562 



NATURE 



[October 13, 1923 



of a type of artificial cell. The dendrograph is also 

 applied further in the study of the growth of trees 

 and minute changes in the volume of the trunk. 



The ecological work includes a study by Dr. Forrest 

 Shreve of the factors influencing the altitudinal dis- 

 tribution of vegetation in Arizona, various observa- 

 tions by Dr. VV. S. Cooper on the strand vegetation of 

 the Californian coast, and on the endemic conifers of 

 the Monterey peninsula. l>r. W. A. Cannon reports 



an atmometer experiment to test the evaj» 

 power of the air m the Karroo and other \<- 

 South Africa, and a study of the transpiring t>ov\. 

 of various Karroo plants, including species of Alo- 

 Gasteria, Cotyledon, and Protea. A si' 

 of observations on Welwitschia near h 

 shows that its transpiring power is very luss . i u. 

 reports of many of these lines of research are either ii 

 preparation or will be published later. K R (i. 



The Mechanics of a Cyclone.' 



A NOTEWORTHY attack is made by Dr. V. H. 

 -^^~*- Ryd, in the paper referred to below, on the 

 problem of the circulation of the air in a cyclone, the 

 source of the energy necessary for its maintenance, 

 and the disposal of the rising air. The cyclone con- 

 sidered is necessarily " ideal," with circular isobars, 

 the pressure at any distance from the centre being 

 determined by an arbitrary formula which gives good 

 results for the body of the cyclone but causes a dis- 

 continuity at the boundary. 



The first part of the paper, entitled " Preliminaries," 

 leads to a series of fundamental differential equations 

 which cannot be integrated, and in the second part 

 a graphical method of solution is developed, by which 

 the air-paths appropriate to the pressure distribution 

 are constructed. This method is next applied to the 

 more important case in which a uniform pressure 

 gradient is superposed on the original circular isobars. 

 The results are shown both as actual paths of air 

 particles and as a synchronous representation of 

 actual wind such as we see in a daily weather chart. 

 In both cases the presentation appears to be in good 

 agreement with Nature. From the air-paths so con- 

 structed it is easy to compute the variation of the 

 surface area occupied by any given mass of air, and 

 consequently the regions of rising and falling air. 

 This is done for the surface, in F"ig. 33, which is re- 

 produced here (Fig. i). The figures represent the per- 

 centage value by which the area changes in an hour, the 

 broken lines referring to contractions {i.e. rising air), 

 and the full lines to dilatations. The region of falling 

 air in the north-west quadrant is of great interest 

 and receives an important verification, with which the 

 author was apparently not acquainted, in a chart 

 showing the distribution of weather with reference to 

 the centre of a depression which crossed England in 

 November 1915 (A. E, M. Geddes in Q.J.R. Meteor. 

 Soc, 43, 1917, p. 15). 



The third part of the paper applies the results so 

 obtained to the construction of a picture of the mechan- 

 ism of travelling cyclones, supported by an actual 

 numerical example. The ordinary temperature dis- 

 tribution results at great heights in a system of open, 

 nearly parallel isobars corresponding with winds of 

 great velocity at the cirrus level ; this is termed the 

 stationary system, and from the nature of its origin 

 it extends with decreasing intensity nearly to ground 

 level. Hence a travelling cyclone can be divided 

 into four parts, namely, the ground stratum ; the 

 lower stratum of the free atmosphere in which the 

 velocity of the wind arising from the stationary 

 system is less than the speed of the cyclone ; the 

 central part of the cyclone, in which these two are 

 equal ; and the higher stratum, in which the velocity 

 of the wind from the stationary system is the greater. 

 The resulting pressure distributions, wind velocities, 

 and vertical motions in these layers are studied, and 

 combined in a description of the circulation of the air. 



' Publikationer fra det Danske Meteorologiske Institut. 

 5 : Meteorological Problems, i : Travelling Cyclones. 

 Pp. iv + i24- (KjBbenhavn: G. E. C. Gad, 1923.) 



NO. 2815, VOL. I 12] 



Meddelelser Nr. 

 By v. H. Ryd. 



The conclusion is that the air which is thrust up i!i 

 the portion of the cyclone with negative coel; 

 cannot escape until it reaches the highest s 

 referred to above ; in this stratum it is cam 

 ward out of the system. The same conclusion 

 to the descending air, which is sucked from t; 

 higher stratum to the ground level. Hence tli' 

 " stationary system " provides the energy of tin- 

 cyclone, and the author considers that in niost cases 



Fig. I. — Variations of areas in a travelling depression. 



the decay of an Atlantic cyclone is due to the dying 

 out or disturbance of the stationar>' pressure field — a 

 conclusion which requires further elucidation. In the 

 last chapter attention is briefly directed to the agree- 

 ment of these theoretical results with actual observa- 

 tions of the direction and speed of cyclones and the 

 vertical and horizontal distribution of temperature, 

 including the existence of a " cold front," which is 

 thus shown to be a consequence, and not a cause, of 

 the formation of a cyclone. 



The paper is highly mathematical in treatment ; 

 this is, of course, necessan,' in a scientific account of 

 new work, but it is unfortunate as being likely to 

 deter the reader without a high mathematical equip- 

 ment, although actually much of it can be read 

 without mathematics. In view of the great interest 

 at present taken in the " polar front " theorj' of 

 cyclones and the importance of this vindication of 

 the older view, it is to be hoped that the author will 

 shortly present us with a more popular account, 

 including more illustrations from Nature. It would 

 have been better to have avoided attaching two 

 different meanings to the symbol R, even though no 

 confusion is caused thereby. 



