Royal Anthropological Institute of Great Britain and Ireland

Prehistoric Monuments in the Outer Hebrides, and Their Astronomical Significance
Author(s): Boyle Somerville

Source: The Journal of the Royal Anthropological Institute of Great Britain and Ireland, Vol.
42 (Jan. - Jun., 1912), pp. 23-52
Published by: Royal Anthropological Institute of Great Britain and Ireland
Stable URL: http://www.jstor.org/stable/2843231 .
Accessed: 25/06/2014 03:22
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].
Royal Anthropological Institute of Great Britain and Ireland is collaborating with JSTOR to digitize, preserve
and extend access to The Journal of the Royal Anthropological Institute of Great Britain and Ireland.
http://www.jstor.org
This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

All use subject to JSTOR Terms and Conditions
23
PREHISTORIC MONUMENTS IN THE OUTER HEBRIDES, AND

THEIR ASTRONOMICAL SIGNIFICANCE.
[WITH PLATES VII-IX.]
BY CAPTAIN BOYLE SOMERVILLE, R.N.
DURING the summer of 1909 my duties took me to the neighbourhood of the Outer
Hebrides; and the wintry gales which marked that season, having driven me more
than once into harbour from hydrographic work in the offing, gave me the chance
of visiting some of the prehistoric monuments, of which so many specimens
remain on these bleak coasts.
In this paper I propose to discuss the surveys of these monuments, with
theodolite and chain, that I was then able to make; and I think I shall be able
to show some interesting, results from the azimuths, or orientations to be found
in them; and to add one more stone to the cairn of evidence which is slowly being
accumulated in support of the astronoinical,-or, perhaps I should say, astrological
-intenition of these aiicient remains.
I am assuming, that the struictures in question are the remnants of edifices
set up for the purposes of religion, or burial, or for both. The long avenues of
stones, anyway, could not have forined part of a dwelling house; which pre-
supposes a roofed place.
Careful investigation seems to show that they were, in some cases, places of
worship, with burials made in and round them afterwards,-churches, as it were,
with graveyards and intramural tombs, as we still see to be customary; or else,
in other cases, mausolea, built solely to contain the dead, though with the further
intention, possibly, of some form of ancestor worship.
The object of the orientationist is to show that this cult, whatever else it was,
was connected definitely with the heavenly bodies: sun, moon, and stars.
The monuments with which I deal here are situated in the Hebrides:-in the
north-western part of the Island of Lewis, and in St. Kilda. Those in the first-
named island are fairly well known, but have never adequately been discussed
froin the astronomnicalstandpoint; while those in St. Kilda are, so far as I can
ascertain, new to science, and are here described for the first time.

24 CAPTAINBOYLE SOMERVILLE.-PPehistoric Monurments
I.-,, THE GREAT CIRCLE" OF CALLANISH.

(Gaelic "Tutrsachan Challantsh.")
This imposing group of forty-eight imegaliths stands on the crest of a small
peninsula near the head of East Loch Roag, on the western coast of the Isle of
Lewis. The surrounding view is of a hilly country, wild and treeless, covered with
heather. Though elevated 100 feet above the sea level, the water of the sea loch
is not visible, except irn a few gleams here and there, as it winds inwards to
Callanish, 7 miles from the open sea.
The accompanying map of the locality (Fig. 1) shows its position, as well as
that of other circles with regard to it; and attention is also called to the plan,
which is reduced from a careful survey, plotted originally on the scale of -i- inch to
1 foot.
/ 1~~~~~~~~~~~~~~~~~0
CTREAT BefRNEiRA 10 v
/ 47~~~~~~~~~~~~~~~~~C
-4At
S
itE ~~~~~~Ag
t?*W
._ '5QA A A
9 /
* ft
FIG. 1.-MAP OF CALLANISH AND DISTRICT.
I may remark in passing, that it is by laying down the relative positions and
attitudes of the stones forming a megalithic group, on a sufficiently large and open
scale, such as the above, that alone is it possible to appreciate their orientation.
And it is equally important to present them with the true north always directed

int the Onter Hebrides and their Astronomical Significance. 25
in the same way on the paper; that is, most conveniently, upwards, as in a rnap.
It is difficult, indeed, to grasp the identity of orientation of two or more separated
monumenits,when they are presented, (as they usually are), with the north and
south mneridianlaid down in any and every direction.
The remains consist, as will be seen (Fig. 2), of the following, chief features:-
A. Two long lines of menhirs, runnincg nearly parallel to one another, a

little eastward of true north. (I shall refer to these as "A east"
and " A west," respectively, when describing them.)
B. A short line of menhirs, lying in an east and west direction.
C. A longer line, lying in a north and south direction.
D. A short line, lying in a direction slightly northward of east.
E. A " circle" of thirteen great menihirs.
Besides these, there are a chambered sepulchre; a very large single menhir
(which dominates the whole group) lying within the perimeter of the "circle "
and three single menhirs without it, to the north-east, south-east, and south-west
respectively, standing at no great distance from it.
Before entering into a discussion of these various parts, it may perhaps be
advisable to give some explanation of the method by which astronoiny is brought
to bear upon the arch.eoloogyof megalithic structures.
When a row of standing stonies, or several rows, as at Callanish, have been
set up in straight lines, it seems rational to suppose that it was done with somle
intention, dependent on the direction in which they lie. This becomes even miore
obvious when it is found that in monuments differing widely in geographical
position, the direction of these rows, or aligninents, is towards the same points of
the horizon. Occasionally the eye is directed by the line of stones to a hill-top,
or a cairn on a hill-summit; sometimes, as at Callanish, no object, natural or other-
wise, lies along the linies of sight in either direction.
But merely to direct the eye towards such an object as a distanit cairn does
not seem a very purposeful proceeding; still less is it so when the line leads the
eye neither to a cairn nor to anything particularly conspicuous on the earthly
horizon. The inferenice is that we must look beyond the terrestrial termination
of the line for some celestial body; and if such, it must naturally be at that
moment of its path when it is either rising or setting behind the horizon of the
observer. As this land horizon is always elevated above the true, or sea horizon,
the body must, at the time of observation, be itself elevated by the same amount.
All heavenly bodies are positioned in the sky by astronomers, according to
their right ascension, and declination. These terms correspond, roughly speaking,
to the co-ordinates by which a spot is located on the earth; namely, to longitude
and latitude respectively. Right ascension governs the, time of risilng of a heavenly
body, declination the bearing on which it rises. lf the latitude of the observer,
the bearing of the heavenly body, and its elevation above the true, or sea horizon,

26 CAPTAIN BOYLE SOMERVILLE.-Prehistoric Monuments
be known, it is easy to calculate the declination of the body, which would be seen
on such a bearing.
Now, the declinations, and riaht ascensions of all heavenly bodies are always
slowly and progressively altering; and consequently no star rises on exactly
the same bearing, nor at the same time on any particular day of the year,
as it did on the same day in any previous year. The alteration is small, but is
known with considerable accuracy for all the principal stars:-for no two stars
have precisely the same " precessional movement " and " proper motion," to which
the changes in position are due. It is not difficult, therefore, having this
knowledge, and having found the declination inferred in the bearing and altitude
of an alignment, to state definitely the name of a star, with the appropriate year,
day, and hour, when the alignment pointed to its rising. This is true also of the
Sun and Moon, but a considerably less definite date is afforded by either of them:
in the case of the Sun, as the rate of its change is so small; and in the case of the
Moon, as it is so irregular. Even in the case of stars, though the movement is,
generally speaking, sufficientlv great, aind sufficiently well known to be able to
obtain a date within perhaps ten or twenty years of the truth, accurate observation
of the alignment is necessary. A date can only be allotted within marginis of, say,
one hundred years in somle cases, where the aligning menihiis have become only
slightly shifted through some cause from their original positions; and where
judgment has had, accordingly, to be exercised as to the correct theodolite reading
of the bearing, of the line. It will be obvious also, that the longer the alignment
is,-the greater the distance is from the observation position to the end menhir of
the line, or to the mountain peak, or cairn, seen against the sky, to which it exactly
directs the eye-the more certain, and the more easily observed will be the bearing
and altitude.
It may be objected that among the great multitude of stars it would be easy
to find several that would have had, at various epochs, the declination ascertained
from aniy particular alignment, and that the dates derived from stuch declination
could thus land you into almost any century. The multitude of stars is, no doubt,
great; but bright stars, and still more brioht stars of particular conspicuousness
(owing to their isolated position in the heavens, or colour, etc.), are remarkably few.
And when the period in which we are to look for a date is limited, as for our
purposes it must be, by that of the Christian era at one end, and the probability
of the date of the beginning of the Stone Agrein these islands, (to which period
these megalithic monumnents are allowed on all hands to belong), at the other,
namely, between 3,000 and 4,000 B.c., and the first two or three centuries A.D.,the
chances of error are still mnorereduced.
In the description that follows I have, however, given the names and dates
of ail the brig,ht stars whose declinations are iinferrable from the " star-lines " that
could have been observed aiong them, within the above limits of time, and discussed
the probabilities as regards the dates thus derived.
I shall deal first with the two longer lines of stones trending to the north-

