382 BOTANICAL GAZETTE [NOVEMBER 
were seen sprouting and sending their tubes through the sporangium 
wall, as is normally the case in Alpanes. 
Figs. 4-7 represent the normal arrangement of the oogonia; the 
hypogynal tube will be noticed in fig. 4.. In countings from my cultures, 
Fics. 4-7.—Fig. 4, simple oogonial branch terminating in an oogonium; fig. 5» 
oogonial filament with one branch; fig. 6, same with two branches; fig. 7, 4 branched 
oogonial filament with terminal oogonium containing four oospores; all X335- 
this tube was present only in about one-sixth of the oogonia, but it 
appears in all cultures and is one of the most distinctive characters of the 
species. The appearance of an antheridial tube, not from an antherid- 
pw, ium, but from the vegetative coenocyte, is a most 
singular occurrence, and would seem to be 4 de- 
generate condition induced by the suppression of 
fertilization. I could find no evidence of fertiliza- 
tion even in cases where a tube was found. In not 
a single case was an antheridium seen below the 
oogonium. 
The arrangement shown in fig. 6, suggesting the 
three balls of a pawnbroker’s shop, occurs so often 
and is so striking as to be one of the best diagnostic 
fic. a cs characters of the species. The lower ae “i 
al oogonium with two tYPically recurved (figs. 5,6), but not always \38- 
projections on surface, Out of many thousands of oogonia seen, a very 
X335- few had one or two short rounded outgrowths from 
