PHYSIOLOGY OF CELL-DIVISION 391 



along the electrical lines of force, and apparently in many in- 

 stances fusing to form strands. There are a number of possi- 

 bilities in the formation of such potential-gradients; these I 

 have partly discussed in previous papers,^" but some further 

 consideration seems desirable at present. The theory of dif- 

 fusion-potentials imphes that any well marked increase in the 

 permeabihty of the plasma-membrane to any intracellular 

 electrolyte distributed uniformly thi'oughout the protoplasm 

 and of which the cation diffuses more rapidly than the anion 

 (e.g., an acid), will form within the cell a potential-gradient, 

 corresponding to the diffusion-gradient, with the central region 

 negative relatively to the peripheral region."*' In other words, 

 for a certain time the concentration of the electrolyte will be 

 higher at the central region than at the cell-surface — ^where it 

 is diffusing into the surrounding medium — and during this 

 time a corresponding potential-gradient must exist. The pre- 

 cise characteristics of such a gradient will depend on the dis- 

 tances concerned, on the relative migration rates of anion and 

 cation through the protoplasm, and on the concentrations. 

 The latter two factors are unknown, so that it is difficult to 

 judge of the adequacy of such a hypothesis; it seems, however, 

 possible that the typical development of a bipolar system of 

 cytoplasmic radiations at the time of division may be due largely, 

 if not entirely, to the appearance of two symmetrical areas of 

 increased permeability centering at the poles and extending 

 toward the equator, as suggested above — the electrical fields 



" Biol. Bull., 1909, vol. 17, p. 188; Amer. Journ. Physiol., 1910, vol. 26, cf. 

 p. 128, Jour. Morph., 1911, vol. 22, pp. 721 seq. 



^1 I do not see why this simple hypothesis should be regarded as unintelligible 

 (cf. Heilbrunn, Biological Bulletin, 1915, vol. 29, p. 184). It consists simply in 

 a recognition that when a diffusion-preventing barrier is removed between two 

 solutions, a diffusion-potential arises corresponding to the diffusion-gradient. 

 Thus (e.g. ) the P.D. between lO"'' normal and 10~^+^ normal HCl is approximately 

 0.05 volt; if two such solutions are separated by a semi-permeable membrane, 

 and the permeability of the latter is then increased, evidently the acid at once 

 begins to diffuse from the more to the less concentrated solution and continues 

 until the concentration is uniform. As long as there is a diffusion-gradient 

 there will be a corresponding potential-gradient. Similarly between the cen- 

 tral and peripheral regions of the cytoplasm at the time of increased permeability. 



THE JOURNAL OF EXPERIMENTAL ZOOLOGY, VOL. 21, NO. 3 



