6. SOUND PRODUCTION AND DETECTION
179
frequency range of most fish is significantly narrower. Although in some
species thresholds can be obtained at frequencies up to 3 or even 5 kHz,
the intensities of the test sounds are in the 20-30 dB pb range, and such
acoustic intensities are probably much higher than sounds normally encountered in the fish‘s environment. The usable frequency range in most
fish, therefore, does not extend beyond about 2 kHz (higher in ostariophysines), and the dynamic range covers a span of about 60 dB. By contrast, human hearing has a usable range of at least 10 kHz and over
100 dB.
Within the narrow frequency and dynamic range of fish, the ability
for discrimination has been found to be remarkably good. The first reports on intensity discrimination were those of Wohlfahrt (1939), who
concluded that this ability was completely lacking in fish, and Dijkgraaf
and Verheijen (1950), who were able to show that the minnow, Phoxinus
laevis, was capable of distinguishing between tones of equal frequency
but different intensity. In another ostariophysine, the goldfish, Curussius
auratus, Jacobs and Tavolga ( 1967) demonstrated that the “just-noticeable-difference” ( i n d ) was from 3 to 6 dB. This compares to a ind of
about 0.5 dB for human hearing ( Licklider, 1951 ) .
The ability of fish to discriminate between different frequencies was
first studied by Stetter (1929), who found that Phoxinus was capable
of discriminating between two tones that were about a minor third apart,
i.e., about a 20% difference in frequency. For the same species, Wohlfahrt
( 1939) obtained a ind of about 6% ( a half tone). This discriminatory
capacity was found to be somewhat temperature dependent (Dudok van
Heel, 1956).
Much of the above work was done with poorly controlled intensities
and crude sound-producing devices (flutes, tuning forks, etc. ) . Using more
accurate sound generators, Dijkgraaf and Verheijen ( 1950) reported that
Phoxinus was capable of discriminating tones with a 3% (one-quarter
tone) difference in frequency. Using more accurately calibrated equipment, standard psychophysical techniques, more animals, and a different conditioning procedure, Jacobs and Tavolga (1968) were able to confirm these data for the goldfish. The mean ind at 200 Hz was 9.4
Hz (4.7%), at 500 Hz the ind was 17.4 Hz (3.5%), and at 1000 Hz the
ind was 50.1 Hz (5.0%). For comparison human subjects possess a ind of
0.448% at 100 Hz and 0.243% at 1000 Hz.
It is significant that species of fish other than Ostariophyi have a much
larger ind for frequency. In two species of mormyrids, Stipetih (1939)
reported a ind of 12-25%, and Dijkgraaf (1952b) stated, that Gobius,
Corvina, and Sargus were capable of discriminating a 9-15% difference in
frequency. Even larger values of 50-100% were given for other nonostario-
179
frequency range of most fish is significantly narrower. Although in some
species thresholds can be obtained at frequencies up to 3 or even 5 kHz,
the intensities of the test sounds are in the 20-30 dB pb range, and such
acoustic intensities are probably much higher than sounds normally encountered in the fish‘s environment. The usable frequency range in most
fish, therefore, does not extend beyond about 2 kHz (higher in ostariophysines), and the dynamic range covers a span of about 60 dB. By contrast, human hearing has a usable range of at least 10 kHz and over
100 dB.
Within the narrow frequency and dynamic range of fish, the ability
for discrimination has been found to be remarkably good. The first reports on intensity discrimination were those of Wohlfahrt (1939), who
concluded that this ability was completely lacking in fish, and Dijkgraaf
and Verheijen (1950), who were able to show that the minnow, Phoxinus
laevis, was capable of distinguishing between tones of equal frequency
but different intensity. In another ostariophysine, the goldfish, Curussius
auratus, Jacobs and Tavolga ( 1967) demonstrated that the “just-noticeable-difference” ( i n d ) was from 3 to 6 dB. This compares to a ind of
about 0.5 dB for human hearing ( Licklider, 1951 ) .
The ability of fish to discriminate between different frequencies was
first studied by Stetter (1929), who found that Phoxinus was capable
of discriminating between two tones that were about a minor third apart,
i.e., about a 20% difference in frequency. For the same species, Wohlfahrt
( 1939) obtained a ind of about 6% ( a half tone). This discriminatory
capacity was found to be somewhat temperature dependent (Dudok van
Heel, 1956).
Much of the above work was done with poorly controlled intensities
and crude sound-producing devices (flutes, tuning forks, etc. ) . Using more
accurate sound generators, Dijkgraaf and Verheijen ( 1950) reported that
Phoxinus was capable of discriminating tones with a 3% (one-quarter
tone) difference in frequency. Using more accurately calibrated equipment, standard psychophysical techniques, more animals, and a different conditioning procedure, Jacobs and Tavolga (1968) were able to confirm these data for the goldfish. The mean ind at 200 Hz was 9.4
Hz (4.7%), at 500 Hz the ind was 17.4 Hz (3.5%), and at 1000 Hz the
ind was 50.1 Hz (5.0%). For comparison human subjects possess a ind of
0.448% at 100 Hz and 0.243% at 1000 Hz.
It is significant that species of fish other than Ostariophyi have a much
larger ind for frequency. In two species of mormyrids, Stipetih (1939)
reported a ind of 12-25%, and Dijkgraaf (1952b) stated, that Gobius,
Corvina, and Sargus were capable of discriminating a 9-15% difference in
frequency. Even larger values of 50-100% were given for other nonostario-