400 Transactions of the Royal Society of South Africa. 
parted with. Such a wet sponge on the leaves will undoubtedly serve 
to retard transpiration, and in this connection the use of the hairs is 
obvious. There remains now only Grassula cymosa, into which eosine 
had penetrated in the experiment previously described. 
Four branches of Grassula cymosa were cut on June 1st. Three had 
their cut ends sealed, one had not. Though still quite healthy, all showed 
decided signs of withering on June 6th. They were then immersed in 
strong, watery eosine solution for twenty-four hours, except their lower 
portions towards the cut ends. In none of them had the leaves regained 
their turgescence on the following day. After washing in water for five 
minutes the parts which had been immersed in eosine were found to be 
suffused with red, the marginal papillae were, on the whole, only faintly 
stained, and in some of the younger leaves hardly stained at all ; the upper 
surfaces of the leaves were fairly evenly stained, in most leaves the under 
surface was also stained evenly, in others only the marginal portions, 
which, on the whole, showed more decided staining throughout. Micro- 
scopical examination of a strongly stained leaf showed the marginal 
papillae very faintly stained, with contents practically colourless, the 
epidermis along the edge of the leaf with stained contents, other portions 
of the epidermis stained very little, the vascular bundles strongly stained. 
In the stem the contents of the epidermis cells and a few of the adjoining 
cells were slightly stained, the vascular bundles were strongly stained, 
and the eosine stain had also slightly diffused into the adjoining tissues ; the 
bulk of the cortex, like the bulk of the mesophyll of the leaf, was unstained. 
On the whole, the microscopic examination was decidedly in favour of 
the conclusion that the eosine solution penetrated through the marginal 
epidermal cells and not through the papillae. Thus these papillae cannot 
be looked upon as water-absorbing organs unless they behave very 
differently in other species. Incidentally it was also shown that they 
cannot be looked upon as water-reservoirs, otherwise the eosine solution 
would certainly have penetrated into them more freely. However, as the 
solution which actually entered in twenty-four hours' immersion did not 
restore the turgidity of the mesophyll-cells it may safely be concluded 
that the power of taking up water possessed by the leaves (and evidently 
to a more limited extent by the stem) under experimental conditions is of 
absolutely no practical consequence to the plant under natural conditions. 
This conclusion was strengthened by a further experiment with 
Grassula cymosa arranged after the manner of an old experiment figured 
by Pfeffer.f A large side-branch of a portion of a plant was immersed in 
* Water-stomata, which are found in other species of Crassula, seem to be absent here. 
Most of the stomata all over the leaves were found open, the guard-cells as a rule slightly 
stained, but the eosine solution had, as a rule, not entered through the stomatal slit. 
t W. Pfelfer, " The Physiology of Plants," Engl. Ed., vol. i. (1900), fig. 15, p. 160. 
