682 



NATURE 



\>^ 



Lk io, 1923 



has shown that the great African lakes, Victoria and 

 Albrrt, show variations of Ifvfl amounting to several 

 f ( . ally in synch m Mi the sunspcjl curve, 



thr .ii.iMii.il of the two ciH ». - ... . urring together. The 

 rainfall in tlic drainage basins docs not show a corre- 

 sponding variation, and it seems probable that the 

 high level at sunspot maximum is due to decreased 

 cvajMjration owing to the lower air temperature. 

 Again, Douglass has found several cases in which 

 the 1,'rovvth of trees, as indicated by the thickness of 

 tlirir iiimiul rings, has v.ni .1 marly in synchronism 

 with the solar cycle ; this is i leurly an index of some 

 more immediate solar meteorological cfTect, whether of 

 thermal origin or not. 



The question as to a possible influence of solar 

 activity on the barometric pressure is one which has 

 received considerable attention. In the case of this 

 element the solar effects must necessarily be more 

 complicated than in the case of temperature, where 

 the variations are likely to be everywhere of the 

 same sign at a given time, though with local differences 

 of magnitude. The total atmospheric pressure upon 

 the earth can scarcely be appreciably affected by the 

 sun's changes, so that if the solar influence increases 

 the pressure in one region, there must be a counter- 

 vailing change in other regions. The difficulty of 

 detecting such effects is clearly much greater than 

 that of demonstrating the temperature changes — 

 itself an exacting task. Any such barometric changes 

 which occur appear to be small, and must be obtained 

 by averaging the results from a number of stations ; 

 if these happen to be distributed across the borders 

 of oppositely-affected regions, the effect sought for 

 ina\ almost or quite cancel out ; in any case it requires 

 extremely detailed research to establish changes of 

 particular sign in different regions, and to ascertain 

 the limits of these regions. 



Such investigations have of late years been pro- 

 secuted vigorously, and not without valuable results, 

 by a number of American meteorologists — amongst 

 others — and are recorded by Mr. Ellsworth Huntington 

 in his new book. The sub-title of this work is "An 

 hypothesis of wcatluT and sunspots " ; it is a com- 

 panion volume to his recent hook on '' Climatic 

 Changes," which dealt mainly with past relationships 

 between the earth and sun, while the present work 

 is concerned with existing connexions. The leading 

 idea of both books is that terrestrial meteorology' 

 depends partly on purely terrestrial conditions, and 

 partly on changes in the solar activity ; the latter 

 " are supposed to act chiefly through variations in 

 barometric pressure and especially in the number, 

 location, and intensity of cyclonic storms." It is also 

 claimed that there is an important solar-acti\ity 

 NO. 2819, VOL. I 12] 



effect on atmospheric electricity. The elucidation of 

 such questions as these is obviously a matter of great 

 interest and significance, and it is very convenient to 

 have a summary of the present state of knowledge of 

 the subject set out as is done in this book. The auth<ir 

 has himself devoted enormous lalx)ur to this kind of 

 investigation, and writes both with enthusiasm and 

 with a wide acquaintance with the literature concerned. 

 But to the reviewer it seems that much more eNidcnce 

 is required before it is safe to accept many of the 

 conclusions which the author regards as established. 

 In particular, the evidence for any i ' .f 



solar activity on barometric pressure _ i' 



electricity seems inadequate seems, however. 



to be a ca.se for a connexion imween sunsp<'' 

 cyclones in certain tropical regions. 



A considerable section of Mr. Huntington's book is 

 devoted to the inverse problem of planetarj' influence 

 upon solar activity. Mr. H. Helm Clayton contributes 

 one of the four chapters in this section, and it is rat In r 

 surprising to see in this chapter what seems to be an 

 error elsewhere expressly pointed out by Mr. Hunting* 

 ton, namely, that the tidal influence of the planets 

 on the sun is inversely proportional to the square of 

 the distance of the planet from the sun. Many at- 

 tempts have been made to relate the sunspot varia- 

 tions to planetary periods, but with doubtful succes-. 

 The period of Jupiter (11 '86 years) is not ver>- different 

 from the mean sunspot period (ii'2 years), but the 

 discrepancy is sufficient to render it ver\' proMdn atic al 

 whether any relationship between the t 

 credited, even when allowance is made lor the dis- 

 turbing influence of the other planets. Mr. Huntington 

 puts forward a hypothesis of electrical influii 

 the planets upon the solar atmosphere, but at pn sent 

 this is almost purely speculative. Such questions may 

 be easier to decide when the nature of sunspots i^ 

 better understood than now. At the moment it is at 

 least a possible \new that the main sunspot variation 

 is due to some intrinsic solar period. S. C. 



Biolog:y and Sociology. 



Essays of a Biologist. By Julian Huxley. Pp. x\ 

 (London : Chatto and Windus, 1923.) 75. 6d. net. 



THIS brilliant book, though somewhat disfigured by 

 overlapping and repetition in certain parts, is 

 one of the most suggestive and enlightening works for 

 the popularisation of science which have appeared fur 

 a long time. It covers a wide field, and Mr. HuxK \ 

 shows himself in it a man of wide interests, many part - 

 and an easy and attractive style of writing. He ha- 

 two serious articles, covering much the same ground, 

 on a new rationalistic conception of God ; a sound 



