274 Evans .— Vegetative Reproduction in Metzgeria. 
gemmae are produced in abundance the growth of the branch soon comes 
to an end and the gemmae become crowded in the apical region (Fig. i). 
In many cases, however, a branch will develop gemmae for awhile and 
then continue its growth normally. The gemmae arise in no definite 
order, large and mature individuals being often scattered among those 
which have scarcely begun their development. There is no evidence 
therefore that they are formed in acropetal succession. 
When a gemma is to be produced one of the marginal cells projects 
beyond its neighbours. In doing this the outer wall of the cell is ruptured, 
but the protruding protoplast is not naked, being covered over by a thick 
layer of transparent gelatinous substance, which perhaps represents a modi¬ 
fication of the inner portion of the original outer wall (Fig. 2, A). The 
deposition of this substance takes place prior to the rupture of the wall. 
Upon the inner surface of the gelatinous 
layer a very thin new wall is soon secrete d. 
The projecting cell thus formed does not 
represent the mother-cell of the future gem¬ 
ma, except in very rare instances. It almost 
invariably divides by a periclinal wall into 
two cells, the outer of which becomes the 
actual mother-cell. The periclinal wall is 
sometimes at right angles to the surface of 
the thallus, but is more likely to be inclined 
to it in such a way that the mother-cell of 
the gemma lies partly over the antical surface 
of the inner cell. The first wall in the mother¬ 
cell extends obliquely outwards from about 
the middle of the periclinal wall. A second 
oblique wall meets the first (Fig. 2, B) and thus forms at the tip of the young 
gemma a wedge-shaped cell, which proceeds to function as a two-sided apical 
cell, and to give rise to two rows of segments in the way so often described for 
Metzgeria fnrcata. The segments soon undergo further divisions (Fig. 2,C), 
and usually one of the two cells first cut off from the mother-cell divides as 
well, thus making the base of the gemma three cells wide (Fig. 2, D). In 
the early stages the gelatinous substance which covered over the original 
projecting cell becomes stretched out into a thinner and thinner layer 
at the apex of the gemma, and before long disappears completely (Fig. 2, 
C and d). 
The separation of the gemmae is brought about by the splitting of the 
walls between the two or three basal cells and the cell of the thallus cut 
off by the original periclinal wall. At the time of separation they vary 
considerably in length, a condition which is doubtless due to the fact that 
germination sometimes begins prematurely. What may be considered an 
Fig. i. M. uncigera. Apex of a 
gemmiparous branch, x 40. 
