274 The American Naturalist. [March, 
Between horizontal glass plates the gastrulation takes place so that 
the crescentic blastopore lip appears upon the edge of the lower side of 
the disk-shaped egg, at any point of this periphery. It then travels, in 
some way not observed, across the lower, flat surface, and closes ata 
point of the periphery diametrically opposite to that whence it started. 
Now in sections it is found that the yolk mass is at first at the end 
near the first position of the blastopore, then shifting, lies at the other 
end. 
If the egg were free and not held fast by the pressing plates this 
shifting of the center of gravity would tend to revolve the egg so that 
its lighter colored part would become uppermost. Meanwhile the head 
fold and medullary folds come in near and along the region traversed 
by the blastopore (they are found upon the flat under side of the com- 
pressed egg) and hence would normally appear upon the upper side if 
this rolling of the egg took place. 
Passing over some other interesting observations we may mention 
those made upon eggs that were forced to develop up-side-down. This 
was done by turning them over, under pressure, after the first or second 
cleavages. The light colored part of the egg thus remains uppermost 
The eggs develop normally at first but finally when gastrulation begins 
the blastopore is irregular in shape and the yolk is asymetrically dis- 
tributed so that very imperfect and monstrous gastrulas result. 
An attempt to separate the first two cells of tritons by drawing à 
loop of fine silk about the constriction between them did not succeed, 
since the two cells remained connected by an isthmus. Yet as they 
were held partly apart some curious modifications in the development 
resulted. The results are, however, very diverse. Each cell may 
cleave anda dumbell-shaped blastula result and eventually a monstrous 
embryo formed half upon one side of the thread, half upon the other 
or chiefly upon one side and partly upon the other. The nervous sy% 
tem may be outlined altogether upon one of the parts kept apart by 
the thread. is 
What may be considered the most important part of the paper 18 
that treating of Hertwig’s repetition of Roux’s experiments upon the 
development of frogs eggs in which one of the first two cells is destroyed 
or injured by needle thrusts. 
Such eggs continue to develop, but produce abnormal embryos. 
Roux maintained that the uninjured half of the egg formed @ half 
blastula, half-gastrula, ete. Hertwig claims that this is not the case 
and figures many sections that support his claim very convincingly. 
