Journal of Apicultural Research

ISSN: 0021-8839 (Print) 2078-6913 (Online) Journal homepage: http://www.tandfonline.com/loi/tjar20

Factors Influencing the Fermentation of Honey in

Mead Production

Keith H. Steinkraus & Roger A. Morse

To cite this article: Keith H. Steinkraus & Roger A. Morse (1966) Factors Influencing the
Fermentation of Honey in Mead Production, Journal of Apicultural Research, 5:1, 17-26, DOI:
10.1080/00218839.1966.11100127

To link to this article: http://dx.doi.org/10.1080/00218839.1966.11100127

Published online: 24 Mar 2015.

Submit your article to this journal

Article views: 4

View related articles

Citing articles: 2 View citing articles

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=tjar20

Download by: [Weill Cornell Medical College] Date: 26 July 2016, At: 16:15
 17
Tournai of Apicultural Research 5(1): 17- 26 (1966)
FACTORS INFLUENCING THE FERMENTATION OF
HONEY IN MEAD PRODUCTION 1
KEITH H. STEINKRAUS AND ROGER A. MORSE 2

Department of Food Science and Technology, New York State Agricultural

Experiment Station, Cornell University, Geneva, N.Y. 14456, U.S.A.
Manuscript received for publication 5th October 1965

SUMMARY
Downloaded by [Weill Cornell Medical College] at 16:15 26 July 2016

Attempts were made to produce a good-flavoured mead by more rapid fermentation than
is usual. Several wine yeasts were tested, and additives of two types, I composed mainly of
inorganic salts and citric acid, and II with a wide spectrum of vitamins and small amounts of
organic and inorganic nitrogen compounds.
Honeys varied in their fermentability, in general light honeys being more difficult to
ferment than dark honeys. But addition of growth factors allowed all honeys tested to be
fermented rapidly, yielding characteristic meads which had none of the unpleasant flavour
associated with long (unaided) fermentation.
A recommended procedure is described, which uses additives I + II and can give a dry
or sweet mead (alcohol content 12% or 14-150!,). Fermentation takes about 2 weeks without
agitation. The production of mead champagne and mead sherry is also described.

INTRODUCTION
Mead (honey wine) has been produced and consumed for thousands of years.
There is extensive literature on methods of producing mead. For example Bergeret
and DeCastro (1943) and Bertullo and Etchandy (1943) used formulae containing
tartaric acid and ammonium phosphate to stimulate yeast growth; Warwickshire
(1952) advised the addition of salts, acid, and vitamin B to stimulate the fermenta-
tion; Pique (1930) discussed a sparkling mead; Filipello and Marsh (1934),
Morquin (1949) and Morse (1953) all included salts and acids in their formulae;
Fabian (1935) used fruit juices, salts and acid as additives to stimulate fermentation
and improve flavour; Lees (1943) discussed the influence of nitrogenous sub-
stances in mead on the distilled alcohol. Morse (1961) has prepared a bibliography
on mead.
Fabian (1935) reported that honey drinks lack body and are too sweet. His
experiments were directed toward the elimination of these deficiencies, which he
felt were best corrected by adding fruit juices to contribute acid and growth factors
for the yeast fermentation. Fabian also added potassium tartrate and ammonium
phosphate in very small quantities (in the range of 0·04%) to diluted honey and
produced a bland alcoholic drink in about 6 weeks. In another formula, he added
ammonium chloride, potassium bicarbonate and sodium phosphate in small con-
centrations (0·04- 0·08%). and produced a mead in about 4 weeks.
Filipello and Marsh (1941) studied the effect of various nutritive salts on the
rate of honey fermentations. They found that the addition of 0·2% cream of tartar,
0·1% ammonium phosphate, 0·5% citric acid, 0·025% magnesium chloride, and
' Approved for publication as Journal Paper No. 1456 by the Director of the New York State
Agricultural Experiment Station, Geneva, N.Y.
• Departmental of Entomology, Cornell University, Ithaca, N.Y. 14850
 18

