THE 

 LONDON, EDINBURGH, and DUBLIN 



PHILOSOPHICAL MAGAZINE 



AND 



JOURNAL OF SCIENCE. 



[SIXTH SERIES.] 



JANUARY 1919. 



I. On Travelling Atmospheric Disturbances. 

 By Harold Jeffreys, M.A., D.Sc* 



IT has been shown by Lamb t that an arbitrary disturbance 

 of the uniform distribution of density in the atmosphere 

 in horizontal layers would ordinarily give rise to a motion 

 of the nature of a wave spreading out from the originally 

 disturbed region, the velocity of propagation being of the 

 same order of magnitude as that of sound. Thus a local 

 variation in mass distribution would be rapidly dispersed 

 over a wide area and the original uniform state restored. 

 Any disturbance of even moderately permanent character 

 must therefore be of a very special type. Now the ordinary 

 cyclone is able to retain its size and pressure distribution 

 about its centre for days, the velocity with which it moves 

 being of the order of twenty feet per second, very much 

 smaller than the speed. of an atmospheric wave. Thus the 

 first question with regard to the nature of a cyclone is, why 

 does it not spread out like an ordinary wave and disperse in 

 an hour or two ? The answer seems to be that a cyclone is 

 of the character of a standing wave, the pressure and velocity 

 distribution being such that for this peculiar kind of dis- 

 turbance the velocity of propagation is practically zero. 

 The realization of this fact has led to the assumption of the 

 well-known " Gradient relation," according to which the 



* Communicated by the Author. 



t "On Atmospheric Oscillations/' Proc. Roy. Soc. lxxxiv. A. pp. 551- 

 572 (1910). 



Phil. Mag. S. 6. Vol. 37. No. 217. Jan. 1919. B 





