OF THE LEAVES OF BRITISH GRASSES. 351 
The conditions under which these wood-grasses naturally grow are decidedly less 
favourable to transpiration than those under which either meadow or waste-place grasses 
are found. On the other hand, there is no excessive supply of water, nor is there any 
difficulty as regards aeration, such as is suffered by the grasses of wet places. The 
conditions for illumination, however, are not so favourable. 
When we examine the leaf-structure of the grasses of this group, we find that it is 
fairly uniform, and it is one that is quite well adapted to such conditions as mentioned 
above. 
In the first place, the leaves are all distinctly thin in comparison with their breadth : in 
the figures this is best seen in Melica nutans, M. uniflora (Pl. 38. fig. 32), Poa nemoralis, 
and Brachypodium sylvaticum, but the feature is quite a constant one, the figure in the 
other cases showing only a small part of the leaf-section, so that the relative breadth and 
thickness of the leaf cannot be judged. This feature, of course, is an adaptation to favour 
transpiration and to secure as much illumination for the leaves as is possible in their 
shaded habitat. 
The ribs on the upper surface of the leaf are either entirely absent, as in Poa nemoralis 
and Melica uniflora (Pl. 38. fig. 32), or they are practically obsolete, as is the case in all 
the other members of the group. Stomata, again, are generally fairly abundant on both 
sides of the leaf, though in most cases they are more abundant on the upper than on the 
lower side. Moreover, the cells of both the upper and the lower epidermis are only feebly 
cutinized, those of the lower being rather more strongly so than those of the upper. 
These features all tend to favour transpiration, or perhaps it is better to say that there 
is an absence of all those structural features which in other grass-leaves tend to reduce 
transpiration. 
Hairs occur abundantly on the upper leaf-surfaces of Melica nutans and Brachypodium 
sylvaticum. Yt is difficult to understand their significance; they can hardly be regarded 
here as a xerophilous feature. 
The stereome is fairly strong, in most cases the larger vascular bundles are girdered at 
least to the lower side, while often even the smaller bundles are girdered to one or to 
both sides of the leaf (Pl. 37. fig. 26). In Melica uniflora most of the bundles are not 
girdered, but, as they all have fairly strong bands of stereome above and below, the point 
is not of importance. This strength of the mechanical tissue is in all probability a 
direct outcome of the thinness of the leaf and its general softness, these characters 
rendering a well-developed supporting system necessary. 
'The motor cells in nearly every case are arranged in very broad bands—in some cases 
they are well marked off from the ordinary epidermal cells, in others they are hardly 
conspicuous. ie 
In Brachypodium sylvaticum the cells of the lower epidermis are strongly cutinized 
and their outer surfaces are flat. These are points in which the leaf of this grass differs 
from that of every other member of the group. 
