Acoustic signalling during courtship in the painted goby, Pomatoschistus pictus

Abstract

Gobies emit sounds during different stages of reproduction, including courtship, pre-spawning events (in the nest) and spawning. The breeding sounds of the painted goby Pomatoschistus
pictus and associated courtship behaviour were recorded in captivity and described for the first time. Males emitted thump-like sounds mainly when displaying alone in the nest and produced drumming sounds outside the nest. Thumps have never been reported for other species of the genus Pomatoschistus. Thumps were short (~80 ms) very-low frequency (below 100 Hz) non-pulsed sounds, whereas drums were longer (hundreds of ms) and consisted of low frequency (~300 Hz) pulse trains. Thump characteristics varied significantly among males but also showed high within-male variability. The frequency of thump emissions and courtship behaviour (total number of courtship displays, lead and nest display) were positively correlated with male size but not with male somatic condition. Thump bursts emitted during nest displays were significantly longer than when emitted with other behaviours. These results suggest that larger males courted females more intensively, both with visual and acoustic displays, than smaller ones.

Full text

J. Mar. Biol. Ass. U.K. (2007), 87, 5682/1–7
Printed in the United Kingdom
Journal of the Marine Biological Association of the United Kingdom (2007)
doi:
Acoustic signalling during courtship in the painted goby,
Pomatoschistus pictus
Gobies emit sounds during different stages of reproduction, including courtship, pre-spawning events (in
the nest) and spawning. The breeding sounds of the painted goby Pomatoschistus pictus and associated courtship
behaviour were recorded in captivity and described for the first time. Males emitted thump-like sounds mainly
when displaying alone in the nest and produced drumming sounds outside the nest. Thumps have never been
reported for other species of the genus Pomatoschistus. Thumps were short (~80 ms) very-low frequency (below
100 Hz) non-pulsed sounds, whereas drums were longer (hundreds of ms) and consisted of low frequency (~300
Hz) pulse trains. Thump characteristics varied significantly among males but also showed high within-male
variability. The frequency of thump emissions and courtship behaviour (total number of courtship displays,
lead and nest display) were positively correlated with male size but not with male somatic condition. Thump
bursts emitted during nest displays were significantly longer than when emitted with other behaviours. These
results suggest that larger males courted females more intensively, both with visual and acoustic displays, than
smaller ones.
INTRODUCTION
Sound production in teleost fishes is especially conspicuous
in the breeding season and is typically related to territorial
defence and mating activities (Myrberg & Lugli, 2006). Fish
sounds are considered to play an important role during social
communication required for successful mating, such as the
expulsion of intruders from the territory where spawning
will take place (Myrberg, 1997), mate attraction (Lugli et
al., 1996) and mate choice (Myrberg et al., 1986). Mating
sounds may also stimulate the gonadal development of
conspecific females (Marshall, 1972) similarly to bird song,
promote synchronous gamete release (Hawkins & Amorim,
2000), and potentially support species recognition (Myrberg
et al., 1978; Amorim et al., 2004).
The family Gobiidae is considered a case study for
the understanding of acoustic signalling in fish (Bass &
McKibben, 2003) because it includes many vocal species
(reviewed in: Lugli et al., 1997; Myrberg & Lugli, 2006)
and it is one of the few fish taxa where sound playback
experiments have been successful (Tavolga, 1958; Lugli,
1997; Lugli et al., 1996, 2004). Playback experiments of
male goby courtship sounds have shown that they are
effective in female attraction (Tavolga, 1958; Lugli et al.,
1996) and facilitate courtship and sound emission in males
previously stimulated with female conditioned water (Lugli
et al., 2004).
Male gobies typically defend nests that are used as spawning
sites and provide parental care during the reproductive
season (Miller, 1986). Sounds emitted during breeding
behaviour have been described for males of several species of
gobies and can be observed during courtship, pre-spawning
(in the nest, while females are inspecting prospective nests)
or spawning events (Lugli et al., 1997; Myrberg & Lugli,
2006). With few exceptions, goby sound emissions can be
grouped into three major sound types: pulsed sounds that
consist of pulse trains repeated at a slow rate and that can be
amplitude modulated, tonal sounds characterised by faster
pulse repetition rates where pulses fuse together resulting
in a sinusoidal-like waveform, and complex sounds that are
made of a combination of the two (Lugli et al., 1997).
