534 LIFE-HISTORIES 



bearing sporangia would be a spore-sac-leaf, while the cylindrical 

 elongating zone from which it arose would now be a true stem. 

 Here would be about as simple a fern as we can imagine; but it 

 would have all the essential features, and it is not inconceivable 

 that higher forms might have been evolved from it. Suppose, for 

 instance, that the sadeaf member forked into two branches, and 

 let one of them be expanded so as to secure as much sunlight as 

 possible and be devoted exclusively to photosynthesis, while the 

 other branch instead of doing much food-making was narrower and 

 developed as many spores as possible from foocl that the expanded 

 branch furnished. Suppose further that the stem lived on from 

 year to year, sending new roots into the earth and new leaves into 

 the air, then our plant would have become like an adder-tongue 

 fern. 



The striking differences between liverworts and ferns of any 

 kind have so imjoressed not a few botanists as to have made them 

 doubt the likelihood of ferns having originated in the manner above 

 suggested; and this doubt has gained strength from the fact that 

 the most ancient fossil ferns are of highly complex organization, 

 being often tree-like in form, and so even less like liverworts than 

 the presumably degenerate ferns with which we are most familiar 

 to-day. Moreover, if modern liverworts are also to be regarded 

 as degenerate plants — a view, as we have seen, for which there is 

 some evidence — the gap which separates them from ferns is even 

 wider. It may well be true that ferns evolved directly from sea- 

 weeds in which a clearly marked alternation of generations had 

 developed as in certain rather highly organized red alga^ living 

 to-day. On this supposition, however, we are still left with the 

 difficulty of imagining the stages through which a seaweed could 

 pass in fitting itself for life on land as a tree. Here fossils cannot 

 help us, for we have none at all intermediate between seaweeds 

 and ferns. Since, however, there are undoubted fundamental re- 

 semblances between a Coleochsete, an Anthoceros, and an Ophio- 

 glossum, these may offer at least a possible clue as to how the great 

 changes in question may have taken place. 



Grape-ferns would be readily derivable from adder-tongues by 

 further branching of the two leaf branches, which in the fertile or 

 sporangial segment might result in each sporangium being borne 

 on a little stalk or branchlet of its own. We may well imagine that 

 wonderful possibilities of development lay before such a type as 

 this as soon as it established itself on the edges of swamps or on land 

 where food and moisture abounded. It could then afford to delay 

 the production of spores imtil it had built a thick, tall stem, by 

 means of leaves made larger and larger year after year and devoted 

 entirely to making food so that a surplus might be stored in the stem. 

 Finally, a very large number of sporangia might be produced upon 

 much-branched spore-sac-leaves; and these, held high in the air, 

 could scatter their spores most effectively. 



