Acoustical behavior of clownfishes (Amphiprion spp.) Akira TAKEMURA. The
acoustical behavior of seven species of clownfishes (Pomacentridae) bred in the
...
NAOSITE: Nagasaki University's Academic Output SITE
Title
Studies on the Underwater Sound―VIII Acoustical behavior of clownfishes (Amphiprion spp.)
Author(s)
竹村, 暘
Citation
長崎大学水産学部研究報告, v.54, pp.21-27; 1983
Issue Date
1983-02
URL
http://hdl.handle.net/10069/30420
Right This document is downloaded at: 2017-10-06T21:51:10Z
http://naosite.lb.nagasaki-u.ac.jp
Bull. Fac. Fish. Nagasaki
Univ., No. 54, 21-27
Studies Acoustical
(1983)
21
on the Underwater
behavior
Sound—VIll
of clownfishes
Akira
(Amphiprion
TAKEMURA
The acoustical behavior of seven species of clownfishes (Pomacentridae) Hubbs Sea World Research Institute was observed. Each specimen emitted of a pulse or a series of pulse sounds.
spp.)
bred in the tanks at some sounds composed
These sounds were divided into three categories by their
frequency components and, among of these species, some differences were found out. Each sound was relatively vigorous.
The fish was utilizing these sounds for protecting their territory on their
life, especially by some activities related with spawning or threat.
The mutualism centridae)
between the clownfish (Poma-
species among them, namely; Amphiprion clarkii,
and sea anemone is already well known.
A. polymnus, A. biaculeatus, A. ocellaris, A. frena-
A pair of adult clownfish forms a territory around
tus, A. melanopus and A. sandaracinos
the sea anemone and their movements
Observations on acoustical
are limited
in this small area. The adhesive eggs of the fish are
were carried out.
laid on the surface of a rock near the sea anemone. Usually, a pair of fish spawn and take care of eggs
are raised.
behavior of these fishes
These specimens had been reared in glass tanks of two sizes, 45 × 60 × 40 (depth) cm and 100 × 300 ×
in their territory every month, however, the spawn-
100 (depth) cm. The sand is placed 15 cm thick on
ing took place more frequently in the laboratory
the
as
bottom
of small
tanks
for filtration.
A
every two weeks. Female vigorously cleans up the
hydrophone was suspended at the middle depth in
spawning ground just before and after spawning.
the center of the tank. A pair of clownfish was kept
However
in each tank.
after
spawning,
females
of the
most
species of this genus hardly take care of eggs. On
Water and air supply were stopped
during the recording to minimize the noise.
the contrary, males eargerly take care of eggs and
It was very difficult to record the correct fre-
do not go away from the eggs through the incuba-
quency or duration of calls of fishes, because the
tion period.
tanks were small and no measure was adopted to
During a series of those behavior, times emit the sound.
Verwey
they some-
(1930), Schneider
lessen the reflection of sound.
(1964) and Allen (1975) described already in their reports about the calls of some species of the fishes.
their sound and behavior.
Moreover, Verwey and Schneider suggested some
Underwater
assumptions on the sound production
Currently
a total
spp. are known
from
of
World
Hubbs
Sea
and Methods
of 27 species the world. Research
between
Sounds
mechanism. The clownfish
Materials
However, it seems
that it is able to assume the relationship
of Amphiprion
were
kept
times
emitted
were
kept
In the aquarium
emitting
Institute,
fishes.
seven
alone
did not emit any sound in the tank,
some
sounds
in a tank.
increased
however, when
The
in proportion
plural
frequency
when they they
some-
specimens of sound
to the number
of
22
A. Takemura: Acoustical behavior of clownfishes
All of these sounds were constituted by pulse 4一
sound and had something in common with the main
ψ■
■膨確
σ壷●
駐
ヒ
Accordingly, the pattern of sound was quite simple.
2.0
t.o
3.0
Tkne in sec.
