December 9, 1922] 



NA TURE 



763 



which lie in columns of five deep ; observations of 

 momentum at the centres of the " white." The 

 changes in any one slab are computed with the aid 

 of the known conditions of the surrounding slabs : 

 hence the calculation for any arbitrary area is limited 

 to the interior slabs, and the area amenable for com- 

 putation diminishes with each step of the process. 

 There is a great amount of original and ingenious 

 scientific speculation and discussion in the description 

 of the process. 



Nearly a hundred separate algebraical symbols are 

 employed. The author sketches a fancy picture of 

 the process of computation going on for the weather 

 of the whole world in a great theatre or forecast- 

 factory in the form of a hollow globe. A spherical 

 orchestra of computers calculates the future weather 

 from the information supplied by 2000 stations under 

 the direction of a conductor at the centre of the globe. 

 In order to keep pace with the weather the orchestra 

 would consist of 64,000 performers on the slide-rule 

 or calculating machine ; and even then, with a space 

 unit of 200 kilometres, phenomena on the small scale, 

 such as tornadoes or local thunderstorms, might be 

 missed. Part of the appeal of the book is for a dis- 

 tribution of stations to be arranged so as to give the 

 process of calculation a better chance than the existing 

 distribution in Europe affords. 



There are twelve chapters of very unequal length. 

 Chapter I. is a brief summary of the contents of the 

 book ; Chapter II. is a simplified example of the 

 method of calculation by finite differences which is 

 to be used. This preliminary canter shows incidentally 

 that a distribution of pressure according to an assumed 

 geometrical law, and a universal geostrophic wind 

 corresponding therewith, lead to the conclusion that 

 a vast system of high pressure over the Eurasian 

 continent, covering one half of the Northern hemisphere 

 east of the meridian of Greenwich, would result in 

 an increase of pressure over England, which lies on 

 the margin. This result is regarded as axiomatically 

 contrary to fact, because " cyclones " are known to 

 pass eastward. We are therefore invited to conclude 

 in passing that the geostrophic idea is inadequate. 

 That is certainly a possibility but not the only one. 

 Since the geostrophic idea is based upon our experience 

 of natural distributions of pressure we might with 

 equal justice conclude that the assumed geometrical 

 distribution is a non-natural one. Or better still, we 

 might say that Mr. Richardson's preliminary canter 

 has given a rigorous dynamical explanation of what 

 is meant by " an anticyclone resisting the advance of 

 a cyclone," a very common statement of meteorological 

 phenomena. The reviewer preserves in memory two 

 natural pictures of an Atlantic cyclone kept at bay bv 

 NO. 2771, VOL. I 10] 



a current from the east and presenting an appearance 

 grotesquely like a revolving ball balanced on a water- 

 jet. A notable feature of our northern winter is a 

 vast anticyclone over Asia which dominates the 

 northern half of the eastern hemisphere like Mr. 

 Richardson's pattern, although the distribution over 

 the other quadrants of the globe is not at all like 

 the pattern. As a matter of experience the anti- 

 cyclone does frequently spread from the east over 

 England. Our weather might not inaptly be described 

 as a conflict between the effect which Mr. Richardson 

 repudiates as contrary to experience and the eastward 

 travel of cyclones which he regards as axiomatic. 

 Not infrequently, the result of the conflict is that 

 the cyclones, instead of going eastward over us, are 

 headed off to the north along the Norwegian Sea — 

 " which nobody can deny." 



In view of our inadequate knowledge of the structure 

 and circulation of the atmosphere caution in drawing 

 conclusions is always desirable, and in this case specially 

 so in the interests of justice, because the alleged failure 

 of the geostrophic principle to anticipate the changes 

 at the surface in Chapter II. reappears in Chapter VI. 

 as the record of a previous conviction, and gets 

 the prisoner another sentence for what is perhaps 

 not his fault. Mathematicians in dealing with the 

 elusive atmosphere are not infrequently inspired by 

 Jabberwocky, 



One two, one two, and through and through, 

 The vorpal blade goes snickersnack ; 



but they ought to make sure that they get the right 

 Jabberwock by the neck before " galumphing back " 

 with his head. 



Chapter III. reinforced by Chapter VII. explains 

 a suitable organisation of what are called co-ordinate 

 differences, the principles of the chess-board or lattice. 

 Chapter IV., a very important one, occupies more than 

 one-third of the whole book. It is devoted to the 

 fundamental equations and the information which is 

 necessary in order to assign numerical values for the 

 variables. It takes the form of about thirty short 

 essays on great subjects, such as the effects of eddy- 

 motion, radiation, conduction of various kinds, the 

 flow of heat to the air from the sea, or from the ground, 

 or from vegetation, the smoothing of observations, 

 and many others. Chapter V. deals with the evaluation 

 of vertical velocity, a very vital subject. Chapter VI, 

 deals with the special conditions for the stratosphere 

 and its equations. Chapter VIII. reviews the numerical 

 operations to be performed and gives the final prepara- 

 tion for Chapter IX., which provides a " full-dress 

 rehearsal " of the process of computation. By its aid 



