844 THE AMERICAN NATURALIST. | [Vor. XXXVI. 
correspondence so close that we can interpret it only in terms 
of cause and effect. 
The dorsal fins are even more conservative in position than 
the ventrals. In the first place they maintain practically the 
same relative distance from one another, measured from the 
anterior rim of the fins! : thus, ina specimen of 21 mm. the inter- 
val measures about 23 per cent, of 40 mm. about 22 per cent, of 
60 mm. 23 per cent, of 82 mm. 25 per cent, of 105 mm. 26 per 
cent, of 197 mm. 27 per cent; in other words, between the two 
fins there is but a slight variation in the interval (say, 5 per cent 
of the entire length) in stages widely different in size (e.g., 
measuring between 20 and 200 mm.). In general, however, we 
note that the growth of the anterior rims of first dorsal and 
ventral are indicated by parallel lines, and similarly the growth 
of the anterior rim of the second dorsal and the posterior rim 
of the ventral. 
SUMMARY. 
The foregoing characters yield what seems to me convin- 
cing evidence that the pectoral fin is subject to changes with 
respect to the gravity center, z.e., physical changes, which cause 
it to become more highly specialized than the ventral fin, 
and that the latter is conservative, after the fashion of the 
unpaired fins. The result of biometric data, it will be seen, 
confirms strikingly the views of the fin-fold theorists, most 
clearly expressed in this particular regard by Wiedersheim in 
his *Gliedmassenskelet." On the other hand, the present study 
yields no evidence that there has ever been a migration of the fins 
in the Gegenbaurian sense; thus, we find no reason to believe 
that the ventral fin is a structure which has shifted its position 
from in front hindward ; it is indeed in about the same position 
fore and aft in the adult as in early embryos (20 mm.). It is 
important, furthermore, that the only fin (leaving anal and caudal 
out of discussion) whose anterior margin passes forward is the 
pectoral, since with this condition is correlated the fact that this 
*In the younger stages the measurement was made from the actual most 
anterior point of the fin; in the later, when a spine appears, from the point where 
the anterior fin rim, if produced, would intersect the dorsal-most line of the body. 
