with oxides and
dried flux. These oxides and deposits can be removed either by quenching the
work in a pickling solution while it is hot (which anneals metals such as
silver, copper, and some alloys of brass) or after it has cooled (to preserve
hardness). A standard pickling solution for silver, nickel silver, copper, and
brass can be made by adding 1 part sulfuric acid to 9 parts water; gold is
pickled in 1 part nitric acid to 19 parts water. When placing any metal alloyed
with copper in pickling solutions, do not use iron tongs, otherwise copper
may plate out on the surface of the workpiece; as a further precaution, if iron
binding wire has been used in the soldering process, make sure that it has
been removed before dropping the work into the pickling solution. Pickles
may be used at room temperature, but they are more effective when heated to
about 180°F (80°C).
Excess solder can be scraped off, though it can also be removed from silver
and gold by immersing the object in the following:

3 parts glacial acetic acid

1 part hydrogen peroxide, 6% (20 vol.).

For soldering of organ pipes see Organ restoration)

References
W. M. Haynes, ed., Handbook of Chemistry and Physics (Boca Raton,
London, New York, 2010).
Herbert Maryon, Metalwork and Enamelling (New York, 1971).
Timothy McCreight, The Complete Metalsmith (Worcester, 1991).
Eric Oberg, Franklin D. Jones, Holbrook L. Horton, and Henry H. Ryffel,
Machinery's Handbook (New York, 2012).
Harold O'Conner, The Jeweler's Bench Reference (Taos, 1988).
www.smarttec.de

Solvents and solvent cleaning

Thorough cleaning should always precede the revival, retouching, and

polishing of varnished surfaces. The solvents used in cleaning works of art
fall into the following categories: linear alkanes (i.e. straight-chain molecules,
also called “paraffinic hydrocarbons”; examples are hexane and heptane);
branched-chain alkanes (also called “isoparaffins”; examples are isohexane
and isoheptane); cycloalkanes (also called “naphthenes,” ring-shaped
molecular structures having only single bonds between carbon atoms;
examples include cyclohexane and cycloeheptane); aromatic hydrocarbons
(unsaturated ring-shaped molecules having alternating single and double
bonds between carbon atoms; examples include benzene, toluene, and
xylene); halogenated hydrocarbons (examples include carbon tetrachloride
and chloroform); alcohols (examples include methyl, ethyl and propyl
alcohol); esters (such as ethyl acetate); ketones (such as acetone, methyl ethyl
ketone); ethers (such as diethyl ether; butyl ether); glycol ethers (such as
cellosolve and butyl cellosolve); amides (such as dimethylformamide); and
amines (such as pyridine, morpholine, and triethanolamine). So-called
“mineral spirits,” “white spirits,” “V.M. & P. naphtha” (varnish makers' and
painters' naphtha), and “Stoddard solvent” are mixtures of aliphatic, aromatic,
and other hydrocarbons. Other solvents and cleaning agents include water,
turpentine, ammonia, soaps, anionic and cationic detergents, nonionic
surfactants, acids, bleaches, chelating agents, and enzymes.
There is a general, though somewhat mistaken, impression that certain
solvents are strong and others are weak. In fact, a particular solvent may be
effective in removing one type of material and completely ineffective in
removing another. Acetone, for example, is generally considered a strong
solvent, yet it has negligible effect upon some water-soluble materials, such
as animal-hide glue. Nevertheless, various systems have been devised for
characterizing solvent strength. One such system, termed the Hildebrand
solubility parameter (δ), ranks solvents according to the square root of their
cohesive energy densities, which allows them to be matched with substances,
such as resins, waxes, and oils. Solubility is also a product of three types of
intermolecular forces: dispersion force, dipolar force, and hydrogen-bonding
force. These forces are present in various degrees in most solvents, and if
they are considered individually, it is often possible to identify a solvent that
will work specifically upon the material one needs to dissolve. A clever
method of representing the three types of force exhibited by individual
solvents is the so-called “Teas graph,” which employs three axes arranged in
an equilateral triangle (see Horie, 1987). The solubilities of individual
solvents can thus be plotted on a Teas graph. Similarly, the solubility of
substances that one might wish to remove from an object can be represented
as regions (termed “solubility windows”) on a Teas graph. Any solvent
located within the substance's solubility window can be used to dissolve it.
The characteristics of solvents used in combination can be ascertained by
computing weighted averages of each type of force; these values can then be
plotted on a Teas graph to determine the effectiveness of that combination of
solvents.

Solvent action

Solvent action is an extremely complex process, involving the diffusion of

the solvent into the solute, swelling, and finally a molecular breakdown of the
solute and its distribution within the solvent. Unless the dissolved solute is
physically removed, the solvent will evaporate and the solute will, in many
cases, re-form as a coherent film. (In some cases reactions occur, and the
solute will not be reconstituted in its original chemical form.) Some of the
problems that may be encountered in solvent cleaning include the swelling of
underlying varnish layers and their retention of solvents. Solvents such as
ethanol and xylene have lower swelling coefficients than acetone and the
chlorinated solvents, while methanol and acetone are less likely to be retained
in varnish films than xylene and cyclohexane.

Selecting a solvent

Volatility is an important characteristic to consider when selecting a

solvent. Depending upon the circumstances, a high rate of volatility can be
either an advantage or a disadvantage: if an aggressive solvent evaporates
very quickly, it may vanish before it has an opportunity to damage an
underlying layer; on the other hand, a highly volatile solvent may evaporate
before it has had time to do its work. Solvents that are slow to evaporate do
not have to be reapplied as often, and this can be an advantage in situations
where the mechanical process of wiping on and removing the solvent may be
injurious. A disadvantage of slow-evaporating solvents is that they stand a
greater chance of being absorbed and retained by underlying layers, and
serious problems may result from this. In general, viscous and branched-
chain solvents will have more difficulty in extricating themselves from a
varnish layer than low-viscosity and straight-chain solvents. Solvents with
moderate to slow evaporation rates may remain in a varnish film for weeks or
longer, and it generally takes considerably more time for a solvent to
evaporate from a varnish layer than it does for it to become absorbed. Diethyl
ether is the quickest-evaporating solvent, so the evaporation rates of other
solvents are compared to it: diethyl ether is rated 1; acetone 1.9; toluene 6.1;
ethanol 7.0; V.M. & P. naphtha 7.1; xylene 9.2; methyl cellosolve 21.1; and
diacetone alcohol 60. Within each broad category of solvent there is a range
of volatilities. For example, if the solubility characteristics of an alcohol are
needed, butyl and propyl alcohol can be selected if a lower rate of
evaporation is required, while ethanol or methanol can be used if a faster rate
is needed (their respective evaporation rates are 19.6, 7.7, 7.0, and 5.2). Of
the commonly used aromatic hydrocarbons (xylene and toluene), toluene
evaporates more quickly than xylene.

Petroleum distillates

Petroleum hydrocarbons include a vast array of solvents having a wide

range of boiling points, vapor pressures, specific gravities, flashpoints, and
other characteristics. Most important, the aliphatic/aromatic proportions are
rarely given on hardware-store varieties of mineral spirits, so it is advisable to
purchase solvents directly from suppliers that deal directly with petroleum
refiners (such as British Petroleum, Exxon-Mobil and Shell) or from
chemical supply houses that can provide specifications. The Shell company,
for example, produces a vast array of petroleum-based solvents having
different compositions and properties, which include the Shellsol “D” series
(“de-aromatized” or high-aliphatic content having less than 1% aromatics);
the Shellsol “A” series (above 99% aromatic content), and the Shellsol “T”
series (isoparaffinic), among others. Stoddard solvent is a grade of mineral
spirits commonly used in dry cleaning that has an aromatic content that is
variously reported as 7% to 25%. “British white spirit” contains about 16–
20% aromatics. Mineral spirits of the isoparaffinic variety tend to have little
odor, and like the aliphatic solvents, relatively low solvent power. Aromatic
or high-aromatic solvents tend to have greater solvent power and may remove
or dissolve certain components of original varnish. One should keep in mind
that most commercial varieties of mineral spirits, sometimes termed white
spirits or turpentine substitutes, are not stable products, and the aromatic
components may evaporate more quickly than the aliphatic ones; therefore
the characteristics of the solvent may change over time. In addition to
selecting a specific grade of mineral spirits that has the desired aromatic
content, one can mix aliphatic solvents with aromatic ones and adjust the
percentage to suit one's needs.
Apart from their utility in removing waxes, oils, grease, and the residue of
some adhesive tapes, aliphatic solvents are useful in another very important
way: they can be used to dilute or restrain more active solvents. By adjusting
the proportion of the active solvent to the restrainer, precise control can be
exercised over solvent action. Sometimes, the active solvent can be applied
directly, allowed to swell the material to be dissolved, and then swabbed off
with a restrainer. When aliphatic solvents or low-aromatic-content mineral
spirits are used to restrain a more active solvent or combination of solvents, it
is advisable to select a grade that has a similar rate of evaporation to the more
active ingredient or ingredients, or a slower rate, otherwise the cleaning
system will become increasingly aggressive as the restrainer evaporates. At
certain concentrations, some combinations of solvents form an azeotrope,
which is defined as a combination of liquids that boil or evaporate without
change in concentration of either of the components. An azeotropic solvent
mixture thus retains its solubility characteristics as it evaporates. Examples of
azeotropes include: acetone 59% / n-hexane 41%; ethanol 49% / heptane
51%; ethanol 96% / water 4%; and 2-propanol 87.7% / water 12.3% (by
weight). A thorough list of azeotropes can be found in the Handbook of
Chemistry and Physics (2010).

Alcohols

Several types of alcohol are commonly used in cleaning: methyl, ethyl,

propyl, and butyl, listed here in order of molecular size and weight (methyl
alcohol, otherwise known as methanol, being the smallest and lightest) as
well as by volatility and polarity (methanol being the most volatile and
polar). Methanol, sometimes called wood alcohol, is poisonous and has an
unpleasant odor, so it is best avoided in favor of ethanol. In the United States,
however, ethanol (potable grain alchohol) is heavily taxed and in many parts
of the country unobtainable by the general public in concentrations of 95%
and up. The alcohol solvent that one finds in hardware stores and scientific
supply houses is ethanol that has been “denatured” (made poisonous) by the
addition of about 10% methyl and/or propyl alcohol, and, in certain grades,
with aviation gasoline. In some countries, denatured alcohol is tinted purple,
and thus should not be used because of the possibility of staining. The
rubbing alcohol sold in drugstores is isopropyl alcohol, otherwise known as
2-propanol, but it is often diluted with water. In its pure form, 2-propanol has
nearly the same solvent parameters as ethanol, but it evaporates more slowly
and is therefore preferred in some cleaning regimens. Alcohols are
hygroscopic and readily absorb moisture from the atmosphere (which reduces
their strength and solvent action). Gelatin (in the form of sheets or capsules)
can be placed in open bottles to absorb water that has been taken up from the
atmosphere. In the eighteenth century, craftsmen “powder-proofed” their
alcohol to determine how well it was distilled (that is, how free it was from
water). The powder-proof test involved wetting black powder with the
alcohol and then attempting to ignite the mixture. The author has ascertained
that 80 proof (40%) alcohol passes the powder-proof test, which suggests that
the solvent strength of alcohol might have been considerably lower than it is
today, though resins such as shellac will not dissolve in alcohol of that
strength. This has ramifications with regard to the formulation of alcohol
varnishes and the development of the technique of French polishing. For non-
aqueous cleaning, the strength of alcohol is generally reduced by mixing with
mineral spirits.

Water

Water is sometimes referred to as the “universal solvent.” Deionized or

distilled water should be used for cleaning if the local tap water is hard or
contains rust or other impurities. Deionized water is made by passing tap
water through a mixture of ion-exchange resins that removes impurities of an
ionic nature (such as salts). Water that is deionized is often pre-filtered to
remove large particles, passed through activated charcoal, and may be treated
by reverse osmosis. After deionization, it may be filtered again to remove
smaller particulate matter (including small grains of the ion-exchange resins
and charcoal used in treating the water), and other suspended materials
generally in the order of 1 to 10 microns. Distilled water is made by boiling
water and condensing the steam, while dissolved matter is left behind in the
retort. Water can be double- or triple-distilled to increase its purity.
 Because of their purity, deionized and distilled waters readily absorb
carbon dioxide and other materials from the atmosphere and from their
containers, which can cause them to become acidic or otherwise reactive. For
this reason, purified water should be stored in glass containers rather than in
plastic ones, which may be permeable to atmospheric gases and contain
extractable chemicals. Ultrapurified water may even leach out contaminants
from borosilicate laboratory glassware. Distilled and deionized water should
be checked with a pH meter (it should read pH 7; absorbed carbon dioxide
will lower the pH appreciably) as well as with a conductivity meter to
determine the presence of impurities (see Water).
The addition of a nonionic surfactant (such as Triton X-100) to water
lowers its surface tension and assists in cleaning operations. After cleaning,
residual surfactants must be cleared (for example, Triton X-100 should be
cleared with distilled or deionized water followed by mineral spirits of
moderate aromatic content). When deciding whether to use water to clean an
instrument, keep in mind that certain makers applied an animal-glue size
under their varnish. If water penetrates the varnish film, the glue may soften,
and the varnish will wash off. Therefore, before an instrument is cleaned with
water, it is a good idea to examine the juncture of the ribs and the belly. If the
varnish shows signs of having been washed off from previous regluing
operations, this may indicate that the instrument was sized with glue and
could be susceptible to further damage by water. Furthermore, wood surfaces
(such as those of violins) are often retouched with watercolor, which may
wash off with water, even if the retouched areas have been overcoated with
varnish or French polish.
Spit cleaning, or using saliva to remove dirt, is an old technique that is still
employed to a limited extent, even in museums. Saliva consists primarily of
water with various proteins (chiefly albumin and mucin), traces of most of
the common amino acids, a significant amount of sodium ions with traces of
other types of ions, and enzymes. The pH of saliva varies from about 6.8 to
7.4. Saliva also contains other materials, such as bacteria and bits and
remnants of one's last meal. For this reason, spit cleaning is not
recommended. Synthetic saliva is available through chemical supply houses,
though some formulations employ an artificial thickener (methyl cellulose)
which must be cleared after treatment. A synthetic saliva can be made by
dissolving 150 mg of mucin and 150 mg of the triammonium salt of citric
acid in 100 ml of deionized water, and used at a temperature of 37°C
(Bellucci, Cremonesi, and Pignagnoli, 1999). The chief attraction of saliva is
its enzymatic action; however, the enzymes present in saliva are available
individually and in purified form. Enzymes are most effective at specific pHs
(which can be maintained with buffers such as Trizma) and at certain
temperatures (lipase, for example, achieves peak activity at around a pH of
8.4 and at a temperature of 45°C). Some useful enzymes include lipase
(which breaks down oils, grease, and fats), pepsin and protease (which break
down proteins, such as egg white and animal-hide glue), and amylase (which
breaks down starches, such as flour-based pastes).

Gelled and emulsified solvents

Recently, there has been much discussion regarding the use of solvents in
gel or emulsion form (Wolbers 2000). Most people are familiar with paint
strippers such as Zip-Strip (Star Bronze Co.), which contains methylene
chloride, ethyl and methyl alcohols, and mineral spirits suspended in a gelling
agent. Gelled or emulsified solvent systems offer certain advantages: the
solvent can be precisely positioned and held in place, even on vertical
surfaces or overhead; evaporation of the solvent is slowed; and absorption of
the solvent is reduced and thus can be more precisely controlled. However, it
is vital that the gelling or emulsifying agent be completely removed after
cleaning, and this may not be as simple as wiping it away with the dissolved
material – it generally involves repeated applications of one or more solvents.
Because gelling agents may be absorbed by the object being cleaned, it may
not be possible to remove them completely (Dorge, 2004). For this reason,
they are not recommended.

Testing solvents prior to use

When cleaning with any solvent, it is best to test it on an inconspicuous

area (under the fingerboard or tailpiece of a violin, or the underside of a piano
lid, for example). A small bit of cotton wool can be wrapped around the end
of a sliver of wood (making, in effect, a miniature cotton swab) and lightly
moistened with the solvent being tested. If the varnish is not affected, the
cleaning solvent may then be applied with larger swabs, which are best rolled
across (as opposed to rubbed against) the surface of the object, or with wads
of cotton wool held with the fingertips. After each pass, the swabs or wads
should be examined for color. If they become black or grayish, dirt is being
removed; if they pick up pale yellow or the color of the varnish, then one has
reached beyond surface dirt to French polish or to an underlying finish.
To remove a nonoriginal varnish or polish from an instrument, one might
start with Shellsol D (a high aliphatic-content solvent), or, if that is not
effective, a high-aromatic-content solvent such as Shellsol A. One can also
combine pure aromatics such as toluene or xylene with an aliphatic solvent. If
100% toluene or xylene is ineffective, then one can continue by adding 10%
ethanol to mineral spirits of moderate aromaticity and gradually increasing
the proportion of alcohol if necessary. If that is ineffective, acetone can be
substituted for the alcohol, or combined with it, again starting with a
concentration of about 10%.

References
Roberto Bellucci, Paolo Cremonesi, and Ginevra Pignagnoli, “A Preliminary
Note on the Use of Enzymes in Conservation,” Studies in Conservation 44
(1999), pp. 278–281.
Pia C. DeSantis, “Some Observations on the Use of Enzymes in Paper
Conservation,” Journal of the American Institute for Conservation vol. 23,
no. 1 (Fall, 1983), pp. 7–27.
Valerie Dorge, ed., Solvent Gels for the Cleaning of Works of Art (Los
Angeles, 2004).
Robert L. Feller, Nathan Stolow, and Elizabeth H. Jones, On Picture
Varnishes and Their Solvents (Washington, DC, 1985).
W. M. Haynes, editor, Handbook of Chemistry and Physics (Boca Raton,
2010).
C. V. Horie, Materials for Conservation (London, 1987).
Knut Nicolaus, The Restoration of Paintings (Cologne, 1998).
Helmut Ruhemann, The Cleaning of Paintings (New York, 1982).
Helen Wilkes, ed., Science for Conservators, 3 vols. (London, 1987).
Richard Wolbers, Cleaning Painted Surfaces (London, 2000).

Specific gravity

Specific gravity (s.g.) is the ratio of the density of a substance to the

density of a reference substance, generally water, which has a specific gravity
of 1 (or 1000 in the UK). Specific gravity for most liquids is ascertained at a
pressure of 1 atmosphere (101.325 kPa) and at a temperature of 20°C.
Specific gravity is often specified in solutions (such as those of salt, sugar,
and ammonia), as well as acids.
The Baumé scales were developed in France by Antoine Baumé in 1768,
and are still encountered, though they were largely replaced by specific
gravity. One scale was intended for liquids denser than water, the other for
liquids less dense than water. They were notated in degrees followed by the
abbreviation for Baumé (B, Be, or Bé). To convert from Baumé to specific
gravity, use one of the two following formulas:

At 20°C temperature

For liquids denser than water, specific gravity (s.g.) = 145/(145 – degrees
Baumé)
For liquids less dense than water, specific gravity (s.g.) = 140/(degrees
Baumé + 130)
A common concentration of ammonia in water of 28–29% (“Spirit of
Hartshorn”), 26° Bé, has a s.g. of 0.897.
To determine specific gravity see under Organ restoration.

Stain removal

(see-also Bleach, Paper, pencil, and ink, Solvents and solvent cleaning,
Textile cleaning, Wood)
Light stains in paper, wood, and textiles can often be removed with 1.5–3%
hydrogen peroxide (see Bleach).
 Copper corrosion stains (green) can be removed with 1–5% ammonium
hydroxide.
Iron stains (gray, black, or brown) can be removed with 5% disodium
EDTA, 2–5% ammonium citrate, or 5% oxalic or citric acid.
Grease (including airborne cooking grease that has settled on objects) can
be removed with dilute ammonia (note that many dyes, especially those
derived from plants, and even the color of shellac, can be dramatically
affected by changes in pH caused by the application of ammonia). Mineral
spirits are actually more effective in removing oil and grease than
ammoniated solutions and alkalis such as soaps.
Oil, fat, tar: mineral spirits; pyridine
Wax and candle residue: mineral spirits
Fly stains: 20% hydrogen peroxide and ethanol, 1:1, or 2% Chloramine-T
(see Bleach)
Tea and coffee stains: 2% potassium perborate or sodium percarbonate;
wash thoroughly after application.
Ink stains: 2% Chloramine-T, 5% oxalic acid, or 2–5% ammonium citrate;
wash thoroughly after application.
Fresh blood stains: 3% hydrogen peroxide.
Enzyme digestion of protein stains: 1 gram pepsin in 25 ml warm
deionized water and 2 drops of hydrochloric acid. Keep moist; may take
several hours. Clear with deionized water.

Staining wood

(see-also Patination)

Darkening boxwood

Nitric acid and deionized water 1:1 (note: always add acid to water); add
two iron nails (this will evolve nitrous oxide which is poisonous, so do this
under a fume hood or outdoors). Apply warm. After the wood has achieved
the desired color, apply warm linseed oil to neutralize residual acid.

Darkening mahogany

Solution of sodium hydroxide.

One traditional method of making mahogany redder or making European
walnut resemble mahogany is to steep alkanet root in linseed oil and apply
the colored oil to the bare wood. Though fugitive, this was often used under
varnish in early nineteenth-century Viennese pianos.

Darkening oak and brazil wood

Ammonia vapors.

Black stain

First apply solution of logwood or tannic acid; then apply solution of

ferrous sulfate. This is also a good stain for darkening ebony.

To simulate oxidation on light-colored woods

Dissolve walnut hulls, coffee, tea, or chicory in water; apply with brush or
rag.

Stradivari's varnish

The findings below are a summary of analysis carried out by the author in
2008 in collaboration with the Objects Conservation Department of the
Metropolitan Museum of Art and the McCrone Laboratory.

1708 violin

Fourier transform infrared (FTIR) spectroscopy detected components of an

oil–resin varnish as well as protein.
 Pyrolysis-gas chromatography/mass spectroscopy (Py-GC/MS) detected
linseed oil, pine colophony, and benzoin.
Raman microscopy detected an unidentified organic red pigment.
Traces of beeswax and shellac were also detected.

1710 violin

FTIR detected components of an oil-resin varnish as well as protein.

Py-GC/MS detected linseed oil, pine colophony, and benzoin.
Raman microscopy detected red ochre pigment.
Traces of beeswax and shellac were also detected.

1736 “Paganini” cello

FTIR detected shellac.

GC/MS detected a drying oil.
No pigments were observed.
Traces of talc or other silicate were detected.

1730 “Pawle” cello

Two distinct layers were detected: in the top layer there were particles of a
black pigment, possibly crushed charcoal or soot; in the bottom layer there
were light brown particles and a few orange–red pigments.
FTIR detected shellac, protein, talc or other silicate, gum ammoniac,
charcoal or soot.
GC/MS detected a drying oil and pine colophony.

1734 violin

Five distinct layers were observed, but they were too thin to distinguish the
components of the individual layers.
FTIR detected calcium carbonate, protein, a drying oil, pine colophony,
and possibly shellac.
GC/MS detected a drying oil and pine colophony.

Stringed-keyboard restoration

(see-also Clavichord maintenance, Disinfection, disinfestation, and

fumigation, Fortepiano maintenance, Harpsichord maintenance, Piano
action regulation and voicing (modern grand), Tuning and temperament,
Wire gauge for early keyboard instruments)
Stringed-keyboard instruments are prone to deterioration because of the
high string tension that they bear (which can range from several thousand
pounds in early wood-framed instruments to twenty tons in modern concert
grand pianos). Fluctuations in temperature and humidity can play havoc with
delicately adjusted mechanical parts and cause swelling, buckling, and the
formation of cracks in casework and soundboards, as well as the gradual
weakening of glue joints. Woodworm and other forms of natural
deterioration, such as rust and corrosion of strings and other metal parts, as
well as mechanical wear from playing, and damage from handling and
transport, also take their toll. To this we must add perhaps an even greater
cause of deterioration: uninformed or inexpert repair and restoration.
If one reviews the restoration work undertaken in museums and in the
private sector since the revival of interest in the harpsichord, clavichord, and
fortepiano (which began in the late nineteenth century) one finds that many
important examples have been severely compromised by repairmen and
restorers, who in many cases removed and replaced original soundboards,
bridges, nuts, wrestplanks, case bottoms, registers, jacks, keyboards, strings,
plectra, key and action cloths, damper materials, and tuning pins in their
efforts to make instruments playable. In other instances, original parts have
been subjected to inappropriate interventions, such as the over-varnishing of
soundboards and the shellacking of painted casework. The carnage has been
so widespread that there has recently been a call for a moratorium on early
keyboard instrument restoration.
The decision to restore

In deciding whether an early keyboard instrument should be restored to

playability, the individual owner, collector, or institution should consult with
scholars and builders who are thoroughly versed in early keyboard instrument
history, and specifically with the instrument's maker or school. The purpose
of this consultation would be to ascertain the instrument's historical
significance, its original character and features, and the nature and sequence
of previous restorations or alterations. In addition, the instrument should be
assessed with regard to its physical condition and ability to withstand
renewed string tension and the wear and tear of use. This may require the use
of X-rays and endoscopy to discover how the case is constructed, the extent
of structural disrepair, and whether there is any woodworm damage.
Analytical tests may be carried out to identify the materials employed in the
instrument's construction, such as the types of wire, woods, glues, varnishes,
paint media, pigments, cloths, and leathers. Measurements should be made of
case and action parts, string lengths, plucking and striking points, and string
diameters, as well as documentation of string gauge markings, plucking
direction and order, and the disposition of registers, stop actions, knee-levers,
pedals, etc. This preliminary study may require partial disassembly to
discover points of weakness and problems that might arise during restoration
and subsequent use.
Once this preliminary work has been completed and the decision is made to
restore the instrument, other choices may have to be considered, such as
whether the instrument should be returned to its original state or whether it is
best restored to some later or intermediate state. For example, a seventeenth-
century Antwerp-school harpsichord originally made with transposing
keyboards that went through a grand ravalement in eighteenth-century Paris
might best be retained in this latter state rather than any attempt being made
to return it to its original configuration. One reason for this approach might
be to make use of the nonoriginal though well-crafted eighteenth-century
keyboards, registers, jacks, bridges, and nuts, rather than replacing them with
“copies” of the long-lost original parts. Similarly, it might be unwise to
remove fine eighteenth-century case painting and gilding in order to reveal a
crude or heavily damaged coat of original paint underneath. In some cases it
may not be so easy to make a decision. For example, sixteenth- and
seventeenth-century Italian harpsichords invariably show evidence of having
had their compasses enlarged and their bridges and nuts repositioned, either
to restore the original string lengths when their keyboards were shifted during
a compass enlargement, or to rescale them for higher or lower pitch. Very
often, there is evidence that bridges and nuts have been moved more than
once, which makes reconstruction of the original string lengths problematic,
and if the nuts and bridges have been moved back to what was assumed to be
their original position in an undocumented restoration, the vestigial evidence
of other nut and bridge positions may confuse matters.
Keyboard restoration work (like any other instrument restoration work)
should only be undertaken by those who have been formally trained as
makers, as they are intimately acquainted with the design and construction
processes and generally possess the requisite skills to undertake repairs and
reconstruct missing parts. Furthermore, they are in the best position to
identify novel approaches taken by the original maker, which should be
preserved as restoration proceeds. Keyboard instrument makers will also
stock many of the materials needed in the reconstruction process, have the
facilities for dimensioning lumber, repair and assembly of casework and
soundboards, and have at hand many of the purpose-built jigs and tools for
fabricating replacement parts. One of the problems encountered by restorers
of early keyboard instruments is the fact that wire, cloth, leather, and various
other ephemeral materials are not manufactured as they were hundreds of
years ago. Any restorer who has restrung an early nineteenth-century
fortepiano with any of the currently available re-creations of early iron wire,
and replaced original hammer leathers and action cloths with modern
materials would be less than honest if he or she claimed to have successfully
restored the instrument to its original state.

Case structure

Before the introduction of metal struts, tubes, and cast frames, keyboard
instruments relied upon wood case structures to support their considerable
string tension. Generally speaking, there are several national styles of case
construction: the classic form of Italian harpsichord consists of a thin-walled
case of cypress wood with the sides built around a softwood bottom and
supported by triangular knees; the more heavily constructed
Flemish/Antwerp school and French harpsichords often have painted heavy-
walled cases made of solid limewood or softwood, with side-to-side
buttressing and a bottom that is nailed or trenailed beneath the case sides;
English construction is similar to that of the Flemish school but often features
fine walnut- or mahogany-veneered oak-core casework; north German
harpsichord makers tended to employ features of other northern European
makers, while south German and Austrian makers adopted some Italian traits;
Spanish and Portuguese makers sometimes employed north German and
Flemish case design, though others exhibit Italian influence. In addition to
certain “national” traits, there were also “city” styles, such as the Neapolitan
style of harpsichord construction featuring recessed case bottoms, and
Viennese fortepiano construction employing brickwork-style interlocking of
internal case buttresses. There are also some strong individual styles, such as
the double-bentside developed by Bartolomeo Cristofori in Florence. In
recent decades a number of museums have made full-scale drawings of
keyboard instruments in their collections. The Comité International des
Musées et Collections d'Instruments de Musique (CIMCIM) has produced an
online catalog of these drawings, most of which are available for purchase.
They are excellent resources for studying different styles of construction.
String tension is responsible for case distortion and ultimately the failure of
glued joints. This distortion is often dramatically evident in English
harpsichords and pianofortes, which invariably have severely cocked cheeks,
as well as in square pianos and clavichords, whose right front corners are
generally pulled upwards. In the case of English harpsichords and
pianofortes, one of the causes of this distortion is the fact that the bottom is in
two sections, the main part extending from tail to belly rail, and the front
section oriented cross-grain beneath the keyboard or keyboards and joining
the cheek and front part of the spine. Because the bottom is divided at the
belly rail, it acts rather like a hinge and thus fails to provide much resistance
against the pull of the strings. Dovetailed joinery generally prevents the case
joints from failing, but the resulting distortion creates problems with string
down-bearing, the function of keyboard and action, as well as stress,
distortion, and the formation of cracks in the soundboard. Ideally, instruments
that are in storage should have their string tension slackened to reduce stress
on their cases. To straighten out case distortion, string tension must be
lowered and the cases allowed to relax. Some restorers clamp cases against a
solid support for a length of time, though because of their inherent structural
weakness, they will eventually retwist once string tension is reapplied. Some
amount of case distortion must be expected in any wood-frame keyboard
instrument, though it should be monitored by measuring the height of the
lifted corner (generally the front right corner of wing-shaped harpsichords,
grand pianofortes, square pianofortes, and clavichords). There is no rule of
thumb regarding the amount of twist that is acceptable, though it should be
considered excessive if it creates insurmountable mechanical problems with
the keyboard and action and prevents reasonable downward pressure of the
strings against the nuts and bridges.
It may be necessary to gain access to the case interior in order to reglue
loose buttresses or soundboard barring (Figures 32a and 32b). One often
encounters rectangular openings that have been sawn in case bottoms to gain
access to the insides, but it is not advisable to make such openings as they
weaken the bottom, which was intended to serve as a structural part of the
case. Removing the bottom is relatively easy with English, French, and
Flemish/Antwerp school instruments. In fact, many of these bottoms were not
even glued on but held in place with trenails or nails, which are easy to locate
and withdraw. Removing the bottom is problematic if the case sides have
been built around it, as was done in most Italian harpsichords and many
German and Austrian fortepianos. In these instances, it may be easier to gain
access to the interior by removing the soundboard, but this too may be
difficult in some early German and Viennese fortepianos in which the
soundboard is overlaid by a massive hitchpin rail that is trenailed or nailed
onto the case structure. Later Viennese fortepianos often have soundboards
that do not extend beneath the hitchpin rail but rest on their own narrow liner,
making them relatively easy to lift out. Again, familiarity with the maker's
work or school is paramount when disassembling case structures. Consulting
technical drawings of similar instruments is instructive, though in some
instances it may be necessary to X-ray the entire case or employ an
endoscope or borescope to determine how it is constructed.
Figure 32a Interior of pianoforte by William Frecker, London, 1799. This
photograph was taken after loose soundboard ribs and case bracing were
reglued.

Figure 32b William Frecker pianoforte after restoration.

Traditionally, early keyboard instruments were glued together with animal-

hide or fish glues, though casein glue was occasionally used as well. Animal-
hide and fish glues remain soluble in water and can be softened through the
application of moisture and heat. Rolled-up cloths that are kept damp with
distilled or deionized water can be placed over glue joints to loosen old glue.
This can be supplemented with heat supplied by tacking irons or vulcanized
rubber mats or strips containing thermostatically controlled heating elements
(such as those manufactured by Watco). Steam-producing devices (such as
the Preservation Pencil and fabric steamers) are also useful for loosening
glue. Glue softens at around 140°F (60°C), though if heat is used to melt
glue, the temperature should be kept below the melting points of resins
commonly used in paint, varnishes, and French polish, which are generally
above 165°F (74°C). When loosening glued joints with moisture or steam, it
is important to protect materials or areas that might be adversely affected by
it, such as metal, paper, and cloth, as well as varnished and painted surfaces.
It is also necessary to exercise patience, as it may take hours or even days to
loosen joints. Old glue joints are often sufficiently deteriorated that parts can
be removed mechanically, such as by working a palette knife into a gap and
running it along the joint. Sometimes glued joints will yield to a sharp blow
delivered by the palm of one's hand. Anhydrous ethanol, when applied to
animal-hide glue joints, has a desiccating effect and will sometimes cause
them to give way – though this technique should not be used if varnish or
paint might be affected by alcohol.
When examining the interior construction of keyboard instruments, note
the direction and overlap of glue drips, as these may provide clues as to the
sequence and method of construction. Old animal-hide glue fluoresces pale
blue under long-wave ultraviolet, whereas freshly applied animal-hide glue
does not fluoresce.
Once the interior is exposed, all joints should be checked for soundness
and reglued if necessary. It may be necessary to add shims or wedges to
tighten up mortise-and-tenon and other joints that have loosened through
compression or shrinkage. When opening up old keyboard instruments, one
often discovers that they have been previously worked on: extra bracing may
have been added, and soundboards may have been repaired with studs, cleats,
straps, and other laminations glued to their under surfaces. One must decide
whether these reinforcements serve any critical function or can be removed.
Optimally, case structures and soundboards should be restored to their
original configurations.
In the past, case bottoms that had severe shrinkage cracks were often
replaced with new ones, frequently made of plywood (which became readily
available after World War One). This material was often used because it was
thought to be more stable and crack resistant, but also because it is easier to
saw out a new bottom from a sheet of plywood than to repair the original or
join up a new bottom of solid planks of wood. One problem with plywood
bottoms is that they do not have the same longitudinal stiffness as planks of
solid wood of the same thickness, so a case fitted with a plywood bottom may
be less stable and have a greater tendency to twist because of string tension.
Be aware that construction marks and scribe lines were often drafted on the
original bottom, and that these marks provide evidence regarding the design
and construction processes, so it is important to preserve this important part
of the case. If the bottom is removed to make repairs, shrinkage cracks can
either be compressed with clamps and glued or filled with shims. Sometimes
shrinkage has been so extreme that the bottom will become significantly
narrower after cracks have been compressed and require the addition of an
extra strip of wood along the spine edge. When reattaching the bottom, it is
advisable to plug the old trenail or nail holes and refit the trenails or nails
either in the plugged holes or at new positions; refreshed anchoring points
will help the bottom resist renewed string tension.
Before closing up an instrument, it is important to make a thorough
documentation of the interior, including photographs and measurements of
the dimensions and location of braces and soundboard barring. The
photographs should include a high-resolution plan-view made with the
camera properly centered and aligned to avoid distortion. While still
photographs offer the best clarity, it is a good idea to make a thorough “tour”
of the case interior with a video camera, as this may pick up details that one
might have missed with a still camera. If time is available, a full-size
technical drawing should be made.

Wrestplank

The wrestplank is often the stumbling block that renders an otherwise well-
preserved instrument unplayable. Tremendous force is placed not only on this
plank of wood, which tends to cause it to bow outward and upward, but also
upon the glue joints that hold it in place. Loose tuning pins can often be
tightened by inserting a strip of paper or a shaving of maple or beech in the
tuning pin hole and reinserting the pin. Cracks running from tuning pin hole
to tuning pin hole may render groups of pins unstable and the instrument
incapable of being tuned. To check for cracks (which may be concealed
beneath an upper lamination or veneer), one can X-ray the wrestplank or use
a borescope fitted with a right-angle mirror to examine each tuning pin hole.
In some instances, fine cracks may be successfully reinforced by sealing the
perimeters of the tuning pin holes with pieces of adhesive tape and filling the
cracks with epoxy. In more extreme cases (such as those with woodworm
damage, wide cracks or that are badly warped), it may be necessary to replace
the wrestplank in order to render the instrument tunable. If this is the case,
serious consideration should be given to preserving the instrument as a study
piece rather than restoring it to playability, as the replacement of this major
structure may result in alterations to the scaling, plucking and striking points,
and the relationships between various structural and mechanical parts. In fact,
many keyboard scholars are suspicious of measurements drawn from
instruments whose wrestplanks have been replaced, and as a consequence
may not include such data in their statistical workups. However, if the
decision is made to replace the wrestplank, any laminations or veneers, as
well as the nuts, should be transferred to the new wrestplank, for in addition
to disguising the new wood below, it will also serve as a positioning guide for
the nuts as well as for drilling tuning pin holes. Obviously, before an
instrument can be restrung and tuned, old or new wrestplanks must be firmly
glued into the instrument, and any laminations and reinforcements (such as
the saddle, gap spacers, etc.) must be securely reattached. The best glue for
this is hot animal-hide glue; modern carpenter glues (such as Titebond,
Elmer's, etc.) are inappropriate for this task because structural elements under
tension that are glued with them have a tendency to slide.
When drilling new holes for tuning pins, it is best to wait until the tuning
pins have been cleaned of corrosion products (see below) before the holes are
drilled and reamed to size. Generally speaking, the holes used to
accommodate the small-diameter tuning pins used in clavichords,
harpsichords and early fortepianos should be around 0.006 in. (0.15 mm)
undersize, while later (early-nineteenth century) fortepianos should be drilled
around 0.008 in. (0.20 mm) undersize. In comparison, modern pianos may be
drilled as much as 0.016 in. (0.40 mm) undersize in the treble and tenor and
about 0.021–0.025 in. (0.53–0.64 mm) in the bass. A torque wrench can be
used to test out various size holes and gripping strengths on scrap stock
before drilling and reaming the new wrestplank. The final diameter of the
holes should be achieved by reaming, which is more accurate than drilling
with conventional twist drill bits and provides cleaner holes. If reamers are
used, the hole should be drilled about six thousandths of an inch (.15 mm)
undersize to accommodate the reamer. If twist drill bits alone are used, make
sure they are sharp, and do not allow the bit to get overheated or clogged with
wood chips, as this may cause the holes to become scored or enlarged.
Drilling speed and feed rate will both affect the final diameter and the
tightness of pins. Reaming can be done by hand with the reamers chucked in
a T-handle tap wrench. While tight pins are important for tuning stability,
making the holes too tight will make tuning difficult (tuning pins may jump
rather than turn smoothly) and may lead to the formation of cracks. Changes
in relative humidity will affect the size of holes drilled in wood; as humidity
increases, holes will shrink, and as humidity decreases, holes will expand.
This should be kept in mind when wrestplanks are drilled in extreme
conditions.

Soundboard

Soundboards are generally constructed of thin planks of quarter-sawn

spruce, fir, or cypress (the latter in the case of Italian harpsichords) and
reinforced on the underside by ribs and in some instances a cutoff bar.
Harpsichords having 4-foot stops generally have a massive hitchpin rail glued
to the underside that holds the hitchpins for the 4-foot strings. The ribs and
cutoff bar serve two functions: they help maintain flatness and also affect
acoustics. Because the grain of ribs and cutoff bars generally runs across that
of the soundboard, there is a constant tug-of-war between these structures as
the humidity rises and falls, which may cause glue joints to separate, often
resulting in buzzing as the soundboard vibrates against the partially detached
ribs or cutoff bar; furthermore, because the soundboard is restrained along its
edges, it may crack during periods of dryness and bulge or buckle during
periods of high humidity. If the soundboard has a tendency to bulge upwards
as it expands, it may come in contact with the strings; if it sinks, there will be
reduced down-bearing of the strings on the bridge, and in some cases the
strings may lift off the bridge entirely (this was sometimes alleviated by
back-pinning the bridge). The soundboards of modern pianos are
intentionally given a slight upward arch (called a “crown”) during
manufacture; however this quickly diminishes due to the downward pressure
of the strings. Because the degree of arching is difficult to control during
manufacture and is essentially ephemeral, some piano makers (such as
Steinway) refuse to release data regarding the amount of arching that they
build into their soundboards, which would certainly be useful to know in
assessing the extent to which it may have subsided or in reestablishing the
crown during restoration. Some harpsichord and fortepiano historians and
makers assert that early makers also “crowned” their soundboards, though
this must be ascertained on a case-by-case basis. The author has observed
burn marks on the interior surfaces of bottoms fitted to several eighteenth-
century Italian harpsichords, which suggest that small fires were set to piles
of wood shavings before their soundboards were glued in. This may have
been done to dry out the soundboards so that when they reabsorbed moisture
from the air, they would expand slightly and create a crown. (Another
rationale for setting these fires would be to warm the wood in order to
prevent the glue from chilling.)
Case distortion brought about by string tension often results in the
compression of soundboards and the formation of severe bulges, cracks, and
the loosening of ribs. Modern piano rebuilders typically replace soundboards
that have cracked or lost their crown. The old soundboards are hammered out
from behind with a sledgehammer and then discarded (note that these scraps
make good stock for shims, cleats, and other parts). Restorers of historic
keyboard instruments should not replace original soundboards and must make
every effort to repair and retain them. A hairline soundboard crack can often
be stabilized by running animal-hide glue into the crack. It is best to do this
during periods of high humidity, when the cracks are narrowest; hopefully,
the glue will hold during periods of dryness. When gluing hairline cracks in
this way, one should use hide glue having a light color so that the glue line
does not show.
Larger soundboard cracks can often be repaired without removing the
soundboard by “shimming,” which involves gluing fitted strips of wood into
the cracks. Soundboard shims should be made of the same type of wood as
the soundboard and inserted in proper relationship with regard to year-ring
orientation (that is, the direction of early-growth to late growth should be the
same for both the soundboard and the installed shim). If the year-ring
orientation is not the same, the shim will reflect light differently than the
surrounding wood, making it virtually impossible to retouch the shim
effectively. It is best to shim cracks when the humidity is low and the cracks
are at their widest. If this work must be done in humid weather, the
instrument can be placed in a room with a dehumidifier to dry it out and thus
widen the cracks prior to shimming.
Before shims are fitted, the cracks should be evened up and given a
moderate taper with a special shimming chisel or by running a knife-shaped
file lengthwise through the crack (very narrow cracks can be tapered with a
similarly shaped jeweler's or escapement file). The shims should be hand-
planed to the same taper as the crack so that when they are pressed in they
make a tight fit. If there is no access to the underside of the soundboard to
trim the part of the shim that might extend below it, the lower part of the
shim should be planed off before it is glued in. To determine how much to
plane off, it is necessary to determine the thickness of the soundboard along
the length of the crack. To do this, pass an L-shaped piece of wire through the
crack, hook it to the underside of the soundboard, and mark the soundboard's
thickness on the section of wire that projects above the crack (or if there is
access for inserting the magnet of a Hacklinger gauge [see Measurement], the
soundboard's thickness can be measured directly). The shim is then
temporarily pressed into the crack and marked where it extends above the
soundboard; the shim may then be removed and planed to the required depth
before gluing it in. After the glue has dried, the part of the shim that extends
above the soundboard is trimmed with a sharp knife or chisel. When leveling
shims, refrain from using sandpaper or scrapers, as they will cut through the
soundboard's patina and alter its thickness.
Before shimming, make sure that the underlying ribs are not loose. If they
are, use a syringe or flexible palette knife to work glue between the
soundboard and the rib or bar, and place weights on the soundboard to press
it against the rib. It may be possible to take advantage of cracks or openings
in the instrument's bottom boards to insert thin strips of wood or metal to
impose upward pressure on the ribs or bars while the glue sets.
If ribs, cutoff bars, and bridges are unglued to the point where they cannot
be effectively reattached by applying pressure from above or through the
bottom, it may be necessary to remove the soundboard to make the necessary
repairs. As with keyboard instruments that have had their wrestplanks
removed or replaced, historians are not terribly interested in instruments that
have had their soundboards removed for repair or replaced with new ones. If
it is deemed necessary to remove a soundboard, every effort must be made to
document its position and string lengths before proceeding. Before removing
a soundboard, markers can be placed along the inside edges of the case sides
to triangulate the positions of various bridge pins and other critical structures.
Positioning pins can also be inserted to assist in realigning the soundboard
when it is glued back in – though any pins or holes made to accommodate
them should be noted in the conservation report so that they are not later
confused with original construction marks.
Soundboard repairs are best carried out on a go-bar deck so that the ribs,
bars, and bridges can be firmly pressed against the soundboard while gluing
them in place. The go-bar deck is also useful in repairing cracks, because
they can be leveled perfectly along their entire length (Figure 33a–d). In
some cases the cracks can be compressed with clamps so that shims are not
required, which is a great advantage if cracks run through painted decoration.

Figure 33a Soundboard of a clavichord by J. C. Meerbach, Gotha, 1799,

prior to restoration.
Figure 33b Soundboard after restoration.

Figure 33c Underside of soundboard before restoration.

Figure 33d Underside of soundboard after restoration.

It is comparatively easy to reglue bridges and ribs in keyboard instruments

that have open bottoms. In regluing the bridge of an open-bottomed
fortepiano by Conrad Graf (Vienna, c.1838), the author turned the piano case
itself into a virtual go-bar deck by clamping beams of wood across the
bentside and spine and applying pressure from above and below by means of
veneer press screws mounted on the beams (Figures 34a and 34b). Use old
glue lines to mark the position of bridges, ribs, and bars, but clean off any old
crystallized glue that might interfere with regluing.
Figure 34a Loose bridge of a Conrad Graf fortepiano, Vienna, c.1838.
Figure 34b Clamping system used to reglue the bridge of the Conrad Graf
fortepiano using veneer press screws to apply pressure from above and below
the soundboard.

When retouching shims, pigmented media, such as watercolors, invariably

dry to a different gloss than the soundboard and thus remain visible. Shims
can sometimes be more effectively retouched with boiled-down and filtered
decoctions of old oxidized wood, walnut hulls, chicory, coffee, or tea. Shims
that interrupt figurative soundboard painting must be retouched with the
appropriate medium (such as watercolor, gouache, or egg tempera). Before
cleaning soundboards, gluing in shims, or making other repairs, soundboard
paintings should be tested to determine whether they are sensitive to water or
other solvents that may come in contact with them; for example, Flemish and
French soundboard painting was often done in a gum medium that is water
sensitive (Ann and Peter Mactaggart, 1977). If shims will run through
soundboard painting that is sensitive to water, it can be temporarily protected
with an application of a removable retouching varnish, such as Soluvar,
Paraloid B-72, or Regalrez (see Retouching). Another approach would be to
use a synthetic adhesive that employs an organic solvent as its vehicle (such
as Paraloid B-72 in acetone). In either case, it is imperative to ascertain the
painting's solvent sensitivity prior to their use.
To summarize, the removal of an instrument's wrestplank, bottom, or
soundboard is a serious matter that should be undertaken only if absolutely
necessary. Undoing glue joints invariably involves some splintering, loss of
wood, and possibly the marring of moldings, finishes, and decorated surfaces.
There is also the possibility of inadvertently changing critical dimensions and
relationships between parts, thereby reducing the value of the instrument
from a historical perspective. Opening up an instrument that is several
hundred years old can be a tremendously exciting experience for a restorer,
though one's eagerness to have a look inside should not influence the decision
to proceed.

Keyboard and mechanism

Keyboards and actions must work flawlessly for an instrument to be

considered playable. In examining many early keyboard instruments over the
years, the author has often been surprised at the amount of wear on key levers
and other action parts. Late eighteenth- and early nineteenth-century English
square pianos, for example, often have ivory key platings worn nearly down
to the underlying wood, whereas modern pianos that are played relentlessly
for decades in conservatory practice rooms rarely exhibit the same degree of
keyboard wear. This is surprising, considering the greater force exerted in
playing the modern piano as well as the extraordinary demands posed by later
repertoire.
When examining keyboards, it is important to determine if they are original
and whether they have been enlarged. Check to see if the rails of the key
frames have been extended or perhaps capped and redrilled (which may
disguise the original positions and sequence of pins for the natural and
accidental keys). Very often the original keys are flanked by later ones that
were added in the bass and treble during compass enlargement. Numbering
on the older keys may provide evidence of the original keyboard range.
Instead of numbering each key, some makers made a diagonal score mark
across the key levers so that if they are removed, they can be readily replaced
in proper sequence. Such score lines may not extend to added keys. Jacks and
other action parts should be laid out to determine their original sequence.
Jacks that were added during a compass extension will invariably reveal a
different quality of workmanship and may have inscriptions that indicate
when and by whom they were added. Check all parts thoroughly for
inscriptions with an ultraviolet lamp in a darkened room. If there has been a
compass extension or alteration, there will be corresponding evidence in
other parts of the instrument, such as the soundboard, wrestplank, nut, and
bridges.
After removing the keyboard from an instrument, the worker should lift the
keys off the keyboard frame. Before doing any superficial cleaning, carefully
examine the key frame, key levers, and key bed for fragments of wire, quill,
damper cloth, and any other materials that might shed some light on the
history and construction of the instrument. In fact, it is a good idea carefully
to collect all of this debris and preserve it for future study. Even grains of
pollen may provide evidence of where the instrument has been. Any
remaining surface dirt can be removed with a lightly dampened cloth, but
refrain from scraping, sanding, using steel wool, or using any abrasive action
on wooden parts. The keyboard frame should be examined for loose joints,
and all guide pins should be checked for tightness and straightened if bent. If
rusted, guide pins may be cleaned up with 0000 steel wool or crocus cloth. If
key levers are warped, they can be straightened through the application of dry
heat by quickly passing them back and forth over an alcohol lamp or hot-air
gun, taking care that the wood does not get scorched, and applying gentle
pressure with one's hands. Leather work gloves can be worn if the wood parts
become too hot to handle. If there are lead weights pressed into holes drilled
in the back ends of the key levers or dampers, check to see if they have
corroded (this corrosion is primarily due to acids in the wood) and are
rubbing against their neighbors. If the corrosion is moderate, the corroded
sections can be removed with a file or sliced off with a chisel. If the corrosion
is extreme, it may be necessary to pick or drive out the old weights and
replace them with new ones. These can be made by casting lead into a sheet
of suitable thickness and then punching out slugs with an arch punch (Figures
35a and 35b). For tight or loose mortises in key levers, see Fortepiano
maintenance, Organ restoration, and Piano action regulation and voicing
(modern grand).
Figure 35a Damper mechanism of the Conrad Graf fortepiano showing
corroded lead weight.
Figure 35b Damper from the Conrad Graf fortepiano shown with replaced
lead weight.

Original key cloths and leathers, if present, are rarely serviceable due to
wear and attack by moths (see Fortepiano maintenance: Figure 9). Any
unevenness or discontinuity in balance rail or back rail cloths or leathers will
make it difficult to level the playing surfaces of the keys. Front rail and/or
back over-rail rail cloths or leathers, if of the wrong thickness, will adversely
affect key dip as well as the function of harpsichord jacks and piano
escapement mechanisms. If present, the original cloths and leathers may
retain enough of their original thickness and texture to serve as guides for
replacements. Cloth of suitable thickness and weave can be dyed to match the
color of the original if necessary; replacement leather must have the proper
thickness (this can be adjusted with a skiving knife or skiving machine such
as the Scharf-Fix) and have similar texture and resilience. If the original
materials are not present, the thickness of replacement key and action cloths
and leathers will have to be determined by trial and error.
When restoring an early keyboard instrument to playability, one alternative
is to make a replica keyboard and action and preserve the original as is.

Harpsichord jacks and slides

It is rare to find an eighteenth-century or earlier harpsichord with a

complete set of serviceable jacks that require only requilling and possibly the
installation of new damper cloths. Very often one encounters instruments that
are missing a few jacks or have several that are broken or missing tongues. A
decision must be made whether the existing jacks should be repaired and the
missing ones replaced with copies, or whether an entirely new set should be
made and the originals preserved as is. In the author's experience, it is
preferable to make copies and preserve the originals, because continued use
does take a toll on original mechanical parts. Tongues, for example, become
brittle over time and may split during requilling or even shatter from the force
of playing. The author regularly worked on a 1744 Kirckman harpsichord
that was used in concerts in a private home; every concert (which invariably
involved several days of rehearsals prior to each event) required the repair or
replacement of mechanical parts that broke in the course of playing. In
situations where an instrument will not be used on a regular basis, or at all, it
may be preferable to replace broken or missing parts with copies and keep the
structurally sound originals in place (see Harpsichord maintenance: Figure
17).
The holes for pivot pins in early jacks are generally not drilled clear
through the jack, so they cannot be pushed out in order to remove the tongue.
The pins were evidently inserted through a hole drilled into one of the tines
of the jack, then directed through the hole drilled in the tongue, the tongue
aligned, and the sharpened point of the pin then pressed into the opposing
tine. The exposed end of the pin was then cut off with a pair of nippers. A
pair of full flush-cutting nippers can sometimes be used to grip the exposed
end of the pin and pull it out, but if it is absolutely flush with the body of the
jack there may be nothing to grab. In this case, try flexing the tines of the
jack – this may cause the exposed end of the pin to extend beyond the side of
the jack so that it can be grasped. Another technique is to slide a razor blade
between the jack and the tongue; if the edge of the razor blade catches the
shaft of the pin and there is enough clearance, the razor blade can be worked
sideways and the pin can be pushed out sufficiently so that it can be grasped
with pliers. Broken boar bristle springs should not be replaced with nylon or
wire; boar bristle can be obtained from makers of bespoke shoes (the author
has a lifetime supply kindly provided by Foster & Son in London). Italian
jacks often have brass leaf springs. If these need to be replaced, a good
source for brass is shim stock, which is obtainable in graded thicknesses from
machinists' suppliers. Brass wire can also be hammered flat or passed through
a rolling mill to convert it into ribbons that can then be cut to the proper
length.

Jack guides or registers

Accurately made jack guides are vital to the working of harpsichords. Not
only must they hold the entire rack of jacks at precisely the correct distance
from their respective strings, but they must permit them to move up and
down without undue play. Furthermore, allowances must be made for
fluctuations in temperature and humidity so that the jacks will not bind in
humid weather or rattle during dry spells. If original jack guides are worn or
for some reason do not support the jacks a uniform distance from the strings,
it may be necessary to make new ones. Some French harpsichords had
leathered slides, though it is very rare for the original leather to be in usable
condition.
An example of a thorny problem that may be encountered when dealing
with jack guides can be found in some English bentside spinets, such as those
made by Baker Harris. His guides were made by planing a groove in a strip
of maple, and then sawing up the grooved strip to produce a series of
parallelopiped-shaped blocks, which were then glued side by side so that the
grooves formed a sequence of mortises for the jacks. This box-guide was then
glued between the belly rail and the wrestplank. This type of guide was
evidently fabricated before the key levers were sawn out of a joined-up plank,
and before the nut was pinned. To mark out the keyboard, the joined-up plank
and key frame were positioned in the key well, and a tool shaped like a jack
was inserted into each of the guide's mortises and used to impress a series of
marks onto the back end of the plank (these marks are still visible if the
padding at the back end of the key levers is peeled back). The marks were
used to lay out the key levers so that when they were sawed out, they would
be in alignment with their respective jacks. In all likelihood, a quilled jack
was moved from mortise to mortise to accurately position the strings when
pinning the nut. In English bentside spinets, the plucking direction alternates
from note to note, so any subsequent shifting of the jack guide relative to the
string band will result in short plectra alternating with long plectra, which is
inimical to even voicing and key touch. What invariably happens with such
instruments is that the jack guides shrink over time, and the parallelopipedal
blocks come apart in groups. If they are then glued back together, and the
reassembled guide is reinstalled, the jacks will no longer be in alignment with
the strings. If this is the case, a new nut will then have to be made or the old
one plugged and repinned; it is to be hoped that the key levers will be able to
accommodate jacks that are slightly out of position. Alternatively, a new jack
guide could be made, though it would be quite a trick to fabricate it in the
original manner and expect it to align perfectly with the strings.
Early harpsichord jacks and fortepiano hammers and escapement parts
were generally exquisitely fabricated; when making copies or replacement
parts, the best way of achieving the same precision and appearance is by
recreating the original manufacturing process. This requires a close
examination of layout and tool marks and the construction of appropriate jigs
and specialized tools (see Woodworking).
See Clavichord maintenance, Fortepiano maintenance, and Harpsichord
maintenance, as well as Wire gauges for early keyboard instruments for
further information on action regulation, stringing, voicing, and
miscellaneous repairs drawn from early nineteenth-century fortepiano
maintenance manuals.
For treatment of active woodworm and other pests, see Disinfection,
disinfestation. and fumigation.
Note: unless adequate climate control (50% humidity and 70°F/20°C) can
be arranged for a keyboard instrument following its restoration, it is fruitless
to restore it, especially in locales where there is considerable fluctuation in
temperature and humidity during the year. For example, an early nineteenth-
century Viennese fortepiano that has been tuned up and regulated in the
summer when the humidity is 90% might experience a drop in pitch by a
third and have its hammer mechanism go completely out of regulation when
the radiators come on in the winter and the humidity drops to 10%. Under
such conditions, one should expect to spend most of one's time tuning and
adjusting the mechanism (rather than playing the instrument), and within a
short period of time, major structural problems would likely develop.

References
General
A. Berner, J. H. van der Meer, G. Thibault, and N. Brommelle, Preservation
and Restoration of Musical Instruments (London, 1967).
Claire Chevallier and Jos van Immerseel, eds., Matière et musique: The
Cluny Encounter (Antwerp, 2000).
Peter Donhauser, ed., Restaurieren, Renovieren, Rekonstruieren: Methoden
für Hammerklaviere (Vienna, 1997).
L'instrument de musique dans les musées: quelle restauration pour quelle
esthétique. Actes du colloque AMS-ICOM Lausanne, November, 1996
(Basle, 1997).
Instruments pour demain: conservation et restauration des instruments de
musique, 9es journées d'études de la Section française de l'institut
international de conservation, Limoges 15–16 juin 2000 (Champes-sur-
Marne, 2000).
Kielinstrumente aus der Werkstatt Ruckers zu Konzeption, Bauweise und
Ravalement sowie Restaurierung und Konservierung: Bericht über die
Internationale Konferenz vom 13. bis 15. September 1996 im Handel-Haus
Halle. Schriften des Handel-Hauses in Halle 14 (Halle an der Saale, 1998).
Per una carta europea del restauro: conservazione, restauro e riuso degli
strumenti musicali antichi: atti del convegno internazionale Venezia, 16–
19 ottobre, 1985. Quaderni della Rivista Italiana de Musicologia, Societa
Italiana di Musicologia (Florence, 1987).
Restauro, conservazione e recupero di antichi strumenti musicali: atti del
Convegno Internazionale Modena 2–4 aprile, 1982). Historiae Musicae
Cultores Biblioteca 40 (Florence, 1986).
Christiane Rieche, ed., Kielinstrumente aus der Werkstatt Ruckers (Halle an
der Saale, 1998).
Gabriele Rossi-Rognoni, ed., Bartolomeo Cristofori: la spinetta ovale del
1690 (Florence, 2002).
Gabriele Rossi Rognoni, ed., Restauro e conservazione degli strumenti
musicali antichi: la spinette ovale di Bartolomeo Cristofori. Proceedings
of the International Workshop organized by the Department of Musical
Instruments of the Galleria dell' Accademia on October 21, 2002
(Florence, 2008).

Technical examination of early keyboard

instruments
William Dowd, “The Surviving Instruments of the Blanchet Workshop,” in
Howard Schott, ed., The Historical Harpsichord, vol. I (Brookline, 1984).
Ann and Peter Mactaggart, “Some Problems Encountered in Cleaning Two
Harpsichord Soundboards,” Studies in Conservation 22 (1977), pp. 73–84.
Grant O'Brien, Ruckers: A Harpsichord and Virginal Building Tradition
(Cambridge, 2000).
Stewart Pollens, “Michele Todini's Golden Harpsichord: An Examination of
the Machine of Galatea and Polyphemus,” Metropolitan Museum Journal
25 (1990), pp. 33–47.
Stewart Pollens, “Flemish Harpsichords and Virginals in the Metropolitan
Museum of Art: An Analysis of Early Alterations and Restorations,”
Metropolitan Museum Journal 32 (1997), pp. 85–110.
Richard Rephann, “A Fable Deconstructed: The 1770 Taskin at Yale,” in
Howard Schott, ed., The Historical Harpsichord, vol. IV (Hillsdale, NY,
2002).
 T
Tap drill bit sizes

The following abbreviations are used in the tables below: NC National

Coarse; NF National Fine; NEF National Extra Fine; NS National Special;
NPT National Standard Taper Pipe Thread.

