[oo8 



A TEXTBOOK OF THEORETICAL BOTANY 



but may often be additionally attached to the epidermis by wedge-like pro- 

 cesses extending inwards between the cells. The thickening of the epidermal 

 walls is always most prominent at the leaf margins, which are obviously the 

 parts most vulnerable to tearing, and here the epidermis may also be supported 

 by collenchyma or sclerenchyma. 



Many species have, however, outer epidermal walls which are calcified 

 or silicified, more commonly the latter. Silicification of the epidermis is 

 particularly strong in Gramineae and in Cyperaceae, as well as in certain 

 Cryptogams such as Equisetiim, and a rash attempt to tear such plants out of 

 the ground may lead to badly cut fingers from the splintered edges of the 

 siliceous covering. 



Fig. 996. — Atriplex liastata. Vesicular hairs from the surface of a 



young leaf. 



Thickening of the outer epidermal walls is often greater in the middle 

 than at the sides of the cells and there may be a definite papilla, centrally 

 placed, or a lens-like thickening in the substance of the wall itself. Such 

 local thickening, combined with the usually domed shape of the wall, gives 

 the epidermal cells the character of condensing lenses, and Haberlandt drew 

 attention to their probable importance in the perception by the leaf of the 

 direction of the incident light, and hence in the adjustment of the position 

 of the leaf with regard to it. He showed that parallel rays of light falling on 

 the upper epidermis of the leaf, are, in many species, condensed to a bright 

 spot (Fig. 997) on the inner cell surface, and that in at least a few cases it 

 was exactly focussed on this inner surface, so that an image of surrounding 

 objects is formed in each cell, in a manner recalling the multiple eyes of 

 insects. He very justly points out that there is no likelihood of any perception 

 of such images by the protoplasmic lining of the cell, and that the operation 

 of the mechanism depends solely on the difl^erence in brightness between 



I 



