56 
As a corollary, one would expect large eche- 
neids to select large hosts or, in their absence, 
to revert to free swimming. The latter appears 
to be the case with Echeneis naucrates, 39-inch 
individuals of which are free-living around piers 
at Eniwetok, Marshall Islands (Strasburg, 1957: 
60). The 640-mm R. remora listed in Table 2 
confirms the first part of the hypothesis, for 
it was one of about three dozen individuals of 
similar size accompanying a 50 to 60-ft whale 
shark, Rhincodon typus, at Maro Reef (north- 
west of the main Hawaiian Islands). Attempts 
to lure this shark to the fishing vessel were un- 
successful because the chopped fish bait was 
intercepted by the darting remoras. Twelve 
remoras v/ere caught by pole and line using tuna 
flesh as bait, but only a single specimen was 
preserved. Its stomach was empty, unfortunately, 
so that no light can be shed on its diet. It is 
doubtful that such a large fish could subsist 
mainly on its host’s ectoparasites (cf Strasburg, 
1959:246). 
Some of the smaller specimens listed in Table 
2 are the smallest reported representatives of 
their species. The 27.1-mm brachypterus is con- 
siderably shorter than Gudger’s ( 1928) 77-mm 
fish, and my 14. 3 -mm osteochir is smaller than 
his 36-mm specimen. Gudger presents few de- 
scriptive data, however, and his specimens seem 
to differ from what are here called brachypterus 
and osteochir . On the basis of pectoral counts, 
Beebe’s (1932) 15 -mm f( Remora remora ” is 
Rhombochirus osteochir, while his 88 -mm speci- 
men is correctly identified as R. remora. 
Beebe (ibid.) also misinterpreted certain 
morphological peculiarities of the lips and jaws 
of his small osteochir . These were stalked cup- 
like structures which he termed "suckers,” and 
which he postulated were used for host attach- 
ment prior to the development of the cephalic 
disk. Actually the "suckers” on the mandible are 
the enlarged fleshy sockets which normally bear 
the fangs, while those on the upper lip are 
merely large pores. The fangs are easily ex- 
tracted with forceps, imparting a sucker-like 
appearance to the supporting tissues. My 14.3- 
mm osteochir has seven outer and two .inner 
fangs on each side of the mandible, making 
a total of 18 sockets. This is a reasonable ap- 
proximation of the 20 "suckers” and four fangs 
which Beebe found in his 15-mm fish. 
PACIFIC SCIENCE, Vol. XVIII, January 1964 1 
The young echeneid’s need for its relatively 
enormous fangs is presently inexplicable. These 
teeth are not deciduous but instead become in- 
conspicuous through overgrowth of the gums ; 
and the appearance of other teeth between them. 
Its intimate association with another species 
could impose serious reproductive restrictions 
on an echeneid. Unless both sexes attached to i 
the same host individual, spawning would be 
limited to times when the host species ag- 
gregated. Echeneids attached to hosts which 
schooled or congregated to feed would have 
many opportunities for mating, but those ac- : 
companying solitary hosts would have to spawn 
simultaneously with them. 
The material at hand was examined for gonad 
maturity and the presence of both sexes on the : 
same host individual. Unfortunately, the speci- 
mens were not always so segregated that the 
latter could be determined. Thus, although ma- 
ture Remora remora and Remoropsis pallidus \ 
were found, it was not certain whether both i 
sexes had been attached to the same shark or j 
marlin. The available Remoropsis brachypterus 
and Phtheirichthys lineatus were few and sex- 
ually immature. 
In contrast to the above, the Rhombochirus 
osteochir data present a relatively clear picture 
of the physical distribution of the sexes. Thir- 
teen marlin and spearfish bore 2 Rhombochirus 
apiece. In 10 cases the 2 were a male and female 
of the same size and degree of maturity (7 
pairs were judged to be ripe, based on abdomi- 
nal distension of the female) . The 11th pair was 
a small female and a fish half her size whose 
sex could not be determined. The remaining 2 
couples consisted of young fish of undetermined 
sex. A 14th host was accompanied by 3 Rhom- 
bochirus, a ripe male and female and a small 
individual, one-fourth the length of the others, 
whose sex could not be ascertained. 
With 11 out of 14 pairs bisexual and in the 
same maturity stage, it would seem that Rhom- 
bochirus can reproduce quite independently of 
its host’s aggregating habits. This does not mean 
that it always does so, for mature single Rhom- 
bochirus were also collected. In such cases, 
however, there is a possibility that these fish 
were remnants of pairs the other members of 
which had detached during the capture of the 
host. 