The emission of drumming sounds during pre-spawning
and spawning activities has been documented in f ive species
of sand gobies belonging to the genera Pomatoschistus and
Knipowitschia (Lugli et al., 1995; Lugli & Torricelli, 1999;
Lindström & Lugli, 2000). Lugli & Torricelli (1999) and
Lindström & Lugli (2000) have proposed that these acoustical
signals could be used in mate choice since females enter
several nests before deciding where to spawn and sounds are
only emitted by the male in the nest. Moreover, sand gobies
often live in sympatry in coastal zones or in freshwater (Miller,
1986) suggesting that possible inter-specific differences in
breeding sounds (e.g. Lugli & Torricelli, 1999) could also
potentially be used in species-specific recognition.
Despite the widespread use and the significance of acoustic
signals in gobies, a detailed and quantified characterisation
of the acoustic repertoire has been carried out for only
one species of the genus Pomatoschistus (Lindström & Lugli,
2000). The present study describes the breeding sounds
of the painted goby Pomatoschistus pictus (Miller, 1973) and
M.C.P. Amorim*‡ and A.S.M. Neves†
*Unidade de Investigação em Eco-Etologia, ISPA, Rua Jardim do Tabaco 34, 1149-041 Lisboa, Portugal. †Unidade de
Investigação em Biodiversidade e Desenvolvimento, Universidade Lusófona de Humanidades e Tecnologias,
Campo Grande 376, 1749-024 Lisboa, Portugal. ‡Corresponding author, e-mail: amorim@ispa.pt

2 M.C.P. Amorim and A.S.M. Neves Courtship sounds in the painted goby
Journal of the Marine Biological Association of the United Kingdom (2007)
associated courtship behaviour for the first time. Sounds are
compared among individuals and associated with particular
behavioural displays.
MATERIAL AND METHODS
Study species
The painted goby Pomatoschistus pictus is an exclusively
coastal benthic species that lives on shallow gravel and sand
substrate areas (Miller, 1986). It is a short-lived small goby
living up to 2 years, attaining a maximum length of 57 mm
total length, and reproducing from approximately February
to July, depending on the location (Miller, 1986). Like other
sand gobies, males build nests under empty valves, such as
Pecten or Cardium, and partially or entirely cover the valves
with sediment leaving only one opening (Bouchereau et al.,
2003). Females enter the nest attracted by male displays
and lay their eggs in a single layer on the ceiling of the
nest, and thereafter leave the male to provide parental care
(Bouchereau et al., 2003). Nest owners actively defend their
nest from intruders both outside and during the reproductive
period (personal observation).
Fish collection and maintenance
Fish were caught from shallow water (water depth: 2–9
m) with the help of hand nets during SCUBA diving in
Arrábida, Portugal (38°26'N 9°06'W), in December 2005
and February 2006. Males and females were housed in
separate aerated stock tanks (11-l) provided with a mixture
of artificial and natural seawater, internal power filters,
sand substrate and shelters. Fish gender was recognised by
examining the external papilla (that is rounded in females
and longer and pointed in males) and the existence of nuptial
colours (Bouchereau et al., 2003). Ripe females could also
be recognised by their swollen bellies. A maximum of eight
fish were kept per stock aquarium. Throughout the study,
including sound recordings, fish experienced an approximate
water temperature of 15–16°C, a natural photoperiod, and
were fed daily with shrimps and clams.
Sound Recording
Recording sessions were carried out in May 2006 in two
25-l experimental tanks (similar to the stock tanks) that were
divided in three compartments of equal sizes by means
of double (one opaque and one transparent) removable
partitions. The experimental aquaria were placed on top of
a 14 cm thick layer of rock wool that proved to be effective
in minimizing the conduction of external noise to the tanks
and therefore improved the quality of fish sound recording.
Each end compartment was provided with one half flower
pot of 4.5 cm in diameter, to be used as a nest, and housed
a single male that was left to acclimatise for a minimum of
eight days before it was used in trials. All males covered the
nest with sand, revealing high breeding motivation (e.g.
Figure 1. Oscillogram and spectrogram of a thump bout
produced by a Pomatoschistus pictus male. This thump bout is made
up of 5 thump bursts of 7, 8, 7, 4 and 3 thumps (from the left to
the right). On top, the detail of the oscillogram and spectrogram
of one thump is depicted. Filter bandwidth 124 Hz, FFT size
512 points, time overlap 50.0%, Hamming window. TI – thump
interval; BI – burst interval.
Figure 2. Oscillogram of a drum burst produced by a Pomatoschistus
pictus male (bottom). The drum burst is made up of two drum
sounds with 114 and 21 pulses and 2981 ms and 611 ms duration
(from left to right). Above, the oscillogram and spectrogram of
the shorter drum is depicted, as well as the detail of one pulse.
