CELLS ОЕ EUCOMMIA ULMOIDES, OLIVER. 247 
rarer occurrence than the similar cells in the cortex, which fact may be connected with 
the splitting away of the pith cells a little later on, so that a hollow stem is formed, with 
septa made up of very much stretched cells of the original pith, in which septa the caout- 
choue cells can be made out. But the origin of the cells is the same, and the pith at 
that early stage has a structure quite like that of the loosely built cortex (fig. 3). 
I was also able to observe a similar origin of the caoutchouc-containing cells in the 
parenchymatous tissue surrounding the vascular bundle of the petiole (fig. 2). Here 
too they arise by a longitudinal division of a cell of this tissue, which, however, here 
consists of more elongate cells more closely set than those of the cortex or pith, both of 
which are tissues with considerable intercellular spaces. 
In the secondary phloem I was not able to observe the actual origin of these cells, as 
the phloem cells are at their commencement more elongate than the cortical tissue, and 
there would be little difference between them and the caoutchouc-containing cells either 
in size or contents at so early a period. I cannot therefore state whether a cambium cell 
divides into two daughter cells, each of which grows out into a caoutchouc-containing 
cell, or whether these cells are each the result of differentiation of a single cambium cell. 
But from the young stages which I was able to observe in the secondary phloem, which 
contained a nucleus, and in which the caoutchouc was only in part formed, I can 
definitely conclude that these caoutchouc-containing cells do originate in the secondary 
phloem, and do not make their way there, as the latex cells of Broussonetia and Ficus 
for example do, according to Chauveaud. 
After the longitudinal division has taken place, the cells begin to grow out, and force 
their ends upwards and downwards through the intercellular spaces of the tissue in which 
they arise. This can be seen from fig. 3 in the case of the pith, and from fig. 4 in the 
case of the cortex. In fig. 4 growth has as yet commenced only at one extremity, but 
from the later stages it will be seen that the cell grows rapidly at both ends (fig. 5). 
The protoplasm, being now distributed over a larger area, becomes much clearer, and at 
the same time a vacuole makes its appearance in the cell, and enables the protoplasm te 
remain applied to the increasing surface. The origin of this vacuole is seen in fig. 4, 
and it is also seen in the more elongate cells of fig. 5. Here, however, № is only indi- 
cated in the wider part of the cell near the nucleus, as in the outgrowing parts the 
superficial view is drawn in order to show the appearance of the larger granules of 
caoutchouc. The lighter protoplasm enables us to see more clearly the large nucleus, 
which remains more or less in its initial position, and indicates the original position of 
the cell. I have never been able to find two nuclei in any of these cells, any appearance 
of two being attributable to some other cell, usually the sister cell lying below the cell 
under observation. 
To whatever length therefore these cells may grow, we must at present assume that 
all growth and other functions are regulated by this single nucleus. This would probably 
account for the fact that these caoutchouc-containing cells never branch like the latex 
cells of the Euphorbiacez, which cells Treub *, Schmidt +, Haberlandt {, and other 
* M. Treub, in Comptes rendus (1879); and in Archives Néerlandaises, t. xv. (1880) рр. 39-60. | 
Т E. Schmidt und Fr. Schmitz, in Sitzb. d. niedrh. Ges. für Natur- und Heilkunde zu Bonn (1879). (E. Schmidt, 
Bot, Zeit. (1882) p. 594). t 6. Haberlandt, * Function und Lage des Zellkerns,’ 1886. 
