262 LIGHT SCIENCE FOR LEISURE HOURS. 



inhabitants resemble the unfortunates described by 

 Dante, doomed 



' a soffrir tormenti e caldi e geli.' 



However, I shall deal first with this feature mean 

 annual temperature as affording a starting-point from 

 which to proceed to other considerations. 



If the surface of the earth were perfectly uniform, or 

 symmetrically distributed into districts of land and 

 water arranged in zones along latitude-parallels, and if 

 the strata of the soil were throughout of like density, 

 radiating power, and elevation, the lines of equal mean 

 temperature would be parallels of latitude. This hypo- 

 thetical condition of things is, we know, very far from 

 representing the true condition of the earth's surface. 

 Land and water are distributed in a manner which 

 hardly presents the semblance of law ; elevations and 

 depressions, not merely of areas of limited extent, but 

 of whole countries, are exhibited in each hemisphere ; 

 and endless diversities of soil, contour, and distribution, 

 disturb that mathematical uniformity and exactness, 

 which could alone produce the co-ordination of climates 

 under latitude-parallels. 



It is to Humboldt that we owe the valuable propo- 

 sition that maps of the world should exhibit parallels 

 of heat, as well as latitude-parallels ; and no atlas is 

 now considered complete without maps in which iso- 

 therms, or lines of equal mean annual temperature, 

 isochimenals or lines of equal winter heat, and iso- 

 therals or lines of equal heat in summer, are exhibited. 