Drill Closest fractional Decimal

Thread size
bit bit inches

0-80 NF in. in. 0.0469

1-64 NC no. 53 - 0.0595

1-72 NF no. 53 in. 0.0595

2-56 NC no. 50 - 0.0700

2-64 NF no. 50 - 0.0700

3-48 NC no. 47 in. 0.0785

3-56 NF no. 45 – 0.0820

4-36 NS no. 44 – 0.0860

4-40 NC no. 43 in. 0.0890

4-48 NF no. 42 in. 0.0935

3 mm-0.50 mm 2.5 mm - 0.0984

⅛-40 NS no. 38 - 0.1015

5-40 NC no. 38 - 0.1015

5-44 NF no. 37 - 0.1040

6-32 NC no. 35 in. 0.1100

6-36 NS no. 34 - 0.1110

6-40 NF no. 33 - 0.1130

6-48 NS no. 31 - 0.1200

4 mm-0.70 mm 3.4 mm - 0.1338

8-32 NC no. 29 - 0.1360

8-36 NF no. 29 in. 0.1360

8-40 NS no. 28 - 0.1405

-24 NS no. 26 - 0.1470

10-24 NC no. 25 in. 0.1495

-32 NS no. 22 - 0.1570

10-32 NF no. 21 in. 0.1590

5 mm-0.90 mm 4.2 mm - 0.1653

5 mm-0.80 mm 4.3 mm - 0.1693

12-24 NC no. 16 in. 0.1770

12-28 NF no. 14 0.1820

 in.

12-32 NEF no. 13 - 0.1850

14-20 NS no. 10 - 0.1935

¼-20 NC no. 7 in. 0.2010

14-24 NS no. 7 - 0.2010

6 mm-1.00 mm 5.2 mm - 0.2047

¼-24 NS no. 4 - 0.2090

¼-28 NF no. 3 in. 0.2130

¼-32 NEF in. in. 0.2188

¼-40 NS no. 1 - 0.2280

7 mm-1.00 mm 6.1 mm in. 0.2401

-18 NC F in. 0.2570

8 mm-1.25 mm 6.9 mm in. 0.2716

-24 NF I - 0.2720

8 mm-1.00 mm 7.1 mm - 0.2795

-32 NEF in. in. 0.2812

9 mm-1.25 mm 7.9 mm - 0.3110

⅜-16 NC in. in. 0.3125

9 mm-1.00 mm 8.1 mm - 0.3189

9 mm-0.75 mm 8.3 mm - 0.3268

⅜-24 NF Q in. 0.3320

10 mm-1.25 8.9 mm in. 0.3503

mm

10 mm-1.50 8.7 mm - 0.3425

mm

10 mm-1.00 9.1 mm - 0.3583

mm*

-14 NC U in. 0.3680

11 mm-1.50 9.7 mm - 0.3818

mm

-20 NF in. in. 0.3906

12 mm-1.50 10.7 in. 0.4212

mm mm

12 mm-1.75 10.5 - 0.4133

mm mm

½-13 NC in. in. 0.4219

12 mm-1.25 10.9 in. 0.4291

mm* mm

½-20 NF in. in. 0.4531

½-24 NS in. in. 0.4531

14 mm-2.00 12.2 - 0.4803

mm mm
 -12 NC in. in. 0.4844

14 mm-1.50 12.7 - 0.4999

mm mm

14 mm-1.25 12.8 - 0.5039

mm* mm

-18 NF in. in. 0.5156

⅝-11 NC in. in. 0.5312

16 mm-2.00 14.2mm in. 0.5590

mm

⅝-18 NF in. in. 0.5781

16 mm-1.50 14.7 - 0.5787

mm mm

-11 NS in. in. 0.5938

18 mm-2.50 15.8 in. 0.5220

mm mm

-16 NS ⅝ in. ⅝ in. 0.6250

¾-10 NC in. in. 0.6562

18 mm-1.50 16.8 - 0.6614

mm* mm

¾-16 NF in. in. 0.6875

20 mm-2.50 17.8 in. 0.7008

mm mm

⅞-9 NC in. in. 0.7656

⅞-14 NF in. in. 0.8125

22 mm-2.50 20.9 - 0.8228

mm mm

⅞-18 NS* in. in. 0.8281

24 mm-3.00 21.4 in. 0.8425

mm mm

1–8 NC ⅞ in. ⅞ in. 0.8750

24 mm-2.00 22.3 - 0.8779

mm mm

1–12 NF in. in. 0.9219

1-14 NS in. in. 0.9375

1⅛-7 NC in. in. 0.9844

1⅛-12 NF 1 in. 1 in. 1.0469

1¼-7 NC 1 in. 1 in. 1.1094

1¼-12 NF 1 in. 1 in. 1.1719

1⅜-6 NC 1 in. 1 in. 1.2188

1⅜-12 NF 1 in. 1 in. 1.2969

1½-6 NC 1 in. 1 in. 1.3438

 1½ in.-12 NF 1 in. 1 in. 1.4219

* These sizes are spark plug taps.

To tap this Use this drill Closest Decimal

size bit fractional bit inches

⅛-27 NPT R - 0.3390

¼-18 NPT 7/16 in. 7/16 in. 0.4375

⅜-18 NPT 37/64 in. 37/64 in. 0.5781

½-14 NPT 45/64 in. 45/64 in. 0.7031

¾-14 NPT 59/64 in. 59/64 in. 0.9219

1-11½ NPT 1 in. 1 in. 1.1562

1¼-11½ 1½ in. 1½ in. 1.5000

NPT

1½-11½ 1 in. 1 in. 1.7344

NPT

2-11½ NPT 2 in. 2 in. 2.2188

References
Edward G. Hoffman, Student's Shop Reference Manual (New York, 1985).
Erik Oberg, Franklin D. Jones, Holbrook L. Horton, and Henry H. Ryffel,
Machinery's Handbook 29th Edition (New York, 2012).
John E. Traister, Machinists' Ready Reference Manual (New York, 1987).

Tapered reamers

Unless noted all measurements are in inches.

American Standard Morse taper finishing reamers

Size Small end Large dia. Flute length

0 0.2503 0.3574 2¼

1 0.3674 0.5170 3

2 0.5696 0.7444 3½

3 0.7748 0.9881 4¼

4 1.0167 1.2893 5¼

5 1.4717 1.8005 6¼

Morse taper reamers are available with straight or left-hand spiral flutes.

Brown & Sharpe taper reamers

Size Small end Large end. Flute length

1 0.1974 0.3176 2⅞

2 0.2474 0.3781 3⅛

3 0.3099 0.4510 3⅜
 4 0.3474 0.5017 3

5 0.4474 0.6145 4

6 0.4974 0.6808 4⅜

7 0.5974 0.8011 4⅞

8 0.7474 0.9770 5½

9 0.8974 1.1530 6⅛

10 1.0420 1.3376 6⅞

Brown & Sharpe taper reamers are available with straight and left-hand
spiral flutes.

Taper pin reamers

Size Small end Large end Flute length

No. 8/0 0.0351 0.0514

No. 7/0 0.0497 0.0666

No. 6/0 0.0611 0.0806

No. 5/0 0.0719 0.0966 1

No. 4/0 0.0869 0.1142 1

No. 3/0 0.1029 0.1302 1

No. 2/0 0.1137 0.1462 1

 No. 0 0.1287 0.1638 1

No. 1 0.1447 0.1798 1

No. 2 0.1605 0.2008 1

No. 3 0.1813 0.2294 2

No. 4 0.2071 0.2604 2

No. 5 0.2409 0.2994 2

No. 6 0.2773 0.354 3

No. 7 0.3297 0.422 4

No. 8 0.3971 0.505 5

No. 9 0.4805 0.6066 6

No. 10 0.5799 0.7216 6

No. 11 0.706 0.878 8

No. 12 0.842 1.05 10

No. 13 1.009 1.259 12

No. 14 1.25 1.542 14

Taper pin reamers taper ¼ in. per foot. They are available with straight
and left-hand spiral flutes.

American National Standard (ANS) die-maker's reamers

Size Small end Large end Flute length

AAA 0.055 0.070 1⅛

AA 0.065 0.080 1⅛

A 0.075 0.090 1⅛

B 0.085 0.103 1⅜

C 0.095 0.113 1⅜

D 0.105 0.126 1⅝

E 0.115 0.136 1⅝

F 0.125 0.148 1¾

G 0.135 0.158 1¾

H 0.145 0.169 1⅞

I 0.160 0.184 1⅞

J 0.175 0.199 1⅞

K 0.190 0.219 2¼

L 0.205 0.234 2¼

M 0.220 0.252 2½

N 0.235 0.274 3

O 0.250 0.296 3½
 P 0.275 0.327 4

Q 0.300 0.358 4½

R 0.335 0.397 4¾

S 0.370 0.435 5

T 0.405 0.473 5¼

U 0.440 0.511 5½

ANS die-maker's reamers taper at a ¾-degree included angle or 0.013 in.

per inch. They are available with left-hand spiral flutes.

Violin/viola/cello/bass peg-hole and endpin reamers

Measurements in millimeters

Size Small end Large end Length

Small 4.0 7.5 105

Standard 5.5 10 135

Cello 8 16 200

Bass 23 36 221

Cello endpin 15 27.5 212.5

Bass endpin 23 26 221

Cello/bass endpin 20 39 305

 Peg-hole reamers are available with straight and left-hand spiral flutes.
Violin/viola peg-hole reamers have a 1/30 taper. Cello peg-hole reamers
have 1/25 taper. Cello and bass endpin reamers have a 1/17 taper. Peg
shapers are available to match the above reamers.
3-degree and 5-degree reamers are available for fitting guitar bridge pins
and harp board pins.

References
Edward G. Hoffman, Student's Shop Reference Manual (New York, 1985).
Erik Oberg, Franklin D. Jones, Holbrook L. Horton, and Henry H. Ryffel,
Machinery's Handbook 29th Edition (New York, 2012).

Temperature

To convert to Celsius from Fahrenheit:

To convert to Fahrenheit from Celsius:

The Réaumur scale was developed in France by René Antoine Ferchault de

Réaumur in 1730, with the freezing point of water at 0° and the boiling point
of water at 80°. This scale was widely used in Europe until the adoption of
the metric Celsius scale around 1790. The Réaumur scale is still used in
measuring the temperature of milk in cheese production in Italy and
Switzerland. The abbreviations °Ré, °Re, or °R are used to denote the use of
this scale.
To convert to Celsius from Réaumur:
°C = °Ré × 5/4
 To convert to Fahrenheit from Réaumur:
°F = (°Ré × 9/4) + 32

Tempering steel

To temper steel, first heat the steel red-hot and quench it in water (brine or
oil are used to quench certain alloys). When quenching an oblong piece of
steel, it should be held straight and immersed quickly to prevent warping.
Quenching brings steel to its maximum hardness and brittleness. It is then
tempered (softened) by carefully reheating it over a column of hot air from a
torch, hot coals, or stove until the desired oxidation color appears. When
tempering a tool with a beveled cutting edge, heat should be applied at the
thickest part of the tool and conducted to the cutting edge. Take care, as the
thinner sections heat up more quickly than the thicker parts, and one does not
want to overly soften the cutting edge. If the reheating is done slowly, one
can observe the oxidation colors advance to the cutting edge and then halt the
tempering at just the right moment by removing the tool from the source of
heat and dipping it into cool water. A properly tempered cutting tool will be
less prone to break or chip while cutting or during impact. When regrinding
cutting edges, it is important not to overheat the tool and upset the temper. If
using motorized grinding wheels, use a light touch and dip the tool in a cup
of water at room temperature to keep the edge cool (see Sharpening tools).

Temper colors

Color °F/°C Tool

Faint yellow 420/216 knives, hammers

Very pale 430/221 reamers

yellow

Light yellow 440/227 lathe and milling cutters

Pale straw- 450/232 twist drills for hard use, razors, delicate
yellow cutting tools

Straw- 460/238 dies, punches, bits

yellow

Deep straw- 470/243 wood and ivory cutting tools

yellow

Dark yellow 480/249 twist drills, large taps

Yellow- 490/254
brown

Brown- 500/260 gravers, wood chisels, axes, taps ½-

yellow inch and over, and dies

Red-brown 510/266

Brown- 520/271 taps ¼-inch and under

purple

Light purple 530/277

Purple 540/282 cold chisels, center punches

Dark purple 550/288

Blue 560/293 screwdrivers, springs, gears, saw blades

Dark blue 570/299

Medium 600/316 scrapers and spokeshaves

blue

Light blue 640/338

 To achieve various degrees of Rockwell C hardness for water-hardening
tool steel, heat the metal to 1450–1500°F (788–816°C) for a few seconds,
then quench in 8% brine. Temper for 1 hour at the following temperatures:

Temperature in °F/°C Rockwell C hardness

200/93 66–67

300/149 64–65

400/204 61–62

500/260 58–59

600/316 54–55

700/371 50–51

800/427 46–47

To achieve various degrees of Rockwell C hardness for oil-hardening tool

steel, heat slowly to 1475–1525°F (800–830°C) hold for a few seconds, then
quench in oil. Temper for 1 hour at the following temperatures:

Temperature in °F/°C Rockwell C hardness

300/149 63–65

400/204 61–62

500/260 58–60

600/316 54–56
 700/371 51–53

800/427 46–48

900/482 43–45

References
Thomas J. Glover, Pocket Ref (Littleton, 2006).
Erik Oberg, Franklin D. Jones, Holbrook L. Horton, and Henry H. Ryffel,
Machinery's Handbook 29th Edition (New York, 2012).
John E. Traister, Machinists' Ready Reference Manual (New York, 1987).

Textile cleaning

(see also Bleach, Cleaning: soaps and detergents, Solvents and solvent
cleaning, Stain removal, Water)
The natural-fiber textiles most often encountered in early musical
instruments are cotton, linen, silk, and wool. The animal fibers (silk and
wool) are primarily protein, while the plant fibers are primarily cellulose.
These four fibers are readily identified under a simple light microscope:
cotton fibers are somewhat flattened and twisted; linen fibers are cylindrical
with spiral or x-shaped transverse nodes; silk is roughly triangular in cross
section, relatively smooth throughout its length, and translucent (raw silk
consists of two connected filaments, while spun silk has had these two
connected fibers separated); wool fibers are roughly cylindrical and have
overlapping cuticle cells resembling scales (the scale patterns vary according
to the animal species; see Teerink, 1991).
Natural fibers deteriorate due to mold and mildew, insect infestation (such
as by moths, which often lay their eggs in wool), excessive heat and dryness
(which cause fibers to become brittle), exposure to ozone, visible light, and
UV (which cause dyes to fade and fibers to weaken, known as “tendering”),
as well as contact with certain chemicals (such as the acidic vapors exuded by
many woods and noxious gases in the atmosphere).
 Textiles should only be cleaned after thorough testing to determine the
strength of fibers and the stability of dyes. Keyboard and action cloths, front-
and balance-pin punchings, and dampers, which are very often made of wool
cloth or felt, may be vacuum-cleaned by placing or stretching a synthetic-
fiber (such as nylon) screen or netting over them to prevent them from being
drawn into the vacuum cleaner and adjusting the suction so that it does not
damage the weave or pull the fibers free (especially with felted materials).
Cleaning with water (see Water) and organic solvents (see Solvents and
solvent cleaning) both present risks. Water causes most natural fibers to relax
and swell, which may be beneficial in terms of restoring a textile's original
tactile qualities, though there may be shrinkage, wrinkling, or shape
distortion upon drying. Water may also leach out dyes and other substances.
Hot water (above 120°F/50°C) should generally be avoided and never used to
wash wool or silk. Dyed silk should not be washed in temperatures higher
than about 60°F (15°C). Because the thicknesses of felted-wool key
punchings and wool key cloths are critical to the proper functioning of
keyboard-instrument mechanisms, it is preferable to “dry,” or solvent, clean
them rather than risk shrinkage that might result from wet cleaning. Velvets
should never be cleaned with water, and nor should weighted or watered
(moiré) silk. However, for most other fabrics, wet cleaning is generally
considered to be safer than dry cleaning, and distilled or deionized water
alone is often sufficient to remove dust and other particulates from them.
Many types of soil (such as salts, sugars, starches, blood, coffee, tea, wine,
etc.) are largely soluble in water, though oils and grease require the addition
of soaps or detergents or the use of organic solvents to release them.

Mold

Relative humidity above 65%, temperatures above 75°F (24°C), and

stagnant air encourage the growth of mold and mildew in fabric-lined
instrument cases and on keyboard action cloths. Spores and hyphae can be
removed with a vacuum cleaner equipped with a HEPA filter; a soft-bristle
brush used in conjunction with the vacuum cleaner?s brush may help
dislodge this material. When vacuuming delicate textiles, open weave nylon
netting should be stretched over the area being vacuumed to prevent fibers
from being pulled from the fabric. The solvents used in dry cleaning will kill
mold and its spores. Vulpex added to dry-cleaning fluid (see below) is
generally effective in removing stains caused by mold, though mild bleaching
(see Bleach) may be required. It should be noted that areas affected by mold
may have become weakened, so care should be exerted when vacuuming, wet
or dry cleaning, or bleaching.

Wet cleaning

Soaps are salts of fatty acids and are made by treating animal fats or plant
oils with an alkali: traditionally caustic potash (potassium hydroxide) or
caustic soda (sodium hydroxide). Soaps are not generally recommended in
conservation work because they tend to form an insoluble scum or film,
especially with hard water, and they are by nature “ionic.” So-called “neutral
soaps” are not in fact pH neutral but simply contain a reduced amount of free
alkali. Many types of soil are acidic in nature, which tends to neutralize the
alkalinity of soapy water, so extra alkali is often added to commercial
laundering soaps to increase their effectiveness; these are known as “built” or
“heavy-duty” soaps, and they are not recommended for conservation work,
especially with delicate fabrics.
Softeners are sometimes added to water to prevent the formation of hard-
water scum by soaps. There are two types of softeners, precipitating (such as
sodium carbonate, commonly known as washing soda) and non-precipitating
(such as Calgon; sodium hexametaphosphate). The first type of softener
reacts chemically with minerals in the water, and the resulting compounds
precipitate or settle out of solution. Non-precipitating softeners also combine
with these minerals but keep them in suspension.
In home and commercial laundering, residual alkalinity is sometimes
neutralized by the addition of acetic acid (white vinegar), sodium bisulfate, or
oxalic acid to the rinse water. However, residual alkalinity left by soap may
in some cases be beneficial in that it offers protection from acidic vapors in
the atmosphere or those exuded by backing or mounting materials (Kajitani,
1989).
Vulpex (potassium methyl cyclohexl oleate) is a spirit soap that can be
added to water (5%) or mineral spirits (see below).
There are three types of detergents: anionic detergents, which have a
negative ionic charge; cationic detergents, which have a positive ionic
charge; and nonionic detergents, which have no ionic charge. All three types
are commonly used: the popular household wool cleaner Woolite contains
both anionic and nonionic detergents, as well as other ingredients; cationic
detergents are often used in shampoos; and nonionic detergents are often
found in dishwashing detergents. Nonionic detergents are generally preferred
for conservation work because of their lower reactivity, especially with
sensitive dyes. Unlike soaps, synthetic detergents do not form scum or film in
hard water; unfortunately, they tend to be much less efficient in dirt removal
than soap. Commercial laundering detergents generally often contain a range
of chemicals in addition to the detergent; these may include soaps, complex
builders and chelating agents (such as sodium tripolyphosphate, sodium
carbonate, sodium citrate, sodium silicate, and EDTA), dirt redeposition
inhibitors (such as carboxymethyl cellulose), bleaching agents (such as
sodium perborate), fluorescent whiteners, disinfectants, fabric softeners,
starches, dyes, and perfumes. In conservation work, some of these additives
may not be desirable, so it is more prudent to compose one's own washing
agents or to use those commercial products that have been tested and
approved by leading textile conservators (see below).
When diluting a detergent for use, it is desirable to know its critical micelle
concentration (CMC) value; unfortunately, this value is often not supplied by
manufacturers. Micelles are groups of detergent molecules that surround dirt
particles and cause them to be suspended in water. The CMC value refers to
the number of grams of the detergent per liter of water. Below the critical
level, the detergent may not be completely effective. By using a multiple of
the CMC value (for example, 5 times the CMC for nonionic detergents and
2–3 times the CMC for anionic detergents), there will be a surfeit of detergent
molecules to reduce the possibility of the solution becoming exhausted
during cleaning. Below are CMCs for several detergents commonly used in
conservation work:

Detergent Type CMC*

Dehypon LS45** nonionic 0.598

Synperonic N*** nonionic 0.190

 Synperonic A7 nonionic 0.822

Synperonic 91/6 nonionic 0.557

Hostapon T anionic 0.438

Orvus WA** anionic 2.907

* Experimentally determined CMC values by Fields et al., 2004.

** Recommended detergents (see below).

*** Ecological issues have led some conservators to replace this once popular detergent with
Dehypon LS45.

Rust stains in textiles may be removed with 1% oxalic acid, 2–10%

ammonium citrate solution, or a saturated solution of disodium EDTA. After
cleaning textiles with chemical solutions, it is important to rinse them
thoroughly in distilled or deionized water. Conductivity and pH meters
should be used to test rinse waters in order to insure that the chemicals have
been removed.
Bleaching (see Bleach) should be carried out separately from general
cleaning, and only on strong fabrics. Silk and wool in particular should not be
subjected to chlorine bleaches (such as Clorox and Purex). In general,
chlorine bleach is best avoided because it is difficult to wash out, and any
residue may continue its bleaching action and break down textile fibers.
Hydrogen peroxide (H2O2) is a safer bleaching agent, as its active agent,
oxygen, dissipates, leaving behind only water. Commercially available
hydrogen peroxide is generally stabilized by the addition of an acid, which
must be neutralized and made slightly alkaline by the addition of a small
amount of alkali, such as household-strength ammonia. The alkali should be
added gradually in relatively concentrated form so as not to overly dilute the
hydrogen peroxide, while monitoring the pH with an electronic meter. A pH
up to about 11 can be used for textiles made of cotton and linen, but
proteinaceous fibers, such as wool, should be washed in a neutral-pH
environment. Because dirt and grime are often acidic, bleaching with
hydrogen peroxide is best done after thorough washing. 3% hydrogen
peroxide is effective for mild general bleaching, through more concentrated
solutions (up to around 10%) may be used locally on spots and strong stains.

General textile bleaching solution (Landi, 2002)

5 g sodium hydroxide
5 g sodium carbonate anhydrous
20 g sodium metasilicate
50 ml of 100 volume (30%) hydrogen peroxide
1 liter of deionized or distilled water.

The above is sufficient to bleach textiles weighing 100 g.

After bleaching, textiles should be thoroughly rinsed with distilled or
deionized water.

Dry cleaning

Dry cleaning is generally more effective than water in the removal of oil
and grease stains. The most frequently used solvent for dry cleaning of
textiles is 1,1,1-trichloroethane (Genklene is a popular brand). 1% Vulpex
spirit soap may be added to 1,1,1-trichloroethane, though this soap must be
cleared with clean, unadulterated solvent afterwards. When dry-cleaning
wool, the natural lanolin may be leached out by the organic solvents, leaving
the fabric brittle and dull in appearance. A small amount of lanolin or
glycerin dissolved in hexanes may be lightly sprayed on the fabric after
cleaning to restore flexibility and sheen.

Drying

Key cloths and punchings (which are generally made of wool cloth or wool
felt) should be blotted after cleaning and left to air-dry. Long cloth strips
(such as key cloths and jack rail cloths) may be pinned to a block of polyester
foam after washing to keep them straight as they dry.

Pressing

Ironing and pressing may be harmful to fabrics, especially those that have a
textured surface. Fabrics that can bear contact with water are best pressed
when slightly damp. Linen and cotton require relatively high heat for
pressing, while synthetic fabrics must be pressed under low heat to prevent
the fibers from melting. Creases in wool cloth can be removed with steam or
ironed under low heat through a dampened cloth. Creases in velvet are best
removed with a fabric steamer; a soft natural bristle brush can be used in
conjunction with steam to straighten or orient the pile. To prevent the
formation of an unwanted sheen on silks, acetates, and rayon, they should be
ironed on the back side. Textured fabrics should be tested to determine
whether they will be adversely affected by heat. Ironing is not recommended
for cloth used in keyboard actions as it tends to alter its thickness and
resilience, which might adversely affect action regulation.
Cabinet upright and giraffe pianos are often fitted with “sunburst” or
pleated silk panels. In some cases, the fabric can be disassembled from its
wooden frame, cleaned, and then remounted. This work should preferably be
done by a textile conservator and upholsterer. If the pleating is complex and
would be difficult to re-create, or the fabric is badly damaged, it may be
preferable to remove and preserve the original framed panel and substitute a
replica.

Recommended cleaning agents for textiles

Orvus WA paste (also available as a powder) is an anionic synthetic

detergent having a nearly neutral pH that is generally mixed with water (1%).
Because of its nearly neutral pH and the fact that it rinses easily, it is often
used in cleaning textiles and carpets.
Dehypon LS45 is a nonionic synthetic detergent that is used in general
cleaning, and is especially recommended for cleaning delicate textiles. A
working-strength solution can be made by adding 0.3 grams to 1 liter of
water.

A “universal” anionic detergent (Landi, 2002)

1.0 g alkyl benzol sulphonate (anionic surface-active agent)

0.05 g carboxymethyl cellulose (dirt redeposition inhibitor)
1 liter of distilled or deionized water.

Another anionic detergent

0.15 g ammonium lauryl ether sulphate (anionic surface-active agent)

0.15 g alkyl benzene sulphonate (anionic surface-active agent)
0.5 g sodium carboxymethyl cellulose
1 liter distiller or deionized water.

An anionic detergent having a nearly neutral pH that can be used to clean wool and silk

0.5 g fatty acid methyl ester α-sulphonate (anionic surface-active agent)

0.5 g carboxymethyl cellulose
1 liter distilled or deionized water.

References
Ágnes Tímár Balázsy and Dinah Eastop, Chemical Principles of Textile
Conservation (Oxford, 1998).
John A. Fields, Andrew Wingham, Frances Hartog, and Vincent Daniels,
“Finding Substitute Surfactants for Synperonic N,” Journal of the
American Institute for Conservation vol. 43, no. 1 (Spring, 2004), pp. 55–
73.
Judith H. Hofenk-de Graaff, “The Constitution of Detergents in Connection
with the Cleaning of Ancient Textiles,” Studies in Conservation 13/3
(August, 1968), pp. 122–141.
Nobuko Kajitani, “Conservation Maintenance of Tapestries at the
Metropolitan Museum of Art, 1987,” The Conservation of Tapestries and
Embroideries: Proceedings of Meetings at the Institut Royal du Patrimoine
Artistique, Brussels, Belgium, September 21–24, 1987 (Los Angeles,
 1989).
Sheila Landi, The Textile Conservator's Manual (Oxford, 2002).
B. J. Teerink, Atlas and Identification Key: Hair of West-European Mammals
(Cambridge, 1991).
Phyllis G. Tortora, Understanding Textiles (New York and London, 1978).

Tinning

To tin copper or brass pins and wire as described in Diderot's Encyclopedie

(1755), boil cream of tartar (potassium bitartrate) in water with tin filings and
the pins or wire to be plated. This form of electrochemical plating produces a
bright, smooth, though extremely thin coating of tin. It is most likely the
method used by English harpsichord and fortepiano makers to plate their
brass bridge pins and hitchpins.

Reference
Denis Diderot and Jean le Rond D'Alembert, Encyclopedie ou dictionnaire
raisonné des sciences, des arts et des métiers, vol. V (Paris, 1755).

Tortoiseshell, horn, and whalebone

Most of the so-called “tortoiseshell” that has been used in making violin
bow frogs and for veneering stringed instruments was not obtained from
tortoises but rather from hawk's-bill turtles. Tortoisehell can be softened by
immersion in boiling water or through the application of dry heat. After
softening, it can be clamped into molds to emboss or raise decorations on its
surface.
To make blocks of tortoiseshell (for example, to fabricate violin bow
frogs), thin plates of tortoiseshell that have been filed, scraped, or sanded flat
so that they fit together can be fused by immersion in brine for several days,
boiling in fresh water, and then clamping them together. Clamping
compresses the tortoiseshell during the fusing process, but soaking the fused
block in fresh water will cause the fused plates to swell back to nearly their
original thickness. Sheets of tortoiseshell may also be joined by scarfing the
edges and clamping them between hot metal clamps or tongs.
The transparent areas surrounding the dark brown spots of tortoiseshell
veneer are often a bright red color. This is generally not the natural color of
tortoiseshell but may be due to the application of paint, varnish, or glue
colored with red pigment, such as vermilion or red lead, to the underside of
the tortoiseshell. This is often done to disguise the underlying wood.
Isinglass and fish glue are recommended for mending or bonding
tortoiseshell to wood.
The author has successfully replaced missing chips and filled fissures in
tortoiseshell with molten shellac that has either been darkened through
heating or colored with pigments to match the surrounding area. After the
shellac has cooled, it can be leveled by rubbing with a block of hard felt
moistened with alcohol. Shellac fills are reputed to shrink and crumble away,
but those that the author made over forty years ago have remained intact and
unchanged in color, transparency, and surface texture.
Horn has similar physical and working properties to tortoiseshell, and the
lighter, more transparent varieties have sometimes been colored or painted on
the underside in imitation of tortoiseshell. The darker type is almost black in
color and was often used to make turned ferrules for woodwind instruments.
In light of the international restrictions on the import, export, and sale of
tortoiseshell, it is advisable to replace missing pieces with synthetic materials.
Polyester, acrylic, or epoxy resins are easily cast into sheets or blocks, and
while in a liquid state, they can be tinted with dyes or pigments to replicate
the color and mottling found in natural tortoiseshell.
So-called “whalebone” is not bone but the baleen of whales, which consists
of horny plates of fused hair that grow in the mouths of certain whales and
are used to filter krill and other edible material from sea water. It is a springy
material that was once used to make such things as corset stays and buggy
whips. In instrument making, it was used to make damper springs of early
English square pianos, violin-bow wrapping, and violin purfling. It can be
softened by the application of heat or soaking in warm water. Again, in light
of international restrictions, various plastics (such as black Delrin) can be
used in place of true whalebone.
Tuning and temperament

(see also Cents conversion, Equal temperament frequency table,

Intervals)

Temperament

Temperament may be defined as the subtle alteration of musical intervals

to form a functional musical scale. Keyboard instruments are not the only
instruments that require tempered tuning – any instrument fitted with frets
(such as lutes, guitars, mandolins, and viols) or tone holes (woodwind
instruments) may make use of tempered scales. Unfretted instruments, such
as those of the violin family, feature an innate flexibility that enables a skilled
player to “finger” the right note. Instruments with movable frets (such as
lutes and viols) offer a certain degree of flexibility in adjusting the
temperament, as do woodwind instruments by virtue of cross-fingering and
the shading of tone holes with the fingertips.
In Western music, there are a number of methods of constructing scales. A
just diatonic scale can be made up of a sequence of intervals based on simple
ratios, such as the whole tone, third, fourth, fifth, sixth, seventh, and octave:

C D E F G

Ratios 1 9:8 5:4 4:3 3:2

relative
to C

Ratios 8:9 9:10 15:16 8:9 9:10

between
notes:

Note that in this type of scale there are two sizes of whole tones, the major
whole tone (9:8) and the minor whole tone (10:9), and one size of semitone
(16:15). The ratios between the notes are determined by subtracting the
smaller interval from the larger. This is done by dividing their ratios or cross-
multiplying them in fractional form. To add musical intervals, their ratios,
expressed as fractions, are multiplied.
In Pythagorean temperament, only the intervals of the octave (2:1) and fifth
(3:2) are employed in building up the other notes of the scale. However, a
major problem encountered with Pythagorean temperament is that a sequence
of twelve perfect fifths exceeds seven perfect octaves by a small amount
termed the ditonic comma, which can be calculated by multiplying 3/2 (the
ratio of the fifth) by itself 12 times, multiplying 2/1 (the ratio of the octave)
by itself 7 times, and then dividing the first figure by the second, or (3/2)12 ÷
27, which yields 531441/524288, or roughly 1.013643265 (the ratio of this
comma). Furthermore, in this system of tuning, the notes derived from an
ascending series of fifths (C–G–D–A–E–B–F♯–C♯–G♯–D♯–A♯–E♯–B♯, etc.)
do not coincide precisely with those produced by a descending series of fifths
(C–F–B♭–E♭–A♭–D♭–G♭–C♭–F♭–B♭♭–E♭♭–A♭♭–D♭♭, etc.). Thus, in the
Pythagorean temperament there are no enharmonics (i.e. tones in a scale that
can serve two functions – for example in equal temperament the piano's black
key between C and D may serve as both C♯ and D♭). In Pythagorean tuning,
its major third has a ratio of 81/64, making it wider than a 5:4 third; similarly,
the Pythagorean minor third has a ratio of 32:27, rendering it narrower than a
6:5 minor third. The Pythagorean diatonic scale has one size whole step (9:8),
and one size half step (256:243):

C D E F G A

Ratios 8:9 8:9 243:256 8:9 8:9

between
notes

The Pythagorean chromatic scale has two types of half steps, the minor
second, such as C–D♭ (256:243), and the augmented prime, such as C–C♯
(2187:2048), the latter being a ditonic comma larger. This can be
demonstrated by subtracting the ratios of the smaller interval from the larger
(i.e. dividing or cross-multiplying), which yields the ratio of the comma
given above.
When tuning keyboard instruments, adjustments are made to form a
workable scale by using divisions of the commas. For example, in so-called
“equal temperament,” each fifth used in building the temperament octave is
narrowed by a twelfth of the ditonic comma. As a consequence, all of the
fourths are widened and the major thirds become very wide. When listening
to a piano tuned in equal temperament (as are most pianos these days),
audiences are accustomed to the cacophony of the out-of tune major and
minor thirds and their inversions, but in earlier times this type of keyboard
temperament was not generally considered acceptable.
The equally tempered scale can also be constructed by multiplying the first
note of the scale and each successive half-step by a factor of . The
semitone divisions of the equal-tempered scale can be thought of as
consisting of 100 “cents,” which form a ratio of :1; thus, the octave
contains 1200 cents (see Cents conversion). Many books and articles on
temperament provide cent deviations for the notes in various tempered scales.
These figures indicate how far the notes deviate from those in the equal-
tempered scale. While useful for comparing the sizes of intervals and for
programming electronic tuners, the cent values of non-equally tempered
intervals can be misleading because equal-tempered notes and intervals are
themselves tempered (for example, an equal-tempered third consists of a nice
round 400 cents, whereas a 5/4 major third consists of 386 cents, or 14 cents
short of an equally tempered third).
A problem encountered with constructing a scale with perfect fifths is that
four perfect fifths exceed two octaves and a 5:4 major third by a small
amount, which can be calculated by multiplying 3:2 (the ratio of the perfect
fifth) by itself four times and then subtracting 4:1 (the ratio of two octaves)
and 5:4 (the ratio of the pure major third). The difference between four
perfect fifths and two octaves plus a major third can be expressed by the ratio
81:80 (1.0125), which is known as the syntonic comma (also known as the
comma of Didymus or Ptolemy). In so-called “meantone temperament,” the
syntonic comma is divided up among certain tempered intervals used to
construct the scale. If the syntonic comma is distributed evenly among the
four fifths C–G, G–D, D–A, and A–E, as it is in so-called “strict quarter-
comma” meantone temperament, the interval C–E becomes a pure major
third; however, unlike the third in the just diatonic scale, it is not composed
of two sizes of whole tones (C–D, 8:9; and D–E, 9:10) because the narrowed
fifths used to arrive at the third C–E produce a D that is the “mean tone”
(having the ratio of :1) between C and E – hence the term “meantone
temperament.” The proof of this is as follows: . In
quarter-comma meantone temperament, there are eleven fifths tempered by ¼
syntonic comma, with the remaining twelfth fifth being wider by 1¾ syntonic
comma (see Terminology below and Intervals). This twelfth fifth is known as
the “wolf fifth,” and it is ordinarily placed between G♯ and E♭, though D♯–
B♭ and C♯–A♭ are other possible positions for it.
Two, three, five, six, and 2/7 divisions of the syntonic comma have also
been used in constructing temperaments, but pure 5:4 thirds do not result. In
third-comma temperament, for example, the minor thirds are pure, though the
major thirds are not. One of the conveniences of tuning quarter-comma
meantone is that after tuning four successive tempered fifths, the accuracy of
their tempering can be confirmed if the resulting third is pure, or beatless.
The rest of the notes in the temperament octave can be tuned by pure thirds.
In equal temperament, the confirmation occurs only after all twelve fifths
have been properly tempered and the final note is a perfect octave above the
starting point. One interesting feature of meantone temperament is that the
pitch-order of sharps and flats is reversed from that of Pythagorean
intonation; that is, in meantone tuning, the sharps are lower than the flats (so
C♯ is lower in pitch than D♭), whereas in Pythagorean intonation the sharps
are higher in pitch than the flats.

Terminology

Beats are periodic fluctuations in amplitude that occur when two notes are
played together and there is a small frequency difference between the two
notes or their coincident overtones (the latter is calculated by cross-
multiplying the fundamental frequencies by the numbers that form the ratio
of the interval). For example, if an A tuned to 440 Hz is played against
another A tuned to 442 Hz, one will hear 2 beats per second (442 – 440 = 2);
if an A tuned to 440 Hz is played along with an E that has been tuned to 658
Hz, rather than as a perfect fifth tuned to 660 Hz, one will hear 4 beats per
second (440 × 3 = 1320; 658 × 2 = 1316; 1320 – 1316 = 4).
 The circle of fifths is a graphic arrangement of consecutive fifths, generally
starting with C and progressing in clockwise order to G, D, A, E, B, F♯, etc.,
and in counterclockwise order to F, B♭, E♭, A♭, D♭, etc. When Pythagorean
intonation is represented, the circle does not close because the twelfth fifth
overlaps the starting point by a small amount termed the Pythagorean comma
– in effect, if one continues, there is a spiral of fifths. Furthermore, the
counterclockwise ratios do not coincide with the clockwise ones. When
closed systems (such as equal temperament) are represented as a circle of
fifths, comma divisions are generally written in between each of the intervals;
in open systems such as meantone, the location of the wolf fifth is usually
indicated.
Circulating temperament is a temperament in which all keys are playable
in which the fifths (and perhaps other intervals) are all tempered to about the
same degree. Though equal temperament fulfills this description, it is not
classified as a circulating temperament.
Closed system is a tuning method consisting of a circle of reasonably
acceptable fifths, such as equal temperament.
Dieses. There are two dieses: the so-called “greater diesis” is the difference
between four pure minor thirds and an octave (which is found between
chromatic pairs, such as C♯ and D♭), and has the ratio of 648:625; the “lesser
diesis,” which is the difference between three pure major thirds and an
octave, has the ratio of 128:125.
Ditonic comma (also called the Pythagorean comma): this is the interval
associated with equal temperament that represents the amount twelve
successive perfect fifths exceeds twelve perfect octaves. Its ratio is (3/2)12: 27
or 531441:524288. The ditonic comma is also the interval between
enharmonic notes in Pythagorean intonation.
Enharmonics are tones in a scale that can be expressed as different notes,
such as C♯ and D♭ and F♯ and G♭ in equal temperament.
Equal-beating temperament is a modern English term derived from the
German gleichschwebende Temperature, as used by Marpurg and others.
Equal-beating temperaments employ two or more intervals that beat at a
similar rate. Though a boon to tuners, fifths (for example) that beat at the
same rate may not be proportionally equivalent in size. The term “equal
beating” should not be confused with “equal temperament.” (See schweben.)
Equal temperament is a closed, regular system in which all of the
semitones are equal and form the ratio of :1. Equal temperament can also
be constructed by diminishing each pure fifth by one-twelfth of a ditonic
comma.
An irregular system is a tuning method in which the fifths vary in size, for
example tunings that include pure as well as tempered fifths.
A just interval (see Intervals) is a beatless interval.
Just intonation is a scale made up of just, or pure, intervals.
Meantone is the mean between the major (9:8) and minor (10:9) whole
tone of just intonation. Because a major and minor tone make up a pure major
third, the meantone can also be expressed as :1.
Meantone temperament is an open, regular system of tuning based upon
divisions of the syntonic comma. Strict quarter-comma meantone
temperament divides the syntonic comma equally among four consecutive
fifths so that the resulting third is pure.
Multiple division involves the division of the octave into more than twelve
notes in order to provide better intonation. Popular divisions have included
19 (Zarlino, 1558), 31 (Vicentino,1555), and 43 (Sauveur, 1701).
An open system is a temperament that does not involve a circle of fifths
and that generally has at least one bad or “wolf” fifth, such as meantone.
A Pythagorean scale is constructed using only two intervals, the fifth (or
its inversion) and the octave.
A Pythagorean third forms a ratio of 81:64, which is a syntonic comma
larger than a pure, 5:4, third (5/4 x 81/80).
Ratio is the quotient of one quantity divided by another.
A regular system is a tuning method constructed with invariable fifths.
Schisma is the difference between a diatonic and syntonic comma.
Schweben is a German term that has come to mean “beating.”
A syntonic comma (also called the comma of Didymus or comma of
Ptolemy) is a small interval associated with meantone temperament that
represents the amount four successive perfect fifths exceeds two perfect
octaves plus a pure major third. Its ratio is 81:80.
Well-temperament is a modern term derived from the title of J. S. Bach's
keyboard composition Das wohltemperierte Clavier (Book I, 1722; Book II,
1738–1742), which consists of forty-eight preludes and fugues written in all
of the major and minor keys. To play this work in its entirety without
recourse to retuning certain intervals requires a closed system of tuning, such
as equal temperament or one approaching it.
A wolf fifth is found in open systems, such as meantone and Pythagorean
temperaments. In meantone temperament, the wolf is an unacceptably wide
fifth; in Pythagorean intonation, the wolf fifth is narrow. (Note that there is a
distinction between a “wolf fifth” and a “wolf tone,” the latter being an
unacceptable sympathetic vibration associated with bowed string
instruments.)

Some useful temperaments for early keyboard instruments

Pythagorean intonation was widely used in the sixteenth century. To

produce an octave that has b♭, e♭, c♯, f♯, and g♯, one can begin the
temperament octave at “a” and tune downward by perfect fifths to e♭, and
then return to “a” and tune upward by perfect fifths until one arrives at g♯
(doubling back by octaves where necessary to produce a complete octave).
This procedure will produce a wolf fifth between g♯ and e♭. By choosing
different starting or end points the wolf can be placed elsewhere, for example
b–f♯. Arnault of Zwolle employed Pythagorean intonation in the fretting of a
clavichord pictured in his manuscript written around 1440. The clavichord
depicted in the intarsia (circa 1476) decorating the studiolo of the Duke of
Montefeltro in the ducal palace in Urbino is also apparently fretted in
Pythagorean intonation with all the fictae tuned as flats (Mondino, 1994).
Quarter-comma meantone temperament is often ascribed to Pietro Aaron
(c.1480–c.1550), but the tuning instructions he provides in his Thoscanello
(1523) do not unequivocally conform to that temperament (Lindley, 1974).
The temperament he describes is tuned in three sections, but only the first is
consistent with the setting of strict quarter-comma meantone. In the first
section he tempers four fifths (C–G–D–A–E) so that they produce a pure
third. However, he begins the second section by stating that the thirds and
sixths are “blunted or diminished” (spuntate overo diminute), which is not
consistent with strict quarter-comma meantone. In this section he also
indicates that the fifths B♭ below F and E♭ below B♭ are to be tuned similarly
to the way F was tuned to C, though he never indicates how that fifth was
tempered. In the final section, he specifies that C♯ is tuned so that it makes a
major third with the A below it and a minor third with the E above it, and that
F♯ and the remaining notes are placed similarly. Strict quarter-comma
meantone is described later in the sixteenth century by Zarlino (1571), and it
is suitable for sixteenth- and seventeenth-century keyboard music. See below
for instructions for calculating the beat rates of the tempered fifths.

Quarter-comma meantone temperament

Tune c1.
Tune c1–g1 narrowed by ¼-syntonic comma.
Tune g1–d2 narrowed by ¼-syntonic comma.
Tune d2–d1 pure.
Tune d1–a1 narrowed by ¼-syntonic comma.
Tune a1–a pure.
Tune a–e1 narrowed by ¼-syntonic comma.
c1–e1 should be a pure, beatless third. If not, readjust the fifths.
Then tune pure major thirds upwards from the above tuned notes:

d1–f♯1
g1–b1
a–c♯1

and pure major thirds downwards:

d1–b♭ (and then up an octave)

 a1–f1

Finally, set e♭1 down from g1 or make it d♯1 by tuning up from b. Set g♯1 up
from e1, or make it a♭1 by tuning down from c2.
To calculate the frequencies of the tempered fifths of quarter-comma
meantone, multiply the frequency of the lower note by 3/2 to find the perfect
fifth and then temper by dividing by the fourth root of the syntonic comma,
(approx. 1.00311). To determine the beat rate, multiply the frequency
of the lower note by 3 and that of the upper note by 2 to determine the
coincident harmonic. The beat rate is the difference between these products.
The thirds (which can be ascertained by multiplying the lower note by 5/4)
that are tuned in this temperament are pure and do not beat. To compute the
frequencies of fifths tempered by - or -syntonic comma, temper by a
factor of (approx. 1.00248) or (approx. 1.00207),
respectively. Fifth-comma temperament is of interest because the tempered
thirds beat at the same rate as the fifths in the root position major triad.
Unfortunately, there is still a wolf fifth.
Around 1775 the composer Padre Antonio Soler attempted to correct
certain well-known problems associated with quarter-comma meantone in his
undated manuscript entitled Theorica, y practica del temple para los
organos, y claves. These problems included the inability to modulate to
distant keys and the presence of the so-called “wolf fifth,” which is the wide
fifth that remains after the others have been tempered. He proposed adjusting
the chromatic semitones by making them the geometric means of their
adjacent notes (for example, F♯ becomes the geometric mean of F and G) –
much in the way that quarter-comma meantone equalizes the size of the two
whole tones forming the third. Soler devised a special monochord fitted with
a fretting template that assisted in tuning this temperament (Figures 36a and
36b; Pollens, 2014).
Figure 36a Padre Antonio Soler's monochord fretting template from his
Theorica y Practica del Temple para los Organos y Claves.

Figure 36b The author using a monochord fitted with Soler's fretting
template to assist in the tuning of a harpsichord.

Jean Denis
 The following tuning instructions described by Jean Denis in his Traité de
l'accord de l'espinette (1650) are thought by some (Panetta, 1987) to
represent meantone temperament and by others (Barbour, 1972) to be
consistent with equal temperament:

f –f1 perfect
f –c1 narrow, tolerable
c1–c perfect
c–g narrow, tolerable
g–g1 perfect
g–d1 narrow, tolerable
f1–b♭ narrow, tolerable
b♭–d1–f1 test; thirds should sound “good”
d1–d perfect
d–a narrow, tolerable
a–a1 perfect
f–a–c1 test
a–e1 narrow, tolerable
e1–e perfect
c–g–c1–e1–g1 test
e–b narrow, tolerable
g–b–d1–g1 test
b–f♯1 narrow, tolerable
f♯1–f♯ perfect
f♯–c♯1 narrow, tolerable
a–c♯1–e1 test
c♯1–c♯ perfect
c♯–g♯ narrow, tolerable
e–g♯–b–e1 test
b♭–e♭ narrow, tolerable
e♭–g–b♭ test
g♯–e♭1 “fault of the tuning”

The remark “fault of the tuning” suggests this interval is the wolf tone of
meantone temperament. However, elsewhere Denis states:
 When we tune the harpsichord to perfection, we narrow all the fifths by
a bit in such a way that the fifth still seems good, even though it is not
pure. And over the sum of fifths, which are twelve in all, the others are
nothing but duplicates. They are all narrowed a bit, making it as small as
you would like. There must be twelve pitches, which is the difference
from the first to the last fifth, and all the fifths must be tempered
equally, all in like manner. (Panetta, 1987)
The tuning of all twelve fifths in this manner is consistent with equal
temperament.

Werckmeister III

This is from Andreas Werckmeister's Musicalische Temperatur (Frankfurt

and Leipzig, 1691; Figure 37). Because this system makes use of two types of
fifths (quarter-comma tempered and pure), it is referred to as an “irregular”
temperament:

c1–g1 narrowed by ¼ comma

g1–d2 narrowed by ¼ comma
d2–d1
d1–a1 narrowed by ¼ comma
a1–e2 perfect fifth
e2–e1
e1–b1 perfect fifth
b1–f♯2 narrowed by ¼ comma
f♯2–f♯1
f♯1–c♯2 pure
c♯2–c♯1
c♯1–g♯1 pure
g♯1–d♯2 pure
d♯2–d♯1
d♯1–a♯1 pure
a♯1–f2 pure
f2–f1
 f1–c2 pure.

Figure 37 Werckmeister III tuning instructions from Andreas

Werckmeister, Musicalische Temperatur (Frankfurt and Leipzig, 1691).

Friedrich Wilhelm Marpurg

Friedrich Wilhelm Marpurg's Versuch über die musikalische Temperatur

(Breslau, 1776) describes an irregular temperament, referred to as Marpurg's
Temperament I, which consists of nine pure fifths and three fifths tempered
by 1/3 ditonic comma. Below is an adaptation of a tuning sequence proposed
by John W. Link (1969):

a1–a perfect octave

a–e1 perfect 5th
e1–b1 perfect 5th
b1–b perfect octave
b– f♯1 perfect 5th
f♯1–c♯2 tempered 1/3 ditonic comma
c♯2–c♯1 perfect octave
c♯1–g♯1 perfect 5th
g♯1–d♯2 perfect 5th
 d♯2–d♯1 perfect octave
d♯1–a♯1 perfect 5th
a♯1– f2 tempered 1/3 ditonic comma
f2–f1 perfect 5th
f1–c2 perfect 5th
c2–c1 perfect octave
c1–g1 perfect 5th
g1–d2 perfect 5th
d2–d1 perfect octave
d1–a1 tempered 1/3 ditonic comma

Because a1 is the starting point, it is only necessary to temper two of the

fifths, making this a very easy temperament to set. The beat rates of the
tempered fifths can be calculated by dividing the pitch of a perfect fifth by
the third root of the ditonic comma, 531441/524288 (1.004527228) and then
comparing their coincident overtones. For example, if the temperament series
were to begin at A=440 Hz, f♯1 would be tuned to 371.25 Hz, and c♯2
tempered by 1/3 comma would beat at 5.02 Hz. The major thirds of the
temperament are equivalent to those in equal temperament.

Kirnberger III (1779)

This is another irregular temperament.

Tune c1 to fork. Then:

c1–e1 pure major third

c1–g1 narrowed by ¼-syntonic comma
g1–d2 narrowed by ¼-syntonic comma
d2–d1 octave
d1–a1 narrowed by ¼-syntonic comma
a1 to e2 narrowed by ¼-syntonic comma
e1 to e2 check that this is a perfect octave; if not, adjust the previously
tuned fifths.
Then tune the following perfect fifths and octaves:

c1–f
f–f1
f1–b♭
b♭–b♭1
b♭1–e♭1
e♭1–a♭
a♭–a♭1
a♭1–d♭1
e1–b1
b1–b
b–f♯1.

Young's temperament

A temperament by Thomas Young as he himself expresses it in the

Philosophical Transactions of the Royal Society (London, 1800):

Make C to E too sharp by a quarter comma, E:G♯ and A♭:C equal, and
F♯:A♯ too sharp by a quarter comma. Major thirds of all intermediate
keys should be tuned more or less perfect, as they approach more or less
to C in order of modulation. The fifths are perfect enough in every
system. In practice, nearly the same effect may be very simply produced
by tuning from C to F, B♭, E♭, G♯, C♯, F♯ six perfect fourths and C, G,
D, A, E, B, F♯ six equally imperfect fifths. If the unavoidable
imperfections of the fourths be such as to incline them to sharpness, the
temperament will approach more nearly to equality, which is preferable
to an inaccuracy on the other side.

Equal temperament

When tuning the temperament octave (for example c1–c2, though some
piano tuners prefer f–f1), one method is to start at c1 and tune a sequence of
upward fifths and downward fourths, very slightly narrowing each of the
fifths and widening each the fourths as one proceeds, tuning back down an
octave when one reaches c♯2, then continuing with upward fifths and
downward fourths until finally reaching c2, which, if one has tempered all of
the fourths and fifths properly, should make a perfect octave with c1. To do
an accurate job, one must keep track of the beat rates (see below).
Beats between intervals occur when there are pitch discrepancies between
the coincident overtones of the two notes. For example, to calculate the equal
temperament beat rate for c1–g1, look up the pitches in the equal temperament
frequency table and multiply the lower note by 3 (261.6256 x 3) and from
this subtract the upper note multiplied by 2 (391.9954 × 2), or 784.8768 –
783.9908 = 0.886. The resultant beat rate is 0.886 beats per second. This rate
is difficult to ascertain by ear, but by multiplying by 5, we get 4.43, or about
4½ beats in 5 seconds, which is easier to estimate. For the fourths, multiply
the lower note by 4 and the upper note by 3 (the inverse of the ratio of the
fourth) and subtract the lower frequency from the higher to find the beat rate.
The beat rates for tuning the temperament octave c1–c2 at A=440 Hz are as
follows:

Approx. beats/5
Tune Interval Beats/second
seconds

c1– g1 5th up 0.886 4½

g1–d1 4th down 1.327 6½

d1–a1 5th up 0.9943 5

a1–e1 4th down 1.4897 7½

e1–b1 5th up 1.1161 5½

b1–f♯1 4th down 1.6722 8½

f♯1–c♯2 5th up 1.2527 6½

 c♯2–c♯1 octave 0 0
down

c♯1– 5th up 0.9385 4½

g♯1

g♯1– 4th down 1.4062 7

d♯1

d♯1– 5th up 1.0534 5½

a♯1

a♯1–f1 4th down 1.5784 8

f1–c2 5th up 1.1824 6

c2–c1 octave 0 0
down

Note that as one moves up the temperament octave, the beat rates for the
fifths and fourths steadily increase. The major thirds in equal temperament
beat very fast, about 10.38 beats per second for c1–e1, again gradually
increasing as one moves up the octave. Though it is impossible to count such
fast beats by ear, one should learn to sense this rate and ascertain whether it
gradually increases as one ascends the temperament octave third by third.
When setting the temperament octave, set the beat rates as accurately as
possible in order to avoid a cumulative error, which will only become evident
when one tunes c2 and checks it against c1.
In his book Tuning (1991), Owen Jorgensen asserts that equal temperament
was not accurately tuned until the twentieth century. He makes this claim
because precise beat rates were never indicated in any instructions for setting
that temperament until the publication of William Braid White's Modern
Piano Tuning and Allied Arts in 1917. In defense of this claim, he cites a late-
nineteenth-century survey conducted by Alexander John Ellis, who used a
device called a tonometer (which consisted of 105 tuning forks) to evaluate
the accuracy of equal temperaments tuned by noted piano tuners, including
those working for the Broadwood firm in London. Ellis discovered that none
of them succeeded in tuning the temperament accurately. The tonometer was
invented by Johann Heinrich Scheibler (1777–1837) and described by him in
a publication entitled Der physikalische und musikalische Tonmesser dated
1834. An English translation of this work was published in 1853. His original
tonometer had 52 forks, but one having 56 tuning forks tuned 4 Hz apart
came to the attention of Ellis and was described by him in a paper read before
the Society of Arts in 1880. In setting or evaluating an equally tempered
scale, Scheibler used his forks to assess the beat rates. In instances where the
beats did not coincide precisely with seconds, Scheibler used a finely
calibrated metronome to calculate the rates. Ellis indicates that sets of equally
tempered tuning forks were available at the time he delivered his paper,
which suggests that certain tuners may have been tuning equal temperament
more accurately than Jorgensen states, and well before 1917.

Adjusting the pitch of tuning forks and electronic tuning devices

Pitch is generally expressed as the vibratory frequency of the A above

middle C (A=440 Hz, 430 Hz, 415 Hz, 392 Hz, etc.); however, the
instructions for tuning various temperaments often begin the sequence with c1
(middle C on the keyboard). If one starts with the C tuning fork that comes
from an equally tempered set of forks tuned for A=440 Hz, in quarter-comma
meantone, the resultant A will be 437.4 Hz, which might present difficulties
if the keyboard instrument is to be tuned with an orchestra pitched at
precisely A=440 Hz. It is often possible to start by tuning the A at 440 Hz
and then reversing the tuning sequence to acquire C, though if one has a
favorite temperament and prefers to start at C, one can adjust the pitch of a
tuning fork to provide the correct starting point. The pitch of a fork can be
raised by filing or grinding down the tops of the tines or lowered by filing
away at the base of the tines with a round or rat-tail file. When hand-filing or
grinding a tuning fork, work slowly and allow it to return to room
temperature (or dip it in a glass of room-temperature water) before checking
its pitch with an accurate electronic tuner. To grind down the tines of forks,
the author prefers to use a 1-inch belt grinder than a grinding wheel, as it
produces flat-topped tines. For example, to determine the pitch of c1 for
quarter-comma meantone temperament that will yield an A at 415 Hz, begin
with a1 to 415 Hz, and calculate the pitch of d1 by dividing 415 Hz by 3/2 and
then tempering that fifth by multiplying by , which yields 277.53 Hz.
Then calculate g in the same fashion (185.59 Hz), and c (124.11 Hz). Finally,
double c to find c1 (248.23). An equally tempered b tuning fork (which might
ordinarily be used to commence tuning at c1 when A=415 Hz) has a
frequency of 246.95 Hz. The pitch of this fork can be raised to 248.23 Hz by
filing or grinding the tops of the tines. If one uses an electronic tuning device,
consult the Equal temperament frequency table to determine a cents value to
offset the frequency of C. Some new tuning “apps” allow A to be set
independently of the temperament, a great convenience.
When tempering the notes in an octave, keep in mind that the general pitch
will affect the beat rates – in other words, the beat rates for the notes in the
temperament octave will not be the same at A=415 as at A=440.

Keyboard temperaments and bowed stringed instruments

Some musicians and theorists have proposed retuning the strings of

members of the violin family so that they will stand in better tune with
keyboard instrument temperaments (see Ritchie, 2012). One might question
the advisability of violating the perfect fifths of a violin's open strings for the
sake of a harpsichord, for example, especially when it so often serves a
subsidiary continuo role and frankly can barely be heard even in relatively
small ensembles. Indeed, Scipione Maffei remarks in the Giornale dei
letterati d'Italia (Venice, 1711) that “the violin, not having keys, can find
everything in its place, and in whatever key make heard perfect notes” and
that harpsichords and theorbos cannot play in tune with the violin, but that
“when used in concert, the ear does not notice it.” In his Versuch einer
Anweisung die Flöte traversiere zu spielen (Berlin, 1752), Johann Joachim
Quantz expresses conflicting views regarding the pairing of violin and
keyboard: he recommends that when harpsichordists accompany instruments
that produce “true ratios” (such as the violin), the harpsichordist should omit
notes that would produce tempered major and minor thirds, or “hide them in
a middle or lower part,” where they would be less likely to “displease the
ear,” though elsewhere he suggests that violinists “will not do badly to follow
the rule that must be observed in tuning the keyboard, namely, that the fifths
must be tuned a little on the flat side rather than quite true or a little sharp, as
is usually the case, so that the open strings will all agree with the keyboard”
(translations adapted from Edward R. Reilly, Johann Joachim Quantz: On
Playing the Flute, 1966). Unfortunately, Quantz does not specify the
keyboard temperament, but presumably was indicating one of the period's so-
called “well-temperaments” that feature fifths that were not as narrow as
those used in strict quarter-comma meantone. However, the widespread use
by many of today's early music ensembles of “Vallotti” temperament (which
includes a sequence of sixth-comma tempered fifths between F–C–G–D–A–
E–B – while the fifths bridging the “black keys” are tuned beatless) does
present difficulties for violinists who are accustomed to tuning their strings in
perfect fifths. Requiring them to adjust their open strings to Vallotti or to
some form of meantone may explain why the string sections of many early
music groups often sound so painfully out of tune.
Many of today's keyboard tuners, technicians, and players rely upon
electronic tuning devices when tuning their instruments. While these devices
may be of some assistance in setting a variety of historic temperaments, I
would suggest that players, in particular, learn to tune by ear, and that they
familiarize themselves with a few basic temperaments. One problem with
relying upon electronic tuners is that the nature of the temperaments remains
a mystery, and many players do not make informed decisions about which of
the electronically preset temperaments to use. For example, when consorting
with members of the violin family, a keyboard player might wish to select a
temperament that favors the perfect fifths connecting C–G–D–A–E (the
popular “Vallotti” temperament found on many electronic tuners does not).
When tuning by ear, precise beat counting is not always feasible or even
necessary – in fact, very few early theorists mention the use of beat rates in
setting temperaments. Exceptions include Dom Bédos (1770), who provided
beat rates for setting meantone temperament, and Sauveur (1701), who used
them to calculate the frequencies produced by organ pipes. Nevertheless,
tuners should at least be aware of the theoretical beat rates. For example, one
can test equal temperament by running up the temperament octave by major
thirds and making sure that the beats progress nicely. Often the extreme bass
and treble of modern pianos and even late fortepianos just do not sound in
tune (due to the inharmonicity of overtones) without a modicum of octave
stretching. While many electronic tuning devices do not support the
technique of octave stretching, or do so in a perfunctory manner, it is easily
applied when tuning by ear. The ear, after all, is the ultimate arbiter, so it
makes sense to tune by ear. Furthermore, one should always have a tuning
fork at hand in the event an electronic tuner's batteries die.

