and Heiminthostachys zeylanica. 39 
those of the other Ophioglossaceae,it projects very slightly from 
the general surface. The large oval mass of spermatocytes 
is surrounded by a layer of flattened cells, derived from the 
cells around that which gave origin to the antheridium. The 
true wall of the antheridium, derived from the outer segment 
of the mother-cell, is, however, different in some respects from 
that of any other known antheridium. It is shown in section 
in Figs. 54 and 56, and in surface view in Fig. 55. For the 
greater part of its extent it is two-layered, a periclinal division 
having taken place in most of the cells, into which it had 
divided by anticlinal walls. A few cells, however, remain 
undivided, so that at these spots the wall is one-layered. 
These single cells are large and prominent both in section 
and surface view, and also stand out in contrast to the others 
by reason of their more deeply staining contents. It is by 
the breaking down of one of these cells that the dehiscence of 
the antheridium takes place. Rarely a single dehiscence-cell 
is found in a nearly central position ; usually two, three, or 
four exist, situated close to the periphery of the antheridial 
wall. When old antheridia are looked at from the surface, all 
these cells appear darker brown, but only one is found to have 
broken down to give exit to the spermatozoids. The latter 
were unfortunately not examined alive, and, as in the case of 
Ophioglossitm pendulum , the examination of those in the anthe- 
ridium has only shown a general agreement in form with those 
of other Ophioglossaceae : it was not found possible to deter- 
mine the number of cilia. 
All the female prothalli which were found had young plants 
or, in the case of two or three, large embryos. A satisfactory 
study of the archegonium was thus impossible, and the fol- 
lowing description is based on the unfertilized archegonia 
borne on the prothalli with the youngest embryos. As fer- 
tilization appears in all cases to arrest the growth of the 
prothallus, the development of the archegonium could not 
be followed. Fig. 57 represents an archegonium in median 
section. The neck projects considerably from the surface of 
the prothallus : the projecting portion consists of four rows 
