October 19, 191 ij 



NATURE 



5 J 5 



In the end I managed to peg out a line, down as much 

 of the axis as I could get at, 523 feet long, from the 

 columns in front of the sanctuary down to the lower end 

 of the Hypostyle Hall ; unfortunately, both the extremities 

 of the line are blocked up. The sanctuary itself is com- 

 pletely filled up with the huge stones of the fallen roof, 

 and the last columns of the Great Hall at the other end 

 are at present built round with stones and bags of sand 

 on account of the repairs being carried out to the neigh- 

 bouring pylon, while the pylon itself is timbered up to 

 prevent its falling, so that the two important points for 

 a survey of this part of the axis cannot be used at present. 



The line at this end of the axis had to be continued by 

 the theodolite alone, as no measures for centring were 

 possible, as it had to be carried through the Rameses 

 Pylon into the Outer Court so far as the standing pillar 

 of Tirhakah in order to get the true bearing of the central 

 line by observations of the Pole Star, that star not being 

 visible from any place on the axis inside the buildings. 



The result of the survey in general quite confirms the 

 data used by Sir Norman Lockver in fixing the date at 

 which the original axis was laid down, viz. about 3700 B.C. 

 ■ — a date which M. Legrain fully accepts on the results 

 of his excavations, as the building of the upper end of the 

 temple has been assigned on archaeological grounds to 

 about this period ; indeed, two statues have been dug out 

 by him, both now in the Cairo Museum, which give direct 

 evidence as to the date. One is a seated figure of Cheops 

 with his cartouche (of the fourth dynasty), the date of 

 which is given in the lists as B.C. 3733, and the other is 

 a headless figure which has been assigned to the work of 

 the third dynasty. These figures were, of course, dug up 

 long after Sir Norman Lockyer's survey ; no older work 

 has been found. 



The height of the hills behind which the sun used to 

 set at the summer solstice, to which the temple was 

 oriented, was takm at 2° 30'. From the spot I was able 

 to climb up to on the stones filling the sanctuary, to what 

 I thought was about the height of an altar, I made it a 

 little more ; but as I had to see the hills through the 

 timbering of the Rameses Pylon, between the struts, I 

 could only measure the small part which was not covered ; 

 if this is the correct height, as I believe it to be, it 

 would make the date of the foundation a little earlier, 

 possibly to the time of the headless statue, which M. 

 Legrain has assigned to B.C. 4000. 



There are a great many difficulties just now in carrying 

 out such an accurate survey as is required to arrive at an 

 astronomical date of any value on account of great work 

 that is being carried on. It ffiust be remembered that 

 the temple is about 1200 feet long, and stands on an area 

 about five times that of St. Paul's, and is divided into 

 numerous halls, corridors, and gateways : hut all these are 

 connected by the axis which runs through the whole build- 

 ing from east to west. This axis, when originally laid 

 down, pointed to the place on the hill at which the sun 

 disappeared behind it on the longest day, and the differ- 

 ence between the place where the sun set then and where 

 it sets now gives the date of the foundation of the temple, 

 the rate of the sun's change in declination being known. 



M. Legrain tells me that in about two years' time he 

 will have cleared out the fallen roof of the sanctuary, and 

 that by that time he hopes the repairs to the Rameses 

 Pylon will be completed and the timbering removed. In 

 that case a unique opportunity will present itself for a 

 survey of the whole of the axis at once from the court 

 behind tin- opened-out sanctuary right down to the 

 Ptolemaic Pylon at (he west ; and this Mr. Dowson, the 

 Direcfor-Gpnt ral of the survev in Cairo, has verv kindly 

 undertaken to have done by the survey officers so snon as 

 the work is completed. Howard Payn. 



20 Hvde Park Place, W., October n. 



A Possible Relation between Uranium and Actinium. 



It is believed fairly generally that actinium has its 

 source in the disintegration of uranium, although it is not 

 a member of the direct line of descent through radium. 