The sound with a long duration such as O.056
−O.065sec was emitted rarely under the special
蓬 ξ。
撃
ッ の の
seldom exceptions of being emitted in a series.
N蚕蕊診§ρ¢
over, most of the sounds were emitted singly with
3 2 i
frequency component of less than 1 kHz. More−
Fig. 1. Underwater sound of A. clarkii
behavior, however, the duration of pulse sound was usually O.02−O.03 sec.
Moreover, those pulse sound emitted by this
The components of frequency of sound were
species was rarely emitted before special move−
very various and all of sounds were divided into
ments as mentioned later and was usually emitted
following three categories;
during or after the violent and special movement.
1. Sound containing the frequency component
Explanations of pulse sound of each species are
from very low frequency such as D. C. to several
given below:
kilohertz, sometimes more than 8 kHz.
1..4〃ψ吻η●onぬ痴ゴ
2. Frequency components limited between 200 −300 Hz and a few kilohertz.
3. Sound composed of the frequency component
Almost all sounds of this species were contained
in the second category. Especially the frequency component were concentrated into 1−2 kHz (Fig.
of more than 1 kHz.
1). The variety of sound of this species was very
Moreover, each category was divided into 5−9
poor, however, the sound was emitted most fre−
subcategories by fundamental frequency, harmonic
quently among the seven species described before.
and so on. H owever, the frequency of emitted
The both sounds of single and a series of pulse were
sound was quite low under the normal behavior and
emitted, and, in generally, single pulse sound was
most of all the frequency components were limited
higher than a series of sound by about 300 Hz.
in low frequency less than 2 kHz and main fre−
Those sounds were heard from the fish not only
quency was about 500 Hz. The clownfish emitted
when they were eating the food either suspending in
the sound composed of single pulse or of a series of
water or floating at the surface of water but also
pulses. The latter sound was emitted by plural
when they were bathing in the sea anemone.
specimens that were staying close to each other and
In addition to this, the single pulse sound with
the sound of series of pulse was not heard when
the frequency component of more than 2 kHz was
each specimen swam or stayed separately.
also recorded.
The clownfishes are taxonomically classified
2. A. Polpmnus
into two groups by the number of stripes of fish’s
This species is large−sized one among the clown−
body and the third category of sound were heard
fish. The sound emitting of this species is very rare
from the group of fish with more than two stripes
and the sound of the second category was not heard
on their body and were not heard from the group of
(Fig. 2). Those rare sounds were cl assified into
fish with less than one stripe. Moreover, the high
three pattem Namely;
frequency and weak pulse sound such as popping
1. Pulse sound with the frequency component of
were heard from every tank, however, it seems that
2.5−4kHz and, sometimes, composed of 4−5
those sounds were caused by nonbiological factor.
pulses.
Namely, the tank was divided into two parts (a part with fish and the other part with no fish) by
styropone plate and same sound was heard from each part of tank.
2. Single pulse and frequency component of D. C. 一4 kHz.
3. Single pulse and frequency component of D. C.
一more than 8 kHz.
23
Bull. Fac. Fish. Nagasaki Univ., No. 54 (1983)
All of these sounds had a strong component at 轟{
鑑嘉監罫ぴ
female, however, this sound was not special sound
障葦‘二り璽皆鍵
sound from their behavior, was emitted from
1
..暇..... 曹 ...舶
The sound which was judging as the threatening
3 a 曝
low frequency.
and was of the same pattern with the other sound
ゆ
3P
2.0
γ㎞彰納 5娼.¢.
and was emitted more strongly. 3. A. biaculeatus
.墾.寮罵灘.・灘饗
e一 ・..・..
