The SOUND Table in FishBase

Contrary to most people’s perception, the underwater environment is by no means a quiet haven.
In addition to the relatively high level of abiotic background noise (e.g. cavitation, wave
movements etc.) invertebrates, fish and marine mammals produce a variety of acoustic signals,
ranging widely in frequency and loudness (Schellart and Popper, 1992). Allusions to the sound
production of fish in the literature date as far back as early Greek and Roman times (Moulton,
1963; Fish and Mowbray, 1970) and, thanks to the technological advances in underwater
acoustics during the 1940’s, the investigation of fish sounds has become an active field of study
over the past decades (e.g. Moulton, 1963; Tavolga et al., 1981). Although the majority of known
fish species still remains to be tested, sonic activity could be demonstrated and recorded in more
than 80 families of fish during the 20th century (Fish, 1948; Fish and Mowbray, 1970). Judged by
the available evidence, it has been proposed that hundreds of species of marine and freshwater
fishes are probably capable of generating acoustic signals (Moulton, 1963). Active vocalisations
of most species are unique and distinctly species-specific (Fish and Mowbray, 1970) and have
indeed been used to distinguish closely related species in some cases (Crawford et al., 1997).

Many fish species have fairly advanced hearing abilities, mainly in the lower frequency ranges
(Hawkins, 1981). There is a moderate but significant concurrence of vocalization bandwidth and
hearing bandwidth for species for which data are available on both, hearing abilities and sound
production, indicating that the improvement of hearing and refinement of vocalizations may have
co-evolved in the context of intraspecific communication (Schellart and Popper, 1992). Recent
findings have shown, however, that the auditory sensitivity of some species may reach into
ultrasonic frequency ranges, thus exceeding the bandwidth of their own vocalisations by far (e.g.
Alosa sapidissima (Mann et al., 1997) or Gadus morhua (Astrup and Møhl, 1993)). It has been
proposed that the development of such extended auditory thresholds may represent an adaptive
response to the use of high frequency echolocating signals of piscivorous mammalian predators,
such as odontocetes (Mann et al., 1997; Astrup, 1999).

Humans across different cultures have exploited sonic abilities of fish for centuries, often to this
present day. Localizing fish by listening to species-specific acoustic signals is an art form that has
been employed in conjunction with fisheries (e.g. Sciaenid and carangid fisheries in Malaya
(Moulton, 1963)). Alternatively, man-made low frequency sounds or noises have been employed
both, as a means to either lure or scare fish into being caught in a variety of gear types ranging
from gill and seine nets to hook and line (Moulton, 1963; Brandt, 1984). In a non-fisheries
context, spawning sounds of some species, such as e.g. Cynoscion regalis and Sciaenops
ocellatus have also been used more recently to map of breeding distributions (Luczkovich et al.,
1998; Sprague et al., 1998).
Sources:

The main source for the information compiled in the SOUND tables is a report about the
comprehensive investigation of the acoustic behaviour of 220 species of North Western Atlantic
fishes published in 1970 by Fish and Mowbray (1970). The majority of fish sound files in this
table stem from the original recordings that accompanied this book.

Please note, that most of the sound samples available on the web page have been amplified and/or
filtered to reduce noise to improve the audio quality. Such modifications and the compressed
audio file format (.mp3) used to reduce file size will result in some alterations of the frequency
distribution and amplitude characteristics of original signal which are not necessarily detectable
by the human ear. To perform any quantitative acoustic analysis of the fish sounds it is therefore
strongly recommended to refer back to the original data.

To supplement the information compiled in the table, we have attempted to include cross-
references to other sound archives, such as the National Sound Archive of the British Library,
based on the information provided to us about their collection. In most cases, however, the actual
recordings are not directly accessible on-line and may be difficult to obtain.

The SOUND Production table

Fields:

Fish may generate acoustic signals either passively or actively. The former type of signals - also
referred to as ‘mechanical’ sounds (Fish, 1954) – are generated as by-products of foraging,
moving or other activities. In contrast, active – or ‘biological’ (Fish, 1954) - sound production
involves the use of organs, which, though initially developed to perform other functions (Fish,
1954), are also especially adapted to generate acoustic signals (Moulton, 1963). It is important to
recognize that both types of signals, active and passive, may possess a biological significance
(e.g. Spheroides maculatus may be attracted to a feeding site by the – passively produced -
chewing sounds of other members of its species (Moulton, 1963), while Holocentrus ascensionis
and other species are known to actively generate sounds during competitive feeding interactions
(Fish and Mowbray, 1970)).

