REPRODUCTION OF THE ZYGNEMACEJ. 435 
means certain, the discovery of the wide distribution of partheno- 
genesis in the vegetable kingdom in classes of plants where the 
normal mode of reproduction is a much higher form of the sexual 
materially weakens the force of his argument. 
lt is worthy of note, as confirming the view here indicated, 
that I could frequently detect a slight difference between the 
protuberances put out by the male and female cells, that from 
the female cell being shorter, but of larger diameter. The divi- 
ding septum is therefore, as a rule, in the half ofthe conjugating 
tube nearest to the female filament. The protoplasm, in passing 
through the connecting tube, appears to press down this dividing 
septum, as shown in fig. 2, so as to make it concave, and causing 
the narrower protuberance to fit into the broader one after the 
fashion of a ball-and-socket joint. During and after conjugation 
the connecting tube gives the impression, in the ordinary lateral 
view, as if it were still divided by a septum; this is due to the 
perforation leaving an annular portion of the septum attached to 
the sides of the tube, like a collar, as may be seen on viewing such 
an orifice from above (see fig. 3). 
Fig. 3. 
Fig. 3. Spirogyra porticalis. Conjugating tube, seen from above. x 165. 
Fig. 4. S. porticalis. Zygospore germinating while still in mother-cell. X 165. 
On two points not connected with the process of conjugation 
l wish to say afew words. The filaments of the Zygnemacew are 
always described and drawn as unbranched. Thisis by no means 
invariably the case. It is not uncommon for a cell to branch ; 
but I have never seen such a branch cut off by a septum. As 
far as I have observed, this tendency to branch is confined to 
non-conjugating cells of female filaments (fig. 1). De Bary (l. c. 
pl. iii. fig. 15) observes the same with respect to Zygnema and 
Mougeotia; and states that here the branch is not only cut off 
by a septum, but is itself septated. 
On the second point I appear to be at issue with previous ob- 
servers. In a well-known drawing of the germination of a zygo- 
2n 2 
