THE QUEST OF PHYLETIC LINES 101 



first place it was assumed that similarity of vegetative texture 

 implies a common origin, allowance not being made for the possibil- 

 ity of parallel development. It is now widely held that the filmy 

 condition of the Hymenophyllaceae is a result of specialisation to 

 a moist habitat, which finds its correlative in many distinct phy- 

 letic lines. Secondly, it was assumed that the parallel between 

 sorus and sporogonium is a real one, though there is an entire 

 absence of any intermediate steps to show that such a transfor- 

 mation as that assumed ever took place. Such a demonstration 

 was especially necessary where the transformation which was as- 

 sumed was so far reaching and fundamental. It is true that the 

 relative positions of the parts named were in correspondence, 

 and that their functions were also alike. But the presumed 

 change involved the origin of the sporangia, which in Ferns are 

 always superficial and of very specialised structure, from the spore- 

 sac which is of internal origin, and shows no sign of similarity to 

 the Fern sporangia beyond the production of spores. To render 

 the suggestion even plausible some structural indication of the 

 steps of the transition in allied forms was essentially necessary. 

 But this evidence was not brought. The hypothesis failed through 

 the absence of any chain of intermediate conditions between the 

 two widely divergent types compared. The modern view is of 

 course that there is no direct phyletic relationship between the 

 Hymenophyllaceae and the Mosses : and that the similarities, such 

 as they are, are the result of adaptation of two distinct spore- 

 bearing phyla to conditions of life which have been similar in 

 their leading traits. 



A third example is seen in the anatomical comparison between 

 the stele of certain species of Lycopodimn, such as L. clavatum, 

 and the polystelic state seen in certain species of Selaginella, such 

 as S. Wildenovii. It was suggested, and widely held until quite 

 recent times, that the more or less distinctly separate xylem-plates 

 of the former resulted from the lateral fusion of steles originally 

 separate as in the latter type. As long as the comparison is based 

 upon isolated transverse sections, the suggestion appears reason- 

 able enough. But the time is gone past for such comparisons as 

 these, and conclusions must be tested upon a much broader basis 

 of fact and of reasoning. 



