French National Institute. 3,7 1 



the layers of wood, which are every year raanifesled under 

 the bark, and that it is lengthened and ramified by shoots, 

 which are merely the development of buds. Each of these 

 new shoots has only a siijgle layer of wood, which is in com- 

 munication with the last of those formed upon the trunk ; 

 and the medullary thread, which occupies the axis of these 

 shoots, conies from the marrow which reigns in the middle 

 of the tree. Naturalists generally suppose that these suc- 

 cessive ligneous layers grow every year under the internal 

 face of the bark. 



Palm-trees, and the other monocotyledonous trees, grow 

 quite differently : the new fibres are d<;veloped in the axis, 

 and not in the circumftrence of the trunk ; they traverse the 

 Avhole length of this axis in, order to expend themselves at 

 the summit of the tree in leaves and flowers. This is the 

 reason why the trunk of the palm-tree increases so imper- 

 ceptibly, particularly below, and in general produces no 

 branches. 



M. Desfontaines, our colleague, has shown that this me- 

 thod of growing is common to nearly all the monoeotyle-- 

 donous plants, and distinguishes them in general from the 

 dicotyledonous. 



But, M. du Petit-Thouars having remarked that the 

 draccBua (trees really monocotyledonous) ramified, as it 

 were, like the ordinary trees, and wishing to account for this 

 phasnomcnon, he ascertained by dissection that the axis of 

 a branch did not communicate with that of the tree, but that 

 the fibres of this branch, when arrived at the place of its 

 junction with the trunk, blow out upnn the latter, diverging 

 like radii ; the lower fibres descend directly ; the upper ones, 

 after having mounted a little, become crooked, and descend 

 also. These trees therefore grow by eoncentrical layers, 

 and, in fact, they become thicker the njoi'e they ramify. 

 Such are the facts, and the following is the system : 



M. du Petit-Thouars, applying these observations to all 

 the trees with eoncentrical layers, concluded that the new 

 layers are not produced by the bark, but the buds ; that their 

 fibres are descending prolongations of these buds, as the 

 bhoots are ascending prolongations. lie thinks that the 

 A a 2 juice 



