Morphometry of Tuna-—SCHAEFER 
119 
The slopes of these regressions do not differ 
significantly from those of the pooled data, the 
statistics for which may be found in Table 2. 
The levels of the lines however, that is, the 
values of x for a given value of y, are less in 
the case of Mr. Marr’s measurements and 
greater in the case of the author’s measure¬ 
ments than would ordinarily be expected from 
random sampling from a population having the 
values estimated from the pooled data. The 
probabilities, in each case, lie between 0.02 and 
0.01. There seems, therefore, to be a real dif¬ 
ference between the two sets of measurements. 
Since the measurements by the two persons were 
made on different groups of fish, there is a pos¬ 
sibility that this difference represents an actual 
difference between the two groups of tunas. 
However, no difference between the two is 
shown by the data on other dimensions; in par¬ 
ticular no difference between the distances of 
snout to insertion of first dorsal, snout to inser¬ 
tion of second dorsal or body depth, one of 
which should reflect any actual difference in 
the distance between the insertions of the first 
and second dorsal fins. Therefore, it seems most 
likely that the difference represents a differ¬ 
ence in the measuring by the two observers, al¬ 
though it is not apparent just how this arose. 
LENGTH-WEIGHT RELATIONSHIP 
In addition to the data recorded in Table 1, 
the lengths and weights were determined for a 
number of specimens over the same size range, 
a total of 93 in all. The regression of logarithm 
of weight on logarithm of length is described 
in Table 2. This relationship corresponds to the 
equation 
W=2.74X10- 8 L 2 - 940 .(3) 
where L is the total length in millimeters and 
W is the weight in kilos. 
TAXONOMIC NOTES 
The yellowfin tuna off Costa Rica appear to 
be assignable to the species Neothunnus wid¬ 
er opterus (Temminck and Schlegel). The speci¬ 
mens examined by me agree closely enough with 
the descriptions of that species in Kishinouye’s 
monograph (1923) and in Godsil and Byers’ 
paper (1944). The latter concluded that all the 
yellowfin tunas of the Pacific examined by them, 
including specimens from Costa Rica, were 
members of this single species. Nichols and 
La Monte’s data (1941) on length of second 
dorsal and anal fins for N. macropterus, which 
they consider to be a synonym of N. albacora, 
given for various fish lengths, fall within the 
expected limits of variation of the values esti¬ 
mated for those same lengths from the regres¬ 
sions in our Table 2. The only exception to 
this is their Portuguese specimen 5 feet long, 
which had dorsal and anal lobes contained 
"2.6 to 2.8 times in the length.” From our data, 
it is estimated that at this length only about 1 
per cent of specimens would have second dor¬ 
sals contained less than 3.9 times in the length, 
or anals contained less than 3.2 times in the 
length. 
Frade (1929, 1931) has found that there are 
rather distinct anatomical differences between 
the Portuguese yellowfin, N. albacora, and the 
Pacific yellowfin, N. macropterus. He has found 
that the air bladder of N. albacora has a large 
dorsal diverticulum, which is not present on the 
air bladder of N. macropterus according to 
Kishinouye’s description, which is confirmed by 
the study of Godsil and Byers (1944). Kishi¬ 
nouye’s Figure Q (page 373) also shows the 
cutaneous artery of N. macropterus arising at 
the level of the 9th vertebra, whereas that of the 
Portuguese yellowfin was found by Frade to 
arise at the level of the 8th. This difference is 
not confirmed by Godsil and Byers, however, 
who also found the cutaneous artery arising at 
the level of the 8th vertebra in their specimens 
of N. macropterus. 
Decision as to whether or not the variety of 
yellowfin tuna from the Hawaiian Islands, 
having very long anal and second dorsal fins at 
larger sizes, is distinct from the more common 
variety, and, if so, whether the difference is spe- 
