570 CORRELATION OF DISTRIBUTIONAL FEATURES. 



Among such species of northern transcontinental range we find 

 chiefly herbaceous and shrubby plants of the evergreen forests, while 

 among those of southern range we find herbaceous plants of facultative 

 seasonal habits, or else palustrine and aquatic forms. We see, there- 

 fore, in each group a set of circumstances which apparently tend to 

 equalize the moisture conditions for these plants — the northern species 

 are subordinate associates of the evergreen forests ; the southern species 

 are active in different portions of the frostless season, according to the 

 seasonal distribution of rainfall, or else they occupy perpetually moist 

 situations. 



VI. CORRELATION OF VEGETATIONAL AREAS WITH GENERALIZED 



CLIMATIC PROVINCES. 



I. INTRODUCTORY. 



The generalized climatic provinces roughly defined from the results 

 of our climatic studies should be of value in comparing the geographic 

 distribution of vegetation with the distribution of different degrees of 

 intensity and duration of climatic conditions throughout the country, 

 and we have therefore carried out such comparisons between the vegeta- 

 tion charts (plates 1 to 33) and the generahzed climatic graphs of 

 figures 18 to 28. Of course, it is not to be expected that any vegeta- 

 tional area will be found to correspond perfectly with any climatic 

 area. Probably the only method by which close areal correlations 

 may be attained lies in the employment of several climatic conditions, 

 as in our two-dimensional systems of climatic provinces. A number of 

 cases have been discovered, however, in which the correspondence 

 between areas of plant distribution and simple climatic provinces is 

 very good. The mention of these will be valuable in the formation of a 

 conception of what sort of plants may be expected to occur in the 

 various climatic provinces. Owing to the complexity of the conditions 

 to be compared and to the varying degrees of precision with which the 

 climatic zones can now be defined — as well as to our own limitations, 

 no doubt — these correlations are but preliminary and very tentative. 



The first observation to be made in beginning these comparisons is 

 one that might have been expected on general grounds, namely, that 

 relatively few of the vegetational areas show any pronounced corre- 

 spondence with any single climatic province, defined by whatever 

 method. The second observation is perhaps a little surprising, after 

 all, considering how crude are our climatic charts, namely, that a num- 

 ber of good agreements have been found. As has been previously 

 mentioned, it appears that correlations between moisture provinces 

 and vegetational areas are more frequent than those between tempera- 

 ture provinces and the same areas. This may perhaps be due to the 

 fact that the range of moisture conditions in the United States is very 

 great (from very arid to very humid), while the range of temperature 



