Sacred Texts  Legends and Sagas  England  Index  Previous  Next 

Stonehenge and Other British Stone Monuments Astronomically Considered, by Norman Lockyer, [1906], at

p. 304



The Original Cult

I HAVE given detailed evidence showing that the first circle builders in Britain worshipped the May-year sun whether they brought it with them or not. This year was used in Babylon, Egypt, and afterwards in Greece. In the two former countries May was the harvest month, and thus became the chief month in the year. The dates were apt to vary with the local harvest time.

The earliest extant temple aligned to the sun at this festival seems to have been that of Ptah at Memphis, 5200 B.C. I have already referred to this temple in relation to the clock-star observations carried on in it.

This approximate date of the building of the temple is obtained by the evidence afforded (1) by the associated clock-star (see p. 298), and (2) by the fact that the god Ptah represented the star Capella, since there is a Ptah temple at Thebes aligned on Capella at a later time, when by the precessional movement it had been carried outside the solar limit. There was also a similar temple at Annu (Heliopolis, lat. N. 30° 10´), but it has disappeared. The light of the sun fell along the axis when

p. 305

the sun had the declination N. 11°, the Gregorian dates being April 18 and August 24.

Another May-year temple was that of Menu at Thebes,


FIG. 63.—Layard's plan of the Palace of Sennacherib discovered in the mound of Kouyunjik. The temple axis, XXXVI., XXXIV., XXIX., XIX. (XXII. is on a lower level), faces the rising of the May sun.
Click to enlarge

FIG. 63.—Layard's plan of the Palace of Sennacherib discovered in the mound of Kouyunjik. The temple axis, XXXVI., XXXIV., XXIX., XIX. (XXII. is on a lower level), faces the rising of the May sun.

[paragraph continues] Az. N. 72° 30´ W. (lat. N. 25°; sun's declination N. 15°; Gregorian date, May 1).

As we have seen (p. 299), Spica had this declination in 3200 B.C., and the coincidence, may have been the reason

p. 306

for the erection, or, more probably, the restoration, of the temple, 1 especially as γ Draconis came into play as a new clock-star at the same date.

FIG. 64.—Layard's plan of the Mound at Nimrood showing its equinoctial orientation.
Click to enlarge

FIG. 64.—Layard's plan of the Mound at Nimrood showing its equinoctial orientation.

The researches of Mr. Penrose in Greece have provided us with temples oriented to the May-year sun. I shall return to them afterwards, as they are later in time than the British monuments.

p. 307

The explorations of Sir H. Layard at Nineveh, lat. 36° N., have shown that the temple in Sennacherib's

FIG. 65.—The Temples at Chichen Itza.
Click to enlarge

FIG. 65.—The Temples at Chichen Itza.

palace, which may have been a restoration of a much older temple, was also oriented to the May sun.

p. 308

It is a pity that our present-day archaeologists do not more strictly follow the fine example set by Sir Henry Layard in his explorations of Kouyunjik. When he had unearthed Sennacherib's palace (700 B.C.) he was careful to give the astronomical and magnetic bearings of the buildings and of the temple which seemed to form the core of them. The bearing is Az. N. 68° 30´ E., giving the sun's declination as N. 16°.

I am enabled by the kindness of Mr. John Murray to give copies of the plans which Sir H. Layard prepared of the excavations both at Kouyunjik and Nimrood, showing the careful orientation which enables us to claim Sennacherib's temple as one consecrated to the May year, while at Nimrood (Babylon) the equinoctial worship was in vogue as at the pyramids.

In association with these plans of Layard's, I give another by Mr. Maudslay of the as carefully oriented temples at Chichen Itza (N. lat. 20°) explored by him. In these temples, of unknown date and origin, the azimuths of two show that the May year was worshipped. 1

p. 309

The May-Year Monuments in Britain.

In the first glimpses of the May year in Egypt we have dates from 5000 B.C. It does not follow that it did not reach Great Britain before about 2000 B.C. because monuments made their appearance about that time. It is clear, also, that with the possibilities of coastwise traffic as we have found it, it might as easily have reached Ireland by then; 2000 B.C., therefore, is a probable date for the May worship to have reached Britain arguing on general principles; we now come to a detailed summary of the facts showing that it really reached Britain earlier.

Alignments in British monuments designed to mark the place of the sun's rising or setting on the quarter-days of the May year have been found as follows:—

I have already shown that it was the practice in ancient times for the astronomer-priests not only to

p. 310

watch the clock-stars during the night, but also other stars which rose or set about an hour before sunrise, to give warning of its approach on the days of the principal festivals.

Each clock-star, if it rose and set very near the north point, might be depended on to herald the sunrise on one of the critical days of the year, but for the others other stars would require to be observed. This practice was fully employed in Britain.

May Warnings.—The following table gives the stars I have so far noted which were used as warners for the May festival.