in the Outer Hebrides and their Astronomical Significance. 27
ward, "A east" and "A west," as giving the most definite results; firstly, on
account of their length, which provides a dependable azimuth, and next because
the northern end of each line is terminated by a high aind conspicuous menhir,
4 or 5 feet taller than any of the other stones of the lines, which, therefore, give
a definite object for observation.'
LINE "A EAST."
I shall first deal with the eastern of the two lines, which consists of eight
stones. In order to determine its azimuth, I chose a spot to the southward,
whence all the stones composing it came into alignmient with the high menhir at its
northern end, and there set up my theodolite. There seemed to be the foundations of
a standing stone at this point, though nio other vestige of it remained; but in any
case, the azimuth was exactly obtainable from here, and is N. 90 49' 30" E. This,
with the hill-crest altitude seen along it of 1? 10', produce a star with declination
320 26' 37" N.
LINE "A WEST."
The western line consists of ten stones. The seven northern ones, including
the tall terminal menhir, are, with one exception, still all in line; but the three
nearest the circle are evidently either displaced or else never belonged to the line;
for it is not now possible to get them into alignment with it. The six stones,
however, that do remnainin situ present a line, which, being produced across the
circle, exactly strikes a tall nenhir which stands somiefeet outside it to the south-
westward; and I have little doubt that this important-looking stonie was the
original southern termination of the linle, and still remainis in position; while its
companion for the soutllern end of the eastern line has disappeared.
The azimuth of " A west " from this position is N. 110 10' 0" E., and this, with
the hill-crest altitude seen along it of 10 14' 50", produce a star with declination
of 32? 28'12" N.; a result so close to that derived from the " A east " alignment that
there is no doubt that both directed to the same star, though, as will be seen, the
two alignments are not parallel, differing as they do by 1" in bearing. The reason
for this is as follows:-The rising path of the star with the above declination
would not be vertically upwards, as seen at Callanish, but on the arc of a circle,
with the pole of the heavens as centre; and thus would have an apparent move-
ment, when first rising, from left to right; so that, as it increased its altitude, so
1 All calculations in connection with the sun in the following pages are worked as for an
observation of the sun's centre.
The declinations of stars, at various epochs, are taken from the tables in the first volume
of Handbuch der .Mathematischenund TechnischenChronoloqie,by Professor F. K. Ginzel,
published in 1906.

28 CAPTAINBOYLESOMERVILLE.-Prehistoric
Aotuwments
it increased its angular distance from the pole. The observer, lookingr along
"A east," which has the lesser altitude and smallor azimuth, was the first to
see the star rising above that part of the hill-crest to wlhich it directed. It was
then still hidden from the observer lookingf along the other linie of stones, but, a
few moments later, the star einerged above the higher part of the hill-crest to
which "A west" was directed, a little to the right of the "A east" spot, and
completed the observation. The same star was, in fact, seeni along each line, but
at slightly differing positions of its path up the heavens.
The bright stars and dates to which this declination (320 27' N.) may refer
are as follows
Capella in 1800 B.C.

Castor in 650 B.C.
Arcturus in 320 B.C.
For reasons to be discussed later, connected with the time of rising of this
star, and also with the date to be derived from line D, it seems -most likely that
the star Capella. at its appropriate date, was the object of observation; but,
astronomically speaking, each of these stars, and dates, has equal probabilities. I
shall according,ly, for convenience, refer to these lines in fulture as the" Capella"
lines; remarking, in passing, that alionments for this star have beenlfound at various
dates in several other prehistoric monuments in Great Britain.
This matter of the date does niot exhaust the initerest of these two lines, for
if each alignnmentbe produced sufficiently far to the southward, it will then be seen
that the centre of the sepulchre and of its circular tumiulus lies exactly on the
ilmiddleline of the avenue formed by them; and besides this, the line of direction
of the stones forming the dividing walls of the two burial vaults of the sepulchre is
parallel to the avenue also. There can thus be little doubt that the double line of
nienhirs and the sepulchre are connected one with another, have the same
astrological intention, anid most likely were erected simuiltaneously. These
connections, I may add, have only become apparent after the survey of the whole
monument had been plotted on a sufficiently large scale, and is an instance of the
scientific value of this method of examination. The fact that the centre of the
tumulus lies on the central line of the aveniue cannot otherwise be appreciated,
nor is the parallelism of the long lines of stones formina the avenue with those of
the sepulchre divisions apparent by casual observation.
The last point of interest concerning the " Capella" lirnes to which I would
direct attention is the connection between them and the outlying stone standing
at about 15 feet to the north-east of the circle. This is evidently a stone of
importance. It is of regular form, with its sides made flat (or particularly chosen
because they were naturally smooth and parallel), and has, when viewed in plaD,
one end square, but the other wedge shaped; the point of the wedge being directed
to the south-westward. If its present attitude in the grouild is that in which

it was originally " planted," it will be seen that it is directed exactly to the tall
menhir standing 10 feet outside, and to the south-westward of the circle, which
I have supposed to be the termiination of the western " Capella" line; and this
affords strong probability that the two stones are in connection one with another.
Standing at this south-western stone, the azimuth of the north-eastern, conmbined
with the altitude of the horizon seen beyond it (at a somewhat mnarkeddip in the
distant hills), produces a declination of 28? 10' 25" N.
When first reaching this result, I was somewhat puzzled as to its meaning,
for I had fully expected that the direction given by the line would turn out to be
for the solstitial sunrise, which would entail a declination of about 240 N.
Obviously, therefore, it could not be a sun-line; nor does this declination belong to
any probablestar in " prehistoric " times except, perhaps, Pollux, in about 1200 B.C.
The onily heavenly body to which otherwise it could refer is the inoon, and in
view of the fact that I hiave obtained a similar declination, along, undoubted,
sig,htlines, in several other monuments in other parts, I venture to put forward the
following suggestion. The moon has a cycle of nineteen years (roughly), within
which it changes its tropical declination from 280 to 180, and back again to 280.
The rising of the full moon, when it is at its northern tropical declination, only
occurs at or near the date of the winter solstice, and if the azimuth of this event,
(full moon rise), were marked when the moon had reached its greatest possible
declination of 28' N., this full moon rise would indicate the beginnings of
periods of nineteen years, and also be closely connectedcwith a definite solar event.
viz., the winter solstice. This period of nineteen years is, as I need not point
out, the Metonic Cycle of ancient Greece, though I do not know how, nlor from
what lunar event, it was there measured. The possibility of the agreement in this
leingth of a calendrical cycle between the two countries is at least of interest, though,
if I am right in my conjectures, it was observed in our islands in 1800 B.C., which
antedates by nearly 1400 years its establishment in Greece, where it is said to have
been initiated in 432 B.C. (see quotation on p. 31).
LINE B.
This line, composed of four menhirs, is constructed in a similar fashion to the
two Capella-pointing lines, namely, of flat slabs placed in the ground in the
direction of the line, and terminated by a taller menhir of nearly square
section.
This end stone, and the next to it, are still upright; the otlhertwo between them
and the circle, lean slightly to the northward, as may be seen in the plhotograph;
which causes the bases, as plotted in the grouind plan, to throw the line a little
out in that direction, but it will be realized that if these two stones were stanrding
upright all four would be exactly in a line. The azimuth of this line, combined
with the altitude of the horizon along it, procducea declination of 0? 35' 19" N.:
that is to say, for sunset at the day of equinox. Of that, there can be no doubt
whatever.