Filter bandwidth 124 Hz, FFT size 512 points, time overlap 50.0%,
Hamming window. DI – drum interval; PP – pulse period.

Journal of the Marine Biological Association of the United Kingdom (2007)
3Courtship sounds in the painted goby M.C.P. Amorim and A.S.M. Neves
Lindström & Lugli, 2000; Svensson & Kvarnemo, 2005).
The middle compartment housed one ripe female.
Approximately 15 minutes prior to sound recordings
aeration was stopped and one opaque partition was removed
to allow one subject male to have visual contact with the
female, as a courtship pre-stimulus. At the start of the 20
min recording sessions, the transparent partition was also
removed allowing the male to interact freely with the female.
Once the recording was complete, the male was removed,
weighed (fresh weight, W) and measured (standard length,
SL), returned to a stock tank and replaced with another
male. A total of twenty males with mean ±SD (range)=29.2
±3.1 (25 to 37) mm SL and 0.39 ±0.09 (0.28–0.57) g W, were
used and recorded. Sounds were registered with a High
Tech 94 SSQ hydrophone (sensitivity –165 dB re 1 V/μPa,
frequency response within ±1 dB from 30 Hz to 6 KHz)
placed just above the rim of the artificial nest (4 cm above
the substrate), and recorded in wav format to a professional
digital sound recorder (Marantz PMD670) directly to a 1
GB IBM microdrive storage card (sampling frequency of
16 KHz and 16 bit resolution). Sounds were subsequently
analysed with Raven 1.2.1 for Windows (Bioacoustics
Research Program, Cornell Laboratory of Ornithology,
Ithaca, NY, U.S.A.). Only the sounds that presented a good
signal to noise ratio were analysed, which were typically
emitted very close to the hydrophone (~2 cm). Only males
emitted sounds. Sounds could be unequivocally attributed to
males because their intensity varied with distance from the
hydrophone and were consistently associated with particular
displays (see results).
Two sound types were registered during the recording
sessions. The most common sound type resembled a thump
and was composed by a long sound wave (~80 ms) that did not
have a pulsed structure (Figure 1). More rarely, drumming
sounds (drums) made up of short (~8 ms) low-frequency
repeated pulses were also heard (Figure 2). The acoustic
pulse is considered the fundamental unit of the drum sound
(Lindström & Lugli, 2000), but thumps could not be divided
into clear units. Thumps and drums were emitted in acoustic
bouts. Thumps were emitted in sequences of several sounds
(thump bursts, Figure 1). Frequently, thump bursts would be
further clustered into groups, which we named thump bouts
(Figure 1). Similarly, drums were also clustered into drum
bursts (i.e. a succession of drum sounds, Figure 2).
To establish an objective criterion to group thump sounds
into bursts and bouts, and drum pulses into sounds (drums)
and bursts, a log-survivor analysis on time intervals between
sounds (in the case of thumps) or pulses (in the case of drums),
was performed following Lindström & Lugli (2000). This is
a simple graphical method for determining the minimum
interval separating successive bouts of events (Martin &
Bateson, 1993), by plotting the cumulative frequency of
the log-time intervals between events (sounds or pulses in
this case) against time intervals. An abrupt switch point
represents the objective estimate of the minimum interval
that distinguishes separate bouts (Martin & Bateson, 1993).
For this purpose, 142 intervals between thumps from the
20 males were used. Thump intervals were measured from
the end of one sound to the onset of the subsequent sound.
Similarly, for drums, 45 pulse intervals (i.e. pulse periods)
concerning sound emissions from the 20 males were used.
Pulse period was measured peak to peak between consecutive
pulses. The visual inspection of the log-survivor analysis for
thumps (Figure 3A) and drums (Figure 3B) confirmed the
presence of a hierarchical organization of time intervals
both in thump and in drum emissions. Intervals larger
than 700 ms separated thump bursts and those larger than
1400 ms separated thump bouts (Figure 3A). Likewise, time
intervals of 31 ms represented the minimum pulse period
within drums, i.e. pulses with periods larger than this value
belonged to different sounds (Figure 3B). These criteria were
used during sound analysis.
