CORRELATION OF DISTRIBUTIONAL FEATURES. 529 



charts, to the exoteric reader, tend in the same direction, to retard real 

 progress. Ecological students should realize that this is not by any 

 means a closed subject, but that it is in a very early, formative stage, 

 and that it requires vastly more critical and original study than has 

 ever been accorded it. Merriam's work formed an excellent beginning 

 and he opened up a new and very important field, but his presentation 

 of the matter was hurried and incomplete, and the later work of his 

 followers has consisted in drawing zonal boundaries on a finer scale on 

 biological evidence without any effort to extend the investigation of 

 the climatic basis of the scheme. The work of Merriam should be 

 regarded as a beginning and the whole field opened by him assuredly 

 deserves elaborate critical study at the hands of ecologists. We do 

 not wish to attempt to substitute any other dogmatic scheme of cli- 

 matic provinces in place of this one, but we do wish to emphasize the 

 fact that some other and much better scheme is to be expected when 

 this subject receives attention such as it deserves. 



V. CONDITIONS THAT PROBABLY DETERMINE THE DISTRIBUTION 

 OF GROWTH-FORMS AND THE ECOLOGICAL DISTRIBUTION OF 

 INDIVIDUAL SPECIES. 



1. GROWTH-FORMS. 



The fundamental vegetational data for use in any study of climatic 

 conditions in relation to the relative abundance of a particular growth- 

 form should be based on a knowledge of the role played by this growth- 

 form in the vegetation of the region involved. Such knowledge is in 

 hand for certain small areas, but is lacking for the great bulk of our 

 region. We have therefore fallen back upon the best obtainable sub- 

 stitute, namely, the securing of distributional data for each of the 

 various species belonging to these growth-forms, and the plotting of 

 their cumulative distribution. 



The distribution of all evergreen broad-leaved trees and of all 

 microphyllous trees of the United States have been superposed so as to 

 show the regions in which these growth-forms are represented by the 

 greatest number of species. The resulting maps doubtless come near 

 to showing the relative importance of these forms in the vegetation, 

 as well as their numerical representation in the flora. The distribu- 

 tion of the eastern deciduous trees has been treated by superposing the 

 ranges of a group of the most common and widespread species, rather 

 than by an attempt to use all of the verj^ numerous trees of this growth- 

 form. The resulting area does not coincide with the Deciduous Forest 

 region of plates 1 and 2, but it comprises the region in which deciduous 

 trees are known to reach their maximum abundance, size, rate of 

 growth, and speed of reproduction, as well as the regions in which 

 they are abundant and successful in the optimum habitats. 



