372 
PACIFIC SCIENCE, VoL XXI, July 1967 
TABLE 3 
Peak Pressure Change and Range of Pressures in Octave Bands of Fish Sounds Before (Normal) 
and After (Operated) Removal of the Muscles and Bones Over the Sound-Producing Muscle 
PEAK PRESSURE 
CHANGE IN 
RANGE OF 
OCTAVE BAND SOUND PRESSURES IN DB 
CONDITION OPERATED FISH 
75-150 
150-300 
300-600 
600-1200 
1200-2400 
2400-4800 
Normal 
+2 db 
16-18 
3-7 
5-6 
1-4 
16-19 
31-34 
Operated 
21-24 
8-9 
4-6 
6 
19-20 
33-36 
Normal 
-l db 
23-26 
11 
4-5 
6-7 
19 
32-33 
Operated 
29-30 
14-15 
6-7 
10 
21-22 
36-41 
Normal 
0 db 
23-36 
8-12 
5-8 
5-8 
17-20 
33-35 
Operated 
25-27 
11-12 
6-7 
5-6 
17-19 
30-34 
Normal 
-1 db 
25-27 
13-14 
7-8 
5 
19 
21-22 
Operated 
27-28 
15-16 
8-11 
6-7 
15-17 
32-33 
Normal 
0 db 
29-31 
14 
6-7 
5 
16 
31-32 
Operated 
28-30 
14 
7-9 
4-8 
16-19 
30-32 
these habitats 
the fish were 
was always observed, 
ever seen swimming 
None of 
in open 
though on 
midnight. 
some nights 
fishing continued until 
water during 
the day. 
Associated 
with the 
Tape re 
wordings in 
the field were carried 
schools of menpachi were groups 
of other 
out in four different areas of Pokai Bay, three 
squirrelfish (H. ensifer, H. xantherythrus, and 
H. spinifer — usually a single specimen), 
aweoweo ( Priacanthus cruentatus and P. 
meeki), moray eels ( Gymnothorax sp.) , car- 
dinal fish ( Apogon sp.) and pipefish ( Syngna - 
thus sp.). Various other diurnal species of reef 
fishes were observed to enter and leave caves 
and ledges. The frequent visits of large schools 
of goatfish ( Parupeneus sp.) did not result in 
production of staccato and grunt sounds when 
the entrances coincided with diel recordings. 
It was not possible to carry out detailed ob- 
servations on the behavior of menpachi in these 
areas, even with scuba gear. A diver’s pres- 
ence resulted in retreat by the fish into darker 
and less accessible areas, accompanied by the 
production of many staccato and grunt sounds. 
Only a few fish briefly investigated the diver 
within the first minute or two after fie ap- 
peared. When the caves or ledges were too 
shallow for backward retreat, the school scat- 
tered to either side or rushed quickly back and 
forth within the confines of the area. 
Many menpachi (both M. berndti and M. 
argyromus ) were caught with hook and line. 
The bait was kept off the bottom, just outside 
the ledge or cave opening. No fish were ever 
caught or took bait during the day. All 57 fish 
caught by fishing during the study period were 
captured between 1930 and 2030 hours, al- 
for a 24-hour and one for a 9-hour period. 
The results are shown in Table 4. Four types 
of sounds were recorded: (1) staccatos, (2) 
grunts, (3) a series of knocking sounds vari- 
able both in intervals between consecutive 
knocks and in number of knocks in a series, 
and (4) growls, consisting of a rapid series of 
sounds lasting from 1 to 4 seconds. Oscillo- 
graphs of these types of sounds recorded from 
laboratory populations are shown in Figure 4. 
Only a few of these kinds of sounds were pro- 
duced after sunset and before sunrise. Knocks 
were the most frequently recorded of all sounds, 
with no obvious peaks in rate of production 
after an initial increase following dawn. In 
one 24-hour recording (April 28-29), there 
was a peak in staccato and grunt sounds at dusk. 
General Behavior of Laboratory Populations 
Laboratory populations confined their day- 
time movements to slow swimming inside the 
cave, with occasional chasing of one fish by 
another. Individual fish occasionally swam out- 
side of the cave for a few seconds. When 
lights were turned off at night, the movements 
of fish could still be detected in the available 
ambient light. Within 5 minutes, the fish were 
swimming rapidly around the tank above the 
cave. Several populations all produced sounds 
in the laboratory at night when recordings were 