 This belief is haspd mainly on the fart that actinium and 

 its products h.nvp a constant ratio to uranium in minerals, 



NO. 2IQO. VOL. 87I 



and since this ratio is very small actinium is supposed to 

 be a branch-product. 



In The Philosophical Magazine for September Mr. G. N. 

 Antonoff describes some experiments in which a new pro- 

 duct is obtained, called uranium Y, and gives strong 

 reasons for the view that it is derived, not from uranium N, 

 but directly from uranium. It is always in a small ratio 

 to uranium X. Antonoff has shown that it is probably 

 a branch-product, and a possible origin of the actinium 

 series. 



The following considerations may indicate how such a 

 branch-product could be formed. They were not thought 

 worthy of mention until the starting point of a branch- 

 series had been found experimentally. If a single atom of 

 uranium begins to disintegrate, it ordinarily leads to the 

 whole radium series, without disturbance from other atoms. 

 But the molecule of uranium will contain at least two 

 atoms chemically combined, and perhaps a large number. 

 An instability arising in one atom may frequently produce 

 a similar instability in a contiguous atom, or even a pro- 

 jection of one atom into another, so that two atoms may 

 break up together and form new combinations. 



The scheduled atomic weight of uranium is 238-5. Two 

 such atoms have a weight 477. If they break up together 

 and form only one substance, it might have a molecular 

 weight equal to that of uranium, or J, j, . . . of that of 

 the combined atoms. On the assumption that three atoms 

 of a substance are formed, its atomic or molecular weight 

 is 159. 



The experiments of Russ appear to give the most trust- 

 worthy value of the atomic weight of actinium emanation. 

 They showed that the thorium emanation is 142 times as 

 heavy as that of actinium. If thorium emits two a 

 particles, both helium atoms, its emanation should have 

 an atomic weight about 224. The atomic weight of 

 actinium emanation thus becomes about 156. 



On a theory of the constitution of the elements given 

 to the British Association by the writer at Portsmouth, it 

 is more likely that the emanation from thorium has the 

 same atomic weight as that from radium, and that 

 actinium emanation has an atomic weight of 152-5. Russ's 

 experiments would then lead to a value very close to this 

 for actinium emanation. It is, of course, difficult in most 

 cases to obtain satisfactory conclusions from such experi- 

 ments on diffusion, but there are strong grounds for think- 

 ing that in this special case the usual sources of error have 

 been minimised. 



If uranium Y be formed in this way, with an atomic 

 weight 159, it may well be the parent of actinium, whether 

 the suggested atomic weight of the emanation be correct or 

 not, and it is not unlikely that at certain stages in the 

 radium series a similar series of branch-products of low 

 atomic weight may be produced. There is evidence of this 

 in the complex product radium C. 



J. W. Nicholson. 



Trinity College, Cambridge, October 11. 



Hot Days in 1911. 



Mr. MacDowall's letter in Nature of October 12 

 (p. 485?, in which he directs attention to certain features 

 in the sequence of annual number of hot days at Green- 

 wich, is interesting. Nevertheless, I think he himself will 

 acknowledge that his example, viz. the summer of 191 1, 

 is a happv one. What his diagram would lead one to 

 expect if one were making a forecast, and not a retrospect, 

 is that the number of hot days in 191 1 would lie between 

 his lower limit of 90 days and an upper limit of about 

 90+130, or 220 days; the value half-way between the two, 

 i.e. 155 days, being the " most probable." Clearly in this 

 case the upper limit, and also the most probable value, may 

 be disregarded, and the lower limit is sufficiently high to 

 be worthy of note. 



The fact is that the dot for 1909, the ordinate of which 

 is sought, lies on the lower edge of the boundary of the 

 area of dots, so that in this case the lower limit gives a 

 close approximation to the truth. If a dot haptens to lie 

 near the upper edge of the area of dots, the upper limit 

 of its range along the vertical of the diagram becomes a 

 close approximation. But since the difference between the 

 upper and lower limits is no fewer than about 130 days, 