0·025% calcium chloride, permitted fermentation to 12- 13% alcohol content in

about 25 days.
Crowther (1960) found it essential to add nitrogen to stimulate the fermenta-
tion; he used diammonium phosphate. Crowther also prepared sherry mead with
good to excellent quality by using a flor yeast in the primary fermentation, and
agitating the fermented wine for 3 weeks to produce the typical sherry flavour.
Maugenet (1964) reviewed the problems involved in the fermentation of honey.
Honey contains mainly fermentable sugars; it is deficient in nitrogen, minerals,
and growth factors that stimulate yeast growth and fermentation. Maugenet recom-
mended the addition of 250 mg. fl. of diammonium phosphate. As a source of
potassium and buffering capacity, he added 250 mg./1. of potassium bitartrate. To
Downloaded by [Weill Cornell Medical College] at 16:15 26 July 2016

adjust the acidity, he added 1·875 g./1. of tartaric acid or 1·750 g./1. of citric acid.
Although citric acid yielded a better flavour, he recommended tartaric, since it was
less easily metabolized by undesirable lactic acid bacteria. Among Maugenet's
strongest recommendations was the use of 25-50 mg. fl. of sulphurous anhydride
or 50- 100 mg. fl. of potassium metabisulphite; these were added to prevent growth
of lactic bacteria. Maugenet recommended meads with a residual sugar content of
5-10%. At a temperature of 20- 25°C.* his fermentations required 20-30 days.
In addition, numerous articles have described recipes for flavoured (spiced1
meads, or metheglins.
Recent patents on the production of honey drinks (Adams & Niesen, 1963a,
1963b) summarized the mead situation quite well: 'Mead, however, has not gained
acceptance as a beverage. Although sweet tasting due to the presence in it of
unfermented sugars, mead has had a generally undesirable flavor. The poor
flavor of mead beverages has been attributed in important part to the excessive
period of time necessary for fermentation of the honey to become complete.' In
these patents, primary importance is made of keeping the yeast cells in suspension
by a slow agitation. This shortens fermentation time and reduces autolysis of the
cells, which these authors indicate is responsible for the undesirable flavours. It
should be noted that these patents cover the fermentation of a honey base to
6 - 9% alcohol, which is then distilled.
It is not unusual for honey fermentations to require 6 months or more if
factors are not added to facilitate growth and fermentation by the yeast, especially
in a light honey base. Efforts in our laboratories have centred upon shortening the
length of the fermentation time by adding factors required for a rapid fermentation,
to produce a clean-flavoured mead, with stability during storage.
The studies reported in this paper include other procedures, besides agitation,
directed toward the production of a desirably flavoured mead, produced by a
rapid fermentation.

EXPERIMENTAL METHODS
The honeys used in these experiments were largely produced in New York
State. Varieties included clover, buckwheat, fall flower (mostly goldenrod), and
* Fahrenheit equivalents of temperature used in this study are:
13• = 55•F. 32" = 90"F. 37• = 98"F. 63•= 145"F.
18" = 65"F. 35" = 95•F. 60• = t40•F. 65" = 150"F.
 19

wild thyme. A major effort was made to use clover as a representative of the
light and buckwheat as a representative of the dark honeys.
Yeasts used included Steinberg 618 and yeast 605 from the Professor Otto
Rahn collection in the Division of Bacteriology, Cornell University, Ithaca, New
York; yeast 223 from the collection in the Department of Food Science and
Technology, Cornell University, Geneva, New York; and the Maury yeast from
England. Yeast 223 is noted for its general desirability for both the primary and
secondary fermentations of grape juices.
The general procedure was to dilute the honey to approximately 25% solids
as measured by a Brix hydrometer or hand refractometer. Various additives in
the form of salts, nitrogen sources and vitamins stimulatory to yeast growth and
Downloaded by [Weill Cornell Medical College] at 16:15 26 July 2016

