Some Aspects of the Biology of the 
Aku, Katsuwonus pelamis, in the Hawaiian Islands 
Vernon E. Brocki 
The aku, Katsuwonus pelamis (Linne), is the 
most important species of fish taken com- 
mercially in Hawaii. It supports the only Ha- 
waiian fishery which supplies a product, 
canned tuna, for export as well as for local 
consumption. It is, on the one hand, a species 
that affords the greatest promise for expan- 
sion of the Hawaiian fishing industry and, on 
the other, one of which little or nothing of 
the life history or biology is known. 
Hence, the scientific staff of the Division 
of Fish and Game of the Territory of Hawaii, 
following an initial study of the length fre- 
quencies of Hawaiian aku by Bonham (1946), 
has been engaged in several projects con- 
cerned with the biology of this fish. Included 
among these projects are further studies of 
length frequencies, studies of the span and 
character of the spawning season, of the na- 
ture of schooling, and of the conduct of the 
fishery with a view toward establishing a basis 
for obtaining a fairly accurate measure of 
catch per unit of effort. The present paper is 
concerned with all these projects except the 
last. 
Beginning in the late summer of 1946, aku 
obtained from the commercial landings were 
sexed and the length measured. The technique 
of sexing will be described later; the measure- 
ment of length obtained was from the tip of 
the snout to the end of the mid-caudal rays. 
This measurement was taken by means of 
calipers similar to those described by Marr and 
Schaefer (1949). In part, measurements made 
1 Division of Fish and Game, Board of Agriculture 
and Forestry, Honolulu, Hawaii. Manuscript received 
November 7, 1952. 
during 1950 and later were taken by marking 
the length on a celluloid strip by punching 
a hole in it. The lengths obtained are, for the 
same fish, very slightly less than the lengths 
obtained by the use of calipers. During the 
winter off-season period, both the number 
of fish measured in a sample and the number 
of samples obtained were affected by the avail- 
ability of fish at the cannery; however, the 
frequency of sampling was chiefly affected. 
The number of fish entering into each of the 
percentage frequency curves plotted in Figure 
1 is shown in Table 1. As individual samples 
usually represented some 100 fish, and some- 
times much less, it is obvious that most of 
the percentage frequency curves of Figure 1 
are composed of a number of samples com- 
bined together. 
In the beginning of the program, estimates 
of the degree of maturity of the fish sampled 
were made by noting the appearance of the 
gonads; however, these estimates did not 
provide a measure of sufficient accuracy to 
show adequately the spawning season of this 
species. For this reason, beginning with the 
summer of 1948, ovaries were removed from 
20 of the fish in each sample and preserved 
in formalin, and subsequently each ovary was 
drained and weighed to the nearest tenth of 
a gram. From each, 50 ova chosen at random 
were measured to the nearest one hundredth 
of a millimeter by means of an ocular micro- 
meter in a compound microscope. 
Beginning with the summer of 1949, a large 
part of the samples taken for length measure- 
ments were taken aboard the aku fishing 
vessels while landing the catch. These sam- 
94 
