370 



CONDUCTING SYSTEM 



was experimentally demonstrated by Schwendener with the aid of 

 staining solutions, and may also be inferred from certain observations 

 made upon plants growing under natural conditions. In many 

 Gramineae and Cyperaceae, for example, the primary cortical tissues of 

 the root die and shrivel up at a comparatively early stage. The central 

 cylinder is thenceforward enclosed only in the endodermis, but never- 

 theless remains quite fresh and continues to perform its functions in a 

 perfectly normal fashion. In a number of roots, e.g. among Gramineae, 

 Cyperaceae, Juncaceae, in many Palms and in various other Monoco- 

 tyledons, the endodermis only becomes impervious [at a late stage], 



Fig. 14S. 



Radial bundle of the root of Allium ascalohicum, in T.S. ; ri, large central vessel, 

 towards which the five hadrome-plates converge ; p, pericycle ; s, endodermis ; 

 d, passage-cells, situated opposite the hadrome-plates. 



when there is no longer any need for freedom of interchange by diffusion 

 between central cylinder and cortex. In such cases the endodermis 

 becomes equally impervious at every point. Among Dicotyledons and 

 Ferns, on the other hand, as well as in the Liliaceae and Orchidaceae, 

 certain circumscribed and histologically differentiated areas of the endo- 

 dermis often remain pervious, even when this layer is fully developed. 

 In such cases the endodermis comprises two different kinds of cells, 

 namely, typical endodermal elements, and interspersed among them the 

 so-called passage-cells, the tangential walls of which are not only un- 

 thickened, but also entirely unsuberised and hence readily permeable. 

 These passage-cells are always placed opposite the hadrome-plates, 

 and in fact abut directly against the first-formed [protoxylem] vessels 



