1 84 RALPH S. LILLIE. 



under what circumstances this possibility is realized is difficult 

 to say. Certainly in most of the instances hitherto investigated 

 (muscle, nerve) the passage of the constant current through the 

 cell produces marked polarization-effects (i. e., sets up imme- 

 diately a strong counter E.M.F.). Nevertheless the above possi- 

 bility should be considered. 



V. GENERAL CONCLUSIONS. 



The foregoing discussion raises the question whether in organic 

 growth the essential structural condition is not the presence of 

 semi-permeable and hence electrically polarized partitions sepa- 

 rating the living substance from its medium, and at which 

 processes of electrolysis may take place. If this is so, the 

 prevalence of the cellular type of organization would be largely 

 accounted for. Organic growth (as well as normal maintenance) 

 would then involve the deposition in the continuous surface- 

 layer of materials which preserve the character of that layer, 

 i. e., its continuity, semi-permeability, and specific physico- 

 chemical composition. It is clear that any layer having such 

 properties would have to form a closed surface; otherwise the dif- 

 ference between the inner and outer solutions could not persist; 

 hence a "cell" mass of protoplasm enclosed by a membrane 

 would be the living unit. A tendency to a continual increase of 

 this surface-layer in an appropriate nutrient medium would 

 then constitute the primary feature of the growth-process. 

 Evidently the continued growth of such a system would involve 

 a decrease in the ratio of surface to volume: and the need of 

 preserving this ratio above a certain minimum may be the condi- 

 tion that has led to the association of cell-division with growth, 

 as Herbert Spencer suggested long ago. According to this con- 

 ception, "living" material would be essentially material at the 

 boundary-surface or under the direct influence of the boundary- 

 surface; definite limits would thus be imposed upon the dimen- 

 sions of the individual living units. In general we find that 

 protoplasm is partitioned freely; i. e., the ratio of surface- 

 protoplasm (membrane-protoplasm) to the total mass is usually 

 large. The exceptions (yolk-laden eggs, muscle cells, certain 

 plant-cells) may be explained on special grounds. 



