570 WILLIAM PATTEN. 



formly, excepting the minute^ faint granules, and the small, 

 darkly-stained nucleolus. 



A spindle metamorphosis was not observed in the embryo, 

 although the nuclei often contained two nucleoli, and were 

 probably undergoing rapid division. The small size of the 

 nuclei, however, rendered them unfavorable for the study of 

 such phenomena. 



These different kinds of nuclei are distributed irregularly 

 throughout the ventral plate, and are so constant in their cha- 

 racters that it is difficult to find stages which might be con- 

 sidered transitional. The first two varieties possibly represent 

 stages preliminary to division. It should be noticed that the 

 ordinary nucleus in a state of division presents no other pecu- 

 liarity than the presence of two nucleoli. 



It will be seen on examining PI. XXXVIb, figs. 8, 9, and 10, 

 that the nuclei of the ventral plate are uniformly located at the 

 inner or deep ends of the cells, while those of the amnion are at 

 the outer ends, thus maintaining in both layers the same mor- 

 phological position. From this we can infer that the amnion 

 is simply a fold of the thickened ventral plate, and the serosa 

 merely the slightly modified blastoderm covering the remainder 

 of the egg, and which gradually extends over the ventral plate 

 in a fold, whose lips finallv meet and fuse, thus producing two 

 completely closed sacs. 



It can be seen, in PI. XXXVIb, fig. 8, when the folds are first 

 forming, that the inner limb {am) has the same thickness as 

 the ventral plate itself, and also that the nuclei have the same 

 morphological position, while, on the other hand, the outer 

 limb {sr) is very thin, being simply a continuation of the blas- 

 toderm. These relations are still perceptible in much later 

 stages (figs. 9 and lOj, even when the membranes are com- 

 pletely formed (fig. 10). Thus, according to this view, the 

 real edge of the ventral plate is not where the amnion joins 

 the ventral plate, but -where the amnion is continuous with the 

 serosa. 



At present I cannot claim any special morphological value 

 for these distinctions^ since all the parts in question are directly 



