78 PLANT MORPHOLOGY 



truth and clearness which should be the aim. The choice is open 

 to us either to make the terminology strictly phyletic throughout, 

 which would certainly be cumbrous, though it would reflect the 

 true position; or, putting phyletic distinctions in the background, 

 to use terms in a more or less comprehensive sense, even grouping 

 together things which we know to have been distinct in phyletic 

 origin. Such a comprehensive sense is conveyed by the expression 

 "homology of organization," which, as Goebel points out, "has 

 only to do with phylogeny in so far as it recognizes a common ca- 

 pacity for development derivable from undifferentiated ancestors" 

 (Organographie, Eng. ed., p. 19). This is indeed a collective 

 term for the results of parallel development; it suffers from the 

 danger of suggesting some ideal type or pattern towards which 

 evolution has tended. 



For my own part I think it matters little what our terminology 

 be, or what the separation of categories of parts, provided we attach 

 clear meanings to the words we use, and select those words as natu- 

 rally conveying those meanings. For instance, if we fully realize 

 that the word "leaf" is used in a sense which is not phylogenetic, 

 but merely descriptive of those lateral appendages on the shoot 

 which are produced exogenously, and in acropetal order, then let 

 it remain, ranking as an expression of "homology of organization." 

 But the appendages thus included may for clearness be conveniently 

 divided into "phyllomes" on the sporophyte, and "phylloids" on 

 the gametophyte, as, indeed, I suggested some years ago. Never- 

 theless, these again are not phyletic unities: they include parts 

 with distinct histories, which have already been recognized in the 

 gametophyte, while for the sporophyte a more advanced state of 

 the science will probably provide definitions. Meanwhile we consent 

 to a compromise in grouping these together: but the only condition 

 upon which this can be safely done is the clear knowledge that this 

 is a compromise by which we secure a certain convenience of de- 

 scription. Moreover, the acceptance of this compromise must not 

 be understood to grant free license to argue from one to another of 

 the forms included, as though they were equivalents: what has re- 

 sulted in one line of descent can at best only throw a side-light on 

 what has happened in another distinct line: and in proportion as the 

 lines involved in a comparison are more remote from one another, 

 their comparison assumes more and more the character of a mere 

 analogy. The danger which our compromise brings with it is that 

 this will not be clearly kept in mind. At all hazards the strict phy- 

 letic view should underlie all present morphological discussion, 

 notwithstanding that, for mere convenience, that view may not be 

 clearly reflected in the classification of the parts. This makes me 

 hope that the compromise is only a temporary concession, and that 



