56 THE EVOLUTION OF CONTINUITY 



" cell-caps/' and their presence is explained as follows : When 

 a cell is about to divide, " a ring-like cushion of cellulose 

 is formed on the inner surface of the cell-wall near the upper 

 end of the cell (Fig. 8) and the nucleus of the cell divides into 

 two. The cell-wall then splits all round just outside the 

 cellulose cushion, and the latter is stretched out so as to give 

 rise to a membrane which is intercalated in the cell-wall 

 (Fig. 8, b). The dividing wall, laid down between the two 

 nuclei, is formed opposite the lower end of the intercalated 

 membrane. The bounding wall of the upper cell therefore 

 consists chiefly of the intercalated membrane ; but there is 

 a portion of the old cell -wall, fitting like a cap, at the upper 

 end of the cell, where it produces a transverse ring-like 

 mark. If the process is repeated, the new cellulose cushion 



a- 



hrctlahi 



lM»'bt'»Tl '• 



Cill ksJrNTiV,). 

 Fig. 8. — The^fonnation^of_cell-caps in /Edogonium. (After Lowson.) 



arises immediately below the previously formed ring. In 

 this way cells frequently come to show a series of ' caps ' 

 or rings at their upper ends (Fig. 8, d)." (Lowson.) 



From all this we can deduce that in iEdogonium cell- 

 division proceeds on the plan suggested in the case of Spiro- 

 gyra ; that temporary cell-arrest is experienced by the 

 proximal result of a cell's division, the distal result alone 

 growing to divide for the time being ; and that none of the 

 uncapped cells in the filament have performed an act of 

 division. And it is to be presumed that when an uncapped 

 cell is released from arrest proximal cell-arrest is a feature 

 in the formation of the resulting intercalated series. 



Thus if we draw for iEdogonium a figure on the lines 

 of Fig. 7, we have a filament in which appear cells provided 

 with a varying number of caps, and between them others 



