228 TABLES AND RESULTS OF THE PRECIPITATION, 



The observed succession of annual amounts of rain-fiill on the Atlantic coast from 

 Maine to Virginia, and in the interior of the State of New York, seems to be 

 governed by the same circumstances or laws, as is evident by a comparison of the 

 curves Type I and Type II ; and indeed Type III, region of the upper Mississippi 

 (Iowa), bears some resemblance to these curves, but is not yet sufficiently de- 

 veloped to be pronounced identical in character. The curve of Type I points to 

 a gradual increase of the average annual precipitation from about 1818 to the pre- 

 sent time; whether or not we have reached the maximum of this long fluctuation 

 or secular change cannot yet be determined, but it is probable that we have passed 

 it. Besides this greater change, which extends over several decades, and which 

 undoubtedly is of a periodic character, there are smaller undulations covering a 

 period of but a few years; these latter waves are superposed upon the larger one, 

 and appear of variable parameter, those near the minimum of the greater period 

 being greatest, and decreasing gradually to the present time. As our records 

 increase in extent this general character of the rain-fall will no doubt receive some 

 modification; the above description answers, however, to its aspect within the 

 period now available for discussion. 



The variations in the rain-fall, from year to year, in the Ohio Valley, and as far 

 west as Missouri, are difl'erent from those just noticed, though they are not of an 

 opposite character. Type IV indicates no secular increase, and the secondary 

 waves appear larger than in Type I; the remarkable period of droughts about 1836, 

 as well as the less conspicuous or relative one about 1855, are common to the two 

 regions. A comparison with Type VII of the Mississippi delta and Alabama 

 . leads to the inference that the law of succession of dry and wet years along the 

 region of the Ohio partakes largely of the character of that on the Gulf coast. On 

 our southern Atlantic coast the distribution is different from any of the above 

 types ; the rain-fall here seems to have been on the decrease (see Type VIII). Type 

 IX, for the coast of California, is too limited in extent to be analyzed. 



The wavelike irregularities presented in Diagram 8 were compared with a curve 

 representing the state of the sun's annual average activity in the production of 

 spots, a phenomenon which may possibly have some indirect connection with the 

 variations in the annual rain-fall. The decided minimum in precipitation about 

 the years 1837-8 corresponds to the decided maximum of solar disturbance at this 

 epoch; but the comparison of the two phenomena about the period 1854-5 leads 

 to an opposite conclusion. This last epoch is one of minimum rain-fall, as well as 

 of a minimum in tlie number of spots occurring in 1855-6. The two curves ap- 

 pear, of course, to coincide for some years about these epochs, yet it is plain that 

 either there is no such connection between the two phenomena, as has b.een sup- 

 posed, or else the accidental and local irregularities in the rain-fall are not suffi- 

 ciently eliminated to allow of the recognition of the law regulating the secular 

 changes. 



Taking a general view of the results of this investigation, it must be admitted 

 that, while they exhibit, in a concise form, the broader features of the laws of dis- 

 tribution of rain in the United States, they must still be regarded as but first 



