514 REPORT OF THE COMMISSIONERS OF 



is 3,265 square miles, with about 585 square miles of water surface included in the 

 lakes and reservoirs above the dam, we have for the period of 1 1 years, from 1885 to 

 1895, inclusive, a mean rainfall of 25.06 inches. The run-offs for this period have not 

 yet been computed for inches on the watershed, but as a general statement we may 

 say 25 inciies of rain a year yields from this area from 4 to 5 inches of run-off. In 

 years of minimum rainfall, when the total is as low as about 20 inches, the run-(_)fif 

 does not exceed 3.5 inclies. The area drained by the Upper Mississippi is now mostly 

 heavy forest with a mean temperature of from 38'' to 40'' F. If the forest were 

 entirely removed there is no reason to doubt but that streams in this area, which now 

 yield some water during the entire summer season, would become entirely dry during 

 a considerable portion of the year, the same as is true of many of the tributary 

 streams of the Muskingum and Des Plaines Rivers. 



The astonishing persistency of the evaporation element for given conditions was, 

 so far as I am aware, first pointed out by Messrs. Lawes, Gilbert and Warrington in 

 their classical paper on the Amount and Composition of the Rain and Drainage 

 Waters Collected at Rothampsted, published in the Journal of the Royal Agricultural 

 Society of England, in 1831. As to why evaporation exhibits such persistency these 

 distinguished authors consider it largely due to the fact that the two principal 

 ctinditions which tletermine large evaporation, namely, excessive heat and abundant 

 rain, \cry rarely occur together ; the result is, especially in the English climate, a 

 balance of conditions unfavorable to large evaporation. In a wet season when the soil 

 is kept well supplied with water there is at the same time a more or less saturated 

 atmosphere with an absence of sunshine, while in dry seasons the scarcity of rain 

 results in great dryness of the soil with scant, slow evaporation. 



The problem of the relation of rainfall to run-off ;ind evaporation has attracted the 

 attention of meteorologists and working engineers for many years, and in P^ngland 

 and some of the other countries of Europe records have been kept as far back as fifty 

 to seventy years ago, and the results tabulated with reference to a solution of this 

 problem. Among other interesting data, a large amount of which has been obtained 

 in England, we may refer to percolation data as derived from the use of drain guages. 

 In the paper by Messrs. Lawes, Gilbert and Warrington already referred to we have a 

 record of rainfall and the percolation through the Rothamsted drain gauges from 1870 

 to 1890. Space will not permit of describing these interesting experiments in this 

 place, but those interested in them may find the whole matter in detail in the original 

 paper of the Journal of the Royal Agricultural Society of England, already referred to, 

 or an abstract may be found in my second report on the survey of the Upper Hudson 

 Valley, included in the annual report of the State Engineer and Surveyor of the State 

 of New York for the year 1896, and about to be published. At Rothampstetl, where 