Thumps were analysed for sound duration (ms) and
dominant frequency (Hz, the frequency where the sound has
maximum energy). The number of thumps in a burst, the
interval between thumps within a burst (ms), the duration
of a thump burst (s), the number of bursts in a thump bout,
the interval between thump bursts (s), the number of thump
bursts per bout, and the duration of a thump bout (s) were
also measured. Durations were measured from the onset
to the end of a thump, a thump burst or a thump bout,
depending on the case. Time intervals (within or between
Figure 3. Log survivorship plots of time intervals for thumps (A)
and drums (B) produced by Pomatoschistus pictus males. The
visual inspection of both plots reveal that intervals of 700ms and
of 1400 ms (indicated by dotted vertical lines) are the maximum
within thump burst and thump bout intervals, i.e. intervals larger
than 700 ms and 1400 ms separate different thump bursts and
different thump bouts, respectively. Likewise, intervals of 31 ms
(indicated by the dotted vertical line) represent the maximum
within-drum pulse period.

4 M.C.P. Amorim and A.S.M. Neves Courtship sounds in the painted goby
Journal of the Marine Biological Association of the United Kingdom (2007)
bursts) were measured from the end of one thump to the
start of the following one. Some of these parameters are
depicted in Figure 1.
Drums were analysed for duration (ms), total number of
pulses in a drum sound, pulse duration (averaged for 10
pulses, ms), pulse period (average peak to peak interval of
10 consecutive pulses within a drum, ms) and dominant
frequency (Hz) (Figure 2). Durations were measured from
the onset to the end of a pulse or a drum. Drum and thump
temporal features were measured from the oscillograms and
dominant frequencies were measured from both the power
spectra and the sonogram (f ilter bandwidth 124 Hz, FFT size
512 points, time overlap 50.0% and a Hamming window).
Behaviour repertoire and recording
Behaviour was observed ad libitum, i.e. with no systematic
constraints (Martin & Bateson, 1993), in stock tanks and
during preliminary sound recordings (two months) to
describe the courtship behavioural patterns of P. pictus.
Females were not seen to court the males. Six distinct male
courtship behavioural categories were described:
– Approach: the male makes a series of jerky jumps towards
the female, while undulating his body and fins. The fins are
erected (pelvics, pectorals and dorsal fins) and the head is
darkened. This behaviour is frequently followed by lead.
– Nudge: the male swims towards the female and nudges
her against her flank.
– Lead: the male moves towards and around the female
and attempts to lead her back to the nest swimming in front
of her fluttering the dorsal and caudal fins.
– Eight display: the male swims rapidly in front of the
female in an eight-figure pattern.
– Nest rubbing: the male turns upside down in the nest
and rubs his belly against the nest ceiling. In other gobies
this behaviour is associated with the preparation of mucus
trails that contain active sperm (Svensson & Kvarnermo,
2005) and pheromones that are used in female attraction
and synchronization of spawning activities (Locatello et al.,
2002).
– Nest display: With the body inside the nest, the male
leaves his head out and makes jerky movements raised on the
pelvic f ins, while making downward thrusts with the head,
with rapid opercula movements. This display is frequently
emitting sounds. This behaviour is displayed mostly with the
male alone in the nest but sometimes also with the female
inside.
To associate sound emissions with particular behavioural
categories, male behaviour and sound emissions were
registered during one of the sound recordings sessions
(one session per male) using the Observer (version 4.0 for
student). Only thumps were registered during behavioural
recordings.
Statistical analysis
Thump acoustic features were compared among males
with Kruskal–Wallis tests. Within-male coefficients of
variation (CV=SD/mean×100) for thump parameters were
calculated to assess the individual stereotypy of acoustic
parameters. The Spearman correlation test was used to
explore the potential relationship between the number
of thump bouts emitted per recording session and thump
acoustic parameters with male traits: SL and condition
factor [CF=(W/SLb)*1000; b is the estimated slope from the
regression between log(SL) and log(W) from the 20 male
group used in the experiments and equalled 1.65].
A χ2 test was used to test for independence of behaviour
and thump production. Adjusted residuals from the χ2 test
were used to assess which behaviours were positively or
negatively associated with the emission of thumps. Similar
analyses were not carried for drums since very few sounds
were registered. Comparisons of thump acoustic parameters
that co-occurred with different courtship displays were made
with Kruskal–Wallis tests. The number of different courtship
behaviours observed per recording session (behaviour
frequency) was correlated with the male traits (SL and CF)
and with total courtship frequency using Spearman rank
correlations.
