163 



At this stage, Fig. 7 B, a divergence of one of the groups of the 

 chordata occurred. In the ancestral Balanoglossus, as in its descen- 

 dants to-day, there was no premature dying out of the lateral and 

 ventral parts of the ring of deuterogenetic tissue, and so it has re- 

 sulted in the very invertebrate worm like Balanoglossus without a "tail". 



On the other hand the dying out of the ventral and lateral parts 

 of the deuterogenetic area of activity has led in all the other groups 

 (omitting from consideration the so-called Diplochordata) to a ridding 

 the hind end of the body of the gut and to the formation of a tail as 

 an important organ of locomotion. Probably to this fact more than 

 any other has been due the present supremacy of the great Vertebrate 

 phylum. 



It would take too much space to indicate in every case the parts 

 of an actual embryo which correspond to the white and shaded por- 

 tions of these diagrams (Fig. 7), but roughly my figure in Kopsch's 

 paper (17) gives it for the bird while for the frog it is still more im- 

 perfectly represented in my paper 1894, Fig. 14, and for the rabbit 

 Fig. 44. It must be recollected (as Hubrecht says) that during the 

 growth of the embryo the parts derived from one area become shifted 

 so as to overlap those formed by the other centre. 



If these two centres, however, are clearly recognised it will be 

 seen that only the gut cavity within the white area, Fig. 7, is strictly 

 speaking "archenteron" and therefore lined by endoderm — and 

 therefore it is only that part of the notochord which is formed from 

 this (if there is any) which can be formed of true endoderm. Every- 

 thing behind this is addition — is post-gastraea and belongs to a later 

 epoch and is not derived from the original germinal layers but from 

 the secondary growth centre. 



Fig. 8 A is a diagrammatic drawing of an embryo of a verte- 

 brate to shew some of the organs, and arranged so as to correspond 

 with the still more diagrammatic Fig. 7C. 



Fig. 8B is less diagrammatic and shews the actual relations in 

 the embryo of an amphibian. Owing to the active formation of the 

 more dorsal part of the deuterogenetic influence, and to the great sub- 

 sequent interstitial growth of the parts derived from this dorsal portion, 

 the primary radial part of the embryo becomes distorted as indicated 

 by the comparison of the lines x — x. 



Fig. 9 indicates similarly the Amniote condition. 



The embryology of the chordates shews clearly that there was a 

 radially symmetrical organism, — a "gastraea" or coelenterate stage; 

 that the single aperture of that time became the anus; there was no 



11* 