References
Pietro Aaron, Thoscanello de la Musica (Venice, 1523).
Murray Barbour, Tuning and Temperament (New York, 1972).
François Bédos de Celles, L'art du facteur d'orgues, 4 vols. (Paris, 1766–
1778).
Jean Denis, Traité de l'accord de l'espinette (Paris, 1650; facsimile, New
York, 1969).
Alexander J. Ellis and Arthur Mendel, Studies in the History of Musical Pitch
(Amsterdam, 1968).
Hermann Helmholtz, On the Sensations of Tone (London, 1885).
James Jeans, Science and Music (Cambridge, 1937).
Owen Jorgensen, The Equal-Beating Temperaments (Raleigh, 1981).
Owen Jorgensen, Tuning (East Lansing, 1991).
H. A. Kellner, The Tuning of My Harpsichord (Frankfurt, 1980).
Johann Philipp Kirnberger, Die Kunst des reinen Satzes in der Musik (Berlin,
1771–1779).
G. C. Klop, Harpsichord Tuning (Garderen, 1983).
Mark Lindley, “Early 16th-Century Keyboard Temperaments,” Musica
Disciplina 28 (1974), pp. 129–151.
Mark Lindley, Lutes, Viols, and Temperament (Cambridge, 1984).
John W. Link, Jr., Theory and Tuning: Aron's Meantone Temperament and
Marpurg's Temperament “I” (Boston, 1972). (Highly recommended, but
unfortunately out of print.)
Scipione Maffei, “Nuova invenzione d'un gravecembalo col piano e forte,”
Giornale dei letterati d'Italia (Venice, 1711), pp. 144–159.
Friedrich Wilhelm Marpurg, Versuch über die musikalische Temperatur
(Breslau, 1776).
Angelo Mondino, “The Intarsia of Urbino,” De Clavicordio (Turin, 1994),
pp. 49–55.
Claude Montal, L'art d'accorder soi-même son piano (Paris, 1836).
Vincent J. Panetta, Treatise on Harpsichord Tuning by Jean Denis
(Cambridge, 1987).
Stewart Pollens, “Soler's Temperament and His Acordante,” Proceedings of
the 13th International Symposium on Spanish Keyboard Music, FIMTE,
2012 (Mojácar-Garrucha, forthcoming).
Johann Joachim Quantz, Versuch einer Anweisung die Flöte traversiere zu
spielen (Berlin, 1752).
Stanley Ritchie, Before the Chinrest (Bloomington, 2012).
Joseph Sauveur, “Système general des intervalles des sons,” Mémoires de
l'académie royale des sciences (Paris, 1701), pp. 403–498.
Johann Heinrich Scheibler, Der physikalische und musikalische Tonmesser
(Essen, 1834).
Johann Heinrich Scheibler, An Essay on the Theory and Practice of Tuning in
General and on Schiebler's Invention of Tuning Pianofortes and Organs by
the Metronome in Particular, translated by Augustus Wehrhan (London,
1853).
Antonio Soler, Theorica, y practica del temple para los organos, y claves
(MS, circa 1775; facsimile published by Sociedad Española de
Musicologia, Madrid, 1983).
Nicola Vicentino, L'antica musica ridotta alla moderna pratica (Rome,
1555).
Andreas Werckmeister, Musikalische Temperatur (Frankfurt and Leipzig,
1691).
William Braid White, Modern Piano Tuning and Allied Arts (New York,
1917).
Gioseffo Zarlino, Istitutioni armonische (Venice, 1558).
 V
Vapor phase inhibitor (VPI or VCI)

Vapor phase inhibitors (or vapor corrosion inhibitors) are chemicals that
sublime and form a protective film on metals. Kraft or wrapping paper
impregnated with these chemicals can be used as a cover or wrapping. To
protect steel, dicyclohexylamine nitrite or ammonium nitrate are often used,
while benzotriazole can be used to protect copper, brass, and bronze.

Varnish

General

There are three basic types of classic varnish: alcohol, drying oil, and spirit
oil. Alcohol varnish consists of one or more resins dissolved in ethanol. This
type of varnish dries quickly by the evaporation of the solvent. Drying-oil
varnish consists of one or more resins dissolved in a drying oil such as
linseed, poppy seed, or walnut oil. Oil varnish often has a spirit oil (such as
turpentine oil, oil of spike lavender, or mineral spirits) added as a diluent. For
this type of varnish to dry, the diluent must evaporate and the drying oil must
oxidize. The oxidation process may be extremely slow and can take months
or even years to complete. Spirit oil varnish consists of one or more resins
dissolved in a spirit oil, such as oil of turpentine. The simplest of this type is
known as turpentine varnish, which consists of pine colophony (rosin)
dissolved in oil of turpentine. As with alcohol varnish, the spirit oil (in this
case turpentine) must evaporate for the varnish to dry.
In addition to the three basic varnishes, there are hybrid varnishes. For
example, a small amount of linseed oil added to a spirit oil varnish to toughen
or plasticize the dried film will produce a hybrid varnish. Similarly, the
addition of a balsam, such as Venice or Strasbourg turpentine, to an alcohol
varnish creates a hybrid due to the spirit oil component of the balsam.
Though the great violin makers of the past did not record their varnish
formulas for posterity, a few lute and violin varnish formulas from the
sixteenth through the eighteenth centuries have been preserved in general
writings on varnish making, and it is worth the violin restorer's while to
become acquainted with them (see below).
Each of the basic varnish types remains sensitive to different types of
solvent. Alcohol varnishes tend to remain sensitive to alcohol, though some
ingredients may cross-link (link polymer chains) over time, a process that
gradually reduces their solubility in alcohol. Shellac is an example of a resin
that crosslinks. Drying-oil varnishes also tend to become less soluble as they
age. Their original diluents, such as turpentine and mineral spirits, may
readily soften or remove them before they have completely dried, but mature,
oxidized oil–varnish films are generally resistant to these solvents and may
require more polar solvents, such as acetone, to dissolve them. In their natural
state, amber and certain fossilized copals are intractable resins that are
resistant to most solvents and oils. To make them soluble, they first must be
“run” (heated to about 572°F [300°C]). After they have been heat-treated in
this way, amber and the fossilized copals are readily dissolved in turpentine
or mineral spirits.

Varnish analysis

A wide variety of analytical techniques can be used to analyze varnish.

Some of the more popular methods include thin-layer, liquid, and high-
performance liquid chromatography, gas chromatography/mass spectrometry,
Raman spectroscopy, Fourier-transform infrared spectroscopy, energy-
dispersive X-ray fluorescence spectroscopy, visible and ultraviolet light
microscopy, polarizing microscopy, and scanning electron microscopy (see
Analysis of materials).

Examination of varnished surfaces by ultraviolet fluorescence and infrared

Musical instrument restorers working outside of major museums may not

have access to the analytical equipment mentioned above to routinely analyze
the varnishes they must clean, revive, or retouch and will have to rely upon
their knowledge and experience as well as a few simple tests to characterize
varnish. One technique is ultraviolet (UV) fluorescence, which is especially
useful in identifying retouched and revarnished areas. Long-wavelength
(approximately 365 nanometers) ultraviolet radiation, which is invisible to
the human eye, causes certain resins, pigments, and dyes to fluoresce in the
visible part of the spectrum. For example, unbleached shellac fluoresces
bright orange; dried linseed oil films produce a yellowish fluorescence; oil
varnishes fluoresce a milky or yellowish color; madder lake fluoresces a dull
orange, while alizarin does not fluoresce; cochineal, carmine, and kermes
lakes fluoresce bright pink; while lac lake (a purplish-red pigment derived
from shellac) does not produce a distinctive fluorescence. Certain driers used
in oil varnishes, as well as metallic oxide pigments used in both oil and spirit
varnishes, absorb UV and appear black. Old animal-hide glue sizes and
grounds fluoresce pale blue, whereas relatively recent applications of animal-
hide glue do not fluoresce. Retouching and revarnishing are invariably
carried out with materials that differ from those used in the original varnish
formulation, and as a result, they often become visible under ultraviolet. It
should be noted that media and colorants cannot be reliably identified by UV
fluorescence because many materials produce similar fluorescence, and
fluorescence colors often change over time.
When UV lights (sometimes referred to as “black lights” or “Wood's
lamps”) are used for viewing fluorescence, their bulbs or tubes should be
mounted in polished metal reflectors rather than in plastic or painted ones, as
the latter may fluoresce sufficient visible light to mask the light fluoresced by
the object under examination. The better grades of UV lights use an exciter
filter (such as a glass Wratten 18A) in front of the mercury vapor lamp or
UV-emitting fluorescent tube to absorb the visible component. UV
fluorescent tubes with integral visible light filters tend to be less effective in
blocking visible light than those placed in housings that are fitted with
supplemental UV filters. Viewing must take place in a completely darkened
room with the ultraviolet light as the only source of illumination. One should
view through UV-absorbing glasses or a UV barrier filter (such as a Wratten
2A) to protect one's eyes from the harmful UV rays. When viewing through a
magnifying lens or stereo microscope, place a barrier filter under the lens or
objective to prevent its optical elements and their cements from fluorescing
and clouding the image.
Infrared (IR) imaging (also called infrared reflectography) may also be
useful for examining varnished and painted surfaces. Special IR viewing and
imaging devices are available, though they are rather expensive. For
information on UV fluorescence and IR photography see Inscriptions, faded.

Microscopy

A conventional light microscope is a powerful tool for examining varnish.

If pigments have been intermixed with varnish, they can often be identified
by their optical properties (McCrone, McCrone, and Delly, 2002). When a
microscope is equipped for epi-fluorescence viewing, varnish layers may be
revealed with greater clarity than with conventional light sources (Figure 38)
and fluorescent stains can be employed to determine the presence of certain
substances. For example, a chip of varnish stained with Sudan Black that
displays a bright orange fluorescence marbleized with black would be
consistent with French polish made with unbleached shellac that was
insufficiently “oiled out” in the final stage of polishing (the non-fluorescing
Sudan Black stains triglycerides and lipids).
Figure 38 Cross section of Stradivari varnish (1734) showing stratification
revealed by epi-fluorescence microscope.

Changes in varnish due to aging

All varnishes undergo changes as they age. Many of the dyes and unfixed
botanical extracts that are used to create yellows, oranges, and reds fade if
exposed to moderate amounts of light (see Illumination levels). Commonly
used botanical colorants include alkanet root, aloes, annatto seed, brazil
wood, campeachy wood, dragon's blood, fustic, gamboge, indigo, logwood,
madder root, orchil, buckthorn berries, quercitron, safflower, saffron,
sandalwood, turmeric (curcuma), and weld. Insect-derived colorants, such as
those found in seedlac, shellac, kermes, and cochineal, are also fugitive.
Though most natural colorants fade, natural resins tend to develop a warm
hue that may over time compensate for the faded colorants. Even the clearest
natural resins like mastic and dammar eventually turn yellow, then orange
and finally brown. This is a gradual process brought about by heat, light, and
oxygen. In addition to color change, a general physical breakdown of varnish
may also occur, resulting in the development of microscopic fissures (causing
a matte appearance), crazing, craquelure, alligatoring, wrinkling, tenting, and
flaking. Aging also causes the refractive index of drying oils to increase (that
of linseed oil increases from about 1.48 to 1.57), causing some pigments
bound within the oil to appear more transparent – an effect that is impossible
to replicate with fresh materials.

Some traditional varnish formulas for musical instruments (see also Stradivari's varnish)

Eastlake (1847) and Merrifield (1849) provide numerous recipes for

varnishes dating from the twelfth century onward. While these works are
principally concerned with the art of painting, they do include varnish
formulas specifically intended for musical instruments. The earliest varnishes
for lutes cited in Merrifield's work are from an early sixteenth-century
manuscript entitled Secreti Diversi (also known as the Marciana Manuscript)
that is preserved in the Library of San Marco in Venice. The first varnish
(number 399 in the manuscript) is described as follows:
 Item, a varnish that spreads like oil, dries quickly, and is very lustrous
and beautiful, appearing like a glass mirror, and which is admirable for
adhering firmly and for varnishing lutes and similar things. Take 1
pound of linseed oil, boil it in the proper manner in a clean glazed
pipkin, add to it half a pound of well-pulverized clear and fine Greek
pitch, and stir and incorporate the whole over a slow fire; then add half a
pound of powdered mastic, and the moment you have done so, withdraw
the pipkin gradually from the fire, because it swells up, and incorporate
the ingredients thoroughly; then replace the pipkin on the fire, and keep
it there until everything is well dissolved and mixed, then some burnt
and pounded roche alum of the size of a nut should be added and mixed
until that also is entirely dissolved and incorporated. Then take the
varnish off the fire and strain it through an old linen cloth. Your varnish
is then made, and it will be found to be beautiful varnish for wood, iron,
paper, leather, and all kinds of painting and works, and for withstanding
water.

The second varnish (number 400 in the manuscript) is described as follows:

Item, a most excellent varnish of mastic for lutes, leather, panels, cloth,
wood, and pasteboard. Take 3 ounces of strained and clear linseed oil
and boil it. Then take half an ounce of mastic pounded and ground, and
add it gradually to the oil, mixing it in such a manner that it may be
entirely dissolved and incorporated with the oil, and that it be properly
evaporated and made into a varnish. Then put in a little pulverized roche
alum at discretion, but sufficient to affect all the varnish; keep it over the
fire until it is entirely dissolved and incorporated with the varnish and
evaporated, after which you may take it off the fire, and strain it through
an old and good linen cloth, then it is finished. But observe that
everything should be done over a charcoal fire and with great care.

The recipe for another varnish used for lutes and other musical instrument
can be found in the early seventeenth-century De Mayerne Manuscript
preserved in the British Library. (Theodore de Mayerne was a prominent
physician and chemist whose knowledge of artists' materials rendered him a
valuable consultant to many important painters, including Rubens and Van
Dyck.)

The true varnish of lutes and viols. Recipe: Take carabé [amber] that is
yellow verging on a reddish color, as much as you like, place it in a
lead-glazed pot on a slow charcoal fire, and stir with an iron rod. It is
heated until it darkens and resembles colophony. The molten resin is
then poured upon paper or onto a marble stone.
To degrease [desgraisser] the oil, filter it through pure linen and place it
in a glazed pot on a fire. Let it boil and skim well. Take a chicken's
feather and dip it into the oil. If it burns, then the oil is not ready.
Continue to boil the oil until the feather does not burn. Filter through
linen.
To prepare the varnish, take a London pint of a chopine de Paris of the
above oil, carabé, prepared as above, pulverized, about 6 ounces, place
on a small fire and stir until all is dissolved. If the varnish is too liquid,
adjust the amount of carabé; if it is too thick, adjust the amount of oil.
The varnish is applied cold and dries with the sun. The varnish is of a
good consistency if it passes through a linen cloth while hot.

Several violin and lute varnish formulas can be found in Johann Kunckel's
Der neu Kunst- und Werck-Schul published in Nuremberg in 1707. “A white,
good Venetian violin varnish” (formula number LXXXII) is prepared as
follows:

Take one pound clear linseed oil and let it boil in a kettle. Take a full
vessel of Beern or Agtstein [amber]. Put in 2 Loth [1 Loth = 1/32 pound]
of cream of tartar and place over a strong charcoal fire. Stir with a hot
glowing iron until fully melted and then pour in the hot oil and stir. Let
cool a little and then add 2 Loth of Silberglett [litharge, or white lead
oxide] and 2 or 3 Loth of Postolin [ground porcelain] of the best and
cleanest variety. Filter through a cloth. The older the varnish the better it
is.

Another varnish from Der neu Kunst- und Werck-Schul (number XC) is
described as “a beautiful light or golden varnish for lutes and violins by
Melchior Schmieden.” It is prepared as follows:

Sandarac, one and a half Loth; fine turpentine that is very light in color,
3 Loth, mastic and Agtstein [amber] each one quintlein [1 quintlein = ¼
Loth]. Mix together in an earthenware container. Filter it through a clean
cloth and add an equal volume of oil of turpentine. Place on a fire. To
prevent evaporation, place in a glass container. If it becomes too thick,
then one can add a little oil of turpentine.

Another formula from the same book (number CXXVIII), termed “a beautiful
glossy violin and wood varnish,” is prepared as follows:

Take the best and lightest gum sandarac, one pound, and Agtstein
[amber], 8 Loth, that have been prepared by steeping in strong lye. Pour
over seven times the quantity of the best rectified spirits of wine. Place
in a double alembic and agitate for two hours. Then place for eight to ten
days in hot ashes or in the hot summer sun. Agitate frequently. The
varnish should develop the color and viscosity of old linseed oil. Pour
through a clean cloth and place in a clean glass with a narrow neck.

A formula (number CLXXVI) for making “an excellent violin and lute
varnish” is prepared as follows:

Take a pound of good Agt- or Pimsen-Stein [amber] and place in a tall

earthen pot. Pour in 2 Loth of good cream of tartar. Place together on a
fire of linden charcoal and blow on it so that it becomes higher and
stronger. Heat an iron rod until it glows and stir the Pimsen-Stein and
the cream of tartar together until the Bimsen-Stein [sic] is melted. In the
meantime, place a pound of linseed oil in a kettle on a tripod over a
charcoal fire. When it boils, add it to the above molten Agt- or Pimsen-
Stein. To make this varnish quick drying, 2 Loth of litharge or 3 Loth of
Porcellin Mehl [pulverized porcelain] can be added.

Kunckel's varnish number XXIV is said to derive from an important violin

maker and to be appropriate for violins and lutes:
 To make this correctly, one must use three glasses. In the first glass one
places gummi-lacca [shellac], 8 Loth, sandarac, 3 or 4 Loth, finely
pulverized, and add 4 fingers of the strongest spirits of wine. Press
through a linen cloth. Place in a quiet place for several days. Then pour
off the clear varnish that rises to the top into another glass. In yet
another glass, dissolve 5 Loth of clean dragon's blood and 3 Loth of
Beern-Wurzel [limonium, or Been rubrum officinarum]. In the third
glass dissolve 3 Loth of colophony, 2 Loth of Aloes succotrini, and 3
Loth Orlenii [annatto]. When all the color is extracted, pour the glasses
together, cover tightly, and let it stand in a quiet place for 8 days. Then
pour off the clean part and let it pass through a clean piece of cloth. If it
is too thin, let some of the alcohol evaporate.

Varnish number 100 is referred to as a “beautiful wood and violin varnish.”

To make it:

one pound of light linseed oil is left to stand for 24 hours in rain water.
Two Loth of common turpentine are heated until it will burn a feather
that is thrust into it. Take 2 Loth of pulverized Agtstein [amber] that has
also been left to stand in rain water, heat it until it is molten, and the
varnish is thick enough.

Varnish number 109, again from the same source, is entitled a “beautiful
violin varnish.” It is made as follows:

Take linseed oil, 6 pounds, colophony or weiss harz, 12 Loth; Aloepatic

[aloe], 3 Loth; dragon's blood, 2 Loth; Glet [litharge] and umber each 8
Loth; mastic, 6 Loth; Mennge [minium or red lead] a half pound. Cook
or boil softly together. If it overflows when it boils, remove it from the
fire. Let it cool and add 6 Loth of turpentine oil so that it will flow. Then
pass through a wool cloth. Be careful of the fire.

Jean Felix Watin's L'Art du peintre, doreur, vernisseur went through

numerous eighteenth-century editions (Paris, 1733, 1755, 1772; Liège, 1778).
One varnish is entitled “Pour les Violons & autres Instruments de Musique.”
It is prepared in the following manner:

Put into a pint of alcohol 4 ounces of sandarac, 2 ounces of gum lac

[shellac], 2 ounces of mastic in tears, and 1 ounce of gum elemi. Place
on a low heat, and when it boils, incorporate 2 ounces of térébenthine.
An instrument that will be handled requires a hard varnish; as a
consequence, one uses a smaller amount of gum lac in grains as a large
quantity results in a powdery [farineux] consistency. Using less
térébenthine, it can take the heat of the hands; gum elemi makes it
harder and substitutes for the térébenthine, which is in a smaller
quantity.

Watin states that térébenthine is a “viscous, sticky, resonous, clear, and

transparent liquid that naturally flows from an incision in the larch, the
térébenthine tree, pine, fir, etc. and other conifers. In Watin's time, the term
“turpentine” referred either to the crude balsam exuded by trees or to the
essential oil that was distilled from the balsam, which left colophony as a
residue. This has led to some confusion among certain modern-day violin-
varnish makers who fail to recognize that the term “turpentine” can refer to
the crude balsam as well as to the essential oil distilled from it, and have
consequently gone to extraordinary lengths to convert oil of turpentine into a
solid resin in making their violin varnish (Fulton, 1988). If Watin's
térébenthine was a viscous material exuded from the larch tree, it would be
what we today call Venice turpentine. The above formula given by Watin is
also found in the anonymous Praktisches Handbuch für Mahler und Lackirer
published in Graz in 1803.
Johann Melchior Cröker's Der wohl anführende Mahler published in Jena
in 1778 describes a ground or sealer that was used under the varnish by a
“violin maker who is a very important man in his profession.” It is made in
the following way:

Gum arabic is dissolved in water. This is mixed with oil and then
cooked until the water has boiled away. (The violin maker) says that this
gives the varnish a beautiful gloss.
Johann Christian Müller provides two violin varnish formulas in his
Praktische Anweisung zum lakkiren published in Leipzig, 1801. The first is as
follows:

One takes gum copal, as much as one wants, and melts it over a fire in a
clean ladle, and then pour it out so that the unclean comes out, and the
copal becomes beautifully clear. Then melt colophonium in a ladle, and
decant it in the same way so that it becomes clean. Then take 1 part of
the copal and 1 part of the colophonium and melt them together in a
ladle over a fire. After this procedure, spoon out the colophonium and
copal; this mass will then easily go into alcohol solution, and will
varnish light and clear.

Müller describes another “three-glass” varnish that is similar to the one

presented above in Der Neu Kunst- und Werck-Schul. It is prepared as
follows:

To make this varnish well and properly, it is best to use three glass
containers. In the first vial, put the most beautiful and finest gum lac
[shellac or possibly seedlac] in grains, 8 Loth, clean sandarac, 6 or 8
Loth, pulverize both together so that it is very fine and pour in 4 fingers
high of the best rectified spirits of wine. Then seal the glass, mix well
for an hour, put the glass for a while in a slightly warm place so that the
gum lac dissolves and the spirit of wine becomes reddish, just like thick
brown beer.
In another vial, take the finest dragon's blood in tears, 10 Loth, red
Beenwürzel 6 Loth, mix well, pour in the best spirits of wine 4 good
fingers high, seal, and leave in a slightly warm place. The spirits of wine
will become colored a deep red, like blood or beautiful red wine.
In the third vial, put beautiful colophonium, 6 Loth, gummi gutt, 4 Loth,
curcume, 2 Loth, mix well together, and pour in the best spirits of wine.
Seal up the glass, and keep slightly warm until a beautiful tincture is
extracted that is golden yellow.
When one needs the varnish, then pour together all the solutions, warm
 slightly, and strain through a linen cloth into a clean glass. This is good
not only for beautiful woodwork, but also for violins, citterns, and lutes.

The Varnisher's Guide by Pierre François Tingry, published in London in

1832, describes the preparation of a varnish for violins and other stringed
instruments, but also indicates that it can be used for furniture of plum wood,
mahogany, and rosewood. The recipe is as follows:

Take gum sandarac, 4 ounces, seedlac, 2 ounces, gum mastic and gum
benzoin, 1 ounce each, pounded glass, 4 ounces, Venice turpentine, 2
ounces, and rectified spirit of wine, 32 ounces. The gum sandarac and
lac render this varnish durable. It may be colored with a little saffron or
dragon's blood.

The following is a varnish formula that is said to have been found in a

notebook written by the violin maker Giacinto Santagiuliana (Vicenza, 1770–
1820), which was handed down to Leandro Bisiach and divulged to Rembert
Wurlitzer (Pollens, 1986):

Sandarac 32 g

Shellac 16 g

Mastic 8g

Ground glass 32 g

Alcohol 250 ml

Elemi 8g

Allow it to stand in the sun for 24 hours, then heat over a low flame to fully
fuse the gums.
The apocryphal “1704” varnish, said to have been used by Stradivari,
contains the following ingredients:
 Seedlac 45 g

Spike oil 9 ml

Elemi 7.5 g

Alcohol 180 g

Another varnish said to have been used by Stradivari and conveyed to

Count Ignazio Alessandro Cozio di Salabue by Father Gaetano Persico of
Cremona is made as follows:

Gum lac 4 oz

Sandarac 2 oz

Mastic 2 oz

Dragon's blood 40 [grains?]

Saffron 1/2 dram

Alcohol 1 pint

Heat until all the resins have dissolved, then add:

Venetian turpentine 4 oz

J. C. Maugin's “orange violin varnish,” from among the many recipes

given in his Manuel du Luthier (Paris, 1869), is made as follows:
In 625 g alcohol, place

Curcuma 23 g
 Saffron 6.5 g

Let sit for 24 hours

Pulverize the following:

Gomme gutte 23 g

Sandarac 60 g

Elemi 60 g

Dragon's blood 30 g

Seedlac 30 g

Dissolve these in the colored alcohol above.

“Varnish of Venice” from André-Jacques Roubo's L'Art du menuisier
(1769–1774) uses the following ingredients:

Sandarac 5 oz

Mastic 2 oz

Elemi 1 oz

Oil of spike 1 oz

Alcohol 1 pint

Heat in water bath, filter.

Details of a “white varnish” for guitar tops is given in J. C. Maugin,
Manuel du Luthier (Paris, 1869).
Heat 500 g Venice turpentine in a pot until it has reduced to a solid resin
upon cooling. Take 45 g of this solidified resin and add it to the following:

Sandarac 90 g

Alcohol 375 ml

Giuseppe Scarampella's violin varnish, as conveyed in a letter to his

brother Stefano, contained the following:

Refined alcohol 80 g

Shellac 8g

Sandarac 16 g

Elemi 4g

Benzoin 4g

[Venice] Turpentine 8g

Dragon's blood 50 cg [centigram?]

Aloe 5 cg

Aniline black 0.05 cg

Tannic acid 0.05 cg

Otto Erdesz (1917–2000)

According to Michael Gerson, an associate of the violin maker Otto

Erdesz, the main component of this maker's violin varnish was Bulls Eye
brand orange shellac. Between coats, he rubbed down the surface with a cloth
moistened with India ink and No-Crawl, a brand of wetting agent (probably
ox-gall). Most of the India ink was then wiped off with a moist cloth before
applying the next coat of shellac. What remained of the India ink subdued the
brassy color of the shellac, producing a very pleasing effect.

French polish (“1½ lb cut”):

Shellac 6 oz

Alcohol quart

Some synthetic resins used to make varnishes

(Paraloid and Acryloid are synonymous trade names.)

Paraloid B-72. Methyl acrylate ethyl methacrylate copolymer. Widely used
as a picture varnish and in concentrated form as an adhesive. It is relatively
flexible. Commonly used solvents include toluene, xylene, and acetone.
Knoop hardness (KHN) 10–11; Glass transition temperature, °C (Tg).
Paraloid B-67. Isobutyl methacrylate polymer. Somewhat harder than B-
72. Commonly used solvents include toluene, xylene, and acetone. KHN
hardness 11–12; Tg 50.
Paraloid B-48N. Methyl methacrylate copolymer. Commonly used solvents
include toluene, xylene, and acetone. Adheres well to bare metal, and is an
especially good protective coating for copper, brass, and bronze. KHN
hardness 11–12; Tg 50.
Paraloid B-44. Methyl methacrylate copolymer. Adheres well to metal and
other materials, harder than B-48N. Often used to protect outdoor metalwork
and sculpture. Commonly used solvents include toluene, xylene, and acetone.
Used in Incralac. KHN hardness 15–16; Tg 60.
Laropal K-80, and Ketone Resin N (discontinued) are polycyclohexanone
resins of low molecular weight. Commonly used solvents include toluene,
xylene, ethanol, and 2-propanol. The advantage of these varnishes is their
excellent penetration of fine scratches and relatively high refractive index.
Their disadvantage is brittleness (tapping with a fingernail may cause a thin
film to craze) and tendency to become detached.
The hydrogenated-hydrocarbon resins such as Regalrez 1094 and Escorez
5300 are considered to be more stable than the polycyclohexanone resins
mentioned above. They dissolve in mineral spirits of low aromaticity and
remain so indefinitely. The addition of a hindered-amine UV light stabilizer
such as Tinuvin 292 (2%) is recommended.

References
Paul Otto Apian-Bennewitz, Die Geige, der Geigenbau, und die
Bogenverfertigung (Weimar, 1892).
Alberto Bachman, An Encyclopedia of the Violin (New York, 1925).
John Broadhouse, The Violin: Its Construction Practically Treated (London).
Ignazio Alessandro Cozio di Salabue, Carteggio (Milan, 1950).
Johann Melchior Cröker, Der wohl anführende Mahler (Jena, 1778).
E. René de la Rie and Christopher W. McGlinchey, “New Synthetic Resins
for Picture Varnishes,” Cleaning, Retouching and Coatings: Preprints of
the Contributions to the Brusssels IIC Congress, 3–7 September 1990
(London, 1990).
Charles Lock Eastlake, Materials for a History of Oil Painting (London,
1847).
Robert L. Feller, Nathan Stolow, and Elizabeth H. Jones, On Picture
Varnishes and Their Solvents (Washington, 1985).
William M. Fulton, Turpentine Violin Varnish (1988).
Edward Heron-Allen, Violin-Making: As It Was and Is (London, 1885).
C. V. Horie, Materials for Conservation (London, 1987).
Johann Kunckel, Der neu Kunst- und Werck-Schul (Nuremberg, 1707).
E. Mailand, Découverte des anciens vernis italiens (Paris, 1859).
J. C. Maugin, Manuel du Luthier (Paris, 1869).
Walter C. McCrone, Lucy B. McCrone, and John Gustav Delly, Polarized
Light Microscopy (Chicago, 2002).
Mary P. Merrifield, Medieval and Renaissance Treatises on the Arts of
Painting (London, 1849).
John S. Mills and Raymond White, The Organic Chemistry of Museum
Objects (Oxford, 2003).
Otto Möckel, Die Kunst des Geigenbaues (Berlin, 1933).
Johann Christian Müller, Praktische Anweisung zum lakkiren (Leipzig,
1801).
Stewart Pollens, “Travelling Apprenticeship,” The Strad 97/1160 (December,
1986), pp. 559–564.
Rohm and Haas website.
André-Jacques Roubo, L'Art du menuisier (Paris, 1769–1774).
Pierre François Tingry, The Varnisher's Guide (London, 1832).
Auguste Tolbecque, L'Art du luthier (Paris, 1903).
Jean-Felix Watin, L'Art du peintre, doreur, vernisseur (Paris, 1772).

Viola da gamba gut strings

See Gut stringsMersenne's Law

Gut string diameters for viols can be calculated from instructions given in
the sections on Gut strings and Mersenne's Law. The following gauges are
adapted from tables provided by Northern Renaissance Instruments, a
supplier of gut strings for a variety of instruments. These diameters differ
somewhat from those calculated using Mersenne's Law. The gauges given
below can be used for preliminary setup, though it may be desirable to alter
the stringing to suit the instrument and player.

6-string pardessus de viole, medium thickness gut, in mm

 1st 0.51

2nd 0.64

3rd 0.86

4th 1.08

5th 1.44

6th 1.92

5-string pardessus de viole, medium thickness gut, in mm

1st 0.51

2nd 0.64

3rd 0.86

4th 1.21

5th 1.92

Treble viol, medium thickness gut, in mm

1st 0.546

2nd 0.686

3rd 0.914

4th 1.14
 5th 1.52

6th 2.03

Tenor viol, medium thickness gut, in mm

1st 0.572

2nd 0.724

3rd 0.965

4th 1.22

5th 1.626

6th 2.159

Bass viol, medium thickness gut, in mm

1st 0.635

2nd 0.813

3rd 1.092

4th 1.372

5th 1.829

6th 2.413

7th 3.226
 Violone, G tuning, medium thickness gut, in mm

1st 1.02

2nd 1.28

3rd 1.71

4th 2.16

5th 2.88

6th 3.85

Violin adjustment

General principles

Violins tend to go in and out of adjustment with changing humidity. High

humidity decreases the stiffness and increases the weight of wood, thereby
diminishing the production of higher frequencies. Dryness increases stiffness
and lightens wood, thereby supporting higher frequencies. Varnish stiffens
wood without adding appreciable weight, thus raising the stiffness-to-mass
ratio. Reducing the thickness of a violin's top or back (as is done when
regraduating an instrument) causes the stiffness of the wood to be reduced by
the third power; that is, if the plate's thickness is reduced by one half, its
stiffness is reduced to one-eighth of the original value: (1/2)3 = 1/8, or 0.125.

Bridge

The violin bridge acts like a low-pass filter, partially blocking high
frequencies, such as bowing noise. The standard bridge blank provides some
latitude in adjusting the tonal characteristics of an instrument or
compensating for certain deficiencies. By cutting away wood or forming an
arch between the feet of a violin or viola bridge, one may be able to brighten
the sound of an instrument. Increasing the diameter of the eyes and heart may
reduce an overly bright or strident tonal quality. The French-style cello bridge
is recommended for bright-sounding instruments, while the Belgian-style
cello bridge, with its longer legs and reduced mass above the heart, tends to
increase sound production in general, which may compensate for cellos that
have a dark sound.

Soundpost

According to Hans Weisshaar and Margaret Shipman (1988), the violin's

soundpost is generally positioned around 2–2.5 mm behind the treble foot of
the bridge, 2.5–5 mm behind the treble foot for violas, and 6.5–7 mm behind
the treble foot for cellos. They suggest post diameters of 6.0–6.5 mm for
violins, 6.5–7 mm for violas, and 10.5–11 mm for cellos. Jacob Augustus
Otto (1875) suggests a much different soundpost position: half an inch (about
12.7 mm) behind the treble foot of the bridge in a good violin, but moved
closer in faulty ones to give greater fullness of tone.
Millant and Millant (1952) provide the following recommendations for
violin soundpost adjustment:

For highly arched instruments use a post of 6.5 mm diameter; for flat
arched instrument use a post of 6.2 mm. If the soundpost is too tight, the
tone will be harsh. Correct by pushing the post inward. If the post is too
close to the bridge, the tone will be too hard. If the violin is thin or
highly arched, place the post closer to the bridge. If the arching is flat or
the thickness great, move it away from the bridge. To give the E string
brilliance, bring the soundpost closer to the bridge or move it towards
the edge of the f-hole. This will be to the detriment to the bass.

Contradicting the Millants' advice, Weisshaar and Shipman suggest placing

the soundpost further behind the bridge in highly arched instruments.
Rather than moving the soundpost, one can shorten it by about ½ mm if the
tone is hard, or make a longer soundpost if the sound is weak. This better
preserves the balance between strings than moving the soundpost to make a
tighter or looser fit. The whistling or squeaking of strings can sometimes be
alleviated by loosening the soundpost (moving it towards the center of the
instrument) or reducing its length. Keep in mind that knocks and vibration
tend to shift soundposts into looser rather than tighter position, so if tonal
quality has deteriorated, moving the soundpost into a tighter position in
accordance with the arching may restore the instrument's sound.

Bass-bar

According to Weisshaar and Shipman, the modern bass-bar of the violin

extends to about 40 mm from the top and bottom of the table, about 11–13
mm in height, and 5.5–6 mm in thickness; that of the viola is about 12–14
mm in height and 6–6.5 mm in thickness; and the cello bass-bar is about 21–
23 mm in height and 10.5 mm in thickness.
Sacconi (1979) recommends positioning the external edge of the bass-bar
about 1 mm inside the edge of the bridge foot of violins and 1.5 mm inside
the edge of the bridge foot of cellos. He establishes its inclination by fixing
both of its endpoints one-seventh the distance between the belly's center line
and the outer edge of the upper and lower bouts.
According to Millant and Millant (1952), the bass-bar helps support the
downward pressure exerted by the strings and facilitates vibration throughout
the entire length of the top. To position the bass-bar, they advise that it must
be 40 mm from the top and bottom edges, but that the inclination is
established by making the top of the bar 18 mm from the center joint and the
bottom 21 mm from the center joint (as opposed to the proportional system
recommended by Sacconi). To make the bar, they start with a strip of sapin
(commonly fir) 270 mm in length and 5 mm wide, and then trim it to length
according to the 40 mm recesses indicated above. A pair of compasses is
used to trace the curve of the top plate's interior onto this strip of wood,
which is then trimmed to shape with a knife. They continue:

The bar is like a spring that helps to resist the tension of the strings on
the table. This bar should not follow the interior curve of the table
exactly. Mark the center of the bar on both sides and adjust the top half
by checking that the bottom part is two to three millimeters above the
 table. Then press down on the bottom and check to see that the top
stands two to three millimeters above the table. Be careful to adjust the
bar so that it stands perpendicular to the edge of the table. Thus the
glued surface of the bass-bar will not be perpendicular to the sides of the
bar. With a smooth file, remove any irregularities. The pressure of the
bar should not be directed immediately under the bridge, but
progressively along the entire top. When gluing, apply the glue very hot.
Start by clamping the extreme ends first. The rest of the bar is then
clamped. After the glue is dry, shape the bar. First mark the maximum
height, 13–17 mm, then cut away the wood up to that height. Then
diminish the extreme ends to 2–3 mm. Continue making the bar lighter
and lighter according to your inclination. Then use a file and glass
paper. Finish by beveling with a knife and rounding the edges with a file
and glass paper.

Not all violin makers “spring” their bars; some prefer to fit them perfectly
without any tension.
See Violin: Baroque fittings and strings for dimensions of early bass-bars.
A great variety of clamps are available for the various gluing operations
involved in maintaining and restoring stringed instruments (Figures 39, 40).
These include garland, spool, and special assembly clamps for gluing the
belly and back plates to the ribs, crack clamps, corner repair clamps, edge
clamps, deep-engagement clamps for cleating and gluing in bass-bars, endpin
rib clamps, and clamps designed for gluing saddles and fingerboards. Some
of the newly designed clamps are made of lightweight materials, such as
aluminum and carbon fiber, which reduce the tendency of clamps to distort
the shape of the workpiece.
Figure 39 Gluing the lower bout of a violin made by Peter Guarneri,
Venice, 1742.

Figure 40 Gluing together a violin top.

For stringing see Figure 41.

Figure 41 Method of attaching violin string to tuning peg. Note how the
string is wound back over the first coil to lock it in place.

References
Roger Millant and Max Millant, Manuel pratique de lutherie (Paris, 1952).
Jacob Augustus Otto, A Treatise on the Structure and Preservation of the
Violin (London, 1875).
Simone F. Sacconi, The “Secrets” of Stradivari (Cremona, 1979).
Hans Weisshaar and Margaret Shipman, Violin Restoration: A Manual for
Violin Makers (Los Angeles, 1988).

Violin: Baroque fittings and strings

Neck
 The Baroque violin featured a tapered neck and wedge-shaped fingerboard
that required the player to change the shape of his or her grip while shifting
up and down the neck. The modern angled-back neck, together with a
thinner, solid ebony fingerboard, provide a nearly parallel glide path for the
left hand. The transition to the modern-style neck and fingerboard was
gradual and began around the third quarter of the eighteenth century (Pollens,
2014). Violins made in earlier periods (including those by Stradivari and his
contemporaries) were modernized to accommodate evolving performance
technique and new repertoire, which require quicker shifts and playing in
higher positions.
Though a number of Stradivari's viola and cello neck patterns are preserved
in the Museo Stradivariano, none of his violin neck patterns survive, and the
few original violin necks that are still mounted on his instruments have all
been reshaped and extended at the heel so that they could be mortised into the
top block. In their 1902 biography of Antonio Stradivari, the Hills state that
they knew of seven Stradivari instruments that retained their original necks,
though they name only five: the 1715 “Alard,” the 1714 “Soil,” the 1721
“Blunt,” the 1724 “Sarasate,” and the 1716 “Messiah.” The 1690 “Medici”
tenor viola and the 1693 “Harrison” violin are two other Stradivari
instruments that retain their original necks – perhaps these were the other two
they had in mind. The altered neck of the “Soil” violin is no longer mounted
on the instrument but is preserved in the Museo Stradivariano (MS no. 128).
Because the “Soil” and other original Stradivari violin necks all have heel
grafts, it is not possible to establish their original lengths with precision.
Despite the lack of direct evidence of the original neck lengths of Stradivari's
violins, the author has deduced their dimensions from original fingerboards
and fingerboard patterns (see below), and it appears that they had virtually
the same effective length (as measured from the nut to the upper edge of the
instrument's belly) as modern necks. The heel grafts on the extant original
necks provide enough extra length to form a dovetailed tenon that was
mortised into the top block (Stradivari's original violin and viola necks were
butted up against the upper rib and held in place with iron nails driven
through the top block). The graft also permitted these necks to be angled back
several degrees further than they were originally, though the neck angle was
primarily increased to compensate for the non-wedge-shaped fingerboard (see
below).
 One significant change made in the design of the modern violin and viola
neck and fingerboard was the extension of the neck foot beyond the upper
surface of the belly to provide what is termed “overstand” (or appui). Today,
this is generally between 6–8 mm in violins and violas and 20–22 mm in
cellos. The Baroque violin and viola neck foot did not project beyond the
upper surface of the instrument's belly; instead, a wedge-shaped fingerboard
provided the requisite height and inclination to the slightly angled neck.
The neck, tailpiece, and bridge of the “Medici” tenor viola (1690) are
original – this remarkable instrument is the only extant example of
Stradivari's work to survive with these fittings intact. The neck has not been
reset, though a wedge has been inserted between the neck and the
fingerboard, presumably to compensate for the neck pulling forward due to
string tension. It should be noted that the fingerboard on this viola is
considerably longer than the pattern for the original fingerboard preserved in
the Museo Stradivariano, which suggests it may be a replacement. Though
inventory and maintenance records for this instrument extend as far back as
1700 (including a major intervention by the violin maker Giuseppe
Scarampella in 1869 that required the removal of the top), there is no mention
of the insertion of the wedge between the neck and fingerboard in any of
these records. It is conceivable that this repair dates from the eighteenth
century or might even have been an adjustment made shortly after the
instrument was delivered to the Medici court in Florence in 1690.
The “Medici” tenor viola neck is not mounted at a 90-degree angle relative
to the upper rib, but is set at approximately 86 degrees. The wedge-shaped
fingerboard produces a composite angle of about 83 degrees, which is within
a degree or two of that used in the modern viola setup. The neck pattern for
this instrument is preserved in the Museo Stradivariano (MS no. 237), but the
dimensions and shape of the neck foot are not clearly indicated, and it would
appear that the pattern was left oversize to provide extra wood for final
fitting; however, a supplemental pattern for the foot (MS no. 240; Figure 42)
also provides a bit of leeway for final trimming (the heel length is 43 mm,
whereas the upper rib width of the “Medici” tenor viola is 39.5 mm) but the
bottom of the foot is cut at an angle of 86º, which matches the original neck
presently mounted on the viola. This confirms that Stradivari did not mount
the neck of this tenor viola in line with the body of the instrument, but tilted it
back 4 degrees. The nut-to-heel length of the original neck is 152.5 mm, and
the body stop is 263.3 mm, which indicates that the neck is about 23 mm
short of the 2:3 mensur (German measurement; the ratio of neck length to
body length). This effectively reduced the string length, undoubtedly to make
this rather unwieldy instrument easier to play. The tenor viola was generally
played in low positions that did not make extensive use of the neck heel for
orienting the hand; thus, it was not as important to maintain the 2:3 mensur
that was used with the violin and contralto viola. The string angle over the
bridge of the “Medici” tenor viola is presently about 158 degrees, compared
to 154 degrees used in a typical modern viola setup.

Figure 42 Stradivari tenor viola neck-foot template showing angled foot.

Like the neck pattern for the tenor viola, the final dimensions for the foot
of the neck pattern for the contralto viola (MS no. 213) are not clearly
indicated, though there are a number of reference marks on it. These include
two pin holes approximately 4 mm from the bottom edge of the foot (these
pin holes may have been made in an early nineteenth-century Museo
Stradivariano installation), a short ink mark about 8.5 mm from the bottom of
the foot (which may represent the point where the neck meets the upper edge
of the instrument's top), and a line drawn roughly parallel to the bottom edge
of the foot and about 15 mm above it (the function of this line is unclear,
though it evidently does not represent a cutoff point as it is too close to the
top of the heel to provide for a full-size button). The cello-style pegbox
cheeks (which originally coincided with the end of the nut and the beginning
of the fingerboard) are denoted on the pattern by a series of inked dots. The
distance between these dots and the short ink mark on the foot (which may
indicate the trimmed dimension) indicates a nut-to-heel length of about 147
mm.
The 1690 “Tuscan” contralto viola is believed to have been made for the
same Medici commission as the tenor viola discussed above. It has a body
stop of 224.5 mm, and if we multiply the body stop by two-thirds, we arrive
at a theoretical nut-to-body length of 149.6 mm, which is close to the distance
between the inked marks on the neck pattern. This indicates that Stradivari
employed a 2:3 mensur in designing this viola (the neck is not foreshortened
as in the tenor viola). The “Tuscan” viola does not have its original neck, and
the new one is presently fitted with a nut mounted in line with the bottom of
the pegbox (that is, well below the cello-style pegbox cheeks). This suits
violists and violinists who find it comfortable to use the chin of the pegbox to
orient their hand in first position. Thus, the mensur of the “Tuscan” is not
presently set at a 2:3 ratio as it was originally designed. Again, the final
dimensions and shape of the heel are not indicated on MS no. 213, but the
supplemental pattern for the foot, MS no. 216, provides a neck angle of 86
degrees, just like that of Stradivari's tenor viola.
Several of Stradivari's cello neck patterns are preserved, and from these we
can clearly see that his cello necks were also angled back. The length of the
neck feet and the presence of scribe lines parallel to the bottom of the feet
suggest that the necks may have been mortised into the top blocks, though
they were apparently reinforced with nails like those of the violins and violas,
though in the case of cellos six nails were used rather than three. From the
patterns, it appears that the neck feet extended beyond the top of the
instrument, very much like the modern cello neck. As with the violin and
viola, a wedge-shaped fingerboard was used. The “Batta” Stradivari cello of
1714 appears to have been made on the B form because the B-form f-hole
positioning template preserved in the Museo Stradivariano (MS no. 272
recto) matches that cello's C-bouts, corners, and f-hole placement. A cello-
neck pattern also marked with the letter B (presumably used to make B-form
cellos) has a nut-to-heel length of 286 mm, which forms a 7:10 ratio with the
“Batta's” body stop of 406 mm. This 7:10 mensur is also used today in setting
up most cellos.
 Three cello neck patterns in the Museo Stradivariano, as well as a
supplementary cello neck-foot pattern marked B (MS no. 279) all provide
about an 84-degree angle relative to the upper rib. Though the overall length
of the foot of the neck pattern MS no. 276 is 158 mm, there is an inked line
that indicates it was intended to be trimmed back to 134 mm. The trimmed
length would provide an overstand of about 6 mm with cellos having an
upper rib depth of 128 mm, which is very likely close to the uncut dimension
of a B-form cello upper rib. Though this is considerably less than the 20–22
mm overstand that is used today, a wedge-shaped fingerboard 29 mm thick at
the neck heel (see discussion of cello fingerboard pattern MS no. 280, below)
would have increased the distance between the playing surface of the
fingerboard and the upper edge of the top plate to about 35 mm, which is
close to that used today. The wedge-shaped fingerboard would have
increased the effective neck/fingerboard angle to within a degree or two of
that used in modern cello setup.

Fingerboard

The fingerboards used in the Baroque were quite different from those used
today. The modern fingerboard is made out of a relatively thin slab of solid
ebony, a hard and extremely dense wood that creates a rigid platform for the
stopped string. As indicated above, Stradivari's violin and viola necks were
not angled back to the same extent as modern necks, nor did they extend
beyond the edge of the top as do modern necks. Instead, the final angle and
elevation of the strings were provided by wedge-shaped fingerboards.
Stradivari's violin fingerboards ranged from about 15 to 18 mm thick at the
neck foot (the modern violin fingerboard is about 8 or 9 mm thick at that
point); his viola fingerboard patterns MS nos. 217 and 241 have markings
indicating a thickness of 23 mm at the neck foot; his cello fingerboard pattern
MS no. 280 has inscribed markings indicating a thickness of 29 mm at the
neck foot.
The feet of Baroque violin and viola necks were rabbeted to clear the
bellies where the tops overlapped the ribs. Baroque fingerboards were
notched at their juncture with this rabbet and were undercut to follow the
arching of the top. Stradivari's fingerboard patterns generally mark the
position of this notch and provide a pair of compass arcs that indicate the
thickness of the fingerboard at that point. Because this notch aligns precisely
with the upper edge of the instrument's top plate, it provides a definitive point
for determining the neck stop, which is needed to calculate the string length
of the instrument for which the fingerboard was intended. Fingerboard
patterns MS nos. 131, 132, 133, and 134 are marked with the form letters P,
G, G, and PG, respectively. A number of violins have been matched with
these forms, and by comparing the distances from the nut to the body, which
are indicated on the fingerboard patterns, with the body-stop lengths of these
instruments, it becomes evident that Stradivari employed a 2:3 mensur in his
violins. For example, the “Soil” Stradivari violin has a body stop of 198 mm,
which requires a neck having a nut-to-body length of 132 mm to provide a
2:3 mensur. We know that this instrument was made on the G form because
the original neck of this instrument, MS no. 128, has a pegbox inscription G.
The G-form fingerboard patterns MS nos. 132 and 133 have a nut-to-body
length of 129 mm, which is just 3 mm short of the theoretical length
calculated using the 2:3 ratio and the “Soil” violin's body stop. Today, full-
size violin necks are generally set at 130 mm, which is only one millimeter
longer than the neck length calculated from fingerboard patterns made for use
with Stradivari's full-size violin forms P and G. This contradicts current
dogma that the “modern” violin neck is significantly longer than those
originally fitted by makers in the Baroque.
An original Stradivari violin fingerboard, MS no. 129, has a core of willow
edged with figured maple and faced with ebony on the playing surface. It is
213 mm long and has a nut-to-body length of 120 mm, so it was presumably
fitted to a small violin having a body stop of about 180 mm. The fingerboard
is 26 mm wide at the nut and 40 mm wide at the bottom; it is 5 mm thick just
beyond the nut and 15 mm thick where it met the violin's top plate, producing
a 4.5-degree angle. The ebony facing is bordered by an inlay of ivory and
ebony, a decorative touch that contrasts with today's solid ebony
fingerboards. The figured maple facings on the sides bear traces of varnish
along the entire length, indicating that Stradivari varnished the necks of his
instruments. Willow is very light and was undoubtedly used as the core to
reduce weight. Though to face and inlay was a labor-intensive process, it
conserved ebony, which was a precious commodity. Baroque tailpieces were
constructed in the same manner.
An important distinction between Baroque and modern fingerboards is the
difference in their lengths. The modern violin fingerboard is about 270 mm
long, whereas Stradivari's patterns and original fingerboards are between 190
and 213 mm long, 207–213 mm being the apparent range for full-size violins
made on the P, PG, and G forms. This length represents the uppermost note
that can be stopped on the fingerboard, which in the case of full-size violins
is shy of an octave and a fifth above e2, or b3 (to stop a string a full octave
and a fifth above a typical string length of 327 mm would require a
fingerboard of 218 mm). Stradivari's template for the fingerboard of the
contralto viola made for the Medicis in 1690 (MS no. 217) is 236 mm long,
whereas modern viola fingerboards range from around 290 mm to 310 mm in
length. His fingerboard pattern for the B-form cello (MS no. 280) has an
overall length of just 424 mm, which would have been adequate for playing
up to the fifth position, which is the uppermost position given in Michel
Corrette's cello tutor of 1741. Thus, the fingerboards fitted in Stradivari's day
were too short to accommodate music written just a few decades later (for
example, Mozart's fifth violin concerto, composed in 1775, extends up to
c♯4), so it is little wonder that no Stradivari violins, violas (with the possible
exception of the “Medici” tenor), or cellos are still equipped with their
original fingerboards.

Strings

Strings are the most ephemeral of the violin's fittings, and very few survive
from Stradivari's time. Two partial sets of strings for a contralto viola and a
cello are preserved in the Museo Stradivariano in Cremona. Three viola
strings (MS no. 222) are sewn onto a sheet of heavy paper that is inscribed
Adi agosto 1727 queste Quattro corde sono la grosezza per finire la viola a
Quattro corde ciovè il contraldo (In the year of our Lord, August 1727, these
four sounding strings are of a size to finish the viola of four strings, namely
the contralto). Three cello strings (MS no. 309) are mounted onto a sheet of
paper that reads Queste sono le mostre del tre corde grosse quella mostra che
sono di budelo va filata è vidalba (These are examples of three thick strings
of gut that are overspun and polished [?]). Overspun strings came into use
around 1660 (see below). Like the cello and viola, violins in Stradivari's day
most likely used an overspun lowest string. The earliest overspun strings
employed a thin wire filament (copper, tinned or silvered copper, or silver)
spun directly over a plain gut core (without the “silking” that is commonly
inserted between the core and the wire filament in the manufacture of most of
today's overspun strings).
No original Stradivari violin strings survive, but a number of attempts have
been made to reconstruct their diameters from various documentary and
iconographic sources. The diameters of violin strings used in Stradivari's day
may be deduced from experiments conducted in 1734 by the violinist
Giuseppe Tartini, which were recounted by the music historian François-
Joseph Fétis (1784–1871) in 1856. According to Fétis:

Tartini found, by experiments made in 1734, that the pressure of the four
strings on the instrument was equal to 63 pounds [63 livres is given in
the original 1856 French edition; l'ancienne livre was equivalent to
489.5 grams, whereas the modern pound is equivalent to 453.6 grams].
It must be observed that the strings of Tartini were smaller than those
with which violins are now mounted, and that his bridge was lower, so
that the angle formed by the strings was considerably less. Twenty years
ago, the first string required a weight of 22 pounds in order to bridge it
up to pitch, and the other strings a little less; so that the total pressure
was, then, about 80 pounds. After 1734, the pitch was raised a semitone,
the instruments were mounted with thicker strings, and the angle which
they formed on the bridge was more acute: hence the necessity of re-
barring the violins. Since then, so excessive has been the rise in pitch,
through the craving for a brilliant sonority, that there is nearly a
difference of a semitone between the pitch of 1830 and that of 1856. If a
new experiment were now made to ascertain the pressure of the four
strings on the belly of a violin, no doubt it would be found greatly
augmented. This enormous weight incessantly tends to effect the
destruction of the old instruments, and demands increased power of
resistance in the bar underneath the bridge. Such is the real cause of the
necessity of substituting for the old, weak bar, in the violins of
Stradivarius, one of stronger proportions.

Fétis was himself a violinist, and in the preface to his published account he
acknowledges the technical assistance provided by the violin maker J. B.
Vuillaume. Though doubts have been cast upon the reliability of Tartini's
measurements of string tension and their conversion to modern units of
weight, the 63 pound (or livre) figure itself has not been disputed. Oddly, 63
pounds (28.6 kg) or 63 livres (30.8 kg) is considerably greater than used
today at higher pitch (a set of medium-weight Dominant strings produce a
total tension of 22.1 kg, or 48.6 pounds); the 80 pounds mentioned by Fétis
would be excessive by today's standards.
Tartini's total string tension (as reported by Fétis) is similar to that given in
Abbé Sibire's La Chélonomie ou le parfait luthier (Paris, 1806; Brussels,
1823). Sibire, whose informant was the French violin maker Nicolas Lupot
(1758–1824), indicates that a violin's strings had a total tension of 64 livres
(19 livres for the chanterelle, or first string, 17 livres for the second, 15 livres
for the third, and 13 livres for the fourth).
Nevertheless, the precise diameters, and hence the distribution of tension
among the strings, remains a subject of debate that has been clouded rather
than clarified by several seventeenth- and eighteenth-century writers, who
indicated that the relative thicknesses of strings should be in proportion to
their relative pitches and that tension should be equal from string to string.
Marin Mersenne (1636), for example, wrote that the violin E string was
equivalent in thickness to the fourth string of a lute, which he indicated was
equal to one-third of a ligne (in the old pied d'roi, or about 0.75 mm). He
further indicated that “the strings will be perfectly proportioned among
themselves when they follow the ratios of the said notes.” Taken literally,
impractical sequences of string diameters would result: for example, starting
with a violin E string of 0.75 mm, it then follows that the A string would be
1.13 mm, D would be 1.69 mm, and G would be 2.53 mm (much too thick);
working backwards from a more reasonably proportioned gut G string of 1.9
mm, D would be 1.27 mm, A 0.84 mm, and E 0.56 mm (much too thin).
Leopold Mozart, however, advocated the same principle of proportionality in
his Versuch einer gründlichen Violinschule (1756):

The violin is strung with four strings, each of which must be of the right
thickness in relation to the other. I say “right thickness,” for if one string
be a little too thick in proportion to another it is impossible to obtain an
even or a good tone. It is true that violinists and violin-makers
frequently judge these thicknesses by the eye, but it cannot be denied
that the result is often very bad. Indeed, one must go to work with the
greatest patience and care if one wishes to string the violin properly and
 in such fashion that the strings have their intervals in the right
proportion to each other, and the right notes lie therefore opposite each
other. He who is willing to take the trouble, can test them according to
mathematical principles. He can take two well-stretched strings, an A
and E, a D and A, or a D and G, each of which is as exactly as possible
of the same thickness throughout. That is: the diameter or cross-section
must be uniform. To each of the two strings equal weights can be
attached. Now, if the two strings have been well chosen, they should, on
being struck, give forth the interval of a perfect fifth, but if one string
sounds too sharp and oversteps the fifth, this is a sign that it is too weak
and a thicker string is then selected; or the string which sounds flat and
is therefore too thick may be exchanged for a thinner string. One must
proceed thus until the perfect fifth is attained and the strings are in
proportion and truly chosen. But how difficult it is to find evenly made
thick strings! Are they not mostly thicker at one end than at the other?
How can one make a sure test with an uneven string? I would therefore
remind you that the choice of strings must be made with the greatest
care and not merely at random.

Despite Leopold Mozart's explicit directions for suspending weights from

strings when selecting gauges, many of today's violinists involved in period
performance practice, as well as historians and makers of gut strings, reject
the idea of equal tension and advocate a system of so-called “progressive
scaling.” Mimmo Peruffo's (1994) research into the diameters of holes drilled
in early lute bridges does in fact confirm that Mersenne's and Mozart's rule of
employing string diameters in strict proportion to pitch (which is equivalent
to equal tension from string to string) was not strictly adhered to, as bass
strings calculated from the pitch relationships between the strings would
result in rather large diameters that could not have passed through the holes
found in early lute bridges. However, it should be pointed out that lutes are
tuned over a greater pitch range than the violin, so string gauges could not
have been increased or decreased in strict proportion to the pitch.
Regarding the use of overspun strings for the lowest pitched strings of
members of the violin family, it is generally believed that gut strings
overspun with thin metal wire first came into use around 1664, the year an
advertisement for them appeared in John Playford's Introduction to the Skill
of Music. Mimmo Peruffo cites a slightly earlier manuscript of Samuel
Hartlib dated 1659 that refers to Goretsky's invention of lute strings covered
with silver wire. However, overspun strings may have been in use at a much
earlier date, as Michael Praetorius's Syntagma Musicum II (1619) refers to
twisted or spun strings used in a d'amore variant of the viola bastarda made
in England:

Jetzo ist in Engelland noch etwas sonderbares darzu erfunden dass unter
den rechten gemeinen sechs Säitten noch acht andere Stälen und
gedrehete Messings-Säitten uff ein Messingen Steige (gleich die uff den
Pandorren gebraucht werden liegen).
(Presently something curious has been invented in England, that under
the ordinary six strings there are in addition eight other steel and spun
brass strings on a brass bridge, just as they would lie on the pandora.)

Praetorius then goes on to explain how sympathetic strings worked, which

leads one to conclude that the addition of sympathetic strings to the viola
bastarda was the “curiosity” and not the fact that they were made of steel and
overspun with brass. The Pandora (or Bandora) was a plucked wire-strung
instrument with scalloped ribs invented by John Rose in England in 1562.
The earliest examples are believed to have had twisted (rather than overspun)
metal strings; however, the German term gedrehete suggests spinning,
turning, as well as twisting, therefore it is possible that Praetorius is referring
to steel-core strings overspun with brass wire.
For equally tensioned strings pulled to a total of 63 pounds or livres (the
total tension measured by Tartini in 1734), we can use Mersenne's Law to
calculate the diameters of each of the gut strings where A=420 Hz, which is
an estimate of the pitch used in Padua, where Tartini lived when he
conducted his string tension experiment (A=420 Hz is about 81 cents below
A=440 Hz):
D = √T/F2L2ρπ

where

D is the diameter in meters

 T is the tension in newtons, in this case (63 lb/4) × 4.44822 newtons per
pound, or 70.06 newtons per string
F is the frequency in Hz, in this case 187 Hz for g, 280 Hz for d1, 420 Hz
for a1, and 629 Hz for e2
L is the length of the strings in meters, in this case 0.327 meters
ρ is the density of gut in kilograms per cubic meter, in this case 1,300
kg/m3.

