Leptocephalus acuticeps — Orton 
193 
in this heterogeneous group. Fortunately, the 
larvae of its type species, C. conger ( Linnaeus ) , 
and of two closely related eels from the At- 
lantic and Mediterranean are known. The latter 
two species have had (and are still having) a 
confused nomenclatorial history, but they are 
identifiable under the commonly used names 
Ariosoma bdearica (de la Roche) and Conger - 
muraena mystax ( de la Roche ) . The similarities 
among the definitely identified leptocephali of 
these three nominal genera provide a basic 
standard for defining true congrid larvae, and 
their differences indicate some of the kinds of 
variational trends that one can expect to find 
in related larvae. I have examined Pacific larvae 
of the Ariosoma balearica and Congermmaena 
mystax groups, but have not yet seen Conger 
larvae. The larvae of this central group of in- 
disputable congrids share essentially the same 
basic format. Its conspicuous features include 
the long, straight, simple gut (without regional 
enlargements or other specializatons ) , and the 
short tail with a well-defined caudal fin that 
typically comprises 6-10 caudal rays. These 
larvae differ among themselves in size, propor- 
tional details, somite counts, and in anatomical 
characters of the sort that I have used above in 
the description of Leptocephalus acuticeps (posi- 
tion of the pylorus, length of the kidney, etc.). 
The quite different color patterns of these three 
kinds of larvae indicate that the true congrids 
have undergone considerable evolutionary diver- 
sification in larval pattern. Conger conger has a 
pair of ventrolateral rows of melanophores, ap- 
parently external, paralleling the gut; a mid- 
lateral row, also apparently external, on each 
side of the body axis; and several large melano- 
phores on each side of the pericardium. Conger - 
mmaena mystax has the ventrolateral and peri- 
cardial pigment, but lacks the midlateral row. 
In both of these larvae, the melanophores are 
relatively large and conspicuous. Ariosoma bale - 
arica differs sharply, both in the pattern itself 
and in the very small size and dense spacing of 
the melanophores. This pattern includes densely 
crowded rows of tiny melanophores externally 
along the middorsal and anterior midventral sur- 
faces, and internally along the top of the gut. 
In place of the simple longitudinal row of large 
melanophores along the midlaterai surface, A. 
balearica has an elaborate lateral surface pattern 
composed of a uniform series of short, oblique, 
parallel black lines just below the midlateral axis 
along nearly the full length of the larva. Each 
of these short black lines consists of a dense row 
of minute melanophores placed lateral to the 
myocomma between two contiguous somites. 
Since the row of cells marks the section of the 
myocomma that lies just below the midlateral 
axis, the row therefore conforms to the oblique 
ventrocaudal orientation of this part of the myo- 
comma. The diagfammatically repetitive com- 
position of the pattern results from the regular 
presence of a row of cells on almost every 
myocomma. 
A typical congrid larva is readily identified as 
such, for it has morphological and color-pattern 
characters that are consistent with the trends 
indicated in this basic group of known larvae. 
For example, the Scripps collection includes 
many eastern Pacific leptocephali that are easily 
allocated to the Congridae, and it is evident that 
they include at least a dozen different kinds 
though few of these can yet be identified with 
named adults. The color patterns of most of 
these kinds of larvae are simple modifications of 
the Conger conger type. 
It is not known whether larvae of all true 
congrids conform strictly to the format of this 
basic group of identified larvae, or how widely 
a larva may depart from this type in morpho- 
logical and pattern characters and still retain 
recognizable evidence of congrid affinity. How- 
ever, some idea of the limits within which con- 
grid larvae might evolve (and, thus, whether 
Leptocephalus acuticeps might belong here) can 
come from study of the problem groups that 
have been interpreted variously by different 
authors. The heterocongrid eels are an instruc- 
tive example. Whether the heterocongrids (the 
garden eels, or tube eels) are best retained in 
the Congridae or interpreted as a separate but 
closely related family is still under debate in the 
literature. Known heterocongrid larvae ( Lepto- 
cephalus magnaghii from the Red Sea, and 
closely similar larvae from the eastern tropical 
Pacific) differ from larvae of the Conger - 
C ongermuraena- Ariosoma complex in some re- 
spects (e.g., higher somite counts, relatively 
shorter gut, and more anterior origin of the 
