PRESIDENTIAL ADDRESS. 



LXIII 



We are now in the position to consider the theories of atmo- 

 spheric circulation. The starting point is naturally the differen- 

 tial heating of the equator as compared with the poles. In 

 Fig. 2, the effect of this heating is illustrated in the case of the 



® 





- Ongmdi surface of 

 2 <?'/■ columns of 

 equal temperature 





-«'*.*'<*' 



■IS. I 



Kkj. 2. 



t,wo masses of air, the one at the equator, the other on the poleward 

 side of it. Both of them, it may be assumed, in the first in- 

 stance, exert precisely similar pressure on the earth's surface, and 

 have similar height, but as the result subsequently of the differen- 

 tial heating effect of the sun, the equatorial mass expands to a 

 greater extent than the mass on the poleward side of it. So far, 

 however, there is no apparent reason why there should be any 

 circulation from the base of one column of air to the base of the 

 other. The greater elongation of the equatorial column has no 

 more increased its weight than the weight of a telescope is in- 

 creased by being pulled out to its full length after it has been 

 fehut up. But now the factor of gravity comes into play. Gravity 

 tends to pull mobile substances towards the centre of gravity in 

 such a way that surfaces of those substances become tangential to 

 the direction of pull. It will be seen that in the two columns of 

 air a and h of Fig. 2, the plane touching their surfaces is no longer 

 a tangent to either of the columns. The equatorial column is 

 practically top heavy, and under the influence of gravity it sways 

 over, and flows on either side of the equator in a poleward direc- 

 tion. As soon as the flow sets in from the top of the higher 