The following gut string diameters were calculated:

e2 = 0.64 mm
a1 = 0.95 mm
d1 = 1.43 mm
g = 2.14 mm.

The G string of the violin would not have been plain gut but rather gut
overspun with fine metal wire. According to Francesco Galeazzi's Elementi
teorico-pratici di musica (1791), the core of the G string was made with una
seconda non molto grossa (a second [string] not too large) and overspun with

L'argento, che comunemente si adopera a questo uso è rame inargentato,

e deve esser sottilissimo. Si adopera con egual successo il rame
semplice, ed anche l'acciajo: ho fatto a bella posta filare dell' argento
fino, ma non vi ho conosciuta differenza dall' argento falso commune, se
non che ei non diventa rosso, ma resta sempre bianco, rilucente, come
fosse sempre nuovo.
(The silver normally used for this purpose is silver-plated copper, and
must be very thin. One can use with equal success copper and even iron.
I purposefully wound some thin pure silver, but saw no difference from
the use of common false silver, except that it does not become red but
stays white and shiny, as if always new.)

Peruffo suggests that wire having a diameter of 0.12–0.13 mm was used for
overspinning, but wire of such fineness was not readily available in
Stradivari's day, or even in the late eighteenth century when Galeazzi wrote
his account of string making. Patrizio Barbieri (1985) has attempted to
reconstruct the diameters of strings found on a harpsichord made in 1559 by
Vito Trasuntino based upon the weights of the strings published in 1767 by
Giordano Riccati (though Riccati did not specify the material from which the
strings were made, which is essential for calculating the string diameters if
only the total weight of the string is known). According to Barbieri's
calculations, the finest strings found on this harpsichord had a diameter of
0.15 mm. The finest gauge of wire generally employed in musical instrument
making was used in the top strings of the 4-foot choir of harpsichords.
Original wire fragments found in the 4-foot choir of a harpsichord by
François Blanchet dated 1733 are 0.17–0.18 mm in diameter, which is
probably a more accurate figure than those suggested by Peruffo and
calculated by Barbieri.
If we substitute a plain gut A string of 0.95 mm diameter for the gut G
string of 2.14 mm diameter, it alone would require a tension of only 13.78
newtons (1.41 kg-force) to bring it to 187 Hz (the pitch of G at A=420 Hz),
which would be insufficient to produce a good-quality sound. A 2.14 mm
diameter gut string 327 mm in length would have a mass of about 1.53
grams, whereas a 0.95 mm diameter string of equal length would have a mass
of about 0.30 grams. Therefore, a mass of 1.23 grams would have to be added
to the 0.95 mm string to bring it up to the requisite tension of 70.06 newtons
(7.14 kg-force). If wound over a 327 mm length of gut, a winding of copper
wire 0.17 mm in diameter would have a length of 6254 mm, and at a density
of 8940 kg/m3 it would provide a mass of about 1.27 grams, yielding a total
string mass of 1.57 grams, very close to and only marginally greater than the
calculated ideal mass.
To compute the length of a winding over a core:
Lw = π(Dc + Dw/2)·(Ls/Dw)

where

Lw is the length of the winding

Dc is the diameter of the core
Dw is the diameter of the winding
Ls is the length of the string.
To calculate the winding's mass, multiply its length by its cross-sectional
area, and then by the density. To simplify this calculation when using lengths
and diameters in millimeters, the density of copper is 0.008940 g/mm3.
Up until the mid nineteenth century, violin strings were made of whole
rather than split sheep gut, and from documentary sources we know that E
and D strings were typically made of 3 and 7 whole guts wound together
(presumably A strings were made of 5 whole guts). Furthermore, gut strings
were not ground down to smooth and regulate size, and thus only a rather
limited selection of gauges was then available. A set of early nineteenth-
century strings made by the highly regarded Ruffini firm in Naples was
measured by William Huggins in 1883. He reported the following:

1st 0.0265 in. [0.67 mm]

2nd 0.0355 in. [0.90 mm]

3rd 0.0460 in. [1.17 mm]

4th 1.41 g [presumably an overspun string, no diameter given]

Surprisingly, this tallies almost perfectly with the string diameters

calculated from Tartini's tension measurements. Strings of these diameters,
with the 4th, or G string, perhaps made up of a 0.90 mm gut core wound with
0.17 mm copper or silver wire, would be an appropriate set of strings for a
Stradivari violin with original fittings, were one ever to be found. Needless to
say, they would also be appropriate for an “authentic” performance of
Paganini's Caprices and Brahms's violin concerto. Holes for the strings in an
original tailpiece used with a Santo Serafin violin, circa 1750, in the Museo
Correr in Venice are 1.95 mm in diameter (measured by the author), which is
not inconsistent with these diameters.
As a point of comparison, the great twentieth-century virtuoso Jascha
Heifetz used the following strings on his “Ferdinand David” Guarneri:

1st Goldbrokat (steel music wire) 0.26 mm

 2nd Pirastro plain gut (14¾ Pirastro gauge) 0.73 mm

3rd Pirastro plain gut (19¼ Pirastro gauge) 0.94 mm

4th Tricolour (16¼ gauge; silver-wrapped gut) 0.815 mm

In conclusion, from Stradivari's patterns, we can see that he mounted the

necks of his instruments at a slight angle. The wedge-shaped fingerboards
used at that time contributed several more degrees, which brought the total
angle of the strings over the bridge very close to that used in modern violin
setup. The 2:3 neck-length to stop-length ratio was a precept of Cremonese
violin making; therefore neck and string length were not increased when
violins were adapted from Baroque to modern configuration. From Tartini's
measurements, we learn that violin string gauges and tensions may have been
considerably greater than used today. As many Strads, Guarneris, and other
fine instruments from the Baroque were regraduated (thinned) in later
periods, stringing regraduated violins with the original, heavier gauges of
strings would be an empty exercise and potentially disastrous, as plate
stiffness is proportional to the cube of its thickness.
Figure 41 shows the proper method of attaching a string to a tuning peg.
When mounting strings, after inserting the string through the peg's hole, make
one complete turn and then reverse direction and wind over this coil in order
to lock the string in place and prevent it from slipping. Continue winding so
that the string comes up close to the wall of the pegbox but does not bear
against it.

Bass-bar

Inscriptions and measurements taken from 50 original bass-bars removed

by the Hill firm from violins, violas, and cellos dating from 1615 through
1834 are noted below.
Violin bass-bar inscriptions:

1. Filippo. Fratello ai Montesecco. 1615. .A. E. H.

2. A. H. Amati. 1621
3. A. & H. Amati. 16_
4. Jacobus. Stainer. 1656.
5. N. Amati, 1665
6. N. Amati. 1671. A. E. H. 1890.
7. F. Ruger
8. Matthias. Albanus. 1674.
9. Strad. period 1680
10. Willems.
11. Ioannes Schorn 1696.
12. M. Albani. 1700.
13. Pasta. 17
14. Jacobs 1702
15. Stradivarius 1703. / Taken. Out. 1896. A. E. H.
16. Stradivari 1704 Betts A. E. H. 1893. .New. Piece. added –
Old.Broken. off / J. B. V. 1859
17. Stradivari. 1719.
18. F. Gobetti period 1720
19. A. Gagliano. 1721.
20. A. Gagliano, 1724
21. N. Cross. 1725.
22. N. Cross.
23. C. Tononi. 1730
24. David Techler 1734
25. S. Seraphin 1736
26. Carlo Bergonzi 1740
27. P. Guarnerius. Venice
28. R. Duke. 1750
29. Januarius. Gagliano. 1750
30. Sanctus. Seraphin. 1757. A. E. H. 1901
31. Carlo Bergonzi period 1760 (The Cozio)
32. Nicolo Gagliano. 1760.
33. J B. Guadagnini. 1760
34. Thomas Balestrieri 1770
35. Gabbrielli 1774
36. George Klotz.
37. Lupot. Orleano
38. Vinaccia.
 39. Lockey. Hill.

Viola bass-bar inscriptions:

40. Taken from Jacobus Stainer 1660

41. M. Albani, 1688. Tenor.
42. J. B. Guadagnini Tenor. 1757.
43. Ferdinandus Gagliano. 1783.
44. V. Panormo/Taken from Panormo tenor 10/4/59
45. V. Panormo. 1800. (Alto)
46. Rocca. 1834. / Guarnerius. Copy.

Cello bass-bar inscriptions:

47. Antonio Catini. Modena 1660. or 8.

48. Antonius. Stradivarius. 1667. Taken out in 1893. A. E.
H./Strad:V:Cello:purchased in Sicily. 1893.
49. A. Gagliano, 1725
50. Januarius Gagliano, 1736

Measurements of bass-bars from the Hill collection

Length Height Thickness Weight Grain

No. Maker/date
mm mm mm g direction

1. Filippo 228 5.6 4.1 1.50 Vertical

Violin Fratello
1615

2. A. & H. 270 5.7 4.3 1.53 Slab

Violin Amati 1621

3. A. & H. 234 6.4 3.9 1.54 Vertical

Violin Amati 16__

4. Jacob 250 6.7 4.9 1.68 Vertical

Violin Stainer
1656

5. N. Amati 235 5.0 4.4 1.19 Vertical

Violin 1665

6. N. Amati 217 6.2 5.0 1.72 Vertical

Violin 1671

7. F. Ruger 263 7.4 4.6 2.13 Vertical

Violin

8. M. Albani 216 4.8 3.3 1.12 Vertical

Violin 1674

9. A. 242 6.8 5.2 1.55 Vertical

Violin Stradivari
c.1680

10. Willems 238 7.0 6.1 2.82 Vertical

Violin c.1680

11. I. Schorn 254 5.6 3.5 1.32 Vertical

Violin 1696

12. M. Albani 234 6.3 5.3 1.69 Vertical

Violin c.1700

13. Pasta 249 5.6 3.7 1.50 Slab

Violin c.1700

14. H. Jacobs 244 6.6 4.3 1.63 Vertical

Violin 1702

15. A. 237 6.5 3.6 1.48 Vertical

Violin Stradivari
1703*
16. A. 249 6.3 4.0 1.88 Slab
Violin Stradivari*
1704

17. A. 241 6.6 4.7 2.09 Vertical

Violin Stradivari*
1719

18. F. Gobetti 252 6.8 4.0 2.16 Vertical

Violin c.1720

19. A. Gagliano 186 6.6 5.8 1.37 Vertical

Violin 1721

20. A. Gagliano 267 7.5 5.3 2.30 Vertical

Violin 1724

21. N. Cross 222 6.5 3.8 1.43 Vertical

Violin 1725

22. N. Cross 254 8.2 3.8 1.86 Slab

Violin

23. C. Tononi 225 4.5 4.4 1.28 Vertical

Violin 1730

24. D. Tecchler 239 5.3 4.0 1.35 Vertical

Violin 1734

25. S. Seraphin 244 6.2 5.3 1.62 Vertical

Violin 1736

26. C. Bergonzi 245 6.8 3.9 1.73 Vertical

Violin 1740

27. P. Guarneri 242 7.2 4.8 1.90 Slab

Violin Venice
28. R. Duke 251 7.9 4.7 2.68 Slab
Violin 1750

29. Januarius 244 7.2 4.3 2.14 Slab

Violin Gagliano
1750

30. S. Seraphin 257 6.8 4.6 2.41 Vertical

Violin 1757

31. C. Bergonzi 262 7.2 4.9 2.77 Vertical

Violin 1760

32. N. Gagliano 251 12.3 6.3 3.17 Vertical

Violin 1760

33. J. B. 249 7.1 4.6 1.78 Slab

Violin Guadagnini
1760

34. T. 277 7.7 4.5 2.46 Slab

Violin Balastrieri
1770

35. Gabbrieli 246 5.0 5.4 1.87 Slab

Violin 1774

36. G. Klotz 251 7.3 4.6 2.74 Vertical

Violin

37. Lupot 256 6.4 3.8 1.83 Vertical

Violin

38. Vinaccia 261 6.5 5.8 1.64 Slab

Violin

39. Lockey Hill 255 10.3 4.9 2.59 Vertical

Violin
40. J. Stainer 320 8.1 6.0 4.22 Vertical
Viola 1660

41. M. Albani 268 5.4 5.5 2.44 Vertical

Viola 1688

42. J. B. 269 8.7 4.4 2.16 Vertical

Viola Guadagnini
1757

43. F. Gagliano 304 8.9 6.6 4.80 Vertical

Viola 1783

44. V. Panormo 273 10.6 4.0 2.88 Vertical

Viola c.1790

45. V. Panormo 250 9.7 6.4 4.74 Vertical

Viola 1800

46. Rocca 1834 292 12.9 5.1 4.74 Vertical

Viola

47. A. Catini 544 13.3 14.8 25.06 Vertical

Cello 1660–1668

48. A. 533 18.3 10.7 25.30 Vertical

Cello Stradivari
1667

49. A. Gagliano 554 12.9 9.5 14.92 Vertical

Cello 1725

50. Januarius 563 17.3 13.4 26.17 Vertical

Cello Gagliano
1736
 * See Figure 43.

Figure 43 Original Stradivari bass-bars.

References
Patrizio Barbieri, “Giordano Riccati on the Diameters of Strings and Pipes,”
Galpin Society Journal 38 (1985), pp. 20–34.
Michel Corrette, Méthode théorique et pratique pour apprendre en peu de
temps le violoncello dans sa perfection (Paris, 1741)
François-Joseph Fétis, Antoine Stradivari, luthier célèbre connu sous le nom
de Stradivarius (Paris, 1856).
Francesco Galeazzi, Elementi teorico-pratici di musica (Rome, 1791).
William Henry Hill, Arthur F. Hill, and Alfred Ebsworth Hill, Antonio
Stradivari: His Life and Work (1644–1737), (London, 1902).
Marin Mersenne, Harmonie universelle: traité des instrumens à chords
(Paris, 1636).
Leopold Mozart, Versuch einer gründlichen Violinschule (1756), trans. by
Editha Knocker as A Treatise on the Fundamental Principles of Violin
Playing (Oxford, 1990).
Mimmo Peruffo, “The Mystery of Gut Strings in the 16th and 17th Centuries:
The Role of Loaded Weighted Gut,” Lute Society of America Quarterly
29/2 (May, 1994), pp. 5–14.
John Playford, Introduction to the Skill of Music (London, 1664).
Stewart Pollens, Stradivari (Cambridge, 2010).
Stewart Pollens, “Some Misconceptions about the Baroque Violin,”
Performance Practice Review 14 (2010), 1–13.
Stewart Pollens, “Transitional Violin Setup,” Journal of the Violin Society of
America: VSA Papers (2014).
Michael Praetorius, Syntagma Musicum II: De Organographia
(Wolfenbüttel, 1619).
Malcolm Rose and David Law, A Handbook of Historical Stringing Practice
for Keyboard Instruments (Lewes and Long Compton, 1991).
Abbé Sébastien-André Sibire, La Chélonomie ou le parfait luthier (Brussels,
1823; repr. Geneva, 1984; first published Paris, 1806).

Violin: transitional setup

What we term “modern” setup with fully inclined neck (generally set at
about 8 degrees) and non-wedge-shaped, solid ebony fingerboard 270 mm in
length, along with a stouter bass-bar, and the now ubiquitous French-style
bridge (with ovoid eyes, kidney-shaped central opening, and slender feet;
apparently developed around 1815) did not come into widespread use until
around the 1830s. The process of moving away from the Baroque setup,
however, was a gradual one that began in the second half of the eighteenth
century, and thus we cannot ascribe a codified set of dimensions as we do
with the modern “orthodox” setup. By the end of the eighteenth century,
violins that had been made in earlier times were having their necks reset and
reshaped and their shorter fingerboards replaced with longer ones. New
bridges had to be fitted in coordination with the increased neck angle, and
bass-bars were being removed and replaced with stronger ones. In 1804 the
violin collector and historian Count Ignazio Alessandro Cozio de Salabue
(1755–1840) wrote to a fellow enthusiast, Count Alessandro Maggi, that
upon purchasing violins in 1774 from Paolo Stradivari (the youngest,
merchant son of Antonio Stradivari) he sent them to the Mantegazza brothers
in Milan (active 1747–1801) for “thinning” – what we today euphemistically
refer to as “regraduation” – and replacement of the original bass-bars (Cozio
di Salabue, 1950). In effect Cozio was beginning the modernization process
on instruments that had been made just fifty years earlier in the Baroque
style. Presumably he also had work done on the necks and replaced the
original fingerboards, as those fitted by Stradivari were of insufficient length
to play all of the music of Mozart, not to mention the works of Beethoven,
Mendelssohn, and Paganini that were written within Cozio's lifetime. As an
amateur violinist, Cozio undoubtedly would have had his instruments refitted
to accommodate the latest repertoire. One of the most remarkable aspects of
the violin has been its chameleonic ability to adapt to different styles of
music and performance practice.
Paganini's 1743 violin made by Giuseppe Guarneri del Gesù, known as the
Cannone, is a notable example of a violin whose original neck was reset and
reshaped around 1828 by Carl Nicolas Sawicki, a Polish-born violin maker
who had settled in Vienna. However, even at this date Sawicki's alteration
was not what we would today consider a “modern” setup. For example, the
original neck was retained but was extended slightly at the foot so that it
could be angled back a few degrees. The extended neck was then renailed to
the top block, just as it had been in Guarneri's time, rather than being
mortised into the block as it is today. The back of the neck was also thinned
and squared off a bit at the heel, presumably to make the instrument easier to
play in higher positions. Sawicki then fitted a new Baroque-style, wedge-
shaped fingerboard, though one that was 262 mm in length, or considerably
longer than Guarneri's original, which was probably around 207–213 mm.
This new fingerboard is a bit shorter than the standard 270 mm length used
today. The tailpiece used by Paganini was 135 mm in length, or about 20 mm
longer than modern ones. Sawicki retained the old bridge (which may be the
original one made by Guarneri del Gesù), and thus the composite inclination
of the reset neck and new fingerboard was apparently adjusted to
accommodate it. However, despite this violin's considerable historical
importance and the vigorous debate as to how its setup may have contributed
to Paganini's virtuosic prowess, it was recently modified to bring it closer to
modern specifications (Carlson, 2004).
Though it is very rare to find violins that preserve either their original
fittings or transitional setups from the Classical and early Romantic periods,
one can occasionally find them in museum storerooms. Two such violins are
preserved in the Museo Correr in Venice: one is attributed to Santo Serafin
(1699–1776), the other to his nephew Giorgio Serafin (1726–1775).
The violin attributed to Giorgio Serafin retains its original neck as well as
its original fingerboard. There is the typical Baroque-style rectangular cutout
in the heel of the neck foot to accommodate the overhanging edge of the top
plate and the fingerboard also has the Baroque notch. The cutout in the heel
has a small rectangular wood fill and the fingerboard notch is mounted about
2 mm behind the upper edge of the top plate). The reason for this
displacement is unclear, as the foot of the neck does not appear to have been
extended and the top plate appears to be positioned properly relative to the
ribs. This unusual feature, however, is found on several violins with original
necks pictured in the Catalogo degli strumente dell' Istituto della Pieta
Venezia (1990). Vestiges of these fills can also be seen in some of the old
necks that have been extended at the foot.
The top block of this violin may be original, and the neck is presently held
in place by a single, rectangular-headed, flat-cut nail that has been driven
through the block and the upper rib. The wedge-shaped fingerboard is in the
Baroque style, with a core of spruce and a veneer of ebony on the upper
surface and maple edging on the sides. Both the neck and the maple facings
on the sides of the fingerboard are covered with the same varnish as the body
of the violin, which was the style in the Baroque period (for example,
Stradivari's necks and the sides of his fingerboards were also originally
varnished).
The neck of the Giorgio Serafin violin is inclined at an angle of about 3
degrees (similar to the inclination used by Stradivari), and the wedge-shaped
fingerboard also forms about a 3-degree angle, providing a composite angle
of about 6 degrees. The distance from the nut to the cutout in the neck foot is
118 mm, while the nut to the nick in the fingerboard is about 119.5 mm.
Because of the present gap between the neck cutout and the upper edge of the
top plate, the present distance between the nut and the upper edge of the top
plate is 121.5 mm. The “stop” of this violin is a somewhat lengthy 200 mm,
which means that a 2:3 mensur (the neck length-to-stop ratio) was not strictly
adhered to. A 2:3 ratio would have required a neck having a nut-to-notch
length of 133 mm.
The length of this original fingerboard is 218 mm, with a width at the nut
of 26 mm and a width at the bridge end of 38.5 mm. Because the neck does
not stand above the upper surface of the top plate (there is no “overstand” as
in the modern neck), the Baroque-style, wedge-shaped fingerboard provides
the needed offset, which in this case is about 8 mm at the upper edge of the
top plate. The bridge end of the fingerboard stands about 12 mm above the
top plate.
The old neck of the violin attributed to Santo Serafin is presently held in
place with a single flat-headed screw driven in through the non original top
block and the upper rib. The slot in the screw head is considerably off-center,
which suggests it is not of very recent manufacture, though it undoubtedly
replaces an earlier nail. The foot has had a wedge-shaped shoe added where it
meets the upper bout, and there is a thin, wedge-shaped piece added between
the button and the back of the heel, which causes the neck to stand proud of
the top edge of the body by about a millimeter and at an angle of about 5
degrees relative to the body. The neck has been thinned slightly and reshaped
at the foot. With the extension, the effective neck length is now 123 mm,
which is about 7 mm short of a 2:3 mensur (the stop is 196 mm).
The non original, transitional fingerboard is made of spruce veneered with
ebony on the upper surface and on the sides – giving it the appearance of a
solid ebony fingerboard. It is 256 mm in length and about 4 mm thick at the
nut, 7 mm thick at the bridge end, 25.5 mm wide at the nut, and 45 mm wide
at the bridge end. This fingerboard lacks the so-called Baroque notch, though
it is wedge-shaped and adds about an additional 2-degree angle, making a
composite angle of about 7 degrees. At the bridge end, the center of the
fingerboard sits about 15 mm from the surface of the violin's top. The overall
curvature is somewhat flatter than today's fingerboards.
The tailpiece of the Santo Serafin violin may be original. It has a walnut
core veneered with ebony, and it has two holes drilled in through the front for
the tailgut. Its length is 104 mm, with an upper width of 38 mm and a lower
width of 22.5 mm. The diameters of the four holes to accommodate the
strings are 1.95 mm (Pollens, 2014).

References
Bruce Carlson, “The ‘Cannone’ and the Original Fittings,” Atti del Convegno
Internazionale di Liuteria Recupero e conservazione del violino Guarneri
“del Gesù” (1743) detto “Cannone”, (Genoa, 2004), pp. 43–50.
Ignazio Alessandro Cozio di Salabue, Carteggio, transcribed and ed. Renzo
Bacchetta (Milan, 1950).
Stewart Pollens, “Transitional Violin Setup,” Journal of the Violin Society of
America: VSA Papers 24/2 (2014).

Violin sizes

It is often necessary to identify the size of violins, violas, cellos, and

double basses according to the modern classification system (¾ size, ½ size,
¼ size, etc.) for cataloging purposes or when ordering stock fittings for them
in the course of restoration, such as bridge blanks, tailpieces, fingerboards,
chinrests, pegs, endpins, etc.
Below is a key for determining the proportional size of such instruments.
Millimeter dimensions are approximate.
Suzuki and sizes are approximately equal to the traditional and
sizes.

Violin

⅞ ¾ ½ ¼ ⅛

Body 356 346 335 310 280 255 230 215 182
length
mm

Viola

Large Medium Small ¾ ½ ¼

Body length mm 430 410 390 356 335 310

Cello
 ¾ ½ ¼ ⅛

Body length mm 755 690 650 580 530

Double bass

¾ ½ ¼ ⅛

Body length mm 1160 1110 1020 940 850

Sources
MENC (Music Educators National Conference).
Henry A. Strobel, Useful Measurements for Violin Makers (Aumsville,
2003).

Violin and viola strings, modern

The following are the silking colors of several popular manufacturers of

violin and viola strings:

E A D G C

Corelli blue black purple yellow red

D'Addario green black yellow red purple

Pirastro green black red brown purple

Prim purple purple green red blue

Super Sensitive red blue gray green red

 Thomastik purple blue green yellow red

Vulcanite and ebonite

Vulcanite is a type of vulcanized rubber that was used in making

woodwind instruments. Ebonite is essentially the same material but with 7–
10% lampblack added to give it the color of ebony wood. Vulcanite and
ebonite were made by kneading between 30% and 50% sulfur with India
rubber or gutta percha and subjecting the mix to heat (approx. 150°C or
300°F) and pressure (about 55 lb) for about four hours. Upon cooling,
vulcanite and ebonite can be sawn, carved, drilled, tapped, and turned. At the
boiling point of water, they can be bent and will retain their shape upon
cooling. At slightly higher temperatures, they can be embossed or molded.
When turning or machining vulcanite and ebonite, the cutting tools should be
kept extremely sharp and held at a moderate rake angle. Soapy water can be
used as a cutting lubricant and coolant. Polishing can be done with emery and
crocus cloth, as well as hard felt laps charged with rottenstone and oil. Low
pressure and speed should be used to prevent the buildup of heat. Flat
surfaces can be leveled with abrasive stones (traditionally a water-of-Ayr
stone) followed by polishing with rottenstone and oil. Lampblack and oil
were sometimes used to deepen the color and to impart a high polish.
Vulcanite and ebonite are fairly stable, and woodwind instruments made
from them tend not to split like those of wood. Because they are fairly brittle
materials, they have a tendency to chip, and if dropped they may shatter.
Exposed surfaces sometimes develop a gray, brown, or greenish color due to
exposure to light, heat, moisture, and ozone, so vulcanite and ebonite objects
should be protected from these elements as well as possible. When the
surface has deteriorated, it also becomes somewhat absorbent and may
darken if exposed to liquids, such as oil, water, and solvents, though the
darkening may disappear as these liquids evaporate. While uncured and
partially vulcanized rubber are vulnerable to many solvents, including
petroleum distillates, vulcanite and ebonite are relatively immune to them,
and after testing, petroleum solvents may be used for general cleaning,
degreasing, removing adhesives, etc. To limit exposure to solvents, it is best
to apply them sparingly and remove them quickly with a soft cloth. It is also
advisable to use solvents that evaporate quickly. The author used hexanes to
remove adhesive tape residue and lipstick from the ebonite head joint of a
flute.

References
John S. Mills and Raymond White, The Organic Chemistry of Museum
Objects (London, 2003).
Ernest Spon, Workshop Receipts for Manufacturers and Scientific Amateurs
(London, 1909).
 W
Water

Distilled water is made by boiling water and collecting the condensate – a

rather slow process employing expensive stills. Double- and triple-distilled
waters are progressively purer. Deionized water is made by passing water
through ion-exchange resins. This is often coupled with filtration, the use of
activated charcoal, and reverse-osmosis membranes to absorb dissolved gases
and other impurities. To determine the purity of water that has passed through
ion-exchange and other filtration systems, pH and conductivity meters can be
used. The Brita water purifiers that are widely available employ a cartridge
containing activated charcoal and ion-exchange resins. Because the flow-
through is rather quick, water treated by this system is not thoroughly
deionized; however, this readily available filtration system may serve as a
stopgap, and water can be run through the filter several times to purify it
further, if necessary. Water filtration and deionization units that are
connected to the water mains are recommended if large quantities of highly
purified water are needed for washing and rinsing. Such units can be fitted
with multiple taps to provide ultra-pure water as well as potable water
(deionized water should not be consumed on a regular basis as it leaches
needed electrolytes from one's system).
Water softening (by using Calgon or sodium hexametaphosphate) does not
produce deionized water but merely replaces calcium ions with those of
sodium.
Highly purified water tends to absorb carbon dioxide from the atmosphere
(as well as other contaminants from containers including those made of glass
and plastic). The absorbed carbon dioxide reacts to form carbonic acid, which
may lower the pH of the water considerably. Materials that are susceptible to
exposure to carbonic acid (such as iron, lead, and lead-tin solders) should not
be soaked or rinsed for prolonged periods in purified water that has been in
contact with the air. Ultra-pure water (which is rarely, if ever, required for
musical instrument conservation) has an electrical resistivity of about 18
mega ohms and a conductivity of 5.5 × 10−6 siemens/m. Electrical
conductivity, pH, and total dissolved solids can be assessed with electronic
meters. Inexpensive hand-held combination pH/electroconductivity/total-
dissolved-solids/temperature testers are currently available for testing tap and
purified water. These should be calibrated with specially buffered testing
solutions that are supplied by the manufacturers of the meters.

Wax cleaning emulsions for finished wood

The following two emulsions were developed many years ago in the
Objects Conservation Department of the Metropolitan Museum of Art for
cleaning and polishing furniture.
Because of the combination of solvents and waxes, the first of these
emulsions has remarkable cleaning action and also supplies a wax finish.

Beeswax 30 g

Ceresin 30 g

Mineral spirits 120 ml

Xylene 60 ml

Ethanol 60 g

Water 120 ml

Triethanolamine 15 ml

1 Heat water in jacket of double boiler.

2 Melt beeswax and ceresin in double boiler.
3 Warm mineral spirits, xylene, and ethyl alcohol together in double
boiler.
4 Mix melted waxes and warmed solvents together away from heat.
 5 Mix water and triethanolamine.
6 Add water and TEA mixture to wax/solvent mixture.

Note: exercise extreme caution while warming flammable solvents, such as

the mineral spirits, xylene, and ethanol used here.
Apply with a soft cloth and polish with a fresh cloth. Do not use on cracked
objects or over loose veneer, as the waxes may interfere with later gluing
operations.
The following is a mild cleaning emulsion for wood (not as effective as
that described above).

Mineral spirits 445 g

Oleic acid 40 g

Triethanolamine 15 ml

Deionized water 500 ml

1 Mix oleic acid and mineral spirits

2 Mix TEA and water
3 Slowly add water/TEA solution to mineral spirits. It will start out
lumpy and thick and slowly become smooth with continued stirring.

Wheat paste

The following recipe is recommended for making a wheat paste for gluing
paper.

500 grams wheat flour

2.5 liters deionized water

Add water gradually until the paste achieves the consistency of cream; warm
in a double boiler for about ten minutes, but do not allow the paste to boil.
Thin with a little water for use. Apply thinly. Dry under pressure with wax or
silicone release paper between the glued paper and applied weight.

Wire gauges for early keyboard instruments

In the old Continental wire gauge systems (such as the so-called

Nuremberg, Berlin, and Vienna systems), the gauge numbers increase as the
diameters decrease. In the English and American systems, the order is
reversed: the smaller the gauge number, the thinner the wire. Before the late
nineteenth century, there was no absolute relationship between gauge
numbers and wire diameters, either from locale to locale, manufacturer to
manufacturer, batch to batch, or over time. For example, original strings
associated with the gauge marking “4/0” on a piano made around 1818 by the
Viennese maker Conrad Graf have a diameter of 0.78 mm, while those on
another piano of his dating from around 1826 have a diameter of 0.93 mm.
Latcham (2000) and Huber (1988) have concluded that there was a shift in
the Viennese gauge system in the 1820s and 1830s, and that the gauge
numbers gradually came to be associated with progressively thicker strings,
hence the disparity noted above. The lack of consistency among gauges of
wire made by different manufacturers was noted in 1811 in an article entitled
“Historische Beschreibung der aufrechtstehenden Forte-Pianos von der
Erfindung Wachtl und Bleyers in Wien” that appeared in the Intelligenz-Blatt
zur allgemeinen musikalischen Zeitung 17 (Leipzig, November, 1811), p. 75.
The variations in diameter were perhaps due to several factors: absence of
rigid standards or the means of enforcing them; imprecise measuring
instruments (the vernier caliper was not invented until 1851 and the
micrometer was not invented until 1848); the fact that the early drawing dies
used to make wire may not have been accurately or consistently sized to
begin with; and the fact that the holes in dies gradually increase in diameter
through wear, which was especially an issue when they were made of steel
rather than of the much harder materials used today (such as tungsten carbide,
ruby, or sapphire). Latcham (2000) postulates that there may not have been
three Continental systems; he suggests that Thomée's (1866) so-named
“Nuremberg” system (see below) was derived from wire found on a single
instrument, just as Huber's (1988) “Berlin” equivalents stem from gauge
numbers and original wire discovered on a particular Graf piano, and those of
the “Vienna” system derive from the caliper formerly used by the Streicher
firm (see below). Latcham posits that the Graf and Streicher firms may have
maintained their own standards rather than relying upon the gauges marked
on spools coming from different manufacturers. Nevertheless, evidence that
there were at least two distinct Continental gauge systems is found in the
equivalency table published in Claude Montal's L'Art d'accorder soi-même
son piano (1836).

Montal's equivalents for the Nuremberg and Berlin gauge systems

Nuremberg gauges Berlin gauges

0 1

2/0 0

3/0 2/0

4/0 3/0

5/0 5/0

6/0 7/0

7/0 8/0

8/0 9/0

9/0 10/0

10/0 12/0

11/0 14/0

12/0 16/0
 Montal's equivalents for the English and Berlin gauge systems

English gauges Berlin gauges

7 4 fort. [strong]

8 3 id. [the same]

9 2 id.

10 1

11 0 fin. [fine]

12 0 fort.

13 2/0

14 3/0 fin.

15 4/0 id.

16 5/0 id.

17 6/0

18 7/0

19 8/0 fin.

20 8/0 fort.

In the interest of historical integrity, restorers should follow the original

gauge numbers found on instruments, if they exist, rather than recalculating
gauge transitions to provide a smoother tension curve (see below). Though
some makers, such as Conrad Graf (see original gauge markings penciled on
the bridge of one of his fortepianos, Figure 34a), consistently marked gauge
numbers on their instruments, others either were inconsistent or did not
bother to mark them at all. If a survey of a maker's work indicates that he or
she did not mark gauge numbers, it is likely that any markings one does find
on a particular example may have been added at a later date by a tuner or
restorer. Such markings may not reflect the original gauges but represent the
gauges used in a later stringing, which might have been carried out with
heavier wire. If original gauge numbers are not present and there are no data
available from another instrument by the same maker, then one must
calculate a stringing schedule. Included here are string gauges taken from a
number of keyboard instruments that can be used for reference, though a
more complete list can be found in Malcolm Rose and David Law's A
Handbook of Historical Stringing Practice for Keyboard Instruments (2005),
which has stringing schedules for scores of instruments as well as tension
curves for various makers and schools of makers.
By employing Mersenne's Law (see Mersenne's Law) and an appropriate
tension curve, one can tailor a stringing schedule for an instrument if original
gauge markings are absent. Because Pythagorean scaling was used in many
early keyboard instruments, one might expect makers to have used the same
diameter of strings (wherever the scaling was Pythagorean) in order to
maintain constant tension, but the early makers discovered that strings
sounded better if their diameters were gradually increased from treble to bass.
Thus, as one descends from the top note, string tensions tend to rise gradually
until the string lengths deviate from Pythagorean scaling (generally around an
octave below middle C); the tensions then diminish sharply for the very
lowest foreshortened strings. Tension curves vary considerably according to
instrument type and from school to school and maker to maker. In some
instruments, for instance, the string diameters may only double from treble to
bass, yielding a relatively “flat” curve, whereas in others they may triple,
yielding a more pronounced curve. If one plots the tensions of each string,
one often finds that the gauge transitions do not always lie on a smooth curve,
which suggests that some makers did not calculate tensions mathematically.
Up until the end of the eighteenth century, the gauge numbers found on most
keyboard instruments appear in whole numbers, though around 1790 one
begins to find markings in half-gauges (Latcham, 2000). The availability of
wire in half-gauges permitted more transition points, and thus smoother
tension curves. For information on calculating the tension of overspun strings
(Figure 44) see Overspun strings.

Figure 44 Overspun string made by the author for a square piano by

Alpheus Babcock, c.1820. The overspinning is an iron ribbon that was
converted from wire by use of a rolling mill. The iron ribbon was
simultaneously annealed and blued at around 600°F in a kitchen oven before
being spun around an iron core.

In 1981, Grant O'Brien published a study on eighteenth-century stringing

practices that presents a graphical system for determining ideal transition
points from iron to yellow brass and from yellow brass to red brass. O'Brien's
method is as follows: on a sheet of semi-log graph paper, string lengths are
arranged on the logarithmic Y axis and pitches on the non-logarithmic X axis.
A series of parallel lines representing “safe” pitches for iron, yellow brass,
and red brass wire are drawn (to calculate the actual breaking frequencies of
different types of wire, see Scaling). Because most harpsichords exhibit so-
called “Pythagorean scaling” (meaning that the string lengths double or halve
on the octave), their plotted string lengths form a straight line on
logarithmically lined graph paper except in the bass, where the strings are
foreshortened (doubling the length down to the bottom note would have made
keyboard instruments impractically large), and in the treble, where they are
often lengthened in accordance with the increased tensile strength exhibited
by thinner gauges of wire (see below). As the string lengths become
foreshortened, the plotted points cross from the iron region of the graph into
the yellow-brass region, and finally into the red-brass region, indicating the
optimal points where one should switch from one type of wire to another.
This graphing technique is helpful when restringing instruments that lack
indications of the transition points. In some instruments the transition points
are denoted by the repetition of the gauge numbers.
O'Brien's graphing technique does not take into consideration the gradual
increase in tensile strength that occurs when wire passes from one drawing
die to the next (a phenomenon termed “tensile pickup”). In designing the
string lengths of their instruments, keyboard instrument makers were
evidently not only cognizant of the weakest gauge of wire that was employed
in a stringing sequence, but were aware that the thinner gauges of wire were
considerably stronger than the heavier gauges. This enabled them to deviate
from Pythagorean scaling in the extreme treble, which offered several
advantages: the treble section of the bridge could be moved back from the
edge of the soundboard onto a more acoustically active area, and more space
was available for harpsichord registers.
Some researchers are of the opinion that wire used in the stringing of
keyboard instruments was not annealed at any point in the entire drawing
process, and that it continued to gather strength from gauge to gauge.
However, a set of eight spools of brass wire associated with a baryton made
by Joachim Tielke and dated 1686 in the Victoria and Albert Museum were
tested by Martha Goodway and Jay Scott Odell in 1984, and they discovered
that the hardness and strength of the wire steadily increased as the diameters
decreased from 1.16 mm through 0.74 mm, then suddenly decreased at 0.73
mm, and increased again through the thinnest gauge, which has a diameter of
0.63 mm. This indicates that in drawing this batch of wire, it was annealed
before it was drawn through the 0.73 mm die. Modern piano wire drawn at
Mapes, Inc. (the principal manufacturer of piano and music wire in the
United States) involves 23 drawing steps to produce 24 gauges of wire from
gauge 25 (0.059 inches) to gauge 12 (0.029 inches). According to Andrew
Wills, the production supervisor of the firm, steel rods are purchased from an
outside manufacturer and are annealed by Mapes to insure uniformity of
hardness before drawing begins. In drawing down the wire, the wire is
annealed three times. Thus, there are four soft-to-hard sequences.
Unlike instruments such as guitars, lutes, viols, and violins, whose strings
are of equal length though of different pitch, keyboard instruments
accommodate most of the notes that their keyboards produce by tailoring the
string lengths according to the principle of Pythagorean scaling. As noted
above, Pythagorean scaling refers to the doubling or halving of string lengths
on the octave, but there is another ratio used in the Pythagorean system
(which includes the construction of the musical scale and tuning), and that is
the fifth, which forms a ratio of 3:2. In many historic harpsichords, one does
in fact find scratch marks or small holes in the soundboards and wrestplanks
corresponding to the placement of the C and F bridge and nut pins. These
marks were evidently used to accurately position the bridges and nuts, and
with the exception of the extreme bass and treble, they were generally spaced
in accordance with the two Pythagorean ratios of 2:1 and 3:2. Ruckers
harpsichords have positioning pins located at the C and F♯ points, with the
F♯s evidently positioned in accordance with meantone temperament rather
than Pythagorean scaling (Koster, 1998). When restringing early keyboard
instruments, one should carefully search for evidence of compass shifts and
scaling alterations, as such changes may render original or earlier gauge
markings inappropriate.

Brass and iron scaling

Michel Corrette stated in his Mâitre de Clavecin (1753) that harpsichords

having c3 string lengths of five pouces (1 pouce equals 27.75 mm; 5 pouces is
equal to 138.75 mm, so c2 would be about 278 mm) should be strung in brass.
However, in 1965 John Barnes proposed that short- and long-scaled keyboard
instruments were not necessarily designed for different types of wire, but
rather for different pitches. (See also Shortridge [1960] and Thomas and
Rhodes [1967], cited below.) Today, however, this theory seems to have lost
traction, and it is widely believed that brass stringing should be used in so-
called “short-scaled” keyboard instruments whose c2 strings extend up to
around 285–290 mm, while “long-scaled” instruments (defined as having c2
string lengths longer than 285–290 mm) should be strung with iron wire (see
Scaling). Around 1970, in accordance with this theory, Edward Ripin
restrung the 1720 Cristofori piano (c2 = 286 mm) in the Metropolitan
Museum of Art with brass wire. Scaling anomalies below c (possibly
resulting from an eighteenth-century compass alteration and the replacement
of the original soundboard in 1938/1939 [Pollens, 1995]) render that
instrument untunable much above a third below Baroque pitch when strung in
brass, which may indicate that that instrument may have been originally
designed for iron wire, despite the short scale. Though there may be a certain
logic to stringing short-scaled instruments in brass and long-scaled
instruments in iron, we must keep in mind that early keyboard instrument
makers did not possess modern engineering degrees, and that certain makers
or schools may have been tolerant of or perhaps preferred strings that were
slacker than those that were “critically stressed.” Furthermore, one should be
skeptical of those who attempt to “calculate” the pitch of stringed keyboard
instruments from the length of their c2 strings, because the “safety factor”
that is built into the scaling of all stringed instruments is an unknown variable
– it may vary between a semitone and a tone and a half, or more (see
Scaling).
While the brass-scale/iron-scale theory may in fact apply to early
harpsichords and clavichords, it breaks down completely when we consider
the fortepiano. Many south German and Austrian fortepianos, for example,
have c2 strings that are about 280 mm in length, yet they were originally
strung with iron wire, with brass wire used for the foreshortened bass strings.
A good number of late eighteenth- and early nineteenth-century English
pianos have slightly longer scaling c2 around 300 mm), though these too were
originally strung primarily in iron wire, again with brass in the bass.
Overspun bass strings were used in many early English square pianos. Some
makers mounted the brass and overspun strings on a separate bass bridge
exhibiting appropriately shorter scaling.

English gauges used in harpsichords and pianofortes

The following gauges were measured and extrapolated from original

strings found on a Broadwood square piano dated 1795:

2 0.007 in. 0.18 mm

3 0.008 in. 0.20 mm

4 0.009 in. 0.23 mm

5 0.010 in. 0.25 mm

6 0.011 in. 0.28 mm

7 0.0125 in. 0.32 mm

8 0.014 in. 0.36 mm

9 0.016 in. 0.41 mm

10 0.018 in. 0.46 mm

11 0.020 in. 0.51 mm

12 0.022 in. 0.56 mm

13 0.025 in. 0.64 mm

14 0.028 in. 0.71 mm

15 0.031 in. 0.79 mm

French gauges used in harpsichords

The first two columns of numbers in the table below are inch and metric
equivalents derived from Grant O’Brien’s “Some Principles of Eighteenth-
Century Harpsichord Stringing and their Application,” The Organ Yearbook
12 (1981) and his current website www.claviantica.com. These are followed
by metric diameters proposed by Denzil Wraight in “Principles and Practice
in Stringing Italian Keyboard Instruments,” Early Keyboard Journal 18
(2000). Despite discrepancies in gauge equivalences found in these and other
sources, some theorize that the gauge system used in France and throughout
continental Europe is the same as the Nuremberg system given below, and
that differences are simply due to inconsistencies in manufacture.

000 0.031 in.; 0.785 mm; 0.689 mm

00 0.027 in.; 0.680 mm; 0.629 mm

0 0.024 in.; 0.605 mm; 0.577 mm

1 0.021 in.; 0.538 mm; 0.524 mm

2 0.019 in.; 0.479 mm; 0.469 mm

3 0.017 in.; 0.426 mm; 0.420 mm

4 0.015 in.; 0.378 mm; 0.375 mm

5 0.013 in.; 0.337 mm; 0.336 mm

6 0.012 in.; 0.299 mm; 0.301 mm

7 0.010 in.; 0.266 mm; 0.269 mm

8 0.009 in.; 0.237 mm; 0.241 mm

9 0.008 in.; 0.211 mm; 0.215 mm

10 0.007 in.; 0.187 mm; 0.193 mm

11 0.0065 in.; 0.167 mm; 0.172 mm

12 0.154 mm
“Antwerp school” wire gauges

These were discovered on a Ruckers harpsichord dated 1618.

00 0.745 mm

0 0.687 mm

1 0.634 mm

2 0.585 mm

3 0.539 mm

4 0.498 mm

5 0.459 mm

6 0.424 mm

7 0.391 mm

8 0.361 mm

9 0.333 mm

10 0.307 mm

11 0.283 mm

12 0.261 mm

Nuremberg gauges

Thomée's (1866) gauges

 9/0 1.10 mm

8/0 1.06 mm

7/0 0.97 mm

6/0 0.87 mm

5/0 0.83 mm

4/0 0.76 mm

3/0 0.66 mm

2/0 0.60 mm

1/0 0.56 mm

1 0.51 mm

2 0.46 mm

3 0.41 mm

4 0.37 mm

5 0.32 mm

6 0.28 mm

7 0.25 mm

Vienna gauges

Kamarsch's (1828) “Vienna” gauges

 These were originally expressed in Zoll. There are gaps in Kamarsch's
original table of actual and calculated diameters; the diameters listed below
have been “fitted” using a drawing ratio of 1:1.120 (see Latcham, 2000).

8/0 1.339 mm

7/0 1.196 mm

6/0 1.068 mm

5/0 0.954 mm

4/0 0.852 mm

3/0 0.761 mm

2/0 0.680 mm

1/0 0.607 mm

1 0.542 mm

2 0.484mm

3 0.432 mm

4 0.386 mm

5 0.345 mm

6 0.308 mm

7 0.275 mm

8 0.246 mm

9 0.219 mm
Original strings found on a Conrad Graf fortepiano, Vienna, c.1826, no. 609

9/0 1.44 mm

8/0 1.30 mm

8/0½ 1.25 mm

7/0 1.14 mm

7/0½ 1.06 mm

6/0 1.01 mm

6/0½ 0.95 mm

5/0 0.89 mm

5/0½ 0.85 mm

4/0 0.79 mm

4/0½ 0.76 mm

3/0 0.72 mm

3/0½ 0.68 mm

2/0 0.64 mm

2/0½ 0.61 mm

1/0 0.59 mm

1/0½ 0.55 mm
 1 0.53 mm

1½ 0.50 mm

2 0.47 mm

3 0.42 mm

Wire gauge caliper used by the Viennese Streicher piano firm, c.1825 (source: Huber, 1988)

8/0 1.39 mm

8/0½ 1.32 mm

7/0 1.25 mm

7/0½ 1.19 mm

6/0 1.13 mm

6/0½ 1.09 mm

5/0 1.05 mm

5/0½ 0.99 mm

4/0 0.93 mm

4/0½ 0.90 mm

3/0 0.87 mm

3/0½ 0.82 mm

2/0 0.78 mm
 2/0½ 0.73 mm

1/0 0.68 mm

1/0½ 0.64 mm

1 0.59 mm

String gauge markings found on some representative instruments

[ ] denotes wire material deduced by the author, though not marked on the
instrument.

Bentside spinet by Baker Harris, London, 1771

Serial no. 93, compass FF–f3 with FF♯ omitted, c2=265 mm. Gauges
marked in pencil on nut.

FF 14

GG 13

GG♯ 12

AA♯ 11

C♯ 10

F 9

A 8

c♯ 7

g 6
 d♯1 5

c♯2 4

b2 3

Harpsichord by Burkat Schudi and John Broadwood, London, 1782, c2=350 mm

Gauges stamped on nut.

8-foot

CC 15

EE♭ 14

FF♯ 13

GG♯ 12

BB♭ 11

C 10

E♭ 9

F♯ 8

c 7

f 6

b 5

c2 4
 4-foot

CC 12

FF♯ 11

AA 10

C♯ 9

F 8

B♭ 7

c♯ 6

f♯ 5

c1 4

Harpsichord by Abraham and Joseph Kirckman, London, 1791

Gauge numbers stamped on wrestplank.

FF 13

GG 12

AA 11

BB 10

D 9

F♯ 8 [brass]
 A 8 [iron]

c♯ 7

f♯ 6

c1 5

c2 4

Harpsichord by Longman and Broderip, London, 1785, c2 = 343 mm

Gauge markings stamped on wrestplank.

8-foot

FF 16 [red brass]

GG 15

AA 14

BB♭ 13

BB 12

C 11

D 10 [yellow brass]

E 9

F♯ 8

A 8 [iron]
 c♯ 7

f♯ 6

c♯1 5

c♯2 4

4-foot

FF 11[red brass]

GG♯ 10

BB♭ 10 [yellow brass]

C♯ 9

E 8

G 8 [iron]

B 7

e 6

c♯1 5

c♯2 4

Harpsichord by François Blanchet, Paris, 1733, c2 = 342 mm.

No gauge markings, but these strings are believed to be original.

8-foot
FF 0.66 mm [red brass]

GG 0.63 mm

GG♯ 0.59 mm

AA 0.54 mm

C♯ 0.48 mm

D 0.47 mm [yellow brass]

F 0.42 mm

G♯ 0.38 mm

B 0.39 mm [iron]

c♯ 0.31 mm

g♯ 0.25 mm

d♯1 0.22 mm

No original wire above c♯2

4-foot

FF 0.51 mm [red brass]

FF♯ 0.48 mm

GG 0.41 mm

AA 0.35 mm

C 0.32 mm
 D♯ 0.32 mm [yellow brass]

A 0.28 mm

A♯ 0.30 mm [iron]

d♯ 0.25 mm

a 0.22 mm

d1 0.21 mm

a1 0.20 mm

d2 0.19 mm

g2 0.18 mm

a2 0.17 mm

c3 0.20 mm [?]

d3 0.20 mm [?]

Harpsichord by Pascal Taskin, Paris, 1786, c2 = 330 mm

Wire gauge markings on nut.

8-foot

EE 0

GG 1

BB♭ 2
 C♯ 3

E 4

G♯ 5

d♯ 6

c1 7

c♯2 8

g♯2 9

Harpsichord (fake Ruckers) by Pascal Taskin, Paris, 1782

The soundboard is fitted with an authentic Andreas Ruckers rose and bears
the date 1636. Gauge markings on wrestplank.

8-foot

EE 00 Rouge (red brass)

GG 0

AA 1

C 2

D♯ 2 Jaune (yellow brass)

F 3

G 4
 A 5

c 5 Blanc (iron)

f 6

a 7

g1 8

e2 9

4-foot

EE 2 Rouge (red brass)

GG 3

AA♯ 4

C♯ 4 Jaune (yellow brass)

E 5

G 6

A♯ 6 Blanc (iron)

d♯ 7

a♯ 8

g♯1 9

c3 10

Harpsichord by Michele Todini, Rome, c.1667

 A♯ 3

B 4

d♯ 5

g 6

c1 7

f♯1 8

c♯2 9

Harpsichord by Bartolomeo Cristofori, Florence, before 1700

GG 3

BB♭ 4

D 5

G 6

c 7

g 8

e1 9

c2 10

Harpsichord by Bartolomeo Cristofori, Florence, 1722

 8-foot

D 5

G♯ 6

d♯ 7

a♯ 8

f♯1 9

d♯2 10

4-foot

B 8

G 9

e1 10

d2 11

2-foot

D♯ 8

A 9

e 10

c1 11

Harpsichord by Bartolomeo Cristofori, Florence, 1722

 8-foot

D 5

G♯ 6

f♯ 7

a♯ 8

f♯1 9

d♯2 10

Cembalo traverso harpsichord by Bartolomeo Cristofori, Florence, before 1700

8-foot

F 5

A♯ 6

g 7

b 8

g1 9

e2 10

Fretted clavichord by C. G. Hubert, Bayreuth, 1756

C–B Gesponnen (overspun)

 c 2

d 3

f 4

a 5

f1 6

g2 7

Note: For other gauges used by this maker, see Vermeij (2000).

Unfretted clavichord by J. A. Hass, Hamburg, 1763, c2=285 mm

8-foot

FF 000

GG♯ 00

BB 0

D♯ 1

F♯ 2

B♭ 3

f 4

c1 5

c2 6
 d♯3 7

4-foot

FF 2

GG♯ 3

C 4

F 5

B♭ 6

Unfretted clavichord by C. Kintzing, Neuwied, 1763

C 1

G 2

B 3

f 4

b 5

f♯1 6

c♯2 7

a2 8

Unfretted clavichord by Carl Lemme, Braunschweig, 1787, c2=266

Markings on key levers.

8-foot

FF–E probably originally overspun (not marked)

F 0 [red brass]

F♯ 0 [yellow brass]

G♯ 1

A♯ 1 [iron]

B 2

d 3

f 4

c1 5

c2 6

f3 7

4-foot

FF 1 [red brass]

GG♯ 2

BB 3

D 4
 F 4 [yellow brass]

F♯ 5

G♯ 5 [iron]

Unfretted clavichord by J. C. Meerbach, Gotha, 1799, c2 = 261

Bass strings overspun; gauge markings on key levers.

8-foot

c 2

e 3

g♯ 3½

a♯ 4

d 4½

f 5

c2 5½

e2 6

d3 7

4-foot unmarked

Square piano by Broadwood, London, 1795, serial number 2998, c2=320 mm

 FF–F Marked “strings” (overspun strings)

F♯ 14 [brass]

G♯ 13

B 12

d 11

f 11 [iron]

b 10

g♯1 9

Set of original overspun strings from a Broadwood square, London, 1801

Brass core in mm/tinned copper open winding in mm.

FF 0.68/0.41

FF♯ 0.64/0.41

GG 0.64/0.43

AA 0.62/0.38

BB 0.55/0.33

C 0.53/0.37

C♯ 0.49/0.32

D 0.48/0.33
 D♯ 0.48/0.34

Fortepiano by Johann Schmidt, Salzburg, c.1780

This instrument is fitted with a pedalboard that extends the compass

downwards from FF to CC. The notes from FF through CC are overspun.

CC 0000000/4

DD 000000/4

FF 00000/8

FF♯ 00000

GG♯ 0000

BB 000

D 00

E 0

c♯ 1

c1 2

a1 3

g♯2 4

Fortepiano by Conrad Graf, Vienna, c.1826, c2=275 mm

Marked in pencil on bridge.

 FF 6/0 [brass]

AA 6/0½

C♯ 5/0

F 5/0½

A 4/0 [iron]

d 4/0½

g 3/0

c♯1 3/0½

f1 2/0

c♯2 2/0½

g♯2 0

d3 ½

g♯3 1

c♯4 1½

Original strings on a Conrad Graf fortepiano, Vienna, c.1838, serial number 2564

Gauges of overspun strings are not marked. Overspun strings have an iron
core and brass winding.

CC overspun, core 1.12 mm/winding 0.59 mm

CC♯ overspun, core 1.05 mm/ winding 0.53 mm
DD overspun, core 1.05 mm/winding 0.53 mm
 DD♯ overspun, 2.01 mm overall diameter
EE overspun, core 0.92 mm/winding 0.45 mm
FF overspun, 1.80 mm overall diameter
FF♯ overspun, 1.55 mm overall diameter
GG overspun, 1.38 mm overall diameter
GG♯ overspun, 1.30 mm overall diameter
AA, marked 8/0, 1.27 mm, brass
AA♯, marked 8/0 ½, 1.22 mm, brass
C, marked 7/0, 1.15 mm, brass
D♯, marked 7/0 ½, 1.05 mm, brass
F♯, marked 6/0, 0.97 mm, brass
A, marked 4/0, 0.90 mm, iron
d♯, marked 4/0 ½, 0.88 mm, iron
a, marked 3/0, 0.83 mm, iron
d♯1, marked 3/0 ½, 0.77 mm, iron
c2, marked 2/0, 0.75 mm, iron
f♯2, marked 2/0 ½, 0.70 mm, iron
c3, marked 0, 0.68 mm, iron
g3, marked 0 ½, 0.66 mm, iron
d4, marked 1, 0.60 mm, iron

Fortepiano by Nannette Streicher, Vienna, c.1807, c2=27.8

FF 7/0 [brass]

GG 6/0

AA 5/0

BB 4/0

D♯ 3/0

F♯ 3/0 [iron]

A 2/0
 c 1/0

f♯ 1

a1 2

f2 3

c♯3 4

g3 5

Fortepiano by Joseph Böhm, Vienna, c.1826

Brass bass strings on separate bridge. Gauges stamped on front edge of

soundboard.

CC 7½ [brass]

DD♯ 6

GG 6½

BB 5

D♯ 5½

G 4

A 3 [iron]

d1 3½

c2 2
 g♯2 2½

d♯3 0

a3 0½

d4 1

Fortepiano with down-striking action by Johann Baptist Streicher, Vienna, 1837, c2=273 mm

CC 12/0

CC♯ 11/0

DD 10/0

DD♯ 9/0

EE 8/0

FF♯ 8/0½

GG 7/0

GG♯ 7/0½

AA 6/0

BB♭ 6/0½

BB 5/0

C♯ 5/0½

D 4/0
 E 4/0½

F♯ 6/0

G♯ 6/0½

B♭ 5/0

c♯ 5/0½

f 4/0

e1 4/0½

a1 3/0

d♯2 3/0½

a2 2/0

d♯3 2/0½

g♯3 0

d4 0½

Claas Douwes' stringing schedules for “Flemish” harpsichords and clavichords

One of the earliest published lists of recommended string gauges and

scalings for keyboard instruments can be found in Claas Douwes, Grondig
Ondersoek van de Toonen der Musijk (Franeker, 1699). Douwes (Friesland,
c.1650–c.1725) was a schoolmaster and organ builder. The instrument length
designations, given in voets (feet) and duimen (“thumbs” or inches), are
believed to be nominal (for a thorough analysis see O'Brien, 1990). Douwes
does not indicate which city's unit of length he was using (see Historical
metrology), though his stringing schedules may be appropriate for
instruments made by the Ruckers family (Ruckers harpsichords and virginals
are often referred to as “Flemish”; however, Antwerp was not in Flanders and
the family's instruments are thus more properly designated as “Antwerp
school”).

For a 6-voets Klavecimbel (harpsichord) with c2 of 14 duimen

C 1 Red [red brass]

D, E 2

F, G 3 Yellow [yellow brass]

A, B [B♭] 4

H [B], c, c♯ 5

d–g 6 White [iron]

g♯–d1 7

eb1–a1 8

b1 [bb1]–e2 9

f2–c3 10

For a 5-voets Klavecimbel with c2 of 13 duimen

C 1 Red

D 2

E 3
 F, G 4 Yellow

A, B [B♭] 5

H [B], c, c♯ 6

d–g 7 White

g♯–d1 8 White

e♭1–a1 9 White

b1 [b♭1]–e2 10 White

f2–c3 11 White

For a 4-voets Klavecimbel with c2 of 10 duimen

C 2 Red

D 3

E 4

F, G 5

A, B [B♭] 6

H [B], c, c♯ 7

d–g 8 White

g♯–e♭1 9

e1–h1 [b1] 10
 c2–g2 11

g♯2–c3 12

For a 3-voets Klavecimbel with c2 of 7 duimen

C 3 Red

D 4

E 5

F, G 6 Yellow

A, B [B♭] 7

H [B], c, c♯ 8

d–a 9 White

b [b♭]–f♯1 10

g1–e♭2 11

e2–c3 12

For a 2-voets, 4-duimen Klavecimbel (no c2 given)

C 4 Red

D 5

E 6
 F, G, A 7 Yellow

B [B♭]–c♯ 8

d–g 9 White

g♯–e♭1 10

e1–c♯2 11

d2–c3 12

For a 5½ or 4-voeten Klavekordium (clavichord; no c2 given)

C 1 Red

D, E 2

F, G 3 Yellow

A, B [B♭] 4

H [B]–d 5

e♭–a 6

b [b♭]–f♯1 7

g1–e♭2 8

f2–c3 9

For a 4½- or 3-voeten Klavekordium (no c2 given)

 C 2 Red

D, E 3

F, G, A 4 Yellow

B [B♭]–c♯ 5

d–g♯ 6

a–f♯1 7

g1–e2 8

f2–c3 9

For a Klavekordium of about 2½ voeten

C 3 red

D, E 4

F–A 5 yellow

B [B♭]–c♯ 6

d–g♯ 7

a–f♯1 8

g1–e2 9

f2–c3 10
 The Talbot manuscript (c.1695) lists wire gauges for a harpsichord,
possibly an English harpsichord with a short-octave compass. The
handwriting of the MS is difficult to read, though several authors have made
efforts to decipher it (Hubbard, 1965; Mould, 1968; Martin, 2009). This
author's reading is as follows:

GG 11 brass or copper

AA 10 or 9

C 9 or 8

D 8

E♭ 7

E 7 or 6

F, G 6 or 5

A, B 5

c 5 or 4

c♯, d 4

to c1 3

then to a2 2

to b2, c3, d3 1

Up through c, gauges 5 or 4, the manuscript is fairly legible, but beyond

that it becomes difficult to read. The transition point to iron is not given.
Michel Corrette's Le Maitre de Clavecin (1753) gives the following string
gauges for the harpsichord and spinet:

FF 00 Red [red brass]

GG 0

BB♭ 1

GG 1 or 2

C 2 Yellow [yellow brass]

D 2

E 2 or 3

F 3

G♯ 4

A 4 or 5

B♭ 5

B 6 Yellow or 4 White [steel]

c 6 Yellow or 4 White

c♯ 6 Yellow or 5 White

d 5 White

e♭ 5 or 6

e 6

g 6 or 7
 g♯ 7

d1 8

c♯2 9

Gall's Clavier-Stimmbuch (1805) provides the following string gauges for

harpsichords, clavichords, and fortepianos.

