MAGNETISM AND CIRCULATION OF THE ATMOSPHERE 115 



It rains more in the valley drained by that river than is evapo- 

 rated from it again. The difference for a year is the volume of 

 water annually discharged by that river into the sea (^S 117). 



At the time and place that the vapor which supplies this im- 

 mense volume of water was hfted by the atmosphere up from the 

 sea, the thermometer, we may infer, stood higher than it did at 

 the time and place where this vapor w^as condensed and fell down 

 as rain in the Mississippi Valley. 



210. I looked to the south for the springs in the Atlantic which 

 supply the fountains of this river with rain. But I could not find 

 spare evaporating surface enough for it, in the first place ; and if 

 the vapor, I could not find the winds which would convey it to the 

 right place. 



The prevailing winds in the Caribbean Sea and southern parts 

 of the Gulf of Mexico are the northeast trade-winds. They have 

 their offices to perform in the river basins of tropical America, and 

 the rains which they may discharge into the Mississippi Valley 

 now and then are exceptions, not the rule. 



211. The winds from the north can not bring vapors from the 

 great lakes to make rains for the Mississippi, for two reasons : 

 1st. The basin of the great lakes receives from the atmosphere 

 more water in the shape of rain than they give back in the shape 

 of vapor. The St. Lawrence River carries off the excess. 2d. 

 The mean climate of the lake country is colder than that of the 

 Mississippi Valley, and therefore, as a general rule, the tempera- 

 ture of the Mississippi Valley is unfavorable for condensing vapor 

 from that quarter. 



212. It can not come from the Atlantic, because the greater 

 part of the Mississippi Valley is to the windward of the Atlantic. 

 The winds that blow across this ocean go to Europe with their 

 vapors ; and in the Pacific, from the parallels of California down 

 to the equator, the direction of the wind at the surface is from, 

 not toward the basin of the Mississippi. Therefore it seemed to 

 be established with some degree of probability, or, if that expres- 

 sion be too strong, with •something like apparent plausibility, that 

 the rain winds of the Mississippi Valley do not, as a general rule, 

 get their vapors from the North Atlantic Ocean, nor from the Gulf 

 of Mexico, nor from the great lakes, nor from that part of the Pa- 

 cific Ocean over which the northeast trade-winds prevail. 



