354 
PACIFIC SCIENCE, Vol. V, October, 1951 
TABLE 8 
Food of Nehu {Stolephorm purpureus) in Honolulu Harbor, Oahu, Based on an Analysis of 21 Stomachs 
ORGANISM 
PERCENTAGE 
FREQUENCY OF 
OCCURRENCE 
AMONG FISH 
EXAMINED 
AVERAGE 
NUMBER TAKEN 
PER FISH 
CONTAINING 
THE ITEM 
PERCENTAGE COMPOSITION 
OF FOOD ITEMS BASED 
ON TOTAL NUMBER OF 
ORGANISMS FOUND 
IN THE STOMACHS 
Crab larvae 
megalopa 
100 
20 
42 
zoeae . 
25 
10 
5 
Ghost shrimps {Leucifer faxonii) 
70 
18 
27 
Shrimps 
adults 
40 
21 
18 
mysis 
15 
10 
3 
Copepods 
40 
6 
5 
Barnacle larvae (cypris) 
10 
2 
TRACE 
crustacean components of the zooplankton, 
and that selection of crustacean types is ap- 
parent in some areas and lacking in others. 
From the standpoint of selection of food 
items, nehu resemble herring more nearly 
than they do sardines. While opinion is di- 
vided on the question of selection of plank- 
ters by sardines, most investigators agree that 
there is little or no selection exercised 
(Kishinouye, 1907; Lewis, 1929; Suyehiro, 
1942), but Parr (1930) believes that phyto- 
plankton is ingested only incidentally while 
sardines pursue zooplankton. Overwhelming 
evidence for selection in the feeding of her- 
ring has been advanced by Moore (1898), 
Hardy (1924), Bigelow and Welsh (1925), 
Bigelow (1926), Jespersen (1928), Savage 
(1931, 1937), Lucas (1936), Wailes (1935), 
and Johnson (1940). It is generally agreed 
that herring feed by active pursuit on sight 
and that adult fish ignore the smallest forms 
of copepods even though they may be the 
most abundant in the plankton. The plankters 
selected showed nice agreement between the 
occurrence in the herring food and in plankton 
samples. These findings are comparable to 
those discovered for nehu. Lebour (1920: 
262) aptly sums up this subject for herring 
and other small plankton feeding fishes, ex- 
cept possibly the sardine, by stating: ". . . 
usually each species of fish selects its own 
favourite food, to which it keeps, indis- 
criminate feeding seldom or never taking 
place. ...” 
Ingestion: To learn how nehu ingest their i 
food, a small school was confined in a display 
tank into which living ghost shrimps were 
then placed. The fish swam quickly to the^ 
shrimps and ate them without slackening 
their swimming speed. When a nehu sighted | 
a ghost shrimp to one side it would make a | 
quick sideways movement in turning to reach |: 
the shrimp. In doing this the whole body j 
partly turns over producing a flash from the 
silvery side when viewed from above. This 
sideways movement and silvery flash are one 
of the characteristic features of a school of 
nehu, and serve to distinguish nehu from 1 
another important baitfish, the iao (Pranesus 
insularum)^ during both day and night baiting 
operations. Captive herring have been ob- 
served to behave similarly (Johnson, 1940: 
392). 
Relation of size offish to size of food: Length- 
frequency plots of nehu in the commercial 
catch for the various baiting areas clearly show 
characteristic differences in the average size 
of fish (unpublished data). Certain of the 
areas involved in this study (Ala Wai Canal 
and Honolulu Harbor) can always be depend- 
ed upon to provide larger fish than the other j 
areas mentioned. The nehu in Kaneohe Bay 
and in Pearl Harbor are characteristically 
small in size, with those in Hilo Bay still 
smaller. Obviously, factors other than food 
supply may operate to regulate the average 
size composition of a fish population, such as 
(1) differences in fishing intensity from one 
