After α Ursae Majoris had become circumpolar in the latitude of Denderah, γ Draconis, which had ceased to be circumpolar, and so fulfilled the conditions to which I have referred, replaced it. Its declination was 58° 52′ N. about 3100 B.C., and it, therefore, could have been watched rising in the axis prolonged of the old temple in the time of Pepi, who restored it then, no doubt on account of the advent of the new star, and is stated to have deposited a copy of the old plan in a cavity in the new walls. Here, then, we have two dates given by orientation of a clock-star temple entirely agreeing with the most recent views of Egyptian chronology. In Dr. Budge’s _History of Egypt_ (iii. 14) the story of the rebuilding of the temple at Annu by Usertsen (2433 B.C., Brugsch) is given from an ancient roll. Supposing this temple built parallel with the faces of the remaining obelisk, γ Draconis would rise in its axis prolonged 2500 B.C., proving that Usertsen did at Annu what Pepi previously did at Denderah, and that the same reason for restoration and even the same star were in question.[121] When the clock-star ceased to be visible in the chief temple other subsidiary temples were subsequently built to watch it. Thus γ Draconis was watched at Thebes from 3500 B.C. to the times of the Ptolemys by temples oriented successively from that of Mut Az. N. 72° 30′ E. to 68° 30′, 63° 30′, and 62°.[122] It is worth while to show that what we know now of the Egyptian methods of observation enables us to carry the matter further, while we gather at the same time that in consequence of the difference of latitude the method employed in Egypt could not be followed in Britain. I showed in the _Dawn of Astronomy_ that several ancient shrines consisted of two temples at right angles to each other (see Fig. 13), one axis pointing high N.E. to observe the clock-star--the worship of Set--the other low N.W. to observe either the sun by itself, or in association with some important star of the same declination as the sun. The temples of Mut and Menu (or Min), and of Amen, with the associated temple M. of Lepsius, at Karnak, are the best extant examples of this principle of temple building. There is evidence that both at Annu and Memphis the same principle was followed, but at Annu one obelisk alone remains, and at Memphis one temple; from these, however, Captain Lyons and myself have obtained sufficient data to enable the original directions of the temple-systems to be gathered. At Denderah, if such a N.W. temple ever existed it has disappeared, but as the monument stands there are still two temples at right angles to each other, but the second one faces S.E. instead of N.W.