I GASTKULATION 41 



alluded t<>, has in these later stages sometimes proceeded so tar 

 t hat that cavity is nearly completely walled in by yolk-cells. 



While the archenteric cavity increases in volume the segmenta- 

 tion cavity UMMMHUS gradually reduced. Normally the latter cavity 

 goes on shrinking until it is finally obliterated but according to 

 0. Schultxe (1887) a certain proportion of eggs show a variation 

 from the normal which appears to be of importance for the inter- 

 pretation of what happens regularly in certain other groups. As 

 seen in Kig. _.", Iv the layer of yolk-cells which separates the 

 archenteron from the segmentation cavity is liable to become 

 extremely thin, and Schultze believes that in certain cases this thin 

 partition breaks down and disappears, so that the archenteric and 

 segmentation cavities are thrown into one. What appears at first 

 sight to be the archenteric cavity of subsequent stages would in such 

 cases be really a complex consisting of the true archenteron fused 

 with the remains of the segmentation cavity. 



If these observations are to be depended upon, they are of very 

 special interest. For, if the confluence of archenteric and segmenta- 

 tion cavity really occurs as an occasional variation in such 

 Amphibians as the Frog, this may be taken as a foreshadowing of 

 the similar phenomenon which has become a normal characteristic of 

 the development of many of the higher Vertebrates. 



It must however be borne in mind that there exists a dangerous 

 source of possible errors of observation, which it is difficult to guard 

 against, namely that when an egg of the stage of Fig. 25, E, is trans- 

 ferred from one fluid to another, as in the ordinary technical pro- 

 cesses which precede section-cutting, violent diffusion currents are 

 set up between the fluid in the segmentation cavity on the one hand 

 and that in the archenteron on the other, and these currents must 

 be very liable to cause rupture in the intervening partition, even 

 when in life this is quite continuous. 



As gastrulation nears its end the circle formed by the gastrular 

 lip becomes gradually smaller. Finally its lateral edges come 

 together so that it takes the form of a short longitudinally placed 

 slit, the remains of the yolk-plug at the same time passing out of 

 sight. The gastrula is now complete. 



As regards the subsequent fate of the slit-like blastopore it may 

 be mentioned that, for the most part, it becomes obliterated by fusion 

 of its two lips. The portions at its two ends, however, remain open 

 as two pores of which the more anterior becomes the neurenteric 

 canal while the posterior becomes, either directly or after temporary 

 obliteration, the anus. 



The process of gastrulation in the majority of Anurous and 

 Urodele Amphibians pursues a course similar in its main features to 

 that of the frog. Detailed studies of the process are, however, 

 urgently needed in those Amphibians which have particularly small 

 eggs and in which therefore gastrulation is less modified by the 

 presence of yolk. 