This species is also larger one among the clown−
膨.
most of sounds emitted by this species was low
e.s
卍ひ
毒
矯 娯 験 φ 陶 齢 串 ﹁
曝
蕊む毒書⑪鉱
奪.≒
暫
o,恩
1.o
LS
’
↑拠鴇 襯 騰《;、
pt・・ ”銑 t
3
\悪/≒蚤べ1趨}懇∵一
..ザ 中
、ノ㌦¶溶㌔﹁
.・で...蜘
・1∴.・:脳.轡.〆・
.、 ご.噸.ぎ:・㌔1一 鳥∴饗監..
F1
心
擢
@lr.111 ”/,.,/ i’f,i.i. ,,},/ “:t.’Nl
﹁ ” .軸 騨. 一 .陣 鵤 鱗 咽 噛 . 馴 σ β .擢
”’
お
弧 ゆ 息.賦
描瓢潔
” 一“ ;一’
餓工X蓬露§﹂﹄
⇔ κ 辱
2
霊
■ r , .㌦ で
B S 4 建
Hz for D. C.一1 kHz sound. 400−800 Hz for D. C.一7
1.5
1.cr
T納憩 議 sec.
frequency less than 1 kHz (200−800 Hz: 200−600 ’
=“⋮
.ψ
, “ 略 擢 惣 帥 岬 昂 擢
sound emitting was also low (Fig. 3). The funda−
mental or stronger frequency component of the
▼ 4 a
fish such as A. Polpmnus and the frequency of
四董渥費§σo鉱
6悼. ,.
ポ i 躍 ’ t
播
辱ρ
O,5
Fig. 4. Underwater sound of A. ocellan’s
Tlpme in sec.
2
遍
3 2 奪 岡葦滋3手差畠 ..
D..........嘱 ﹄.....臨
曹
ponent from 5−8 kHz and has strong component at
奪. ’ を
茎
一噸
kHz sound). Moreover, the sound which has frequency com−
.で
鰍・ぽ 鵡畿姥ρ
il一
6 kHz was very rarely heard, but, the source of this
sound was not clear. 2ρ 3,¢
4. A. ocellan’s
When the pair form their tenitory and live in a
口li淘豪.魂1妊.角器溺翻3
sea anemone, this species assume a conservative attitude and are always bathing in a sea anemone
1:ll一...唱..ぴ四一.二
and hardly go away from it. Moreover, the sound emitting of this species is also not so frequent (Fig.
麟、㍉噛ぜへ感_.嫁、......、.鷺.幣属..旨.端..ド..、..:.白油.j
4).
q5 1.e 恥 鞠鷺 恥 薯㏄,
These sounds were classified into three patterns
Fig. 2. Underwater sound of A. Polpmnus
by frequency range and pattern. Namely; 1 . Single pulse sound with very narrow frequency
9齢
range (1.25−2.8 kHz).
闇 Ψ 鱒
3 2 婁
望盤聯ぎ器尋置
2. Low frequency sound composed of the fre− quency component of less than 2.5 kHz. 要
9
φ
α贈
嵐
町⇒
頃サ
伽
養.
頑
な ト
@§ ”
3 . Sound with high frequency component of more than 3 kHz and, sometimes, make a series of
Ire ” ’ ” ’ 21e ’ 3LQ T㎞6 1貸 鷲C.
Fig. 3. Underwater sound of A. biaculeatzts
pulse continued for about a second. Frequently,
those sounds are heard during they are pecking
w・﹂毒.馨
D7㌦
e.s
了㎞e勧 sec,
灘
.F
。.
﹂..
O.5
甑 . 、 此.
}
灘讐 ▼
蝿 岬
ひ.
㍉弧卸.〆一
κ
ぴ ざ ゆコ コ
(
1
Dが.騨4.、ゴ..、
サ、
(
画
ド勢
. 髄垂蕊♂§﹄
辱7
軸
繭
撫
メ
か、 駈
飢
︷
魯脇 κ 避
の4甑掌一
㍗
寧
t?
郁G
同︵
q 曜 冒 騨 騨 鴨 刷 O り 軸 胴
N工菖蕊套ロ詮﹄
レ
砂営
層
雑.