RESULTS
Acoustic repertoire
During recording sessions all males actively courted the
females when they were given free access to them. Sound
production was frequent during courtship interactions and
two sound types could be distinguished, the thump and the
Thump parameters N Mean SD Range Rangeabs CV H P value
Thump duration (ms) 400 81.4 3.97 74.7–88.8 50–101 13.4 49.70 <0.001
Peak frequency (Hz) 400 83.2 9.70 71.3–104.9 37.7–134.6 32.9 41.03 <0.01
Burst duration (s) 400 4.06 1.32 1.71–6.88 1.40–22.35 69.0 84.64 <0.001
No. of thumps 400 7.3 1.98 3.6–11.7 2–30 64.2 76.65 <0.001
Thump interval (ms) 400 584.9 23.10 535.3–622.5 265–912 17.8 13.79 NS
Bout duration (s) 200 8.45 3.49 1.16–13.64 0.35–41.83 75.2 54.22 <0.001
No. of bursts 215 1.96 0.53 1.43–3.33 1–8 59.1 18.53 NS
Burst interval (ms) 171 2.01 1.72 1.05–8.47 0.92–24.41 31.7 102.58 <0.001
Table 1. Means, SDs, range, within male coefficient of variation (CV) for thump parameters among 20 painted goby Pomatoschistus pictus
males. Twenty thumps and an average of 11 thump bouts were considered per male. Descriptive statistics is based on male means except for absolute
range values (rangeabs) that concern all data. SL - standard length; CF - condition factor [(weight/SL1.65)*1000]; CV – within male coefficient of
variation [(SD/mean)*100]. H values are the results of Kruskal–Wallis tests comparing thump parameters among males.

Journal of the Marine Biological Association of the United Kingdom (2007)
5Courtship sounds in the painted goby M.C.P. Amorim and A.S.M. Neves
drum. Thumps were relatively short (~80 ms) non-pulsed
sounds of very low frequency (Figure 1, Table 1). Typical
peak frequencies were below 100 Hz and most sound energy
was under 1 kHz. Thumps were emitted in bursts of 4 to 12
sounds that lasted on average 4 s. Thump bouts were made
up of a succession of typically 1 to 3 bursts and could last
up to 42 s.
Thump parameters differed significantly between males,
with the exception of thump intervals within a burst and
the number of bursts in a bout (Table 1). Coefficients of
variation (CV) showed that within male variability for all
thump parameters was very high, with thump duration
being the least variable characteristic (CV=13.4%). Thump
acoustic parameters were not related with male traits (SL
and CF) except for burst duration that was negatively
related with CF (Spearman correlation: N=20, R=-0.47,
P=0.04) and the number of thump bursts in a bout that
decreased with male SL (Spearman correlation: N=20, R=-
0.46, P=0.04); all other tests: R=-0.43–0.31, P>0.05). The
number of thump bouts emitted per recording session was
positively correlated with male SL (Spearman correlation:
N=20, R=0.52, P=0.02) but not with his condition (N=20,
R=0.047, P>0.05).
Drumming sounds were heard only rarely. On average,
only two sounds (range = 1–5) were recorded per male (total
N=33 sounds). This sound type consisted of low frequency
pulse trains with mean dominant frequencies around 300
Hz [mean ±SD (range)=297.9 ±69.9 (156–431)]. Most of the
acoustic energy of drums was in the range of the 60–500 Hz
(Figure 2). Typical drums had durations of approximately
663 ms [662.9 ±464.1 (211–2981)], with 29 pulses [28.5
±17.8 (9–114)] and pulse periods of 23 ms [22.7 ±2.8 (19–
31)]. Pulses were short, with mean durations of 8.0 ms [±1.8
(6–14) ms]. Drums were usually emitted singly but drum
bursts of up to 3 sounds were heard, lasting up to more than
4 s.
Association with behaviour
Pomatoschistus pictus males courted the females by series
of vigorous displays that included exhibiting erected fins,
darkening parts of the body, leading the female to the
nest, exhibitions in the nest and making sounds. During
the 20 behaviour recording sessions, 1244 behavioural acts
were registered from which 154 were accompanied by the
emission of thumps. All behavioural categories except nudge
were observed to co-occur with sound production (Figure
4). The emission of thumps was significantly associated
with the behaviour nest display (Chi-square test: χ2=448.83,
df=6, P<0.001, Figure 4). While emitting thumps, males
made downward thrusts with the head while closing the
opercula rapidly. Although drums were not registered
during behaviour recording sessions, this sound emission
was only observed when the male was outside the nest near
the female with erected dorsal fins, raised on the pelvic fins,
and quivering his body.
