THE MATERIAL INCOME OF PLANTS 307 



them. Yet the share of the rest of the protoplast in this discrimination 

 is not to be overlooked; and since it is impossible to analyze the action 

 of each part, we may for convenience consider the protoplast as a mem- 

 brane between the vacuole and the outer world. But for substances in 

 the protoplast 'itself the ectoplast may act alone. 



Selective action. The chemical composition of the cytoplasm being 

 almost wholly unknown and doubtless variable, no clear statement can 

 be made as to the mode of its discriminative action. It is known only ' 

 that it allows many substances to pass through readily and debars others; 

 and further, that some substances, which are usually denied passage, 

 are permitted to pass under other conditions. These relations are best 

 explained by the theory that solubility in the membranes is prerequisite 

 to osmosis. If so, a change in composition of the cytoplasm might ac- 

 count for the change in permeability that is observed on occasion. 



It is quite possible that local differences in the composition of the cytoplasmic 

 membranes (a sort of mosaic composition) may permit the passage of different sub- 

 stances at different places. 



Variable selection. The welfare of the organism is largely dependent 

 on the discriminative action of the cytoplasmic membranes, for sub- 

 stances requisite to food-making are allowed to enter; and foods are not 

 permitted to diffuse out and be lost. Chemical transformations of the 

 most varied kind occur within the plant, both among the substances that 

 enter it and are elaborated into foods, and also among the foods that 

 are assimilated. Of course each change in chemical nature changes the 

 relations of the substance to the protoplast and may modify thereby its 

 diffusibility through it. Moreover, without known chemical change, 

 the mere presence of one solute may greatly modify the behavior of 

 another, either by changing the membranes, or by its direct influence 

 upon the other solute. With membranes capable of change, and solutes 

 capable of change, and the almost unknown extent of the influence of 

 one solute on another, the complexity of the phenomena of osmosis 

 has almost baffled investigation hitherto, but some hopeful progress has 

 been made recently in the discovery of factors determining the permea- 

 bility of protoplasm. 



It cannot be too strongly emphasized that the " selection " above described has 

 in it no element of choice, nor does it depend upon the " needs " of the plant. On 

 the contrary, it is purely physical, and depends solely upon the mutual relations of 

 the substances (membranes and solutes) which the conditions bring into contact. 



