1915-16.] The Dynamics of Cyclones and Anticyclones. 1 77 
energy from potential to actual and back again to potential gives the 
explanation of the lowering of the isothermal layer over cyclones, and its 
rising over anticyclones. 
The increase in velocity of cyclonic air also shows itself in other ways 
when passing from higher to lower pressures. As the air in cyclones is 
rising it is at the same time passing from higher to lower pressure, and in 
so doing is gaining velocity. This seems to account for the increase in 
velocity with height observed in cyclones. Up to the height of the cirro- 
cumulus clouds its centrifugal force is not sufficient to hold back the higher 
pressure surrounding it; but by the time it has risen above these clouds 
its velocity is so great that its tangential force enables it to overcome the 
surrounding pressure and to flow outwards towards the high-pressure area. 
The increase of velocity with height in cyclones is not simply a ques- 
tion of the retardation of the lower layers by friction with the earth’s 
surface, as the increase in velocity goes on up to heights far above that to 
which we would expect the earth’s surface to have any influence. Of 
course, all this expansion is accompanied by a loss of temperature ; but, as 
the observations show, the loss of volume from this cause is more than 
compensated for by the fall in pressure. 
There is another point to which I should like to call attention. If the 
table of the temperatures at the different elevations in the upper air, already 
referred to, is examined, it will be seen that, while the temperatures in the 
lower part of the cyclone differ from those in the anticyclone, at the ele- 
vation of 10 kilometres they have the same temperature ; and that, while 
the temperature over the cyclone ceases to fall, that over the anticyclone 
continues to fall to a considerable amount. One naturally asks, Why this 
difference ? I have already pointed out why the stratosphere should be 
lower over the cyclone than elsewhere ; but why should the air over the 
anticyclone at higher elevations than the cyclone be colder than that over 
the cyclone ? If I may be allowed to hazard an explanation, I would 
suggest that the top of the anticyclone is formed of the air which has 
escaped from the upper part of the cyclone and flowed towards the high* 
pressure area, and in so doing has been forced upwards — its energy of motion 
being converted into potential energy of elevation — and at the same time 
has expanded and fallen in temperature; the top of the anticyclone being 
thus in a certain sense the top of the cyclone. According to the generally 
received opinion, the air which rises in the cyclone comes down in the anti- 
cyclone, and this suggestion as to the passage of the air between the two 
systems seems to explain the process. 
The lower temperature at the top of the anticyclone and its greater 
VOL. xxxvi. 12 
