572 MOSSES AMD FERNS chap. 



In the few Ferns in which perennial prothallia are formed, e. g., 

 Gymnogramme -triangularis, G. (Anogramme) leptophylla, the 

 behaviour of the gametophyte is precisely the same as in the 

 Liverworts. 



Coulter has suggested that the determining factor in the 

 development of the leafy sporophyte has been photosynthesis or 

 "chlorophyll work." He sees no reason why such a structure 

 as the leafy sporophyte may not have arisen non-sexually in 

 response to the need for increased chlorophyll activity, quite 

 apart from the production of spores. The spores would find 

 more favourable conditions upon a leafy shoot than upon the 

 thallus. 



It is doubtless true that the production of a large leafy 

 shoot would be advantageous in increasing the output of spores ; 

 but why this leafy shoot should not have developed gradually 

 from the sexually produced sporophyte of some bryophytic 

 prototype, as there is the strongest evidence that it has done, 

 is not made clear. The development upon the leaves of the 

 sporophyte of spores of the same type as those of the lower 

 Archegoniates is entirely comprehensible if it is admittted that 

 the sporophyte of the Fern is descended from the leafless sporo- 

 phyte of some ancestral Bryophyte; but it is very hard to 

 explain if we assume that there is no genetic connection between 

 the spores of Bryophytes and Pteridophytes. 



According to Coulter's hypothesis, the leafy sporophyte 

 originated by budding comparable to that of the leafy shoot of 

 a Moss from the protonema, or the apogamously produced spo- 

 rophyte of a Fern. The leaves were originally purely vegeta- 

 tive organs, and the development of sporangia was secondary. 

 The germination of the asexual spores and the zygote are 

 assumed to have been the same, each giving rise to a thallus 

 upon which arose secondarily the leafy shoot. 



If such were really the course of development, it is strange 

 that no trace of the thallus-stage has persisted in the embryo- 

 sporophyte. The only structure which could possibly be so 

 interpreted is the suspensor in Lycopodium and Selaginella, 

 which most morphologists would hesitate to consider of such 

 nature. 



The statement (Coulter (i), p. 56), "Perhaps such a tend- 

 ency {i. e., the elimination of the thallus portion of the zygote 

 product) is no more difficult to understand than the fact that 