Date or dates


Pleiades (R)


Merry Maidens

Pleiades (R)
Antares (S)


The Hurlers

Antares (S)
Pleiades (R)



Pleiades (R)



Pleiades (R)



Pleiades (R)



Pleiades (R)


Longstone (Tregaseal)

Pleiades (R)


(R) = rising. (S) = setting.

It is convenient here to give a list of the May warning stars found by Mr. Penrose in Greece, as it shows that the same stars were observed for the same purpose.

p. 311





B. C.

Archaic temple of Minerva

Pleiades (R)

+  7° 50´

April 20


Hiero of Epidaurus, Asclepieion

    „    (R)

+  9  15

  „   28



    „     (R)

+  9  58

  „   26


Older Erechtheum

Antares (S)

- 14  31

  „   29


Temple of Bacchus

Pleiades (R)

+ 10  35

  „   29



Antares (S)

- 16   0

May  6



    „     (S)

- 16  45

  „    7


The warning stars at Athens were the Pleiades for temples facing the east, and Antares for temples using the western horizon.


August warnings.—Sunrise at the August festival was heralded by the rising of Arcturus, which, as we have seen, was also used as a clock-star. The alignments and dates given in the Arcturus table therefore hold good for August. At the Hurlers, where the hill over which Arcturus was observed fell away abruptly, we find Sirius supplanting Arcturus as the warning star for August in 1690 B.C.


November warnings.—So far I have discovered no evidence that any star was employed to herald the November sun. There may be two reasons for this. In the first place the November festival "Halloween" took place at sunset and the sun itself could be watched, no heralding star being necessary.

Secondly, the atmospheric conditions which prevail in Britain during November would not be conducive to the making of stellar observations at the horizon, and only risings or settings were observed with regard to the quarter-days.

p. 312

February Warnings.—In the same way that Arcturus served the double purpose of clock-star and herald for the August sun, so did Capella serve to warn the February sun in addition to its use at night. The alignments and dates given in the Capella table will therefore hold good for its employment at the February quarter-day.


The Solstitial Year Monuments.

In Egypt generally, the solstitial worship followed that of the May and equinoctial years. The religion of Thothmes III. and the Rameses was in greatest vogue 2200-1500 B.C.

We find little trace of it in Greece proper, though Mr. Penrose has traced it in Calabria and Pompeii, and in some of the islands.

The solstitial cult was born in Egypt; it is a child of the Nile-rise. I have shown in my Dawn of Astronomy that the long series of temples connected with the solstice may have commenced about 3000 B.C.; but for long it was a secondary cult; it was parochial until the twelfth dynasty, say 2300 B.C. Egypt's solstitial "golden age" may be given as 1700 B.C., and her influence abroad was very great, so that much travel, "coastwise" and other, may, be anticipated. It is for some centuries after the first date that the introduction of the solstitial worship into Britain may be anticipated. It, for instance, is quite probable that the pioneers of this worship should have reached Stonehenge in 2000 B.C.

p. 313

The solstitial alignments found by Mr. Penrose in Greece are as follows:—

We find plentiful evidence that the worship of the solstitial sun such as was carried on in Egypt at Karnak and at other places 1 was introduced into Britain some time after the May-year worship was provided for in the monuments.

Although some of the alignments already discovered are in all probability solstitial, the variation of the sun's solstitial declination is so slow and takes place between such narrow limits that a most careful determination of the actual azimuths and of the angular heights of the various horizons must be made before any definite conclusion as to dates can be arrived at. The necessity for this care is illustrated in the paper on Stonehenge 2 communicated to the Royal Society by Mr. Penrose and myself in 1891; where, after taking the greatest precautions, the resulting date. was, in doubt to the amount of 200 years in either direction.

p. 314

So far Stonehenge is the only temple at which these observations have been made, so that for the other alignments contained in the following list no dates can yet be given.

I cited an alignment at the Hurlers which marked the rising point of Betelgeuse. This star warned the summer solstice sunrise at about the Hurlers' date. So far, however, I have not yet found any suggestion of its use elsewhere.

At Shovel Down and Challacombe on Dartmoor there are avenues pointing a few degrees west of north. The sight-lines along these avenues would mark the setting-point of Arcturus at the time that that star (setting) warned the rising of the sun at the summer solstice; but this use cannot be considered as established, as Arcturus would scarcely set before its light was drowned in that of the rising sun. The absence of

p. 315

darkness in high summer in these latitudes and the bad weather in the winter may both be responsible for so few alignments for the solstices.


The Equinoctial Year Monuments.

The equinoctial pyramid and Babylonian cult in vogue in Egypt in the early dynasties (4000 B.C.), with the warning stars Aldebaran (March) and Vega (September), was represented in Greece at a much later period. The facts for Greece, according to Mr. Penrose, are as follows:—

In Britain equinoctial alignments are not wanting, but so few have been traced that I have reserved them for future inquiry.


306:1 See Dawn of Astronomy, p. 318.

308:1 The temple conditions are approximately as follows:—

313:1 Dawn of Astronomy, p. 78.

313:2 Proc. Roy. Soc., vol. 69.

Next: Chapter XXX. The Life of the Astronomer-Priests