198 DISCUSSION OF THE DESICCATION QUESTION. 



regions, over which precipitation no longer has any features of regularity, 

 biat may be expected at any season of the year. These regions are the areas 

 of constant rains, so called. As to the way in which precipitation would 

 take place over the temperate and polar regions if the earth were entirely 

 covered by water, it is not easy to form a precise idea. That it would be 

 moderate in amount, as compared with that of the tropical regions, there can 

 be no doubt, and that it would be distributed over the year, and not limited 

 to one season, is also clear ; moreover, there is every reason to believe that 

 it would not be concentrated over special districts, leaving others nearly 

 rainless, as is the case at present. Exactly how much variation there would 

 be in the rain-fiiU, as we approached the immediate vicinity of the pole, it is 

 not easy to say; but that there would be considerable increase in the amount 

 it is reasonable to assume, as Avill be explained farther on. 



In point of fact, what would be the normal distribution of precipitation 

 if the earth's surface were entirely covered by water is entirely changed in 

 character by the land masses, wliich produce this effect by their form and 

 size, as well as by the manner in which the lowlands, the plateaus, and the 

 mountain ranges which make up the continents are situated with respect to 

 the oceans and to each other. 



Statistical tables of the rain-fall of various stations throuirhout the world 

 show that outside of the tropics the regions wdiere the precipitation is large 

 are very limited in area and pecidiarly arranged Avith reference to the land 

 masses; to this statement may also be added another, — which, in fact, is 

 almost a corollary of the other, — to the effect that the areas of small rain- 

 fall are very large, and evidently closely dependent for their position on the 

 form and size of the continents. An inspection of the rain-chart* will illus- 

 trate these assertions better than can be done in words. Some explana- 

 tion of these peculiarities may be here permitted, since the subject of the 



* On Wojeikof's rain-chnrt of the earth the areas where the precipitation exceeds 1,200 millimeters (a little 

 over forty-seven inches) are designated by fine lines. The regions thus indicated occupy but a very small space 

 outside of the tropics. In North America, for instance, there is a belt along the coast of the United States extend- 

 ing north as far as Chesapeake Bay ; another in the Mississippi Valley reaching not quite to the southern edge of 

 Lake Michigan (including the prairie region, often considered as treeless on account of deficiency of moisture); and 

 finally, another very narrow strip along the Pacific Coast, extending from the northern edge of California north 

 to the parallel of 60°. In South America there is only a small area in the Brazilian Coast Mountains, reaching no 

 farther soutli than the parallel of 30° ; and another narrow belt on the southwest side of the southern end of Chili 

 and along the coast of Patagonia. In Asia the smallness of the aiea of even moderately large precipitation out- 

 side of the tropics is still more striking, since only a part of China, along the valley of the Yang-tze, and the 

 southeastiTn side of the .Japanese Islands, is thus designated. 



