482 ETHEL SARA BARTON ON THE 
of plurilocular sporangia. The recollection of this led me to 
examine very closely the sections of Soranthera which showed 
the stage immediately previous to the shedding, to see if in any 
way this was connected with the formation of fruit. In all the 
sections I had noticed separate, single-rowed outgrowths con- 
sisting of 6 or 7 cells arising from the same cell as the assimila- 
tive filaments (Pl. 23. fig. 6). On comparing these with the 
plurilocular sporangia of Chordariacee, notably Leathesia, I am 
led to the belief that they are possibly the plurilocular sporangia 
of Soranthera—although I have of course not seen the escape 
of the spores, all the material being preserved in spirit. It is 
hardly strange that empty sporangia have not been seen, since 
they probably drop off immediately after the escape of the 
spores. 
Another point of resemblance between these supposed pluri- 
loeular sporangia of Soranthera and those of Leathesia is that 
in both cases they grow all over the thallus and are not confined 
to sori. 
In order to make clear the further development of 5. ulvoidea, 
it is necessary to go back to the early stages of the plant and 
trace the growth of the cryptostomata. Hitherto the only 
genera in Enceliacee which have been shown to possess these 
bodies are Colpomenia and Hydroclathrus (Miss Mitchell, in Phye. 
Mem. pt. ii. 1893, p. 53); but investigation shows that they exist 
in S. ulvoidea from almost the earliest stages of the plant. 
Even in these, the young hairs of the future cryptostoma can be 
recognized, growing each from a small oval cell below the surface, 
similar to the cells which bear the assimilative filaments at the 
periphery (Pl. 23. fig. 7). The presence of the smaller cells 
deep down in the thallus shows that the origin of the cry pto- 
stomata is the same as in Colpomenia and other algee, where ig 
are formed from a single initial cell or group, the growth o 
which is arrested. The young hairs are close together among 
the surrounding tissue, but their typical basal growth and naro ' 
form enable one to distinguish the position of the future cyp M 
stoma before any sort of hollow has been found. As the p " 
grows the hairs, growing quicker than the assimilative m 
push out above the surface, and owing to their length are se*? 
io be found unbroken in sections of the plant. 
are 
In Colpomenia and Hydroclathrus the eryptostomata 
