466 SCIENCE PROGRESS. 



them constitutes the primary medullary rays, while the 

 conjunctive tissue outside the bundles is the pericycle, often 

 less obvious than in the root, but characterised by the same 

 power of forming new tissues and organs. 



Now the views just explained are essentially different 

 from those of De Bary. The latter author does not recog- 

 nise the central cylinder in the stem at all (except incident- 

 ally, in discussing the apical development). In the case 

 of the root he describes the central cylinder as a single 

 vascular bundle (comparable therefore to any one of the 

 numerous collateral bundles of a typical Phanerogamic 

 stem). Hence there is no attempt to bring out the 

 homology between root and stem, while in Van Tieg- 

 hem's system the recognition of this homology gives the 

 clue to the whole anatomy. The central cylinder of the 

 root is beyond question directly continuous with that of the 

 stem, and not with any single bundle. The individual 

 xylem and phloem-strands of the root are continuous with 

 those constituting the vascular bundles of the stem, though 

 of course differently arranged. The mistaken opinion that 

 the central cylinder of the root is a single vascular bundle 

 arose from the exaggerated importance attached to the idea 

 of the vascular bundle. As a matter of fact such bundles, 

 in the sense of conjoint " liberoligneous " strands, do not 

 exist, as such, in the root ; the groups of xylem and phloem 

 are here separate from one another. In order to force the 

 root into the vascular bundle scheme it was necessary to 

 make a single bundle of its whole central cylinder, and 

 thus the entire morphology of the vascular tissues was 

 obscured. 



The limitation of the central cylinder in the stem pre- 

 sents much greater difficulties than in the root. This is 

 owing, in the first place, to the intimate relation between 

 the vascular system of the stem and that of the leaves. In 

 the great majority of Phanerogams, and in some Cryptogams, 

 all the vascular bundles are " common " to leaf and stem, 

 so that every bundle in the central cylinder of the stem is 

 directly continuous with one which passes out into a leaf. 

 Hence at every node the continuity of the main stele is 



