846 THE AMERICAN NATURALIST. 
If, now, we figure the matter up, we find that the distance 
the gill openings pass tailward (contrasting embryos vary- 
ing from about 20 to 300 mm.), added to the distance which 
the anterior rim of the pectoral passes headward, makes the 
considerable sum of 20 per cent of the length of the animal. 
And from this it follows that there occurs in this region a 
process of “shortening up," which is surely enough to account 
for the presence of anastomosing vessels, blended muscle bands, 
plexus of nerves, and “collectors.” The presence of such com- 
plicated structures,! especially in the region of the anterior or 
posterior rim of a (pectoral) fin, is therefore to no little degree a 
product of the mechanical needs of fin migration. 
COLUMBIA UNIVERSITY, October 17, 1902. 
! Another complicating feature to be considered in this connection is the dispo- 
sition of segments arising in the caudal region and growing cephalad. In the 
earlier embryos (15 to 20 mm.) the total number of segments behind the last gill arch 
is enclosed within a space measuring about 80 per cent of the entire length; later 
specimens (300 mm.) include the segments within a space of 76 percent. The size 
of the embryo at which the formation of new segments ceases has not been deter- 
mined in the present form, nor is it as yet ascertained in kindred forms. 