30 CAPTAINBOYLE SOMERVILLE.-Prehistoric Monuments
LINE C.
This line is composed, at present, of five menhirs; but, by the gaps in it,
evidently comprised, originally, at least three more.
From the southernmost of the stones still remainiing in place, the linle runs
true north, within 1?, up to the great slab, about 14 feet hiigh, which stands
within the " circle," anid as all three of the lines already discussed terminate in
larger and distinctive pillar stones than those forming the rest of the linie, it seems
more likely that the eye was intended to be directed in this line, also, towards
the important stone, rather than from it to the southward, where there is no trace
of such a menhir.
I should add that near the southern end of line C there is a collection of
enormous boulders, so enormous that their heaping together can scarcely be other
than natural, though there is a sort of symmetry in their disposition, which renders
an opposite view permissible. This group of boulders lies exactly on the
continuation to the southward of line C. and, standing on the flat surface their
top affords, one can suppose oneself to be at the observing position for an azimuth
of true north, along the line to the great central menhir; but this is the inerest
conjecture, and has no great importance in itself, astronomically speaking. I am
not prepared to say how this alignment to the true north was obtained by the
ancienlts. Polaris (a Ursa, Minoris), the so-called " Pole Star " of our times (though
it is not yet actually polar), had a declination of 681' N. at the date of the
monument given by the " Capella" lines (1800 B.C.); that is to say, it circled rounld
the nortlh pole of the heavens at a distance from it of 21-0. The ancients may
have had sufficient kniowledge to divide equally the distance between the eastern
and western points of the circumpolar path of Polaris, in order to obtain its polar
centre; or, facing the other way, a north and south line may have been laid out, by
findincgthe meridian position of the sun (that is, when it bore due south), through
the shortest length of the shadow thrown by the great menhir; but, whatever
nmethodwas employed, the fact remains that the line of stones under discussion
does present a practically true north and south bearing.
LINE D.
The stones of this line are somewhat disarranged from their original regularity,
though not seriously, and it is not difficult to obtain a mean alignment, which shall
include all the four stones very fairly. The azimuth of this alignment, combined
with the altitude of the horizon seen along it, produce a declination of 6? 43' N.
which refers to the following stars aniddates, viz.
Pleiades rising in 1750 B.C.
Spica in 1270 B.C.
a Arietis in 1130 B.C.
Aldebaran in 800 B.C.

FIG. 2.
E V
a:~~~~~J
0110
_ IU
)
H 4 # V!) 1
.F.; ADL; oFs all7)~~L

zy
Ma; ou I~-
t X g(16)>s446to
t (l5 X(5)/~~~~~~~
ffi

FIG. 2.
0~~~~~~~~~~~~~?
0 W-
0~~~~
D bD~~~~~~~~
E c~~~~~~~~~~~~~~~
i1 ~/ 0;
I~~~~~~~. (17)-_____ 19
F-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~l
' IO
/ z ng* <
,.;~~~~-
QD;^ X1)
I 4 1 1(5i 1'
| ; M x (6)1}-2^ _ _~~~~~~~~~~~~L
0.1 . <it.\\s 4;h ~ ~

4) 4i
J~~~~~~~4v~
XL5/(3)
.I
(10)
'2w EPU
4--- ~
( 5 / 4 e >~~~ior8 *
LINE B. ~~* ~ ~ (O
Equinocriai (23) (2.2) (ZI) (20) 4(t c e,

WU -t I
g MEN~~~~(7
Equlnoctt(tt~~~~~~~~~~~~~~~9
-. (z6 A
v1@_.E:.t_ t _ _ [ [ ! | .%t~~~~~(27

Zs
/, ()
- X - 4
A ~ I / to2
(50)~~~~~0
A, VW~~.,%,
I ~ ~~ ~~~~~~-I sloul_<*re,1Xto
r5
S / ,,jb4si.iri / ,;,-eS -,
-~~~~ ~io
of 6) ~ -- - (6

- - -(z6)
SCALE
10 (HORIZONTAL,& VERTICAL)
feet 0 1o 20 30 40 50 100
EHEE
X
I3LEEC
Wll C
GAL
?- SEPULCHRALTJM
wiT
0 -Turn
{aaGrcelic:
ShewingPlaniof the,5ronet
VWas 0F faces of
ecCh abre-asWof i1
all betm6ontWe-same
-f~ |j It

SCALE
40 50 (HORIZONTAL,& VERTICAL) 100 150feet
s DES
HEBRI
4
I3LE OF LEWIS
CGALLANIS H
ZzlI% SEPULC[HRALTJMULUS
WTHlsTIGLNT6R (Gl.E'
z Ciaelic:-t Tur5ach-anChallanish.')
1909
Shewing Plan of the, 5rones at prezent 6round-level, and

Views o faces of Jthosbe forming AlNgnmenth,
CiCh abare-as- of iks plotte-4 pos0ton in plan:
all betnmon tWe, same verticazl aid horizontal scal,.
;; 7 \ I 7/70- d5n" haW nUMbCliell C ,wV

- cnvr CO-f rWferePvce,.

in the Outer Hebrides antd their Astronomical Significance. 31
The astronomical probabilities of each of these four stars is equal in degree;

but it should be remarked that a Arietis is not at all conspicuous in the heavens,
either by position or brilliancy; though it must not be forgotten either that this
star, or rather the constellation to which it belongs, had -great importance in the
astronomy of eastern countries in early days; marking as it did the sun's entry at
springtime on a new year-the termi,indeed, persists to our own day; the sun still
being said to be at " the first point of Aries," when it is at the spring equinox,
though the actual coincidence of a Arietis with the equinox has not occurred
since the fourth century B.C.
Taking the Pleiades date of 1750 B.C., it would make the building of this line
practically contemporaneous with the two long Capella lines, which, fromnthe
similarity in size and description of their stones, is in the highest degree probable,
whatever their date. I would even go farther, and say that the Pleiades star-date
forms a valuable check on the accuracy of the Capella star-date for the founding
of the monument. But the Spica date of 1250 B.C. and the Aldebaran date of
800 B.C. are just as astronomically probable, though neither is supported by the
date given by another line, in the same way as that of the Pleiades.
I would now call attention to the fact that line D, the " Pleiades line "-if I
may so name it-and the "Equinoctial line'" (line B), if produced towards one
another, meet at a point on the central line of the tuinulus covering the sepulchre,
at its southern edge which is exactly equidistant, namely, in each case, 69 feet
6 inches from the terminal stones of lines B and D (see Fig. 2). A single position
is thus afforded from which an observation can be made along both these lines: a
position, too, which is symmetrical with the ttimulus and the two Capella lines,
lying as it does on the central line of the avenue formed by them. This connection
between the equinox and the rising of the Pleiades is of particular interest in the
light of the followino well-known quotation from Diodorus Sicutlus concerning the
Hyperboreans." The first part of the quotation, bearing on the Metonic Cycle,
has already been cited (vide sypra).1
QUOTATION.
(Diodorus Siculus, ii, 47, ed. Didot, p. 116.)
"It is also said that in this Island (i.e., that of the Hyperboreans) the moon
appears very niear to the earth; that certain eminences of a terrestrial form are
plainly seen upon it; that the god (Apollo) visits the Island once in a course of
nineteen years, in which period the stars complete their revolutions, and for this
reason the Greeks distinguish the Cycle of nineteen years by the name of the
Great Year. During the season of his appearance the god plays upon the harp,
1 It is, I understand, by no means certain that the " Island " of the "iyperboreans"
refers, definitely, to Britain; but the quotation clearly points to some race inhabiting a country
northward of Greece, where solar and stellar observations, of a religious character, were made
in early days.

32 Monuments
CAPTAIN]BOYLESOMERVILLE.-Prehistoric
and dances every night from the Vernal Equinox to the rising of the Pleiades,
pleased with his own successes."
At the date and latitude to whclih the above quotation refers (which I take to
be 400 B.C.,and 380 00' N.-the latitude of either Sicily or Athens) this period
would ainounit to 46 days, thus continuinT from the Equiinox up to " May-day " of
the "agricultural year"; buit at Callanish, in 1750 B.C. (the date obtained from
the " Pleiades " line), this remarkable group of stars rose " heliacally," that is to say,
so as to be observablein the morning twilight (or at abouLt40 minutltes before
sunrise) on April 10th. There would thus be, at Callanish, at the obtained date,
twenty nights after the sun's passage of the equinox, for the "dancing," before
the rising of the Pleiades became visible in the dawning of the day.
The change in the time of rising of the Pleiades on any particular day from
year to year, at the epoch in question, amounted to about five minutes in one
hundred years, and the alteration in bearing at rising only 1? in the same time, so
that the "heliacal" rising of this conspicuous star-cluster would follow the
equinox by twenty days for a considerable period of time.'
THE GREAT MENHIR.

