580 CORRELATION OF DISTRIBUTIONAL FEATURES. 



medium temperature conditions and with the more humid conditions 

 of the semihumid moisture province. It does not extend to the bound- 

 ary of the humid province at any point. 



Silphium laciniatum (plate 25) shows a clearly outlined geographical 

 area of distribution, occupying the Missouri-Mississippi-Ohio Valley 

 as far north as the Grand River in Michigan, as far west as western 

 Kansas, and as far east as the Appalachian Mountains. It does not 

 occur either in the West or east of the Appalachians. Within these 

 geographical limits the distributional area of this plant corresponds to 

 the following two-dimensional climatic provinces: 



(a) Warmer two-thirds of the cool semihumid, west of Appalachians. 



(6) Small portion of cool semiarid, the warmer, more humid part of this 

 province. 



(c) Medium semihumid, west of Appalachians. 



(d) Eastern (more humid) half of medium semiarid. 



(e) Most of warm semihumid (all but a small area in Georgia). 

 (/) Coolest portion of warm semiarid (Texas). 



Ig) Western half of warm humid (Louisiana to Florida). 



This Silphium appears not to extend into the very warm temperature 

 province to any considerable extent. It occupies the more arid part of 

 the semihumid and the more humid part of the semiarid, within the 

 warm and medium provinces and the warmer part of the cool province. 

 Many other examples might be given showing the use of two-dimen- 

 sional climatic provinces, supplemented by geographical data, in 

 climatically describing vegetational areas for purposes of comparison. 

 Indeed, any vegetational area may be so described after the requisite 

 two-dimensional chart has been once prepared. But the four cases 

 considered above should be sufficient to demonstrate the investiga- 

 tional value of this general method. If the relations holding between 

 climatic conditions and plant activity receive the attention that they 

 deserve from ecologists and climatologists, this method, with improve- 

 ments, should prove very useful. Especially should this be true for 

 studies of agricultural and forest climatology, which is just beginning 

 to attract serious attention in this country. 