Stringing for a small quilled instrument, measuring 5 Schuh (feet), 6

Zoll (inches) in length, 2 Schuh, 7 Zoll in width. Compass AA–e3

AA 1

D♯ 2

G 3

c 4

f 5 steel

h 6

f♯1 7

f2 8

Stringing for an ordinary quilled instrument. Compass FF–f3

FF 0

BB 1
 D♯ 2

G♯ 3

c♯ 4

f♯ 5

c♯1 6

g♯1 7

f2 8

Stringing for an ordinary clavichord. Compass C–e3

C 00 brass

D 0

F 1

c 2

d♯ 3

a 4

d1 5

g1 6

c2 7
 g2 8

Stringing for an ordinary clavier. Compass CC–e3

C 00

D 0

G 1

c 2

d♯ 3

a 4

d1 5

g1 6

c2 7

g2 8

Stringing for a fortepiano with a longer octave in Cammertone.

Compass FF–f3

FF 000

GG 00

AA 0
 C 1

D♯ 2

G 3

B 4

E 5

d♯1 6

g2 7

Stringing for a fortepiano with longer octave. Compass FF–g3

FF 000

GG 00

AA 0

C 1

D 2

F 3

A 4

d 5

a 6

f1 7
 f2 7 steel

Stringing for a fortepiano with an overall length of 6 Schuh. Compass

AA–d3

AA 00000 brass

BB 0000

D 000

G 00 steel

d 0

f 1

g 2

a 3

h 4

f1 5

Stringing for a Mozartschen fortepiano. Compass FF–g3

FF 0000000

GG 000000

C 00000
 C♯ 0000

E 000

G 00

H 0

e♯ 1 steel

g♯ 2

d1 3

g♯1 4

e♯2 5

h2 6

References
John Barnes, “Pitch Variations in Italian Keyboard Instruments,” The Galpin
Society Journal (1965), pp. 110–116.
Michel Corrette, Le Maitre de Clavecin (Paris, 1753).
Claas Douwes, Grondig Ondersoek van de Toonen der Musijk (Franeker,
1699).
Franz Josef Gall, Clavier-Stimmbuch oder Deutliche Anweisung wie jeder
Musikfreund sein Clavier-flügel, Forte-piano und Flügel-fortepiano selbst
stimmen, repairiren, und bestmöglichst gut erhalten könne. Nebst einer
Nachricht von einigen neuerfundenen musikalischen Instrumenten des
herrn Joseph Wachtl (Vienna, 1805).
Martha Goodway and Jay Scott Odell, The Metallurgy of 17th- and 18th-
Century Music Wire (Stuyvesant, NY, 1987).
Frank Hubbard, Three Centuries of Harpsichord Making (Cambridge, MA,
1965).
Alfons Huber, “Saitendrahtsysteme im Wiener Klavierbau zwischen 1780
und 1880,” Salzburger Museum Carolino Augusteum Jahresschrift 34
(Salzburg, 1988), pp. 193–222.
Karl Kamarsch, Jahrbücher des Kaiserlichen königlichen polytechnischen
Institutes in Wien 13 (Vienna, 1828).
John Koster, “Toward the Reconstruction of the Ruckers' Geometric
Methods,” Keilinstrumente aus der Werkstatt Ruckers zu Konzeption,
Bauweise und und Ravalement sowie Restaurierung und Konservierung
(Halle an der Saale, 1998), pp. 22–47.
Michael Latcham, The Stringing, Scaling, and Pitch of Hammerflügel Built in
the Southern German and Viennese Traditions 1780–1820 (Munich, 2000).
Darryl Martin, “The Native Tradition in Transition: English Harpsichords
circa 1680–1725,” Aspects of Harpsichord Making in the British Isles: The
Historical Harpsichord 5 (Hillsdale, 2009).
Claude Montal, L'Art d'accorder soi-même son piano (Paris, 1836).
Charles Mould, “James Talbot's Manuscript, VIII. Harpsichord,” The Galpin
Society Journal 21 (March, 1968), pp. 40–51.
Grant O'Brien, “Some Principles of Eighteenth-Century Harpsichord
Stringing and Their Application,” The Organ Yearbook 12 (1981), pp.
160–176.
Grant O'Brien, Ruckers: A Harpsichord and Virginal Building Tradition
(Cambridge, 1990), pp. 284–295.
Paul Poletti, “The Interpretation of Early Wire Gauge Systems – Fixed
Diameters or Proportional Relationships,” Matière et Musique: The Cluny
Encounter (2000), pp. 201–226.
Stewart Pollens, The Early Pianoforte (Cambridge, 1995).
Malcolm Rose and David Law, A Handbook of Historical Stringing Practice
for Keyboard Instruments (Lewes and Long Compton, 2005).
John Shortridge, “Harpsichord Building in the 16th and 17th Centuries,”
United States National Museum Bulletin 225/15 (1960), pp. 93–107.
W. R. Thomas and J. J. K. Rhodes, “The String Scales of Italian Keyboard
Instruments,” The Galpin Society Journal 20 (1967), pp. 48–62.
H. Thomée, “Untersuchungen über Draht- und Blechlehren,” Zeitschrift des
Vereines Deutscher Ingineure 10 (1866).
Koen Vermeij, The Hubert Clavichord Data Book (Bennebroek, 2000).
Denzil Wraight, “The Stringing of Italian Keyboard Instruments c.1500–
c.1650,” PhD thesis, Queen’s University Belfast (1997).
Denzil Wraight, “Principles and Practice in Stringing Italian Keyboard
Instruments,” Early Keyboard Journal 18 (2000).

Wood

See Dendrochronology and the dating of woodDisinfection,

disinfestation, and fumigationRelative humidityWoodworking
It has been estimated that there are over fifty thousand species of trees and
woody plants, with some 680 native to North America and approximately
1000 found in Europe; however, comparatively few species are generally
employed in making a particular type of musical instrument, so if one is
familiar with the general history of the instrument, it is relatively easy to
learn to identify many if not all of the woods used in its construction.
Monographs about makers and their work, conservation reports, and museum
catalogs can be consulted to learn which woods were customarily used. For
example, the flute maker Theobald Boehm wrote that he preferred two types
of wood, cocuswood and granadilla from South America, for the bodies and
head joints of his flutes, while he eschewed ebony and boxwood, which he
thought appropriate only for cheap instruments. Thus, when one encounters a
wood Boehm flute it will probably be made of either cocuswood or
granadilla, which can be readily distinguished. Stradivari made his internal
top, bottom, and corner blocks, as well as his liners, of willow as opposed to
spruce, which was more commonly used by violin makers, including those
active in Cremona. The Ruckers/Couchet dynasty of harpsichord and virginal
makers working in Antwerp during the sixteenth and seventeenth centuries
used relatively few varieties of woods to make their instruments: generally
beech for the jacks and registers, bog oak or bog chestnut for the black
accidental keys, cherrywood for the bridges, oak for the wrestplanks and
stands, poplar for the cases and internal framing, and spruce for the
soundboards, lower jack guide, and wrestplank veneer. The Blanchet and
Taskin workshop in Paris used a wider variety of woods: poplar, lime
(basswood), and occasionally pine for the case and framing, lime for the key
levers, fir or spruce for the soundboards, service woods (sorbus, alisier, or
cormier) for the jacks, holly for the tongues, beech for the cheek blocks,
fruitwood, service wood, beech, and walnut for the bridges and nuts, and oak
and beech for the wrestplanks. Some makers occasionally turned to less
commonly used woods; for example, one may occasionally find an Italian
flute made of olive wood, or a cello having a pine top or chestnut back and
sides. One-off instruments constructed by amateur makers are often made of
unconventional woods. For example, one might come across an Appalachian
dulcimer made of sassafras wood or an American farm-built piano with a
hemlock soundboard.
Most of the commonly used woods can be readily identified by visual
means without the aid of magnification or a microscope, though in certain
instances microscopic examination may lead to definitive identifications or
distinctions between likely candidates. Unfortunately, it is often necessary to
remove a sample from the instrument so that it can be examined under the
microscope. In certain instances this will not impair or disfigure an
instrument (for example, when sampling the wood used in the internal
bracing of a keyboard instrument), though with other types of instrument
sampling must be ruled out (for example, when identifying woods used in
making woodwind instruments or fine stringed instruments). Many studies of
musical instruments simply provide broad characterizations of the woods
used in their construction. For example, William Dowd's superb 1984
monograph on the Blanchet workshop identifies the wood used to make
bridges and accidental keys simply as “fruitwood.” In his introduction to his
monograph he points out that “short of carving off samples to be sent to
wood laboratories for scientific examination (a process that would probably
banish the perpetrator from a museum or collection for life), one must depend
on one's eye and judgment … Under two-hundred years of oxidation and dirt,
euphemistically called patina, similar species are frequently
indistinguishable.” On the other hand, some studies ostensibly provide the
genus and species of woods, but these attributions may be little more than
guesswork.
In identifying woods, the microscope may in fact be no more reliable that
the unaided eye, and in some cases its use may be misleading. For example,
in attempting to distinguish between spruce and fir, one searches for resin
canals, which are found in spruce but not in most varieties of fir; however,
resin canals may not be present in a small sample of spruce placed on a
microscope slide, resulting in the sample being misidentified as fir. Many
years ago the author sent a chip of wood from the original soundboard of the
1720 Cristofori piano to the US Forestry Service for microscopic
identification. The Forestry Service was not informed that the sample came
from an Italian instrument, and they identified the wood as Atlantic white
cedar. It seemed unlikely that an instrument maker working in Florence at
that time would have imported wood from the New World (especially as
Cristofori submitted bills to the Medici court for cypress wood grown locally
and in Crete; Pollens, 2013), so another sample was submitted along with
information about the origin of the instrument from which it came. The
Forestry Service identified that sample as Mediterranean cypress. Evidently,
Mediterranean cypress and Atlantic white cedar are similar when viewed
under the microscope.
Identifying woods under the microscope involves slicing or exposing
radial, tangential, and cross sections, which are then mounted on a glass slide
for examination. In some cases, a razor blade can be used to slice off thin
layers (the sample can be softened using steam or immersion in water), but an
embedded specimen sliced by a microtome does a better job at producing
sections that are thin enough for examination under the microscope by
transmitted light. One then consults reference works that provide keys for
identifying the salient features of various species, such as: J. D. Brazier and
G. L. Franklin, Identification of Hardwoods: A Microscope Key (London,
1961); H. A. Core, W. A. Côté, and A. C. Day, Wood Structure and
Identification (Syracuse, 1979); R. Bruce Hoadley, Identifying Wood
(Newtown, 1990); A. J. Panshin and Carl de Zeeuw, Textbook of Wood
Technology (New York, 1980); Fritz Hans Schweingruber, Anatomie
europäischer Hölzer (Bern and Stuttgart, 1990); and E. A. Wheeler, P. Baas,
and P. E. Gasson, IAWA List of Microscopic Features for Hardwood
Identification (Leiden, 1989). For example, to distinguish between European
walnut (Juglans regia) and American black walnut (Juglans nigra), one
searches for gashlike pits and crystals in longitudinal parenchyma (see
below), which are found in the American species but are not typical of the
European one. A microscope equipped with epi-fluorescence illumination (or
even a common hand-held “black light”) is useful for this work, as certain
types of wood exhibit distinct fluorescence. For example, holly fluoresces a
pale blue-gray color under long-wave ultraviolet, which can help distinguish
it from boxwood, which has a yellowish fluorescence.
It is sometimes possible to deduce where an instrument was made by
identifying the woods from which it is constructed. For example, if the
bottom and internal structural parts of a keyboard instrument are made of
yellow poplar or red oak, and the soundboard is of Sitka spruce (all of which
are native to North America and are not generally used by European makers),
it is likely that the instrument was made in North America rather than Europe.
However, one should keep in mind that lumber and veneers have been traded
for centuries, and species native to one continent have been introduced to
others, so identifying the region from which a wood originated may not
necessarily indicate where the instrument was made. For example, one can
find sixteenth-century French court furniture made of Brazilian rosewood,
and the fine mahogany used to make English harpsichords and pianofortes in
the eighteenth and nineteenth centuries came from Cuba and the Honduras.
The Italian violin maker Giuseppe Scarampella (1838–1902), who worked in
Paris between 1862 and 1865, wrote in 1889 that he preferred American
wood because of its large pores, and he believed that Stradivari used it as
well.
With export restrictions on the tropical hardwoods (for example Brazilian
rosewood), musical instrument manufacturers are turning to a wide variety of
African and South American woods that were rarely used prior to these
restrictions.

Some hardwoods used in musical instrument making

African blackwood (Dalbergia melanoxylon). Used in making modern

clarinets and oboes; distinguished from ebony by its open pores.
Alisier or alizier (see Service wood).
Amaranth or purpleheart (Peltogyne purpurea). Bright purple when
freshly cut, but turns brown with age. Used in decorative inlay of piano
cases.
Amboyna (Pterocarpus indicus). Used as a veneer and decorative inlay
in piano cases.
Apple wood (Malus pumila; Prunus serotina). Similar in appearance to
pear wood; occasionally used to make harpischord jacks.
Ash (Fraxinus excelsior, European or white ash; Fraxinus Americana,
Fraxinus nigra, black ash). Used in keyboard instrument case
construction; so-called Hungarian ash has a distinctive wavy figure, and
was often used to veneer pianos and reed organs.
Basswood (Tilia spp., European linden; Tilia europaea, European linden
or lime; Tilia Americana, American basswood or linden). Light tan, very
even texture; often used to make the key levers of modern pianos.
Beech (Fagus sylvatica, common European beech; Fagus grandifolia,
American beech). Excellent steam-bending properties. Bridges of some
English keyboard instruments are made of this wood; it was often used
to make jacks and tongues, as well as box slides in Italian harpsichords;
it was sometimes used to make violin purfling. It was rarely used to
make the backs, sides, and necks of violins.
Beefwood (Grevillea striata).
Berry wood (see Service tree).
Birch (Betula pendula, European silver birch; Betula pubescens,
European white birch; Betula alleghaniensis, American yellow birch;
Betula papyrifera, paper birch). Light tan color; often used in making
drum shells. The beautifully figured birch inlays found in fine early
American furniture and piano name boards are often mistaken for
satinwood.
Bois de rose (Dalbergia maritima, Dalbergia louvelii). Not to be
confused with wood from the tulip tree, this wood is yellow or pink
striped with red and is often used cross-grain in decorative edgework and
inlay.
Bois de violet, violetwood or kingwood (Dalbergia cearencis).
Occasionally found on elaborately veneered nineteenth- and early
twentieth-century French pianos.
Boxwood (Buxus sempervirens, European boxwood; Gossypiospermum
praecox, West Indian boxwood). Light golden color, hard, smooth
texture; often used in making the natural key platings of Italian
harpsichords, as well as the bodies of woodwind instruments such as
flutes, oboes, clarinets, and recorders. English harpsichord makers used
it to make treble jack tongues.
Briar root (Erica arborea). Used in key tops of some early Flemish and
German harpsichords.
Bulletwood (Manikara bidentata).
Cherry (Prunus serotina). Used to make early woodwind instruments.
Frequently used in solid and veneer form in German and Austrian
fortepiano cases, as well as the bridges of Flemish and Antwerp school
harpsichords.
Chestnut (Castanea dentata, Castanea vulgaris). Tan, rough, porous
texture. Key levers of Italian harpsichords and early pianofortes are often
made of this wood; it was rarely used to make the backs and sides of
cellos. It is often mistaken for oak, but lacks the conspicuous rays of oak
in radial section.
Cocobolo (Dalbergia retusa). An orange-brown wood sometimes used
in making woodwind instruments and keyboard accidentals.
Cocuswood (Brya ebenus). Dark brown, striped. This wood has an even
texture that takes a high polish; it was used in the nineteenth century to
make high-quality woodwind instruments.
Cormier (see Service wood).
Coromandel wood (see Macassar ebony).
Deal. The term “deal” is often used generically to denote any cheap,
coniferous wood. Early keyboard instrument case framing is often
identified as deal.
Ebony (Diospyros ebenum, Diospyros crassifiora). Black with gray or
tan streaks, and a smooth texture that takes high polish. Used to make
woodwind instruments, bowed string instrument fingerboards, nuts,
saddles, tailpieces, chinrests, and pegs, guitar bridges, and piano keys.
Sometimes used to make violin purfling.
Elm (Ulmus hollandica, Dutch elm, Ulmus procera, English elm; Ulmus
Americana, American elm). Used in veneer and structural case parts.
Figured elm is often mistaken for oak, but it lacks the conspicuous rays
that are typical of quartered oak.
Fig bark. Light tan-colored strips peeled from the fig tree. Early
Brescian bowed string instruments often have purfling made of this
material.
Grenadilla (Dalbergia melanoxylon). Also known as African
blackwood or mpingo. Used to make woodwind instruments.
Harewood. Wood (usually maple) that is dyed gray. Used in marquetry.
Hemlock (Tsuga canadensis). Rarely used as soundboard wood.
Holly (Ilex aquifolium and Ilex opaca). White, smooth texture. Used in
making harpsichord tongues and inlay.
Hornbeam (Carpinus betulus). Light-colored, very hard, close-textured
wood. Used to make action parts of European pianos and the soles of
wooden planes.
Jujuba (Frangula alnus, Frangula azorica). Also known as alder
buckthorn. Used by some early Italian violin makers to make pegs.
Kingwood or violetwood (Dalbergia cearencis). Brownish purple,
striped, swirls. Dense, takes high polish. Used to make woodwind
instruments and veneer of nineteenth- and early twentieth-century
pianos.
Kiri (Paulownia imperialis). Used to make the soundboards of the
Japanese koto.
Lignum vitae (Guaiacum officinale). Hard, heavy wood (specific
gravity 1.3 – heavier than water), used in parts of some modern piano
actions. Woodworking mallets are often made of this wood.
Lime (see Basswood).
Linden (see Basswood).
Locust, West Indian (Hymenaea courbaril). Cabinet-grade wood used
by French piano makers.
Macassar ebony (Diospyros melanoxylon). Occasionally used in case
veneers. Not solid black like Diospyros ebenum, but highly figured with
yellow and yellow-brown streaks.
Mahogany (Swietenia macrophylla, Swietenia mahogoni). Used in solid
and veneer form by English and European keyboard instrument makers.
Eighteenth-century mahogany imported from Cuba and West Indies
became the most popular case wood and veneer in England when
indigenous walnut grew scarce in the mid eighteenth century. So-called
African mahogany (Khaya ivorensis), which is increasingly being used
to veneer modern piano cases, is not a true mahogany.
Maple (Acer campestre, field maple; Acer hyrcanum, Balkan maple;
Acer opalus, Italian maple; Acer platanoides, Norway maple; Acer
pseudoplatanus, great maple). “Tiger stripe” or “fiddleback” maple is
commonly used to make the backs, sides, and necks of bowed string
instruments. Maple is also used to make the wrestplanks of keyboard
instruments and mechanical parts of modern pianos. Renaissance
recorders and flutes were often made of maple.
Mountain ash (see Service wood).
Mulberry wood (Morus alba). Coarse-grained wood used to make the
bodies of such instruments as the Turkish saz, kemençe and Afghan
rabāb.
Oak (Quercus alba, American white oak; Quercus robur, European oak;
Quercus rubra, American red oak). Case wood and framing of keyboard
instruments, organ pipes. Bog oak (Quercus robur) is oak that has been
unearthed from peat bogs and may be partly petrified. Is is black in color
and was often used to make the accidental keys of Flemish harpsichords
and virginals.
Olive (Olea europea). Yellowish brown streaked with brown, having a
hard, close texture. Rarely used to make woodwind instruments as well
as the key platings of Italian harpsichords.
Palisander (see Rosewood, Brazilian).
Palo santo or guaiac wood (Bulnesia sarmienti).
Pear (Pyrus communis). Tan to pinkish tan, fine, close texture. Used in
making harpsichord jacks. See also Swiss pear and Service wood. Pear
dyed black was often used in violin purfling.
Pear, Swiss. This is service wood that has been steamed to give it the
darker color of true pear wood. Used to make harpsichord jacks and in
marquetry.
Pernambuco (Caesalpinia eschinata). Used to make violin bows. The
wood of lesser quality or lacking flame is sometimes referred to as brazil
wood.
Plane tree (see Sycamore; Platanus occidentalis).
Plum (Prunus domestica). Used in making recorders and other early
woodwind instruments. Used as case wood and veneer in German and
Austrian fortepianos.
Poplar (Liriodendron tulipifera, American yellow poplar, also known as
the tulip tree; Populus Alba, white poplar; Populus nigra, black poplar).
Used in Italian, Flemish, and French harpsichords' case construction,
including internal bracing, key frames, etc. Occasionally used for
making the backs of violas and cellos. Used by many violin makers,
including Stradivari, for purfling.
Rosewood, Brazilian (Dalbergia nigra). Rich reddish to purplish brown
color, often with variegated black markings. Lovely aroma of roses when
sawn. Used to make the backs of classical guitars and in veneer form for
piano casework. Exportation from South America is now prohibited.
Rosewood, East Asian (Dalbergia latifolia). Similar in color to
Brazilian rosewood, but generally striped and lacking variegated black
markings.
Satinwood (Chloroxylon swietenia). Golden color with rich mottled
figure resembling satin or watered silk. Strong coconut aroma when
sawn.
Service wood (Sorbus domestica, Sorbus terminales). Also known as
sorbus and mountain ash. The French berry-bearing trees alisier (Sorbus
terminales) and cormier (Sorbus domestica) are varieties of service
wood; alisier is light grayish-pink, while cormier tends to be a darker
pink. Service wood was often used in France for making harpsichord
jacks, and it is sometimes used to make violin pegs. It is often mistaken
for pear. Steamed service wood develops a darker color and the
appearance of pear and is consequently referred to as “Swiss pear.”
Snakewood (Piratinera guianensis) or Leopard wood. Dark red with
markings resembling those of a snake or leopard's spots. Darkens with
age, reducing the contrast of these markings. Pre-Tourte-period stringed-
instrument bows were often made of this wood.
Sycamore (Platanus occidentalis; plane tree). Also known as lacewood
because of the appearance of its broad rays. A North American wood.
Sycamore maple (Acer pseudoplatanus). In the UK, the maple used in
violin making is often called sycamore. A European wood.
Teak (Tectona grandis). Used to make the backs of the Chinese pipa.
Tulip tree (see Poplar).
Tulipwood, Brazil (Dalbergia frutescens and Dalbergia decipularis).
Variegated rose-colored wood often used in inlay. Not to be confused
with American yellow poplar, also known as the tulip tree.
Violetwood (see Kingwood).
Walnut, American or black walnut (Juglans nigra). Used in solid and
veneer form for keyboard instrument casework. American walnut is
darker and has larger pores than European walnut.
 Walnut, European (Juglans regia). Solid and veneer forms for
keyboard instrument casework and some mechanical parts (such as
Italian harpsichord jacks, box slides, key levers).
Whitewood (see Basswood and Poplar).
Willow (Salix retusa). Used to make internal blocks and linings of some
Italian violins. Used as the core of some Italian Baroque violin
fingerboards and tailpieces.
Wutong (Firmiana platanifolia). Used to make the soundboards of the
Chinese qin (chin).
Zi (Catalpa kaempferi). Used to make the bottoms of the Chinese qin
(chin).

Softwoods

Cedar, Atlantic white (Cedrus libani, cedar of Lebanon;

Chamaecyparis thyoides; Juniperus virginiana, Eastern red cedar).
Occasionally used to make soundboards, case parts, key levers, and
hammer shanks of keyboard instruments.
Cedar, Spanish (Cedrela odorata). Used to make classical guitar necks.
Also called cigarbox cedar. Often mistaken for mahogany.
Cypress, Mediterranean (Cupressus sempervirens). Light tan color.
Mediterranean cypress was used in making Italian harpsichord cases and
soundboards, as well as flamenco guitars. This wood has a pungent,
peppery aroma when sawn.
Fir (Abies alba, true fir; Pseudotsuga menziesii, Douglas fir). Somewhat
stiffer than spruce. Used in keyboard instrument framing and
soundboards.
Larch (Larix decidua), Soundboard wood.
Pine (Pinus lambertiana, sugar pine; Pinus strobus, Eastern white pine;
Pinus sylvestris, Scots pine; many other species). Used in keyboard
instrument framing, soundboards, and infrequently for violin tops.
Spruce (Picea abies, Norway spruce; Picea engelmannii, Engelman
spruce; Picea sitchensis, Sitka spruce). Soundboards and tops of violins
are generally made of spruce, as are the internal blocks and linings of
many violins. Sitka spruce is often used in the US to make piano
soundboards. Also used for keyboard instrument framing and key levers.
Yew (Taxus baccata, European yew; Taxus brevifolia, Pacific yew). The
 European species was widely used to make the backs of lutes and
chitarroni. Individual ribs often include the heartwood and sapwood
(termed shaded yew) to accentuate the appearance of the multi-ribbed
structure.

Macro structure of wood

Most species of trees that grow in temperate climates form annual growth
rings (or year-rings). When the trunk of a tree is viewed in cross section, at
the center one finds the pith, followed by dark annual rings forming what is
called the heartwood, then lighter-colored rings termed the sapwood,
followed by the cambium and inner bark, and then the outer bark. Some
species of trees exhibit a clear distinction between the heartwood and
sapwood, while in others the sapwood may consist of only one or two year-
rings, or it may be indistinguishable from the heartwood. Only the cambium,
inner bark, and some cellular structures of sapwood are living tissue. The
heartwood does not contribute to the conduction of water or storage of
nutrients, but serves primarily as the structural support of the tree. Some
craftsmen deliberately discard the sapwood of some species because of its
lighter color and use only the heartwood, while others make use of the wood
right out to the last year-ring. As indicated above, the bodies of many early
lutes and chitarroni were made of so-called shaded yew, which displayed
both heartwood and sapwood as a decorative element. A dendrochronological
study of the wood used by Giuseppe Guarneri indicated that the last year-ring
of the spruce tops of a number of his violins often dated just a few years
before the date on the instrument's label, which indicates that this maker did
not discard the sapwood (see Dendrochronology and the dating of wood).
Knots are sections of branches that have become overgrown by the trunk of a
tree as it grows. Figure is a distinctive pattern (such as “tiger-stripe” or
“fiddleback” maple) that results either from the tree's normal structure or
from environmental disturbances. For example bear-claw figure results from
indented growth rings, while bird's-eye figure is created by indented wood
fibers that form numerous small circular disturbances. Bird's-eye figure is
commonly found in sugar maple (Acer saccharum), an American species.
Many European-made violins, such as those made by Joachim Tielke and
Jacob Stainer, feature backs and sides of bird's-eye maple, and thus it is
widely believed that they used New World maple.
Micro structure

Fibers. These are elongated, thick-walled, small-diameter cells that have

pointed ends. These provide structural support in hardwoods.
Lumen. A cavity enclosed by a cell wall.
Parenchyma. These are thin-walled cells used to carry and store
nutrients. Parenchyma are oriented longitudinally as well as radially.
Pit. A circular opening in the wall of a cell through which fluids pass.
Generally directly opposite a similar opening in an adjacent cell.
Pore. An exposed opening created when one cuts across a hardwood
vessel.
Rays. These are bands of cells extending radially through the tree trunk.
Resin canals. These are tubular structures that carry resin, which is
exuded by epithelial cells surrounding the canals.
Tracheids. Softwoods consist mainly of tracheids, which are aligned
longitudinally as well as radially. They conduct nutrients and provide
structural support for the tree.
Tyloses. These are walls of parenchyma that have collapsed into resin
canals.

Physical characteristics of wood

Because wood is an anisotropic material, it shrinks and swells at different

rates across its longitudinal, radial, and tangential axes in response to
changing temperature and humidity. The longitudinal shrinkage rate is
generally very small (typically 0.1–0.3%), so soundboards that are restrained
around their perimeters or by ribs running across the grain are likely to crack
or warp as the humidity changes. As woodwind instruments shrink due to
loss of moisture, their bores tend to become ellipsoidal due to differences in
radial and tangential shrinkage rates.
Two types of moisture are found in wood: bound water (water that is
chemically bound to cell walls) and free water (water that is found within
lumens, or cell-wall cavities). When live trees are chopped down, the green
wood may be saturated with water, which can contribute 30–200% of the
wood itself. Wood was traditionally seasoned by air-drying, but in modern
times it is often “kiln-dried,” which means that it is gradually brought from
its green state to a particular moisture content in an oven at around 214–
221°F (101–105°C). Kiln-dried softwood is generally brought to a moisture
content of 10–20%, while hardwood and softwoods used in cabinet making
and millwork are generally brought to a moisture content of about 6–8%. As
wood dries out during seasoning, it shrinks at different rates depending upon
the grain direction:

Shrinkage (%) of some commonly used woods

Radial Tangential
Wood
shrinkage shrinkage

Ash (Fraxinus excelsior) 5.0 8.0

Beech (Fagus sylvatica) 5.8 11.8

Boxwood (Buxus 5.0 9.0

sempervirens)

Cherry (Prunus serotina) 3.7 7.1

Chestnut (Castanea dentate) 3.4 6.7

Ebony (Diospyros ebenum) 5.4 8.8

Fir (Abies alba) 3.8 7.6

Holly (Ilex opaca) 4.8 9.9

Larch (Larix decidua) 3.3 7.8

Mahogany (Swietenia 3.0 4.1

macrophylla)

Maple (Acer saccharum) 4.8 9.9

Oak (Quercus robur) 4.3 8.9

 Pine (Pinus sylvestris) 4.0 7.0

Poplar (Liriodendron 4.6 8.2

tulipifera)

Rosewood (Dalbergia 2.9 4.6

nigra)

Spanish cedar Cedrela 4.2 6.3

odorata)

Spruce (Picea abies) 3.6 7.8

Walnut (Juglans nigra) 5.5 7.8

Note: Longitudinal shrinkage of the above woods ranges from 0.1–0.4.

Sources: www.jp.europeanwood.org and Wood Handbook: Wood as
an Engineering Material, 1999

Unfortunately, the tabulated shrinkage percentages that one finds in

numerous reference works (from which the above table was derived) are from
the green to kiln-dried state, which is of little utilitarian value to the
conservator, who must cope with the swelling and shrinkage of old,
thoroughly seasoned wood due to seasonal and daily changes in relative
humidity and temperature. After seasoning, wood reaches an equilibrium
moisture content (EMC), which means that at a given temperature and
relative humidity it neither gains nor loses water from the atmosphere. For
example, many woods have an EMC of about 9.2% at a temperature of 70°F
(21.1°C) and a relative humidity of 50%. At 40°F (4.4°C) and 50% humidity,
the EMC is 9.5%, while at 100°F (37.8°C) and 50% humidity the EMC is
8.7%. Thus, the amount of water maintained in the wood decreases as the
temperature rises, even if the relative humidity remains constant. There is
also a hysteresis (or lagging) phenomenon with regard to the absorption and
desorption of moisture; that is, if wood from a dry environment (say, 20%
relative humidity) is placed in a more humid environment (such as 50%
relative humidity), it will stabilize at a lower EMC than it would if it were
taken from a more humid environment (70% relative humidity) and placed in
the same 50% relative humidity. Because of this phenomenon, the EMC
values given in most reference books represent an average between the higher
and lower percentages of moisture content that are reached when wood is
dried and moisturized.
When wood absorbs or desorbs moisture it not only swells or shrinks, but
may also undergo shape distortion called warping. Wood that is quarter-sawn
(that is, sawn radially from the trunk) does shrink and swell across the grain
(year-rings) with changing humidity, but to a lesser degree than wood that is
slab-sawn (sawn tangentially from the trunk). This is the primary reason why
quarter-sawn wood is often used for making the tops and soundboards of
violins and other instruments; furthermore, slab-sawn wood also has a greater
tendency to warp, resulting in bowing, cupping, or twisting. If wood is sawn
at a particular EMC and the relative humidity subsequently drops, slab-sawn
lumber has a tendency to cup against the curvature of the year-rings; if the
humidity subsequently rises, the wood tends to cup in the opposite direction.
The direction and degree of bowing and twisting are difficult to predict. So-
called “reaction wood,” which is often caused when areas of the trunk are
compressed or in tension due to the weight of branches, has a tendency to
twist during drying. Turnings and holes bored in wood will shrink and swell
under changing humidity if they are concentric with the year-rings, though
they will also exhibit shape distortion if they are not concentric with the year-
rings.
It is unfortunate that many musical instruments are designed in such a way
that structural and acoustic parts are glued or fastened across the grain of
other parts. For example, piano, guitar, and lute soundboard ribs typically run
across the grain of the soundboard, so when the humidity changes, the
soundboard tends to expand or shrink at a different rate than the ribs. This
may cause the soundboard to bulge, crack, or come unglued from the ribs.
The rib structures of violin-family instruments tend to restrain their bellies
and backs, especially where the ribs cross the year-rings of these plates. In
dry weather, the bellies in particular have a tendency to develop cracks
because they are prevented from moving as they shrink. To prevent the
formation of cracks, the animal-hide glue that is used to glue the bellies to the
ribs is deliberately thinned with water to reduce its strength, so that the glue
joint will fail before the wood cracks. This is why violin bellies often have to
be reglued at the upper and lower bouts when the seasons change.
Instruments made of wood are preferably stored, exhibited, and played in a
stable, climate-controlled environment (see Handling, storage, and
transporting of musical instruments). Museums attempt to stabilize the
environment with specially designed climate-controlled cases (see Humidity
control) or with coordinated heating, air-conditioning, humidifying, and
dehumidifying equipment. Instruments that are not kept in climate-controlled
environments are subject to the vagaries of seasonal and daily changes in
temperature and humidity. In New York, for example, though indoor
temperature may be maintained year round at around 70°F (21°C) with
central heating and air conditioners, relative humidity may vary from 10% to
90%. Musicians who travel with instruments often go from one temperature
and humidity extreme to another, which places extraordinary stress on their
instruments.
When making and restoring musical instruments, one must be cognizant of
the fluctuations in temperature and humidity that an instrument is likely to
face. For example, if you are working in a humid environment and are about
to glue a soundboard back into a keyboard instrument that is destined for a
drier environment, consider placing the soundboard in a drying cabinet (such
as a box equipped with a heating element or even several incandescent light
bulbs) to reduce the amount of moisture in the wood before gluing it back. As
the soundboard regains moisture after gluing, it may bulge slightly, though in
the drier environment, the bulge will subside and there will be less tendency
for cracks to develop. Some Italian harpsichords have burn marks on the
inside surface of the instrument's bottom that look as though a small pile of
wood shavings had been set aflame inside the case. It is possible that the
soundboard was warmed over this little fire just before the maker glued it in.
Different wood species exhibit different moduli of elasticity, and though
many textbooks give specific figures for each type of wood, in fact most
species exhibit a range of elasticities. It is generally recognized that the stiffer
a piece of wood, the faster sound waves will travel through it. Some
instrument makers believe that stiffer wood produces better-sounding
instruments and more responsive bows, so they employ a Lucchi meter,
which measures the speed of sound in wood. This device is widely used to
sort and grade pernambuco wood used in making violin bows as well as
spruce used to make violin bellies.
The art conservation literature is awash with articles and books devoted to
the treatment of waterlogged wood that has been dredged up from the ocean
floor (see proceedings of the 1970, 1981, and 1984 conferences listed below).
The removal of salts and the controlled drying of wood that has been
saturated with sea water are problems that are fortunately rarely encountered
when restoring musical instruments. Many of the treatments that are used in
this type of conservation work involve the impregnation of woods with
substances such as polyethylene glycol, sorbitol, and even sugar. Furniture
and wood sculpture conservators are keen on impregnating worm-eaten wood
with various synthetic resins, waxes, and glues; however, it is inadvisable to
impregnate acoustical or mechanical parts of musical instruments with these
or any other substances. A possible exception would be the reinforcement of
the odd woodworm gallery with products such as Xylamon LX Hardening;
however, if an instrument is completely riddled with woodworm galleries and
is too weak to bear the stresses imposed by the string tension function of
playing, the instrument should probably be retired from use rather than
impregnating it or replacing the weakened wood with fresh material.
Furthermore, wood surfaces that were intentionally left bare by the maker
should never be oiled, sized, waxed, or varnished, as such treatments alter
acoustical characteristics as well as appearance. Even a wash coat of glue size
or shellac will stiffen a soundboard considerably, and thus raise its resonant
frequency. Keep in mind that any type of impregnation or coating is
irreversible.

References
Theobald Boehm, The Flute and Flute Playing (New York, 1964).
J. D. Brazier and G. L. Franklin, Identification of Hardwoods: A Microscope
Key (London, 1961).
Conservation of Stone and Wooden Objects: 1970 New York Conference
(London, 1970).
Albert Constantine Jr., Know Your Woods (New York, 1959).
H. A. Core, W. A. Côté, and A. C. Day, Wood Structure and Identification
(Syracuse, 1979).
William Dowd, “The Surviving Instruments of the Blanchet Workshop,” The
Historical Harpsichord 1, Howard Schott, editor (Brookline, 1984), pp.
17–107.
R. Bruce Hoadley, Understanding Wood (Newtown, 1980).
R. Bruce Hoadley, Identifying Wood (Newtown, 1990).
Romeyn Beck Hough, The Wood Book (Cologne, 2002).
A. J. Panshin and Carl de Zeeuw, Textbook of Wood Technology (New York,
1980).
Stewart Pollens. “Bartolomeo Cristofori in Florence,” The Galpin Society
Journal 66 (March, 2013), pp. 7–42.
Proceedings of the ICOM Waterlogged Wood Working Group Conference,
Ottawa 1981 (Ottawa, 1982).
Fritz Hans Schweingruber, Anatomie europäischer Hölzer (Bern and
Stuttgart, 1990).
Pierre Verlet, French Furniture of the Eighteenth Century (Charlottesville
and London, 1991).
Waterlogged Wood: Proceedings of the 2nd ICOM Waterlogged Wood
Working Group Conference (Grenoble, 1985).
E. A. Wheeler, P. Baas, and P. E. Gasson, IAWA List of Microscopic
Features for Hardwood Identification (Leiden, 1989).
Wood Handbook: Wood as an Engineering Material (Madison, 1999).

Woodwind instruments

Wind instruments that are made of wood or ivory tend to shrink as they
age. This does not affect their length as much as it causes the diameters of
bores to become smaller and ovoid. These changes adversely affect tuning,
pitch, and voicing. Aside from warping, which it is sometimes possible to
correct through the use of heat and moisture, the only means of correcting
tuning and intonation problems that arise from shrinkage would be to re-ream
the bore, alter the undercutting of tone holes, change the dimensions of flute
embouchures, and reshape the windways and labia of recorders. Such
interventions have long been considered ruinous to historic woodwind
instruments; for example, in 1791 Johann George Tromlitz railed against the
reboring of shrunken or warped joints of flutes. Nevertheless, many examples
of early flutes found in museum collections show evidence of bore
alterations, the enlargement of embouchures, and the recutting of tone holes.
Some otherwise competent makers have used electronic tuners calibrated for
equal temperament when retuning and voicing historic woodwinds, thereby
obliterating evidence of their original temperaments. Similarly, the pitches of
many historic woodwind instruments have been altered to adjust them to
modern early music “standards” (such as A=392, A=415, etc.).
About the only conservation procedures that can be sanctioned are
superficial cleaning, removing tarnish and corrosion from keys, springs, pivot
rods, screws, etc., straightening bent keys, making replacements for missing
ferrules, end caps, and bell rims (Figures 45a–45c), replacing broken or
missing springs, relapping tenons, repadding keys, adjusting the position of
flute end corks, and any tuning or voicing that can be accomplished by
applying removable materials, such as applying bits of beeswax to the edges
or undersides of embouchures and tone holes or pieces of removable tape to
the bores. New blocks can be made for recorders if the damaged originals can
be safely pressed out. All old parts (including old pads, lapping thread, and
broken keys, springs, etc.) should be saved and documented in conservation
reports (see Conservation reports).
Figure 45a Graves & Co. clarinet, Winchester, New Hampshire, c.1825,
before restoration.
Figure 45b Turning an ivory ferrule for the Graves & Co. clarinet.
Figure 45c The Graves & Co. clarinet after restoration.

Disassembly for cleaning

The tenons of most early wood and ivory flutes were originally wrapped
with linen or cotton thread that was held in place and lubricated with
beeswax, a mixture of beeswax and tallow, or goose grease; the London
makers Hawkes & Son recommended fresh butter or petroleum jelly for this
task (see Spon, 1909). However, substances such as butter, tallow, and goose
grease turn rancid and tend to harden over time, and thus are not
recommended. A combination of beeswax and petroleum jelly (a 2:1
proportion works well) can be used to lubricate thread-wrapped or corked
tenons, as well as the metal tenons of modern flutes.
After many years, woodwind instrument joints may become difficult to
disassemble because the old lubricants have thickened and become sticky.
The usual method of separating the joints is to gently pull as one twists back
and forth, taking care not to wrap one's hands around keys or other
mechanical parts that might get bent or broken, and disengaging or removing
any bridge keys before twisting. If the lubricant has solidified, it may help to
inject mineral spirits or a combination of mineral spirits and xylene to loosen
the joint. Another technique is to place the instrument in a warmer and drier
environment or apply a moderate amount of heat from a hairdrier or hot-air
gun, which will cause the tenon to shrink and the socket to widen – though
these techniques should be used with caution on wood and ivory instruments,
as they may lead to cracking. (For repairing broken tenons, see below.)
When rewrapping tenons with thread, Johann George Tromlitz (1791)
suggests starting at the free end of the tenon and wrapping the waxed thread
tightly and evenly towards the body of the joint. He then advises winding
backwards and forwards diagonally. This diagonal back-and-forth winding
helps hold the thread in place and prevents it from bunching up and
potentially locking the two joints together.
The removal of metal pivot pins that guide the keys in early wood or ivory
woodwind instruments is often difficult because the pins have corroded or the
wood or ivory has shrunk around them. These pins (which are generally made
of brass or nickel silver) often have a hammered or bent-over extension that
was intended to assist in removing them; however, if a fingernail or piece of
hardwood is not sufficiently strong to pull them out, the conservator may
have to resort to using pliers; these should have brass jaws or, better yet, a
lining of leather or cloth electrical tape to protect the pivot pin from being
damaged. Another approach is to use a pin-punch or a metal rod of slightly
smaller diameter than the pivot pin (such as a drill bit mounted backwards in
a pin vise) to push out the pivot pin. If that method is used, make sure the rod
does not slip and damage the surrounding wood. It is also a good idea to wrap
the head of the pin vise with tape to prevent it from damaging the body of the
flute should the pivot pin give way suddenly.
A set of jeweler's screwdrivers is generally sufficient for disassembly work
up through the Boehm system, though special long-blade screwdrivers
(available through woodwind instrument suppliers) are sometimes required to
reach screws in posts that are set far back from the ends of the joint. Short
pivot screws and the long rods used as pivots for tubular hinge keys are often
very difficult to remove due to corrosion. Penetrating (low-viscosity) oil or
mineral spirits may assist in loosening seized screws, though often, the slots
in the head of these screws and rods have become so worn or damaged that
the blade of a screwdriver cannot be properly seated. In this case, special
screw removal tools (available from woodwind instrument suppliers) can be
used to cut a new slot in the head. These are pushed against the head of the
screw and worked back and forth until a footing for a screwdriver is made. If
this does not work, or the head has broken off the end of the screw, a small
hole can be drilled in the broken end and a miniature screw extractor inserted.
Such screw extractors can be fabricated by grinding a carbon or high-speed
steel drill bit to a four-cornered pyramid, which is then tapped into the hole
made in the broken screw until it grabs. The free end can be held in a pin
vise, which is turned to unscrew the pivot screw or rod. As a last resort, a
broken or immovable screw can be drilled out with a carefully centered drill
bit just slightly smaller than the screw. For this operation, it is advisable to
use a lathe or drill press in conjunction with a post jig (or some means of
centering the head of the screw relative to the drill bit). The thin remains of
the screw can then be removed from the post, key, or hinge tube with a pair
of jeweler's tweezers.

Removal of broken springs

 Most woodwind instruments employ springs to keep keys, rings, and vents
open or closed, as the case may be. Early woodwind instruments (such as
one-key flutes) used slightly curved leaf springs that were often riveted to the
key; needle springs, which are generally mounted in holes drilled in metal
posts, were developed by Buffet in Paris in 1837 and have been in general
use ever since. Flat springs are still used in many instruments today, and
these are generally held in place with rivets or small screws. To replace a
broken leaf spring that is riveted in place, the part of the broken spring that is
attached to the key must be removed. If the broken part is carefully worked
free of the mushroomed head of the rivet, with any luck there may be enough
of the rivet left to reattach a new spring. If not, the remnants of the old rivet
will have to be filed off or driven out of the key and replaced. If the spring
was held in place by a soldered-on post, a new one will have to be soldered in
place. Woodwind instrument repair suppliers sell replacement springs to fit
just about every type of instrument. Many come preformed, but if they are
supplied flat, they must be bent into shape with a pair of pliers. For bending
back the tip of a leaf spring, the main part of the spring can be held with a
pair of pliers (which serves as a heat sink) so that only the exposed tip can be
fully softened by heating it red-hot and then allowing it to cool slowly.
Moderate curves in the body of the spring can be formed by pulling the
spring at an angle through a pair of round-jawed pliers. Flat springs can be
fabricated from flat stock or made by hammering down or passing wire
through a rolling mill. If made of steel, before it is attached to the key it
should be heated red-hot, quenched, and then annealed to spring temper (blue
oxidation color; see Tempering steel). After a small hole is drilled to admit
the pin or rivet, the pin or rivet is hammered down to secure the spring.
Replacing leaf springs that are held in place with screws is generally
straightforward. Myriad screws are available, though for early woodwind
instruments it may be necessary to turn them on a lathe (see Metalworking).
Needle springs generally engage their keys by being hooked to a slotted or
notched catch. To remove these springs, they must first be unhooked from the
key with a crocheting needle or a special tool. Needle springs are generally
tapered, so broken ones must be firmly grasped with pliers and pushed out
through their posts. A spring-punching pliers or a pair of pliers that has had a
slot milled or ground in one of its jaws (which permits the end of the broken
spring to pass through) can be used to drive out a needle spring that has
broken close to the post. If the spring is broken at the post, it can be tapped
out with a hammer and driver made from a needle spring that has had its
point ground down slightly. When replacing needle springs, they must be
securely pressed into their posts. A pair of pliers with a slot in one jaw
(which allows the free end of the spring to pass through) can be used to press
the spring in place. Some repair manuals (see References below) recommend
using a pair of diagonal cutters to press in the spring; however, the cutting
faces of the jaws will very likely mar the post, so this technique is not
recommended.

Adjusting the length of hinge tubing

Boehm system woodwind instruments generally have groups of keys,

rings, and linkages mounted on silver or nickel-silver hinge tubes that run on
steel pivot rods. The rods have a screw slot at one end and are threaded at the
other for attachment to posts. In wood-bodied instruments, like clarinets and
oboes, the posts are screwed into the wood, while on metal-bodied
instruments, such as flutes and saxophones, the posts are soldered to the
metal tubing. With wood-bodied instruments, shrinkage (caused either by
aging or dryness) may cause the hinge tubing to bind as the posts are drawn
closer together. This is easily remedied by use of special tubular hinge
cutters, which have a pilot that is inserted in the tube to establish a perfect
perpendicular orientation for cutters that shave off material from the end of
the tube as the tool is rotated by hand. If the tubing has worn down and
become loose, it can be extended slightly by using a pair of swedging pliers
(i.e. pliers with a hole drilled across the jaws so that moderate pressure can be
exerted against the tubing as the pliers are rotated back and forth). If this
technique does not lengthen the tubing sufficiently, or if the worn edge is
very close to a key or ring and thus too short to be lengthened by this method,
an extension can be silver-soldered onto it. The extension can be formed
either with a length of tubing that has the same diameter and wall thickness
as the original or by soldering on a length of solid rod, which is then drilled
out to the internal diameter of the attached tubing. One method is to turn
down the end of the solid rod on a lathe so that this stub fits snugly within the
hinge tube. This assists in aligning and holding the hinge-tube extension in
place during soldering. The hinge tube should first be squared off with a
tubular hinge cutter so that it fits squarely against the shoulder of the new
piece of solid rod. After soldering, the solid rod is then drilled out to the same
inner diameter as the hinge tube.
Rod keys are keys mounted on solid rods rather than tubes. These are
generally countersunk and held in place by pointed pivot screws mounted in
posts. If the rod is loose, the pivot screw can be turned in with a screwdriver
to tighten it. If the head of the pivot screw cannot advance further, the
counter-bore in the post for the pivot screw's head can be deepened with an
appropriately sized pivot screw countersink. If the rod key is tight, the hole in
the rod can be deepened slightly with a cutting tool shaped like the point of
the pivot screw.
Loose posts can sometimes be tightened by unscrewing them and placing
some powdered pumice or a bit of paper in the threaded hole; the post will
bear against the pumice or paper so that it can be tightened and maintain its
proper orientation relative to its pivot rod.

Broken tenons

For historic instruments that are not in regular use (such as those in
museums and collections), the best method is to attempt to glue the broken
tenon back in place, as this preserves the original tenon and bore. Epoxy glue
can be used for this repair, though animal-hide glue can also be used,
especially if the instrument is not subjected to condensed moisture from
playing. Replacing a broken tenon with a plug of new wood is a delicate and
complex operation that requires an accurate lathe. To do this, the joint must
be held by a mandrel and the bore very accurately centered. The broken tenon
is turned off (with care taken to preserve the length of the joint) and a boring
bar or fly cutter is then used to bore out the joint to provide a socket for a
new piece of wood that will serve as the tenon. If possible, this socket should
not intrude upon a tone hole. After the new piece of wood has been carefully
turned to the correct diameter, it is glued into the socket. The new tenon must
then be bored and reamed to the precise diameter or taper to match the
dimensions of the original tenon. Obviously, if the bore of the joint has not
been centered perfectly, or if it is out-of-round, there will be some difficulty
in achieving perfect continuity between the old joint and the new tenon. In
some cases, it may be necessary to leave the machine-made bore somewhat
undersize and to finish the work by hand using fine scrapers or abrasive
paper. A special tenon-and-socket cutting tool is also available for boring out
the joint. To use this tool, it is mounted in the headstock chuck of the lathe,
while the broken joint's bore is aligned by a pilot extending from the tool and
a live center mounted in the tailstock. As the joint is advanced against the
tool's two cutters by the tail-stock hand wheel it simultaneously cuts the
socket for the new tenon and refaces the broken end of the joint.

Repadding

The flat pieces of soft leather used in padding early woodwind instruments
were traditionally attached to the keys with molten sealing wax, which was
made with a combination of wax, resins, and pigments (see Sealing wax).
After the old pad and sealing wax have been removed (either by scraping
mechanically or by letting the key sit in acetone until the sealing wax has
dissolved), the key is inverted over an alcohol lamp (if acetone has been used
to remove the sealing wax, the key should be thoroughly dried before
subjecting it to an open flame) and heated sufficiently to melt enough sealing
wax to lightly cover the underside of the key where the new pad is to be
attached. The pad of leather is then pressed down onto the molten sealing
wax and the key is allowed to cool. After cooling, the pad and excess wax are
trimmed around the key with a sharp knife. An alternative to using this
traditional method is to glue the pad with viscous Paraloid B-72 (dissolved in
acetone), or even with modern hot-melt glues. In selecting leather for the pad,
it is important to use the correct thickness so that the leather forms a good
seal when the key is released. A bookbinder's skiving machine (such as the
Scharf-Fix) is the perfect device for reducing the thickness of skins, as it can
be finely adjusted to produce just the right thickness. It is not advisable to
bend the keys of early woodwind instruments in order to create a good seal
between the leather pad and the body of the instrument because old metal
keys are often brittle and may crack if bent. Tromlitz recommends fine,
white, downy leather for flute pads. This suggests the use of tawed (alum-
cured) skins, which is surprising, as they do not hold up well if they become
wet (water dissolves the alum leaving the skin vulnerable to decomposition).
Soft, oil-cured skins also make good pads. The hair side generally makes a
good seal with the hole. Tromlitz advises against oiling pads, as the oil
eventually becomes sticky causing the pad to stick to the hole. Old, sticky
pads can sometimes be revived by swabbing them with mineral spirits to
remove gummy deposits. A light dusting with finely powdered talc may solve
the problem of sticky, noisy keys.
Around the beginning of the nineteenth century, woodwind instrument
makers began using so-called “salt spoon” keys, which had hemispherical
key cups. These were fitted with “purse pads” (also called “stuffed pads”),
which were generally made of soft, white tawed kid or calfskin that was sewn
around a wad of lamb's wool. Thread was first stitched around the perimeter
of a disc of leather, a ball of wool was placed in the middle of the disc, and
the thread was then drawn up to close the pad. The “purse” was generally not
fully closed up, leaving the central area open and the lamb's wool partially
exposed. Like flat leather pads, these were cemented onto the key with
sealing wax. The sealing wax often permeates the wool to some extent.
With the development of the Boehm and other later keying systems, the
flute's “salt-spoon” key cups gave way to the modern flat cups that often
employ a small screw to secure the pads rather than sealing wax. The pads
designed to fit these new keys are flat rather than purse-shaped and generally
consist of kidskin or one or two layers of bladder or goldbeater's skin
wrapped around a flat wool-felt punching backed with a cardboard disc.
Flutes fitted with closed-hole keys generally have their pads pierced to admit
a small screw that secures the pad to the cup. A metal washer or plastic “tone
booster” is sometimes held in place by the screw. Open-hole or French model
flutes have a metal bushing or grommet that holds the pad in place. Open-
hole pads, including those used in oboes, must be punched to create an
opening, and special self-centering punches are available for this purpose.
Pads that are screwed to their cups are sometimes shimmed with full or
partial discs or rings of paper or cardboard to effect a perfect seal.
In repadding modern woodwind instruments, it is important to use pads of
the correct diameter and thickness. Woodwind instrument supply houses
generally stock a wide variety of diameters, thicknesses, and types of pads for
flutes, clarinets, oboes, bassoons, etc. If the old pads are of the proper
thickness, they can be used as guides. Modern flutes generally use bladder-
skin pads that are held in place with screws and washers. To replace them,
the keys must be removed and new pads screwed on. The pressure of the
washers may cause the thin bladder skins to become wrinkled; if so, they can
be moistened slightly and ironed with a key slick that has been heated slightly
with an alcohol lamp. The slick should be heated just enough to generate
steam, but not to the point where it scorches the bladder skin.
 Some woodwind instruments (such as oboes) use cork pads. These can be
purchased or fabricated from sheets of cork with sharp arch punches. They
can also be sliced from a turned cork rod or tubes using a very sharp knife or
straight razor. In either case, it is important to use fine-quality cork, as any
holes, pits, or other flaws will prevent a tight seal. As with conventional pads,
it is imperative to use cork of the correct thickness. When fabricating cork
pads, it is advisable to bevel the edges or backs slightly so that they fit snugly
into the key cup. To do this, the pad can be temporarily attached to a dowel
with double-stick tape, and the dowel twisted between one's fingers as the
cork is held at an angle against a sheet of fine sandpaper.
To repad keys that employ pads that are “floated in” with French cement or
shellac, the keys must first be disassembled from the instrument. The old
pads are then removed by heating each cup over an alcohol lamp and picking
the pad out with a needle spring. The old cement or shellac should be
removed with alcohol or acetone, or scraped clean if the key is lacquered or
has some other protective coating that needs to be preserved. To repad a key,
the new bladder-skin pad should first be pricked on the side with a needle
spring (this allows air and moisture to escape when the pad is heated by the
molten shellac). Shellac flakes are placed in the key cup (or stick shellac is
melted into the cup), the key cup is then heated over an alcohol lamp, and the
new pad is pressed in. The key is then remounted on the instrument and the
new pad is checked to see if it seats well. This is done with a narrow strip of
cigarette paper or bladder skin that has been glued to a matchstick or thin
sliver of wood. The paper or bladder skin is pulled between the pad and the
tone-hole seat while the cup is held down (either by pressing the key or
releasing it, as the case may be). The strip of paper or bladder skin acts like a
feeler gauge to determine whether there is even pressure all around the pad. If
not, the cup can be reheated until the shellac softens. To prevent the alcohol
lamp's flame from burning the wood body of the instrument, rotate the joint
so that the key cup is nearly on its side, but facing downward slightly so that
the flame reaches only the key cup and not the body of the instrument. When
the cement softens, a key slick is inserted between the pad and the tone hole,
and the cup is pressed down to reorient the pad. Finally, the pads can be
clamped in place overnight with special spring clamps; this helps seat the
pads. Another way of checking to see if all the pads are sealing is to cork or
tape up one end of the joint, cover all the finger holes, vents, etc., and blow
through the other end. It should be easy to detect the presence of leaks.
Late woodwind instruments often have cork pads under the touches. In
replacing these, it is important that they be neatly trimmed around the
perimeter of the key, that they fit squarely against the body of the instrument
when the key is depressed, and, most important, that they are trimmed to the
correct thickness, as this controls the height of the key pad above the hole. A
pad that is raised to the wrong height when the key is depressed can
adversely affect tuning or cause “stuffiness” if it does not open sufficiently.
Again, assuming that the old pads are the correct thickness, they can be used
as guides for cutting new ones. Cork is best trimmed with a razor-sharp knife;
it is sometimes helpful to moisten the blade with water. Because cork is so
soft and compressible, to make a smooth edge, most of the excess cork
should be cut back before the final slicing cut is made. A sandpaper file can
also be used for final shaping, but one should be careful not to abrade the key
or round off the bevel angle.
The end cork of flutes must be accurately positioned relative to the
midpoint of the embouchure or blow-hole to insure proper tuning. Generally,
the distance from the end of the cork to the center of the embouchure is equal
to the diameter of the bore at the center of the embouchure. This is about 17–
17.5 mm for modern flutes and 8–8.5 mm for piccolos. The swab sticks
supplied with flutes and piccolos often have a mark on them for checking the
position of the end cork. When removing end corks, they should be pushed
out in the direction of the tenon (because the head joint tapers outward
towards the tenon end of the head joint. New corks can be turned on a lathe.
They are sometimes screwed onto ivory or metal devices that thread into the
end caps; these devices permit the cork to be adjusted, and some have marks
engraved on them for adjusting the cork's position (Figure 8). Modern flutes
employ a metal face plate that is attached to a threaded rod that passes
through a hole drilled in the cork and is secured by a stopper nut. This nut can
be tightened so that the cork bulges out a bit to provide just the right fit. End
corks should be lubricated with cork grease (such as a mixture of beeswax
and petroleum jelly).
Modern flutes, oboes, clarinets, bassoons, and other woodwind instruments
are generally fitted with a number of tiny screws that permit adjustments
between keys and other parts that are interconnected. For example, in the
oboe, one screw regulates the pressure exerted by the B♭ and C vent keys in
the top joint, which together are opened by the F♯ key. The pressure exerted
by these two vent keys can be checked by pulling a strip of cigarette paper or
bladder skin across the closed pads; if one pad is found to be exerting greater
pressure than the other, the appropriate adjustment screw can be turned with a
watchmaker's screwdriver to equalize the pressure. (Providing step-by-step
instructions for regulating every type and size of woodwind instrument is
beyond the scope of this handbook. See References below for published
guides to adjusting modern woodwind instruments; strangely, firms such as
Selmer, Buffet, Lorée, Powell, etc. do not publish or supply service manuals.)
Dents in modern metal flutes can be pressed out very much like the dents
in brass wind instruments (see Brass instruments). Special mandrels are
available (or can be turned on a lathe) for removing dents from flute bodies
and head joints. Tenons in metal flutes can be expanded with adjustable
mandrels or shrunk with swedging dies. See Soldering for information on
soft- and silver-soldering. Metal flute ribs, tenon rings and plates are
generally soft-soldered, while posts and keys are generally silver-soldered.
Cracks in wood and ivory woodwind instruments have traditionally been
stabilized by pinning. This is an irreversible procedure and is not
recommended for repairing historically important woodwind instruments.
Pinning and crack filling are usually disfiguring in ivory, boxwood, maple,
and other light-colored woods, though these techniques can often be
successfully employed when repairing dark woods such as African
blackwood, cocuswood, ebony, granadilla, and kingwood. For pinning, brass
and nickel-silver wire were generally used in the past, but stainless steel,
titanium, and carbon fiber rods are better choices as they do not corrode
(which can cause additional splitting of the wood). To pin cracks, a series of
small-diameter holes is drilled through the wall of the instrument across the
crack, sometimes in a zigzag pattern. In “blind” pinning, the holes are not
drilled completely through the wall of the instrument. Pinning wire is
sometimes lightly threaded to give it a better grip. If the wire is threaded, one
end can be chucked in a hand drill or drill press and driven into an undersize
hole that has been drilled in the instrument's wall (the diameter of the drill bit
should be the root diameter of the threaded wire; see Metalworking). Before
the wire is driven in, a notch is sawn or filed across the wire so that when it is
fully screwed in, the saw mark or notch will be positioned just within the
hole, so that the wire can be broken off beneath the surface of the instrument
by bending it back and forth. This will leave a recess that can be filled with
wax, sealing wax, stick shellac, or other compound that has been
appropriately tinted to match the surrounding wood. If the wire is left flush
with the hole, it can be painted to match the wood, but this often chips or
wears off leaving the bright end of the metal wire clearly visible.
Crack fillers should be reversible with solvents that will not affect the
wood or finish of the instrument. Wax, wax–resin mixtures, sealing wax,
stick shellac, or some modern adhesives and compounds can be used. Cured
epoxy is difficult to remove and should not be used on historically important
examples. Fillers can sometimes be effectively disguised by coloring them
with filings or shavings from the same type of wood. Fillers should be left a
bit proud of the crack while they set; after hardening they can be carefully
scraped or taken down with various grades of abrasive paper or cloth until
they are perfectly level with the surrounding wood. With historically
important instruments, if a decision is made to fill cracks, it may be best to
use fillers (such as wax or resins) that can be leveled and polished with
solvents rather than by mechanical or abrasive means.
It should be noted that it is improper to adapt keys to accept modern-style
pads (such as sawing off original “salt-spoon” cups and soldering on modern
Armstrong cups). (Figure 46). Assuming that one can secure the proper types
of leather, bladder skin, lamb's wool, etc., historic-style pads are relatively
easy to make. Though it is generally advisable to follow historic practice
when setting up and adjusting musical instruments, some of the old nostrums
(such as Hawkes & Son's recommendation for the use of butter to lubricate
tenons cited above) are obviously not suitable from a conservation
standpoint.
Figure 46 Steps in making a key from the firm of H. Bettoney made for a
didactic display at the Metropolitan Museum of Art, c.1900, starting with a
rod of nickel silver, rolling, hand-forging, and final assembly.