6 4 2
耀零
A. Takemura: Acoustical behavior of clownfishes
り ρ 咽
8
欝.
24
1.5
毒ρ
了㎞e 酌 鱒。,
2
直
奪ρ 2.0 3,0 Tkue in sec,
1.e 2.Q s.o マ醗騰 軌 彩●c,
Fig.5. Underwater sound of/1.ノ地πα伽s
Fig. 6. Underwater sound of A. melanoPus
6
噛
丁9苓
餐一
source of sound because several kinds of these fishes are living in the same tank.
争
2
6 ” ▼ 噛 0
ご&諏g鑑
一 齢
4
・勘
師工鰯
纂
・順 .
, 7 醒
juvenile tank. But, we can not distinguish the
’i
O ﹁
8
In addition to this, the neutral sounds among
騨 7
up on the surface of spawning ground.
the above−mentioned sound are heard from the
毒
縛 ・ ﹂
登番
儲エ鱗忌恥り§σ働慮
こ
毒. ソ罫∵岬.鞠轡讐:「騨.警軍.’// 2
3 2 婁
− 精 ,
N釜召二り奉&塞
3 2 −
}
:.,....,....,,,“....“’L.S,S.L.!L,!,,jLev...,....“ii!“itl
4
.飾
4
,
o.5 te 婁.5 Tkne in sec.
5. A. frenatzcs
Fig. 7. Underwater sound of A. sandaracinos
The frequency of sound emitting of this species is also not so high but following two types of sound
anemone, swims towards a male and stop just in
were heard (Fig. 5). Namely;
front of the latter. Although this kind of movement
1. Sound with stronger frequency component at 5 −6kHz.
2. Sound with stronger frequency component at
was observed rather frequently, sound emitting was not recorded very often. And, sometimes, this type of sound was observed when a female bite a male at
300−600 Hz.
throat or abdomen to cause the latter cramping all
This sound was, sometimes, composed of some
over. After being bitten by a female, the male
pulses.
usually started to peck madly up the surface of a
6. A. melanoPus
rock and emitted the sound.
The following two kinds of sound were heard
While a vigorous sound during eating the feed
from this species (Fig. 6). Namely;
was heard in other species, this species did not emit
1. Sound composed of a very wide frequency
such sounds.
range (D. C.一8kHz) without any stressed fre−
7. A. sandaracinos
quency components.
The sound of this species was composed of
2. Sound composed of low frequency less than 1
single or a series of pulse sound with stressed
kHz with stressed frequency component at 250
frequency component at 300−600 H z and were
−500 Hz.
composed of less than three pulses with wide fre−
It is very doubtfull whether the former sound is
quency range (200−3500 Hz) with two or three
caused by this species. On the other hand, the latter
stressed frequency components (Fig. 7)
sound was often emitted in a series and a female
A fish daShed toward the other fish floated in
used to emit this sound when it get out of a sea
water and the sound was, sometimes, recorded j ust
Bull. Fac. Fish. Nagasaki Univ., No. 54 (1983)
bef6re passing each other.
Moreover, this sound was also recorded when
25
rested in the sea anemone as before. の
On the other hand, a male pecked the surface of
two fishes were floating in water in parallel in such
rock and emitted a high frequency pulse sound
a species as A. melanoPus. The same sound was not
sometimes. The pecking behavior was observed in
always recorded at the same behavior on each
afemale as well. However, female pecked more
specles.
frequently in earlier time, while male began to peck
more frequently later. This movement was some.