Anotlher fact observable in connection with the tumuluis and its sepulchre is
that the great menhir-standing within the " circle," though not at its centre-is
erected due west from the centre of the tumulus: that is to say, that at sunset on
the day of equinox, the shadow of the great stone wouLldlie exactly across the
grave, and along the entranlcealley that leads to it from the eastward. It is also
to be noted that this important stone, which stands about 14 feet high, forms one
of the edginig stones to the sepulellral tumulus, and gives the general effect of the
" gravestone " of modern days. It is also placed in the ground with its long side
in the same direction as the chamber stones of the tomb itself, and therefore (vide
ante) with the " Capella " lines "A east" and "A west."2 These facts seem to
establish a connection between these lines, the sepulchre and the menhir, though
it is, of course, quite possible that the grave was made subsequentlyto the erection
of the stone, that advantage was taken of its position, and (possibly) of its sanctity,
to make a grave in such a position that it might form the grave-stone. But
probability seems to point the other way.
THE SEPULCHRE.
As to the sepulchre itself, it is of the shape shown in the plan. A description
of it, when first excavated, will be found in the Proc. Scot. Ant. Soc., vol. iii. At
1 It may be of passing interei,t to note that another Celtic nation, viz., the Welsh, regard
the Pleiades as " auspicious "; for their name for the group, " y saith seren sirio]," signifies
"the seven cheering (or solacing) stars."
2 Mr. A. L. Lewis points out that from a position near the centre of the eastern face of
the great menhir the summer soistitial sunrise would be observable over the outlying menhir
to the north-east of the circle ; and it is quite possible that this menhir fulfilled this
" aligning " purpose as well as that described on page 29 (vide cnte).

in the Outer Hebrides and their Astronontical Significance. 33
present it forms a sort of well in the highest part of the tumulus, with sides of dry
masonry about 3 feet high, its earth floor being on the natural ground level, or
nearly so. It is divided into two portions by a pair of endwise planted flagstones,
each about 3 feet high, and the entrance, which is at the eastern side, is closed by
three somewhat similar stones. Two loose slabs lying near seem to have once been
connected with it in some way, perhaps as roofing, or as flooring.
E. THE " CIRCLE."
The great circle, which is, at first sight, the most striking feature of the whole
group, being composed of thirteen huge stone slabs, 10 to 12 feet in height, carries
to-day no actual evidence (astronomical) of its date, or purpose. It is, in the first
place, not, strictly speaking, a " circle " at all, but a sort of irregular oval, with its
eastern side flattened, while the stones composing it are themselves not equidistant,
fronmone another.
It is hard to account for this shape, unless perhaps there were difficulties in
providing foundations for such large and weighty stones at equal distances apart,
and on an exactly circular line-a line which could so easily have been traced oin
the ground by using a string, attached to a pointed stone as " compasses."
It seems obvious, in any case, that the circle was not erected at the samne
time as the rest of the monument. Not only are the stones of which it is
composed of a different size and character from those of the alignments, but also
it is, as mentioned above, quite unsymmetrically placed both to the sepulchre, the
long double line of stones, and all the short cross-wise lines. Even the Great
Menhir, which is of appearance and size somewhat similar to the stones forming
the " circle," does not stand in any obvious geometric relation to it, and is quite
considerably distant from any spot that might be named the " centre "-though this
spot does not actually exist for so irregular a figure.
If symmetry of attitude is any 'evidence of a connected design, it seems that
the " circle " is not an integral part of the remainder of the Callanish monument,
and the only conjecture that I can offer as to its purpose, (and that, it must be
confessed, not a very probable one), is that it was deliberately placed with the
intention of effacing the worship, or at least the visible connection of the dead with
the sun, moon, and stars, which seems to have been the purpose of the alignments,
the Great Menhir, and the sepulchre. Yet this seems to be a cumbrous method of
obliteration, unless some superior and different worship invested the circle itself;
and what that worship was still remains to be decided.
The last feature that remains to be noticed, though it has nothing to say to
orientation, is the relics of heaps and circles of small stones on the eastern side of
the circle. These seem to be ancient-their present shapes and positions may be
seen on the ground plan--but they now merely exist as a sort of pattern, lying on
the surface of the soil.
There are also the remains of a ring of small stones, a few inches high, filling-
VOL. XLII. D

Vernal Summer Autumnal Winter

Equinox. Solstice. Equinox. Solstice.
h f 16 m h mly J,u e s W&V 0

I2I g 1, 1 v8 .1 zi No
2, <A 2
>W
s, , II X / / /1 II II / .~~~~~~~~I
I
_ -- A- - iI I ll '/ '/ 1i __________
5~ ~~~~~~ __
g_~~~~~~~J
z- 13 (11111111 )+I'11111111
7 thIrough
1outthesame
A G p at same LatitIude
o
D
If
1IAI
t IiI1 I11
1I
IJoW III I 1II I1111l r -0
Tl\11/111
111
121
z1+1 11I 1Ift11f
IA FI7____\_
; NF tW 3 T IlilSl IIIII,Y 1/111111f1111 IlNlf5t IF

{^~~ ~~~EL0AE
IJ 1 llI ( i l $ I1 1 r E b;n
n t0 Et0 44110;/T 11i!ilLillitltXlI

IG. 3.-Diagramof the conditions
-0 ol1111111 1 of DA LIGH,TIIGT N DAKNS
S- thouhotPMya
t
Jant 2.1 z I[IA
S # g~TGIE DAE thoghu th sam peid at sam Latitude

in the Outer-Hebrides and their Astronomical Significance. 35
the spaces between the tall stones of the " circle " and defining its outline, but now
only to be traced in the western-part of the ring.
The enquiry so far refers us to certain stars and their dates, as obtained from
their declinations. But there is another point to be considered, namely, the time
of year at which the inferred stars thus rose, at the epochs to which they belong.
Without entering deeply into astronomical facts, it should be stated that while
stars rise on the same bearing (practically), on every day throughout the year, their
time of rising alters, for they appear about four minutes earlier on the horizon of
the observer on each successive occasion. It will thus be realized that the rising
of any particular star can only actually be seen during a certain part of the year,
namely, for the period that it rises between sunset and sunrise ; or rather
between the end of evening twilight and the beginning of morning twilight.
The accoinpanying diagram (Fig. 3) represents in pictorial form the varying
conditions of daylight, twilight, and darkness, throughout a year at the latitude of
Callanish. The diagonal lines running across the diagram show, at their points of
intersectioii with the others, the actual time of rising of the star whose name they
bear, at different times of the year; namely, of Capella in 1800 B.C., of the Pleiades
in 1750 B.C., and of Spica in 1270 B.C.
I will deal first with Capella in 1800 B.C. It will be seen in the diagram that at
that date it rose just before sunrise at the winter solstice. It rose in darkness thence-
forward, earlier and earlier each night, until the middle of May: wheni it rose about
one hour after sunset. Subsequently to that it rose in daylight until the following
winter solstice; but it was visible every night, at some part of its course.
This connection between Capella-rise and the winter solstitial sunrise is
importarit; for the observation of the rising of the full moon every nineteen years
(discussed at p. 29) always takes place at this (the Saturnalian) time of the year;
and the " heliacal" rising of Capella would thus have " warned " both these great
solar and lunar events. Hence, perhaps, the greater length, and the important
terminal nenhirs given to these two lines of stones.
An interesting fact in connection with Capella, as seen frorm Callanish in
1800 B.C., must now be stated.
In this year, at that latitude, Capella performed its path round the pole at a
distance from it of 570 33'. At Callanish, the north pole of the heavens is elevated
580 12' above the horizon; so that Capella, when at the lowest point of its path,
was some 39' above the horizon, and thus never set below nor rose above it: it was
"circumpolar," as it is termed., But the skyline of the hills towards which the
long lines of stones is directed, is elevated 1? 14' 50", so that Capella was obscured
from sight by the hills, when at the lower part of its course, just as much as if it
had sunk below the sea horizon; and thus its "rising" actually couild have been
observed, only it was above a hill horizon, instead of a sea horizon.
By about 1700 B.C., a hundred years after the date that I have assigned to the
erection of these lines of stones, the declination of Capella would have altered
D 2

36 CAPTAIN Monuments
BOYLESOMERVILLE.-Preh'istor'ic
sufficiently to cause this apparent "rising" to cease; and Capella would then
always be in sight at night, circling round the pole. Alignments for Capella have
been found in several prehistoric monuments, and it is conjectured that its circum-
polar movement may perhaps have been utilized for calculating the time at night,
just as the attitude of the constellation of the Great Bear in its swing round the
Pole Star is used as a " shepherd's clock " at the present day.
As regards the other star alignment, that to the eastward, it may have been
the Pleiades, in 1750 B.C. as stated above, or, with equal astronomical probability,
the star Spica (a Virginis) in 1270 B.C. If the Pleiades, this group rose in bright
twilight-too bright for observation-at 5.37 a.m., or about twenty minutes before
the sun, on March 21st, but was visible as a rising body from April 10th onwards,
until the middle of August, when it rose just as the sun set. For the rest of the
year the rising of the Pleiades took place in daylight, and was consequently
invisible.
If the star, on the other hand, was Spica in 1270 B.C., it rose in that year at
one and a half hours before the sun at the autumnal equinox, thus "warning"
that event, and as the opposite radiating arm of the " cross " to this we are now
considering pointed to the sunset of the same date, there seems some possibility
that Spica, (with its appropriate date), was the heavenly body for which this align-
ment was laid out; or, in the course of the centuries, it may have succeeded the
Pleiades as a date-fixer for the " dancing" of Apollo. Spica was visible as a rising
body during the winter months, from September 21st to the miiddleof February,
when it rose at about one hour after sunset. During the rest of the year its rising
took place in daylight.
SUMMARY.
Before summarizing the particular inferences to be derived from each feature