fermentation were added. The diluted honey base was inoculated with 0·5 - 10%
by volume of yeast, the strain depending upon the type of experiment. Fermenta-
tion temperature was generally l8°C. However, temperatures as high as 37°C.
and as low as 13°C. were tried in certain tests. All experiments, except the flor
sherry and bottle champagne, were carried out anaerobically, using bubblers to
release the carbon dioxide produced.
Light clover honey consists principally of sugars and water; it lacks nitrogen,
phosphate, and vitamins stimulatory to yeast growth and fermentation. Fermenta-
tions in the absence of sufficient nitrogen and phosphate are prolonged; insuf-
ficiency of vitamins is less serious, especially if sufficient nitrogen is present, as
most yeasts have the ability to synthesize their vitamin requirements. However,
if the vitamins are supplied to the yeast, the fermentation time is decreased, and
it was therefore decided to supply growth stimulants in two forms. The first was
through use of Formula I, containing principally inorganic salts and citric acid:
ammonium sulphate 1·0 g.
potassium phosphate (KJ>O,) 0·5 g.
magnesium chloride 0·2 g.
sodium hydrogen sulphate 0·05 g.
citric acid 5·0 g.

6·75 g.

This quantity was added to 1 litre of diluted honey base.

Formula II contained a wide spectrum of vitamins and small amounts of both
organic and inorganic nitrogen stimulatory to yeast growth:
biotin 0·05 g.
pyridoxine 1·0 g.
meso-inositol 7·5 g.
calcium pantothenate 10·0 g.
thiamine 20.0 g.
peptone (Roche) 100·0 g.
ammonium sulphate 861·45 g.
1 000·00 g.
 20

It was added at 0·25 g. per litre of honey base.

The effects on the rate of fermentation of Formulae I and II were studied
separately and in combination. The effects of individual components of the
mixtures were also studied. The effect of pH was investigated by adjusting the
pH level from 2·9 to 6·0.
To produce a sparkling mead, clover honey mead fermented to an initial
alcohol content about 10% by volume was refermented in the bottle after the
addition of 2% sugar (in the form of clover honey) and fresh yeast inoculum
(yeast 223 and yeast 618).
To produce a flor type sherry, clover honey mead with an alcohol content
approximately 13% was inoculated with a flor sherry yeast, Saccharomyces
Downloaded by [Weill Cornell Medical College] at 16:15 26 July 2016

oviformis (strain 31), and refermented.

In each experiment the final mead was clarified and subjected to stability and
organoleptic tests. Storage tests were made on some of the meads, by holding the
bottles for up to 2 years and watching for cloudiness and I or precipitates.
Based upon the various investigations, a recommended procedure was
developed for the production of mead with a flavour resembling a good sauterne,
which can be produced from a clover honey base by a rapid fermentation.

RESULTS AND DISCUSSION

It was found that honeys vary considerably in their fermentability. Light
honey, as represented by clover, was more difficult to ferment and required more
additives in the form of vitamins, minerals and nitrogen than did dark honey, as
represented by buckwheat or goldenrod. However, not all batches of clover honey
reacted in the same way; some permitted a very slow fermentation, others allowed
practically no fermentation without the addition of supplementary growth factors.
With the addition of growth factors, all honeys tested could be fermented
rapidly, yielding meads characteristic of the types of honey from which they were
derived, but lacking certain unpleasant flavours believed to result from typically
long fermentations carried out in the absence of sufficient yeast growth factors.
Clover honey yielded a mild-flavoured, sauterne-like mead which was accept-
able to the majority of tasters participating in organoleptic tests; there was only a
hint of clover honey flavour. Buckwheat honey yielded a dark-coloured, rather
heavy-flavoured mead which was less acceptable than clover mead to the average
taster. Other honeys yielded meads reflecting their particular flavour characteristics.
The addition of both Formulae I and II produced the most rapid fermentation
of a clover honey base, using yeast 618 (Fig. 1). Using a 25% solids base and a
0·4% yeast inoculum, the alcohol content reached 12% in less than 2 weeks.
Formula I alone, containing mineral salts and a source of ammonia nitrogen and
phosphate, permitted a somewhat slower but still rapid fermentation in the
absence of Formula II. Adding to clover honey only (NH,)2 SO, and K 3PO, (the
two most important yeast foods in Formula I) resulted in about the same rate of
fermentation as the entire formula. Formula II was much less effective in pro-
moting the fermentation of clover honey in the absence of Formula I. In the
purest clover honey, which permitted practically no fermentation, Formula II by
itself stimulated fermentation slightly, but in other batches of clover honey
 21