Acoustical Behavior during Reproductive Period
times observed up to five hours before spawning
but the remarkable change in behavior was not Acoustical behavior during reproductive period was observed on A. ocellan’s, A. frenatzrs and A.
observed. About five hours before spawning, female move−
sandaracinos. The behavior of these fishes resem−
ment became dull and low frequency pulse sound
bled each other. lt was necessary to observe on
which may be emitted by the female was heard
every species, however, it seemed that the other
frequently and gradually. H:igh pitched sound
species behaved as almost the same as above−
which may be emitted by male were also heard
mentioned three species judging from some brief
sometimes. Pecking movement of a male on the
observation.
su㎡ace of rock became vigorous from aboI覧f two
It was easy to recognize the anival of spawning
hours before spawning. This movement was con−
period from the degree of fatness of abdomen and
tinued until just before spawning. In case of∠L.
the projection of spawning duct of a female and
sandaracinos this movement was carried out either
these characteristic features became distinct three
simultaneously or altemately with the staying fish
or four days before spawning.
on the lookout for the disturbance. If they were
They did not go away from a sea anemone and
disturbed, they stopped this movement and started
were bathing in it almost always. And the move−
to take a lookout in the sea anemone. Most of the
ment like “tail digging” was observed sometimes.
disturbance had no effect on this behavior in case of
From about 30 hours before spawning, this move−
/Locellari’s.
ment wa$ executed vigorously and they became
Just before spawning, the cleaning of spawning
rubbing their side of belly against the rock near the
ground by pecking was more rigorously carried out
sea anemone. This behavior, however, was not
by the female than the male. The female used to
continued for so long a time and theY began to
rub the tip of her spawning duct against the rock
behave like before. Moreover, they sometimes
and pecked the surface of rock for 15minutes
behaved in unusual manners, namely, going away
before spawning, as if she was making sure of the
from the sea anemone, pecking up the surface of the
completion of spawning ground. In case of、4.
rock and going around as if they were looking for
sandaracinos this movement was not observed.
the spawning ground. A series of high frequency
Moreover, the male and female of A.0661彪廊
pulse sound became to be heard more frequently at
emitted individually the low or high frequency
28 hours before spawning, but, it was not clear
pulse sound just before spawning, then she began to
whether these pulses were emitted by this fish or
spawn. However, these sounds were not so closely
not.
connected with spawning, because these sound were
Female began to rub her abdomen against the
not always heard and were the same as usual sound.
rock (spawning ground) and cramp her body, and
These sound were not heard from∠4. sandaracinos.
emitted a strong and low frequency pulse sound
The spawning was carried out in the following
sometimes. This behavior lasted for 20 minutes.
manner:The female spawned and sticked the eggs
followed by a 10−minutes interruption. After re−
in a queue,3−4cm in length, on the surface of rock;
peating these movements two or three times, they
after a short rest she reversed and began to lay eggs
26
A. Takemura: Acoustical behavior of clownfishe$
parallel and close to the first queue; this was re−
same attacks by female when male was madly
peated until the spawning ground, 4−5 cm in diam−
pecking on the surface of the rock. The same sound
eter, were filled with egg mass leaving a marginal
was sometimes used for threat, although the degree
space of O.5−1cm wide. After the spawning,’
??│
of threat of A. polymnus was more vigorous.
male and male take care bf eggs. Female became
A. clarleii emitted the vigorous threatening
gradually less eager to take care of eggs. On the
sound against the angelfish Holacanthus sp., which
other hand, male and female of A. sandaracinos
was put into the same tank. This sound was also a
continued to take care of eggs.
single pulse sound and had very wide frequency
From beginning to end of spawning, sound were
component (D. C.一8 kHzS) and moreover some of
hardly heard and A. ocellan’s did not emit any
frequency component was lasting for morethan one
sound for a long time after spawnlng. On the other
second depending on the resonance of tank.
hand, A. sandaracinos emitted continuously and
Threatening sound was mainly emitted after the
constantly a weak and low frequency sound from
fish found the invader in near place, but threat was
the beginning of cleaning of spawning ground.
frequently canied out by some kind of sound.