The Sound production field contains two types of information: a.) if a fish has already been
tested for sound production and b.) if it has been found to – actively or passively – produce one or
several types of acoustic signals. If a species has been tested, detailed information about the
experimental set-up etc is provided in the SOUND Experiments table

To differentiate between different acoustic signals, early categorization efforts used descriptive
sound types to classify the recorded fish sounds. Unfortunately, however, although most people
agree that a ‘thump’, a ‘bump’ and a ‘knock’ are different sounds, what some may call a ‘bump’
may be considered to be a ‘thump’ by others. Despite attempts to standardize such descriptive
The SOUND Tables 2
sound type definitions (e.g. Fish and Mowbray, 1970), a large degree of subjectivity remains
associated with the qualitative classification approach. Consequently, analytical methods such as
oscillograms (two dimensional representation of a sound plotting the change in amplitude of a
waveform over time) and spectrograms (three dimensional representation of a sound plotting the
change in frequency and amplitude/loudness over time) are generally employed today to
quantitatively describe and compare acoustic signals. A real-time oscillogram of the acoustic
signal can be viewed when playing the sound file using the settings recommended on the web
page. In the future this will supplemented by a pre-generated spectrogram of the signal.

Despite the limited objectivity of qualitative descriptions, the sound types produced by a certain
species - if mentioned in the original reference – is also included in the record as it may allow a
coarse categorization and comparison of species based on their sounds.

The swim bladder is the most commonly used sound production organ in fish for active
vocalizations. This membranous air-filled sac fulfills several other functions, most importantly
regulating the animals’ buoyancy in the water, in addition to playing a role in the auditory system
of fishes (Tavolga et al., 1981; Schellart and Popper, 1992). In the context of sound production,
the swim bladder can either function as the actual sound generator itself or as an amplifier for
sounds generated by other body parts including e.g. the pectoral girdle, fin rays, various other
bones or the incisors (Schellart and Popper, 1992). Most commonly, however, sounds amplified
in such a way are generated by pharyngeal teeth (patches of mosaic-like denticles located far back
in the pharynx) which most fish species possess (Tavolga, 1964).

When sounds are generated ‘passively’, as a by-product of other activities, the involved body
parts, such as teeth (feeding sounds) or the entire body or the tail (swimming or hydrodynamic
sounds) could essentially be regarded as the ‘sound production organ’ (Fish, 1954).

The sonic mechanism is the actual underlying physical mechanism that generates the sound. Fish
actively produce sounds using a variety of different mechanism, one of the predominant ones
being the vibration of the air-filled swim bladder through sudden contraction of the skeletal
musculature, as associated with e.g. rapid acceleration, evasive movement or a startle reaction.
The resulting ‘escape sounds’ are a type of acoustic signals which the majority of fish with swim
bladders seem to be able to produce (Fish and Mowbray, 1970).

Alternatively, swim bladder vibrations may be initiated by means of specialized muscles

sometimes working in conjunction with modified bones. These muscles may be embedded
directly in the wall of the swim bladder (‘intrinsic musculature’ e.g. Triglidae) or may exist in the
form of ‘drumming’ muscles extrinsically associated with the swim bladder (e.g. Sciaenidae)
(Moulton, 1963; Fish and Mowbray, 1970). Furthermore, externally induced vibration of the
swim bladder in some species may be achieved by means of rhythmic beating of finrays against
the body surface (e.g. Balistes carolinensis) (Fish, 1954). The type of sound actually generated by
the vibrating swim bladder is mainly determined by the size and the shape of the swim bladder

The SOUND Tables 3

(Fish, 1954). Generally, however, these sounds are low-pitched, guttural and drum-like (i.e.
‘grunts’, ‘groans’, ‘thumps’, ‘knocks’, ‘clucks’, ‘booms’ or ‘barks’) with fundamental
frequencies ranging between 25 – 250 Hz and an upper frequency of 800 Hz (Fish and Mowbray,
1970; Tavolga, 1971) and are quite loud (Fish, 1954). In contrast, sounds associated with the
release of gas bubbles from the swim bladder, such as sometimes produced by Physostomi fish
species, which have maintained a pneumatic duct, are generally relatively faint (Fish, 1954).

