400 MOSSES AND FERNS CHAP. 



upper part of the prothallium as the antheridium. In spite of 

 his statement that the development of the male prothallium has 

 little in common with the true Filices, his figures of Azolla are 

 extraordinarily like the simple male prothallia that sometimes 

 occur among the Polypodiacese. 



In my earlier studies of the male gametophyte, the second 

 prothallial cell (Fig. 1234, #), described by Belajeff, was over- 

 looked, but subsequent examination of my preparations showed 

 that it was present. 



The subsequent divisions correspond to BelajefFs account. 

 In the middle cell of the antheridium two nearly vertical walls 

 are formed, which with the top cell (cover cell) completely 

 enclose the central one. The cover cell recalls in form and 

 position the same cell in the antheridium of the Polypodiacese, 

 but is formed here previous to the separation of the central cell. 

 In one of the lateral cells a horizontal wall is formed, so that 

 the sperm cells are surrounded by five parietal ones. The cen- 

 tral cell now divides by a median vertical wall, and each of the 

 daughter cells twice more, so that eight sperm cells are formed, 

 as in Salvinia. The prothallium remains embedded in the sub- 

 stance of the massula, and the spermatozoids probably escape 

 by the softening of the outer part of the latter. In Salvinia 

 the prothallia project beyond the sporangium wall, and are 

 easily detached. 



The antheridium of the Salviniaceae does not closely re- 

 semble that of any other group. Azolla differs less from the 

 homosporous Ferns in this particular, and shows some resem- 

 blance to the Hymenophyllaceae in the arrangement of the 

 parietal cells. Occasionally a triangular opercular cell occurs 

 in Azolla, which recalls that in Osmunda. 



The -Female Prothallium 



The macrospores of Azolla filiculoides are borne singly in 

 the sporangia. The spores only germinate after they have 

 been set free by the decay of the indusium, the upper part of 

 which, however, persists as a sort of cap. The decay of the 

 sporangium wall and indusium exposes the curious tuberculate 

 epispore, with its filamentous appendages, which serve to hold 

 the massulse, which are firmly anchored to them by their 

 peculiar hairs (glochidia) with their hooked tips. This is evi- 



