22 J. T. CUNNINGHAM. 
equator of the yolk the total area of the invaginated layer in 
the non-embryonic part of the germinal ring is continually 
‘decreasing in proportion as the embryo increases. There is 
LO reason to suppose that cells once formed in the blastoderm 
are again absorbed; it is much more probable that the cells 
in the invaginated layer are continually multiplying, and the 
only way of accounting for the layer is that as it slides over 
the yolk concrescence of the two halves of the ring takes 
place at the posterior end of the embryonic rudiment; the 
decrease in the total area of the non-embryonic part of the 
invaginated layer is thus caused by the continual addition of 
some of it tothe embryonic part. The gradual folding together 
and concrescence of the diverging limbs of the germinal ring 
is represented in fig. 26. If this concrescence does not take 
place after the blastoderm has passed the equator, the cells of 
the invaginated layer in the non-embryonic part of the ring must 
either be absorbed, or the invagination process must be reversed 
and the ring unfolded, both of which suppositions are improb- 
able. The necessity for the process of concrescence is not so 
evident during the period before the edge of the blastoderm 
reaches the equator of the ovum, but the process is probably . 
continuous from the beginning of the growth of the blasto- 
derm over the yolk. Fig. 23 is intended to picture the way in 
which the concrescence takes place, the embryonic rudiment 
(a B) continually increasing in length as the limbs of the 
germinal ring are brought together at the point 8B. No notch 
like that which is seen at B in this figure is usually visible in 
the real ovum, but a small indentation was observed in two 
embryos by Agassiz and Whitman (1, p. 74) in the same 
position. 
The theory that in Teleosteans and Elasmobranchs the 
embryo was formed by the concrescence of the edge of the 
blastoderm was propounded by His and Rauber (5, 6, and 7) 
some years ago. Their arguments in support of it were 
founded chiefly on the relation of the medullary folds in 
Elasmobranchs to the diverging limbs of the ring. The 
theory was opposed by Balfour (‘ Comp. Emb..,’ vol. ii, p. 254), 