YEAST 618, 80°F /X

X F"ORMULA I I. n __x. . .
12
0 F"ORMULA I
D IIJ P04 I. (NH4 )z S04 ..x--- --
,,,"'""
,.
10
,. /

,.X
..J
0 'f/
J: I
0 I
~ 8 I
I

< I
I
1- I
z
w
I
6 I
u I
I
Downloaded by [Weill Cornell Medical College] at 16:15 26 July 2016

a: I
w
0..
4
I
,.
I
I

I
I
I
X
2

2 4 6 8 10 12 14 16 18
DAYS
FIG. 1. Effect of various additives on clover honey fermentation

Formula II alone failed to show stimulation of fermentation rate. The control (no
additives) in Fig. 1 showed no fermentation at 18 days and reached an alcohol
content of only 6% in 54 days.
Buckwheat honey, without additives, fermented much more rapidly than clover
honey (Fig. 2). This suggests that dark honeys contain more natural yeast nutrients

4 6 8 10 12 14 16 18
DAYS
FIG. 2. Effect of growth factors on fermentation of buckwheat honey
 22

than light honeys do. With buckwheat honey, Formulae I and II together again
produced the most rapid fermentation, but Formula II by itself resulted in some
stimulation. Formula I by itself gave a higher rate of fermentation of buckwheat
honey than Formula II alone. Ammonium sulphate added to buckwheat honey
slightly stimulated the rate of fermentation. The other components of Formula I
added separately showed no positive effect upon the rate of fermentation.
The pH of the honey, which was different in different batches of honey,
influenced the rate of fermentation considerably. After the addition of Formula I
containing citric acid, the pH was often below 3·0. A clover honey fermentation
at pH 2·9 progressed much more slowly than one at pH 3·7 (Fig. 3). It was found
Downloaded by [Weill Cornell Medical College] at 16:15 26 July 2016

12 YEAST 618, 80°F

10
.J
0
I
0 8
u
.J
<{

~ .6
w
u
oc
:t 4

0o"L----2~~-4~---6~---8~--~~~o--~1~2--~14~--~~6~--~

DAYS
FIG. 3. Comparison of clover honey fermentation rates at two pH levels

(Fig. 4) that a pH of 3·7- 4·6 was desirable for honey fermentation; a pH of 3·7
was high enough to permit rapid fermentation but still low enough to inhibit
growth of undesirable bacteria.
The effect of size of inoculum upon the rate of fermentation is demonstrated
in Fig. 5. A 10% (by volume) inoculum gave a more rapid initial fermentation
than 0·4%. but the advantage was not generally maintained after 10-12 days. At
a low temperature (e.g. 13°C.), a 10% inoculum produced a more rapid fermenta-
tion than a 1% inoculum.
The yeasts varied in their ability to ferment honey. Of those tested, strain
618 (Steinberg) gave consistently rapid fermentations and yielded a mead with
relatively high alcohol content, with a fine flavour and better than average stability
on storage. The Maury yeast produced a very rapid fermentation, but there was
a haze problem, the mead being very difficult to clarify and stabilize. Yeast 605
appeared to be very close to strain 618 as far as desirability for honey fermentation
is concerned. Yeast 223, which is excellent for grape juice fermentations and
 23

YEAST 618 p pH 4.6

" pH 3.7
12 ""
'" pH5.0

, ,, pH3.0

,,
10
...J
0 ,
,
,,
J:
0
u
...J 8
,,
<
1-
z
w 6
u
a:
Downloaded by [Weill Cornell Medical College] at 16:15 26 July 2016

w
a..
4

10
DAYS
FIG. 4. Effect of initial pH on buckwheat honey
fermentation

12

10
...J
0
J:
0 8
~
<
1-
z 6
w
u
a:
w 4
Q.

