THE GAMETOPHYTE GENERATION OF THE PSILOTACE^E. 
97 
then penetrating the cell walls and entering the cell below. This was especially 
noticeable in older rhizoids whose apices had become injured. Fig. 5 is a representa- 
tion of a few superficial cells of the prothallus bearing such rhizoids. Here it will 
be seen that the fungus traverses the entire length of the rhizoids and enters the 
internal cells of the prothallus. It seems highly improbable that the direction of 
growth of these hyphee is away from the food supply in the living prothallial cells. 
It seems more probable that it is directed towards the host cells, and that they have 
entered the apex of the rhizoid from the surrounding soil. These infected rhizoids, 
showing the great length of the fungal hyphse, offered excellent conditions for 
observing the latter and their non-septate nature. The infection of the fungus is not 
limited in its distribution to certain definite zones, as in Lycopodium* The infection 
is fairly uniform throughout. In its saprophytic habit and its subterranean habitat 
the prothallus of Tmesipteris does recall that of Ijycopodium, but structurally there 
is no real resemblance. 
The Antheridia. 
The number and distribution of the antheridia seemed to vary considerably in 
the numerous prothalli examined. There was no evidence to show that they were 
confined to any particular region ; on the contrary, they were found on every region 
of the surface. They were, however, almost invariably accompanied by archegonia. 
The distance between these two reproductive bodies was never very great. In some 
specimens there were only two or three antheridia present, but in others they were 
quite numerous, as many as forty having been counted on a single prothallus. In 
the latter cases they were very closely set together and projected out as minute 
spherical beads from the surface. The antheridium is much larger than the arche- 
gonium, as indicated in figs. 1 and 2. They are nearly spherical in form and extend 
out quite conspicuously from the surface. In this respect the antheridia of Tmesip- 
teris are in great contrast to those of Lycopodium, t Equisetum, t Ophioglossum, t 
and Botrychium, t in which the antheridium invariably develops below the surface of 
the prothallus. 
In regard to the development of the antheridium in Tmesipteris an interesting 
and fairly complete series of stages was obtained. From the early stages it would 
seem that the antheridium begins as a superficial cell. A very young stage is shown 
in fig. 7. This is from a median section, showing three wall cells which envelop a 
large inner cell, whose nucleus has just undergone division. The two nuclei thus 
formed are very large, and their chromatin stains deeply. They stand out very 
conspicuously as compared with the nuclei of the wall cells and other vegetative 
prothallial cells. The cytoplasm surrounding them is also very dense and granular. 
In fig. 8 we have represented a median section of a young antheridium slightly older 
* Bruchmann, H., l . c . 
t Bruchmann, H., l . c . ; Campbell, D. H., 1913 ; Jeffrey, E. C., 1898. 
