DEVELOPMENT OE TEACIIINUS TTPEKA. 287 



have not been able to trace it tbrougb all its phases. A strand 

 of cells extends from the point where the future anus will open, 

 upwards and inwards to meet the mesenteron, and the proctodeum 

 is iDrobably invaginated along this short strand of cells, but no 

 lumen is formed up to at least four days after hatching. The 

 constriction in the alimentary canal which is shown in figs. 27 & 

 29 is not, as might be at first supposed, evidence of the junction 

 of the proctodeum and the mesenteron. It is in reality a fold in 

 the mesenteron itself produced by a bending down of its apex 

 in consequence of the development of the embryo, and is more 

 marked in some individuals than in others. It perhaps may be 

 said to mark ofl" the cloacal part of the mesenteron. The mesen- 

 teron, which has widened out up to the point where the liver 

 arises, suddenly narrows, as seen in side view, and extends as a 

 comparatively narrow tube into the region of the head. The 

 changes which take place in the respiratory section of the mesen- 

 teron are exceedingly difficult to follow in the living embryo, 

 owing to its being curled up within the shell, and I propose to 

 leave a discussion of this portion until I have made an investi- 

 gation of stained specimens and sections. 



The invagination of the stomodeum does not seem to take place 

 until after the embryo has left the shell. Three pairs of bran- 

 chial clefts were observed on the eighth day. The first pair, the 

 hyomandihular, is developed about the time the heart begins to 

 pulsate. 



The heart has the simple form shown in fig. 21 when pulsation 

 first begins, and the lumen is not very well marked. Its wider 

 part indicates where the future venous end will be. Its develop- 

 ment and separation of a ventricle will best be understood by a 

 comparison of figs. 19-26. 



The eyeballs begin to be pigmented some hours before the 

 first specimens of a batch are hatched, and those which are 

 hatched out later have them fully and darkly pigmented before 

 leaving the egg. 



After the blastopore has closed, the tail is developed as a /ree 

 proloDgation of the vertebral column ; and as soon as it has fairly 

 left the yolk, the first folding of the epiblast to form the conti- 

 nuous embryonal dorsal and ventral fin is to be seen. After the 

 pulsation of the heart begins, the development of the tail is more 

 rapid, and its gradual extension over the yolk will be easily fol- 

 lowed by a comparison of the accompanying figures, up to 'the 



