I assumed, without historical data, that this view was acted on in Egypt, at the temple of Menu; Mr. Penrose found, with historical data, that it was actually acted on in Greece at the Parthenon. To show that we are at all justified in this view we must study the association of gods with temple worship, and look for temples in different azimuths erected at different times if the god is a star; and we can run the star home if the dates fall in with the star’s precessional change. Thus there is reason for supposing that the god Ptah and the star Capella were associated. There is a temple of Ptah at Memphis, Az. N. 77° 15′ W., hills 50′, decl. N. 11°, star Capella, date 5200. In the rectangular system at Memphis, then, α Ursae Majoris was watched in one temple and Capella in the other at that date. There is also evidence that the god Menu was associated with the star Spica. In the temple system of Mut at Thebes, in 3200 B.C., γ Draconis was used as a clock-star in one temple, while the setting of Spica was watched in the other. If a temple is erected to the sun with no specially named cult, it may be a sun-temple pure and simple, not connected with star worship because there was no star with the proper declination at the time. In Greece temple-building was carried on at a much later time, so late that perhaps water clocks were available, so that we should not expect to find many clock-star temples in that country. As a matter of fact there is only one, of which the data, according to Mr. Penrose, are as follows:-- N. Decl. Star. Date. Thebes, The City of the Dragon +54° 28′ γ Draconis 1160 It will be seen that the star used in Greece was the last clock-star traced in the Egyptian temples. _Britain._ I now come to Britain. So far as my inquiries have gone, these clock-star observations were introduced into these islands about 2300 B.C. In my statement concerning them I will deal with the astronomical conditions for lat. 50° N., as it is in Cornwall that the evidence is most plentiful and conclusive. In that latitude and at that time Arcturus, decl. N. 41°, was just circumpolar with a sea horizon, and therefore neither rose nor set. Capella, decl. N. 31°, when northing was 9° below the horizon, so that it rose and set in azimuths N. 37° E. and N. 37° W. respectively; it was therefore invisible for a long time and was an awkward clock-star in consequence. [Illustration: FIG. 61.--Arcturus and Capella as clock-stars in Britain. AB = sea horizon. A′B′ = horizon 3° high. ] Fig. 61 represents diagrammatically the conditions named, the circumpolar paths of Arcturus and Capella being shown by the smaller and larger circle respectively. _A B_ represents the actual sea horizon and _A′ B′_ a locally raised horizon 3° high, whilst the dotted portion of the larger circle represents the non-visible part of Capella’s apparent path.