The acoustic features of thumps that co-occurred with
different courtship behaviours were compared. Thumps
bursts were signif icantly longer (Kruskal-Wallis test:
H=37.84, P<0.001; Figure 5A) and with a higher number
of sounds (Kruskal–Wallis test: H=31.12, P<0.001; Figure
5B) when associated with nest displays than with other
behaviours (approach, lead, eight display and nest display;
nest rubbing was excluded from the analysis due to the small
sample size). All other thump features did not differ between
associated behaviours (Kruskal–Wallis tests, H=1.19–5.59,
P>0.05).
Figure 4. Percentage of courtship displays that were
accompanied by the emission of thumps in Pomatoschistus pictus
males. Total number of occurrences per recording sessions is
depicted on top of the histogram bars. Eight – eight display.
Figure 5. Mean ±SE of the duration of thump bursts (top) and
the number of sounds per thump burst (bottom) observed in
sounds emitted by Pomatoschistus pictus males during different
courtship displays. Eight – eight display.

6 M.C.P. Amorim and A.S.M. Neves Courtship sounds in the painted goby
Journal of the Marine Biological Association of the United Kingdom (2007)
Courtship intensity and male traits
Frequencies of total courtship behaviour, lead and nest
display were positively correlated with male SL (Spearman
correlation: N=20, total courtship, R=0.63, P=0.002; lead,
R=0.72, P<0.001, nest display, R=0.67, P=0.001), but not with
male CF (R=-0.15–0.19, P>0.05). The frequency of all other
behaviours were not correlated with male traits (R=-0.25–
0.44, P>0.05). When males courted females more intensively
(i.e. showed a higher total courtship frequency), the visual
displays that increased more markedly were approach,
followed by lead and nest display (Spearman correlation:
N=20, R=0.69–0.85, P<0.001; all other behaviours: nest
rubbing, R=0.47, P=0.04; nudge, R=0.52, P=0.02; eight
display, R=0.33, P>0.05). As expected, the frequency of
sound production was also positively correlated with total
courtship performed by each male (Spearman correlation:
N=20, R=0.79, P<0.001).
DISCUSSION
Painted goby males emitted two sound types during early
stages of courtship. The most common sound, the thump,
was emitted when the male was displaying alone in the nest,
but also while courting the female outside the nest. Drums
were emitted more rarely and only outside the nest, when the
male was stationary near a female, quivering and displaying
the fins. Thumps were short (~80 ms) very-low frequency
(below 100 Hz) non-pulsed sounds, whereas drums were
longer (hundreds of ms) and consisted of low frequency (~300
Hz) pulse trains. Similarly to other sand gobies (Lindström
& Lugli, 2000), sounds of Pomatoschistus pictus were organised
in groups (sound bursts): thumps were organised in thump
bursts and thump bouts (groups of bursts) and drums into
drum bursts.
Thump acoustic characteristics varied significantly
among painted goby males, with the exception of thump
intervals and number of bursts in a bout. However, thump
parameters showed low levels of stereotypy, especially in the
number of thumps in a burst, burst duration, number of
bursts in a bout and bout duration (CV>59%). Similar results
have been found for the drums of P. minutus (Pallas, 1770)
that also show inter-male differences and high within male
coefficients of variation, in particular for sound duration
and the number of pulses per sound (CV>60%) (Lindström
& Lugli, 2000). In addition, sounds produced by males of P.
minutus differ in the sound pressure level, which is positively
correlated with male body length, and in P. marmoratus,
drum pulse period also varies among males, suggesting that
drums emitted by Pomatoschistus males could be used in mate
choice based on male size or other male qualities (Lugli &
Torricelli, 1999; Lindström & Lugli, 2000). Intra-specific
variability in drums was not studied in P. pictus due to the
small sample size.
Thump duration was the least variable acoustic feature in
P. pictus, which could be limited by the mechanism of sound
emission. Tavolga (1971) suggested a shared mechanism of
sound emission for Bathygobius soporator (Valenciennes, 1837),
Neogobius melanostomus (Pallas, 1814) and Gobius jozo (= Gobius
niger Linnaeus, 1758), that consists in the rapid ejection of
strong jets of water through the opercula, occurring during
downward thrusts of the head (also see Stadler, 2002).
Thump-like sounds that also accompany head shakes have
also been described for Tridentiger obscurus males (Temminck
& Schlegel, 1845) (Kishi, 1979), suggesting that this sound
type and associated mechanism of sound production is
shared by several goby species, including P. pictus.
