THE MOVEMENT OF WATER IN PLANTS. 99 



to serve as a reservoir upon which the conducting cells of the alburnum 

 can draw. 



We are now in a position to give a connected account of 

 the transpiration-current. The water absorbed by the root- 

 hairs filters under pressure from the innermost parenchymatous 

 cells of the root into the walls of the adjacent wood-cells : at 

 the same time water is being withdrawn by the transpiring 

 leaves from the wood of the stem, and the demand is met by 

 the conduction of water upwards from the root through the 

 lignified cell-walls. It may happen that the quantity .of 

 water forced in a given time into the wood of the root is 

 not so great as that which is transpired by the leaves. Under 

 these circumstances the plant becomes flaccid and begins to 

 wither. On the other hand it may happen that the quantity 

 of water forced into the wood of the root in a given time is 

 greater than that which is transpired by the leaves. Under 

 these circumstances the water filters into the cavities of the 

 cells and vessels of the wood of the root and collects there. 

 This latter condition is the one which obtains, for instance, in 

 plants in the spring, and gives rise to the phenomenon of 

 " bleeding," which we have already considered. 



In the course of the very numerous experiments which have been 

 made on the filtration of water through wood, it has been found that 

 satisfactory results can only be obtained when, in the first place, the 

 water is quite pure (distilled), and in the second, the cut surface is quite 

 fresh. If the cut surface is allowed to become dry, the filtration is much 

 retarded. It has been found, further, that in long-continued filtration- 

 experiments, the rate of filtration gradually diminishes : if, however, a 

 thin layer be removed from the surface of the piece of wood which has 

 been in contact with the water, and the experiment be then resumed, the 

 rate of filtration is again considerable. This gradually increasing resist- 

 ance to filtration is due, as von Hohnel has shewn, to the fact that the 

 cut surface becomes covered with a layer of mucilaginous substance, 

 derived partly from the cells which have been necessarily mutilated, and 

 partly from Bacteria which grow and multiply in it. This is also the 

 reason why cut branches wither even when their ends are placed in water. 



The most trustworthy observations which we have as to 

 the rate at which the transpiration-current travels in the wood 

 are those of Sachs. His method consisted in supplying the 



72 



