148 



JVA TURE 



[July 29, 192: 



Letters to the Editor. 



[The Editor does not hold himself responsible for 

 opinions expressed by his correspondents. Neither 

 can he undertake to return, or to correspond with 

 tin- writers of, rejected manuscripts intended for 

 this or any other part of Nature. No notice is 

 taken of anonymous t ommunications.'] 



The Intensity of X-ray Reflection. 



In a letter on " The Intensity of X-ray Reflection 

 from Powdered Crystals " (Nature, July 8, p. 38), Prof. 

 A. H. Compton and Mr. N. L. Freeman have described 

 some measurements which they have made on the 

 reflection of X-rays by rock-salt, and have compared 

 their results with figures published by us in the 

 Philosophical Magazine for July 1921. Quantitative 

 measurements of the amount of radiation reflected 

 by the crystal are of great interest, since it is probable 

 that an estimate of the way in which the electrons 

 are distributed throughout the atom may be based 

 on them. 



In the case of the reflection by the (100) planes of 

 rock-salt, they find a large discrepancy between our 

 figures and theirs. The point at issue may be put 

 as follows. By comparing the incident and reflected 

 monochromatic X-ray beams, and applying formula; 

 calculated by Darwin and others, it is possible to 

 estimate the amount of X-ray energy scattered by 

 an atom in certain directions in terms of that scattered 

 by a single electron under the same conditions. As 

 the angle through which the rays are scattered is 

 increased, the amplitude of the diffracted wave falls 

 off, owing to interference between the waves scattered 

 by' the electrons distributed around the centre of the 

 atom. In the case of the (100) reflection by rock-salt, 

 we found that at the angle at which the reflection 

 takes place the effect of the 28 electrons in a pair 

 of sodium and chlorine atoms is reduced to an effect 

 20-1 times as great as that of a single electron (p. 7, 

 Phil. Mag., loc. tit.). The ratio 20-1/28 is that 

 indicated by V in this note by Prof. Compton and Mr. 

 Freeman. 



We obtained an estimate for V 2 by measurements 

 made with a single rock-salt crystal. Prof. Compton 

 and Mr. Freeman determined the quantity by the 

 powder method, measuring the amount of radiation 

 diffracted by a layer of powdered rock-salt spread on 

 a plate, irradiated by an X-ray beam made homo- 

 geneous by previous reflection. They obtain a value 

 of 1A 2 . equal to 0-64, correct to 10 per cent., and quote 

 our results as giving i/- 2 = o-43, a much lower value. 



It may well be that our estimate is too low, for the 

 reasons given by Prof. Compton in his note, which 

 have been fully discussed recently by Darwin (Phil. 

 Mag., May 1922). By using the powder method, the 

 extinction effect, which increases the absorption of 

 the rays passing through at the reflecting angle, is 

 partly or wholly eliminated. We believe, however, 

 that an error has been mad© in quoting our value for 

 f 2 as being 0-43. We obtained 

 20-1 

 "28 

 >A 2 = 0-515. 

 If this be the case, the discrepancy is not so great 

 as Prof. Compton and Mr. Freeman believe. They 

 used the molybdenum K„ line, of wave-length 0-708 

 4 U., whereas we used the rhodium K a line of wave- 

 length 0-615 A.U. The value of V 2 , however, is 

 independent' of the th for a spectrum of 



given order. 

 ' Their estimate for the value of >p- in the case of the 



(100) reflection is of the very greatest interest. The 

 measurements made with a single crystal are difficult 

 to interpret for the strong reflections, on account 

 of extinction, though they are probably accurate for 

 the weaker ones. On the other hand, it is just these 

 values for ■+'- at small angles which are of the greatest 

 importance, as indicating the distribution of electrons 

 in the atom, particularly in its outer confines. 



W. L. Bragg 

 R. W. James. 

 Manchester University, July 13. 



Geology and the Primitive State of the Earth. 



I have no wish to depreciate the importance of 

 the work of those geologists who have examined the 

 older known rocks of the earth's crust and decided 

 that all of them have been altered since they first 

 became solid ; but I question whether their results 

 afford the evidence concerning the primitive state 

 of the earth that is suggested by Prof. A. P. Coleman 

 and Prof. G. A. J. Cole (Nature, June 17 and July 8). 

 The deepest boring yet made is one of 2-5 km., and 

 there are few over a kilometre in depth ; that is the 

 extent of the crust accessible to geologists. Yet the 

 average thickness of sediments over the continents 

 is probably some kilometres, and in many places it is 

 practically certain that the thickness reaches tens of 

 kilometres. In addition, we have vast thicknesses 

 of intrusive and eruptive rocks. Would it be surpris- 

 ing if all rocks in their primitive state have been 

 denuded away or buried to an inaccessible depth ? 



Even if it is considered that a piece of primitive 

 crust, still unaltered, has had an appreciable chance 

 of being discovered and examined, I do not see that 

 the results actually found afford any evidence for the 

 cold earth theory as against the hot earth theory. 

 The primitive rocks formed by slow accretion of 

 solid planetesimals would be in a perfectly character- 

 istic and recognisable condition ; no such rock has, 

 so far as I have heard, been discovered, and therefore 

 Prof. Coleman's arguments, if they are sound, would 

 be just as injurious to the planetesimal theory as to 

 the hot earth theory. 



The temperature of the surface of the earth is at 

 present maintained almost wholly by solar radiation, 

 and not by conduction from the interior. The rate 

 of conduction of heat from the interior is of order 



2 x io" 6 cals/cm. 2 sec. The solar constant is about 



3 x io -2 cals/cm. 2 sec. Even if the earth had a solid 

 crust only a mile thick, the surface temperature 

 would still depend wholly on the sun's radiation. 

 Thus it would become cool at the surface at a very 

 early stage in the solidification, and it is not surprising 

 that the oldest rocks should show evidence of water 

 at the surface. 



I have published in the Month!}' Notices of the 

 Royal Astronomical Society, vol. 77, 1916, pp. 84-112, 

 and in the Proceedings of the Royal Society, vol. 100, 

 IC121, pp. 122-123, arguments that I believe to be 

 fatal to the form of the planetesimal hypothesis that 

 attributes to solid planetesimals any cosmogonical 

 importance, and postulates an earth that has always 

 been solid. No supporter of this hypothesis has yet 

 published any repl} to, or acknowledgment of, these 

 criticisms. Harold Jeffreys. 



St. John's College, 

 Cambridge, July 10. 



a-Particles as Detonators. 



The observation recorded by Mr. Henderson in a 

 letter to Nature of June 10, under the above heading, 

 appears to involve possibilities of the greatest prac- 



NO. 2752, VOL. I 10] 