of Tursachan Challanish, I wish to point out a general inference, which should set
at rest any doubts as to the reality of the factors of astronomy and orientation in
this ancient monument.
For what is Orientation, or Direction ? What is the mieaning of North, South,
East, or West? How did we derive these fixed points to start with ?
Not from any local or national origin; for pure Direction is entirely
independent of locality; it is, in fact, derived solely from the movements of the
"heavens." It is only by reference to the positions of the stars, sun, or mooln,that
Azimuth, or true Direction, exists: there is no other meaning in the term.
So that when, in the monument just described, we find two lines of megaliths
laid out on absolute "cardinal points," viz., West and South, the setting and
nooning points respectively, we realize that it can only have been accomplished by
some reference to the heavenly bodies: (magnetic compasses, and their divergences
from the true meridian presumably being unknown in those days).
And if these two lines of stones could so be laid out (and I think that no one,

in the Outer Hebricles anid their Astronomical Significance. 37
can suppose that their directions are due to mnerechance), the inference is irresistible
that the others also are astronomically aligned, though for what purpose we are
not able definitely to say.
We may summarize, therefore, the results of the preceding investigation as
follows:
(1) There is a single point, situated on the southern edge of the tumulus
covering the now exposed sepulchre, from which an observer finds
himself aligned by lines of megaliths for the equinoctial sunset along
line D, for the rising of Capella along the central line of the avenue
formed by lines " A east" and "A west," and for the rising of
the Pleiades along line D: the two latter events during the epoch
1800 to 1750 B.C.
(2) Line C is laid out on a true north and south line with the Great
Menhir erected at the western edge of the tumulus. All the align-
ments, therefore, are connected with the sepulchre as their point of
origin.
(3) Evidence possibly exists, from the directioniof the line joining the two
outlying menhirs, of the observation of moon-rise when it is Full
Moon at the extreme northern tropical point of its path, occurring
every nineteen years; thus marking that epoch.
(4) The Great Circle, besides being constructed of megaliths of a different,
and much larger type from the others, is placed entirely
asymmetrically to the renainder of the group: and is, therefore,
probably a later, and possibly an alien constructioni, intended to
invalidate, or to mar the astrologically auspicious qualities of the
alignments.
II.-THE SMALLER CIRCLES AT CALLANISH AND VICINITY.
There are at least four circles standing in the neighbourhood of the great and
complex monument described above; and though these are all in a much ruined
state, some reference to them may be of interest.
Their positions and attitude towards the great circle can best be realized
from the map of the district appended. Two of them are quite accessible to the
tourist who makes the expedition to Callanish from Stornoway, as they are
close to the road, and meet the eye at once, each standing on its little
enlinence.
I have numbered these, for convenience of reference, as (1) and, (2). Circle
No. (3) is also at no great distance from the road, but is more out of the beaten track,
being about a inile beyond the Garry na Hine Hotel, on the side road that leads
past it, to the southward.
Circle No. (4) is farther off, and is most easily reached by boat, as it entails a
very roundabout journey by road to get anywhere near it.

38 CAPTAINBOYLE SOMERVILLE.-Prehtstoric Monutments
HEBRIDES
6 fee.
a'*@
eackato z (LEWis)
CL LANS
/eLL ArN _ tSH
- C ircia?)
f ~~ \ ~ ~~~~
II$o
I SC- L E,
X s >~~~~~f{e
of p / D
/1
oftA
V;w fom cie, r S A
_ .iler/ / Uk3
N i
SCALL E
-C.~~
Vi ,fron ) __4rn.,
Cente,oFI. corn*
ti
%
7V 4 -
FIG. 4.

in the Outer Hebrides andl their Astronomical Signifcance. 39
CIRCLENo. (1). (FIG. 4.)

This monument at present consists of a circle of about 56 feet in diameter,
coinposed of eleven monoliths (three of which have fallen), enclosing what appears
to have been a " cove,,"of which four stones remain in place. The " circle " is not
a regular figure, as it now stands, but six of the stones forming it, which remain
erect, stand on a truLlycircular line; the other three erect stones being placed a
few feet inside and outside the circle as derived from the first named. Judging
by the distances between the stones still standing, the whole circle was probably
originally formed by fifteen stones.
Whether the irregular ones have been thus placed purposely, or so as to avoid
the necessity of cutting a hole for a foundation beneath the surface soil into rock,
or whether the irregularity is due to displacement through earth-movements since
the circle was erected, is not possible to decide from surface appearances, though
excavation might partly decide it.
The seeming irregularity might point, possibly, to there having originally been
an inner and an outer " wall" of great stones, enclosing between them an earth
bank, somewhat in the fashion of those described by Mr. A. L. Lewis as occurring
at Longh Gur, in County Limerick, Ireland. It may be recalled (see p. 33) that there
was the remnant of an enclosing " wall " of small stones, filling the spaces between
the monoliths of the Great Circle ; so that the idea is not foreign to the
neighbourhood.
The "cove," as I conceive it, is also rather irregular; but seems to have been
directed more or less towards the sunset of May Day, which is known to have bad
importance as a festival day in pagan times. There are no traces of a burial
within this circle, though it is not impossible that one so existed in former times.
There are several cairns visible on the hill-crests around the horizon. I was
informed by a native of the place that marny,if not all, of these are quite modern;
for the building of cairns forms a sort of pastime for the boys tending sheep on the
hill-sides; and without actually visiting each of them, it would be impossible to
say which had an archleological import, and which not; though it is quite
possible that some are ancient, and intended to indicate an astrollomical
alignment.
No outlying stone or row of stones now exists near this circle from which to
derive any astronomical purposes in connection with it.
CIRCLE No. (2). (FIG. 5.)

This circle stands on a small plateau a little below the first, and is in an even
more ruinous state, as only five of its megaliths remain standing, and three fallen.
The original nunmberwas probably thirteen. Those in situ staind on the circum-
ference of a circle 67 feet in diameter.
A ruined sepulchre, of irregular shape, which once was probably circular, is
placed, not in the centre of the circle, but inclined towards the north-eastern

40 CAPTAIN BOYLE SOMERVILLE.-Prehistoric Mlfonuments
HXEBRIE55E
CALLANI5H
H. ? s. 909.
IX -
0
0
IoO' 0 ~ - 0 v
\\\'\tO~~~ Cew,foctve of (9 1 0 r ,4
~~~~~~~~~~~
~~. ca
"
*o,
VI-
o.~~
? fel0 o o 5?
\~~~~~~~~~0 56 /O Oc\
\Oh ~~~~~~O
SI
( c-awr t0 Sca IE)
COC. -
L -U ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
5CfTLE
(Ro->zo-n-3 o,-
6t rrci
FTG. 5.

in the OutterHebrides and their Astronomicat Significance. 41
quadranit. There is no trace left of any burial cist at its centre. The ring of small
stones, 7 to 8 feet wide, that indicates the site is about 25 feet in diameter, and
stands a few inches above the soil.
Any connections between this circle and the many cairns on the surrounding
hill-tops labour under the same doubtfulness as those described under Circle (1),
and are not worth entering into; but the centre of the first-named circle is exactly
on the line of sunrise of May Day, from the centre of the second; and one must
conclude that this is probably intentional. I cannot find any connectioin of
astronomical imlport between either the first or second circles with the Great
Circle; which is more strange, since it is easily visible, and, of course, very
conspicuous from each of them; appearing as a regular forest of tall stones cresting
the ridge.
Circle No. (2), besides being somewhat larger in diameter than No. (1), is also
composed of more remarkable stones; those remaining being from 6 to 9 feet in
height; that lettered "A" in the plan being especially conspicuous, with a wide
base, tapering to a pointed top. From the centre of the circle this stone lies
exactly on the line of summer solstitial sunrise, to mark which may possibly have
been its original intention.
CIRCLENo. (3). (FIG. 6.)
Circle No. (3) is at about a mile along the road that turns down past the Garry
na Hine Hotel from the, main Stornoway road. It is on the right-hand side, at
about 150 yards from the road, near, but not actually on, the summlit of a heathery
hill, named Sron a' Chail (? " The Hag's Nose ").
There is yet another circle, on the hill-side almost opposite this one, on the
left-hand side of the road, from which it is distant 500 yards; but I was prevented
from visiting it, owing to a sudden bad turn in the weather, through which my
exainination of No. (3) circle was completed only under great difficulties, in wind
and drenching rain.
No. (3) has five stones remaining in its circumference, which was originally
formed, perhaps, by seven. Three of these are placed on the circumference of a
circle 36 feet in diameter; the other two on a concentric circle 42 feet in diameter
with the line joining them directed towards the Great Circle of Callanish, easily
visible, distant two miles. In the north-western quadrant there is the remains of
an oval band of small stones, forming an enclosure, in the middle of which is a
small standing stone slab, 2 feet 6 inches high, which is planted in the ground
on a line N. 220 W. and S. 22? E., or roughly, also in the direction of the Great
Circle; but this is all that remains of the burial cist. The soil is inarshy, and
there is an accumulation of peat, three or four feet deep, surrounding the circle,
which has been removed within its circuit to the original grouild level. The stones
composing the circle are all large and vary, from 7 to 10 feet in height. That
marked " A," which directs the eye from stone D across the circle to the Great

Circle, is especially noticeable:-a massive pillar stone, 8 feet in height, with a

pyramidal top and well-trimmed sides and allnles.
HEBRIDES
LEvwIs
C.ALLAN ISt
A :)
~~,V
: i919-
IN~~~~~~~~~~~
> \
l ;~~~~~~~~~~~~~~~~ne of Equinm*0talSwwr'
s
~~~~~~~~~~~~~rv4inj
sr-o leSrv4n3
6xit-r)o
a_\v%nS ( I IN
to Same sc.1iem
A
View, fromn ceute, 4 Stones COrnpo$ng1 DCircl E
( drawnt- rscale-1)
fHosrl;oi"2 a'd erl,;cco/

