1902] RISE OF THE TRANSPIRATION STREAM 167 



ral explanations of its mechanism are conceivable. A sugges- 

 tion by Pfeffer, that an unequal distribution within the cell of its 

 osmotically active matter could cause a stream through the cell, 

 has sometimes been adopted as the best solution. But to main- 

 tain this inequality would use as much, or rather more, energy 

 than is required to keep up the osmotic pressure present and to 

 lift the water however far it is forced ; and this energy would 

 presumably be a net loss to the plant. Detmer (1877) and C. 

 Kraus regarded root pressure as a function of tissue tensions; 

 but the tensions to be observed do not justify this view; and if 

 they did, the solution would not be ultimate, for how should the 

 tissue tensions arise and be maintained ? Under the most advan- 

 tageous conditions external to the active cells, a rhythmic 

 variation in their permeability would result in a root pressure 

 corresponding to the loss of osmotically active matter. But 

 this would make an exceedingly poor machine, unless, as in the 

 original scheme of this kind, by Godlewski, an unequal distribu- 

 tion of the permeability works with the periodic variation. And 

 unequal permeability of different sides of the cell constitutes a 

 perfectly sufficient explanation without any periodicity. 



The greatest pressure must be exerted on the side of the cell 

 where resistance to the passage of the solute is greatest. By 

 virtue of this greater pressure when the cell can contain no more 

 water, water will be taken up there, and given off on the other, 

 more permeable side, with at least as much of the solute as 

 would exert an osmotic pressure equal to that under which the 

 water escapes. Artificial cells illustrating this mechanism have 

 been constructed by Sachs (1882) and myself ( 1900). 

 far as its diffusion out of the tracheae into the cortex is resisted, 

 the matter dissolved in the sap in the tracheae exerts an osmotic 

 pressure against the surrounding cells, and to that extent a 

 root pressure can then be maintained without any subsequent 

 loss of solutes by the living cells. It seems probable to me 

 that this suggestion represents the mechanism of the bleeding 



' Hofmeister's artificial cell (1S58 : 11 ; 1862 : 145), sometimes refered to as illus- 

 trating this action, had aU its membranes alike. 



' In so 



