112 The Botanical Gazette. [March 
peared to be an antherozoid at its tip, but the trichogyne was 
so very delicate and the antherozoid so small that the relation 
between the two structures could not be satisfactorily studied. 
Even if such stages as are shown in fig. 3 were common their 
small size seems to the writer to preclude the possibility of 
determining what takes place at the time the antherozoid is 
applied to the trichogyne. 
The single cell or each of the two cells of the procarpic 
branch below the carpogonium contains a single well defined 
thallus cell (fig. 3 and 4, t) to which the pro- 
them is multinucleate. It is probably always united to neigh 
boring thallus cells by strands of protoplasm, although such 
connections may not always appear in sections which are net- 
essarily cut in a single plane. The attachment to the inter- 
nal filament is usually by a broad strand of protoplasm, an 
not infrequently one of the bulb cells found along the internal 
filaments occurs opposite this point (fig. 3, 4). ; 
The trichogyne withers quickly and the upper portion dis 
appears very rapidly. The swollen portion at the base (cat 
pogonium) remains somewhat longer (figs. 5 and 6) but takes 
the stain very faintly and finally disappears, and the lowet 
cells of the procarpic branch are left attached to the thallus 
cell. The condition of the procarp is then either that show! 
in fig. 7 or fig. 8; that is, it consists of one or two cells, 
probably containing as a rule only a single nucleus, attach? 
to a thallus cell which is multinucleate. At this time tht 
At this point we had best consider the changes which talt 
place in the tissue around the cells of the procarp after 
directly around that which bears the procarpic branch grae® 
ally assume a different character. The protoplasm become 
very much denser and the nuclei increase in number and 
larger in diameter. These thallus cells are evidently w# 
Hauptfleisch has called the auxiliary cells. There are 
