LECT. V1IJ.] STRUCTURE OF HERBACEOUS ROOTS. 433 



are very conspicuous in these cells. I have not 

 been able to detect any fasciculi of proper vessels 

 in roots of this description, and I am, therefore, 

 disposed to believe that these tubular cells are not 

 only the reservoirs, but the conductors of the 

 proper juice. That they are endued with con- 

 tractility, and communicate freely with each 

 other, is evident: for, by making a horizontal 

 section of the root, the exudation of juice is much 

 greater than can be contained in the range of cells 

 which is divided ; and, by placing a longitudinal 

 slice of the root under the microscope, we find 

 the cells of several successive ranges empty and 

 shrunk. That this longitudinal communication 

 is regulated by valves, or something of a similar 

 nature, is probable; for, the exudation of the 

 juice is much more considerable on the divided 

 surface, and the shrinking of the cells extends to 

 more distant ranges, in the portion of the divided 

 root which remains attached to the plant, than in 

 that which is separated from it. 



The disposition of the parts of the root is more 

 varied in biennial than in annual roots. Taking 

 the root of Burdock, Arctium lappa, as an ex- 

 ample, we find that, in the first season of its 

 growth, or in the seminal plant, it consists of a 

 thick cellular bark, the cells of which are irre- 

 gular hexagons arranged in concentric circles 

 around a large central part, which is composed 



VOL. I. F F 