The number of thump bouts and courtship behaviour
(total courtship, lead and nest display) observed per recording
session were positively correlated with male size but not
with male condition. Thump acoustic parameters were not
related with male traits except for thump burst duration and
the number of bursts in a bout that decreased with male
CF and SL, respectively. The smaller number of thump
bursts per bout found in larger males probably represents a
trade-off with the concomitant increase in acoustic activity.
A similar trade-off has been described in other fish species
(e.g. Thorson & Fine, 2002), suggesting the existence of
physiological constraints on sound production.
Thump acoustic parameters also varied with the associated
visual display. Thump bursts were significantly longer and
with a higher number of sounds when associated with nest
display than with other behaviours. These acoustic variables
were however highly correlated. Nest display seems to signal
a high motivation for mating as it significantly increases
with total courtship frequency and male size, and it is
associated with longer emissions of thumps. Taken together,
these results suggest that larger males court females more
intensively than smaller ones, both in terms of visual and
acoustic displays. This indicates that courtship activity
(including sound production) and male size may be under
sexual selection in P. pictus. Consistent with this suggestion,
females of P. minutus and other gobies chose mates on the
basis of multiple traits including courtship intensity (e.g.
Forsgren, 1997; Takahashi & Khoda, 2004) and male size
(e.g. Lindström, 1992; Forsgren, 1997). Goby females are
thought to prefer males with higher courtship intensity
because of their higher genetic quality and parental abilities,
as displaying is energetically costly and correlated with
parental care (e.g. Svensson et al., 2004; Takahashi &
Khoda, 2004). Male size is also an important trait in the
mating success of sand gobies. Larger P. minutus males are
more successful in nest competition, provide more parental
care and are preferred by females (e.g. Lindström, 1992).
Thumps emitted by P. pictus males at a higher rate and in
longer bursts, as observed in larger and more active males
during nest displays, could thus signal male quality and
perhaps parental abilities.
Thump-like sounds have never been reported for
Pomatoschistus spp. or other sand gobies, but drums are
similar to the breeding sounds emitted by other congeneric
species and by Knipowitschia spp. (Lugli et al., 1995; Lugli
& Torricelli, 1999). Although drumming emission was not
registered for P. pictus with both male and female in the nest,
the extension of sound production throughout spawning
cannot be ruled out since no female was observed to spawn
in the present study.
In conclusion, the ability that the painted goby shows
to emit two different sound types associated with different
courtship behavioural patterns, and the positive correlation
found between courtship intensity (including sound
production) and male size, indicates that acoustic signalling

Journal of the Marine Biological Association of the United Kingdom (2007)
7Courtship sounds in the painted goby M.C.P. Amorim and A.S.M. Neves
may play a major role in mate assessment and male mating
success in P. pictus, as suggested for other sand gobies.
The authors would like to thank Frederico Almada, Ines
Tojeira, António Vitorino and Gonçalo Jesus for the help with fish
collection. Miguel Simões also advised on behavioural recording
with the Observer. MCPA thanks FCT, Portugal (grant POSI
SFRH/BPD/14570/2003) for financial support.
REFERENCES
Amorim, M.C.P., Knight, M.E., Stratoudakis, Y. & Turner, G.F.,
2004. Differences in sounds made by courting males of three
closely related Lake Malawi cichlid species. Journal of Fish Biology,
65, 1358–1371.
Bass, A.H. & McKibben, J.R., 2003. Neural mechanisms and
behaviors for acoustic communication in teleost f ish. Progress in
Neurobiology, 69, 1–26.
Bouchereau, J.-L., Houder, V., Marques, A., & Rebelo, J.E.,
2003. A new distribution record and the reproductive strategy
of Pomatoschistus pictus adriaticus (Pisces: Gobiidae) in the
Mediterranean Sea. Journal of the Marine Biological Association of the
United Kingdom, 83, 1157–1161.
Forsgren, E., 1997. Mate sampling in a population of sand gobies.
Animal Behaviour, 53, 267–276.
Gerald, J.W., 1971. Sound production during courtship in six
species of sunfish (Centrarchidae). Evolution, 25, 75–87.
Hawkins, A.D. & Amorim, M.C.P., 2000. Spawning sounds of the
male haddock, Melanogrammus aeglefinus. Environmental Biology of
Fishes, 59, 29–41.
Kishi, Y., 1979. Social behaviour of the goby, Tridentiger obscurus.
Hiyoshi Science Review, 15, 127–146.
Lindström, K., 1992. Female spawning patterns and male mating
success in the sand goby Pomatoschistus minutus. Marine Biology,
113, 475–480.