Relatively loud, higher pitched ‘rasping’ or ‘scratching’ sounds are produced by the grinding of
incisors or – more commonly - the stridulation of pharyngeal teeth (Fish, 1954). In some species
(e.g. Pomadasyidae) these stridulatory sounds may be amplified considerably by a resonating
swim bladder located in close proximity to the pharyngeal teeth (Moulton, 1963; Fish and
Mowbray, 1970). Other types of sonic mechanisms include the stridulation of different body parts
against each other (e.g. some Syngnathidae) or the spasmodic contraction of heavy skeletal
muscles resulting in a low-pitched droning, like the hum of an electric motor (e.g. Cottidae) (Fish
and Mowbray, 1970).

In the case of passive or “mechanical” sound production, the sonic mechanism is generally the
friction between body parts during foraging activities (Moulton, 1963). Alternatively, the sudden
movement of a fish or a whole school in water may result in recordable compression waves with
frequencies up to 1 kHz (Fish and Mowbray, 1970; Schellart and Popper, 1992). One of the
loudest known fish sounds, produced by Pogonias cromis and frequently heard from boat decks
probably falls into the category of mechanical sounds as is believed to be produced by fish
beating their tails against the bottom of vessels to rid themselves of parasites ((Gunther, 1880) in
(Moulton, 1963)).

Active sound production in fish may vary considerably with time and space. For example many
fishes become most loquacious in the breeding season and at night time, following a seasonal
and/or diurnal cycle (Fish, 1954; Fish and Mowbray, 1970). Although the vocabulary of most fish
species is very limited (Fish and Mowbray, 1970) and the acoustical signals are mostly lacking in
fine graded complexity, it seems certain that sounds do serve a purpose in communicating gross
information about environmental and internal physiological states (Winn, 1964). This basic
information is most likely encoded in the repetition rate or duration of the signal or the intervals
between signals or its the harmonic frequency distribution (Winn, 1964)

The specific behavioural context in which a given sound was produced is documented in the
corresponding field in the table. As already indicated by the examples above, fish may actively
vocalise for a variety of different purposes. The behavioural contexts during which sounds are
produced may be roughly divided into three major categories: i. startle and escape reactions (e.g.
Arius felis or Bagre marinus (Tavolga, 1960; Fish and Mowbray, 1970)) ii. agonistic inter- or
intraspecific interactions, such as competitive feeding (e.g. Haemulun melanurum (Fish and
Mowbray, 1970) or Hippocampus erectus (Colson et al., 1998)) or territorial defense (e.g.
Pomacentrus partitus (Myrberg, 1997) Gaidropsarus mediterraneus (Almada et al., 1996)
The SOUND Tables 4
Opsanus tau (Gray and Winn, 1961) and iii. sounds produced in a reproductive context, such as
basic species identification and discrimination (e.g. some closely related Momyridae species
(Crawford et al., 1997)), courtship (e.g. Hippocampus erectus (Fish, 1954; Fish and Mowbray,
1970)) or spawning (e.g. many Sciaenidae (Fish, 1954) or more specifically Hypoplectrus
unicolor or Scarus iserti (Lobel, 1992)). In addition, there is some evidence that acoustic signals
may also be produced for non-communicative purposes, such as the orientation in a new
environment, i.e. a simple form of echo-location (Vincent, 1963). A classic example of this are
the clicks produced by Hippocampus erectus specimen when introduced into a new environment
(Fish, 1954; Fish and Mowbray, 1970). Spontaneous sound production, i.e. without an apparent
external stimulus, is fairly rare in fish, the so-called ‘boatwhistle’ of Opsanus tau being one of the
few exceptions (Winn, 1964).

