ANATOMY OE LEAVES. 309 



the tipper surface. They are never situated on the 

 costae, nor on the edges of the leaf. 



But these demonstrations make us acquainted with 

 the superficial aspect only of the foliar apertures ; 

 placing under the microscope a very thin vertical slice 

 of a leaf of the Clove Pink (Dianthus carynphyUus), 

 cut in the direction of the axis of the leaf, we find 

 that the aperture which is thus divided in its longitu- 

 dinal diameter, is a short cylindrical tuhe penetrating 

 completely through the cutis, and terminating in a sac, 

 which is impressed with a vesicle that appears to com- 

 municate with the oblong cells immediately beneath 

 the cutis. But although the aperture penetrates the 

 cutis, there is no opening through the epidermis, 

 which, on the contrary, enters into the tubular part of 

 the pore, and lines it throughout. In another slice of 

 the same leaf, cut so as to divide one of the apertures 

 in its cross diameter, we may perceive that the vesi- 

 cle appears to be double ; from which it is probable 

 that it is this vesicle, seen through the transparent sub- 

 stance of the cutis, which gives the appearance of the 

 shield in the superficial view of the aperture. As we 

 find that, in the superficial view of these apertures, 

 the character varies considerably in different plants, 

 so this form of the tube and the vesicle is also vari- 

 ously modified ; but the general character is nearly 

 the same, with a very few exceptions, throughout the 

 vegetable kingdom. 



Decandolle considers that the cuticular apertures 

 are connected with the ultimate ramifications of the 

 vessels of the leaf; and, if it be true, that the cuticu- 

 lar meshes are formed by lymphatic vessels, which ter- 

 minate on one hand in the larger vessels of the leaf, 

 and on the other, in the vescicular circles surrounding 

 the fundus of the aperture, this opinion must be cor- 

 rect. 



