86 Geometrical relation of Nuclei 



problem is necessarily further complicated by the interaction of 

 the numerous force centres, though probably this effect is less 

 where yolk is abundant than where it is absent. 



Some deny the occurrence of any form of invagination in Rana. 

 With them I would agree, except in so far as the withdrawal of 

 the yolk plug inwards from the blastopore (anus of Rusconi) to 

 form the gut floor at a later period may be called an invagination. 

 The cases of the Urodela and the Cyclostomata appear to be rather 

 different. Here it looks as though there were at a rather earlier 

 period a distinct inrolling of the lower pole to form the ventral 

 wall of the archenteron really very much like the withdrawal of 

 the yolk plug just mentioned. Could not the mathematician 

 decide whether this can be or must of mechanical necessity be so ? 

 Could not the mathematician decide whether, on the hypothesis 

 of universal cell attraction put forward above, the combined effect 

 of attraction between cell and cell could account for the alleged 

 inrolling of the yolk mass in certain cases ? 



To return for a moment to the differentiation zone alluded to 

 in the last paragraph but one. 



After this has disappeared from the surface, that is to say, after 

 the complete formation on the surface of the circular blastopore or 

 anus of Rusconi, what is the subsequent history of this differen- 

 tiation zone? Does the differentiation cease? Or does it con- 

 tinue till it dies out by coalescence of all parts of the zone, during 

 which time a complete layer is differentiated as gut epithelium? 

 That it continues for a while, causing a split as suggested in my 

 paper, 1894, I am confident, but it is a moot point whether it is 

 not in some cases brought to an end by the confluence of the split 

 produced thereby with the original segmentation cavity as main- 

 tained by Hertwig, etc. 



Brachet's careful work on Siredon and Rana temporaria makes 

 it probable that confluence is the rule in the former as it seems to 

 be in the Gymnophiona (Brauer) and may or may not occur in the 

 latter. But in either case the true archenteron is of protogenetic 

 origin and is the immediate result of the simple working of forces 

 above postulated, guided by extrinsic factors. Probably this could 

 be demonstrated mathematically. 



So also in Amphioxus the gut cavity is brought into being by 



