VIII.] SAP. 121 



transference of this fluid to the existence of some molecular 

 relation established between the substance of the cellulose 

 thickening of the cell-walls and the particles of water by 

 which it is freely permeable, the precise nature of which we 

 are not yet in a position to define, but which may be 

 currently described as capillary. Whatever explanation 

 may be ultimately adopted, the fact is clear that the chief 

 upward transference of fluid takes place through the wood- 

 cells of the vascular bundles, which collectively form in 

 Dicotyledons the sap-wood or alburnum, and that this 

 transference during active spring-vegetation is very rapid. 

 The direction of the, current is mainly determined by the 

 constant evaporation from the leaves, which necessarily 

 tends to render their cell-contents denser, so that the water 

 taken up by the surface-cells and hairs of the root-fibrils is 

 impelled upwards cell by cell, to restore the equilibrium, 

 until it reaches the leaves and other tissues exposed to 

 evaporation. The course which the ascending sapy as it is 

 termed, takes has been usually supposed to be through the 

 cells forming the younger layers of wood, the vessels assisting 

 when the current is rapid. This, however, is not absolutely 

 determined. The elaborated food- products, prepared in 

 the green organs under the influence of light, are generally 

 admitted to descend chiefly through the inner layers of 

 bark. Indeed, a rude experiment may be regarded as 

 strongly confirming this view. If you remove a ring of bark 

 from the stem of a tree, or bind it very tightly round with 

 a strong hoop, no wood will be formed below the ring or 

 hoop. On the other hand, a considerable thickening will 

 take place immediately above it. 



1 6. I have spoken of leaves as capable of absorbing gases, 

 especially carbonic-acid gas, and probably also vapour, from 

 the atmosphere. 