However, it seemed that this sound was also not
When we snap the wall of tank by a finger nail, we
connected directly with spawning behavior.
can make a strong pulse sound such as white noise resembling the sound of this fish. A. frenatus took
Threat, Attack a皿d Acoustical Behavior
a threatening behavior against this sound. Namely,
they dashed, looked around and cramped their body There was relatively clear relationship between
and frequently emitted the threatening sound.
the sound and threatening behavior. The clownfish
When this stimulus was frequently given to this
would drive away any invaders from their territory.
fish, the reaction became weak. However, when
It was observed in the tank of A. ocellaris that,
this stimulus was given after an interval, their
cramping their body, they attacked and killed an
threatening movement was provoked again. lt
invader. In case of・4.ノ’renatzcs, male and female,
seemed that this phenomenon was especially re−
especially male, dashed and bit the invader and
markable during a reproductive period and same
ノ
then, sometimes, cramped their body and emitted
phenomenon was observed also in A. sandaracinos.
sound. All of sound were pulse sound, however, the
In case of threat, it was most common to bite
sound that looked like threatening sound with
the invader, but the threat was sometimes carried
cramping had a long duration(0.056 sec), a very
out by means of the gesture (including the sound).
narrow frequency range of about 500 Hz and with
These threatening movement was limited in their
some harmonic.
territory, but the sphere of these movement was
These behaviors were continued until the.in−
reduced in case of a bigger invader.
vader got out of sight. At this time, female also
Observing the behavior and each sound emitted
behaved the same action, but not so actively. It
by clownfish, it seemed that the pattern of sound
was observed that female tried to bite male to get
was not so important and the meaning included in
him back to spawning ground. Very high frequency
sound was variable by the circumstance when the
sound was also rarely heard at that time, but the
sound was emitted and the way of sound proj ection
source of sound was not clear..4, melanopus also
(intensity, duration, repetition rate and etc.).
emitted the same threatening sound with a long duration(0.064 sec)and the sound had sometimes
Mechanism of Sound Production
some harmonics. In this case, female bit the male at throat or abdomen and male emitted a series of
lt seemed that the mechanism of sound produc−
pulse while male was cramping his body. The same
tion among clownfishes was similar, because the
sound were heard from male which was done the
difference among sounds emitted was very little
Bull. Fac. Fish. Nagasaki
Univ., No. 54 (1983)
and the movement during emitting sound was also
William
similar.
of Hubbs
When we observed the movement of fish
carefully during they were emitting sound, we could sometimes find out the special movement.
E. Evans,
suggestions
and kind
director,
Research
and the stuff
Institute
for their
advices.
This References
of mouth or oper-
cle. Frequently, bubble was projected from opercle.
1) Allen, G. G.
Teeth, air bladder and so on may be taken into consideration
executive
Sea World
movement was mainly connected with taking food, consisting of a quick movement
27
(1975) .
The
Anemonefishes,
second ed., T. F. H. Publications Inc. Ltd., N. J.,
as source of sound, but it seems that
199—239.
these are not closely connected with sound projec-
2) Schneider, H. (1964) . Z. Morphol. Okol. Tiere, 53, 454—474.
tion as mentioned by Verwey (1930) and Schneider
3) Verwey, J. (1930). Treubia, 12(3-4),
(1964) .
305-366.
Acknowledgements
The
author
wishes to express
his thanks
to Dr.
水 中音 の 研 究―Ⅷ ク マ ノ ミ属 魚 類 の 音 響 生 態 につ い て 竹
村
陽
Hubbs Sea World Research Instituteで 飼 育 中 の クマ ノ ミ属の魚類7種 類 の音 響 生態 を調 査 した.各 種 と もパ ル ス 音 又 は一 連 のパ ル ス音 を発 す る が,こ れ らの発 生音 は周 波 数 成分 に よ って大 き く3つ の カテ ゴ リー に分類 され,種 に よ り幾 分 相違 点 が 観 察 され た.発 生 音 は か な り強 い音 で あ り,主 に テ リ トリー を守 るた め威 嚇や 産 卵 時 の行 動 中に 聞 か れ る.
28