174 
Proceedings of the Royal Society of Edinburgh. [Sess. 
VII, — The Dynamics of Cyclones and Anticyclones. Part III. 
By John Aitken, LL.D., F.R.S. 
(MS. received December 22, 1915. Read January 24, 1916.) 
In 1900* I communicated to this Society a paper on the above subject. 
Since that date a great deal of information has been obtained by means 
of free balloons carrying instruments which recorded the temperature, 
humidity, and pressure of the air up to great elevations. Much of this 
new knowledge seems to contradict our previous ideas, and does not seem 
to fit into the old convectional theory that cyclones are formed by the 
rising of the hot, moist air from the surface of the earth; their energy 
being due to their temperature and to the heat liberated by the condensa- 
tion of the water vapour in them. We are told by those who have studied 
the bearing of the new knowledge on our atmospheric circulation that 
the old theory is “utterly untenable.” Their reasons for this conclusion 
are, first, that the recent investigations show that the air is colder in 
cyclones than in anticyclones; second, that the isothermal layer is lower 
than the mean over cyclones, while it is higher than the mean over anti- 
cyclones. At first sight these discoveries seem to shatter the convectional 
theory, but before we come to any conclusion I should like to present 
certain facts which it appears to me will require to be considered before 
we scrap our old ideas. 
Let us take the first objection to the convectional theory, namely, the 
lower temperature in cyclones. Now, while temperature plays a most 
important part, it is not the only factor in determining the circulation ; 
the pressure has also much to do with the density of the air. The 
air in the cyclone is under less pressure than the air in the anti- 
cyclone ; it has therefore less weight per unit of volume. Air expanded 
by reduction of pressure tends to rise just as air that has been expanded 
by heat. 
Let us now see how far the expansion due to reduction of pressure 
affects the question of convection. In the Journal of the Scottish Meteoro- 
logical Society for 1914 there is a paper on cyclones and anticyclones 
by W. H. Dines, F.R.S., in which there is an excellent abstract of these 
upper-air observations, given in tabular form. From that table I find 
1 Trans. Roy. Soc. Edin., vol. xl, part i (No. 7). 