Recording sounds in the field is quite difficult, partly because of the background noise level in
these circumstances, but mainly due to the problems of matching the recorded signals to specific
individuals, if a number of animals and/or species are present. Therefore the sonic abilities of a
certain species have often been tested under controlled experimental conditions or while a
specimen is caught in fishing gear rather than in situ. Furthermore, since spontaneous
vocalizations in these circumstances are rare, sound production was sometimes initiated by using
artificial stimuli such as manual or electrical stimulation (Fish and Mowbray, 1970). Although
these experiments may not establish the natural occurrence of sound for a specific species, they
may appraise the sonic ability of a species. Moreover, experiments have shown such artificially
stimulated sounds to be very similar to naturally occurring alarm calls (Fish and Mowbray, 1970).
The use of artificial stimuli was nonetheless documented in the behavioural context field.

If provided in the original source, information about the circumstances under which the recording
was made is given in the remarks field. This includes details about the specific external
experimental set-up, such as the recording environment (i.e. field pen or enclosure, wooden tank,
aquarium etc.) as these have an impact on the quality of the recording (e.g. acoustic recordings
made in closed space will include reverberations of the walls, which may interfere with the
targeted signal etc.). Furthermore, the number of specimen vocalising and/or present at the time
of recording is given if such information was available. Sometimes it is specifically noted in the
source, whether or not sounds were ever produced spontaneously or if sound production
associated exclusively with artificial stimulation. If mentioned this information was included in
the remarks field as it may provide some indication of the likelihood of encountering the recorded
sounds in the wild.

All acoustic recordings are affected to some extent by distortion introduced through the particular
sensitivities of all parts of the recording system, which need to be taken into account when trying
to analyse or interpret a recorded sound. For the technical specifications of hydrophones and
preamplifiers please refer back to the original sources, as these could not be included in
appropriate detail in this table.

The SOUND Tables 5

In many cases sounds are recorded using filters, often employed to reduce the noise level in the
recordings. Depending on the specific attributes of a filter, certain frequency ranges in the
recording will be blocked out completely or at least dampened considerably. Filters can either cut
out all frequencies above (“lowpass”) or below (“highpass” filters) a certain specified threshold.
Alternatively, so-called “bandpass” filters allow the selective recording of pre-defined frequency
ranges with an upper and lower set boundary. As all of these filters will obviously have a
significant impact on the resulting frequency distribution of the recorded signal, the use of filters
is reported in the remarks field, if mentioned in the source.

During the digitization and extraction process necessary to make sounds available on FishBase,
the recordings were sometimes amplified and/or an additional noise reduction filter was
employed, to improve the overall audio quality. Again, this will have had an effect on the
acoustic properties of the final signal and is therefore documented in the table.

Acknowledgements:

We are very grateful to Kenneth Hinga, University of Rhode Island for granting us the permission
under his copyright to use the Fish and Mowbray (1970) fish sound recordings. We are also
indebted to Arthur N. Popper, Director of the Neuroscience and Cognitive Science Program,
Department of Biology, University of Maryland, USA for generously allowing us produce a
digital copy of his copy of original Fish and Mowbray (1970) tapes.

References
Almada, V. C., Amorim, M. C. P., Pereira, E., Almada, F., Matos, R., and Godinho, R., 1996. Agonistic behaviour and
sound production in Gaidropsarus mediterraneus (Gadidae). Journal of Fish Biology, 49(2):363-366.
Astrup, J., 1999. Ultrasound detection in fish - a parallel to the sonar-mediated detection of bats by ultrasound-sensitive
insects? Comparative Biochemistry and Physiology, Part A 124:19-27.
Astrup, J., and Møhl, B., 1993. Detection of intense ultrasound by the cod Gadus morhua. Journal of Experimental
Biology, 182:71-80.
Brandt, A. v., 1984. Fish catching methods of the world. Farnham : Fishing News, 418 pp.
Colson, D. J., Patek, S. N., Brainerd, E. L., and Lewis, S. M., 1998. Sound production during feeding in Hippocampus
seahorses (Syngnathidae). pp. 221-229. Environmental biology of fishes, The Hague, 51.
Crawford, J. D., Cook, A. P., and Heberlein, A. S., 1997. Bioacoustic behavior of African fishes (Mormyridae):
Potential cues for species and individual recognition in Pollimyrus. Journal of the Acoustical Society of America,
102(2):1200-1212.
Fish, M. P., 1948. Sonic fishes of the Pacific. Proj. NR 083-003, Contr N6 ori-195, t.o.i. between ONR and Woods
Hole Oceanographic Institution, Technical Report, 2.
Fish, M. P., 1954. The character and significance of sound production among fishes of the Western North Atlantic.
Bulletin of the Bingham Oceanographic Collection, 14(3):1-109.
Fish, M. P., and Mowbray, W. H., 1970. Sounds of Western North Atlantic Fishes -
A Reference File of Biological Underwater Sounds. The John Hopkins Press, Baltimor & London, 204 pp.
Gray, G.-A., and Winn, H. E., 1961. Reproductive ecology and sound production of the toadfish, Opsanus tau.
Ecology, 42(2):274-282.
Gunther, A. C. L. G., 1880. An Introduction to the Study of Fishes. Adam and Charles Black, Edinburgh.

