578 CORRELATION OF DISTRIBUTIONAL FEATURES. 



5. TWO-DIMENSIONAL CLIMATIC PROVINCES. 



Although a number of the various vegetational areas shown on the 

 charts of plates 2 to 33 have been shown to be more or less precisely 

 comparable with geographically corresponding climatic areas, it never- 

 theless appears that such satisfactory correspondence is the exception 

 rather than the rule. The climatic conditions concomitant with the 

 vegetation areas that fail to show such simple correlations as have been 

 mentioned in the preceding paragraphs require a more complex mode of 

 description, at least until the proper simple climatic indices may be 

 discovered. The most thoroughgoing subdivision of the country 

 into climatic provinces, which has been attempted in our studies, 

 is that based on the two-dimensional provinces. The use of these 

 smaller climatic areas makes it possible to describe any vegetation 

 area not simply correlated with either moisture or temperature prov- 

 inces alone, in terms of both moisture and temperature conditions 

 together. Such a description is clearly climatic and may lead to further 

 correlations, but this method soon encounters limitations, since, as 

 has been pointed out, there are frequently several two-dimensional 

 provinces with the same dimensions or characteristics, and these can 

 not as yet be simply distinguished on a climatic basis alone. For the 

 present, and in the comparisons mentioned below for illustration, it 

 seems best to fall back upon geographical terms, in order to distinguish 

 such climatically similar but geographically distinct areas. This 

 method frankly begs the entire question of correlations; it furnishes, 

 wherever it is employed, nothing more than a geographical description 

 of the details of configuration with which it deals. It is, however, not 

 to be resorted to until the climatic description of the vegetational area 

 in question is as complete as possible, so that the resulting description 

 always bears much more climatic information than would a purely 

 geographic description. The latter sort of description is quite useless, 

 as far as our purposes are concerned, for it merely states that the given 

 plant or vegetation type occurs where it is. 



In the following paragraph we present descriptions of several vege- 

 tational areas, following the method just outlined. For the two- 

 dimensional climatic provinces we shall here employ only the chart 

 formed from the length of the average frostless season and from the 

 precipitation-evaporation ratio (fig. 19). The cases considered are 

 set forth here simply to illustrate the use of this method of interpreta- 

 tion; we have not yet proceeded far enough with this more complicated 

 aspect of the subject to be able to arrive at any very promising general- 

 ization. The distribution areas chosen for discussion here are taken 

 from eastern forms among those that fail to show intelligible relations 

 to the simple climatic provinces of moisture and temperature. 



In the case of the evergreen broad-leaved and microphyllous trees 

 (plate 3), attention has been called to the fact that the distribution of 



