ORIGIN OF TllK WOOD. 107 



buds diflfer from the seed in no important respect. They send 

 up the ascending axis in the form of a branch, and tlie descend- 

 ing axis in the form of fibers, which united form the stem, and 

 tei-miiiate in the earth, modified in some respect by the celhilar 

 system, forming what, taken as a whole, is called the wood. 

 The various varieties of wood are pi-oduced by the action of tlie 

 cellular system. Of this w^e may be convinced by removing a 

 ring of bark of one tree, and supplying the place of the removed 

 portion by a ring from a different tree of the same fomily, and 

 the wood formed under the strange portion will be of the same 

 kind as the tree from which it was taken, but the wood above 

 and below will remain unaltered. The same is the case with 

 gratis. The graft will always remain unaltered, while the wood 

 of the stock remains unaffected by the graft. 



190. Some uncommon examples are upon record, which go 

 to confirm the position. In the Pandanus, the stem near the 

 ground is extremely slender; higher up it is thicker, and gives 

 out aerial roots, which act as props to the plant by entering the 

 earth ol)liquely. The aerial roots are what would have com- 

 posed a part of the stem, had they remained bound by the cel- 

 lular substance into one cylinder, but from some cause they 

 separate and subtract so much from the mass of the stem near 

 the root. Professor Lindley describes another instructive case 

 on this subject in the Barbacenia from Rio Janeiro, recently 

 discovered. 



It consists of a central portion similar to common monoco- 

 tyledonous stem, but this column is surrounded by bundles of 

 vascular fibers, the bundles having no connection with each 

 other, corresponding, in the opinion of the professor, to the 

 aerial loots of the Pandanus. The putting forth of roots from 

 buds when planted, as in the case of the Multicaulis, appears 

 easily explicable on this principle. These will put forth roots 

 • iimediately from the bud, and in all essential points agree ex- 

 actly with the germination of the seed. A complete bud of a 

 Multicaulis will germinate, when all the substance of wood and 

 bark is removed, which does not enter into the composition of 

 the bud. The vitality of most buds seems to be much less, and 

 in some cases cannot be made to germinate at all when re- 

 moved from the parent stock. Whence come the roots from a 

 bud, if they are not the direct production of the bud ? When 

 the bud is removed from the parent, the fibers it would other- 

 wise have sent through the trunk become proper roots at once. 



What of buds ?— 190. What of the Pandanus ? Barbacenia ? What of 

 the buds of tlie multicaulis ? What does it prove ? 



