March 17, 1921] 



NATURE 



67) 



Prof. Humphreys is known to us all as an 

 accomplished physicist who is not averse from 

 mathematical reasoning-, with a wide range of 

 knowledge ; a cautious and rigorous thinker, a 

 ;Competent critic, a clear writer, and a shrewd 

 observer who is well acquainted with the inherent 

 difficulty of associating the unconditioned or un- 

 controlled phenomena of the atmosphere with the 

 carefully conditioned and completely controlled 

 experiments of the physical laboratory. He is, 

 )erhaps, best known to us as having been the 

 irst to ofifer an explanation on a deductive basis 

 of the separation of the atmosphere into tropo- 

 sphere and stratosphere, which appeared almost 

 at the same time as Col. Gold's memoir in the 

 Proceedings of the Royal Society, and as 

 having constructed a very useful diagram of the 

 chemical composition of the atmosphere at dif- 

 ferent heights, also arrived at deductively, which 

 is reproduced in Prof. Willis Moore's "Descrip- 

 tive Meteorology " and in the work now under 

 review. It is none the less interesting- because 

 Dr. Chapman and Mr, Milne have suggested to 

 the Royal Meteorological Society that the hydro- 

 gen which occupies so large a part of the diagram 

 should be left out. 



The results of the assiduous study of the pheno- 

 mena of the atmosphere from the point of view 

 which is characterised by the two examples just 

 given cannot fail to be of interest and import- 

 ance for meteorology and meteorologists. They 

 range over an extraordinarily wide field. The 

 mechanics and thermodynamics of the atmo- 

 sphere, including the average meteorological con- 

 ditions of the surface and the upper air, the 

 physical aspects of their changes, the composition 

 of the atmosphere, insolation and radiation, atmo- 

 spheric circulation, evaporation and condensation, 

 rain and raindrops, fogs, clouds, thunderstorms 

 and lightning, form only the first part. It includes 

 a very good chapter on winds adverse to aviation. 

 The second part is devoted to atmospheric elec- 

 tricity and auroras, and the third to atmospheric 

 optics, a very acceptable section in view of our 

 lack of a summary of the subject in English books. 

 The fourth deals with factors of climatic control, 

 and comprises a penetrating discussion of the prin- 

 cipal theories of glacial epochs, with a remark- 

 ably novel and effective discussion 'of the possible 

 or probable effects of vulcanism. 



The book is fully illustrated with many excel- 

 lent diagrams and photographs excellently repro- 

 duced. The pictures of the succession of recorded 

 volcanic eruptions are quite fascinating. Prof. 

 Humphreys may be congratulated on having re- 

 ceived from the Franklin Institute such effective 

 assistance in that important side of the presenta- 

 NO. 2681, VOL. 107] 



tion of a subject which is largely dependent upon 

 the success of its illustrations. 



Apart from the general excellence of the book 

 and the presentation of its material, the parts 

 which impress one most on reading them for the 

 first time are the chapters on thunderstorms and 

 lightning, atmospheric electricity and auroras, 

 and atmospheric optics, as examples of close 

 physical reasoning, and the chapters on factors of 

 climatic control as an example of reasoning of a 

 more general character. 



Where there is such a wealth of subject details 

 can scarcely be regarded, but one or two points 

 attract attention. There is nothing in the index 

 under the letter "U," and the reader is left to 

 draw his own conclusions about the units of the 

 physics of the air, which, in the author's country 

 as in ours, involve a question of real importance to 

 progress in science. It must be remembered that 

 the study of the atmosphere appeals not only to 

 students in physical laboratories where intricate 

 questions about units are all in the day's work, 

 but also to persons outside who care little or 

 nothing for the co-ordination of the various parts 

 of the subject, and to whom any references in 

 unfamiliar units are an unmitigated bore. Such 

 questions should, therefore, be treated in a 

 manner that leaves no room for uncertainty. 

 On p. 30, in a discussion of temperature changes 

 under variations of pressure, Prof. Humphreys 

 tosses ^ = 981 into a mixture of />'s and T's with 

 scarcely any warning to his readers, but on p. 33 

 he makes use of Dt> as the equivalent of pressure 

 p where D is the density of mercury and b the 

 barometric height in millimetres ! The explana- 

 tion of that cryptic equation affords quite a good 

 exercise for the student of physics, but it is not 

 the same as ^- = 981. It is not quite fair to his 

 readers to subject their intelligence to this kind 

 of gymnastic, and when physical reasoning has to 

 be addressed to unprofessional, as well as to pro- 

 fessional, physicists there is really no alternative 

 but to have a coherent and consistent system of 

 units and to stick to it. The longer the step is 

 postponed, the worse for us. One offence against 

 the life-long habits of a reader may be condoned 

 if it is sufficiently pressed, but no one can expect 

 pardon for two such within three pages of the 

 same book. 



On p. 43 the author expresses his preference 

 for "isothermal region" as against "strato- 

 sphere " as a name for that part of the atmo- 

 sphere of which the characteristic feature is that 

 there is no change of temperature with height. 

 This is really astonishing, because to regard the 

 "isothermal region" as really isothermal would 

 be destructive of the whole plan of the structure 



