444 



NATURE 



[August 2, 191 7 



intra-corpuscular. We are surprised also to find 

 no reference to amoebic dysentery ; with many 

 cases now coming from abroad the practitioner is 

 quite likely to meet with this disease. While 

 p)ointing out these few slight blemishes, we can 

 cordially recommend this book as on the whole a 

 simple and reliable guide to clinical bacteriology 

 and pathology. 



Soil Conditions and Plant Growth. By Dr. E. J. 

 Russell. Third edition. Pp. viii + 243. (Lon- 

 don : Longmans, Green, and Co., 191 7.) Price 

 65. 6>d. net. 

 This book is, as the title implies, concerned with 

 the relationship between soil and plant. After 

 an introductory historical account of the subject 

 the author describes the constitution of the soil 

 and the various factors of plant growth. In the 

 development of these topics and of the question of 

 the relation of the plant to its soil environment, 

 the reader is kept constantly in touch with the' 

 best original work at home and abroad. 



The author has made numerous additions in his 

 third edition, and has considerably expanded those 

 portions treating of the biological conditions in 

 soils. He has also added a chapter on the 

 colloidal properties of the soil, in which he brings 

 the reader abreast of the recent researches and 

 disputes of Continental workers, as well as the 

 latest Rothamsted work on the interaction of 

 dilute acids and soil colloids. 



It is superfluous to say that the book is well 

 written. There is an ample bibliography, which 

 should be invaluable to the investigator in any 

 branch of the subject. 



The study of the relationship between soil and 

 plant is exceedingly complex. Progress can only 

 be made by studying the soil in every aspect of 

 importance to the plant. It is a pity that so much 

 labour should have been expended during past 

 years in haphazard manurial trials, designed to 

 instruct the farmer, but yielding generally 

 3. scanty harvest of accurate information. Soil 

 investigators owe a debt to Dr. Russell and his 

 predecessors at Rothamsted for enlarging the 

 study of the soil into a respectable field of scientific 

 activity. G. W. Robinson. 



Lezioni di Antr apologia. By Prof. Fabio Frassetto. 



Vol. iii. Pp. xiii + 422. (Bologna: Mareggiani, 



1917.) Price 20 lire. 

 In the thirteen lectures contained in this volume 

 Prof. Frassetto covers that part of his course 

 which is devoted to the limbs — their evolution, 

 development, and morphology. In his lectures 

 dealing with the methods which are to be applied 

 to the measurement of bones and to the exact 

 estimate of their anthropological characters he 

 has introduced much that is new and valuable. 

 All through these lectures is reflected that spirit of 

 mutual understanding which has existed between 

 the anatomists of Italy and England since Harvey's 

 time. There is no better summary of the contri- 

 butions which British anatomists have made to 

 physical anthropology than is to be found in these 

 clearly written and excellent lectures by Prof, 

 Frassetto. A. K. 



NO. 2492, VOL. 99] 



LETTERS TO THE EDITOR. 



[The Editor does not hold hunself responsible for 

 opinions expressed by his correspondents. Neither 

 can he undertake to return, or to correspond with 

 the writers of, rejected manuscripts intended for 

 this or any other part of Nature. No notice is 

 taken of anonymous communications.] 



The Radiation of the Stars. 



I AM sorry that the only reply I can make to Prof. 

 Eddington's remarks in Nature of July 5 is that his 

 amended equation seems to me neither to be true nor 

 to lead to his supposed laws. But perhaps I may be 

 permitted to offer a few remarks on the general 

 problem. 



Suggested sources of stellar energy fall into two 

 broad classes — gravitational and electrical, the latter 

 including chemical and radio-active sources. Of the 

 former the Helmholtz contraction is by far the most 

 powerful source of energy ; the contraction of our sun 

 from a state- of infinite rarity would provide energy 

 for about 20,000,000 years of radiation at the present 

 rate. 



An upper limit to the capacity of electrical sources 

 can be calculated in a way I have not seen elsewhere. 

 Our sun's mass is 2 x 10"^ grm-j its radiation about 

 4 X 10^^ ergs, per second. Thus to radiate for 20,000,000 

 years (or, say, 7x10^* sec.) at its present rate, each 

 gram of matter must provide on the average 14 x lo''' 

 ergs of energy. A gram of matter contains 3 x 10-^ 

 negative electrons, so that the average electron must 

 provide 4-7x10-* ergs of energy, corresponding to a 

 fall through a potential difference of 10 electrostatic 

 units, or 3000 volts. This is the energy of falling 

 from infinity to a distance of only 4-7 x lo""* cm. from 

 a nucleus loe, and so is probably much greater than 

 any energy actually available from changes of elec- 

 trical structure; it is, of course, enormously greater 

 than any known ionisation potentials. 



It accordingly looks as though the Helmholtz con- 

 traction will provide much more energy than any 

 other source, and we must apparently adjust our 

 views to the time-scale set by the contraction theory. 



If this, is accepted, it will be obvious that the calcu- 

 lation of stellar temperatures and emissions of energy 

 cannot be a steady-state problem at all. I do not see 

 how, in any case whatsoever, any new knowledge can 

 be gained from calculations which assume the star to 

 be in a steady state, for the calculated late of emission 

 can come to nothing but the previously assumed rate 

 of generation of energy inside the star. For pro- 

 gress to be made, this big and difficult problem must, 

 I think, be attacked on dynamical, and not on statical, 

 lines. 



July 7- 



I HOPE I may state my grounds for disbelief in 

 Prof. Eddington's results more clearly. The results 

 are readily combined in the one result that the total 

 emission depends only on M, and varies as M. Prof. 

 Eddington claims to obtain this result in two ways. 

 In his original paper he assumes {M.N., .vol. Ixxvii., 

 p. 20) a rate of generation 47rMe, and after much 

 calculation obtains a result which, on introducing the 

 omitted constants, reduces (I.e. p. 29, equations (29) 

 and (25)) to exactly 4a-Me, proving my point. Again, 

 in Nature of June 14, Prof. Eddington clainjs to ob- 

 tain a result which I do not understand, but which is 

 necessarily contradictory to the foregoing, since the 

 emission cannot now involve e at all. He apparentljr 

 says: — "Tell me the mass of a star and I will t^ 

 you its output of radiation without knowing' the rate 

 at which energy is being generated in its interior; 1 



