82 
PACIFIC SCIENCE, Vol. VIII, January, 1954 
template. However, whether or not such an 
intermediate group is discovered eventually, 
it seems probable that a more thorough knowl- 
edge of the internal and external characters of 
brotulid genera will make necessary a rather 
drastic revision of any classification that has 
hitherto been erected. 
GEOGRAPHIC AND BATHYMETRIC 
RELATIONSHIPS OF HAWAIIAN 
DEEP-WATER BROTULIDS 
The fact that none of the five Hawaiian 
species of deep-water brotulids can be cer- 
tainly allocated to hitherto described forms 
is striking. It could indicate an endemic Cen- 
tral Pacific brotulid fauna. In the light of 
shallow-water fish distribution— including 
that of Brotula multibarbata^ which ranges 
from South Africa to Hawaii — this seems im- 
probable. Or it might indicate that the Mauna 
Loa lava flow has brought to the surface 
wide-ranging brotulids that live in habitats 
heretofore unsampled. However, in view of 
the fact that most of the other described 
brotulids have also been taken only once, it 
most probably serves merely as a further dem- 
onstration of how inadequately brotulids have 
been collected. If this last is the correct ex- 
planation, the inference is that the Brotulidae 
will eventually prove to be one of the larger 
fish families. 
Turning to genera, Volcanus has been hith- 
erto unknown. The distribution of the species 
presumably related to the Hawaiian species 
of the other three genera is shown in Table 3. 
Aside from pointing out the wide extent and 
variety of these generic distributions, little 
can be said. Cataetyx hawaiiensis suggests af- 
finities with the tropical West American fauna 
and Fycnocraspedum armatum with that of the 
Indian Ocean, but in the present state of 
knowledge such suggestions are purely ten- 
tative. Perhaps a more curious aspect of the 
Hawaiian forms is their almost complete lack 
of affinity with the best known of brotulid 
faunas — that of the Philippines (Radcliffe, 
1913). 
TABLE 3 
Distribution of Species Related to the Hawaiian 
Deep-Water Brotulids 
SPECIES 
LOCALITY 
DEPTH 
Fathoms 
Diplacanthopoma 
hrachysoma 
Brazil 
350 
rivers-andersoni. . . . 
Arabian Sea 
947 
rankeps 
Andaman Sea, 
Gulf of Aden 
405-600 
alcock i 
Andaman Sea 
490 
jorduni 
Galapagos 
385 
brunnea 
Philippines, 
Arabian Sea 
375-1,050 
nigripinnis 
Natal 
700 
Cataetyx 
messieri 
Chile 
345 
rubrirostris 
California 
205-359 
Pycnocraspedum 
squamipinne 
Bay of Bengal, 
Zanzibar 
130-250 
Regarding depth distribution, Table 3 in- 
dicates Pycnocraspedum to be a shallower- water 
genus than Cataetyx or Diplacanthopoma. In 
line with this, the Hawaiian specimen of Pyc- 
nocraspedum was among the earliest of the 
fishes collected after the lava flow entered the 
water. It also maintains a brownish coloration, 
whereas the others range from gray to blue 
black. 
REFERENCES 
Alcock, a. W. 1889- On the bathybial fishes 
of the Bay of Bengal and neighbouring 
waters obtained during the seasons 1885- 
1889. Ann. and Mag. Nat. Hist. VI, 4: 
376-399, 450-461. 
■ — 1895. On a new species of viviparous 
fish of the family Ophidiidae. Ann. and 
Mag. Nat. Hist. VI, 16: 144-146. 
1895.^ Illustrations of the zoology of the 
royal Indian marine surveying steamer ''Inves- 
tigator.'' Fishes. Calcutta. [The plates in this 
work were issued at various times between 
1892 and 1909; those referred to in the 
text seem to have appeared about 1895.] 
1898. A note on the deep-sea fishes, | 
with descriptions of some new genera and j 
species, including another probably vivi- 
