XII LEPTOSPORANGIAT.E HETEROSPORE^ 383 



The macrospores germinate most promptly in the early 

 autumn, and in California, where this species is abundant, this is 

 probably the natural time for germination. As the first stages 

 of germination take place within the completely closed spore, 

 it is difficult to tell precisely just when it begins. So nearly as 

 could be determined, the first division may take place within 

 two or three days, and the whole development be completed 

 within a week. 



A section of the ripe spore, still within the sporangium, 

 shows its contents to be nearly uniform, and much like that of 

 Isoetes. The nucleus is here at the apex of the spore cavity 

 and not conspicuous. It is somewhat elongated and stains but 

 little. No nucleolus can be seen. 



The first sign of germination is an increase in the size 

 of the nucleus, which becomes nearly globular, and a small 

 nucleolus becomes evident. At the same time the cytoplasm 

 about it becomes free from large granules and indicates the 

 position of the mother cell of the prothallium. This upper part 

 of the spore cavity is now cut off by a nearly straight transverse 

 wall, and this small lenticular cell becomes the prothallium. 

 The granules in its cytoplasm are finer than those in the large 

 basal cell, and the nucleus stains strongly and shows a large 

 nucleolus. The nucleus of the lower cell remains in the upper 

 part, and is much like that of the prothallial cell. 



The first division wall in the upper cell is vertical and 

 divides it into two cells of unequal size. In a prothallium 

 having but three cells, the second was also vertical, but in 

 others it looked as if it were horizontal, which Prantl ^ states is 

 the case in Salvinia. From the upper of the cells formed by 

 the first horizontal wall the first archegonium arises. If the 

 horizontal wall forms early, the primary archegonium is nearly 

 central, but if two vertical walls precede it, its position is 

 nearer the side opposite the first cell cut off. In the few cases 

 where successful cross-sections of the very young prothallium 

 were made, the archegonium mother cell was decidedly triangu- 

 lar, showing that it was formed by three intersecting walls, as 

 in Isoetes. It divides into an outer and inner cell, the latter, 

 as in Isoetes^ giving rise at once to egg and canal cells, without 

 the formation of a basal cell. 



Up to this point the exospore remains intact ; the central 



^ Prantl (4), p. 427. 



