﻿and Fluctuations. 441 



tions. In both hemispheres the normal west wind blows in ac- 

 cordance with this law ; and the low barometer to the left or 

 right, as the case may be, does not necessarily imply an inde- 

 pendent rotatory motion of the air. In the same way the Trade- 

 Winds of the Atlantic pass by the low pressure of the Doldrums 

 according to the law, but without inclining towards or circling 

 round it ; and though it would seem highly probable that this 

 low pressure adds strength to the Trades in some parts of the 

 ocean, it evidently does not produce any tendency to that cen- 

 tripetal movement the theory of which is now so much in 

 favour. As I believe that the whirls of air, when they do occur, 

 are due mainly to geographical peculiarities, in which I would 

 include the characteristics of the prevailing winds, so also I be- 

 lieve that the direction in which the whirl turns is guided 

 entirely by the geographical conditions under which it is formed ; 

 but the discussion of this point is foreign to my present sub- 

 ject. It is, however, worth noticing that if, in any part of the 

 world where the west winds blow, a whirl is formed which re- 

 volves the wrong way, the air, which in the first instance sup- 

 plies it, is carried from a higher latitude to a lower, and must, 

 at least sometimes, when thrown into an area of central calm, 

 have both its temperature and humidity increased. I have shown 

 how, on a very large scale, the system of winds on the equato- 

 rial side of each oceanic basin thus revolves the wrong way 

 round a centre of high pressure; and it would seem possible 

 that other centres of high pressure round which the winds oc- 

 casionally circle, in what Mr. Galton has aptly designated 

 " anticyclones," are formed in the same way. We cannot, 

 however, lay much stress on this possibility ; for since increase 

 of temperature increases evaporation, and increased evaporation 

 again lowers the temperature, the vagaries shown by the ther- 

 mometer are frequently extremely puzzling ; and the difficulty 

 of establishing a satisfactory comparison between observations 

 made in different places, at different hours, and at different 

 heights above the sea-level has not yet been overcome. And it 

 must be borne in mind that if the place of relative low pressure 

 lies to the left, the place of relative high pressure must lie to 

 the right ; and the appearance of an anticyclone may be shown 

 on the chart by the near approach of two, or occasionally even 

 three centres of depression*. 



The conclusion, then, to which this examination into the phe- 

 nomena of cyclonic storms leads us is, that the centre of low 

 pressure is formed, not by any meteorological agency, but by 

 the circular rush of a body of air according to the known prin- 



* For an example of this, see the Chart for March 28, 1868 (Morning), 

 in the Atlas des Tempetes. 



