44 K- M ITS ü KU RI; 



of the epiblastic layer shows in this section a median 

 indentation which goes upwards and the mass below 

 seems to become dirccth/ continuous witli the yolk-plug 

 — an arrangement different from what obtains in g. It 

 would be reasonable to suppose that the primitive streak 

 ends posteriorly at the point Avhere the arrangement as in 

 g (i.e., the fusion of the epihiast proper with the mass 

 below) ceases and that the arrangement as in /(- represents 

 the yolk-plug standing u]) from below like Series I, / or 

 Series III Ic. The limit is very difficult to define, as we 

 must consider all this as existing in a rudimentary condi- 

 tion, but I should say it is present about two sections 

 in front of /;. 

 i is the third section behind //. The epiblast is now entirely 

 independent of the mesoblast, but there is still a faint 

 median elevation (indicated by two stars) which is the 

 continuation of the 37olk-plug. The mesoblast also sends 

 up a median projection towards the epiblast. 



I think, it is easy to see from this series that the course 

 of events is very much less primitive in Triomix than in the other 

 two species. 



Series V. 



(PI. V.) 

 Chelonia caouana. 



An cmhnjo with 9 mesohlastic somites ; taken out 9^ J 2 dai/s after 



deposition, represented in Fig. 5 (PI. ].), 5a (PL II.) é ob (PI. I.) 



(BU ^>) 

 This is much older than ;iny of the series above described and 

 is one of the most important. The yolk-plug has now receded a 

 considerable distance backwards. 



