SEGMENTATION OF THE OVUM OF THE SHEEP. 237 



fig. Vg, PI. 18 be as likely an interpretation of his fig. 36 as 

 that given by his diagram B^ p. 106 ? 



This stage is at once succeeded by one in which a rapid 

 increase in the mass occurs at this point, which Duval con- 

 siders to be entirely due to the activity of the outer layer. 

 So great is the increase that it is impossible to distinguish 

 between it and the hypoblast ; there is "une soudure des deux 

 feuillets avec engrenetnent des cellules de Pun dans celles de 

 I'autre.^' Is it decisively proved that the intermediate cells 

 take no part in this increase ? I hardly think it is, and suggest 

 that his fig. 41 is equally open to the interpretation given by 

 my fig. V^. 



The subsequent history tends to support my interpretation. 

 In speaking of the increase in bulk of the general embryonic 

 mass Duval says, p. 430, " La cavite amniotique du Murin 

 prend naissance d^une maniere singuliere. II se forme d'abord 

 un epaissisement massif de Pectoderme, et ce processus rapelle 

 la masse amniotique pleine du cochon d'Inde ; mais au lieu 

 que cette masse se creuse, comme chez ce rongeur, d'une 

 cavite centrale close de tons cotes, elle se disloque irreguliere- 

 ment, chez le Murin, et s'ouvre h la surface, de Poeuf, figurant 

 une bourse largeraent etalee, dont les bords se relevent alors 

 selon le type classique de replis amniotique et produisent 

 I'occlusion de I'amnios par leur rapprochment et soudure." 



He then explains how after the roof of the amniotic cavity 

 has been separated off in small fragments from the solid floor 

 (figs. V5, Vg, PL 18) the true amnion is eventually formed by the 

 advance and fusion of the free edges of the true epiblastic mass 

 as shown in diagram Vy. 



On my interpretation, the rapid growth of cells which 

 produces the ''amniotic mass'' is due chiefly to the inner mass 

 cells, which cause the rupture of the overlying trophoblast 

 cells which now grow rapidly for a time, and form irregular 

 pieces which ultimately are thrown ofi" and lost, leaving the 

 epiblast exposed on the surface. It has no connection at all 

 with the amnion formation, which is brought about subse- 

 quently by folds of the true epiblast, and not by trophoblast. 



VOL. 41, PAKT 2. — NEW SERIES. R 