feeX to 20
i, 1 .<2 =
FIG. 6.
The following astronomical events are observable from this circle on the
surrounding horizon line, as seen from its centre, by bearings of objects; two of
which are of an artificial nature. I do not take into account cairns, for the
reasons before stated-
(a) Two standing, stones on the skyline of a hilltop at some considerable
distance, (more than a mile), which may possibly form part of a
ruined circle. Bearing 89? 51', elevation 0? 20' 00", giving
declination 0? 12' N., or sunrise on the day of the Equinox.
(b) Large boulder on a skyline hilltop, distant about one mile. Bearing

in the Outer Hebrides and their Astronomical Signitficance. 43
1220 13', elevation 00 34' 30", givinig a declination of 15? 57' S., or
sunrise on November 6th.
(c) A small hilltop plateau, a few miles away, which is almost exactly
level with the distant skyline horizon of hills as seen from the
circle. Bearing 299? 03', elevation 0? 06' 30", giving a declination
of 140 50' N., or sunset on May 1st.
NOTE.-AS this hilltop is not artificially marked, this alignment
should perhaps be looked upon as fortuitous: but the
astrognomical fact is undoubted; and as this was anciently an
important calendrical or religious date, it probably was an
intentional alignment. I may renmarkthat I have found in
other places this arrangement of an alignment, at which the
near and distant horizon of hilltops appear superposed or
nearly so; or, sometimes, of a near horizon cutting off, as it
were, the summit of a distant hill.
CIRCLE No. (4). (FIG. 7.)

The only other megalithic monument in the neighbourhood of Callanish that
I was able to visit is situated on the edge of a perpendicular cliff, about 50 feet
high, on the north side of the narrow channel separating Great Bernera Island
from the main island (Lewis). Four stones remain out of what were probably
originally seven. Of these three are standing, the other having fallen over,
apparently more or less recently; yet, ruined as it is, this monument possesses
singular interest from the fact that it is not a " circle," as for convenience I have
styled it, but a semicircle, with a diameter of about 68 feet (see plan).
It is, of course, possible that the cliff may have subsided since the date of
the erection of the whole circle, though there is no appearance, by fallen debris,
or of a new surface to the rock precipice below, (for it is of rock), that this has
taken place. But if there has been a subsidence, it is at least remarkable that
the straight edge of the cliff-top, after the occurrence, should form practically the
exact east and west (true) diameter of the circle, so that the centre of the remain-
ing semicircle should be on a small irregular hummock of rock, a foot or so inside
the edge.
There is no trace on the ground surface of a burial within its limits.
There are observable from the centre of the semicircle several interesting
astronomical alignments, all solar.
I must here make a short digression to refer to a feature which I have found
in several places in connection with the outlook round the horizon of ancient
monuments, not only at this particular one, but elsewhere. This is the occurrence
of large and massive boulders of rock on the sky-lines of the hills surrounding the
monument, but not necessarily on their summits; and as I have found that many
of them are on particular alignments of the rising or setting of the heavenly
bodies, sun, moon, or certain stars, I cannot but suppose that some, at least, have