00 2 4 6 8 10 12 14
DAYS
FIG. 5. Comparison of clover honey fermentation rates using 10% and
0·4% inocula

champagnes, did not generally yield as high an alcohol content as 618 in honey
fermentations.
The effect of temperature on the rate of fermentation of clover honey contain-
ing both Formulae I and II by yeast 618 is shown in Fig. 6; the rate of fermenta-
 24

tion rises progressively with temperature up to 35°C.; at 37° the rate is slightly
slower. However, at 32° and above, fermentation stopped before the sugar was
entirely fermented. At 24 o there was a sufficiently rapid fermentation, with
complete conversion of the sugar into alcohol in approximately 2! weeks, using a
yeast (618) inoculum at 0·4% by volume. At 27° (Fig. 1) the fermentation was
complete in 2 weeks.

YEAST 618 ,u--o-' 75°F

12 .,.,."' p
-~ ,It"- 68°F
...~.:- JJ- - - - :.R:--'0
Downloaded by [Weill Cornell Medical College] at 16:15 26 July 2016

10 ,... .... .o...-~-....-::.""' ........ "' 90°f

..J
~~~~-~-~~~-~-~-~95°:---~-
0
J:
0 8 ///..... ,.~"'
;t/ 98°F
~ / .
<{
I

~
/
6 _Jf
w
u
a: , ---
---
~ 4 v
/
/
/
/
I
I
I
I

-/

6 8 10 12 14 16
DAYS
FIG. 6. Effect of temperature on clover honey fermentation

Thus it can be concluded that with added growth factors, a selected yeast,
proper pH, and a temperature of 24- 27°, even light-coloured honey such as clover
can be fermented to an alcohol content of 12-13% (by volume) in approximately
2 weeks, without agitation.
Although more rapid fermentations can be achieved at higher temperatures,
as shown above, there is a belief among some wine makers that the use of lower
temperatures and a longer fermentation gives a better flavour in the resulting wine.
It has not yet been shown that this is so with clover meads, but when speed of
fermentation was less essential, we used a fermentation temperature of l8°C.
Based upon our laboratory studies, a pilot-plant procedure was developed for
the production of clover honey mead in approximately 40-gallon lots in 55-gallon
oak barrels [1 gallon = 3·79 1.]. The process was designed to yield a dry, light,
almost colourless mead, devoid of harsh or bitter flavour, and with good stability
in the bottle. The procedure consisted of the following steps :
I. Clover honey was diluted to 21% solids with water. Crystallized honey was
heated to 60- 65°C. to facilitate solution.
2. The following yeast nutrients were added (per gallon of diluted honey):
 25

citric acid 18·9 g.

ammonium sulphate 4·65 g.
potassium phosphate (K 3PO,) 1·9 g.
magnesium chloride 0·7 g.
peptone 0·1 g.
sodium hydrogen sulphate 0·2 g.
thiamine 20·0 mg.
calcium pantothenate 10·0 mg.
inositol 7·5 mg.
pyridoxine 1·0 mg.
biotin 0·05 mg.
Downloaded by [Weill Cornell Medical College] at 16:15 26 July 2016

These were approximately equivalent to Formulae I + II.

3. The pH was adjusted to 3·7- 4·0 with sodium hydroxide or hydrochloric
acid.
4. When cooled to about 27o, the 40 gallons were placed in the 55-gallon
barrel, inoculated with 0·5% by volume of actively growing culture of yeast
618, and sealed with a bubbler.
5. The fermenting mead was incubated at 18oC.
6. The mead was allowed to age in the barrel for about 6 months.
7. It was then decanted and filtered through Celite 503 or a similar filter aid.
8. Total acidity was adjusted to 0·6% with citric or tartaric acid.
9. The mead was pasteurized at 63° for 5 minutes and bottled while hot.
The above process yielded a dry mead with an alcohol content of about 12%
by volume. Starting with an initial solid content of 25%, the mead had an alcohol