If an old woodwind instrument has not been played regularly, it is

advisable to “break in” the instrument by gradually increasing the playing
time, beginning with perhaps ten minutes the first day and gradually
increasing playing time to about an hour over a period of a week or more.
This allows the moisture level in the wood to build up slowly. It is important
to keep in mind that condensed moisture drips down the bore, where it is
adsorbed, while the interior and external surface remains comparatively dry;
this creates stresses that may result in cracking or warping. Woodwind
instruments were traditionally swabbed with oil to repel moisture. Almond oil
was often used for this purpose, though Tromlitz felt it had too little body and
recommended rapeseed oil (now called canola oil in the USA). So-called
“non-drying” vegetable oils are apt to become rancid, whereas “drying oils,”
such as linseed or walnut oil, quickly harden and form a varnishlike film,
which may not be desirable. Mineral oil (not traditionally used) avoids the
problems of rancidity and is nonhardening. Oiling was traditionally done
periodically – not before each use. Tromlitz advises using a feather to
distribute a light film of oil around the bore. Woodwind instruments that are
no longer played should not be oiled.

References
Anthony Baines, Woodwind Instruments and Their History (New York,
1962).
Theobald Boehm, The Flute and Flute-Playing in Acoustical, Technical, and
Artistic Aspects, trans. Dayton C. Miller (New York, 1964).
Erick D. Brand, Band Instrument Repairing Manual (Elkhart, 1978).
Adrian Brown, The Recorder: A Basic Workshop Manual (2009).
Ernst Ferron, The Clarinet Revealed (Paris, 1996).
J. James Phelan and Lillian Burkart, The Complete Guide to the Flute and
Piccolo (Shirley, MA, no date).
Johann Joachim Quantz, Essay of a Method for Playing the Transverse Flute,
trans. Edward R. Reilly (New York, 1966).
Georgina M. Rockstro, A Treatise on the Construction, the History, and the
Practice of the Flute, Including a Sketch of the Elements of Acoustics and
Critical Notices of Sixty Celebrated Flute-Players (London, 1928).
Ronald Saska, A Guide to Repairing Woodwinds (Glenmoore, 1997).
Carl J. Sawicki, A Method for Adjusting the Oboe and English Horn
(Fredericksburg, 1986).
Ernest Spon, Workshop Receipts for Manufacturing and Scientific Amateurs
vol. III (London, 1909), pp. 207–210.
Johann George Tromlitz, Ausführlicher und gründlicher Unterricht die Flöte
zu spielen (Leipzig, 1791).
Johann George Tromlitz, The Virtuoso Flute-Player, trans. Ardal Powell
(Cambridge, 1991).
Yamaha Band Instruments Repair Manual (Hamamatsu, n.d.).
Woodworking

Because modern woodworking machinery and manufacturing techniques

affect the design, structure, and appearance of objects made using them, it is
essential that traditional woodworking skills and techniques be employed
when restoring early instruments as well as in making copies of them.
One of the greatest problems facing those who must work with wood is
reducing lumber to usable dimensions. Up until modern times, there were
specialized tradesmen who would saw up lumber to a cabinetmaker's or
instrument maker's specifications. In the Encyclopédie, ou dictionnaire
raisonné des sciences, des arts, et des métiers (Paris, 1751–1772) (edited by
Denis Diderot and Jean le Rond D'Alembert) and the Encyclopédie
méthodique (Paris, 1782), there are engravings of pairs of workers resawing
lumber using a frame saw. Today, instrument makers generally must saw up
lumber themselves. Fortunately, most of this work can be done with
relatively inexpensive motorized band saws, table saws, and radial-arm saws,
with final dressing done with motorized thickness planers, jointers, and belt
sanders. However, wood “machined” in this fashion often produces a sterile
effect, and parts made this way often will not harmonize with old woodwork,
so it is often necessary to employ traditional tools and techniques to achieve
satisfactory results.

Marking out

There are essentially two ways of achieving dimensional accuracy when

using hand tools: “marking-out” (or “laying out”) and the use of jigs.
Marking-out involves the use of straight edges, rules, dividers, proportional
dividers, sectors, compasses, try squares, miter squares, sliding bevels,
protractors, marking and cutting gauges, etc. to lay out the design directly on
the piece of wood, using either a sharp point or a knife blade rather than a
pencil. On the inner surfaces of early harpsichord and fortepiano soundboards
and bottoms, as well as the interior surfaces of soundboards of lutes, guitars,
and violins, one can often see vestiges of the scribe lines that were used in
preliminary shaping as well as for positioning parts that were glued in place
(such as soundboard barring and structural braces). These scribe lines can
also be found on keyboards, where they were used to accurately position the
mortises for front and balance pins, and on small action parts, such as jacks,
where they were used to mark out the positions of pivot pins, slots for
tongues, plectra, etc. After the dimensions were marked on the wood, the
cutting and shaping operations commenced. Unlike a pencil line, a scribed
layout line partakes in the shaping process and lends neatness and precision
to the work.
Jigs are especially useful when many identical parts must be accurately
fabricated, such as harpsichord jacks and piano action parts. A jig may be as
simple as a miter box fitted with a fence or stop to facilitate the accurate,
repetitive positioning of pieces of wood from which parts are sawn, or it
might be a purpose-built device for positioning the workpiece or a tool
relative to the workpiece. For example, a planing jig (with guide rails for the
sole of the plane set at a specific height) can be used to accurately dimension
and taper the thickness of harpsichord jacks. A drilling jig may be used for
sequential or repetitive drilling operations to assure that groups of holes are
accurately positioned relative to one another.

Wood-shaping techniques

André-Jacques Roubo's L'Art du Menuisier (1769–1774) and the

Encyclopédie contain a wealth of information about general woodworking as
well as the crafts termed menuiserie (solid-wood furniture making) and
ébénisterie (veneered-furniture making) (Figure 47).
Figure 47 Woodworking tools, including rabbet and bullnose planes, small
dovetail saw, gouges, chisels, spokeshave, knives, and scrapers.

Sawing

The frame saw has been used for making straight cuts in wood since at
least medieval times. It uses a relatively narrow and flexible blade that is
mounted between two end pieces that are connected by a central beam. The
blade is brought under tension by a twisted wire or a cord. The bow saw is a
smaller version generally fitted with a narrower blade that is used for sawing
curved shapes. In both the frame and the bow saw, the blade can be rotated so
that the frame can clear the work; nevertheless, the size of the frame may
limit the depth of the cut.
The so-called hand saw uses a wider blade with a handle at one end. The
blade is broad at the handle and tapers towards the far end. Though the blade
is fairly thin and somewhat flexible, its width allows it to follow a reasonably
straight line, and because there is no framework surrounding the blade (as in
the frame or bow saw), it is not limited with regard to the depth of the cut.
Hand saws are generally 22 to 26 inches in length and come with either
chisel-shaped teeth for ripping with the grain or knife-shaped teeth for cross-
cutting. A slightly smaller version, called the panel saw, comes in lengths of
20 to 22 inches and generally has finer cross-cut teeth. Common metric sizes
are 300 mm, 350 mm, 500 mm, 550 mm, and 600 mm. The panel saw was
chiefly designed for sawing plywood, but it can be used on solid lumber as
well.
Several techniques exist for sawing a tree trunk into lumber: in plain-sawn
wood the year-rings are tangential or nearly tangential to the face of the
board; in quarter-sawn wood the year-rings are perpendicular to the face of
the board; in rift-sawn wood the year-rings are at 30 to 60 degrees relative to
the face of the board.
Back, dovetail, tenon, and bead saws have their non-cutting edge
reinforced with a heavy folded-over length of brass or steel, which adds
rigidity to the relatively thin blade. These saws are used for cutoff work,
mitering (often in conjunction with a miter box), and cutting dovetails and
tenons. They come in a variety of sizes and numbers of teeth per inch. Most
are equipped with cross-cutting teeth, though various models are now
available with rip-cutting teeth, which are designed for sawing tenons with
the grain. So-called gentleman's saws are scaled-down versions of dovetail
and tenon saws. The “blitz,” or model-maker's saw has very fine teeth, which
are useful when sawing out old parchment hinges in square piano escapement
jacks, for example.
Coping and scroll saws employ a U-shaped frame to hold a very narrow
blade under tension. They are used in sawing very fine and tight curves,
mostly in thin material. The scroll saw has a deeper frame than the coping
saw, permitting it to reach further. In earlier times, treadle-operated scroll
saws and fretsaws were used for sawing out intricate designs in wood and
veneer. The jeweler's saw is a more delicate version of the coping saw.
Jeweler's saw blades are primarily designed for working with metal, though
they can be used to cut designs in thin pieces of wood (such as veneer), as
well as mother-of-pearl, tortoiseshell, and ivory. When using a jeweler's saw
for cutting out small, intricate shapes, it is helpful to use a jeweler's slotted
bench pin with a V-shaped opening to support the work. This clamps onto the
edge of the workbench and projects from it, providing a platform for the
workpiece to rest on, while the V-shaped cutout permits the blade to be
brought up to the workpiece. The saw is held vertically and operated with one
hand while the workpiece is held against the platform and positioned and
moved against the blade by the other hand. When sawing veneer, a traditional
technique employed a “donkey,” which was a bench equipped with a foot-
operated vise for supporting the veneer (generally backed up with a piece of
wood called a “waster”). A foot-operated vise enabled the workpiece to be
quickly repositioned relative to the saw, which was hand-operated (see
section on Veneering below).
Keyhole, pad, and compass saws have narrow blades that come to a point.
They are useful for starting straight or curved cuts in the middle of wide
boards, for example, cutting a circular opening in a panel that is too deep for
the reach of a bow saw.
To sharpen and condition wood saws, the teeth should first be leveled with
a single-cut flat file, then set with a tool called a saw-set, which bends
alternating teeth to the left and to the right by an amount determined by the
number of teeth per inch (saw-sets employ a graduated anvil to adjust the
amount of deflection), and finally sharpened with a triangular (“three-
square”) file using a saw clamp to hold the blade straight and prevent it from
vibrating while filing. Rip saw teeth are filed at a right angle to the blade,
while cross-cut teeth are generally filed to a point at around a 65° angle to the
blade. Properly set and sharpened teeth not only saw much faster and cleaner,
but they are less likely to bind in the saw cut or wander from the line or
direction that one wishes to follow (this is also true of motorized band saw
blades). The set given to most back and tenon saws by the manufacturers is
often excessive and produces a wide, rough cut; reducing the set will often
produce cleaner, more accurate cuts.

Splitting

Sections of logs can be easily split with a hatchet, or more accurately with
a froe and mallet. This was the traditional way that violin makers (or their
wood dealers) produced perfectly quarter-sawn wedges of spruce and maple
for their instruments. There are other advantages to splitting wood: no wood
is lost as sawdust, it is much quicker, and it is less arduous than sawing.

Planing

Generally, the next operation after sawing is planing. Planes come in a

great variety of types and sizes. The blade of a standard bench plane is about
2 inches in width, and is mounted at around 45° to the sole. The blade is
beveled on the underside, which effectively reduces the cutting angle to about
20°. The blade generally does not extend the full width of the sole, but
protrudes from a rectangular opening in the sole called the mouth. Bench
planes come in a variety of lengths, the most common being the smoothing
plane (the shortest is generally around 9 inches in length), the jack plane
(generally around 14 inches in length), and the jointer, or try plane (the
longest having soles up to around 26 inches in length). Common metric sizes
are 245 mm (smooth plane), 355 mm (jack plane), and 560 mm (jointer
plane). The smooth plane is used for edge planing as well as for smoothing
the surface of boards. If it is used to level broad surfaces, the blade should not
be ground perfectly straight across its entire width, but relieved slightly at the
corners so that they do not extend beyond the sole and dig into the surface of
the workpiece. Scrub planes are about the same length as smooth planes but
have curved blades and wide mouths. These are used for rough removal of
wood when reducing the thickness of boards.
Bench planes of machined cast iron made by companies such as Stanley
and Record after the venerable Bailey patent (1867) generally have rather
thin blades that are stiffened by an extra plate of metal called a “chip
breaker.” This is held on by a screw, and its position should be adjusted so
that the edge of the chip breaker is about a 1/16 inch (2 mm) from the edge of
the blade. The chip breaker works in conjunction with the opening in the sole
to help break off the ribbon of wood that is shaved off by the blade. If chips
tend to get caught under the chip breaker, the bottom edge of the chip breaker
is not making firm contact with the top of the blade. To correct this, flat-file
(see Metalworking) the bottom edge of the chip breaker so that it fits snugly
against the blade. When adjusting metal bench planes, the frog (a wedge-
shaped device that supports the blade) should be positioned so that the mouth
(the opening in front of the blade) is not considerably greater than the
thickness of the shaving produced by the plane. When planing across the
grain, it is often helpful to use a toothed blade to prevent the grain from
tearing.
Wooden bench planes, similar to those used before the development of the
Bailey cast-iron planes, are still being manufactured, and many old ones are
still usable. Most have their blades held in place with a simple wood wedge
(as did the early ones), though some modern ones are fitted with a screw
device for advancing the blade as well as a lever for leveling the blade (very
much like the devices found in metal planes). In the author's experience, the
screw and leveling mechanisms found in these wooden planes do not work as
precisely as those found in the cast-iron planes. To adjust the position of the
blade in wood planes that have a simple wooden wedge, use a hammer or
mallet to tap the heel (back end) of the plane body (or the blade itself) to
extend the blade in order to deepen the cut, or the “strike button” at the top or
the front of the plane to retract the blade in order to make a lighter cut. One
advantage of the wooden plane over the cast-iron type is that the shape of the
sole can be easily refined, either by making it perfectly flat or by removing
material in front of or behind the blade to reduce friction (this is often done in
wooden Japanese planes). General flattening can be done with another plane,
while slight indentations in front of or behind the blade can be made with a
scraper (see section on Scrapers below). If the soles of cast-iron planes are
not true, they can sometimes be flattened with sharpening stones (an arduous
process), though regrinding in a machine shop may be necessary.
When planing the edge of a piece of wood, a shooting-board can be used to
maintain perpendicularity between the surface and edge of the board being
planed. The shooting-board consists of two equal lengths of wood that are
glued to one another so that one board overlaps the other. If the shooting-
board is clamped flat on the workbench, the plane is turned on its side with
the cutting edge oriented vertically. The lower board provides a platform for
one side of the bench plane, while the upper board supports the workpiece so
that the edge to be planed is brought into contact with the plane's blade. The
shooting-board can also be clamped vertically in a workbench vise with a
jack or jointer plane clamped against it in an inverted position with the blade
pointed up. When positioned this way, long boards can be pushed across the
inverted plane, with the upper section of the shooting-board serving as a
perpendicular fence, similar to that of a motorized jointer. Before planing the
edges of boards, a square should be used to check perpendicularity between
the sole of the plane and the surface of the shooting-board that guides the
workpiece. Precise resawing and planing to thickness are required when
making laminated inlays, such as those pictured in Figure 48.
Figure 48 Preliminary fitting of replacement inlays in a Broadwood square
piano, London, 1801. Approximately fifteen feet of inlay were missing from
this instrument along with brass bolt covers and other fittings (see Mold
making).

Block planes (around 6 to 7 inches in length, or 150 mm) often have blades
set a lower angle (between 12° and 20°) than bench planes, which makes
them better suited for planing end-grain. To prevent fragments of wood from
splitting off the edge when planing end-grain, the back edge of the workpiece
should be supported by or clamped against a block of wood.
Rabbet planes have blades that are generally mounted at 25° to the sole and
extend out to the sides of the plane body, which are accurately machined
perpendicular to the sole. These are designed for forming or truing rabbets
and dadoes (flat-bottomed grooves) and come in a variety of widths
corresponding to the widths of commonly used dadoes and grooves (½ in., ¾
in., 1 in.), in different lengths, and with different length noses. These sole
widths harken back to a tradition that persisted in woodworking beyond the
adoption of the metric system, and planes having these dimensions are still
manufactured by the English Stanley and Record firms. The shoulder rabbet
plane has a long nose that helps guide the plane when making grooves; the
bullnose plane (which has a very short nose) and chisel plane (which has no
nose) are principally used for forming stopped dadoes and planing into
corners. The mouths of shoulder and bullnose rabbet planes can generally be
widened or narrowed by loosening the front of the plane and adjusting a set
screw.
Compass planes have an adjustable, curved sole for planing concave or
convex surfaces.
Molding planes, traditionally made of wood with shaped soles and
matching blades, were once available in a staggering variety of
configurations for cutting tongues, grooves, beads, ovolos, ogees, fluting,
reeding, fillets, quarter and half rounds, and window sashes. An eighteenth-
or nineteenth-century carpenter or cabinetmaker might have dozens of these
wood molding planes in his tool box or workshop. In the late nineteenth
century, the Stanley firm developed the combination plane, which was sold
with sets of cutters (45 and finally 55 in number in its last, most complex
version). It dispensed with the shaped sole and instead employed two parallel
runners (one of which could be raised or lowered by a screw device) that
served as guides for the cutters and prevented them from digging too deeply
into the wood. These planes were fitted with adjustable fences, depth stops,
and cutting spurs, making them highly versatile and ultimately less expensive
than dozens of molding planes. Though a bit troublesome to set up and often
frustrating to use (the bladelike runners are sometimes not as effective as a
fully shaped sole), these planes are especially useful in restoration work, as
cutters can be ground to any shape and used with this type of plane. The
Clifton firm has fortunately reintroduced the long-discontinued Stanley 55
plane.
When grinding and sharpening the blades of planes (see Sharpening tools),
consider the type of wood to be planed. If the plane will be used primarily
with softwoods, the primary bevel can be ground at around 25°; if hardwoods
or woods of a tough or abrasive nature are being planed, 30° to 35° may be
preferable. In any case, the secondary bevel should be kept very narrow.
Repetitive sharpening and honing will gradually widen the secondary bevel;
when it reaches several millimeters in width, the primary bevel should be
reground.
One of the myths regarding hand planes is that they generate a perfectly
straight edge or flat surface. This is not the case because the blade projects a
bit below the flat sole, thereby generating a slight curve in the workpiece.
(The motorized jointer is not set up this way, as the “sole” consists of two
separate plates, one in front of the rotating cutter and the other behind it. The
plate in front of the cutter can be raised and lowered to adjust the depth of
cut, while the plate behind the cutter is maintained level with the very top of
the rotating blades. Because the sole is on two levels, the planed edge formed
by the cutter will be perfectly straight.) Despite this fault of the hand plane,
there is actually an advantage to using it when preparing the edges of long
boards that will be glued together: the slight convexity of the planed edges
brings the ends of the boards together and assures a tighter joint.

Violinmaker's planes

Violinmaker's planes, or finger planes, are very small planes having either
flat or curved soles, the latter being used in shaping the arching of tops and
backs after preliminary gouging. They are generally between 1 inch and 2
inches (25 mm to 50 mm) in length. So-called palm planes are a bit larger
(about 3½ inches, or 90 mm) and are useful for general planing small work.

Japanese planes

Generally made of wood, these are intended to be pulled towards the

worker rather than pushed away, though they can also be pushed in the same
way that Western planes are used. One feature of Japanese planes that makes
good sense is the location of the blade, which is toward the rear of the sole
rather than in front. This allows the plane to be more securely positioned and
steadied against the surface being planed. The soles of Japanese planes are
often slightly relieved between the front end of the sole and the mouth and
behind the back of the mouth and the back end of the sole. This is done with
a special scraper and serves not only to reduce sliding friction but also to
prevent tear-out.

Spokeshave

The spokeshave is a variant of the plane having a very short sole and
handles on either side to help draw it across the workpiece. Spokeshaves are
made with flat or convex soles, as well as half-round and radial shapes.
Spokeshaves are excellent for shaping and smoothing the curved surfaces of
such objects as cabriole legs, as well as harpsichord, clavichord, and
fortepiano bridges.

Drawknives

These handle somewhat like spokeshaves but lack a sole to limit the cut.
They are useful for the rough forming of curved surfaces, such as cabriole
legs.

Chisels

These simple, though essential tools come in a variety of widths, lengths,

and thicknesses. The firmer chisel is the strongest, with its blade formed of
rectangular stock, and is often driven with a mallet. Mortise chisels are
generally narrow in width but have long, heavy blades that give them great
strength and allow them to be hammered deeply into wood (for example,
when making a mortise for a lock). The bevel-edged chisel is not as heavily
built as the firmer and mortise chisels, and it is generally not used with a
mallet. Its beveled upper face allows it to be used to clean up dovetails,
though it is also an excellent general-purpose chisel. The butt chisel is a
shorter version, and the paring chisel is a longer version. The paring chisel is
available in a cranked version having an offset handle that allows the entire
back of the blade to slide along the workpiece, which is very useful when
trimming waste from soundboard shims. The skew chisel has its cutting edge
angled relative to the length of the blade. For fine work, especially with
softwood, the primary bevel of chisels should be ground at about a 20° angle;
when used to cut hardwood, especially if a mallet is used, an angle of 25–30°
is appropriate. When grinding, sharpening, and honing, the back of the chisel
should be kept perfectly flat (see Sharpening tools).

Gouges

These come in different widths and radii of curvature (or “sweeps”), as

well as with straight, curved, and back-bent shapes. The sweeps are graded
by number, in the so-called “London pattern”; numbers 1 and 2 are straight
and skewed chisels (though they differ from standard woodworking chisels in
that the cutting edge is beveled on both sides), with numbers 3 through 11
moving progressively from shallow to deeply curved. Strangely, each number
does not refer to a particular radius, but rather to a sequence of curves of
different widths (generally from about 1/16 to 1½ inches, or 2mm to 30 mm)
that nest within each other. The “Continental” numbering systems employed
by German and Swiss manufacturers differ in slight detail (for example, the
German “Two Cherries” brand uses the number 1 to indicate both straight
and straight skewed gouges). In addition to curved gouges, other useful
shapes include veiners and V-shaped parting tools. Most carving gouges have
their primary bevel ground on the underside of the blade and are chiefly used
for scooping out material (such as the undersides of violin top and back
plates) and generating convex curves (such as the volutes of violin scrolls);
these are called in-cannel gouges. Out-cannel gouges are beveled on the
inside of the curve and are primarily used to generate or clean up concave
curves. Guidance for regrinding, sharpening, and honing gouges is provided
in the section on Sharpening tools.

Turning

Turning may be done freehand on a woodturning lathe, though a

metalworking or machine lathe can also be used for turning wood and ivory.
(See Metalworking, especially the section on turning threads, for further
information.) There are two basic types of cutting tools: scraping tools
(which are generally supported on a tool rest that is adjusted to align the
cutting edge with the workpiece's center of rotation) and turning gouges
(which are held at a higher angle and slice off material). Shaped tools can be
used to form rings, beads, and other decorative devices. (Figures 8 and 49).
Figure 49 Turned bead-and-reel molding of rosewood for a Broadwood
pianoforte. The short piece is an original molding. Several running feet were
missing from the instrument.

Scrapers

Scrapers are flat sheets of steel whose edges are ground and sharpened at
either a 90° or 45° angle, and then have a burr raised that serves as the cutter.
For fine instrument work, the 45° bevel, or even a more acute angle, works
better because the burr is sharper. Many books advise simply filing the edge
prior to raising the burr, but a finer cutting edge can be achieved if the bevel
is properly ground and sharpened prior to raising the burr. The burr is raised
by holding a hardened steel burnisher at several degrees' inclination to the
bevel and giving a firm stroke; several more strokes are then given,
increasing the burnisher's angle a few degrees with each stroke. When drawn
across the surface of wood, the scraper removes extremely thin shavings and
produces a surface that is superior to sanding. For cabinet work and finishing
large flat areas such as tabletops, soundboards, lids, etc., the scraper plane
produces a glistening surface that is without equal. For work on violins and
other instruments having delicately curved surfaces, very thin, flexible curved
scrapers are used. These also function at their best if their cutting edges are
ground at an angle prior to sharpening and raising the burr. Sets of seven
scrapers, which generally include two straight, three curved, and two
elliptical scrapers, are sold by violin makers' suppliers; however, these can be
reground to different shapes and sizes. In violin making, the scraper is used
not only for refining the surface finish but also in the subtle contouring of the
arching and edgework and in fluting the back of the pegbox. These thin
scrapers can be flexed between the fingertips to alter the radius of the cut.
The scraper can also be used to replicate fine hardwood moldings. The shape
of the molding is first carefully traced onto the corner of a scraper blade,
which is then ground or filed to shape with jeweler's files. After turning the
burr, the blade is then attached to a fence (such as a block of hardwood) that
guides the scraper along the edge of the board or strip of wood that is being
used to make the molding. Even rather deep molding profiles can be formed
in this way. This is an excellent method of fabricating missing molded key
fronts for English square and grand pianos, for example. It also works
splendidly when making moldings in very hard materials such as ebony and
ivory (Figure 50). Scrapers with toothed cutting edges can be used to roughen
wood surfaces prior to gluing.

Figure 50 Scraped moldings. The top molding, of ebony, was made for a
sixteenth-century organ case. Below it is a maple molding to be cut into key
fronts for an English square piano.
Knives

Though sets of chip-carving knives are commonly used by wood sculptors,

toy, and model makers, the knife is not much used in most types of
instrument making with the exception of violin making. Violinmakers' knives
are generally obtained in sets having blades of various widths. They consist
of a long, ovoid handle, generally of wood, with a thin blade running through
it. The blade projects about 1–2 inches (25–50 mm) from the handle, and it is
either ground straight, at about a 30° angle or less relative to the back of the
blade and coming to a sharp point, or it may be curved up to the tip. Violin
makers use their knives to refine the shape of the belly and back plates after
they have been sawn out, while narrow-bladed knives ground to an acute
angle are used to shape the f-holes and adjust the openings in bridges. Violin
makers often make their own knives from blade blanks (which are generally
tapered slightly to facilitate the grinding of bevels) and fabricate their own
handles from scraps of figured maple. Most violinmaker's knives are beveled
on both sides, but for some work, it is advantageous to have blades ground on
the left or right side (the opposing side being left flat). Violinmaker's knives
are generally kept razor sharp.

Rasps, files, and floats

Rasps have sharp, pointed teeth that grind away at the wood, while files
have parallel rows, or two sets of crisscrossing rows of sharp protrusions that
shear off the wood. Floats employ a series of parallel rows, sometimes
curved, that act like a series of plane blades. Whereas rasps are specifically
designed for shaping wood, files can be used on metal as well as wood. Rasps
come in various shapes: flat, half-round, and round, and in various cuts (fine,
medium, and coarse). Files come in a great number of shapes and cuts (see
Metalworking). Floats are used by makers of wooden planes for truing and
adjusting the dimensions of throats and the seatings for blades.

Drilling holes

First a distinction in terms: drills are the tools that turn bits (a twist drill bit
should not be referred to simply as a “drill”). Ordinary twist drill bits have
their points ground at a 59° angle and are designed for drilling metal, but the
small sizes (1/2 inch or under) can be used to drill wood, especially if they
are driven by a high-speed electric hand drill or a drill press. An egg-beater-
type hand drill, a push-drill employing an Archimedean spiral, and even a
hand-turned pin vise can be used to drill holes with smaller-size twist drill
bits. For larger holes, brad-point, multi-spur, and Forstner bits are preferred
because they have a cutting spur or rim that scores the perimeter of the hole,
thereby reducing chipping and tear-out and producing a cleaner hole. The
old-fashioned bit-and-brace has a special chuck for holding solid-center auger
or Jennings-type double-spiral twist bits. These bits have a central point that
has spiral grooves that dig in under pressure from the brace and help pull the
bit through the wood. The Jennings-pattern bits run in size from 1/4 inch to 1
inch, and are numbered from 4 to 16 (the numbers refer to the diameter in
sixteenths of an inch). A similar type, called “solid-center auger bits” are
available in sizes ranging from 1/4 inch (6.35 mm) through 1½ inches (38
mm). Spade bits run up to about 1½ inches in diameter, while expansion bits
can be adjusted to produce holes up to 3 inches in diameter.
To drill a hole for a wood screw, specially designed, tapered wood-screw
bits, with adjustable stop collars and countersinks, are available in sets
covering the most commonly used sizes of wood screws. If these are not on
hand, one can use ordinary twist drill bits, but at least two sizes of bits are
required plus a countersink if a flat-headed screw is used. Wood screws
generally have a wide, unthreaded section just below the head (termed the
shank), and a tapered threaded section extending down to the pointed tip.
When connecting two pieces of wood with a wood screw, the top piece is
drilled out to provide clearance for the wider, unthreaded shank so that it can
be pushed through, while the lower piece is drilled undersize so that the
threaded section will be able to cut into the wood, thereby drawing the two
pieces of wood together as the screw is tightened. If the upper board is drilled
to the same diameter as the lower piece, the screw may bind at the top end
and fail to pull the two pieces of wood together effectively. To drill a hole for
the threaded part of the screw (which will also serve as a pilot for the larger-
diameter bit used for the shank), the following table can be used as a guide
(recommended bits are from the numbered and lettered series):

Wood screw Shank Hardwood Softwood

size bit thread thread

0 52 58 69

1 47 55 65

2 42 53 59

3 37 51 56

4 32 48 55

5 30 44 53

6 27 42 52

7 22 38 50

8 18 35 48

9 14 32 46

10 10 30 44

11 4 29 43

12 2 27 39

14 D 20 34

16 I 16 31

18 N 10 29

20 P 3 26

24 V D 18
 When drilling, start with the bit for the threaded section of the screw. If it
is necessary to drill to a certain depth (for example, when drilling holes for
piano lid-hinge screws where one does not want to drill clear through the
lid!), mark the bit with a piece of masking tape to serve as a visual indicator
of when to stop drilling. When using a drill press, the depth stop can be used
to limit the travel of the bit. After this hole is drilled, select a bit to provide
clearance for the shank, and mark it with a piece of masking tape (or readjust
the drill press's depth stop) so that the drilled hole will not extend beyond the
shank when the screw is inserted. Finally, if a countersink is required, drill
that as the last step. Keep in mind that it very easy to make a countersink too
deep, so it is best to employ the “cut and try” method. When countersinking
wood with a hand-held drill, single-flute countersinks work well; if a drill
press is used, multi-fluted countersinks can be used.
Reamers are used to refine the surface of holes and to produce accurately
dimensioned holes, such as those intended for tuning pins and pegs. Special
tapered reamers and matching peg shapers are available for violin pegs
having a taper rate of 1/30, for cello pegs having a taper rate of 1/25, and
cello endpins having a taper rate of 1/17 (see Tapered reamers). Straight
reamers are available in fractional and numbered twist drill sizes as well as in
a special numbering system that is used when drilling wrestplanks for modern
piano tuning pin sizes. Adjustable reamers are also available. Metalworking
reamers that have multiple straight flutes tend to chatter in wood; however,
spiral reamers work beautifully in wood, and are recommended when fitting
violin, viola, and cello pegs.

Power tools

Though most conservation work can be accomplished with hand tools, the
following power tools are also recommended: the band saw (useful for
sawing curves and ripping lengths of wood, as well as for resawing planks of
moderate dimensions), the scroll saw (for finer work, useful for cutting small
irregularly shaped inlays), the drill press (for accurate drilling of holes),
stationary disc and belt sanders (for finishing and shaping), as well as wood,
machine, and jewelers' (or model makers') lathes.

Clamps
 Probably the most useful accoutrements of the musical instrument
conservator (after the glue pot and a pile of rags) are clamps. For work on
large keyboard instruments, such as harpsichords and fortepianos, bar and
pipe clamps of 8-feet (2.4 meters) length and greater are often necessary for
casework. Deep-engagement bar clamps are also useful. Iron C-clamps (with
full-length screws for tightening) or fast-acting deep-engagement clamps with
shorter tightening screws should be stocked by the dozen in lengths of 1 foot,
2 feet, and 3 feet, though if purchased in metric lengths, the standard sizes are
300 mm, 400 mm, 600 mm, 800 mm, and 1000 mm. Cam clamps with cork
facing are very handy, and it is generally useful to have several dozen of
these in the 1-foot and 2-feet lengths (300 mm and 600 mm). Jorgenson
hand-screw clamps come in a variety of sizes and are very useful because
their jaws can be set at an angle, with full clamping force applied over their
large wood faces. One disadvantage of Jorgenson clamps is that the
manufacturers soak the wood jaws in crude oil, and even after many years,
this oil is readily absorbed by wood parts that are being clamped. Thus, it is
necessary to place a nonabsorbent barrier (such as aluminum foil, wax paper,
mylar sheets, etc.) between the clamp jaws and the work.
Violin makers' suppliers sell a vast array of specialized clamps for gluing
together and repairing violins, violas, and cellos. These include spool,
garland, and specially designed assembly clamps for gluing the belly and
back plates to their rib structures, and crack-repair clamps that can span
across the upper, lower, and center bouts to draw cracks together (see Figures
39 and 40). For working on other sizes of stringed instruments, such as treble,
tenor, and bass viols, it is often necessary to fabricate one's own clamps; for
example, the wooden parts of spool clamps can be turned on a lathe, and
carriage bolts and wing nuts can be used for the screw-tightening mechanism.
Also available are myriad sizes of lightweight C-clamps for gluing cleats,
special clamps for drawing together cracks at the edges of belly and back
plates, rib-corner clamps, clamps for gluing in saddles, and clamps for
applying pressure under the fingerboard when regluing bellies. Guitar
makers' suppliers offer special deep-engagement clamps for gluing loose bars
through the sound hole (see Figure 11a). The longer clamps have leveling
screws that prevent the weight of the clamps from distorting the soundboard.
When gluing ribs and bridges to large keyboard instrument soundboards, as
well as those of lutes and guitars, the best clamping system consists of go-
bars and a go-bar deck. This involves a flat, rigid surface supported several
feet above a bench top (if a workbench is about 3 feet (1 meter) high and the
workshop ceiling is no higher than about 8 feet (2.5–3 meters), one may be
able to use the ceiling to brace the go-bars). The go-bars are strips of supple
wood (such as spruce or pine) that are flexed between the overhead support
and the workpiece. In addition to assisting in gluing on ribs and bridges, the
go-bar deck is helpful when gluing soundboard cracks, as downward pressure
can be maintained along their entire length in order to hold them in alignment
while side-to-side pressure is exerted by clamps.

Workbenches

The standard woodworkers' bench (such as the Ulmia brand) is generally

equipped with a wood-jawed shoulder vise on the left end of the bench and a
tail vise on the right, the latter equipped with mortises and bench dogs for
clamping work to the table. In addition to a woodworkers' bench, one should
have a large table with a well-braced, perfectly flat top, and a go-bar deck
overhead. This can be used for assembly and repair of large instruments, such
as harpsichords, harps, etc. Adequate light should be available; most violin
makers set up narrow benches against windows facing north to facilitate
accurate retouching (see Retouching).

Glue

Hot animal-hide glue is generally used in the making, repair, and

restoration of musical instruments, from the largest pipe organs down to
violins. Most of the new synthetic glues (see Glues, pastes, and other
adhesives) should never be used where hot animal-hide glue was traditionally
employed because they often provide no greater strength and generally pose
difficulties if the joint must be reopened at a later date. Instances where
synthetic glues may be appropriate are in the repair of woodwind
instruments, where condensed moisture might weaken animal-hide glue, and
where gaps must be filled, such as cracks running between tuning pins (see
Stringed-keyboard restoration) or in the tables of organ windchests (see
Organ restoration).
If the glue is in granular form (see Glues, pastes, and other adhesives), it
must be soaked in cold water until all of the grains absorb water and swell up
(this can take as little as an hour, but an overnight soaking is best). If the glue
is purchased in sheets or cakes, these should be wrapped in cloth and broken
up with a mallet or hammer. About a finger's depth of water should be added
to an ounce or two of swollen glue, and the jar containing the glue should be
placed in a water bath that is heated to about 140°F (60°C) until the glue
liquifies. An electric glue pot can be used, or a hotplate with thermostatic
control. To determine if the glue is the right consistency for most gluing
operations, dip a brush into the glue and raise it about a foot above the glue
pot – the glue should form a continuous stream. Glue brushes should not have
metal ferrules, which might rust and discolor the glue. For some gluing
operations (such as gluing the bellies of violins to the ribs), the glue can be
diluted somewhat. Animal-hide glue will gradually thicken due to
evaporation of water, so it is helpful to cover the glue pot with aluminum foil,
breaking a hole through it of sufficient size to admit the glue brush.
Overheating glue (for example, permitting it to boil) will reduce its strength.
When gluing large surfaces, especially in cold environments, it is important
to keep the wood surfaces and glue warm until the clamps are applied. If the
glue “chills,” it becomes viscous, preventing the parts from making good
contact. In the old days, the work was warmed next to an open fire or stove;
today, heat lamps can be used. Commercial hide and fish glues that are liquid
at room temperature are not recommended for use in instrument making or
restoration because they are considerably weaker than hot animal-hide glues
and are susceptible to reverting to a jellylike state under high humidity.
When gluing end-grain, first size it with dilute hot animal-hide glue and
allow it to harden. If this raises the grain, use a sanding block or flat-file to
smooth the surface. This process can be repeated if the wood is especially
porous. The sized end-grain can then be glued more securely because it
absorbs less of the full-strength glue.
To glue center joints or to glue long boards along their edges without
clamps, one can use the “rubbed-joint” technique. Begin by sizing both edges
to be joined with dilute animal-hide glue and allowing it to dry. Check the fit
and correct if necessary, and then apply full-strength glue. Bring the edges
together, sliding them back and forth until the glue begins to gel. Check the
alignment before setting the work aside to dry.
When gluing old cracks, they first must be scrupulously cleaned, otherwise
the glued crack will be visible. Take care not to damage the edge of the crack
by breaking off fibers of wood or flakes of varnish or paint. Warm water can
be used to loosen old glue and dirt that have collected in the crevice;
however, if polish, wax, or other solvent-borne materials have become
entrapped, the appropriate solvents will have to be used to remove these
materials before gluing. If cracks have been open for a long time, the exposed
surfaces may have darkened through oxidation. If this is the case, it may be
necessary to bleach the crack with hydrogen peroxide, oxalic acid, or other
bleach (see below and Bleach).
Just prior to gluing, it is important to pre-arrange clamps and leveling
devices so that they can be quickly assembled as soon as the glue has been
applied and before it has had a chance to chill. Small cleats may be
temporarily glued to the underside (alternating on either side of the crack) to
facilitate leveling. Pieces of clear acrylic plastic can also be interposed
between the clamps and the workpiece to help level cracks; animal-hide glue
does not adhere to this plastic and one can observe the crack during clamping
and gluing operations. The surfaces of acrylic pieces should be relieved along
the edges with a file or sandpaper so that they will not mark the wood.

Veneering

Veneering refers to gluing a thin layer of wood (usually a fine, cabinet-

grade wood such as mahogany or walnut) to the surface of more massive
wood (usually a cheaper or more common wood, such as oak or pine).
Traditionally, hot animal-hide glue was used in veneering, but modern
cabinetry is often veneered using synthetic glues and contact cements, which
create an immediate bond and do not require the use of clamps to secure the
veneer to the underlying wood. When veneering using animal-hide glue, it is
generally necessary to provide pressure over the entire surface to achieve
good adhesion. This requires the use of numerous clamps, cauls, or veneer
presses, especially when large surfaces (such as keyboard instrument lids) are
being veneered. Another traditional method of gluing down veneer employed
a special veneer “hammer” (shaped somewhat like an adze, but with a
smooth, rounded over “blade”), which was drawn back and forth over the
veneer to press it down until the glue took hold. Veneer hammers are once
again available, though wooden rollers are often used today to accomplish the
same ends, especially when quick-setting glues, contact cements, or veneers
with adhesive backings are used. A vacuum table can also be used in certain
veneering and laminating operations.
Marquetry and parquetry are types of decorative veneering. Marquetry
involves the inlaying of various shapes (such as scroll, floral, and folial
patterns, as well as pictorial scenes) that have been cut out with fine jeweler's
saws or knives, while parquetry refers to the arrangement of purely geometric
shapes (such as triangles, squares, diamonds, and bands) that have been cut
out by sawing or by use of a knife and straightedge; woven, lattice, and trellis
designs are examples of parquetry. Special paper veneer tape is sometimes
used to hold sections of veneer together prior to gluing them down.
Veneered panels often employ matched veneers to cover large areas or for
decorative effect. Book matching, sliding and slip matching, sunburst, four-
piece, and diamond matching are some of the techniques used to join veneers.
Modern veneer manufacturing generally involves slicing the veneer rather
than sawing it. While this technique may produce the same types of figure as
traditional sawed veneer, some veneers are produced by peeling the wood
from rotating logs, which produces a figuration (such as seen on plywood
used in home construction) that one does not encounter in early instruments.
When selecting veneer for repairs, the restorer must match the grain, its
orientation, and its figure with that of the workpiece.
Missing inlays can be fabricated by taking an impression of the area where
the missing piece had been by placing a piece of tracing paper over it and
rubbing the perimeter with a fingertip covered with pencil lead or powdered
charcoal. This impression can then be glued to a piece of veneer and used to
guide the saw or knife in cutting out the replacement. If the piece is sawn out,
use a jeweler's saw blade with sufficiently fine teeth that at least two of them
come in contact with the edge of the veneer. For typical modern US veneers
(which are generally in. or about 0.9 mm in thickness), a 4/0 saw blade
(66 teeth per inch) is fine enough, while European veneers (which are
generally thinner, around in. or about 0.6 mm thick) require a 7/0 saw
blade (84 teeth per inch). Traditional veneering used hand-sawn veneers that
were considerably thicker, and to replace missing pieces, it is necessary to
saw veneer of the proper thickness; it may also be possible to achieve the
proper thickness by stacking thin veneers. A motorized band saw is useful for
resawing lumber to the required thickness, and a motorized scroll saw fitted
with a jeweler's blade can then be used to saw out small pieces of inlay from
the resulting veneer. Scroll saws that operate by vibrating the blade rather
than by making long up-and-down strokes produce better results, especially
with very thin veneer. When sawing out small pieces and complex shapes, it
is helpful to support the veneer with a thicker piece of wood. In early times, a
stationary device upon which the worker sat, termed a “donkey,” was used to
hold the veneer while pieces were sawn from it using a frame saw; the veneer
was held in place by a wooden vise that was operated by a foot pedal – this
freed up the worker's hands and enabled him or her to quickly reorient the
veneer in relation to the saw. Long inlay strips and decorative edging can be
cut with a knife and straightedge or with special inlay and strip cutters.
Inlaid pieces (such as those depicting the petals of flowers) can be shaded
by dipping their edges into hot sand. Veneers were often dyed, especially
when used to make figurative or scenic depictions. Traditionally, various
colors were achieved with decoctions of botanicals such as alkanet wood,
brazil wood, buckthorn berries, indigo, logwood, and weld, or with chemicals
such as verdigris and copperas (the latter was used in combination with
logwood extract to make the black strips found in some violin purfling).
Today, synthetic dyes are often used. It should be pointed out that many of
the traditional botanical and modern synthetic dyes used to color veneers are
fugitive, and the colors we observe in old woodwork generally have faded
considerably. For example, the walnut veneer used on many Viennese
fortepianos was often stained a deep reddish color to make it look like
walnut; the folded-back sections of lids and keywells may retain much of this
color, whereas areas of the case that were constantly exposed to light, such as
the bentside and cheek, are generally faded.
If veneer is detached along an edge or if it can be lifted up without damage,
the old glue should be washed off with warm water before reattaching the
veneer with fresh glue. If the veneer has blistered or separated from the
underlying structure, the old glue can sometimes be reinvigorated by rubbing
over the loosened veneer with an electrically heated tacking iron or clothing
iron set to a low temperature (animal-hide glues soften at around 140° F or
about 60° C), though keep in mind that many finishes, such as French polish,
will melt at this temperature. If large areas of veneer need to be reglued, it
may be necessary to remove the existing varnish or polish to make an
effective repair. If the separation is not at an exposed edge, a fine cut can be
made along the grain with a razor blade in order to admit glue, or a syringe
with a fine needle can be used to inject glue between the veneer and the
underlying wood. A roller or veneer hammer can be used to press the veneer
down until the glue sets (overnight or twenty-four hours is generally
sufficient for complete drying).
If corners or edges of veneers are damaged or missing, do not clean up or
cut away the damaged area with a straight cut in order to fit a replacement
piece, as such a repair will be noticeable. Instead, make an irregularly shaped
cut or cut along the figuration, and fit the replacement piece using the method
described above. Holes and burnt areas should also be filled with irregularly
shaped pieces. Another technique is to feather-in the repair by beveling the
perimeter of the damaged area and chalk-fitting the replacement piece to fit;
after leveling the fill with a scraper or sanding block, an almost invisible
repair will result.
Masking tape is often sufficient in holding down decorative edgework,
banded inlays, and stringing during gluing, obviating the need for complex
clamping arrangements. When using individual clamps on small areas or
when making repairs, always place softwood blocks (whose edges have been
relieved with a file or plane to prevent the blocks from making indentations)
or pieces of cardboard between the workpiece and the clamp faces to prevent
them from scratching, denting, or otherwise damaging the surface. Wax paper
or silicone release paper can be placed between the workpiece and the glue
blocks or cardboard to prevent them from becoming glued to the surface of
the veneer. Before final clamping, remove excess glue with a damp cloth.
After the glue dries, wipe down the veneer with a damp cloth to remove any
glue that may have dried on the surface, as glue drips and areas that have
been impregnated with glue will show up in later finishing stages, such as
staining and varnishing.

Stain removal

Bleach for stained wood: for mold and mildew stains, 3% hydrogen
peroxide can be used. Because hydrogen peroxide is generally stabilized by
the addition of acid, a small amount of alkali (such as household-strength
ammonia) must be added to release the oxygen. The ammonia should be
added drop by drop until a pH of about 9 is achieved (this can be monitored
with an electronic pH meter).
For iron-gall ink and rust stains, a 5% solution of oxalic acid or 10% citric
acid can be brushed over the stain; neutralize with sodium carbonate solution
(check with pH meter); and rinse with distilled or deionized water.
For oil, grease, and tar stains, apply mineral spirits, xylene, or pyridine.
Fuller's earth or French chalk may be effective in absorbing fresh oil and
grease stains. These powders can be applied over the stain, left overnight, and
then brushed or vacuumed off.
For wax and candle grease, apply mineral spirits, xylene, or
trichloroethane.
For fly stains, apply 3–6% hydrogen peroxide mixed with ethanol 1:1.
For tea and coffee stains, apply 2% potassium perborate with a brush, rinse
with distilled or deionized water, and expose to mild UV or sunlight for an
hour or two.
For ink stains (other than India ink), apply 2% Chloramine-T (followed by
antichlor; see Bleach), 5% oxalic acid, or 10% citric acid, rinse with distilled
or deionized water.
Embedded or ground-in dirt in both finished and unfinished wood can
sometimes be removed by rubbing with plastic erasers (avoid those
impregnated with abrasives, such as those used to remove ink inscriptions) or
with a draftsman's cleaning pad (this is a cloth sack filled with pulverized
rubber). The draftsman's cleaning pad is twisted or tapped to release the
pulverized rubber, and the pad is then rolled over these bits of rubber to clean
the soiled surface. After cleaning, remove all traces of the pulverized rubber,
as the sulfur it contains may have a corrosive effect upon certain metals.

References
Denis Diderot and Jean le Rond D'Alembert, eds., Encyclopédie ou
dictionnaire raisonné des sciences, des arts et des métiers (Paris, 1751–
1772).
Paul N. Hasluck, The Manual of Traditional Wood Carving (London, 1911).
Albert Jackson, David Day, and Simon Jennings, The Complete Manual of
 Woodworking (New York, 1994).
W. A. Lincoln, The Complete Manual of Wood Veneering (New York, 1984).
David Middleton and Alan Townsend, Marquetry Techniques (London,
1993).
N. Nosban and W. Maigne, Nouveau Manuel complet de l'ébéniste (Paris,
1887).
André-Jacques Roubo, L'Art du menuisier (Paris, 1769–1774).
 Select bibliography
Acoustics
Backus, John. The Acoustical Foundations of Music (New York and London,
1977).
Bacon, Francis. Sylva Sylvarum: or A Natural History in Ten Centuries
(London, 1627).
Benade, Arthur H. Fundamentals of Musical Acoustics (New York, London,
and Toronto, 1976).
Beranek, Leo L. Acoustics (New York, Toronto, and London, 1954).
Cremer, Lothar. The Physics of the Violin (Cambridge, MA, and London,
1981).
Dipper, Andrew. ‘Librem Segreti de Buttegha’: A Book of Workshop Secrets:
The Violin and Its Fabrication, in Italy, Circa 1725–1790 (Minneapolis,
2013).
Fletcher, Neville H. and Thomas D. Rossing. The Physics of Musical
Instruments (New York, 2010).
Forsyth, Michael. Buildings for Music: The Architect, the Musician, and the
Listener from the Seventeenth Century to the Present Day (Cambridge,
MA, n.d.).
Hall, Donald E. Musical Acoustics: An Introduction (Belmont, CA; 1980).
Helmholtz, Hermann L. F. On the Sensations of Tone as a Physiological
Basis for the Theory of Music (New York, 1954).
Hunt, Frederick Vinton. Origins in Acoustics: the Science of Sound from
Antiquity to the Age of Newton (New Haven and London, 1978).
Hutchins, Carleen M., ed. Musical Acoustics, Part 1: Violin Family
Components (Stroudsburg, 1975).
Hutchins, Carleen M., Musical Acoustics, Part 2: Violin Family Functions
(Stroudsburg, PA; 1976).
Hutchins, Carleen M. and Virginia Benade, eds. Research Papers in Violin
Acoustics 1975–1993, 2 vols. (Woodbury, NY; 1997).
Hutton, Charles. Recreations in Mathematics and Natural Philosophy
(London, 1803/1840).
Mahillon, Victor-Charles. Elements d'Acoustique Musicale et Instrumentale
(Brussels, 1874; reprinted Brussels 1984).
Maupertuis, Pierre Louis. “Sur la forme des instruments de musique,”
Histoire de l'Academié de Mathématique et de Physique (Paris, 1726), pp.
215–226.
Mersenne, Marin. Harmonie universelle (Paris, 1636).
Miller, Regis. et al. Wood Handbook: Wood as an Engineering Material
(Madison, WI; 1999).
Morse, Philip M. Vibration and Sound (New York and London, 1936).
Nederveen, Cornelis J. Acoustical Aspects of Woodwind Instruments
(Amsterdam, 1969).
North, Francis. Philosophical Essay of Music Directed to a Friend (London,
1677).
Peluzzi, Euro. Tecnica Costruttive degli Antiche Liutai Italiani (Florence,
1978).
Rayleigh, Lord. The Theory of Sound, 2 vols. (London: vol. I, 1877; vol. II,
1878).
Regazzi, Roberto. Il Manoscritto Liutario di G. A. Marchi (Bologna, 1986).
Saunders, Frederick A. “The Mechanical Action of Violins,” Journal of the
Acoustical Society of America 9 (1937), pp. 81–98.
Saunders, Frederick A. “Recent Work on Violins,” Journal of the Acoustical
Society of America 25 (1953), pp. 491–498.
J. C. Shelleng, “The Violin as Circuit,” Journal of the Acoustical Society of
 America 35 (1963), pp. 326–338.
J. C. Shelleng, “Acoustical Effects of Violin Varnish,” Journal of the
Acoustical Society of America 44 (1968), pp. 1175–1183.
J. C. Shelleng, “On Vibrational Patterns in Fiddle Plates,” Catgut Acoustical
Society Newsletter 9 (1968), pp. 4–10.
Sibire, Abbé. La Chélonomie ou le parfait luthier (Paris, 1806; Brussels,
1823).
Tartini, Giuseppe. Trattato di musica (Padua, 1754).
Vitruvius, The Ten Books on Architecture (New York, 1960).

Analysis of materials, authenticity, and dating

Alexander, John Henry. Universal Dictionary of Weights and Measures,
Ancient and Modern: Reduced to the Standards of the United States of
America (Baltimore, 1850).
Baillie, M. G. L. Tree-Ring Dating and Archaeology (Chicago, 1982).
Baillie, M. G. L. A Slice Through Time: Dendrochronology and Precision
Dating (London, 1995).
Bion, Nicolas. The Construction and Principal Uses of Mathematical
Instruments; Translated from the French of M. Bion, Chief Instrument
Maker to the French King by Edmund Stone (London, 1758).
Brazier, J. D. and G. L. Franklin. Identification of Hardwoods: A Microscope
Key (London, 1961).
Brommelle, N. S. and Garry Thomson, eds. Science and Technology in the
Service of Conservation (London, 1982).
Cavalli, E. Tables de comparaison des mesures, poids et monnaies anciens et
modernes (Marseilles, 1869).
Clarke, F. W. Weights, Measures, and Money of All Nations (New York,
1876).
Constantine, Albert, Jr. Know Your Woods (New York, 1959).
Core, H. A., W. A. Côté, and A. C. Day, Wood Structure and Identification
(Syracuse, 1979).
Derrick, Michele R., Dusan Stulik, and James M. Landry. Infrared
Spectroscopy in Conservation Science (Los Angeles, 1999).
Goodway, Martha and Jay Scott Odell. The Metallurgy of 17th- and 18th-
Century Music Wire, The Historical Harpsichord 2 (Stuyvesant, NY;
1987).
Hoadley, R. Bruce. Understanding Wood (Newtown, CT; 1980).
Hoadley, R. Bruce. Identifying Wood (Newtown, CT; 1990).
Hough, Romeyn Beck. The Wood Book (Cologne, 2002).
Hutton, Charles. Recreations in Mathematics and Natural Philosophy
(London, 1840).
Jaffé, H. L. C., J. Storm van Leeuwen, and L. H. van der Tweel, eds.,
Authentication in the Visual Arts: A Multi-disciplinary Symposium;
Amsterdam, 12th March 1977 (Amsterdam, 1979).
Jones, Mark, ed. Fake? The Art of Deception (Berkeley and Los Angeles,
1990).
Jones, Mark, Why Fakes Matter: Essays on Problems of Authenticity
(London, 1992).
McCrone, Walter. Judgement Day for the Turin Shroud (Chicago, 1996).
McCrone, Walter C., Lucy B. McCrone, and John Gustav Delly. Polarized
Light Microscopy (Chicago, 2002).
Mills, John S. and Raymond White. The Organic Chemistry of Museum
Objects (Oxford, 2003).
Moens, Karel. “De Viool[bouw] in de 16de en 17de eeuw, I,” Musica
Antiqua 2/1 (1985), pp. 24–26.
Moens, Karel. “De Viool[bouw] in de 16de en 17de eeuw, II,” Musica
Antiqua 2/2 (1985), pp. 38–41.
Moens, Karel. “De Viool[bouw] in de 16de en 17de eeuw, III,” Musica
 Antiqua 2/3 (1985), pp. 85–90.
Moens, Karel. “De Viool[bouw] in de 16de en 17de eeuw, IV,” Musica
Antiqua 2/4 (1985), pp. 123–127.
Moens, Karel. “Authenticiteitsproblemen bij Oude Strijkinstrumenten,”
Musica Antiqua 3/3 (1986), pp. 80–87.
Moens, Karel. “Authenticiteitsproblemen bij Oude Strijkinstrumenten,”
Musica Antiqua 3/4 (1986), pp. 105–111.
Moens, Karel. “Authenticiteitsproblemen bij Oude Strijkinstrumenten,”
Musica Antiqua 4/1 (1987), pp. 3–11.
Moens, Karel. “Renaissance Gambas in het Brussels Instrumentenmuseum:
Vragen rond Toeschrijvingen, Verbouwingen en Authenticiteit,” Bulletin
van de Koninklijke Musea voor Kunst en Geschiedenis Jubelpark Brussels
66 (1995), pp. 161–237.
Moens, Karel. “Problems of Authenticity in Sixteenth-Century Italian Viols
in the Brussels Collection,” The Italian Viola da Gamba: Proceedings of
the International Symposium on the Italian Viola da Gamba: Magnano,
Italy, 29 April–1 May 2000 (Solignac, 2002), pp. 97–114.
Moens, Karel. “Les ‘Violons’ de Charles IX,” Musique, Image, Instruments 5
(2003), pp. 71–96.
Moens, Karel and Klaus Martius, “Wie Authentisch ist ein Original:
Untersuchungen an Zwei Alten Streichinstrumenten des Germanischen
Nationalmuseums Nürnberg,” Concerto 6 (1988), pp. 15–21.
Mondino, Angelo and Matteo Avalle. Manual of Dendrochronology Applied
to the Dating of Musical Instruments (Cremona, 2005).
Mondino, Angelo and Matteo Avalle. Nuova Procedura di Dendrodatazione
con Synchro Search Versione 2.2.1 (2010).
Panshin, A. J. and Carl de Zeeuw. Textbook of Wood Technology (New York,
1980).
Pollens, Stewart. “Le Messie,” Journal of the Violin Society of America 16/1
(1999), pp. 77–101.
Pollens, Stewart. “Messiah on Trial,” The Strad 112/1336 (August, 2001), pp.
54–58.
Pollens, Stewart. “Messiah Redux,” Journal of the Violin Society of America
17/3 (2001), pp. 159–179.
Pollens, Stewart. Stradivari (Cambridge, 2010).
Pollens, Stewart, Pollens, Stewart et al., “The Messiah: A Panel Discussion,”
Journal of the Violin Society of America 17/3 (2001), pp. 181–222.
Rendell, Kenneth W. Forging History: The Detection of Fake Letters and
Documents (Norman, OK and London, 1994).
Ripin, Edwin M. The Instrument Catalogs of Leopoldo Franciolini
(Hackensack, NJ; 1974).
Rowland, Ingrid D. The Scarith of Scornello (Chicago and London, 2004).
Schweingruber, Fritz Hans. Anatomie europäischer Hölzer (Bern and
Stuttgart, 1990).
Striegel, Mary F. and Jo Hill. Thin-Layer Chromatography for Binding
Media Analysis (Los Angeles, 1996).
Stuart, Barbara. Analytical Techniques in Materials Conservation
(Chichester, 2007).
Verlet, Pierre. French Furniture of the Eighteenth Century (Charlottesville,
VA and London, 1991).
von Vega, Georg. Natürliches, aus der wirklichen Grösse unserer Erdkugel
abgeleitetes … Maß-Gewichts- und Münz-System (Vienna, 1803).
Wheeler, E. A., P. Baas, and P. E. Gasson. IAWA List of Microscopic
Features for Hardwood Identification (Leiden, 1989).
Wood Handbook: Wood as an Engineering Material (Madison, 1999).
Zupko, Ronald Edward. French Weights and Measures before the Revolution
(Bloomington, IN and London, 1978).
Zupko, Ronald Edward. Italian Weights and Measures from the Middle Ages
to the Nineteenth Century (Philadelphia, 1981).
Zupko, Ronald Edward. A Dictionary of Weights and Measures for the
British Isles: The Middle Ages to the Twentieth Century (Philadelphia,
1985).
Zupko, Ronald Edward. Revolution in Measurement: Western European
Weights and Measures since the Age of Science (Philadelphia, 1990).