Horezon
-
a4nJ Vaer4 I HEBRSIDES LE
""CALE
0 f~ eI 10 140 0 GREAT BERNERA
~~~~~~~~~~~~~~~~
C
It~~~~~~~~9
A~~~
A~L~k K O4raciij
N
View,frorn c.v%r?e, of srStgO
(4rawn $evv~,
Corv~~po1~~?~j
> -
coe-.
/
14,,.,,,,,,,,,,%~~~~~~~~~~~~~~~~~~~~4
ce,4t I _ h ;/'
\ ,t,.:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
c0iglea
t'6a v
FIG. 7.

actually so been placed, in order to form the terminal point of the alignments, or
else that the monument itself-the point of observation-has been erected so that
any particularly conspicuous hill-crest boulder should be on the correct bearing
from it. It does not follow that because every boulder visible from a monument
on the sky-line is not on an astronomical alignment, that the occurrence is
fortuitous when it does so happen; for it is quite possible that while some of
those in sight from a particular monument are in connection with it, the other
boulders may easily be connected with other monuments in the vicinity, each of
which would, of course, require some alignment mark; and it would be very
improbable that the boulders that would suit one point of observation would be
correctly placed for another. Boulders perched on a hill-crest are not in them-
selves very common objects; and when, from the central point of a circle, one is
seen conspicuously standing out as a dark point against the bright sky, and when
its bearing and elevation produces, (for example), the solstitial declination, the
inference is almost irresistible that it was placed there in order to indicate the
sunrise, or sunset, of that date; or, as I suggest above, if the boulder shotuld have
been too massive for transportation (even for those who could carry into position,
and erect, the immensely heavy monoliths of the great circle of Callanish), that
the required monument was placed so as to take advantage of the naturally placed
boulder.
I have already referred to such a boulder "fore-sight " (as we may term it,
from its resemblance to the uses of the fore-sight of a rifle or gun) when discussing
No. 3 Circle, and I know of other examples at Carloway, (not far from Callanish)
at Lochboisdale, (in the Hebridean Isle of South Uist), at St. Kilda, and in the
Lough Swilly district in Coulnty Donegal, Ireland.'
I will now continue my remarks on the Semicircle No. 4, and enumerate the
astronomical events observable in the alignments to hill-crest boulders seen from
its central spot:
(a) Large boulder on hilltop, bearing 600 21', elevated 1? 30' 00",
producing declination 16? 22' N., or sunrise on May 6th.
(b) Prominent boulder, bearing 740 01', elevated 1? 04' 00", producing
declination 8? 54' 49/" N., or
Rising of Pleiades in 1350 B.C.
Spica ,, 1700 B.C.
,, ,, a Arietis ,, 740 B.C.
Aldebaran,, 310 B.C.
(NOTE.-The two last stars must be looked upon as doubtful, as
producing dates possibly too recent for the style of monument;
though not necessarily so.)
1 The latter is actually propped up with smaller stones, in order to make it stand up
prominently. Its artificial attitude is undoubted, though I was unable to find any trace of the
monument with which, presumably, it once was connected.

46 CAPTAINBOYLESOMERVILLE.-Prehtstoric
Moniqitents
(c) Large boulders heaped into a cairn on hilltop, and, from their size,
evidently ancient, bearing 880 04', elevated 0? 46' 00", producing
declination 1' 33' N. Sunrise of March 25th and September 19th
was thus indicated; no doubt intended for the Equinoxes, being
about three days off in each case.
(d) Boulder (not very conspicuous), bearing 1240 45', elevated 10 27' 00",
producing declination 16? 15' S., or sunrise of November 8th.
(e) Prominent boulder on hill-crest, bearing 180? 10', elevated 10 22' 00",
producing the true meridian of the circle (due south).
(J) Conspicuous boulder on hill-crest, bearing 2180 00', elevated 10 22' 00",
producing declination 23? 18' 37" S., or sunset on December 16th
and 28th, namely, six days before and after the day of winter solstice.
(The hills were too hazy further on to the westward to permit of
observations in that direction.)
III.-ISLAND OF ST. KILDA.

The Island of St. Kilda, which is separated by about forty miles of stormy
water from the nearest part of Harris, in the Hebrides, was evidently inhabited
in prebistoric times, (judging by the remains now to be described), in spite of its
remoteness and difficulty of access. That they were able to sustain life in that
desolate spot seems to point almost inevitably to the existence of the use by the
inhabitants of domestic animals, such as sheep, for food; and, indeed, a small flock
of "aboriginal " sheep, which greatly differ both in size and colour from the
modern variety, persist there to the present day; being preserved by the owner of
the islands on Soay, one of the three islets forming the St. Kilda group.
The boats of prehistoric times must have been of a superior build also to
have reached the island at all, and the fact of their employment argues some
considerable advance in the art of navigation.
I have discovered the remains of at least two prehistoric structures on the
Island of St. Kilda, and I have no doubt that the stones of others are built into
the walls of the houses of the little settlement of about seventy souls, who now
inhabit the place.
The two monuments of which I speak, which, so far as I am aware, are new
to science, are dolmens. They are situated at the south-eastern corner of
St. Kilda, about 200 yards from one another. No. 1 is on the very edge of the
sea cliff, about 350 feet high; indeed, in a sense, it overhangs the cliff, for it is
erected on the top of a sort of " pulpit " of rock that projects from the top of the
otherwise perpendicular cliff, while No. 2 is in a less giddy position, several feet
inside the edge, and on comparatively level grassy ground.
DOLMEN No. (1). (FIG. 8, AND PLATE VIII, VIEWS A AND B.)
This dolmen consists, as may be seen in the photograph and plan, of a
triangular, flat-faced slab of stone, one side of which is supported on the natural

surface of the rock, which slopes upwards; the other propped on two other flat
slabs, in such a manner that the top surface is roughly level; the whole being
erected on the top of the projection of the cliff edge, to which I have just referred.
The space left beneath the capstone is only a few inches in height above
the rock surface and altogether too limited in area for the placing in it of a body,
or even of a jar of cremated remains. The object of the dolmen appears to have
been entirely for observation purposes.
The capstone is, as has been said, triangular, with two long, nearly equal
sides, and a shorter base. Standing at the centre of the base, (Plate VIII, View A),
it is seen that the apex is directed towards a hilltop less than a mile away, which
is marked-obviously by the hand of man-with a pointed boulder, supported on
others, in the manner seen in the photograph (Plate VIII, View B). This presents
a near view of it, but is taken from the same aspect. There is no doubt, when
standing at the base of the dolmen, and looking along its length to the point, that
this remarkable object is that to which an observer is intended to cast his vision.
The bearing is 309' 22', and the altitude 8? 44'. This produces a declination
of 280 32' N., and possibly refers to the setting nioon, at its extreme tropic
declination, marking an epoch of nineteen years (see p. 29, where a similar
declination, obtained at the Great Circle of Callanish, is discussed).
I should add, however, that (with equal astronomic probability) it might
refer to the following star, namely, Pollux, in 1200 B.C.
DOLMEN No. (2). (PLATE IX, VIEWS A AND B.)
This dolmen is of a somnewhatdifferent character from that just described. A

large number of biggish boulders have been assembled into a shape roughly oval in
plan, and 25 feet in length. They are not heaped one on the other, but placed
near one another on a piece of rising ground, near the edge of the cliff. At the
southern end of the " oval," that is, at its highest part, a large rough boulder is
supported, at a height of about 1 foot 6 inches from the ground, on two of the
boulders of the group, and at its northern end by a natural projection of the rock
surface. The space thus roofed over is about 6 feet long, 3 feet wide, and 1 foot
6 inches high-of a suitable size and shape, therefore, to contain a body; though
inlihumationwould not be possible, the ground having a flat rock surface (Plate IX,
View A).
The capstone lies with its long side in the direction of the main axis of the
" oval ": and looking along it to the southward, the eye is directed towards an object
so fantastic and remarkable that, placed as it is on the very edge of a tremendous
precipice, 450 feet high, I cannot suppose it to be other than natural, though it lhas
all the appearance of beina artificial.
The photograph, (Plate IX, View B), which was taken at a distance of about
30 feet, gives an excellent representation of this strange " window," formed by the
supporting of a great slab of stone, 8 feet 6 inches long, 7 feet 5 inches wide, 2 feet

48 Monuments
CAPTAINBOYLESOMERVILLE.-Prehistoric
6 inches thick, on two natural supports of rock about 7 feet high. Not more
than a couple of inches of holding keep it in its place at either end; yet it has
stood, one must suppose, thus, for many centuries at least, the terrific fierceness of
the squalls, and pressure of the winter gales, for which St. Kilda is so renowned.
The natives have a name for this object, which, pronounicedas it is, somethiing
like " slain deal," I gather (from a Gaelic dictionary) to be "sleamhain diollaith "-
the "slippery saddle "-which is a fairly descriptive title.
D0
ts~~~~~~~~~~~~~~~~~~~~~L /\ro
-f0
Elevaron X|SPFor
>0Ar~aw base- of Nr ;
9 f eetot 2 30 ^
I I I I I II y .
(P/a,7 and ClavarIoP)
FIG. 8.-ST. KILDA ISLAND.
Dolmen I.-Aug. 1909.
The photograph of the dolmen, which is taken in the direction in wbich it is

pointing,, shovvs the great window as being exactly in the same line, and is obviously
the object used as a " foresight" the heavenly body being seen through it.
The bearing of this line is 165? 50', and the altitude 90 0a', which produce a
declination of 220?13' S.
One star to which this could be referred is Sirius (a Canis Majoris) in about
2500 B.C. There is nothing inherently improbable in this (except that the date

seems uniusually remote for a British monument), for it is known that this star
was employed in ancient Egypt, to " warn " the summer solstice at a date anterior
to that given above. The only other star is Rigel (a Orionis), in about 2300 B.C.,
and, astronomically speaking, this is equally likely.
But what appears to me to be as more probable than either of these is that
the alignment was for the sun, at about eighteen days before the winter solstice.
In various parts of Wales, such an anticipation of the actual solstitial date has
been discovered in several monuments, and the celebrated " Friar's Heel " stone at
Stonehenge seems to have been erected so as to give warning by several days of
the actual day of the solstice.'
?
sJ b o
-q~~~~~~~~~4~C
cP0 hi~~~~
5CAxLE. C/
o fear '? 20 304
FIG. 9.-ST. IECILDA ISLAND.

"The Fairy House.')-August, 1909.
The covering flagstones at presenlt ex:posedare- a few inches below the
soil-level ; and are supported on a dry-walling about 3' 6" bigh, over an
earth floor: the whole thus being subterranean.
IA relic of this " anticipation "is said, to remain to us in the date of Advent Sunday,
which falls at about tbree weeks before Christmas Day. Just as the latter is considered to be
the CbristiaDizedversion of the pagan midwinter (solstitial) festival ; so mnaywe suppose that
Advent supplanted a pagan day on which preparation for the great feast may have beguii.
