Lag a E ERRORE E E o ET AE RN NET 
OF THE LEAVES OF BRITISH GRASSES. 321 
all strongly cuticularized; moreover, they do not correspond to the position of the 
water-storing tissue which is found in certain Egyptian desert-grasses and described by 
Volkens (1887). 
The “ motor cells” serve the function of bringing about the folding or rolling which 
takes place in nearly all grass-leaves in dry weather, and serves to protect the upper 
surface of the leaf against excessive transpiration. When the leaf is folded or rolled, 
only the lower surface (or part of it) is exposed, and this, as has been seen above, 
generally possesses a stronger cuticle, fewer stomata, Ze, than the upper, and is thus 
less liable to excessive transpiration under dry conditions. The change is probably 
brought about by an alteration in turgidity of the motor cells, causing them to contract 
and thus shortening the upper surface of the leaf. Kerner *, however, considers that in 
some cases, at least, these ** motor cells” are not strong enough for this purpose, and 
suggests that the change is brought about by the swelling of certain thick-walled cells 
situated under the lower epidermis. This to me seems hardly satisfactory. He also 
suggests that the “motor cells” function as a kind of pneumatic cushion during the 
folding of the leaf, and also, in part at least, as water-absorbing cells, a view which 
seems to be based on an exaggerated idea of the feebleness of the cuticularization of 
their outer walls (Pl. 40. figs. 57, 75). 
The motor cells differ from the ordinary epidermal cells chiefly in their greater depth 
(Pl. 40. fig. 57, 75). They are much more narrowed at their outer than at their inner 
end, where they touch the assimilating parenchyma. They pass gradually into the 
ordinary cells of the epidermis, and are sometimes hardly distinguishable from them; in 
other cases, however, as in most species of Poa, in Triodia decumbens (Pl. 40, fig. 59), 
Arundo Phragmites, Hierochloé borealis (Pl. 40. fig. 66), &c., the difference is very 
marked. The cells are certainly not conspicuously more thin-walled than the ordinary 
cells of the epidermis (PI. 40. figs. 55, 57). 
The motor cells are arranged in bands running longitudinally along the whole length 
of the leaf. In leaves where the upper surface is more or less ribbed a band of motor 
cells is generally found at the base of each groove (Pl. 37. fig. 17; Pl. 39. fig. 50, &e.). 
Where the upper surface is perfectly smooth we usually find two bands of motor cells, 
one on each side of the midrib. This is the case, for instance, in most species of Poa, 
Glyceria (Pl. 38. figs. 37, 39), Catabrosa aquatica (Pl. 38. fig. 38). In Dactylis 
glomerata, alone among British grasses, there is only one band of motor cells, and it 
stretches right across the midrib (Pl. 38. fig. 31). 
Long silky hairs are not common on grass-leaves. They do occur, however, in certain 
roadside grasses, as in Holcus lanatus (Pl. 39. fig. 43), H. mollis (fig. 42), Bromus 
sterilis (Pl. 37. fig. 27), &c., and there can be little doubt that here they serve to prevent 
excessive transpiration by retaining a layer of more or less water-saturated air about the 
leaves. They may serve also in preventing the stomata, &c. being blocked by dust, 
a danger to which these roadside grasses are peculiarly liable in dry weather. 
Much more common are shorter, stiffer hairs; but these vary considerably in abund- 
ance and arrangement, also probably in function. In xerophilous grasses, both of the 
* Kerner, ‘ Natural History of Plants,’ Engl. ed. (1897), pp. 339-345. 
