168 



"gastrulation" is the formation of a cavity, as the direct result of the 

 primary growth centre, which cavity represents the gut cavity of the 

 original radially symmetrical animal. To this alone the term arch- 

 enteron should be applied, and to the lining of this cavity alone the 

 word endoderm. In Amphioxus and many invertebrates it is brought 

 about by actual invagination ; in Amphibia by a splitting due to the 

 continuance of the segmentation process among the more superficial 

 of the inner segmentation spheres; in Amniotes by the accumulation 

 of fluid among the segmentation spheres. Any rolling over of cells 

 such as may occur in, for instance. Amphibia while the lip of the 

 "blastopore" is forming is due to the activity of the secondary growth 

 centre and so should not be reckoned as "gastrulation". 



On this view the subgerminal cavity is archenteron for it is wholly 

 protogenetic, while the so called gastrulation canal of a reptile is 

 wholly due to the secondary growth centre, that is, it is deuterogenetic 

 and is more suitably named "neurenteric" canal. 



Beard in certain of his interesting papers has referred to the 

 two growth centres in Vertebrate embryos and sees in them confirm- 

 ation of his idea of a sexual generation formed upon an asexual 

 larva, or phorozcon as he terms it. 



Quite apart from the merits or demerits of the view of alternation 

 of generations I would point out that experiment shews conclusively 

 that the definitive animal, i. e., "adult" contains parts due to both 

 growth centres. It is quite impossible to separate the "yolk sac" from 

 the anterior part of the adult developmentally ; they are derived from 

 the same thing, both are due to protogenesis. 



Postscript, Mr. Sedgwick has kindly drawn my attention to 

 his remarks in his recently published volume of "The Students Text- 

 book of Zoology" p. 4 and in his article on Embryology in the "Times", 

 supplement to the Encyclopaedia Britannica in which he lays stress 

 upon Graham Kerr's statement that the medullary folds envelop the 

 blastopore in Lepidosiren (Phil. Trans., 1900). This would not seriously 

 militate against the views expressed in the foregoing article, for the 

 crux of the problem is the mouth formation and the position of the 

 growing point relative to the pre-existing part of the embryo; while 

 the envelopment of the blastopore by nerve tissue would accord with 

 the idea so generally held of a ring of nervous tissue round the 

 gastrsea mouth. 



But similar statements have been made with reference to other 

 vertebrates, and an exactly comparable figure was given by Marshall 

 for the frog (Vertebrate Embryology, p. 113), and I cannot feel quite 



