358 POPULAR LECTURES AND ADDRESSES. 



The result is about 6 gramme 1 - water- centigrade 

 units of heat per annum. Now it would take 

 about 79 such units to melt a centimetre of ice 

 per annum. So that is all that underground 

 heat would do ; in other words, it would go 

 for nothing in respect to retarding the increase 

 of the ice-cap. This shows that if we were to 

 have this change of temperature and this double 

 snowfall for several hundred years, there would 

 be a very sensible addition to the quantity of 

 ice, and a very sensible depression on the water 

 elsewhere. But if the Antarctic ice-cap were to 

 be greatly increased it would lower the water, 

 diminish the circulation, and tend to cool the 

 Arctic Ocean. Therefore, from this considera- 

 tion alone, we should expect glaciation in the 

 northern hemisphere simultaneously with an aug- 

 mentation of the Antarctic ice-cap. 



But by far the most potent influence for altering 

 the climate in an)- part of the world is oceanic 

 circulation. The sea is the great carrier of heavy 

 goods. Our atmosphere of air, with its pressure 

 of fifteen pounds per square inch, corresponds to 



