FUNCTION.] RESPIRATION. 203 



leaves to such purposes as those just, mentioned. It has been 

 already shown, in speaking of the anatomy of a leaf, that in 

 most cases it consists of a thin plate of cellular tissue pierced 

 by air vessels and woody tissue, and enclosed within a hollow 

 empty stratum of cells forming epidermis. Beneath the upper 

 epidermis the cells of the parenchym are compactly arranged 

 perpendicular to the plane of the epidermis, and have but a 

 small quantity of air cavities among them. Beneath the 

 lower epidermis the parenchym is loosely arranged parallel 

 with it, and is full of air chambers communicating with the 

 stomates. The epidermis prevents too rapid an evaporation 

 beneath the solar rays, and thickens when it is especially 

 necessary to control evaporation more powerfully than usual ; 

 thus in the Oleander, which has to exist beneath the fervid 

 sun of Barbary, in a parched country, the epidermis is 

 composed of not less than three layers of thick-sided cells. 

 To furnish leaves with the means of parting with superfluous 

 moisture, at periods when the epidermis offers too much 

 resistance, there are stomates which act like valves, and open 

 to permit its passage : or when, in dry weather, the stem does 

 not supply fluid in sufficient quantity from the soil for the 

 nourishment of the leaves, these same stomates open them- 

 selves at night, and allow the entrance of atmospheric moisture, 

 closing when the cavities of the leaf are full. In submersed 

 leaves, in which no variation can take place in the condition 

 of the medium in which they float, both epidermis and stomates 

 would be useless, and accordingly neither exists. For the 

 purpose of exposing the fluids contained in leaves to the 

 influence of air, the epidermis would frequently offer an 

 insufficient degree of surface. In order, therefore, to increase 

 the quantity of surface exposed, the tissue of the leaf is 

 cavernous, each stomate opening into a cavity beneath it, 

 which is connected with multitudes of intercellular passages. 

 But, as too much fluid might be lost by evaporation in parts 

 exposed to the sun, we find that the cells of the upper stratum 

 of parenchym only expose their ends to the epidermis, and 

 interpose a barrier between the direct rays of the sun and 

 the more lax respiring portion forming the under stratum. 

 It is not improbable, moreover, that those cells which form 



