July 7, 1904] 



NA TURE 



225 



THE MECHANICS OF THE ATMOSPHERE.^ 



THE motion of the atmosphere at any time is 

 admitted to be so complicated that any approach 

 to a workable representation of it must necessarily 

 be by steps. The motion at any time must be re- 

 garded as a temporary' divergence from the average 

 motion, and the question naturally arises, What is the 

 nature of the average state of motion about which the 



actual state of motion fluctuates? We may approach 

 the solution of this question in either of two ways ; 

 we may find out what the motion actually is or we 

 may find what the forces are which, so far as we can 

 tell, cause the motion, and trust to our knowledge of 

 dynamics to compute the average 

 motion from the average forces. 

 As regards the latter method, it 

 may be said that the dynamics of 

 an elastic fluid moving on a 

 rotating spheroid, however interest- 

 ing, is beset with an extraordinary 

 number of temptations to error, and 

 the more humble ambition of try- 

 ing to find out W'hat the motion 

 really is, although painfully labo- 

 rious, has advantages which may bo 

 compared with the advantages 

 which walking has as compared 

 with the use of a flying machine. 



In the early 'seventies of the last 

 century. Clerk Maxwell set a 

 question in a Cambridge examin- 

 ation to which I owe the inspira- 

 tion of a number of lectures and 

 examination questions. It was 

 this : — " Show how by observ- 

 ations of the motion of a body the 

 resultant force acting upon it may 

 be determined," and he added the 

 luminous rider (I quote from 

 memory), " A fish weighing lo lb. 

 swims through the water with a 

 uniform velocity of lo miles per hour, always in the 

 same direction ; find the resultant action of the water 

 on the fish." As soon as one begins to think of 

 answering these questions, and in particular of apply- 

 ing them to the relation between the controlling forces 

 of pressure and the motion of the atmosphere, one 



1 iJ.-ised upon a paper on the " General Circulation of the Atmosphere in 

 Middle and Higher Latitudes," read before the Royal Society on June -■ 

 by Dr. W. N. Shaw, F.R..S. 



NO. 1810, VOL. 70] 



realises what pitfalls await the unwary. The most 

 obvious remark in relation to the first question is that 

 the motion at any instant tells us absolutely nothing 

 whatever about the forces acting. Unless observ- 

 ations sufficient to determine the change of motion 

 have been dealt with, nothing about the cause of 

 motion is known. Yet, in spite of this rudimentary 

 fact of dynamics, obvious enough when it is stated, I 

 cannot help wondering how many 

 students of elementary dynamics 

 ever really get rid of the notion that 

 if you find a body moving in a 

 certain direction you must look for 

 a force in that direction too ; we are 

 surrounded with examples to the 

 contrary, but the study _ of 

 dynamics, being mainly deductive, 

 usually passes them by. 



In meteorology it is impossible 

 to avoid the consciousness of 

 temptation to the converse error of 

 expecting to find the motion of air 

 in the direction of the recognised 

 forces. The most obvious force is 

 that due to pressure, and who can 

 resist the temptation of thinking 

 that the flow of air from a high- 

 pressure area to a low-pressure area 

 must be the dominant feature of 

 atmospheric motion ? Yet the one 

 great inductive statement in con- 

 nection with meteorology, Buys 

 Ballot's law, warns us that if we 

 trust to the direction of forces to 

 indicate the direction of motion we shall certainly be 

 misled. Motion along isobars, perpendicular to the 

 gradient, is a closer representation of the actual state 

 of things than motion along the gradient, along, that 

 is to say, the direction of resultant forces. 



ncURE 2 . ISOBARS AT THE LEVEL OF 4000 METRES FOR JANUARY . 

 HANN* Re PROOUCTION OF THEORtClNAt DIAGRAM BY TBSStBENCOEOORT. 



There is no doubt that if we could arrest for a time 

 the motion of the atmosphere, without altering the 

 pressure, and let the air start again from rest, the 

 direction of initial motion would be along the pressure 

 gradients from high to low, but we have to deal with 

 an atmosphere that has been moving for countless 

 ages, and all that existing forces do is to maintain 

 or disturb the average, or steady motion ; if "in those 



