j; 
2 
Sees 
1876.] Plain, Prairie, and Forest. 585 
From other parts of Mr. Foster’s chapters on the origin of 
prairies, in his work already cited, it would appear, however, 
that he considers the deficiency of rain in the winter months to 
be the essential cause of the absence of forests. He remarks as 
follows:1 “ A region where the annual precipitation is slightly 
in excess of twenty inches, I infer from observation, is unfavora- 
ble to the growth of trees, even were this moisture equally dis- 
tributed, but where three fourths of it is precipitated during the 
spring and summer, the grasses flourish and mature to the exclu- 
sion of arborescent forms.” This seems also to be the theory 
advocated by Dr. Newberry, from whom we quote as follows : ? 
“Those who know anything of the climate of the prairie belt 
know that it is characterized by a deficiency of winter rain and 
snow, and by occasional though rare seasons of excessive dry- 
ness. The want of winter rains to deeply saturate the ground 
gives to the superficial hibernating grasses, which may be said to 
live upon the almost copious summer rains, an advantage over 
trees equal to a victory.” 
Let us now examine this question in the light of the Smithso- 
nian rain-tables and Mr. Schott’s discussion of them. This very 
point is taken up, under the head of Annual Fluctuation in the 
Rain-Fall, annual fluctuation, as Mr. Schott explains it, mean- 
ing the “ changes from month to month.” He gives the typical 
curve for the “ region embraced in the Hudson River Valley, 
Vermont, and Northern and Western New York,” as derived 
from an ageregate of five hundred and sixty-four years of ob- 
servation. He also gives from the records of one hundred and 
fourteen years the curve for “the Upper Mississippi from Fort 
Madison, Southern Iowa, to Fort Ripley, Central Minnesota, 
and including part of Wisconsin.” One of these regions, as will 
be readily seen, is a region of forests, the other of prairies. It 
would be difficult to select two districts in this country of equal 
area more characteristically situated for showing the difference 
between the rain-fall of a wooded and of a prairie country. Yet 
we find Mr. Schott declaring that the two curves thus obtained, 
and representing typically the distribution of the rain-fall through- 
out the year, “do not materially differ.” In each case there are 
maxima, one about the beginning of the summer and the 
other at its end, and also a principal minimum about the begin- 
mng of February. The only difference between the two types 
1 Mississippi Valley, page 101. 
2 Geology of Ohio, i. 30. 