Conservation and craft techniques

Andersen, Paul-Gerhard. Organ Building and Design (New York, 1969).
Apian-Bennewitz, Paul Otto. Die Geige, der Geigenbau, und die
Bogenverfertigung (Weimar, 1892).
The Art and Science of Gilding: A Handbook of Information for the Picture
Framer (Rochester, 1909).
Audsley, George Ashdown. The Art of Organ Building (New York, 1905).
Bachman, Alberto. An Encyclopedia of the Violin (New York, 1925).
Balázsy, Ágnes Tímár and Dinah Eastop. Chemical Principles of Textile
Conservation (Oxford, 1998).
Ball, Philip. Bright Earth: Art and the Invention of Color (New York, 2002).
Barbieri, Patrizio. “Giordano Riccati on the Diameters of Strings and Pipes,”
Galpin Society Journal 38 (1985), pp. 20–34.
Barnes, John. “Pitch Variations in Italian Keyboard Instruments,” Galpin
Society Journal 18 (1965), pp. 110–116.
Barnes, John. “The Specious Uniformity of Italian Harpsichords,” in Edwin
M. Ripin, editor, Keyboard Instruments: Studies in Organology
(Edinburgh, 1971).
Bellucci, Roberto, Paolo Cremonesi, and Ginevra Pignagnoli. “A Preliminary
Note on the Use of Enzymes in Conservation,” Studies in Conservation 44
(1999), pp. 278–281.
Birren, Faber. Principles of Color (Atglen, PA; 1987).
Boeck, Phileas. Die marmorirkunst (Vienna, Pest, Leipzig; 1896).
Bovin, Murray. Jewelry Making for Schools, Tradesmen, Craftsmen (Forest
Hills, NY; 1971).
Bovin, Murray. Centrifugal or Lost Wax Jewelry Casting for Schools,
Tradesmen, Craftsmen (Forest Hills, NY; 1995).
Bovin, Murray. Silversmithing and Art Metal (Forest Hills, NY and New
York, 1995).
Brady, George S., Henry R. Clauser, and John A. Vaccari, Materials
Handbook (New York, 1997).
Brand, Erick D. Band Instrument Repairing Manual (Elkhart, IN; 1978).
Brannt, William T. The Metal Worker's Handy Book of Recipes and
Processes (London, 1896).
Briggs, N. R. The American Tanner (New York, 1802).
Broadhouse, John. The Violin: Its Construction Practically Treated
(London).
Brommelle, Norman S. and Perry Smith, eds. Conservation and Restoration
of Pictorial Art (London and Boston, 1978).
Brommelle, Norman S. and Perry Smith, Urushi (Marina del Rey, 1988).
Brown, Adrian. The Recorder: A Basic Workshop Manual (2009).
Callister, William D, Jr. Material Science and Engineering: An Introduction
(New York, 1997).
Calnan, C. N. Fungicides Used on Leather (Northampton, UK; 1985).
The Care of Ivory, Technical Notes on the Care of Art Objects no. 6 (Victoria
and Albert Museum, London, 1971).
Carlyle, Leslie. The Artist's Assistant (London, 2001).
The Carver and Gilder (London, 1864).
Church, Arthur H. The Chemistry of Paints and Painting (London, 1915).
Conservation of Stone and Wooden Objects: 1970 New York Conference
(London, 1970).
Coutrait, J. -P. Nouveaux Trucs et procédés du bois (Paris, 1991).
de Carle, Donald. Practical Clock Repairing (London, 1994).
de Carle, Donald. Practical Watch Repairing (London, 1995).
de la Rie, E. René and Christopher W. McGlinchey. “New Synthetic Resins
for Picture Varnishes,” Cleaning, Retouching and Coatings: Preprints of
the Contributions to the Brusssels IIC Congress, 3–7 September 1990
(London, 1990).
Denis, Jean. Treatise on Harpsichord Tuning by Jean Denis, trans. and ed.
Vincent J. Panetta (Cambridge, UK; 1987).
DeSantis, Pia C. “Some Observations on the Use of Enzymes in Paper
Conservation,” Journal of the American Institute for Conservation vol. 23,
no. 1 (Fall, 1983), pp. 7–27.
Dietz, Rudolf and Franz Rudolf. Das Intonieren von Flügeln (Frankfurt,
1968).
Donhauser, Peter, ed. Restaurieren, Renovieren, Rekonstruieren: Methoden
für Hammerklaviere (Vienna, 1997).
Dorge, Valerie, ed. Solvent Gels for the Cleaning of Works of Art (Los
Angeles, 2004).
Drayman-Weisser, Terry, ed. Gilded Metals: History, Technology and
Conservation (London, 2000).
Eastlake, Charles Lock. Materials for a History of Oil Painting (London,
1847).
Ellis, Alexander J. and Arthur Mendel. Studies in the History of Musical
Pitch (Amsterdam, 1968).
Elliston, Thomas. Organs and Tuning: A Practical Handbook for Organists
(Sudbury, Suffolk, 1898).
Feisner, Edith Anderson. Color Studies (New York, 2001).
Feller, Robert L., Nathan Stolow, and Elizabeth H. Jones. On Picture
Varnishes and Their Solvents (Washington, D.C., 1985).
Fernbach, R. Livingston. Glues and Gelatin, a Practical Treatise on the
Methods of Testing and Use (New York, 1907).
Ferron, Ernst. The Clarinet Revealed (Paris, 1996).
Field, George. Chromatics or, an Essay on the Analogy and Harmony of
Colours (London, 1817).
Field, George. Field's Chromatography, or, Treatise on Colours and
Pigments as Used by Artists (London, c.1869).
Fields, John. A., Andrew Wingham, Frances Hartog, and Vincent Daniels.
“Finding Substitute Surfactants for Synperonic N,” Journal of the
American Institute for Conservation vol. 43, no. 1 (Spring, 2004), pp. 55–
73.
Finegold, Rupert and William Seitz. Silversmithing (Radnor, PA, 1983).
Fiorentino, P., M. Marabelli, M. Matteini, and A. Moles. “The Condition of
the ‘Door of Paradise’ by L. Ghiberti: Tests and Proposals for Cleaning,”
Studies in Conservation 27/4 (November, 1982), pp. 145–153.
Forsén, Sture, Harry B. Gray, L. K. Olof Lindgren, and Shirley B. Gray.
“Was Something Wrong with Beethoven's Metronome?,” Notices of the
American Mathematical Society 60/9 (October, 2013), pp. 1146–1153.
Frank, George. Adventures in Wood Finishing (Newtown, CT; 1981).
The French Polisher's Handbook, with a Section on Gilding and Bronzing by
“A Practical Man” (London, n.d.).
Fried, Henry, B. Bench Practices for Watch and Clockmakers (Fairfax, VA;
1997).
Fulton, William M. Turpentine Violin Varnish (1988).
The Gilder's Manual (New York, 1876).
Goodrich, Ward. The Watchmakers' Lathe (Fairfax, VA; 1999).
Goodway, Martha and Jay Scott Odell. The Metallurgy of 17th- and 18th-
Century Music Wire (Stuyvesant, 1987).
Grondona, Stefano and Luca Waldner. La Chitarra di Liuteria (Sondrio,
 2001).
Hamilton, Donny L. Methods for Conserving Archaeological Material from
Underwater Sites (www.nautarch.tamu.edu, 2010).
Harding, Rosamond E. M. The Metronome and Its Precursors (Henley-on-
Thames, 1983).
Hasluck, Paul N. Handbook of Knotting and Splicing (London, 1904).
Hasluck, Paul N. Metalworking Tools, Materials and Processes for the
Handyman (London, 1904).
Hasluck, Paul N. Woodfinishing, Comprising Staining, Varnishing, and
Polishing (London, Paris, New York, and Melbourne, 1906).
Hasluck, Paul N. Manual of Traditional Wood Carving (London, 1911).
Haynes, Bruce. A History of Performing Pitch: The Story of “A” (Lanham,
MD and Oxford, 2002).
Haynes, W. M., ed., Handbook of Chemistry and Physics (Boca Raton,
London, and New York, 2010).
Helmholtz, Hermann. On the Sensations of Tone, trans. Alexander J. Ellis
(New York, 1954).
Hemstock, A. On Tuning the Organ (London, 1924).
Heron-Allen, Edward. Violin-Making: As It Was and Is (London, 1885).
Hofenk-de Graaff, Judith H. “The Constitution of Detergents in Connection
with the Cleaning of Ancient Textiles,” Studies in Conservation 13/3
(August, 1968), pp. 122–141.
Hoffman, Edward G. Student's Shop Reference Handbook (New York, 1985).
Horie, C. V. Materials for Conservation (London, 1987).
Howe, Alfred H. Scientific Piano Tuning and Servicing (Clifton, NJ; 1976).
Hubbard, Frank. Harpsichord Regulating and Repairing (Boston, 1963).
Huber, Alfons. “Saitendrahtsysteme im Wiener Klavierbau zwischen 1780
und 1880,” Salzburger Museum Carolino Augusteum Jahresschrift 34
 (Salzburg, 1988), pp. 193–222.
Hughes, Richard and Michael Rowe. The Colouring, Bronzing and
Patination of Metals (London, 1982).
Hunter, Dard. Papermaking through Eighteen Centuries (New York, 1930).
Hurst, Geroge. Painters' Colours, Oils, and Varnishes: A Practical Manual
(London, 1901).
Igrec, Mario. Pianos Inside Out (Mandeville, LA; 2013).
Jackman, James, ed. Leather Conservation: A Current Survey (London,
1982).
Jackson, Albert, David Day, and Simon Jennings. The Complete Manual of
Woodworking (New York, 1994).
Kajitani, Nobuko. “Conservation Maintenance of Tapestries at the
Metropolitan Museum of Art, 1987,” The Conservation of Tapestries and
Embroideries: Proceedings of Meetings at the Institut Royal du Patrimoine
Artistique, Brussels, Belgium, September 21–24, 1987 (Los Angeles,
1989).
Kelly, Ashmun. The Standard Grainer Stainer and Marbler (Philadelphia,
1923).
Kibbe, Richard R., John E. Neely, Roland O. Meyer, and Warren T. White.
Machine Tool Practice (Englewood Cliffs, NJ; 1995).
Kite, Marion and Roy Thomson. Conservation of Leather and Related
Materials (Oxford, 2007).
Koster, John. Keyboard Instruments in the Museum of Fine Arts, Boston
(Boston, 1994).
Kottick, Edward. The Harpsichord Owner's Guide (Chapel Hill, NC; 1987).
Kühn, Hermann. Conservation and Restoration of Works of Art and
Antiquities vol. 1 (London, 1986).
La Niece, Susan and Paul Craddock, Metal Plating and Patination: Cultural,
Technical and Historical Developments (Oxford, 2003).
Lamb, Trevor and Janine Bourriau, eds. Color, Art, and Science (Cambridge,
1995).
Landi, Sheila. The Textile Conservator's Manual (Oxford, 2002).
Latcham, Michael. The Stringing, Scaling, and Pitch of Hammerflügel built
in the Southern German and Viennese Traditions 1780–1820 (Munich,
2000).
Laurie, A. P. The Painter's Methods and Materials (London, 1960).
Lemmer, Geoffrey Michael, “The Cleaning and Protective Coating of Ferrous
Metals,” Bulletin of the American Group – The International Institute for
Conservation of Historic and Artistic Works, 12/2 (1972), pp. 97–108.
Lewis, Walter and Thomas Lewis. Modern Organ Building (London, 1939).
Lincoln, W. A. The Complete Manual of Wood Veneering (New York, 1984).
Mactaggart, Ann and Peter Mactaggart. “Some Problems Encountered in
Cleaning Two Harpsichord Soundboards,” Studies in Conservation 22
(1977), pp. 73–84.
Mactaggart, Peter and Ann Mactaggart, Practical Gilding (Welwyn, Herts.,
1984).
Mailand, E. Découverte des anciens vernis italiens (Paris, 1859).
Maryon, Herbert. Metalworking and Enamelling: A Practical Treatise on
Gold and Silversmiths' Work and Their Allied Crafts (New York, 1971).
McCreight, T. The Complete Metalsmith (Worcester, MA; 1991).
Merk, Linda. “The Effectiveness of Benzotriazole in the Inhibition of the
Corrosive Behavior of Stripping Reagents on Bronzes,” Studies in
Conservation 26 (1981), pp. 73–76.
Merrifield, Mary P. Medieval and Renaissance Treatises on the Arts of
Painting (London, 1849).
Middleton, David and Alan Townsend. Marquetry Techniques (London,
1993).
Millant, Roger and Max Millant. Manuel pratique de lutherie (Paris, 1952).
Mills, John S. and Raymond White. The Organic Chemistry of Museum
Objects (Oxford, 2003).
Mills, John S., Perry Smith, and Kazuo Yamasaki, eds. The Conservation of
Far Eastern Art (London, 1988).
Möckel, Otto. Die Kunst des Geigenbaues (Berlin, 1933).
Monette, L. G. The Art of Organ Voicing (Kalamazoo, MI; 1992).
Moore, R. The Artizans' Guide and Everybody's Assistant (Montreal, 1874).
Nicolaus, Knut. The Restoration of Paintings (Cologne, 1998).
Nicholson, Henry. The Organ Manual for the Use of Amateurs and Church
Committees (Boston, 1879).
Nosban, N. and W. Maigne. Nouveau Manuel complet de l'ébéniste (Paris,
1887).
Oberg, Erik, Franklin D. Jones, Holbrook L. Horton, and Henry H. Ryffel.
Machinery's Handbook 29th Edition (New York, 2012).
O'Brien, Grant. “Some Principles of Eighteenth-Century Harpsichord
Stringing and Their Application,” The Organ Yearbook 12 (1981), pp.
160–176.
O'Conner, Harold. The Jeweler's Bench Reference (Taos, 1988).
Oddy, W. A. “On the Toxicity of Benzotriazole,” Studies in Conservation 17
(1972), p. 135.
Organ, R. M. “The Reclamation of the Wholly-Mineralized Silver in the Ur
Lyre,” Application of Science in Examination of Works of Art (Boston,
1965), pp. 126–144.
Organ Voicing and Tuning (Cincinnati, 1881).
Otto, Jacob Augustus. A Treatise on the Structure and Preservation of the
Violin (London, 1875).
Pawson, Des. The Handbook of Knots (New York, 1998).
Pearson, Colin. Conservation of Marine Archaeological Objects (London,
 1987).
Peruffo, Mimmo. “The Mystery of Gut Strings in the 16th and 17th
Centuries: The Role of Loaded Weighted Gut,” Lute Society of America
Quarterly 29/2 (May, 1994), pp. 5–14.
Petherbridge, Guy. Conservation of Library and Archive Materials and the
Graphic Arts (London, 1987).
Pfeiffer, Walter. The Piano Key and Whippen (Frankfurt, 1967).
Phelan, James. J. and Lillian Burkart. The Complete Guide to the Flute and
Piccolo (Shirley, MA, n.d.).
Piano Action Handbook (Seattle, 1971).
Plenderleith, H. J. and A. E. A. Werner. The Conservation of Antiquities and
Works of Art (Oxford, 1979).
Poletti, Paul. “The Interpretation of Early Wire Gauge Systems – Fixed
Diameters or Proportional Relationships,” Matière et Musique: The Cluny
Encounter (2000), pp. 201–226.
Pollens, Stewart. “Travelling Apprenticeship,” The Strad 97/1160
(December, 1986), pp. 559–564.
Pollens, Stewart. “Early Nineteenth-Century German Language Works on
Piano Maintenance,” Early Keyboard Journal 8 (1990), pp. 91–109.
Pollens, Stewart. “Recipe for Success,” The Strad 120/1429 (May, 2009), pp.
34–38.
Pollens, Stewart. “Some Misconceptions about the Baroque Violin,”
Performance Practice Review 14 (2010), 1–13.
Pollens, Stewart. “Transitional Violin Setup,” Journal of the Violin Society of
America: VSA Papers (2014).
Proceedings of the ICOM Waterlogged Wood Working Group Conference,
Ottawa 1981 (Ottawa, 1982).
Pujol, Emilio. Escuela razonada de la guitarra (Buenos Aires, 1934).
Rendell, Kenneth W. Forging History: The Detection of Fake Letters and
 Documents (Norman and London: University of Oklahoma Press; 1994).
Riffault, Jean René Denis, Armand Denis Vergnaud, and Claude Jacques
Toussaint. A Practical Treatise on the Manufacture of Colors for Painting
(Philadelphia, 1874).
Rogers, Allen. Industrial Chemistry: A Manual for the Student and
Manufacturer (New York, 1931).
Rose, Malcolm and David Law. A Handbook of Historical Stringing Practice
for Keyboard Instruments (Lewes and Long Compton, 2005).
Ruhemann, Helmut. The Cleaning of Paintings: Problems and Potentialities
(New York, 1982).
Sacconi, Simone F. The “Secrets” of Stradivari (Cremona, 1979).
Saska, Ronald. A Guide to Repairing Woodwinds (Glenmoore, PA; 1997).
Sawicki, Carl J. A Method for Adjusting the Oboe and English Horn
(Fredericksburg, TX; 1986).
Scheibler, Johann Heinrich. Der Physikalische und Musikalische Tonmesser
(Essen, 1834).
Scheibler, Johann Heinrich. An Essay on the Theory and Practice of Tuning
in General and on Schiebler's Invention of Tuning Pianofortes and Organs
by the Metronome in Particular, trans. Augustus Wehrhan (London, 1853).
Sease, Catherine. “Benzotriazole: A Review for Conservators,” Studies in
Conservation 23 (1978), pp. 76–85.
Segerman, Ephraim. “Shelley's Guitar and 19th Century Stringing Practices,”
FoMRHI Quarterly 67 (April, 1992), pp. 41–42.
Selwitz, Charles and Shin Maekawa. Inert Gases in the Control of Museum
Insect Pests (Getty Conservation Institute, 2003).
Shortridge, John D. “Italian Harpsichord Building in the 16th and 17th
Centuries,” Contributions from the Museum of History and Technology,
United States National Museum Bulletin 225, Paper 15 (Washington, DC,
1960), pp. 93–107.
Sidaway, Ian. Color Mixing Bible (New York, 2002).
Spon, Ernest. Workshop Receipts for the use of Manufacturers, Mechanics
and Scientific Amateurs, 5 vols. (London and New York, 1896).
Spon, Ernest. Workshop Receipts for Manufacturers and Scientific Amateurs,
4 vols. (London, 1909).
Šrámek, Jiří, Tove. B. Jakobsen, and Jiří B. Pelikán. “Corrosion and
Conservation of a Silver Visceral Vessel from the Beginning of the
Seventeenth Century,” Studies in Conservation 23/3 (August, 1978), pp.
114–117.
Staining and Polishing: The Woodworker Series (Philadelphia and London,
n.d.).
Stambolov, T. The Corrosion and Conservation of Metallic Antiquities and
Works of Art: A Preliminary Survey (Amsterdam, n.d.).
Steinway Grand Key and Action Regulation (New York, 1949).
Stokes, J. The Cabinet-maker and Upholsterer's Companion (Philadelphia,
1889).
Teerink, B. J. Atlas and Identification Key: Hair of West-European Mammals
(Cambridge, 1991).
Thomas, William R., and John J. K. Rhodes, “The String Scales of Italian
Keyboard Instruments,” Galpin Society Journal 20 (1967), p. 48.
Tolbecque, Auguste. L'Art du Luthier (Paris, 1903).
Töpfer, Johann Gottlob. Lehrbuch der Orgelbaukunst, edited by Paul Smets
(Mainz, 1955–1960).
Tortora, Phyllis G. Understanding Textiles (New York and London, 1978).
Traister, John E. Machinists' Ready Reference Manual (New York, 1988).
Tuck, D. H. Oils and Lubricants Used on Leather (Northampton, 1983).
Vermeij, Koen. Tuning and Maintenance of the Clavichord (Amsterdam,
n.d.).
Villus, Aive and Mart Viljus. “The Conservation of Early Post-Medieval
Period Coins Found in Estonia,” Journal of Conservation and Museum
Studies 10/2 (2013), pp. 30–44.
Walton, Harry. Home and Workshop Guide to Sharpening (New York, 1967).
Waterer, John. John Waterer's Guide to Leather Conservation and
Restoration (London, 1972).
Waterlogged Wood: Proceedings of the 2nd ICOM Waterlogged Wood
Working Group Conference (Grenoble, 1985).
Weisshaar, Hans and Margaret Shipman. Violin Restoration: A Manual for
Violin Makers (Los Angeles, 1988).
Wharton, Glenn, Susan Lansing Maish, and William S. Ginell. “A
Comparative Study of Silver Cleaning Abrasives,” Journal of the
American Institute for Conservation 29/1 (Spring, 1990), pp. 13–31.
White, William Braid. Piano Tuning and Allied Arts (Boston, 1946).
Wilder, Tom, ed. The Conservation, Restoration, and Repair of Stringed
Instruments and Their Bows (Montreal, 2010).
Wilkes, Helen, ed. Science for Conservators, 3 vols. (London, 1987).
Wolbers, Richard. Cleaning Painted Surfaces (London, 2000).
Wolfenden, Samuel. A Treatise on the Art of Pianoforte Construction (Old
Woking, 1975).
The Woodworker Series: Staining and Polishing (Philadelphia and London,
n.d.).
Wraight, Denzil. “The Stringing of Italian Keyboard Instruments c. 1500–c.
1650,” PhD diss., Queen’s University Belfast (1997).
Yamaha Band Instrument Repair Manual (Hamamatsu, 1988).
Young, Ronald D. Contemporary Patination (San Rafael, CA; 1988).
Zycherman, Linda A. and J. Richard Schrock. A Guide to Museum Pest
Control (Washington, D.C., 1988).
Ethics and general considerations
Appelbaum, Barbara. Guide to Environmental Protection of Collections
(Madison, WI; 1991).
Atti della giornata di studi sul restauro liutario (Cremona, 1976).
Barclay, Robert, ed. Anatomy of an Exhibition: The Look of Music (Ottawa,
1983).
Barclay, Robert L. The Preservation and Use of Historic Musical Instruments
(London, 2005).
Barclay, Robert L., ed. The Care of Historic Musical Instruments (Edinburgh,
1997).
Barclay, Robert, Florence Gétreau, Friedemann Hellwig, Cary Karp, Jeannine
Lambrechts-Douillez, and Frances Palmer. CICIM Recommendations for
Regulating the Access to Musical Instruments in Public Collections (1985).
Barnes, John. “Does Restoration Destroy Evidence?,” Early Music 8/2
(1980), pp. 213–218.
Bartolomeo Cristofori: la spinetta ovale del 1690, ed. Gabriele Rossi-
Rognoni. Il luogo del David Restauri 3 (Florence, 2002).
Berner, A., J. H. van der Meer, G. Thibault, and N. Brommelle, Preservation
and Restoration of Musical Instruments (London, 1967).
Code of Ethics and Guidelines for Practice (American Institute for
Conservation, 1994).
Gétreau, Florence. “Restaurer l'instrument de musique: l'objet sonore et le
document sont-ils conciliables?” Geschichte der Restaurierung in Europa
2 (Worms, 1993), pp. 145–154.
ICOM Code of Ethics (International Council of Museums, 2013).
Karp, Cary. “Restoration, Conservation, Repair and Maintenance: Some
Considerations on the Care of Musical Instruments,” Early Music 7/1
(January, 1979), pp. 79–84.
Kielinstrumente aus der Werkstatt Ruckers zu Konzeption, Bauweise und
 Ravalement sowie Restaurierung und Konservierung: Bericht über die
Internationale Konferenz vom 13. bis 15. September 1996 im Handel-Haus
Halle. Schriften des Handel-Hauses in Halle 14 (Halle an der Saale, 1998).
Knell, Simon, ed. Care of Collections (London and New York, 1994).
L'Instrument de musique dans les musées: quelle restauration pour quelle
esthétique. Actes du colloque AMS-ICOM Lausanne, November, 1996
(Basle, 1997).
Instruments pour demain: conservation et restauration des instruments de
musique, 9es journees d'etudes de la Section française de l'institut
international de conservation, Limoges 15–16 juin 2000 (Champes-sur-
Marne, 2000).
O'Brien, Grant. “The Conservation of the 1690 Spinetta Ovale by Bartolomeo
Cristofori,” Restauro e Conservazione degli Strumenti Musicali Antichi:
La Spinetta Ovale di Bartolomeo Cristofori (Florence, 2008), pp. 137–150.
Per una carta europea del restauro: conservazione, restauro e riuso degli
strumenti musicali antichi: atti del convegno internazionale Venezia, 16–
19 ottobre, 1985. Quaderni della Rivista Italiana de Musicologia, Societa
Italiana di Musicologia (Florence, 1987).
Pollens, Stewart. “Curt Sachs and Musical Instrument Restoration,” The
Musical Times 130/1760 (October, 1989), pp. 589–594.
Pollens, Stewart. “Flemish Harpsichords and Virginals in the Metropolitan
Museum of Art: An Analysis of Early Alterations and Restorations,”
Metropolitan Museum Journal 32 (New York, 1997), pp. 85–110.
Pollens, Stewart. “Early Alterations Made to Ruckers, Couchet, and
Grouwels Harpsichords in the Collection of the Metropolitan Museum of
Art,” Kielinstrumente aus der Werkstatt Ruckers zu Konzeption, Bauweise
und Ravelement sowie Restaurierung un Konservierung (Halle an der
Saale, 1998), pp. 136–170.
Restauro, conservazione e recupero di antichi strumenti musicali: atti del
Convegno Internazionale Modena 2–4 aprile, 1982). Historiae Musicae
Cultores Biblioteca 40 (Florence, 1986).
Restauro e conservazione degli strumenti musicali antiche: la spinetta ovale
 di Bartolomeo Cristofori, ed. Gabriele Rossi-Rognoni. Proceedings of the
International Workshop organized by the Department of Musical
Instruments of the Galleria dell' Accademia on October 21, 2002
(Florence, 2008).
Sandwith, Hermione and Sheila Stainton, The National Trust Manual of
Housekeeping (London, 1995).
Standards in the Museum Care of Musical Instruments 1995 (Museums and
Galleries Commission, 1995).
Stolow, Nathan. Conservation and Exhibitions (London, 1987).
Thompson, John M. A., ed. Manual of Curatorship: A Guide to Museum
Practice (Oxford, 1994).
Thomson, Garry. The Museum Environment (London, 1978).

Musical instrument studies

Andersen, Paul-Gerhard. Organ Building and Design (New York, 1969).
Apian-Bennewitz, Paul Otto. Die Geige, der Geigenbau, und die
Bogenverfertigung (Weimar, 1892).
Aspects of the Historical Harp; Proceedings of the International Historical
Harp Symposium, Utrecht, 1992, ed. Martin van Schaik (Utrecht, 1994).
Audsley, George Ashdown. The Art of Organ Building (New York, 1905).
Bachman, Alberto. An Encyclopedia of the Violin (New York, 1925).
Baines, Anthony. Woodwind Instruments and Their History (New York,
1962).
Baines, Anthony. Brass Instruments: Their History and Development
(London, 1976).
Barbour, Murray. Tuning and Temperament (East Lansing, MI; 1951).
Barnes, John. “Pitch Variations in Italian Keyboard Instruments,” The Galpin
Society Journal (1965), pp. 110–116.
Barnes, John. “The Specious Uniformity of Italian Harpsichords,” in Edwin
 M. Ripin, ed., Keyboard Instruments: Studies in Organology (Edinburgh,
1971).
Bavington, Peter. Clavichord Tuning and Maintenance (2007).
Boehm, Theobald. The Flute and Flute Playing in Acoustical, Technical, and
Artistic Aspects, trans. Dayton Miller (New York, 1922).
Boehm, Theobald. The Flute and Flute-Playing in Acoustical, Technical, and
Artistic Aspects, trans. Dayton C. Miller (New York, 1922/1964).
Bordas, Cristina and Luis Robledo. “José Zaragozá's Box: Science and Music
in Charles II's Spain,” Early Music (August, 1998).
Broadhouse, John. The Violin: Its Construction Practically Treated
(London).
Chevallier, Claire and Jos van Immerseel, eds. Matière et Musique: The
Cluny Encounter (Antwerp, 2000).
Coates, Kevin. Geometry, Proportion and the Art of Lutherie (Oxford, 1985).
Coelho, Victor Anand, ed., The Cambridge Companion to the Guitar
(Cambridge, 2003).
Cohen, Albert. “A Cache of 18th-Century Strings,” The Galpin Society
Journal 36 (March, 1983), pp. 37–48.
Dobronić-Mazzoni, Rajka. The Harp (Zagreb, 1989).
Dowd, William. “The Surviving Instruments of the Blanchet Workshop,” The
Historical Harpsichord 1, ed. Howard Schott (Brookline, MA; 1984), pp.
17–107.
Ellis, Alexander J., and Arthur Mendel. Studies in the History of Musical
Pitch (Amsterdam, 1968).
Forsén, Sture, Harry B. Gray, L. K. Olof Lindgren, and Shirley B. Gray.
“Was Something Wrong with Beethoven's Metronome?”, Notices of the
American Mathematical Society 60/9 (October, 2013), pp. 1146–1153.
Happiness is a Contented Harp: A Manual on the Care and Regulation of the
Harp (Chicago, 1970).
Harding, Rosamond. The Piano-Forte: Its History Traced to The Great
Exhibition of 1851 (Cambridge, 1933).
Hargrave, Roger “The Working Methods of Guarneri del Gesù and Their
Influence upon his Stylistic Development,” in Giuseppe Guarneri del
Gesù, II (London, 1998).
Heyde, Herbert. Musikinstrumentenbau 15–19. Jahrhundert Kunst-Handwerk
Entwurf (Wiesbaden, 1986).
Hill, William Henry, Arthur F. Hill, and Alfred Ebsworth Hill. Antonio
Stradivari: His Life and Work (1644–1737) (London, 1902).
Hill, William Henry, Arthur F. Hill, and Alfred Ebsworth Hill. The Violin-
Makers of the Guarneri Family (1626–1762) (London, 1931).
Hubbard, Frank. Three Centuries of Harpsichord Making (Cambridge, MA,
1965).
Koster, John. “Toward the Reconstruction of the Ruckers' Geometric
Methods,” Keilinstrumente aus der Werkstatt Ruckers zu Konzeption,
Bauweise und und Ravalement sowie Restaurierung und Konservierung
(Halle an der Saale, 1998), pp. 22–47.
Latcham, Michael. “The Pianos of Johann Andreas Stein,” Zur Geschichte
des Hammerklaviers; 14. Musikinstrumentenbau-Symposium in
Michaelstein am 12. und 13. November, 1993, ed. Monika Lustig
(Michaelstein, 1996), pp. 15–49.
Latcham, Michael. “Authenticating and Dating the Pianos of Anton Walter,”
in Peter Donhauser, ed., Restaurieren, Renovieren, Rekonstruieren:
Methoden für Hammerklaviere (Vienna, 1997), pp.67–82.
Lindley, Mark. Lutes, Viols and Temperaments (Cambridge, 1984).
Lundberg, Robert. Historical Lute Construction (Tacoma, WA; 2002).
Martin, Darryl. “The Native Tradition in Transition: English Harpsichords
circa 1680–1725,” Aspects of Harpsichord Making in the British Isles: The
Historical Harpsichord 5 (Hillsdale, NY; 2009).
Mondino, Angelo. “The Intarsia of Urbino,” De Clavicordio (Turin, 1994),
 pp. 49–55.
Mould, Charles. “James Talbot's Manuscript, VIII. Harpsichord,” The Galpin
Society Journal 21 (March, 1968), pp. 40–51.
Myers, Arnold. Historic Musical Instruments in the Edinburgh University
Collection (Edinburgh, 1990).
O'Brien, Grant. Ruckers: A Harpsichord and Virginal Building Tradition
(Cambridge, 1990).
O'Brien, Grant. “The Development of an Idea: From the Design to the
Instrument,” in Gabriele Rossi-Rognoni, ed., Bartolomeo Cristofori: La
Spinetta Ovale del 1690 (Florence, 2002), pp. 62–79.
Palmer, Susan, and Samuel Palmer. The Hurdy Gurdy (London, 1980).
Pollens, Stewart. “Michele Todini's Golden Harpsichord: An Examination of
the Machine of Galatea and Polyphemus,” Metropolitan Museum Journal
25 (1990), pp. 33–47.
Pollens, Stewart. “Early Nineteenth-Century German Language Works on
Piano Maintenance,” Early Keyboard Journal 8 (1990), pp. 91–109.
Pollens, Stewart. The Violin Forms of Antonio Stradivari (London 1992).
Pollens, Stewart. The Early Pianoforte (Cambridge, 1995).
Pollens, Stewart. “Flemish Harpsichords and Virginals in The Metropolitan
Museum of Art: An Analysis of Early Alterations and Restorations,” in the
Metropolitan Museum Journal 32 (New York, 1997), pp. 85–110.
Pollens, Stewart. Stradivari (Cambridge, 2010).
Pollens, Stewart. “Some Thoughts on the Tuning of the Early Three-String
Violin,” The Galpin Society Journal 64 (March, 2011), pp. 61–66.
Pollens, Stewart. “Bartolomeo Cristofori in Florence,” The Galpin Society
Journal 66 (March, 2013), pp. 7–42.
Pollens, Stewart and Henryk Kaston. François-Xavier Tourte Bow Maker
(New York, 2001).
Pollens, Stewart, et al. Giuseppe Guarneri del Gesù (London, 1998).
Quantz, Johann Joachim. Essay of a Method for Playing the Transverse
Flute, trans. Edward R. Reilly (New York, 1966).
Rensch, Roslyn, The Harp, Its History, Technique and Repertoire (New
York, 1971).
Rephann, Richard, “A Fable Deconstructed: The 1770 Taskin at Yale,” The
Historical Harpsichord 4 (Hillsdale, 2002).
Rockstro, Georgina M. A Treatise on the Construction, the History, and the
Practice of the Flute, Including a Sketch of the Elements of Acoustics and
Critical Notices of Sixty Celebrated Flute-Players (London, 1928).
Romanillos, José. “The Classical Guitar,” in Charles Ford, ed., Making
Musical Instruments: Strings and Keyboard (New York, 1979).
Romanillos Vega, José and Marian. The Vihuela de Mano and the Spanish
Guitar (Guijosa, 2002).
Sacconi, Simone. F. The “Secrets” of Stradivari (Cremona, 1979).
Schwarz, Kerstin. “Bartomomeo Cristofori: Hammerflügel und Cembali im
Vergleich,” Scripta Artium 2 (Autumn, 2001), pp. 23–67.
Shortridge, John. “Harpsichord Building in the 16th and 17th Centuries,”
United States National Museum Bulletin 225/15 (1960), pp. 93–107.
Smithers, Don L. The Music and History of the Baroque Trumpet before
1721 (Carbondale and Edwardsville, IL; 1988).
Steiner, Thomas, ed. Instruments à Claviers: Expressivité et Flexibilité
Sonore (Bern, 2004).
Tarr, Edward, H. “The ‘Bach Trumpet’ in the Nineteenth and Twentieth
Centuries,” in Michael Latcham, ed., Music of the Past: Instruments and
Imagination (Berne, 2006), pp. 17–48.
Thomas, William R. and John J. K. Rhodes. “The String Scales of Italian
Keyboard Instruments,” The Galpin Society Journal 20 (1967), pp. 48–62.
Töpfer, Johann Gottlob. Lehrbuch der Orgelbaukunst, ed. Paul Smets
(Mainz, 1955).
Tromlitz, Johann George. The Virtuoso Flute-Player, trans. Ardal Powell
(Cambridge, 1991).
Vermeij, Koen. The Hubert Clavichord Data Book (Bennebroek, 2000).
von Gontershausen, H. Welcker. Der Flügel (Frankfurt am Main, 1856).
Whitworth, Reginald. The Electric Organ (London, 1930).
Wolfenden, Samuel. A Treatise on the Art of Pianoforte Construction (Old
Woking, 1975).
Wraight, Denzil. “The Stringing of Italian Keyboard Instruments c. 1500–
c.1650,” PhD dissertation (Queen's University Belfast, 1997).
Zur Geschichte des Hammerklaviers: 14. Musikinstrumentenbau-Symposium
in Michaelstein am 12. und 13. November, 1993, ed. Monika Lustig
(Michaelstein, 1996).

Historical writings (pre-1900)

Aaron, Pietro. Thoscanello de la Musica (Venice, 1523).
Adlung, Jakob. Musica Mechanica Organoedi (Berlin, 1768).
Agricola, Georgius. De Re Metallica, trans. Herbert Clark Hoover and Lou
Henry Hoover (New York, 1950).
Alexander, John Henry. Universal Dictionary of Weights and Measures,
Ancient and Modern: Reduced to the Standards of the United States of
America (Baltimore, 1850).
Apian-Bennewitz, Paul Otto. Die Geige, der Geigenbau, und die
Bogenverfertigung (Weimar, 1892).
Bacon, Francis. Sylva Sylvarum: or a Natural History in Ten Centuries
(London, 1627).
Bagatella, Antonio. Regole per la costruzione de' violini, viole, violoncelli, e
violoni (Padua, 1782).
Baillot, Pierre Marie François de Sales. L'art du violon (Paris, 1835).
Bédos de Celles, François. L'Art du Facteur d'Orgues (Paris, 1766–1778).
Bion, Nicolas. The Construction and Principal Uses of Mathematical
Instruments; Translated from the French of M. Bion, Chief Instrument
Maker to the French King by Edmund Stone (London, 1758).
Biringuccio, Vannoccio. De La Pirotechnia, trans. Cyril Stanley Smith and
Martha Teach Gnudi (New York, 1959).
Boeck, Phileas. Die Marmorirkunst (Vienna, Pest, and Leipzig: 1896).
Boehm, Theobald. Die Flöte und das Flötenspiel (Munich, 1871).
Brannt, William T. The Metal Worker's Handy Book of Recipes and
Processes (London, 1896).
Briggs, N. R. The American Tanner (New York, 1802).
Broadhouse, John. The Violin: Its Construction Practically Treated
(London).
The Cabinet-Maker's Guide; or Rules and Instructions in the Art of
Varnishing, Dying, Staining, Japanning, Polishing, Lackering, and
Beautifying Wood, Ivory, Tortoise-shell, and Metal (Concord, NH; 1827).
The Carver and Gilder (London, 1864).
Cavalli, E. Tables de comparaison des mesures, poids et monnaies anciens et
modernes (Marseilles, 1869).
Cellini, Benvenuto. The Treatises of Benvenuto Cellini on Goldsmithing and
Sculpture, trans. C. R. Ashbee (New York, 1967).
Cennini, Cennino D'Andrea. The Craftsman's Handbook: The Italian “Il
LIbro Dell' Arte,” trans. Daniel V. Thompson Jr. (New York, 1954).
Chevreul, M. E. The Principles of Harmony and Contrast of Colors and
Their Application to the Arts (London, 1859).
Corrette, Michel. Méthode théorique et pratique pour apprendre en peu de
temps le violoncello dans sa perfection (Paris, 1741).
Corrette, Michel. Le Mâitre de Clavecin (Paris, 1753).
Cozio di Salabue, Ignazio Alessandro. Carteggio (Milan, 1950).
Cröker, Johann Melchior. Der wohl anführende Mahler (Jena, 1778).
Denis, Jean. Traité de l'accord de l'espinette (Paris, 1650).
Diderot, Denis and Jean le Rond D'Alembert. Encyclopédie ou dictionnaire
raisonné des sciences, des arts et des métiers (Paris, 1751–1772).
Dieudonné, Carl and Johann Lorenz Schiedmayer. Kurze Anleitung zu einer
richtigen Kenntniss und Behandlung der Forte-pianos in Beziehung und
Erhalten derselben, besonders derer, welche in der Werkstatte von
Dieudonné und Schiedmayer in Stuttgart verfertigt werden (Stuttgart,
1824).
Dossie, Robert. The Handmaid to the Arts (London, 1764).
Douwes, Claas. Grondig Ondersoek van de Toonen der Musijk (Franeker,
1699).
Dowland, Robert. Varietie of Lute-Lessons (London, 1610).
Eastlake, Charles Lock. Materials for a History of Oil Painting (London,
1847).
Elliston, Thomas. Organs and Tuning: A Practical Handbook for Organists
(Sudbury, Suffolk, 1898).
Euclid: The Thirteen Books of The Elements, trans. Thomas L. Heath (New
York, 1956).
Fétis, François-Joseph. Antoine Stradivari, Luthier célèbre connu sous le nom
de Stradivarius (Paris, 1856).
Field, George. Chromatics or, An Essay on the Analogy and Harmony of
Colours (London, 1817).
Field, George. Field's Chromatography, or, Treatise on Colours and
Pigments as Used by Artists (London, c.1869).
Galeazzi, Francesco. Elementi teorico-pratici di musica (Rome, 1791).
Galilei, Vincenzo. Dialogo della musica antica et della moderna (Florence,
1581).
Gall, Franz Josef. Clavier-Stimmbuch oder Deutliche Anweisung wie jeder
Musikfreund sein Clavier-flügel, Forte-piano und Flügel-fortepiano selbst
stimmen, repairiren, und bestmöglichst gut erhalten könne. Nebst einer
Nachricht von einigen neuerfundenen musikalischen Instrumenten des
Herrn Joseph Wachtl (Vienna, 1805).
The Gilder's Manual (New York, 1876).
Helmholtz, Hermann. On the Sensations of Tone (London, 1885).
Heron-Allen, Edward. Violin-Making: As It Was and Is (London, 1885).
Hutton, Charles. Recreations in Mathematics and Natural Philosophy
(London, 1803/1840).
Kamarsch, Karl. Jahrbücher des Kaiserlichen königlichen polytechnischen
Institutes in Wien 13 (Vienna, 1828).
Kircher, Athanasius. Musurgia universalis (Rome, 1650).
Kircher, Athanasius. Phonurgia nova (Kempten, 1673).
Kirnberger, Johann Philipp. Die Kunst des reinen Satzes in der Musik (Berlin,
1771–1779).
Kunckel, Johann. Der Neu Kunst- und Werck-Schul (Nuremberg, 1707).
Lemme, Karl. Anweisung und Regeln zu einer zweckmässigen Behandlung
englischer und teutscher Pianoforte's und Klaviere/nebst/einem
Verzeichnisse der bei dem Verfasser verfertigen Sorten von Pianoforte's
und Klavieren, von Karl Lemme, musikalischen Instrumenten macher und
Organisten in Braunschweig (Braunschweig, 1802).
Mace, Thomas. Musick's Monument; or Remembrancer of the Best Practical
Musick, both Divine, and Civil, That Has Ever Been Known, to Have Been
in the World (London, 1676).
Maffei, Scipione. “Nuova invenzione d'un gravecembalo col piano e forte,”
Giornale dei letterati d'Italia (Venice, 1711), pp. 144–159.
Mahillon, Victor-Charles. Elements d'Acoustique Musicale et Instrumentale
(Brussels, 1874; reprinted Brussels 1984).
Marpurg, Friedrich Wilhelm. Versuch über die musikalische Temperatur
(Breslau, 1776).
Maugin, J. C. Manuel du Luthier (Paris, 1869).
Maupertuis, Pierre Louis. “Sur la forme des instruments de musique,”
Histoire de l'Academié de Mathématique et de Physique (Paris, 1726), pp.
215–226.
Merrifield, Mary P. Medieval and Renaissance Treatises on the Arts of
Painting (London, 1849).
Mersenne, Marin. Harmonie universelle (Paris, 1636–1637).
Montal, Claude. L'art d'accorder soi-même son piano (Paris, 1836).
Moore, R. The Artizans' Guide and Everybody's Assistant (Montreal, 1874).
Mozart, Leopold. Versuch einer gründlichen Violinschule (Augsburg, 1756).
Moxon, Joseph. Mechanick Exercises (London, 1677).
Müller, Johann Christian. Praktische Anweisung zum Lakkiren (Leipzig,
1801).
Nachersberg, Johann Heinrich Ernst. Stimmbuch oder vielmehr: Anweisung
wie jeder Liebhaber sein Clavierinstrument selbst repariren und also auch
stimmen könne (Breslau and Leipzig, 1801).
Nicholson, Henry. The Organ Manual for the Use of Amateurs and Church
Committees (Boston, 1879).
North, Francis. Philosophical Essay of Music Directed to a Friend (London,
1677).
Nosban, N. and W. Maigne. Nouveau Manuel complet de l'ébéniste (Paris,
1887).
Organ Voicing and Tuning (Cincinnati, 1881).
Otto, Jacob Augustus. A Treatise on the Structure and Preservation of the
Violin (London, 1875).
Piemontese, Alessio. De' Secreti del Reverendo donno Alessio Piemontese
 (Venice, 1555).
Playford, John. Introduction to the Skill of Music (London, 1664).
Praetorius, Michael. Syntagma Musicum: De Organographia (Wolfenbüttel,
1619).
Quantz, Johann Joachim. Versuch einer Anweisung die Flöte traversiere zu
spielen (Berlin, 1752).
Rayleigh, Lord. The Theory of Sound, 2 vols. (London: vol. I, 1877; vol. II,
1878).
Rees, Abraham. The Cyclopaedia, or, Universal Dictionary of Arts vol. 24
(Philadelphia, n.d.).
Regazzi, Roberto. Il Manoscritto liutario di G. A. Marchi (Bologna [1786],
1986).
Riffault, Jean René Denis, Armand Denis Vergnaud, and Claude Jacques
Toussaint. A Practical Treatise on the Manufacture of Colors for Painting
(Philadelphia, 1874).
Romain, A. Nouveau Manuel Complet de Fabricant de Vernis de Toute
Espèce, (Paris, 1888).
Rose, Algernon. Talks with Bandsmen (London, prior to 1896).
Roubo, André-Jacques. L'Art du Menuisier (Paris, 1769–1774).
Sauveur, Joseph. “Système general des intervalles des sons,” Mémoires de
l'académie royale des sciences (Paris, 1701), pp. 403–498.
Scheibler, Johann Heinrich. Der Physikalische und Musikalische Tonmesser
(Essen, 1834).
Scheibler, Johann Heinrich. An Essay on the Theory and Practice of Tuning
in General and on Schiebler's Invention of Tuning Pianofortes and Organs
by the Metronome in Particular, trans. Augustus Wehrhan (London, 1853).
Sibire, L'Abbé. La Chélonomie ou le parfait luthier (Paris, 1806; Brussels,
1823).
Soler, Antonio. Theorica, y practica del temple para los organos, y claves
 (MS, circa 1775; facsimile published by Sociedad Española de
Musicologia, Madrid, 1983).
Stokes, J. The Cabinet-Maker and Upholsterer's Companion (Philadelphia,
1889).
Streicher, Andreas. Kurze Bemerkungen über das Spielen, Stimmen und
Erhalten der Forte-piano, welche von Stein in Wien verfertiget werden
(Vienna, 1802).
Tartini, Giuseppe. Trattato di musica (Padua, 1754).
Thon, Christian Friedrich Gottlieb. Ueber Klavierinstrumente, deren Ankauf,
Behandlung und Stimmung. Ein nothwendiges handbuch für jeden Besitzer
dieser Art Metallsaiteninstrumente (Sonderhausen, 1817).
Thomée, H. “Untersuchungen über Draht- und Blechlehren,” Zeitschrift des
Vereines Deutscher Ingineure 10 (1866).
Tingry, Pierre François. The Varnisher's Guide (London, 1832).
Töpfer, Johann Gottlob. Lehrbuch der Orgelbaukunst (Weimar, 1855).
Tromlitz, Johann George. Ausführlicher und gründlicher Unterricht die Flöte
zu spielen (Leipzig, 1791).
Vicentino, Nicola. L'antica musica ridotta alla moderna pratica (Rome,
1555).
Vitruvius [Marcus Vitruvius Pollio]. The Ten Books on Architecture, trans.
Morris Hicky Morgan (New York, 1960).
von Gontershausen, H. Welcker. Der Flügel (Frankfurt am Main, 1856).
von Vega, Georg. Natürliches, aus der wirklichen Grösse unserer Erdkugel
abgeleitetes … Maß-, Gewichts- und Münz-System (Vienna, 1803).
Watin, Jean-Felix. L'Art du Peintre, Doreur, Vernisseur (Paris, 1772).
Werckmeister, Andreas. Musikalische Temperatur (Frankfurt and Leipzig,
1691).
Werckmeister, Andreas. Orgelprobe (Quedlingburg, 1698).
Zarlino, Gioseffo. Istitutioni armonische (Venice, 1558).

Temperament
Aaron, Pietro. Thoscanello de la Musica (Venice, 1523).
Barbour, Murray. Tuning and Temperament (New York, 1972).
Bédos de Celles, François. L'art du facteur d'orgues, 4 vols. (Paris, 1766–
1778).
Denis, Jean. Traité de l'accord de l'espinette (Paris, 1650; facsimile, New
York, 1969).
Denis, Jean. Treatise on Harpsichord Tuning by Jean Denis, trans. Vincent J.
Panetta (Cambridge, 1987).
Helmholtz, Hermann. On the Sensations of Tone (London, 1885).
Jeans, James. Science and Music (Cambridge, 1937).
Jorgensen, Owen. The Equal-Beating Temperaments (Raleigh, 1981).
Jorgensen, Owen. Tuning (East Lansing, MI; 1991).
Kellner, H. A. The Tuning of My Harpsichord (Frankfurt, 1980).
Kirnberger, Johann Philipp. Die Kunst des reinen Satzes in der Musik (Berlin,
1771–1779).
Klop, G. C. Harpsichord Tuning (Garderen, 1983).
Lindley, Mark. “Early 16th-Century Keyboard Temperaments,” Musica
Disciplina 28 (1974), pp. 129–151.
Lindley, Mark. Lutes, Viols, and Temperament (Cambridge, 1984).
Link, John. W., Jr. Theory and Tuning: Aron's Meantone Temperament and
Marpurg's Temperament “I” (Boston, 1972). [Highly recommended, but
unfortunately out of print.]
Maffei, Scipione. “Nuova invenzione d'un gravecembalo col piano e forte,”
Giornale dei letterati d'Italia (Venice, 1711), pp. 144–159.
Marpurg, Friedrich Wilhelm. Versuch über die musikalische Temperatur
(Breslau, 1776).
Montal, Claude. L'Art d'accorder soi-même son piano (Paris, 1836).
Peluzzi, Euro. Tecnica Costruttive degli Antiche Liutai Italiani (Florence,
1978).
Pollens, Stewart. “Soler's Temperament and His Acordante,” Proceedings of
the 13th International Symposium on Spanish Keyboard Music, FIMTE,
2012 (Mojácar-Garrucha, forthcoming).
Quantz, Johann Joachim. Versuch einer Anweisung die Flöte traversiere zu
spielen (Berlin, 1752).
Ritchie, Stanley. Before the Chinrest (Bloomington, IL; 2012).
Soler, Antonio. Theorica, y practica del temple para los organos, y claves
(MS, circa 1775; facsimile published by Sociedad Española de
Musicologia, Madrid, 1983).
Werckmeister, Andreas. Musikalische Temperatur (Frankfurt and Leipzig,
1691).
White, William Braid. Modern Piano Tuning and Allied Arts (New York,
1917).

General reference works

Boalch, Donald H. Makers of the Harpsichord and Clavichord 1440–1840
(Oxford, 1995).
Brady, George S., Henry R. Clauser, and John A. Vaccari. Materials
Handbook (New York, 1997).
Clinkscale, Martha Novak. Makers of the Piano 1700–1820 (Oxford, 1993).
Clinkscale, Martha Novak. Makers of the Piano 1820–1860 (Oxford, 1999).
Haynes, W.M., ed. Handbook of Chemistry and Physics (Boca Raton, 2010).
Hopfner, Ruldolf. Wiener Musikinstrumenenmacher 1766–1900 (Tutzing,
1999).
Karp, Cary, ed. The Conservation and Technology of Musical Instruments
(Marina del Rey, CA; 1992).
McCauley, Christopher J., ed. Machinery’s Handbook (New York, 2012).
Vannes, René. Dictionnaire Universel des Luthiers (Brussels, 1951).
von Lütgendorff, Willibald Leo Freiherrn. Die Geigen- und Lautenmacher
vom Mittelalter bis zur Gegenwart (Frankfurt, 1913).
Waterhouse, William. The New Langwill Index: A Dictionary of Musical
Wind-Instrument Makers and Inventors (London, 1993).
de Wit, Paul. Geigenzettel alter Meister vom 16. Zur Mitte des 19.
Jahrhunderts (Leipzig, 1910).
Wittkower, Rudolf. Architectural Principles in the Age of Humanism (New
York, 1971).
 Index
“Alard” violin by Stradivari 9, 341
“Ferdinand David” violin by Guarneri 348
“Harrison” violin by Stradivari 341
“Lady Blunt” violin 341
“Messiah” violin 20, 47, 341
“Sarasate” violin by Stradivari 9, 341
“Soil” violin by Stradivari 9, 341, 343
“Medici” contralto viola by Stradivari 8, 254, 342
“Medici” tenor viola by Stradivari 341–342, 344
“Tuscan” contralto viola by Stradivari 342
“Batta” cello by Stradivari 343
“Paganini” cello by Stradivari 286
“Pawle” cello by Stradivari 286

Aaron, Pietro 318, 326

abalone 220
abrasives 25–26, 31, 188, 206, 208, 211–212, 216, 275
acetic acid 25, 33, 49, 52, 57, 189, 246, 280, 311
acetone 25, 31, 72, 121, 124, 138, 183–184, 188, 224, 266, 271, 275, 280–
282, 284, 295, 327, 335, 402–403
acid finish 115
acoustical measurement 201
acoustics 4–5, 16, 153, 421
acrylic butyl-ester 125
acrylic emulsion paint 119
acrylic paint 188
acrylic resins 25, 37, 121, 124–125, 189, 314
Acryloid 21, 121, 335
Acryloid B-44 21, 25, 36–37, 125, 335
Acryloid B-48 245
Acryloid B-48N 25, 31, 36–37, 183, 277, 335
Acryloid B-67 125, 335
Acryloid B-72 119–121, 125, 138, 184, 188, 335, 402
action cloth viii
adhesives 121, 192
African blackwood 389
Agateen no. 27 25, 36–37, 183, 245, 277
Ageless Eye 59
Ageless Z-2000 59–60
Agtstein see amber
air resonance 4, 6, 153
Albers, Josef 261
alcohol 51–53, 57, 115–116, 125, 192, 217, 258, 266–267, 271, 280, 282,
333–335
alcohol varnish 327
Alexander, John Henry 158, 166, 177, 422, 432
alginate 218
aliphatic resin glue 121, 125
aliphatic solvents 49, 54, 58, 258, 264, 271, 280, 282, 284
alisier 387, 391
alizarin 260
alkaline glycerol 21
alkanes 55
alkanet root 260, 286, 328
alkyl benzene sulphonate 313
alligatoring 329
alloying 204, 208
almond oil 405
aloes 328, 331
alpha brass 24, 204–205, 210
alpha-beta brass 24
alum 49, 51, 53, 56, 189–190, 243, 329–330
alumina 25–26, 31, 52, 275
see also aluminum oxide
aluminum 27, 49, 51–52, 207, 209, 215, 272
aluminum oxide 27, 31, 51–52, 272, 274
Amati, Andrea 104
Amati, Nicolo 8, 348, 350
amber 185, 327, 330–331
American coin silver 275
American Institute for Conservation 1, 28, 104, 284
American Musicological Society 107
American Society for Testing and Materials 268
amides 280
amines 280
ammonia 22, 24, 26, 49, 51–54, 184, 206, 221, 241, 245, 266–267, 276, 280,
285–286, 312
see also ammonium hydroxide
ammonium chloride 32, 54, 56, 120, 230, 277
ammonium citrate 184, 285, 312
ammonium hydroxide 49, 285
ammonium lauryl ether sulphate 313
ammonium nitrate 327
ammonium sulfide 180, 243
ammonium thiosulfate 275
amplitude 11, 13, 317
amylase 284
ancient theaters 4
aniline dye 49, 70, 260, 262
animal-hide glue 44, 51, 60, 121–124, 184, 188, 224, 230–231, 244, 247,
281, 284, 289–292, 328, 395, 401, 417
anionic detergents 43–44, 280, 311–313
annatto 328, 331
annatto seed 328
annealing 29–30, 205–206, 212, 215, 276
antichlors 22, 242–243, 311, 420
antimony chloride 22
Antwerp 126, 157, 288, 379, 387
Antwerp-school harpsichords 289, 389
Antwerp-school wire gauges 363
Appalachian dulcimer 387
Araldite 124
arithmetic proportion 252
Arkansas stone 274
Armstrong cups (woodwind instruments) 404
Arnault of Zwolle 318
aromatic hydrocarbons 125, 280
aromatic solvents 49, 58, 125, 258, 266, 271, 280, 282–284
arsenic 203
arsenic trisulphide 188
ARSTAN 47
Art Sorb 59–60, 177
ash 50, 55, 57, 109, 133, 389, 391
asphaltum 185
Atlantic City Convention Hall organ 222
audio oscillator 201
audio spectrum analyzer 4, 201
authentication 19
automatic instruments 139
AYAA, AYAB, AYAC, AYAF, AYAT 124
azeotrope 282
azo dyes 260