content in the range 14- 15%.
For a mead champagne, it was preferable to use a clover honey base diluted
to about 18-19% solids at the start of fermentation. Formulae I and II were
added to the diluted honey. This yielded a mead (cuvee) with an alcohol content
about 10% by volume, which was better for champagne manufacture than one
with higher alcohol content. The steps in making the champagne were as follows :
Sucrose (2% by weight) was added to the cuvee with the following yeast
nutrients (per gallon):
peptone 0·1 g.
thiamine 20.0 mg.
calcium pantothenate 10·0 mg.
inositol 7·5 mg.
ammonium sulphate 0·86mg.
pyridoxine 1·0 mg.
biotin 0·05mg.
These ingredients (similar to Formula II) were added to ensure sufficiency of yeast
vitamins for the fermentation in the bottle.
The cuvee was inoculated with a 7% (by volume) yeast inoculum made of
equal quantities of yeast 223, a champagne yeast, and 618, the yeast adapted to
 26

honey fermentation. The inoculated cuvee was bottled in champagne bottles,

sealed with metal caps, and incubated at 18 o. Within 21 days, the alcohol content
had increased to 11·9% (by volume) and carbon dioxide pressure to 4·6
atmospheres (at room temperature). Standard disgorging procedures were used to
remove the yeast cells. The mead champagne produced from clover honey had a
pleasant flavour and retained its carbonation very well. There was a greater
tendency to form a head than is generally found in grape champagnes.
Clover honey mead was made into a light sherry by refermenting it with a
flor sherry yeast, Saccharomyces oviformis (strain 31). The acetaldehyde content
was raised from 48 to 190 mg./1. in 48 hours by passing the mead through a glass
column filled with ceramic tile pieces inoculated with the yeast. Initial content of
Downloaded by [Weill Cornell Medical College] at 16:15 26 July 2016

the mead was 10·4%. which was increased (fortified) to an alcohol content of 13%
(by volume) before commencing the sherry fermentation. The sherry produced
from mead had a definite flor sherry flavour that many tasters preferred to the
original mead flavour. Because of its low acidity, mead offers an excellent base
for the production of sherry.

REFERENCES
ADAMS, S. L. & NIESEN, G. V. (1963a) Honey beverage and process for making it. U.S. Pat.
3,100,705
- - (l963b) Beverage from honey and process for making it. U.S. Pat. 3,100,706
BERGERET, G. & DECASTRO, J. A. (1943) [Honey wine.] Revta Asoc.lng. agron., Muntev.
15 : 67-70
BERTULLO, W. A. & ETCHANDY, A. M. (1943) [A comparative study of the fermentation of
honey must using wine yeasts and bee yeasts.] Revta-Asoc.lng. agron., Montev.
15 : 71-73
CROWTHER, R. F. (1960) Mead. Rep. hart. Exp. Stn Prod. Lab. Vineland 98 : 101
FABIAN, F. W. (1935) The use of honey in making fermented drinks. Fruit Prod. J. Am. Fd.
Mfr. 14 : 363-366
FILIPELLO, F. & MARSH, G. L. (1934) Experiments with mead. Am. Bee J. 74 : 537-538.
- - (1941) Honey wine. Fruit Prod. J. Am. Fd. Mfr. 41 : 78-79
LEES, P. M. (1943) [The influence of nitrogenous substances on the characteristics of the
distilled alcohol.] Revta Asoc.lng. agron., Montev. 15 : 79-83
MAUGENET, J. (1964) Hydromel. AnnlsAbei//e 7: 165-179
MoRQUIN, M. (1949) Le bon hydrome1 chez soi: sa fabrication raisonnee. Apiculteur
93: 135-147, 165-175, 195-198, 220-221
MoRSE, R. A. (1953) The fermentation of diluted honey. Cornell University, Ithaca: Thesis
- - (1961) Annotated bibliography on honey wine (mead). Cornell University, Ithaca:
unnumbered report
WARWICKSHIRE, H. (1952) Notes on making mead. Br. Bee I. 80 : 634