upon a plant influence it otherwise than merely through evapor- 

 ation, chiefly by stimulating or retarding its activities. 



The region here under discussion furnishes abundant con- 

 firmation of the writer's earlier conclusions concerning the cause 

 of the treelessness of the prairies, — indeed it presents the prob- 

 lem in a form of unusual interest. For that reason some effort 

 has been made to work it out in greater detail, though only the 

 more general conclusions can be presented within the limits of 

 this paper. 



In the first place we find the prairie here again upon surfaces 

 most exposed to the summer winds and afternoon sun, while the 

 groves are found upon the sheltered abrapt slopes, chiefly in 

 close proximity to the lakes and rivers, as has been noted. It 

 would appear that if the chief cause of the prairie is to be found 

 in the evaporating agencies that make for drouth there should 

 be a preponderance of forest in the vicinity of the numerous 

 lakes of this region, for surely there is much vapor rising from 

 these water surfaces, and the increased humidity of the air should 

 make tree growth possible. And there is a greater amount of 

 forest here than would be ordinarily found or expected in a re- 

 gion of such comparatively slight and gradual variation in 

 topogi-aphy. To a person traveling on West Okoboji lake, for 

 example, it must appear that a large part of the region is for- 

 ested, especially on the east side of the lake. Yet the forest forms 

 a mere fringe, for the total forest area, while no doubt increased 

 by the more favorable atmospheric conditions, is after all much 

 less than that of the prairie, — a fact which is strikingly brought 

 out by the map, on which the native forest areas are represented 

 to scale as nearly as possible. At first sight a puzzling feature 

 of this region is found in the fact that in many places the prai- 

 rie extends almost to the water's edge. But an analysis of the 

 entire region reveals a striking consistency in the distribution of 

 forest and prairie along the lake and stream bodies, and more- 

 over that distribution decidedly confirms the conclusions hereto- 

 fore reached. 



The location of the groves in this region on, or in the shelter 

 of, abrupt slopes has already been emphasized. These slopes 

 may be either on the leeward or the windward side with refer- 

 ence to the prevailing southerly and southwesterly winds of 

 summer. Where they are on the leeward side, as at the Labora- 



16 