Babcock, Alpheus viii, 360

Bach, Johann Sebastian 318
back check 249
back saw 407
backfalls 234
Bacon, Francis 6–7, 15, 421, 432
badger blenders 126
Badura-Skoda, Paul 251
Bagatella, Antonio 9, 255, 432
balance mortise 233
balanced action, organ 233
balling-out die 35
ballpoint pen 186
balsam 185, 327, 331
bandora 6, 346
banjo 244–245
Barber, Stephen and Sandi Harris viii
Barbieri, Patrizio 347, 351
Barbour, Murray 42, 320, 326, 429, 434
Barclay, Robert L. 1–2, 142, 428
barite fillers 241
Barnes, John 1–2, 269–270, 361, 386, 423, 428–429
Baroque pitch 222, 251, 270, 361
Baroque trumpet 15, 17
Baroque violin 340–341, 343
Baroque violin fingerboards 343
Baroque violin fittings 156, 340
Barrier brand gloves 267
bars, vibrations in 14
Baskerville, John 241
bass-bar viii, 338–339, 348, 352
bassoon 402, 404
bassoon stop 112
basswood 387, 389–390, 392
Baumé, Antoine 285
Baumé scale 285
bead saw 407
bead-and-reel molding viii, 412
bear-claw figure 393
beat rates 318, 323–325
beating reeds 232
beats 13, 217, 317, 323–324
bebung 40
Beccaria, Giovanni Battista 70
Bechstein 105
Becke-line test 259
Bédos de Celles, François 229–230, 238, 432
beech 291, 387, 389
beeswax 32, 49, 191, 213, 215, 271, 286, 357, 399, 403
Belgian honing stone 274
bell iron 29
bell wire 35
Benade, Arthur H. 11, 15–16, 421
bench grinder 271
bending brake 208
bending jigs 208
bentside spinet 366
benzene 27, 49, 51, 280
benzoin 115, 286, 332
benzol sulphonate 313
benzotriazole 21, 25, 27, 36–37, 276, 327
Bergonzi, Carlo 254, 349
Berlin Physical Institute 10
Berlin valve 33
Berlin wire gauge system 357
Bernard, Edward 171
Bertrand, Nicolas 105
Bessemer process 182
beta brass 24, 210
Bettoney, H., firm of wind instrument makers viii, 405
BEVA 124
B-form cello of Stradivari 343–344
Bion, Nicolas 158, 177
bird’s eye maple 393
bird’s-eye figure 393
Biringuccio, Vannoccio 182–183
Bisiach, Leandro 332
bistre 243
black light 328
black polishing 70
blackwood 20, 390, 404
bladder skin 402–404
bladder-skin pads, woodwind instrument 402
Blanc, Charles 260
Blanchet, François 347, 369
Blanchet, family of harpsichord makers 387–388
bleach 21–22, 184, 243, 267, 312, 419–420
bleaching paper 241
block plane 111, 409
blow-horn anvil 29
bluing 22
boar bristle springs 151, 300
Bodleian Library, Oxford 130
Boehm, Theobald 10, 16, 20, 387, 396, 399, 402, 405, 430, 432
Boethius, Anicius Manlius Severinus 4, 252
bog oak 387, 391
Böhm, Joseph 378
bole 117–121, 123
Bologna, Biblioteca Comunale 9
bone black 242, 262
bone glue 121–122, 124
Boosey & Co viii, 35
borax (sodium tetraborate) 32, 50, 124, 190, 278, 280
borescope 227, 289, 291
boric acid 32, 208, 280
bouging metal 212
bows, violin 136–138, 184, 201, 220, 246, 264, 314, 389, 391, 396
boxwood 23, 246, 285, 387–389, 404
Brahms, Johannes 348
branched-chain alkanes 280
brass 12, 23–26, 30–31, 33, 36–37, 126, 138, 203–205, 207, 209, 212, 270,
272, 314, 327, 335, 346, 359, 361, 368, 404, 410
brass instruments 15, 24–25, 28, 32–33, 35, 135, 137, 193, 206, 208, 210–
212, 251, 277–279
brazil wood 43, 286, 328, 391
brazing 31–32, 34–35, 279–280
breaking frequency of strings 133, 269
breaking pitch 194
breaking tension 154
Brealey, John 45
bridge 7, 41, 121, 124, 140, 179, 253, 293, 296, 348, 361, 387
bridge pin 314
bright dips 26
Brita water filter 356
Britannia silver 275
British Library 330
British Museum 45, 191, 276
British Petroleum 282
Broadwood viii, 112, 218, 324, 362, 375, 410
Broadwood, John 367
Broadwood square pianos viii, 218, 362, 375, 410
bronze 23, 25, 37, 112, 126, 132, 204, 207, 327, 335
bronze disease viii, 25, 37, 135
Brown, Mrs. John Crosby 105–106
Brown, William Adams 106
Brown & Sharpe taper 305
buckthorn berries 189, 260, 329
Buffet, firm of woodwind instrument makers 400, 404
buffing 31, 208, 212, 245, 272
buffs 212
Bulls Eye shellac 334
burnisher 20, 28, 30–31, 120, 208, 232, 236, 274, 413
burnishing 119, 208
burnt sienna 118, 126–127
burnt umber 126–127, 246, 260, 262
butyl cellosolve 280
butyl methacrylate 125

cabinet upright piano 313

cabriole legs 411
calcium carbonate 32, 50–51, 53–54, 58, 117, 220, 242, 286
calcium hydroxide 50, 53–54, 56, 241–242
calcium oxide 50, 53, 55, 121, 220, 244
calcium sulfate 51, 53, 55, 117
calfskin 112, 402
Calgon (sodium hexametaphosphate) 21, 27, 37, 50, 311, 356
calipers 197–200, 211
Cameron, Rod 246
Cammerton(e) 251, 385
campeachy wood 328
canola oil 405
carabé 330
carat 207
carbon black 188
carbon fiber 404
carbon tetrachloride 267, 280
carbon-14 dating 46
carboxymethyl cellulose 311, 313
carmine lake 189
Carnegie Hall 251
carrying cases 24, 137–138, 140–141
carving gouges 271, 273
casein 124, 289
casein glue 124
casting pliers 231
cationic detergents 43, 280, 311
cedar 107, 388, 392
Celcon 150
Cellini, Benvenuto 120–121, 219, 432
Cellofas 125
cellosolve 271, 280
Cellugel 192
cellulose nitrate 50, 53, 55, 125
Celsius, Anders 171
Celsius temperature scale 171, 308
cembalo traverso 372
Cennini, Cennino D’Andrea 117, 121, 219, 432
cents 38, 42, 182, 316, 325, 346
cents conversion 38
cents equivalents 39
cerium oxide 31
Cerrobend 34
chalk 25, 27, 36, 51, 57, 115, 229, 247, 275–276
chalk-fitting 210, 418
chamber pitch 251
chamois 244
charcoal 139
chasing 208
Chatterton compound 232
check height 108, 110–111, 249
cherrywood 109, 387
chestnut 387
Chevreul, M. E. 260, 264
Chevrier, André Augustin viii, 132
chicory 246, 286, 294
chiff 235–236
chisels 271, 273–274, 308, 407, 411
chitarrone 392–393
Chladni, Ernst 10
Chloramine-T 242–243, 285, 420
chlorine bleach 22, 266, 312
chloroform 280
choir pitch (Chorton) 251
chroma 261
chrome plating 27, 51, 71
chromium oxide 28, 208, 272–273
chucks, lathe 216
Church, A. H. 260
cinnabar 188, 260
ciphers, organ 226–227
circle of fifths 317
circulating temperament 317
citric acid 21, 243, 284–285, 420
cittern 6
clamps 124, 132, 224, 228, 277, 416
clarinet viii, 15, 252, 389, 397, 400, 402, 404
clavichord viii, 39–43, 294, 318, 362, 373–375, 381, 383–384
cleaning emulsion for wood 356
Clorox 51, 312
closed system (temperament) 317
cochineal 189, 260, 328–329
cocuswood 20, 387, 404
COFECHA 47
coin silver 32
colles d’os 122
colles de nerfs 122
colles de peaux 122
collets 215
colophonium 332
color circle 260
color rendering index 264
color temperature 263
color theory 260
Comité International des Musées et Collections d’Instruments de Musique
288
comma of Didymus 316, 318
comma of Ptolemy 316, 318
compass plane 410
compass saw 408
compo viii, 44–45, 51, 121, 144
conservation reports 45
constant Q transform 4
Conté, Nicolas Jacques 242
Continental wire gauge systems 357
Convention on International Trade in Endangered Species of Wild Fauna and
Flora (CITES) 107
copal 185, 259, 332
coping saw 408
copper 17, 21, 23, 25–26, 32–34, 37, 50–52, 54, 56–57, 70–71, 131, 147,
149, 188, 203–207, 221, 229, 275–276, 278, 314, 327, 335, 344, 347–
348
copperas 50–52, 58, 70
cork 33–34, 108, 402–403
cormier 387, 391
Corrette, Michel 344, 351, 361, 382, 386
Cosmoloid H80 32, 192
cotton 184, 310, 312–313, 399
Coulomb, Charles Augustin 269
couplers, organ 136, 234
Cozio di Salabue, Count 254, 333, 335, 432
cracks 30–31, 121, 227–228, 244, 258, 289–293, 395, 404, 417
craquelure 329
crazing 114, 329
cream of tartar (potassium tartrate or bitartrate) 32, 51, 189, 276, 314, 330–
331
creases 211
Cremer, Lothar 11, 16
Cremona 155, 253, 255, 387
Cremonese braccio 155
Cremonese oncia 155
Cremonese violin making 155
Cristofori, Bartolomeo 3, 19, 106–107, 156–157, 288, 301, 361, 371–372,
388, 396, 428–429
critical micelle concentration 311
crocus 28, 52, 197, 208, 232
crocus cloth 297, 355
Cröker, Johann Melchior 332, 335
Crosby Brown Collection of Musical Instruments 105
Crowell, William viii, 228
crowning of soundboards 292
cryogenic treatment of brass 205
cupric acetate 120
cupric nitrate 245
curcuma 329, 332
cutoff bar 292–293
Cuttersil 144
cyanide 55, 71, 266–267
cyanoacrylate glue 125
cycloalkanes 54, 280
cycloeheptane 280
cyclohexane 280–281
cypress wood 47, 107, 288, 292, 388, 392

dammar 329
dampers 112, 140, 152, 190–192, 250, 297, 315
damping board 40
dapping 208
de Mayerne, Theodore 330
De Mayerne Manuscript 330
decibel 13
deckle edge 241
Deconet, Michele 19
deerskin 111
Dehypon LS45 44, 275, 310, 312
deionized water 22, 126, 184, 187–192, 207, 221, 229, 241–242, 258, 275–
276, 283–285, 289, 312–313, 356–357, 420
Delrin 150, 315
dendrochronology 20, 46–48, 393, 422
Denis, Jean 158, 160, 177, 314, 320–321, 326, 406, 424, 427, 432–434
dent balls 29, 229
dent removal tools viii, 29
Dentmaster 30
dents 26, 28–30, 206, 210–211, 228–229, 235, 258, 404
Derham, William 9
detergents 43, 188, 229, 266, 280, 310–312
dezincification 24, 206
diacetone alcohol 281
dial test indicators 200
diatonic scale 315–316
dichroism 261
dicyclohexylamine nitrite 327
Diderot, Denis 158, 160, 177, 314, 406, 420, 432
die-maker’s reamers 306
dieses 317
diethyl ether 52, 280–281
Dieudonné, Carl 109
dimethylformamide 72, 280
Dipper, Andrew 9, 16, 421
disinfection 58
disinfestation 58, 135, 301
display cases viii, 137, 405
distemper 126–128
distilled water 283, 356
ditonic comma 315–317
dogfish skin 112, 117
Dominant strings 345
Dossie, Robert 243, 271, 432
Douwes, Claas 154, 379, 386, 432
dovetail saw 407
Dowd, William 301, 388, 396
Dowicide 1 60, 192
Dowicide A 60
Dowland, Robert 116–117, 133–134, 432
dragon’s blood 328, 331–332
draw filing 210
draw-bench 34
drawknife 411
drawplate 30–31, 34, 209
drill bits 60, 208–209, 215, 274, 302, 414
drill press 208
drilling 200, 208–209, 215, 227, 246, 291, 414
drop black 70, 126
drum planer 228
drums 15, 106, 137, 141, 244–245, 389
Du Hamel, Jean Baptiste 9
Duco Cement 125
ductility 204–205
Dutch Broadcasting System 256
dyes 17, 21, 43, 49, 135, 191, 241–242, 244, 246, 259–260, 262–263, 285,
310–311, 314, 328
dyes, fugitive 135, 260, 286, 329

early keyboard instrument temperaments 318

earth pigments 189, 260
Eastlake, Charles Lock 329, 335, 424, 432
ébénisterie 406
ebonite 355
ebonizing 70
ebony 70, 112, 286, 340, 343–344, 355, 387, 389–390, 404, 413
echoes 10
EDTA 25–26, 36, 126, 183–184, 190, 221, 232, 275, 285, 311–312
egg tempera 119, 121, 294
electrical conductivity 356
electrical resistivity 356
electrochemical plating 314
electrocleaning 30, 70, 275
electro gilding 117
electrolytic cleaning 267
electrolytic reduction 183, 190
electronic tuners 4, 42, 46, 201, 250, 316, 325, 397
electroplating 30, 70–71, 120, 276
electroplating solutions 71
elephant ivory 184
Ellis, Alexander J. 251–252, 324, 326, 424–425, 430
Elmer’s glue 125, 291
Elvace 125
emery 28, 52, 208, 212, 232, 275, 355
endoscope 289
endpin 246
energy-dispersive X-ray analysis 17
English bentside spinet 300
English harpsichords 288–289
English keyboard instruments 389
English piano actions 111
English pianos 362
English square pianos 315, 362, 375
English wheel 30
English wire gauges 362
engraving 31, 36, 210, 212
enharmonics 41, 315, 317
enzyme cleaning 284
epi-fluorescence illumination 18
epi-fluorescence microscopy viii, 18, 115, 328–329, 388
epoxy 31, 72, 124, 227–228, 291, 314, 401, 404
equal beating temperament 317
equal temperament 38, 41, 116, 270, 315–318, 320, 323–324, 397
equal temperament frequency table 73–103
equilibrium moisture content 394–395
Erdesz, Otto 334
erh-hus 244
escapement mechanisms 297, 300
escapements 111
Escorez 259, 335
esters 280
ethanol 25, 32, 44, 52, 54, 114–115, 119, 121, 124, 184, 192, 224, 242–243,
249, 258, 260, 264, 281–282, 284–285, 290, 327, 335, 357, 420
ethics 1–2, 104, 428
ethnographic instruments 135, 139, 141, 245
Ethulose 125
ethyl acetate 49, 121, 280
ethyl vinyl acetate resins 124
ethylene vinyl acetate co-polymer 125
ethylenediamine tetra-acetic acid (EDTA) 25, 37
Euclid 253, 255, 432
European Union classification of hazardous substances 266
eutectic 204, 230, 278
eutectic point 204
Evasol 125
Everclear grain alcohol 115
Exxon-Mobil 282
Eytelwein, Johann Albert 171

Fagotzug 112
Fahrenheit temperature scale 308
Faraday, Michael 70
fatigue 204
faux finishes 126
feather 139, 141, 150
feeler gauges 200
felt 112, 135, 143, 186, 249, 310, 313, 402
felt-tipped pen 186
ferric acetate 246
ferric chloride 22
ferrous sulfate 51–52, 58, 286
Fétis, François-Joseph 344, 351
fiddleback maple 391, 393
file, screw-slot 215
files 209–210, 212, 274, 414
filters, UV and IR 18, 180, 334
fingerboard 70, 116, 130, 141, 253, 269, 340, 343–344, 348
fingerboards, Baroque 343–344
finite element analysis 4
fir 109, 245, 292, 338, 387–388, 392
fire extinguishers 267
fire safety 267
firescale 32, 206, 276, 278, 280
fish glue 121–125, 289, 314
Fitzpatrick, Horace 27
flamenco guitars 392
Flamsteed, John 10
flap valves 223
Flemish harpsichords 126, 288–289, 294, 301, 379, 389
Fletcher, Neville H. 11, 16
floats 414
Florence, Italy 16, 19, 371–372, 388
Florentine soldo 156
flute viii, 10–11, 15, 20, 104, 108, 135, 246, 275, 387, 389, 396–397, 399–
400, 402, 404
flute end cork 403
flux 31–33, 70, 267, 277–278, 280
flux, solder 33, 230
fly stains 285, 420
Fodara Vedla Alm chronology 48
Fogliani, Ludovico 252
Folkes, Martin 171
Food and Drug Administration 268
force gauges 201
forging metal 20, 187, 210
formaldehyde 49–50, 192
formic acid 25, 275
fortepiano viii, 60, 108, 111, 141, 251, 289, 296, 362, 365, 376–378, 383, 385
fortepiano hammer 111, 300
fortepiano maintenance manuals 109
Fourdrinier, Henry 241
Fourier transform 4
Fourier-transform infrared spectroscopy 17, 286, 327
foxing 241
frame saw 406–407
Franciolini, Leopoldo 19–20
Frankfort black 243
Frecker, William viii, 289
free reeds 221, 232
free-cutting brass 204, 209
French Baroque pitch 251, 269
French cement 403
French Classical organ 233
French harpsichords 289, 300
French polishing 70, 114–116, 188, 258, 283–284, 290, 328, 334
French Revolutionary (or Republican) calendar 38
French Royal Academy 8
French soundboard painting 294
French standard silver 275
French wire gauges 362
frequency 11, 13–15, 39, 133, 217, 235, 269, 324
frequency counter 4, 201
fret law 41–42, 253, 270
frets 116, 156, 194, 315
fretting 42, 116, 156, 318–320
fuliggine 52, 246
fumigation 58
fundamental 11, 14–15
furniture beetle 58
fusing, tortoiseshell 50, 314
fustic 328

Gafurio, Franchino 252

Galeazzi, Francesco 347, 351
Galilei, Galileo 5
Galilei, Vincenzo 5, 116–117
Gall, Franz Josef 109, 386
gallic acid 23, 180
gamboge 328, 332
gas chromatography/mass spectrometry 16
gas lamps 24
Gasparo da Salo 46
Gassendi, Pierre 9
gauges, string 108, 198, 211, 335, 346, 348, 382
gelatin 244
gelled solvents 284
Genklene 52, 312
geometric proportion 252
German silver (nickel silver) 15, 24–26, 30–31, 36, 138, 207, 221, 280, 404
Germanisches Nationalmuseum, Nuremberg 187
Gerson, Michael 334
gesso 117–118, 121
gesso grosso 117
gesso sottile 117–118
Giertz Obergurgl chronology 48
gilder’s cushion 119
gilder’s knife 119
gilder’s tip 118–119
gilding viii, 23, 70, 116–118, 120–121, 185
gilding, foil 70
gilt brass 25
giraffe piano 313
glair 52, 120
glass-paper 114–115
Gleichläufigkeit 48
gleichschwebende Temperature 317
Globally Harmonized System of Classification and Labelling of Chemicals
266
gloves 21–22, 28, 136, 188, 228–229, 233, 238, 267, 297
glue pot 123
glues 16, 121–122, 124–125
gluing end-grain 417
glycerin 192, 313
glycol ethers 280
go-bar deck 293, 416
Goermans, Jean 105
Goethe, Johann Wolfgang von 260
gold 12, 15, 17, 23–24, 32–33, 49, 70–71, 118, 120–121, 126, 188, 203–205,
207–208, 262, 280
gold, white 70
gold leaf 117–120
gold-plated silver 207
gold solder 31–33, 279
goldbeater’s skin 224, 244, 402
Goodway, Martha 182–183, 203, 269–270, 361, 386, 422, 425
goose grease 112, 399
gouache 294
gouges 412
Graf, Conrad viii, 111, 157, 293, 296, 357–359, 365, 376–377
grain painting 126
graining combs 126–127
graining rollers 126
granadilla 387, 404
graphite 53, 112, 151, 208, 219, 225, 242
gravers 120, 210–211, 308
Graves & Co. viii, 397
Greek pitch 329
Greenway, Henry viii, 144
grinding 207, 211, 271–275, 409–410, 412, 414
grit size 128
Grouwels, Lodewijck 106
Guarini, Guarino 254
Guarneri, Giuseppe viii, 5, 186, 255, 264–265, 339, 348, 350, 393
Guarneri, Peter viii
guitar viii, 105–106, 116, 121, 126, 130–132, 178, 194, 253, 269, 273, 315,
334, 361, 391–392
guitar-makers’ clamps 416
gum ammoniac 286
gum arabic 120, 126, 230, 243, 332
gut strings 13, 116, 130, 133–135, 142, 145, 149, 156, 194–195, 203, 269–
270, 335, 344, 346–348
gut strings, tensile strength 133
gut strings, twist rate 195
gutta percha 51, 53, 232, 355

Haag, Sabine 184

Hacklinger gauge 198, 200, 293
Hall, Donald E. 11, 16
Halley, Edmond 10
halogenated hydrocarbon 266, 280
Halon 267
Hamamatsu Museum viii
hammer coverings 111, 191, 251
hammer drop 249
hammer leather 192
hammer let-off 108, 111, 249
hammer shank 113, 249
hammer-beak leather 112
hammers 29, 211–212, 250, 308
hand saw 274
handling musical instruments 135
harmonic oscillator 11
harmonic proportion 252
harmonic series 15
harmonica reeds 183
harmonics 11, 15, 235–236
harmonium 221, 232
harp viii, 7, 44, 126, 138, 142–145, 155
harper’s knot viii, 142–143
harpsichord 1, 7, 60, 104–106, 128, 150–153, 157, 183, 255, 268, 298, 320–
321, 347, 360–362, 367, 369–372, 379, 382–383, 389
harpsichord, plucking order 151–152
harpsichord plectra 150–152, 190, 300
Harris, Baker viii, 153, 300, 366
Hartlib, Samuel 346
Hass, Johann Adolph 373
Hawkes & Son 399, 405
Haynes, Bruce 251
Haynes company 275
Hayter, Charles 260
hazardous substances 266
heat sinks 30
heat treatment of metal 30, 204–206, 228
Heifetz, Jascha 348
Helmholtz, Hermann von 10–11, 14, 16, 153, 252, 326, 421, 425, 433–434
Helmholtz resonator 10–11, 14, 153
hemlock wood 387
heptane 53, 280, 282
hermaphrodite calipers 198
hexane 280, 282
hexanes 115, 137, 191, 313, 355
Heyde, Herbert 36, 157, 177, 430
high performance liquid chromatography 17
Hildebrand solubility parameter 281
Hill firm of violin dealers 341, 348
Hills, firm of violin dealers 37, 72
historical metrology 153
hitchpin 292, 314
hitchpin loops 109, 113
hitchpin rail 289, 292
HMG glue 125
hog-bristle mottlers 126
holly 387–388
hologram interferometry 4
honing 211, 272–274, 411–412
Hooke’s law 206
horn 314
horsetail, Equisetum 117
hot-air gun 28
hot-melt glues 125, 402
Houston, R. A. 260
Howard, John 171
Hubbard, Frank 153, 382, 386, 425, 430
Huber, Alfons 357–358, 366, 386, 425
Hubert, Christian Gottlob 373
Huggins, William 348
humidity 37, 59–60, 111–112, 121–122, 135–141, 152, 156, 177, 187–188,
191, 238, 241, 244, 247, 257, 269, 287, 292–293, 300–301, 337, 393–
395, 417
humidity control 177
Hungarian ash 389
hurdy gurdy 178
Hutchins, Carleen Maley viii, 11, 15–16, 179, 421
Hutton, Charles 9–10, 16, 158, 162, 171, 177, 421–422, 433
Hxtal 124
hydraulic blowout press 35
hydraulic draw press 209
Hydrocal 218, 271
hydrocarbon resin 125
hydrochloric acid 32, 54, 57, 190, 277, 285
hydrogen embrittlement 71
hydrogen peroxide 21–22, 33, 55, 184, 241, 280, 285, 312, 417, 420
hydroxyethyl cellulose 125
hydroxypropylcellulose 125, 192

IBM typewriters 186

illumination levels 135, 178, 184, 191
Incralac 21, 25, 36–37, 245, 335
India ink 70, 242, 334, 420
India rubber 355
India stone 274
indigo 242, 328
infrared 17, 180, 328
infrared photography 180, 328
infrared reflectography 242, 328
ink 21, 180, 186–187, 241–243, 260, 285
ink stains 420
inscriptions 19–20, 45, 180, 186, 225, 241–242, 297
Intelligenz-Blatt zur allgemeinen musikalischen Zeitung 357
International Council of Museums 1, 104
International Tree Ring Database 48
ion-exchange resins 283
Irish harp 6, 142, 147
iron 12, 22, 24, 53, 135, 182–183, 203–205, 207, 210, 229, 246, 250, 268,
285, 330, 347, 356, 359–361, 368, 382
iron oxide 188
iron sulfate 70
iron-gall ink 180, 186, 241–243
iron-gall ink stains 420
irregular system (temperament) 317
irrigated windchest 227
isinglass 53, 123–124, 244, 314
isoparaffinic solvents 280, 282
Italian harpsichords 124, 287–289, 292, 389, 392, 395
ivory 15, 44, 107–108, 112, 114, 116, 123, 126, 135, 178–179, 184, 215, 246,
297, 344, 397, 403, 408
ivory, yellowing 179
ivory black 70, 185, 242, 262, 271

jack guides 300

jack rail 152
jacks viii, 150–152, 297, 299–300, 387, 389, 391–392
Jade 403 and 711 125
Japan gold size 119
Japanese planes 409, 411
japanning 114, 116, 185
Javel water 21
Jean le Rond D’Alembert 158, 160, 177, 314, 406, 432
Jecklin disc 256
Jennings wood boring bits 275, 414
jeweler’s saw 212, 215, 220, 408
jigs 35, 226, 273, 277, 400, 406
Jorgensen, Owen 324, 326
Jorgenson clamps 416
just intonation 317

Kamarsch, Karl 364, 386, 433

kapsels 112
karat 118, 207
Kaston, Henryk 246, 431
Kelly, Ashmun 126–128, 425
kermes 189, 260, 328–329
ketones 280
key cloths 40, 108, 111, 113, 248, 297, 310, 313
key dip 39, 108–110, 152, 233, 249, 297
key lever 39, 247
keyboard 157, 201, 222, 233–234
keyboard instrument restoration 287
keyboard instruments 48, 106, 135–136, 138–139, 253, 268–269, 273, 287,
316, 325, 357, 360
keyboard octave span 157
keyhole saw 408
keys 33, 135, 225, 387
keywork 233
killed acid flux 32, 277
kingwood 389, 404
Kintzing, Christian 374
Kircher, Athanasius 6, 16, 252, 254–255, 433
Kirckman, Abraham and Joseph 367
Kirckman, Jacob 152
Kirkman, family of harpsichord makers 106
Klucel 125, 192
knives 118, 273, 413
knurls 211
Kodak Color Dataguide 261
Koster, John 270, 361, 386, 425, 430
koto 390
Kremer Pigmente 124
Kunckel, Johann 330, 335
Kundt, August 10
Kunsthistorisches Museum, Vienna 184

labels 19–20, 186–187, 271

lac lake 328
lacquer 187–188
lacquered instruments 135, 139, 141, 188
lactic acid 228
laid paper 241
lake pigments 17, 189, 260
lampblack 53–54, 242, 262, 355
lanolin 312
lapping 211
larch 47
Laropal K-80 335
Lascaux 360 125
laser printers 187
Latcham, Michael 16, 357–359, 364, 386, 426, 430
latex gloves 137–138, 266
lathe 33–35, 132, 200, 208–209, 211–216, 220, 229, 246, 400–401, 403–404,
412, 416
Law, David 240, 352, 359, 387, 427
Le Blon, J. C. 260
lead 23, 28, 35, 50–58, 189–190, 204–205, 207, 209, 222, 228–229, 231–232,
298, 314, 356
lead acetate 128
lead pencil 242
lead/tin alloys 189
lead weights 297
lead white 126–127, 330
leather 24, 44, 54, 112, 120–121, 125, 135, 178, 190–192, 224–227, 288, 330
leather dressings 191
leathering of piano hammers 111–112
Leman, Anatoli 5
Lemme, Carl 239, 374
Lemme, Karl 109, 111, 433
Library of Congress 192
Library of San Marco in Venice (Libreria Marciana) 329
Librem Segreti de Buttegha 9, 16, 421
lime 50–51, 53–54, 56–57, 112, 121, 124, 190, 212, 220, 244, 387, 389
limewood 288
Lindley, Mark 117, 255, 318, 326, 430, 434
linear alkanes 280
linen 310, 312–313, 399
linseed oil 23, 44, 52, 54–55, 70, 114, 128, 137, 246, 285–286, 327, 329–331
lipase 284
litharge 330–331
litmus reaction 43
liver of sulfur 245
logwood 70, 286, 328
long-pattern violin 8–9, 254
Lorée, firm of woodwind instrument makers 404
lost-wax process 217
Lot, Louis 275
lubricants 36, 137, 139, 192–193, 399
Lucchi meter 4, 201, 396
Lundberg, Robert 117, 430
Lupot, Nicolas 9, 345
lute viii, 6–7, 48, 54, 106, 116–117, 121, 124, 133–134, 193–196, 253, 269,
273, 315, 330, 346, 361, 392–393
Lyons (gut strings) 133, 193–194
lyre from Ur 276

Mace, Thomas 133–134, 193, 196, 433

machinability 204
mackintosh rubberized cloth 224
madder lake 189, 260, 328
madder root 189, 328
Maffei, Scipione 325–326, 433, 435
Mahillon, Victor and Joseph 16, 35, 421, 433
mahogany 47, 109, 126, 135, 286, 288, 332, 389–390
malleability 204–205
mammoth ivory 184
mandolin 156, 194, 253, 269–270, 315
mandrel 29–30, 34–35, 208–211, 216, 229, 236
manometer 221, 237
Mapes, Inc., piano wire manufacturer 361
maple 47, 109, 124, 245, 291, 344, 390, 392, 404, 413
marbling 126
Marchi, Giovanni Antonio 9, 16, 421
Marciana Manuscript 329
Markert, Frederick 105–106
marking-out 406
Marpurg, Friedrich Wilhelm 317, 326, 433, 435
Marvelseal 59
massicot 271
mastic 54, 115, 128, 329–332
Mathesius, Johann 242
Maugin, J. C. 333–335
Maupertuis, Pierre-Louis 8, 16, 421, 433
Maxwell, James Clerk 261
McCrone, Walter 19–20, 328, 335, 422
McCrone Laboratory 286
meantone 181, 317
meantone fretting 116
meantone temperament 41, 316–318, 320, 325, 361
Medici court 8, 254, 341–342, 388
Meerbach, Johann Christian viii, 294, 375
megilp 54, 128
melodeon 221, 234
melting points of metal 28, 32, 204, 228–229, 278–279
membranes, vibrations in 15
Mendel, Arthur 251
Mendler-Schramm harpsichords 105
mensur 342–343
menuiserie 406
mercury amalgam gilding 70, 117
mercury gilding 119–120
mercury-gold amalgam 117, 120
Merrifield, Mary P. 329, 335, 426, 433
Mersenne, Marin 5–6, 9, 16, 116–117, 134, 195–196, 203, 252, 269, 336,
345–346, 351, 421, 433
Mersenne’s Law 4–5, 14, 130, 134, 203, 269, 359
mesures usuelles 154
metal hardness 36, 182, 204–206, 208, 280, 308–309, 361
metal spinning 35, 213
metal test strips 207
metallurgical analysis 206
metallurgical microscope 206
metallurgy 203
metalworking 29, 208, 272
methanol 52, 54, 58, 280–282
Methocel 125
methyl acrylate ethyl methacrylate copolymer 335
methyl cellosolve 281
methyl cellulose 283
methyl ester α-sulphonate 313
methyl ethyl ketone 280
methyl hydroxycellulose 125
methyl methacrylate copolymer 335
1-methyl-2-pyrrolidinone 72
methylene chloride 25, 54, 72, 267, 275, 284
metric system 60, 134, 153, 156–157, 160, 162, 197, 202–203, 214
metronome 217, 324
Metropolitan Museum of Art 1, 35, 43, 45, 105–106, 141, 207, 219, 286, 356,
361, 405
microcrystalline wax 27, 32, 36, 183
micrometer 199
microphones 256–257
microscope 17, 310, 328, 387–388
Millant, Roger and Max 338, 340, 426
milling 211
Milliput 124
mineral oil 28, 54, 137, 193, 405
mineral spirits 28, 44, 60, 116, 121, 125, 182, 188, 191–192, 242, 258, 264,
266–267, 280, 282–285, 311, 327, 335, 356–357, 399, 402, 420
Minikins 133, 193–194
minium 331
Mish’s silver polish 275
miter box 406–407
modern pitch 251
modulus of elasticity 206, 395
Moebius oils 36, 193
Mohs hardness 204
mold and mildew stains 419
mold making 44, 121, 217
molded key fronts 413
molding, bead-and-reel viii
molding, scraped viii, 413
molding planes 410
molybdenum disulfide 112
Mondino, Angelo 41, 43, 48, 318, 326, 422, 430
monoammonium phosphate 267
monochord viii, 4, 8, 156, 252–253, 269, 319–320
Montal, Claude 326, 358, 386
Montefeltro, Duke of 318
Montucla, Jean 9
morpholine 280
Morse taper 305
mother-of-pearl viii, 112, 185, 220
mounts 138
Mowilith 125
Moxon, Joseph 182–183, 433
Mozart, Leopold 134, 345–346
Mozart, Wolfgang Amadeus 344, 351
MS2A resin 259, 335
Müller, Johann Christian 332, 335
Munsell, Albert 261
Munsell color system 260
muriatic acid 189
Musée de la Musique, Paris 276
Museo Correr, Venice 348
Museo Stradivariano, Cremona viii, 130, 155, 253, 341–344
musette de cour 214
music boxes 139
Music Educators National Conference (MENC) 354
music wire 182, 270, 361
musical intervals 156, 252–254, 315
mylar 29

Nachersberg, Johann Heinrich Ernst 109

National Association of Glue Manufacturers 122
National Bureau (or Institute) of Standards 197
National Fire Protection Association 266, 268
Neapolitan style of keyboard design 288
neck-foot template viii, 342
New York Philharmonic 251
Newton, Isaac 9, 260
newton, unit of force 13, 134, 203
Newton-Laplace Law 12
nickel 23–25, 34, 70–71, 203–205, 221, 405
nickel plating 25
nickel silver 15, 24–26, 30–31, 36, 138, 207, 221, 280, 404
nicking of organ pipes 235–236
nitric acid 23, 25, 37, 49, 54, 57, 120, 187, 207, 245–246, 280, 285
nitrile gloves 137, 266
nitrocellulose 37, 125, 183, 188, 245
nitrocellulose lacquer 188, 277
nominal pitch 251
nonionic detergents 25, 43–44, 280, 311–313
nonionic surfactants 25
North, Francis 7–8, 16, 184, 283, 421, 433
north light 263
Northern Renaissance Instruments 130, 335
Norton sharpening stones 274
numbering of keys 297
Nuremberg wire gauge system 357, 364
nut galls 70
Nye oils 36, 193
nylon 142, 151, 300, 310
nylon strings 270

oak 22, 47, 109, 126–127, 180, 286, 288, 388, 390–391
Oakite 43, 54
oboe 104, 252, 389, 400, 402, 404
O’Brien, Grant 1, 3, 104, 156–157, 177, 255, 269–270, 301, 359–360, 379,
387, 426, 429–430
Occupational Safety and Health Administration 268
ochre 32, 54, 118, 126, 246, 260, 262
Oddy test 138, 221, 277
Odell, Jay Scott 182–183, 203, 269–270, 361, 386, 422, 425
Office de Radiodiffusion-Télévision Français 256
oil 17, 50–51, 54, 57–58, 115, 119, 185, 243, 259–260, 286, 309, 327, 332
oil, grease, and tar stains 420
oil curing of leather 190
oil gilding 119
oil varnish 263, 327
oiling woodwind instruments 137
oils 16–17, 36, 190, 260
oilstones 274
open system (temperament) 318
Optisil 144
Orasol dyes 260
orchil 329
organ 136, 221–225, 227, 233–235
Organ, R. M. 276–277
organ bellows 190, 192, 222–225
organ pipe metal 207, 222–223, 228, 230–231, 233, 236, 238
organ pipes 5, 14, 135, 154, 189, 207, 210, 221–222, 225–231, 233, 235–238,
268, 270
organ restoration 221
organ voicing tools 229
organ wind pressure 221–223, 227
organ wind supply 222
organ windchest 225
Orléans, France 9
ormolu viii, 25, 219
orpharion 6
orpiment 188
Orvus WA 44, 312–313
Ostwald, Wilhelm 261
Otto, Jacob Augustus 338, 340
overspun strings 110, 238, 344, 346–348, 359, 362, 373–377
over-varnishing 115
oxalic acid 25, 184, 243, 285, 311–312, 417, 420
Ozanam, Jacques 9
ozone 191, 245, 310, 355

Pacific Silver Cloth 139, 276

padding woodwind instruments 397, 401–402
Paganini, Niccolò viii, 264–265, 286, 348
Paintings Conservation Department, Metropolitan Museum of Art 45
pallet valves 225, 227, 234
pallet-and-slider windchest 221
pandora (bandora) 6, 346
pantometer 253
paper 21–22, 120–121, 125, 177, 186–187, 241–243, 284–285, 290, 330, 357
paraffinic hydrocarbons 280
Paraloid 125, 335
parchment 112, 119–121, 123, 137, 177, 224–225, 243–244
parchment glue 123, 244
pardessus de viole 336
Paris, France 9, 55, 114, 154, 160, 275, 287, 369–370, 387, 389, 400, 433
pastiglia 44
patent gold leaf 119
patina 29, 31, 36–37, 245, 293, 388
patination 23, 72, 121, 245–246
peg dope 247
peg shapers 247, 415
pegbox 19, 70, 116, 246–247, 270, 342
pegbox inscriptions 343
peg-hole reamers 307
peg-hole reamers 307
pegs viii, 2, 138, 201, 246–247, 391, 415
Peluzzi, Euro 9, 16, 421, 435
pencil 187, 241–242
pepsin 284–285
percussion instruments 137, 141, 244
perfect pitch 252
Perinet valve 33
pernambuco wood 396
Persian berries 189
Persico, Gaetano 333
perspiration 23, 37, 136, 220, 228, 275
Peruffo, Mimmo 133–134, 346–347, 351, 426
Peter Cooper glue grades 122
Peterson strobe tuner 201
petrolatum 36, 57
petroleum jelly 32, 119, 193, 238, 277, 399, 403
pH meter 22, 43, 241–242, 283, 312, 356, 419
phase 14
Philosophical Magazine 171
phosphoric acid 183, 276
phosphorus 182
piano moving 139–140
piano regulating tools viii, 248
piano technician’s tools 248–249
piano wire 108, 205, 361
pianoforte viii, 106, 109, 141, 185, 212, 244, 248–249, 251, 360–361, 390,
413
piccolo 403
pickling 31, 33, 206, 280
pied d’ roi 345
pigments 17–18, 21, 58, 70, 119, 121, 126, 188–189, 241, 246, 259–260, 262,
271, 314, 328–329
pine 51–52, 56, 115, 245, 387, 392
pine colophony 271, 286, 327
pinning cracks in woodwind instruments 404
Pio, Stefano 19
Pirastro 131, 134, 348
pistons 26, 33–34, 137, 211
Pistoy Basses (gut strings) 133, 194
pitch 5–6, 11, 35, 42, 55, 133, 136, 138, 142, 150, 154, 156, 193–195, 198–
201, 203, 222, 227, 235–238, 251–253, 269, 288, 324, 345, 347, 396
plain-sawn wood 407
planes 273, 275, 407–408
planishing metal 211–212
plaster of Paris 44, 117, 210, 218, 271
plates, vibrations in 5, 15
platinum 15, 49, 203–204
Playford, John 346, 351
Plexiglas 12, 138, 244
Plextol 125
pliers 211
plum wood 109, 332
plywood 36, 138, 290, 407
pokeweed berries 243
polarized light microscopy 18
polarizing microscopy 18, 327
polishing 114, 116, 212, 245, 355
Pollens, Stewart 3, 16–17, 19–20, 106–107, 113, 184, 187, 253–255, 270,
301, 319, 326, 332, 335, 340, 351, 361, 387–388, 396, 423, 427, 429–
430, 435
Pollio, Marcus Vitruvius 4, 16, 254, 420, 433
polycyclohexanone resins 335
polyester resin 314
polyethylene 138–139
polyethylene glycol 396
polymethyl methacrylate 138
Polyolefin heat-shrink tubing 138
polyurethane 258
polyurethane resin 125
polyvinyl acetate 121, 125
polyvinyl acetate resins 124
polyvinyl alcohol 266
poplar 70, 126, 387–388, 391–392
poppy seed oil 327
potassium aluminum sulfate 189
potassium carbonate 54–56, 133, 189
potassium dichromate 56, 245
potassium ferricyanide 52, 55, 58, 180
potassium hydroxide 50, 54–55, 57, 133, 311
potassium lactate 191–192
potassium nitrate 30, 51, 56, 58, 120, 276
potassium oleate soap 191
potassium perborate 285, 420
potassium permanganate 51, 55, 246
powder-post beetle 58
powder-proofing of alcohol 283
Powell, Verne Q. 275, 404, 406
power tools 416
Praetorius, Michael 251–252, 270, 346, 352, 433
Preservation Pencil 244, 289
prestant 237
Prest-O-Lite torch 32, 277
principal 237
2-propanol 52, 192, 282, 335
proportion 43, 153–154, 156, 203, 252–254, 338, 345
proportional divider 156, 233, 253
protease 284
protractor 198, 201
Pujol, Emilio 131–132
Purex 312
purfling 70, 315, 389–391
Purkinje effect 262
purse pads, woodwind instruments 190, 402
pyridine 280, 285, 420
pyrolysis-gas chromatography/mass spectroscopy (Py-GC/MS) 286
Pyroxylin glue 125
Pythagoras 4, 252
Pythagorean comma 181, 317
Pythagorean scaling 39, 253, 268, 359–361
Pythagorean temperament 41, 116, 315–318

qin 392
Quantz, Johann Joachim 325–326, 405, 431, 433, 435
quarter-comma meantone temperament 316, 318
quarter-sawn wood 395, 407
quartz halogen lamps 180
quercitron 329
quicking 120
quilling harpsichords viii, 150–151, 298–299

rabbet planes 410

rabbitskin glue 117–118, 120, 123–124
radiocarbon dating 46–47
rag paper 241–242
raising metal 210, 212
Raman microscopy 17, 286
Raman spectroscopy 327
rapeseed oil 405
rasps 414
Rayleigh, Lord (John William Strutt) 10, 16
Raymond Russell Collection, University of Edinburgh 104
reamers 30, 212, 246–247, 249, 292, 305, 307–308, 415
Réaumur temperature scale 308
Record, tool maker 409
recorders 389
recording, sound 256
red brass 23, 203, 360
red lead 271, 331
red ochre 286
red rot 24, 191–192
reed pipes 231
Rees, Abraham 157–158, 171, 177
refractive index 189, 259, 261, 329, 335
refractometer 259
regal 221–222, 234
Regalrez 116, 125, 258–259, 295, 335
Regazzi, Roberto 9, 16, 255, 421, 433
registers 104, 150–152, 234, 300, 387
regraduation 6, 348
regular system (temperament) 318
regulation, keyboard instruments 110, 151, 234, 247, 249–250
Reilly, Edward R. 325, 405
Reinert caliper 200
Renaissance pitch 251
Renaissance wax 192
repoussé 208
rescaling 288
resins 17, 115, 125, 258–259, 290, 328
resonance 14
resonant frequency 6, 14, 153, 235
retouching 189, 258–259, 263, 294, 328
revarnishing 115
reverse osmosis 283, 356
Rhodes, John J. K. 269, 271, 361, 387, 428, 431
rhodium 70–71
Rhoplex 125
Riccati, Giordano 347
rice paste 124, 188, 192
rifflers 212
rift-sawn wood 407
Ripin, Edwin M. 19–20, 270, 361, 423, 429
Ritchie, Stanley 325
Roberson’s orange 189
Rochelle salt 21, 27, 56, 126, 221
Rockwell hardness 204, 309
rollers 226, 234
rolling mill 212, 232, 300, 360, 400
Roman gut string makers 134
Rome 70, 154, 159, 371
Rood, Ogden 260
Rose, John 346
Rose, Malcolm 240, 359, 387
Rosenbaum collection viii
rosewood viii, 107, 126, 135, 332, 389, 391, 412
rosin 32, 44, 51, 56, 110, 178, 193, 210, 259, 264–265, 271, 277
Rossing, Thomas D. 11, 16, 421
rotary valve 33–34, 211
rottenstone 23, 25, 56, 185, 355
Roubo, André-Jacques 114, 116, 334–335, 406, 420, 434
rouge 25, 31, 53, 56, 119, 212, 275–276
Royal Society 9, 171, 323
rubbed-joint 417
rubber 24, 36, 50, 52–53, 58, 132, 139, 275, 355
rubber bands 138
Rubens, Peter Paul 330
Ruckers, family of harpsichord makers 3, 105, 128, 157, 255, 270, 361, 363,
379, 387
Ruffini firm of gut string makers 348
rust 110, 135, 139, 182–183, 283, 287, 297, 312
rust stains 420

Sacconi, Simone F. 338, 340, 427, 431

Sachs, Curt 3, 105–107
sackbutt 120
saddle soap 191
safety equipment 22, 266
safflower 329
saffron 329, 332
sal ammoniac (ammonium chloride) 32, 56, 230, 277
saltpeter (potassium nitrate) 30, 51, 56, 58, 119, 246, 275
salt-spoon keys, woodwind instrument 208, 402
sandalwood 329
sandarac 115, 243, 258, 330–332
sandpaper 10, 33, 117, 251, 273, 293, 297, 401, 403–404, 418
Santagiuliana, Giacinto 332
sassafras wood 387
Saunders, Frederick A. 11
Sauveur, Joseph 5, 318, 326, 434
Savart, Félix 5
saw blades 213, 274, 309, 408
sawing 212, 407
saxophone 35, 400
scaling 194, 222, 235, 268–270, 291, 346, 361
scanning electron microscopy 17, 206, 327
Scarampella, Giuseppe 334, 341, 389
Schantz, Johann 112
Scharf-Fix skiving machine 223, 298, 402
Scheibler, Johann Heinrich 324, 326
Schiedmayer, Johann Lorenz 109, 432
Schiffermuller, Ignaz 260
schisma 181, 318
Schmidt, Johann 376
Schmied, Johann Georg viii, 18
Schmieden, Melchior 330
Schreger lines 184
Schudi, Burkat 367
Schweben 318
Schweingruber Obersaxen chronology 48
scouring rush 117
scrapers 213, 274, 278, 412
scratches 31, 36, 114, 116, 212, 258, 335
screw extractor 399
screw gauges 200
screw pitch gauge 214
screws 198, 200, 214–215, 404
screw-thread measurement 199
scroll saw 408
sealing wax 152, 271, 401, 404
season cracking 24
Seccotine 56, 125
sector viii, 156, 253–254
seedlac 260, 271, 329, 332
Segerman, Ephraim 130, 132
Sellas, Matteo 134
Senovilla, Miguel Angel viii, 131
sepia 243
Sepp Leaf Products 123, 244
Serafin, Santo 19, 348
1704 varnish 260, 333, 350
shaded yew 392
sharpening 271, 274, 409–410, 412
sharpening stones 274
sharpening tools 271
shears 213
sheepskin 111–112, 119, 223
sheet metal 29, 35, 208, 210, 212–213
Shell 282
shell gold 120
shellac 43, 49, 57, 70, 114–115, 119–120, 126, 185, 188, 258, 271–272, 283,
285–286, 314, 328–329, 331–332, 334, 396, 403–404
Shelleng, John C. 11
Shelley, Percy Bysshe 130
Shellsol solvents 264, 282, 284
shimming 112, 293
Shipman, Margaret 337, 340, 428
Shortridge, John 269, 271, 361, 387, 427, 431
shrinkage of wood 6, 224, 227, 290, 393–394, 396, 400
Sibire, Abbé Sebastien-André 9, 16, 345
silica gel 59, 177, 247
silica gel, preconditioned 177
silicon carbide 31, 272, 274
silicone casting rubber 217
silicone dental impression material 144
silicone oils and greases 36, 193
silicone rubber 217
silicones 136, 258
silk 131, 149, 310, 312–313
silk strings 148
silver 15, 17, 30, 32–33, 54–55, 70–71, 118–120, 138–139, 149, 188, 203–
205, 207–208, 210, 212, 262, 267, 275–277, 280, 344, 346–348, 405
silver cleaning 275
silver dip 276
silver nitrate 37
silver plating 30, 275–276, 344, 347
silver point 242
Silver Shield 267
silver solder 28, 31–32, 35, 278–279
silver-gilt 120
sinking metal 212
skiving leather 223–224, 298, 402
slides 33, 36, 138–139, 193, 251, 389, 392
sling psychrometer 257
slunk 243
small-hole gauges 200
smalt 271
Smithsonian Institution 192
smooth-hammering 30
snarling iron 30
soap 43, 50, 54, 56, 128, 190–191, 247, 264, 266, 280, 285, 310–312
Socher, Johann 187
soda process, paper manufacture 241
sodium bicarbonate 49, 56, 183, 267
sodium bisulfate 33, 54, 56, 311
sodium bisulfite 242
sodium borohydride 21–22, 242
sodium carbonate 43, 49, 52, 54, 56, 58, 71, 183, 189, 311–312, 420
sodium carboxymethylcellulose 125
sodium chloride 276
sodium dithionite 21–22
sodium hexametaphosphate 27, 50, 311, 356
sodium hydroxide 22, 50, 54, 56, 58, 71, 126, 183, 275–276, 286, 311–312
sodium hyposulfite 23, 53
sodium metasilicate 312
sodium percarbonate 21, 285
sodium sulfite 22
sodium thiosulfate 22, 49, 242, 275
solder 17, 31–33, 228–230
solder, hard 31–33, 279
solder, 60/40 277
solder, soft 30–31, 33, 276
soldering 31–33, 206, 229–230, 267, 276–277, 280, 401, 404
soldering, hard 32–35, 278, 280
soldering, soft 32, 34, 277–279
soldering iron 32–33, 116, 229, 277
Soler, Padre Antonio viii, 319–320, 326, 434–435
Soluvar 119–121, 258, 295
solvent action 281
solvent cleaning 280
solvents 280–281
Solvol Autosol 27
Sonneck, Oscar 106
Sorb-It 177
sorbus 387, 391
sound intensity 13
sound intensity level (SIL) 13
sound power 13
sound power level (SWL) 13
sound pressure level (SPL) 13
sound wave 10–11, 13
soundboard 15, 107, 109, 112, 124, 135, 142–144, 220, 244, 273, 289–290,
292–293, 295, 387–388, 392, 416
soundboard barring 289, 291–293, 395
soundboard painting 294
soundboard ribs 135, 289, 292–293, 343, 395, 416
soundpost 6–7, 138, 140, 337–338
south German fortepiano 362
specific gravity 207, 230–231, 285, 390
spectrophotometer 262–263
spectrum analyzer 201
speed of sound 4–5, 9–14, 153, 201
spelter 31–32, 208, 279
spike lavender, oil of 327
spinet viii, 106, 152–153, 156, 382
spirit-oil varnish 327
spit cleaning 283
spokeshave 407, 411
Spon’s Workshop Receipts 24–26, 44–45, 70, 116, 125, 134, 180, 185, 189,
192, 216, 399
spotted metal organ pipes 230–231
spring-chest, organ 225
spruce 47, 124, 245, 273, 292, 387–388, 392, 396
square piano 110, 112, 115, 187, 288
squares 33, 201, 234
squares, organ mechanism 234
squirrel-hair mop 119
Staatliche Instrumenten Sammlung, Berlin 105
stain removal 285, 419
Stainer, Jacob 350, 393
staining wood 285
stains 21–22, 184, 241, 285, 312
standing wave 14
Stanley, tool maker 409–410
starch pastes 125
Staub, Wolfgang 105
steam, use of in disassembly 289
steam-bending of wood 389
steel 12, 22, 110, 132, 135, 147, 182–183, 203, 205, 207, 210, 270, 308–309,
327, 346, 383, 385
steel oxidation colors 272, 308, 400
steel rule 198
steel violin E strings 270
steel wool 70, 297
Steininger, Johann Gottlieb 112
Steinway 140, 154, 157, 292
stereo microscope 48
sterling silver 32, 70, 275–276
stickers 225–226, 233–234
Stockholm tar 232
Stoddard solvent 44, 264, 280, 282
storage of musical instruments 24, 135, 137
Stradivari, Antonio viii, 5–6, 8–9, 17, 20, 70, 105, 130, 155–156, 186–187,
253–255, 260, 263, 270, 329, 333, 341–344, 348–351, 387, 389, 391
Stradivari, Giacomo 260
Stradivari “grand-pattern” violin 156
Stradivari’s golden period 9
Stradivari’s long-pattern violin 8–9, 254
Stradivari’s varnish 286
Strasbourg turpentine 327
Streicher, Andreas 109
Streicher, Johann Baptist 378
Streicher, Nannette 377
Streicher, Nannette & Sohn 157
Streicher family of piano makers 358, 366
stress and strain 205
stress-cracking 26, 221
string density 203
string diameters proportional to pitch 346
string gauges 40, 107, 109, 131, 134, 147, 198, 210, 334, 344, 348, 382
string tension 5–6, 40, 109, 111, 134, 138–139, 203, 238, 287–289, 292, 341,
345–346, 348, 359
stringed instruments 60, 106, 114, 136, 138, 140, 244, 253, 269, 325, 362,
388
stringing 109, 130, 152, 193, 384
strings 8, 14, 110, 117, 133–135, 142, 144, 148, 193–196, 269–270, 343,
361–362, 365, 369
strings, overspun 359
stuffed pads, woodwind instruments 402
Sudan black 115, 328
sulfate process, paper manufacture 241
sulfite process, paper manufacture 241
sulfur 22, 24, 53–55, 137–139, 141, 206, 217, 276, 355
sulfur dioxide 24, 191, 275
sulfuric acid 25, 30, 33, 49–52, 54, 57–58, 71, 115, 190, 276, 280
Super Glue 125
suspended action, organ 233
swaging 214
swedging pliers 211, 401
sympathetic vibration 6–7, 318
Synchro Search 47–48
synthetic clock and watch greases 36
synthetic resins 335
syntonic comma 41, 181, 316–319, 322

tacking iron 258

tailpiece 138, 140, 179, 348
tailpieces, Baroque 344
Talbot manuscript 382
talc 36, 57, 193, 286, 402
tallow 57, 190, 210, 230, 399
tamiso 117
tangents 39–43
tannic acid 189, 242, 246, 286
tanning, leather 112, 190
tap tone 5
tap water 22, 25, 37, 183, 190, 229, 275–276, 283
taper pin reamers 305
tapers 201, 246, 305, 415
taps and dies 214
tar 51–52, 55, 246, 285
Tartini, Giuseppe 13, 16, 344–346, 348, 421, 434
Taskin, Pascal 301, 370, 387
tawed leather 112, 190–191, 223, 225, 402
tawing of leather 190
Taylor, Brook 203
tea and coffee stains 420
Teas graph 281
Teflon 266
telescoping gauges 200
Tellervo 47
temper colors of steel 205, 272, 308
temperament 41, 72, 171, 237, 268, 315–316, 318, 322–323, 434
temperature 10, 25, 56, 60, 105, 135–136, 223, 238, 244, 257, 269, 278, 287,
300–301, 308, 393–395
tempering steel 205, 272, 308–309
tenon saw 407
tenons, woodwind 137, 139, 399, 401, 405
tensile pickup 205, 268, 360
tensile strength 24, 133, 156, 182, 194, 204, 268–269, 360
térébenthine 331
tetraethylammonium borohydride 22
tetramethylammonium borohydride 22
textile cleaning 310
textiles 22, 43–44, 52, 111, 121, 125, 135, 178, 285, 288, 290, 310, 312
theorbo 134, 270, 325
Thermo Lignum Warmair process 25
thermostatically controlled heating elements 289
thin-layer chromatography 17, 19
thiourea 276
third-comma temperament 316
Thomas, William R. 11, 16, 113, 196, 238, 255, 260, 269, 271, 323, 361, 387,
428, 431–432
Thomée, H. 358, 364, 387, 434
Thon, Christian Friedrich Gottlieb 109
thread-chasing 214
threading 214
three-octave span 157
thumper rail 233
Thymol 60, 123
Tielke, Joachim 361, 393
Tillwich oils 36, 193
timpani 244
tin 23, 28, 50, 52, 54, 203–205, 207, 228–231, 277, 314
tin chloride 189
Tingry, Pierre François 242–243, 332, 335
tinning 230, 277, 314, 344
Tinuvin 292 335
titanium 204
Titebond 125, 291
Todini, Michele 301, 371
Toledo 253
toluene 125, 188, 264, 266, 280–281, 284, 335
tool steel 309
Töpfer, Johann Gottlob 230, 238, 270, 428, 431, 434
torch, jeweler’s 32–33, 205, 208, 246, 277–279
tortoiseshell 107, 314, 408
Tourte, François-Xavier 17, 275
trackers 226, 233–234
transporting musical instruments 135
transposing keyboard 251, 287
Trasuntino, Vito 347
traveling wave 14
1,1,1-trichloroethane 44, 52, 191–192, 312, 420
triethanolamine 280, 357
Trinity College, Cambridge 6
tripoli 53, 112, 212
trisodium phosphate 32, 280
Triton X-100 25
Tromlitz, Johann George 396, 399, 402, 405–406, 431, 434
trumming 216
trumpet viii, 8, 15, 18, 20, 34–35, 120, 209–210
trumpet, Baroque 206
trumpet marine 8
trundles 234
Tucson format 47–48
tuning 15, 109, 132, 136, 222, 226–227, 237, 251, 269, 315, 320, 324, 396,
403, 433
tuning cones 236, 238
tuning fork 250, 324
tuning pin coils 138
tuning pins 40, 109–110, 142, 182–183, 291, 415
tuning slides 28, 137
turmeric 329
turning viii, 108, 215, 397, 412
turpentine 57, 115, 127–128, 185, 264, 271, 280, 282, 327, 330–331
turpentine varnish 327
Tylose MH300 125
Tyndall, John 10
typewriter 186

Ulmia workbenches 416

ultraviolet lamps 180
umber 126
Underwriters Laboratories 268
universal bevel 198, 201
urea resin glue 125
urushi 187, 189
US Department of the Interior Director’s Order no. 210 107
US Forestry Service 388
UV (ultraviolet) 245, 328
UV barrier filter 328
UV exciter filter 328
UV fluorescence 17–18, 180, 231, 243, 290, 297, 328, 388
UV fluorescence microscopy 327
UV fluorescence photography 328
UV lamp 40
UV light 21, 25, 36–37, 180, 188, 191, 310, 328, 335, 420

V.M. & P. naphtha 280–281

Vallotti, Francesco Antonio 325
Vallotti temperament 325
valve, brass instrument 28, 33–34
valves 15, 28, 33, 36, 137, 139, 193, 251
Van Cortlandt House Museum viii
Van Dyck, Anthony 330
Vandyke brown 126–127
vapor phase inhibitor 21, 139, 327
varnish viii, 16–18, 37, 44, 51, 54, 57, 114–116, 118, 121, 128, 179, 185,
189, 217, 258, 260, 262–264, 277, 280, 284, 290, 327–332
varnish analysis 327
varnish for lutes 329–330
varnish for lutes and violins 330–331
varnish for violins 331
varnish, Venetian violin 330
varnish of Venice 334
varnished wood 258
Vatican 9
vellum 243–244
velvet 313
veneer viii, 109, 116, 139, 291, 296, 387, 389–392
veneer-press 293
Venetian red 126
Venice 154, 253, 339, 433
Venice turpentine 271, 332, 334
Venice-Catlins 133, 193–194
verdigris 271
verditer 52, 271
vermilion 57, 127, 262, 271, 314
vernier caliper 197, 357
vernier protractor 246
vibration microscope 10
vibrational modes 15
Vicentino, Nicola 253, 318, 326, 434
Victoria & Albert Museum 184
Victory wax 192
Vienna 57, 111, 141, 296, 376–378
Vienna valve 33
Vienna wire gauge system 357, 364
Viennese fortepianos 219, 288–289, 301
Viennese piano action 111
vihuela 253
vinegar 70, 246, 267, 311
viola 6–7, 59, 105, 179, 253–254, 269, 338, 342–343, 348
viola bastarda 346
viola d’amore 7
viola da gamba 6–7, 20, 106, 116, 140, 194, 253, 269, 315, 330, 335–336,
361
viola necks 341
violin 1, 4–8, 11, 15, 17, 46–48, 126, 134, 140, 155, 194, 200, 245, 247, 253–
255, 263–264, 269–270, 283, 315, 325, 337, 339, 343–345, 348, 361
violin adjustment 337
violin bow frogs 314
violin bow hair 138, 178
violin bridge 337
violin forms 8–9, 155–156, 254, 318, 343–344
violin neck patterns 270, 341
violin necks 341, 343
violin sizes 354
violin strings 344–345
violin varnish 259–260, 263, 327, 330–334
violin / viola strings, modern 355
violinmakers’ clamps 416
violinmaker’s planes 411
violoncello 6, 8, 105, 178, 337, 342–344, 348
violoncello bridge, Belgian style 337
violoncello bridge, French style 337
violoncello fingerboard 343
violoncello neck 343
violoncello neck patterns 342
violone 337
virginals 6–7, 152
Vitruvius 4, 16, 255, 422, 434
voicing 150, 222, 226, 232, 235–237, 249, 251, 300, 396
voicing, piano 111, 150, 247, 251, 433
voicing jack 237
Volta, Alessandro 70
von Gontershausen, H. Welcker 111, 113
von Vega, Georg 171, 177
von Wolf, Nathanael Matthaeus 171
Vuillaume, Jean-Baptiste 5, 245, 345
vulcanite 355
Vulpex 44, 191, 264, 311–312

walnut 47, 109, 127, 135, 286, 288, 387–388, 390, 392
walnut hulls 246, 286, 294
walnut oil 327, 405
Watco 289
water 12, 207, 283, 310, 356
water gilding 117, 119–120
watercolor 119, 127, 178, 188, 244, 260, 294
waterlogged wood 396
watermarks 241
water-of-Ayr stone 355
waterstones 274
Watin, Jean-Felix 114, 116, 331, 335, 434
wavelength 10–11, 13–14
wax 16, 25, 32, 35, 55, 128, 152, 192–193, 217, 219, 258, 271, 285, 401
wax seals 271
wax stains 420
Weisshaar, Hans 337, 340
weld 329
well temperament 318
Werckmeister, Andreas viii, 238, 321–322, 326
Werckmeister III viii, 321–322
West System epoxy 124
Westphalian process 182
whalebone 107, 314–315
wheat paste 124, 357
whispering room 6, 14
White, William Braid viii, 248, 251, 324, 326
whiting 25, 32, 44–45, 50–52, 55, 58, 112, 117, 184, 230, 247
Williams, Jane 130
Wills, Andrew 361
wind instruments 15, 36, 137–138, 141, 200
wire 198, 204, 268, 277, 288, 314, 344, 347, 360
wire drawing 182
wire gauges 357
wire gauges, half 359
Wittkower, Rudolf 254–255, 435
wolf fifth 39, 316–319
wolf tone 201, 318
Wolfenden, Samuel 268, 271, 428, 431
wood 15, 22, 44, 46, 49, 55, 57–58, 114, 116, 121, 125, 127–128, 135, 138,
177–178, 201, 231, 245, 247, 258, 260, 285, 330, 387, 389
wood, dating of 46
wood, kiln drying 393
wood, macro structure 392
wood, micro structure 393
wood, physical properties 393
wood, seasoning of 393
wood identification 388
wood planes 409
wood screws 414
wood shrinkage 40
wood pulp paper 241
wood-grain painting 126
Wood’s lamp 328
woodwind instruments 16, 23, 108, 136–137, 190, 214, 251, 314, 355, 388–
391, 393, 396
woodwind instruments, oiling 405
woodworking 406
woodworm viii, 25, 58–59, 135, 287, 291, 301, 396
wool 44, 275, 310, 312–313, 404
Woolite 311
work hardening of metal 205
wove paper 241
Wraight, Denzil 271, 387
Wratten 2A and 2E filters 180, 328
Wratten 18A filter 180, 328
Wratten 87 and 87C filters 180
wrestplank 121, 291, 295, 370, 387
Wurlitzer, Rembert 332
X-ray diffraction 24, 206, 327
X-ray fluorescence 18, 120, 206–207, 275
X-ray fluorescence spectroscopy 17, 231
Xylamon 396
xylene 25, 36–37, 51, 72, 120–121, 125, 264, 266, 277, 280–281, 284, 335,
357, 399, 420
xylophone 14

yellow brass 268, 360

yellowing of ivory 184
Young, Thomas 171, 260, 323
Young’s modulus 14

Zarlino, Gioseffo 253, 255, 318, 326, 434

Zecchi 124
Zenti, Girolamo 106
zinc 24, 26, 32, 50–51, 53, 56–58, 203–207, 221, 228, 277–278
zinc chloride 277
Zip-Strip 284
zirconium/aldehyde leather 191