200 NINTH ANNUAL REPORT OP 



this country. We have abeady alluded to the frequency of south and 

 southwest winds over the United States in summer. These currents, 

 derived, as we have said, from the trade winds of the tropics, contain 

 a large amount of moisture, and are the source of the great fertility of 

 the Mississippi valley. Extensive rains and thunder storms only occur 

 after the south wind has prevailed for some time. I am inclined to 

 believe that this vast aerial current, which flows below the upper west- 

 erly current, is much deeper in the southern than in the northern States. 

 Mr. Phelps's observations, near the mouth of the Rio Grande, show that 

 in his district the thunder clouds drift in the southeast current ; and I 

 have been informed that they follow the same direction in Louisiana 

 and Alabama. But in the higher latitudes, as for example that of 

 Washington, the thunder clouds all move in the direction of the upper 

 current from a point or two north or south of west. The moisture 

 which forms these clouds is no doubt derived from the lower stratum, 

 whose upper surface is exposed to the erosion of the upper current, 

 constantly sweeping a portion away towards the Atlantic. The action 

 of the lower moist current rising into the upper and forming cloud, 

 and perhaps part of the upper descending, is no doubt the principal 

 propelling power of the surface wind. President Dwight gives a very 

 good description of the thunder storms of New England, and refers to 

 points which should be specially borne in mind. On one occasion he 

 says : " The meridional line upon which I stood was crossed by the 

 storm several miles to the south. During the whole day the wind had 

 blown from the south, and continued to blow in the same direction on 

 the surface throughout the afternoon, without a moment's intermission. 

 But had the wind," says he, " which carried the cloud when it passed 

 over the meridian swept the surface, the wind for a time, at least, must 

 have been entirely stopped. This, however, was not the fact even for 

 a moment." On another occasion, in 1809, a "thunder storm passed 

 over New Haven, from the northwest, with great rapidit3^ It con- 

 tinued, as I judge, from an hour to an hour and a half But though 

 the clouds moved rapidly to the southeast, a soutlmest wind blew the 

 whole of that day and widle the thunder storm ivas overhead icith great 

 violence.'*'' 



The increase in the strength of the southwest wind, as the clouds 

 drifted overhead from the northwest, shows that the lower and upper 

 currents mixed together ; that as in the case of the squall mentioned 

 by Humboldt, a part of the lower stratum of air rising formed cloud, 

 and that a part of the upper stratum descending to the earth was car- 

 ried away by the prevaihng surface wind from the southwest. This 

 propelling power, I think, cannot be doubted in thunder storms ; but I 

 imagine the same principle must be taken into account as a source of 

 motion to the southwest surface winds whenever clouds form in the sky 

 and drift in the upper current. Thunder squalls drifting from the west, 

 as I am also informed by Professor Espy, often draw in the stratum of 

 air from the east immediately above the earth's surface ; while the 

 falling rain under the clouds forces the air outwards towards the east, 

 and it rises immediately before the rain, carrying up dust and other 

 materials. A slight westerly breeze then usually springs up and fol- 

 lows the cloud just a little in the rear. 



In fact, from the very irregular nature of the winds of the lower 



