420 MISS E. 8. BARTON ON NOTHEIA ANOMALA. 
cell, thus forming a layer which lines the cavity. Each of these 
cells produces a hair in the young eryptostoma, thus forming a 
thick tuft of long hairs which protrude through the mouth and 
are plainly visible to the naked eye in the young branches of 
the plant (PI. 12. fig. 5). 
In some of the cryptostomata, however, a cell which lies at 
the base of the depression, instead of produeing a hair, grows up 
into a small round protuberance, and after division proceeds to 
cut off from the apex successive cells to form a young branch 
(Pl. 13. fig. 6). These apical cells are exactly similar to those 
of the primary thallus, as would be expected. As growth in 
length takes place, the lower cells begin to divide up to form the 
various layers of tissue, and this transverse division keeps pace 
with a corresponding division of the cryptostoma-cell, which 
originally gave rise to the branch. So that before the young 
branch is sufficiently advanced to protrude through the opening 
of the cryptostoma its base has become quite broad, and gives 
the appearance of having originated from a group of cells rather 
than from a single cell (Pl. 13. fig. 7). The remaining cells of 
the eryptostoma continue to produce hairs as usual. 
In Dr. Gruber's description of the young cryptostoma and 
the origin of the branch he says that when the hairs fall off a 
flask-shaped cell is left at the base, which probably gives rise to 
the young branch. But the hairs do not fall off in the very 
early stages of the cryptostoma, and it is in these stages that 
the first sign of the branch is seen. There seems to me no 
doubt that the branch arises direct from one of the cells of the 
lining layer. 
Since the branching takes place in this way from the erypto- 
stomata, it follows that branches arise from all sides of the 
thallus irregularly ; indeed I have seen in one section three 
eryptostomata with a branch growing out of each. In one 
instance of an old eryptostoma I have seen the centre occupied 
by the base of a thick branch and at the side there was beginning 
to shoot up another small branch, showing that one cry ptostoma 
can produce more than one branch, and that these need not 
necessarily spring from the centre. I have failed, however, to 
find a ease in whieh two branches have succeeded in growing to 
maturity from the same eryptostoma. 
All this time the cells lining the cavity of the eryptostoma 
have continued to divide, as well as the thallus-cells surrounding 
