35 1 
the Leaf of Welwitschia mirabilis . 
cotyledon and of the young leaf is much the same as that of the adult leaf, 
but the walls are less thickened. In my material of the cotyledon I have 
noticed that only the upper portion of the dorsal wall is lignified (Fig. 5). 
The figures and descriptions of the stoma given by Sykes ( 19 , p. 182, 
PI. XVII, Figs. 2, 3) do not seem to be correct ; her Fig. 2 only represents 
a section cut near the end portion of the stoma, ‘ M * in this figure being not 
the ventral wall of the guard cell. ‘ M ’ in her Fig. 3 is very puzzling, 
because, in fact, ‘ Z ’ in this figure corresponds to ‘ M * in Fig. 2. 
The development of the stoma has been traced in the young leaf, and 
particularly in the cotyledon. An initial cell divides longitudinally into 
Text-fig. 3. A portion of epidermis from cotyledon, showing successive stages in the 
development of the stoma, x 285. 
two. One of the daughter-cells again divides in the same way, so that 
there are three cells formed. The two lateral ones become subsidiary cells, 
while the central one will be the mother-cell of the stoma. The mother- 
cell divides longitudinally and forms the two guard cells. Thus a stoma 
with two parallel subsidiary cells is formed from one single initial cell. 
This is the simplest and probably the typical case (Text-fig. 3). One or 
both of the subsidiary cells may further divide transversely, longitudinally, 
or sometimes obliquely (Fig. 4, Text-fig. 4). 
It has been observed occasionally that in the three-celled stage, which 
is the result of two successive divisions of the initial cell, a lateral cell, 
B b 2 
