FOURTEEN TELEOSTEAN FISHES AT BEAUFORT, N. C. 391 
behind the beginning of the posterior fourth of the body. Heart action is evident 
slightly in advance of the auditory canal, but a blood stream is not visible. 
The newly hatched A. epsetus apparently is more active than most other forms 
that have been hatched in the laboratory at Beaufort. It most frequently descends 
to the bottom of the vessels used in hatching the eggs, but it may swim to the sur- 
face or occupy any part of the available space. In a period of 24 hours after hatching 
the yolk is mostly absorbed, the fish has reached a length of about 4.0 millimeters; 
and in the laboratory, since feeding experiments have failed, the fish usually dies about 
this time. There has been some advancement in the development of the mouth; 
the fin fold remains continuous; the head is still slightly deflected; and no chroma- 
tophores have appeared, the larvae still being highly transparent. (Fig. 9.) 
Specimens 5.0 millimeters long. — Specimens of this length and until the anal fin 
is sufficiently developed to admit the enumeration of the rays, at a length of about 
9 to 10 millimeters, apparently can 
not be separated definitely from A. 
mitchilli. The body remains very 
slender at a length of 5.0 millime- 
ters, and without chromatophores, 
except for a row of about five very small elongate ones situated near the ventral 
outline posterior to the head. Muscular rings remain quite evident and in addition 
to the rings, cross striations are present. The mouth is large; and the gape is quite 
oblique, extending to the eye. The alimentary canal is plainly visible and largely 
separate from the body; that is, it does not appear to be covered by the body wall, 
being attached loosely to the trunk by connective tissue only. 
Superficially, the intestine gives the appearance of being coiled. Upon closer 
examination and dissection it is found, however, that the alimentary tract is almost 
a straight tube and the “coils” are muscular rings in the thin body wall. In some 
specimens the dorsal and anal fins already have become rather indefinitely differen- 
tiated at a length of 5.0 millimeters. The fins do not become evident, however, until 
a length of at least 6 millimeters is reached, and generally the rays can not be defi- 
nitely enumerated until a length 
of about 10 millimeters is at- 
tained. The pectoral fins are 
just becoming evident at a length 
of 5 millimeters although rays 
are not developed. Ventral fins 
are not visible but the caudal fin is partly developed. The notochord is bent slightly 
upward posteriorly, extending into the partly developed caudal fin. (Fig. 10.) 
Specimens 10.0 millimeters long. — The body has become slightly less slender, the 
mouth is still terminal and oblique, and the gape extends somewhat past the anterior 
margin of the eye. Ventral fins are still missing; the dorsal and anal are now suffi- 
ciently well developed to admit of a fairly accurate enumeration of the rays. Since 
A. epsetus rarely has more than 20 rays in the anal fin, whereas A. mitchilli generally 
has about 26, the two species may now be definitely separated by this character. 
The caudal fin is well developed and definitely forked. The notochord is bent 
abruptly upward posteriorly, Ending at the base of the upper rays of the caudal fin. 
Muscular rings are still faintly visible, at least in the caudal region, and pigmentation 
consists of a few dark points on the median line of the chest and along the ventral 
Figure 10. — Anchoviella epsetus. From a specimen 5.6 millimeters long 
Figure 9. — Anchoviella epsetus. Newly hatched fish, 3.6 millimeters long 
