of the Stem of the Phanerogamia. 403 



asserts that the vascular bundles originate in the primary paren- 

 chyma [Urparenckyma) of the embryo^ and increase in number 

 solely by ramification, and that no new independent vascular 

 bundles can be formed in the plant. 



The dogma thus generally expressed is decidedly erroneous. 

 We may admit it for the vascular bundles belonging to one axis, 

 originating by the development of a common connected cam- 

 bium-sheath : for no one now entertains any doubt of the want of 

 foundation of the view of Du Petit Thouars and his followers, 

 that the vascular bundles originate in the leaves and run down in 

 the stem to the points of the roots ; for the vascular bundles enter- 

 ing the leaves from the stem are developed from below upwards. 

 But it is another question, and, as it appears to me, one not yet 

 solved by existing researches, whether in all plants the primary 

 vascular bundles lying at the circumference of the pith are con- 

 nected together, or whether there do not exist plants in which 

 the young vascular bundles which run to leaves situated higher 

 up, are formed between the older, without entering into con- 

 nexion with them, and whether the mutual connexion of them 

 is not effected by the subsequently produced woody layers. We 

 shall, however, leave this point out of consideration. 



On the other hand, there can be no doubt that the roots of 

 Monocotyledonous stems possess their own bundles, which only 

 subsequently enter into connexion with the vascular bundles of 

 the stem, and (at least in many cases) cannot be derived from 

 the same cambium-layer which was the place of origin of the 

 latter. The Palms and Pandanese are the most favourable plants 

 for the investigation of these conditions. In the former, the 

 observations of myself, Mirbel, and Karsten agree tliat the roots 

 originate under the rind, in that layer of the stem through which 

 the most external fibrous bundles run — therefore in a layer where, 

 in a fully developed Palm-stem, no peripheral cambium-layer 

 any longer exists. Here there is formed locally for each root a 

 nucleus of new cambium-tissue, by the transformation of paren- 

 chyma-cells existing long before ; and in these nuclei originate 

 the vascular bundles, which on the one side grow forward with 

 the roots and form the wood of the roots, well known to be very 

 different from the vascular bundles of the stem, — on the other 

 side spreading out like a tuft, and penetrating an inch deep into 

 the parenchyma of the stem, become entangled with the vas- 

 cular bundles of the stem, and apply their extremities upon 

 them. One must of course give up any idea that these vascular 

 bundles grow into the cellular tissue of the stem in the manner 

 of a root growing into the earth ; they can only be formed by 

 the transformation of particular portions of the parenchyma of 

 the stem into cambial tissue, and the development of the latter 



