It seems probable that the great circle was one of the first erected, and the fact that, like Stonehenge, it had an avenue, but that, unlike Stonehenge, the avenue was directed towards the May and not the June (solstitial) sunrise further, I think, suggests that the May worship was considered the most important and was the first provided for.
There is reason for supposing that the great circle was at all events built before the S.W. one. The great circle is situated at a lower level than the S.W. one. The angular elevation of the hills over which Arcturus rose would appear, therefore, to be higher from the great than from the S.W. circle. Arcturus has been reducing its declination for centuries in consequence of the precessional movement. It would therefore rise gradually in a greater azimuth, that is, nearer the east. An observer in the centre of the great circle, to follow this more easterly rising over the quoit, would have to change his position gradually to the westward. But there was another way. The original direction could be nearly maintained if the observation were made at a higher level near the original line, as then the relative elevation of the rising-place would be reduced.
This is what possibly was done, and this indeed may be the vera causa of the building of the S.W. circle.
This view of the possible function of the “quoit” is, of course, strengthened by the fact that we find traces of high northerly alignment in other stone circles. I have already shown that there are such alignments in Cornwall.
The “quoit” is nearly on a level with the great circle, while the hills rise behind it. It has been suggested that it would have been more useful on the top of the hill, but this suggestion cannot be accepted for a moment if it were used in the way I have indicated. On a dark night it would have been invisible, and it also would have prevented the observation of star-rise if it were truly aligned. Being comparatively near the circle it could easily have been illuminated at the critical time, and thus have anticipated the bright line micrometer of more modern times.
So far I have found no obvious use for the avenue attached to the N.E. circle. The conditions are:
| Az. | Height of Hills. | Dec. | ||
|---|---|---|---|---|
| Morrow. | Lockyer. | Morrow. | Morrow. | Lockyer. |
| S. 83° 52′ E. | S. 79° E. | 1° 40′ | 3° 52′ 30″ S. | 5° 49′ 30″ S. |
With regard to this N.E. circle, in relation to the large circle, the data are as follows:
| Az. | Height of Hills. | Dec. N. | ||
|---|---|---|---|---|
| Morrow. | Lockyer. | Morrow. | Morrow. | Lockyer. |
| N. 53° E. | N. 51° E. | 1° 5′ | 22° 43′ 50″ | 23° 48′ 46″ |
As Mr. Morrow states, the choice of centre of the circle may alter the azimuth obtained by as much as “a few degrees,” but the value obtained from the Ordnance map is, definitely, N. 51° E., and with the height of hills determined by Mr. Morrow this would suggest that the N.E. circle was really erected to provide the alignment, from the centre of the great circle, or from the Cove, to the summer solstitial sun, about the year 870 B.C., Stockwell’s values for the obliquity being taken. This result is the more striking as it gives a date for the substitution of the June for the May worship at Stanton Drew, which is in full accordance with that obtained for the similar change at Stenness.