The SOUND Tables 6

Hawkins, A. D., 1981. The hearing abilities of fish. pp. 109-138. In: W. N. Tavolga, A. N. Popper, and R. R. Fay, eds.,
Hearing and sound communication in fishes, Springer Verlag, New York Heidelberg Berlin.
Lobel, P. S., 1992. Sounds produced by spawning fishes. pp. 351-358. Environmental biology of fishes, The Hague, 33.
Luczkovich, J. J., Johnson, S. E., and Sprague, M. W., 1998. Using sound to map fish spawning: Determining the
seasonality and location of spawning for weakfish and red drum (Family Sciaenidae) within Pamlico Sound, NC.
Journal of the Acoustical Society of America, 103(5):3000.
Mann, D. A., Lu, Z., and Popper, A. N., 1997. A clupeid can detect ultrasound. Nature, 389(6649):341.
Moulton, J. M., 1963. Acoustic behaviour in fishes. pp. 655-687. In: R.-G. Busnel, ed. Acoustic behaviour of animals,
Elsevier Publishing Company, Amsterdam - London - New York.
Myrberg, A. A., Jr., 1997. Sound production by a coral reef fish (Pomacentrus partitus): Evidence for a vocal,
territorial "keep-out" signal. Bulletin of Marine Science, 60(3):1017-1025.
Schellart, N. A. M., and Popper, A. N., 1992. Functional aspects of the evolution of the auditory system of
actinopterygian fish. pp. 295 - 323. In: D. B. Webster, R. R. Fay, and A. N. Popper, eds., The evolutionary biology
of hearing, Springer-Verlag, New York, Berlin, Heidelberg.
Sprague, M. W., Luczkovich, J. J., and Johnson, S. E., 1998. Using fish sounds to identify spawning activity of
weakfish (Cynoscion regalis) and red drum (Sciaenops ocellatus) in nature. Journal of the Acoustical Society of
America, 103(5):3001.
Tavolga, W. N., 1960. Sound production and underwater communication in fishes. pp. 93-136. In: W. E. Lanyon and
W. N. Tavolga, eds., Animal sounds and communcation, American Institute of Biological Sciences, Washington,
D.C., Publication No. 7.
Tavolga, W. N., 1964. Sonic characteristics and mechanisms in marine fishes. pp. 195-212. In: W. N. Tavolga, ed.
Marine Bio-Acoustics, Pergamon Press, Oxford, London, New York, Paris.
Tavolga, W. N., 1971. Sound production and detection. pp. 135-205. In: W. S. Hoar and D. J. Randall, eds., Fish
physiology, Academic press, New York, 5.
Tavolga, W. N., Popper, A. N., and Fay, R. R., 1981. Hearing and sound communication in fishes. Proceedings in Life
Sciences. Springer-Verlag, New York, 608 pp.
Vincent, F., 1963. Acoustic signals for auto-information or echolocation. pp. 221-227. In: R.-G. Busnel, ed. Acoustic
behaviour of animals, Elsevier Publishing Company, Amsterdam - London - New York.
Winn, H. E., 1964. The biological significance of fish sounds. pp. 213-231. In: W. N. Tavolga, ed. Marine Bio-
Acoustics, Pergamon Press, Oxford, London, New York, Paris.

Kristin Kaschner.

Cite as:
Kaschner, K. 2012. The SOUNDS table in FishBase. In: Froese, R. and D. Pauly. Editors.
2012. World Wide Web electronic publication, www.fishbase.org, version (08/2012).

The SOUND Tables 7