VOL. XLII. E

50 CAPTAIN BOYLESOMERVILLE.-Prehistoric
Monu,ximents
THE FAIRY HousE. (FIG. 9, AND PLATEVIII, VIEW a.)

There is also on St. Kilda another relic of the prehistoric age, in the shape of
an allee couverte, named "The Fairy House," which was unaearthed,so I was
informed by one of the inhabitants, about forty years ago, when the soil above it
was being dug over for sowing. It yielded, my informant stated, a bronze weapoll
of some sort, soine burnt bones, and some broken pottery, which, he understood,
were in a museum in Scotland. The name (" The Fairy House ") seems to have
been invented purely for the sake of the tourists, and conveys no archaeological
connection with an old-established sanctity.
The pbotograph (Plate VIII, View c), which is taken from the south-eastern,
or entrance, end shows the direction in which the allde is built, namely, towards a
very marked dip in the high hills behind it, which I think is probably intentional. If
so, the bearing of 313' 56' seen along it, combined with the altitude of 170 55',
produce a declination of 370 41' N., which was that of Capella in 780 B.C.,or of
Arcturus in 1180 B.C.,either of which may be accounted, on other grounds, to be a
probable date. This declination refers to no other bright star at any other date.
About two-thirds of the roofing stones of the allde are exposed, and are about
a foot below the present surface of the ground; the remainder are still covered up.
The height of the ilnterior is of an average of 4 feet, and the width is, about the
same. The sides which support the slab roofing are formed, not of large stones on
end, but of a sort of dry walling, of stones of various size, but none very large.
The photograph hardly shows what is an interesting feature, namely, the
remains of a sort of semicircular walling, of the same nature as the interior,
which faced the two sides of the entrance.
Enough of this still remains to prove its former existence, though a good deal
of it has been removed; having proved to be too handy to the village, the " street"
of which is only 50 yards away, to escape being built into outhouses, and so forth.
I mention this particularly as the semicircular entrance seems to be a marked
feature of the Sardinian allees couvertes, or " Tombs of the Giants," recently
described in Mernnon,by Mr. Duncan Mackenzie.'
I do not know whether other " Eirde Houses " (for that is the Scotch name for
these subterranean tombs or dwellings) have this characteristic, besides the
specimen on St. Kilda, but I look forward to examining other specimens, with this
point in view,
There is at St. Kilda the remnainsof a " broch," situated on a point guarding
the bay in which is the village, and extremely difficult of access. I mention this
"(fort" or " dun " (as it is named in Gaelic) chiefly on account of its association
with the all6e couverte," The Fairy House "; for a " broch " or " dun " seems to be
of very much the same style of building as the nuraghe of Sardinia, which are
1 ienmnon, 1908, " The Tombs of the Giants, and the Nuraghe of Sardinia in their West
European relatioins."

in the OuitterHebrides and their Astronomical Significance. 51
always found in the vicinity of the " Tombs ofthe Giants" (see Mr. Mackenzie's
paper in femnon before mentioned); and this double similarity between the
Scotch and the Sardinian towers and burial-places (or dwellings), both in actual
appearance and construction, as well as in the association of this particular form of
tower with this particular form of burial, becomes the more striking.
In conclusion, I would again bring forward the importance of examining the
prehistoric monuments of this, as well as other countries, for the evidence of
orientation. No one with any knowledge of surveying and astronomy can deny
that alignments to definite points of the horizon do exist in some of our prehistoric
monuments, at least; if not in all. Speaking for myself, I have thus examined fifty-
five separate megalithic objects in different parts of Ireland and Scotland, and in all
this number there have only been six in which I could not find evidence of
orientation. The ethnological importance of this is, of course, very great, arguing,
as it does, the existence of a knowledge of the movements of the heavenly bodies,
which implies an unexpected degree of culture among the inhabitants of these
islands at an early date ; as well as a connection, at all events intellectual, with
the western inhabitants of the Continent, to press the results no further.
The external character of the megalithic remains in all countries is so similar
that it would not be surprising to find that an equally similar internal character
of " religious " import should be inherent in their construction; and though
identity of religion does not, of course, necessarily imply identity of race, it at
least points to a missionary impulse from a single source of origin, into which it
would be of great interest to enquire. So elaborate a science as astrology, if it can
be proved to exist, could scarcely spring up spontaneously among a barbarous
people; and in our cloudy climate would be unlikely to rise at all; the heavenly
bodies being so rarely capable of continuous observation: unless, indeed, the
meteorological conditions of 4,000 years ago were extremely different from those of
to-day. If it is the case that such a cult once existed in these climes, we must, 1
think, look to a sunnier, less humid country than our own; such as that East,
whence the British religion of to-day has sprung, as the place of origin for that of
prehistoric days in these islands.
It is to be remarked, further, that the orientation of these ancient buildings in
Britain is to the rising of the sun on definite days of the year (associated with the
rising of certain stars, and of the moon), and that these days are the festival days
of the early Eastern Mediterranean nations: equinoctial, solstitial, " half quarter
days " (when the sun rises at a point of the horizon exactly half-way between that
of the solstices and equinoxes ; namely, with a declination of 160 30' N. or S.).
These dates (and consequently the direction in which the rising or setting of the
heavenly bodies marking them), having become of religious observalnce,it would not
be remarkable to find as we do find, that the tombs of the dead-perhaps we
should say the Temples of the Dead-should be laid out also, in one or other of
these " auspicious " bearings ; and the cult of the dead thus be associated with the
cult of the heavens.
E 2

52 Monuments,&c.
CAPTAINBOYLESOMERVILLE.-Prehistoric
" Half quarter days," mentioned above, which are known nowadays as
Candlemnas, May Day, Lammas, and All Hallows respectively, formed, as Sir Norman
Lockyer has pointed out in " Stonehenae," the important dates of the " agricultural
year"; and their employment, therefore, denotes the probability of the nation
to be in a pastoral, or more or less settled, condition; so that, if the star dates to
be derived from the monuments we have been discussing have any reality, (and I
find it impossible to doubt it), this in itself reveals at how early a date the state
of settled life was established in these islands.
A fairly high condition of mental culture is arguable also, as I think anybody
who has had to do with the manipulation and moving of heavy weights will agree,
from the very bringing together and erection of the enormous stones that were
employed in the making of circles, alignments, and dolmens. And when to this
had to be added the skilful choosing of the site, so that the required risings and
settings of the sun, moon, or stars should take place behind a definite hill-summit,
(though, as I have pointed out, it was often necessary artificially to mark the
azimuths), then, I think, it must be allowed that our ancestors of prehistoric
tiines, though, unfortunately, they had not the art of writing, have left behind them
a record of ability, both of brain and body, for which they have not yet had proper
credit.

Journal of the Royal Anthroyologieal Institute, Vol. XLII, 1912, Plate VII.
(F)-THE GREAT CIRCLE, SEEN FROM S.W., WITH GREAT MENHIR; LATTER A
LITTLE TO LEFT OF CENTRE OF VIEW. SEE PP. 32 AND 33.
(A)-GENERAL VIEW OF LINES "A. EAST AND "A. WEST,

TAKEN FROM N. SIDE OF GREAT CIRCLE. SEE P. 25.
(D)-LJINE B. SEE P. 29.

(Note.-The menliir in foreground is onie of those
foriming Great Circle.)
(B)-LINE A. EAST." SEE P. 27.
(E)-GENERAL VIEW OF GROUP FROM HEAP OF BOULDERS TO

SOUTHWARD, SHOWING LINE C IN LINE WITH GREAT MENHIR
IN CIRCLE (SEEN JUST TO LEFT OF TELEGRAPH POLE), WITH
(C)- LINE AA. WEST."
This content downloaded from 194.29.185.25
SEE P. 27.
on Wed, 25 Jun 2014 03:22:28 AM
LINES B AND D TO LEFT AND RIGHT RESPECTIVELY. SEE P. 30.
(B)-ST. KILDA ISLAND. DOLMEN
(C)-ST. KILDA I
(A)-ST. KILDA ISLAND. DOLMEN (1).
PREHISTORIC MONUMENTS IN THE OUTER HEBRIDESI AND THEIR ASTRONOMICAL

;! i~~~~~~~~~~
_ n ! 7 ~~~'K
(A-G EA VIE OF DOMN() TKLA IH"INO"I AKRUDT HC

IT IS DIETD TE MA SENI TNIGWT IH ADO H ASOE
(B)-NEA
PREHISTORIC MONUMENTS IN THE OUTER HEBRIDES, AND THEIR ASTR


Royal Anthropological Institute of Great Britain and Ireland

Uploaded by

Copyright:

Available Formats

Royal Anthropological Institute of Great Britain and Ireland

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Royal Anthropological Institute of Great Britain and Ireland

Uploaded by

Copyright:

Available Formats

Prehistoric Monuments in the Outer Hebrides, and Their Astronomical Significance

Author(s): Boyle Somerville

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

PREHISTORIC MONUMENTS IN THE OUTER HEBRIDES, AND

[WITH PLATES VII-IX.]

BY CAPTAIN BOYLE SOMERVILLE, R.N.

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

I.-,, THE GREAT CIRCLE" OF CALLANISH.

FIG. 1.-MAP OF CALLANISH AND DISTRICT.

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

A. Two long lines of menhirs, runnincg nearly parallel to one another, a

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

LINE "A EAST."

LINE "A WEST."

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

Capella in 1800 B.C.

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

.F.; ADL; oFs all7)~~L

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

Equinocriai (23) (2.2) (ZI) (20) 4(t c e,

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

Shewing Plan of the, 5rones at prezent 6round-level, and

CiCh abare-as- of iks plotte-4 pos0ton in plan:

all betnmon tWe, same verticazl aid horizontal scal,.

;; 7 \ I 7/70- d5n" haW nUMbCliell C ,wV

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

The astronomical probabilities of each of these four stars is equal in degree;

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

THE GREAT MENHIR.

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

E. THE " CIRCLE."

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

Vernal Summer Autumnal Winter

h f 16 m h mly J,u e s W&V 0

; NF tW 3 T IlilSl IIIII,Y 1/111111f1111 IlNlf5t IF

n t0 Et0 44110;/T 11i!ilLillitltXlI

Jant 2.1 z I[IA

S # g~TGIE DAE thoghu th sam peid at sam Latitude

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

Before summarizing the particular inferences to be derived from each feature

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

II.-THE SMALLER CIRCLES AT CALLANISH AND VICINITY.

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

CIRCLENo. (1). (FIG. 4.)

CIRCLE No. (2). (FIG. 5.)

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

( c-awr t0 Sca IE)

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

CIRCLENo. (3). (FIG. 6.)

This content downloaded from 194.29.185.25 on Wed, 25 Jun 2014 03:22:28 AM

Circle, is especially noticeable:-a massive pillar stone, 8 feet in height, with a