273 
Parathunnus in Hawaiian Waters— -B rock 
length by the method of least squares, when 
projected for the sizes smaller than those in- 
cluded in the Hawaiian data, indicates that the 
smaller fish would have, relatively, a much 
longer pectoral fin. It would seem futile, there- 
fore, to diagnose tunas by such characters as 
fin lengths without first examining the fins 
throughout the size range of a species. 
Table 2 contains measurements and counts 
of 20 specimens made in the Honolulu fish 
markets during the early summer of 1948. The 
method of measuring and counting described 
by Godsil and Byers (1944: 125-128) was 
followed. All measurements were made with 
large calipers with one fixed and one sliding 
arm, held parallel. As may be noted in Table 2 
characters listed were not determined for all 
fish. This was because the measurements were 
made during an auction of the fish, the only 
practical time and place in which to measure 
them, and fish would occasionally be purchased 
and butchered before measurements could be 
completed. Similarly, it was not possible to 
determine the sex of all fish since sexes were 
determined by observation while a dealer 
butchered his recently acquired merchandise. 
The computations for the regression lines 
for the various characters given in Table 2 have 
not been given here, since such comparisons 
of regression lines can be made with comparable 
data obtained for other localities and by statis- 
tical methods that seem appropriate. It does 
not seem worthwhile to suggest the design of a 
statistical scheme of analysis here by computing 
part of it, especially since most workers would 
prefer to take the field data as given in Table 2 
as their starting point. 
Table 1 and Figure 1 show that the largest 
landings of Parathunnus are made during the 
winter months. Many of the fishermen alter 
their gear during this period to increase the 
catch of this species. The Hawaiian catch of 
Parathunnus is taken on flagline or longline 
gear. This is an unanchored set line with hooks 
at approximately 30-fathom intervals and floated 
by buoys in deep water. The hooks are attached 
to leaders up to 20 fathoms long in fishing for 
Neothunnus during the summer months. In 
fishing for Parathunnus, and Neothunnus too, 
during the winter months these leaders are 
often lengthened several fathoms. For a de- 
tailed description of longline gear see Shapiro 
(1948: 40-44). Parathunnus, aside from the 
occasional capture of small individuals, is rarely 
taken by surface fishing techniques such as 
trolling or fishing with live bait. 
This information would imply that Para- 
thtmnus is not a surface fish but that, at least 
during daylight, it feeds in the layers below 
20 fathoms. The large eye characteristic of the 
species would lend weight to such an assump- 
tion. Kishinouye’s discussion (1923: 444-445) 
of the habits of the species likewise indicates 
that it is not a surface fish but may approach 
nearer the surface at night. 
Parathunnus, as taken in the Hawaiian flag- 
line fishery, is a large tuna. Examples under 80 
pounds are rare; the average weight of the 
specimens landed during January to September, 
1948, for example, was 157.8 pounds. The 
maximum weight of this species landed in Ha- 
waiian waters approaches or exceeds 300 pounds. 
Kishinouye, however, gives a maximum weight 
of 86 kilograms (190 pounds) for Japanese 
examples, but as he states that a fish of this 
weight would be about 2 meters long, it is 
probable that his estimate of the maximum 
weight is in error. A 2 -meter long specimen 
with the body proportions of the Hawaiian 
fish would weigh about 334 pounds. 
The fact that Parathunnus is rarely taken by 
surface fishing methods has some interesting 
connotations. The present fishing grounds for 
the species in the Pacific are those grounds on 
which flagline fishing gear is employed. Al- 
though in Hawaii Parathunnus is from one-half 
to two-thirds as important as Neothunnus in 
the landings of large tunas, in Japanese land- 
ings (Shapiro, 1948, table 8) in some years, 
it was far more important than Neothunnus. 
Aside from the two specimens discussed by 
Godsil and Byers (1944: 105-119) and one 
