6 THE VEGETATION OF THE UNITED STATES. 



observational experiences, and, indeed, whether such agreement 

 would actually yield substantial advantages. 



The classification of vegetations is thus seen to be by no means an 

 easy task, and it is no marvel that there is lack of agreement among 

 those interested in it. Plant communities are usually made up of many 

 species, and these species are usually of distinct floristic relationship, 

 or dissimilar geographic range, and of varied physiological require- 

 ments and behavior. Furthermore, a particular community does not 

 range far without the acquisition of new members and the loss of old 

 ones. The classification of such a complex material requires the adop- 

 tion of arbitrary standards, and consequently leads to an unnatural 

 system or else to one which is only of local appHcability. 



Physiographic ecology affords a genetic basis for the classification of 

 vegetation, and the logic of such a system has much to recommend it. 

 Physiographic criteria are not, however, of universal applicability for 

 the classing of vegetation, and lead to unnatural interpretations of 

 vegetational phenomena in many regions, particularly in the tropics, 

 in deserts, and in regions with diversified geological and soil conditions. 

 Furthermore, physiographic ecology is incHned to lead to a false sense 

 of satisfaction with the assembled results of the study of a series of 

 successions, and has often failed to stimulate a study of the physical 

 causes which underlie the observed successional phenomena. 



Several Scandinavian and German botanists have rightly urged 

 that the classification of plant communities should be based upon the 

 recognition of certain distinctive types or forms of plants. These 

 "biological types," "life-forms," "vegetation forms" or "growth- 

 forms" are recognized without regard to phylogenetic relationships, 

 and serve to distinguish such groups as deciduous broad-leaved trees, 

 evergreen coniferous trees, perennial grasses, sclerophyllous shrubs, 

 etc. Several systems of growth-forms have been proposed, each more 

 elaborate and complete than the next preceding. In so far as these 

 systems represent an attempt at a physiological classification of plants 

 they are highly commendable, and must lie at the foundation of physi- 

 ological plant geography. 



Unfortunately, however, our knowledge of the physiology of plants 

 is chiefly based on the behavior of a small number of kinds of plants, 

 mostly cultivated forms growing under sets of conditions at best only 

 partially controlled or measured, and the existing classifications of 

 growth-forms have been based on the inference that the gross anatomi- 

 cal features of plants are an index of their major physiological char- 

 acteristics. We have not yet secured enough evidence to test this 

 i nference in more than a general way. We know that there are marked 

 differences between the annual march of transpiration and photo- 

 synthesis in evergreen conifers and in deciduous broad-leaved trees, 

 and that there is no contradictory significance in the fact that such 



