xi LEPTOSPORANGIATJE HETEROSPOREJE 435 



form. Before this curvature is very pronounced, however, in 

 the sporophyll, a protuberance arises upon its inner face, a short 

 distance above the base (Fig. 254, A). This originates from a 

 single cell, which functions for some time as an apical cell, and 

 causes the young sporocarp to project strongly from the leaf, of 

 which it is simply a branch, somewhat analogous to the spike in 

 Ophioglossum. It may, perhaps, be better compared to a fertile 

 leaf segment of Ancimia, as it has been shown by Johnson (2), 

 that the mother cell of the young sporocarp arises from the 

 margin and not from the face of the leaf. 



It has at first the form of a blunt cone, but soon upon the 

 side turned toward the leaf a slight prominence appears (Fig. 

 254, B, L), and about the same time two similar lateral ones are 

 formed. As in the sterile part of the leaf growth is stronger 

 on the outside, and the young sporocarp bends in toward the 

 leaf, so that the position of fertile and sterile segments is very 

 like that in the young sporophyll of Ophioglossum. The apex 

 of the sporocarp rudiment, together with the three lobes, en- 

 close a slightly depressed area, which becomes the top of the 

 sporocarp. The four prominences (including the original 

 apex of the fertile segment) are beyond question to be consid- 

 ered leaflets, which remain confluent except at the top. A little 

 later a slight depression or pit forms at the base of each lobe 

 and the central area at the top. These pits are separated later- 

 ally by the coherent edges of the leaflets, which extend to the 

 axis of the sporocarp and are continuous with it. As the 

 young fruit enlarges, the depressions deepen owing to the 

 elongation of both leaflets and the axial tissue, which forms a 

 sort of central columella (Fig. 254, D). Thus are formed 

 four deep cavities, separated laterally by the united margins of 

 the leaflets, and corresponding to the much more numerous 

 "canals" described by Russow and Johnson in the fruit of 

 Marsilia; like these they at first open at the summit by a pore, 

 and a study of longitudinal sections shows clearly their strictly 

 external origin. 



From his study of P. globulifera, Johnson (2) concludes 

 that all four lobes of the sporocarp are of lateral origin. He 

 was able to trace the origin of each sorus to a single marginal 

 cell in each of the four segments of the young sporocarp. Sec- 

 tions of the young sporocarp of Marsilia at this stage (John- 

 son (i), Figs. 22, 23) resemble to an extraordinary degree 



