THE WEATHER, AND WEATHER PROPHETS. I53 



it with little obliquity in tropical regions, while their 

 incidence on those near the poles is always very oblique, 

 and during the half of each year null; it is obvious that 

 its surface must be very unequally warmed. The cook, 

 to use a homely illustration, knows full well that, how- 

 ever good her fire, the two ends of her joint will be 

 under-roasted when the middle is done brown ; unless 

 slie apply a couple of concave reflectors on her spit to 

 throw some of the lateral heat upon them. As a matter 

 of fact, no one needs to be told that it is so ; and that 

 the intertropical regions of the globe are very hot, and 

 the polar, habitually very cold. The average annual 

 temperature at the equator is about 84 Fahr., while in 

 the colder regions near the North Pole it is as low as 

 5 Fahr., or 27 below freezing. The diflference would 

 be much greater were there no sea, or even were the 

 whole surface initially moist soil. Whatever that initial 

 moisture, it would soon dry ^^from the warmer portions, 

 to settle down in snow or hoarfrost on the colder; after 

 which the dried portions would grow hotter and hotter. 

 Every one knows what a cooling power there is in the 

 evaporation of water. So long as a vestige of moisture 

 were present, the temperature of the soil could never, at 

 at all events exceed, however it might fall short of, that 

 of boiling water : but when once completely dried off, 

 there would no longer be a limit to the possible increase 

 of temperature ; since there would then be no circulation 

 or return of moisture to the part once dried. How this 

 circulation is kept up under the existing circumstances 

 is what we must now explain : and first of all how it 



