PHYSICAL FACTORS IN PLANT DISTRIBUTION 65 



What we have in such a system as the growth-forms of Drudc 

 is a confessedly rough but eminently natural grouping of plants 

 according to their size and form, their manner of branching, the 

 character of their foliage, the seasonal behavior of their fohage, 

 and other features which stand in intimate relation to their 

 gross physiological behavior. Some of the criteria used in sepa- 

 rating the growth-forms appear to have only a morphological 

 value, as that which places the palms in a separate category 

 from other broad-leaved evergreens; many other criteria seem 

 to lay special emphasis on the water relations, as those that 

 have to do with succulence and the aphyllous condition. We 

 are not in a position at present to know how much physiological 

 significance there may be in the differences between the mono- 

 cotyledonous stems of palms and the dicotyledonous wood of 

 other broad-leaved evergreens. Neither are we in a position to 

 appreciate how much the features of succulence and aphylly 

 may affect functions other than those of water-loss, although the 

 work of Richards to which I have referred gives some suggestion 

 of the relation of succulence to respiration. In the main the 

 criteria upon which Drude has based his system are of undoubted 

 physiological value, and many of them are of such a nature as to 

 be related to more than one set of major functions. 



In brief, the system of growth forms of Drude is a rough at- 

 tempt, on the basis of observational data, to classify plants accord- 

 ing to their ecological characteristics. This is merely another way 

 of saying that it and the other systems are the first expression 

 of our need for a thoroughgoing physiological classification of 

 plants. It is needless to say that such a grouping of plants must 

 be based on experimental work, and that it must have no bias 

 regarding the greater importance of water relations, of tempera- 

 ture conditions, of root distribution, of respiratory features, or 

 of photosynthetic behavior. To be faithful to the facts, and 

 to be useful as a tool — which is the chief value of any classifica- 

 tion — this one must be a well balanced attempt to arrange plants 

 according to a natural anatomical and physiological system, 

 even in the absence of anything akin to phylogenetic relationship 

 to serve as a guide. 



