PERMEABir.ITV OF SURFACE LAYER OF CELLS. 2C>3 



plasma-membrane ; the latter is, temporarily at least, thinned and 

 its permeability will therefore be increased. This increase in 

 permeability is, on the present view, the primary event in the initia- 

 tion of cell-division. 



Why should such increase in permeability --granting that this 

 is the primary change induced by the above various forms of 

 treatment initiate so apparently complex a process as mitotic 

 cell-division ? The main factors in producing this effect are in my 

 opinion two : first, a disturbance of chemical equilibrium due to 

 an increase in the rate at which certain metabolic products (prob- 

 ably chiefly carbon dioxide) are lost from the cell ; this is an 

 effect similar to that which, on the general theory of stimulation 

 outlined above, underlies the chemical effect of stimulation in 

 muscle ; the precise effect of such a change will of course vary 

 from cell to cell. Second, a definitely localized increase in the 

 general surface-tension of the cell in consequence of a loss or 

 lowering of the electrical surface polarization. 



If the entire cell be regarded as a drop of fluid with a distinct 

 tension at its boundary surface, it is apparent that a relative increase 

 in surface-tension over the general surface of the two hemispheres 

 combined with a relative decrease at the equator would result in a 

 drawing of the material toward the two former areas and away 

 from the equatorial region : a division of the drop into two might 

 thus result. Robertson has recently drawn attention to this pos- 

 sibility, and by a simple but ingenious experiment has shown 

 that a floating oil droplet may be made to divide symmetrically 

 into two by locally lowering the surface-tension along its equator 

 by means of a thread containing alkali or soap solution. 1 



1 Robertson, Archiv fur Entwicklun^smechanik, 27, 1909, p. 29. 



It would seem that the effect of locally increasing the surface-tension would vary 

 with the mobility or fluidity of the substance composing the system. If the surface 

 layer possessed a high degree of viscosity approaching solidity an increase in surface- 

 tension over an equatorial band-like area encircling the cell might produce constric- 

 tion just as would occur, e. g., in an inflated rubber ball if the tension were suffi- 

 ciently increased about its equator. This was Biitschlr s supposition, which was later 

 favored by Erlanger and Conklin ; and by myself in an early paper. Dividing cells, 

 however, act very plainly like fluid systems (at least in many cases, e.g., starfish eggs), 

 and there are many theoretical grounds for preferring the above point of view (see 

 below). If the egg behaves like a drop of fluid the distribution of surface-tension 

 assumed above would produce the observed change of form. 