Lindström, K. & Lugli., M., 2000. A quantitative analysis of the
courtship acoustic behaviour and sound patterning in male sand
goby, Pomatoschistus minutus. Environmental Biology of Fishes, 58,
411–424.
Locatello, L., Mazzoldi, C. & Rasotto, M.B., 2002. Ejaculate of
sneaker males is pheromonally inconspicuous in the black goby,
Gobius niger (Teleostei, Gobiidae). Journal of Experimental Zoology,
293, 601–605.
Lugli, M., 1997. Response of male goby, Padogobius martensii, to
aggressive sound playback following pre-experimental visual
stimulation. Behaviour, 134, 1175–1188.
Lugli, M. & Torricelli, P., 1999. Prespawning sound production in
mediterranean sand-gobies. Journal of Fish Biology, 54, 691–694.
Lugli, M., Pavan, G., Torricelli, P. & Bobbio, L., 1995. Spawning
vocalizations in male freshwater gobies (Pisces, Gobiidae).
Environmental Biology of Fishes, 43, 219–231.
Lugli, M., Pavan, G. & Torricelli, P., 1996. The importance of
breeding vocalizations for mate attraction in a freshwater goby
with a composite sound repertoire. Ethology, Ecology and Evolution,
8, 343–351.
Lugli, M., Torricelli, P., Pavan, G., & Mainardi, D. 1997. Sound
production during courtship and spawning among freshwater
gobiids (Pisces, Gobiidae). Marine and Freshwater Behaviour and
Physiology, 29, 109–126.
Lugli, M., Pavan, G. & Torricelli, P., 2004. The response of male
freshwater goby to natural and synthetic male courtship sound
playback following exposure to different female sexual stimuli.
Ethology, Ecology and Evolution, 16, 55–70.
Marshall, J.A., 1972. Inf luence of male sound production on
oviposition in female Tilapia mossambica (Pisces, Cichlidae).
American Zoologist, 12, 663–664.
Martin, P. & Bateson, P., 1993. Measuring Behaviour. An Introductory
Guide, 2nd edn. Cambridge: Cambridge University Press.
Miller, P.J., 1986. Gobiidae. In Fishes of the north-eastern Atlantic and
the Mediterranean. Vol. 3 (ed. P.J.P. Whitehead et al.), pp. 1019–
1085. Paris: Unesco.
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, 1017–1025.
Myrberg, A.A., Jr & Lugli, M., 2006. Reproductive behaviour and
acoustical interactions. In Communication in fishes. Vol. 1 (ed. F.
Ladich et al.), pp. 149–176. Enfield: Science Publishers.
Myrberg, A.A., Jr., Spanier, E. & Ha, S.J., 1978. Temporal
patterning in acoustical communication. In Contrasts in Behaviour
(ed. E.S. Reese and F.J. Lighter), pp. 137–179. New York: John
Wiley & Sons.
Myrberg, A.A., Jr., Mohler, M. & Catala, J., 1986. Sound
production by males of a coral reef fish (Pomacentrus partitus): its
significance to females. Animal Behaviour, 24, 923–933.
Stadler, J.H., 2002. Evidence for a hydrodynamic mechanism of
sound production by courting males of the notchtongue goby,
Bathygobius curacao (Metzelaar). Bioacoustics, 13, 145–152.
Svensson, O. & Kvarnemo, C., 2005. The importance of sperm
competition risk and nest appearance for male behavior and
female choice in the sand goby, Pomatoschistus minutus. Behavioral
Ecology, 16, 1042–1048
Svensson, O., Nyman, A. & Kvarnemo, C., 2004. Costly courtship
or dishonest display? Intensely displaying sand goby males have
lower lipid content. Journal of Fish Biology, 64, 1425–1429.
Takahashi, D. & Kohda, M., 2004. Courtship in fast water currents
by a male stream goby (Rhinogobius brunneus) communicates the
parental quality honestly. Behaviour, Ecology and Sociobiology, 55,
431–438.
Tavolga, W.N., 1958. The signif icance of under-water sounds
produced by males of the gobiid fish, Bathygobius soporator.
Physiological Zoology, 31, 239–271.
Tavolga, W.N., 1971. Sound production and detection. In Fish
Physiology. Vol. 5 (ed. W.S. Hoar and D.J. Randall), pp: 135–205.
New York: Academic Press.
Thorson, R.F. & Fine, M.L., 2002. Crepuscular changes in
emission rate and parameters of the boatwhistle advertisement
call of the gulf toadfish, Opsanus beta. Environmental Biology of Fishes,
63, 321–331.
Submitted 2 January 2007. Accepted XXXX.