Content uploaded by José Miguel Simões
Author content
All content in this area was uploaded by José Miguel Simões
Content may be subject to copyright.
1 23
Behavioral Ecology and
Sociobiology
ISSN 0340-5443
Volume 65
Number 4
Behav Ecol Sociobiol (2011)
65:707-716
DOI 10.1007/
s00265-010-1072-3
Stereotypy and variation of the mating call
in the Lusitanian toadfish, Halobatrachus
didactylus
1 23
Your article is protected by copyright and
all rights are held exclusively by Springer-
Verlag. This e-offprint is for personal use only
and shall not be self-archived in electronic
repositories. If you wish to self-archive your
work, please use the accepted author’s
version for posting to your own website or
your institution’s repository. You may further
deposit the accepted author’s version on a
funder’s repository at a funder’s request,
provided it is not made publicly available until
12 months after publication.
ORIGINAL PAPER
Stereotypy and variation of the mating call in the Lusitanian
toadfish, Halobatrachus didactylus
Maria Clara P. Amorim &José Miguel Simões &
Vitor C. Almada &Paulo J. Fonseca
Received: 8 April 2010 / Revised: 13 September 2010 / Accepted: 4 October 2010 /Published online: 19 October 2010
#Springer-Verlag 2010
Abstract Signal attributes should show different degrees of
variability depending on the information to be conveyed.
Species identity is usually associated with stereotyped
features of a signal, whereas other types of information
such as individual quality and motivation are associated
with signal plasticity. Lusitanian toadfish males form
aggregations during the breeding season and emit a tonal
advertisement call (the boatwhistle) to attract mates to their
nests. We test the hypothesis that the boatwhistle can
convey information both on individual identity and moti-
vation by checking how signal parameters vary with time.
We study how the physical (tide level) and social (calling
alone or in a chorus) environments and male calling rate
affect this advertisement signal and how all these external
and internal factors (environment, social and male motiva-
tion) blend to modulate the Lusitanian toadfish’s advertise-
ment call. Boatwhistles of each male were very stereotyped
in short periods of time (minutes), but intra-male signal
variability greatly increased in a longer time scale (days).
Nevertheless, significant differences among males could
still be found even in a long time scale. Pulse period was
the acoustic feature that most contributed to discriminate
among males. Tide level and male calling rate modulated
boatwhistle characteristics, and there was a differential
effect of tide on call attributes depending on male calling
rate. Social acoustic environment only affected calling rate.
These results suggest that inter-individual differences in call
characteristics and call plasticity may mediate both male–
male assessment and mate choice.
Keywords Acoustic communication .Individuality .Signal
plasticity .Batrachoididae .Teleost fish .Tide effects
Introduction
Animals use acoustic signals to convey different types of
information such as species identity, individual identifica-
tion, condition, sexual or aggressive motivation and
territorial ownership. The variability of spectral and
temporal features contained in different calls or in a single
call may convey different types of messages. For example,
signal attributes that convey a species’identity show little
variation among individuals within a species. Conversely,
acoustic features that convey individual identity evidence
strong stereotypy within an individual but show larger
variation between individuals (Bee et al. 2001). Addition-
ally, features such as signal repetition rate or intensity may
indicate motivation or condition and thus provide funda-
mental information for mate choice and for male–male
interactions (Bradbury and Vehrencamp 1998).
In species in which males defend territories in complex
spacing systems to obtain mates, such as in male calling
aggregations, advertisement acoustic signals should in-
form receivers about species and sex identity, sender’s
location, motivation and individual quality (Bradbury and
Vehrencamp 1998). These different messages can be
communicated in some species by large and complex
Communicated by V. Janik
M. C. P. Amorim (*):J. M. Simões :V. C. Almada
Unidade de Investigação em Eco-Etologia,
Instituto Superior de Psicologia Aplicada,
Rua Jardim do Tabaco 34,
1149-041 Lisbon, Portugal
e-mail: amorim@ispa.pt
P. J. Fonseca
Departamento de Biologia Animal e Centro de Biologia
Ambiental, Faculdade de Ciências da Universidade de Lisboa,
Bloco C2, Campo Grande,
1749-016 Lisbon, Portugal
Behav Ecol Sociobiol (2011) 65:707–716
DOI 10.1007/s00265-010-1072-3
Author's personal copy
acoustic repertoires while other species may exhibit more
limited repertoires and rely on the plasticity of certain call
features (e.g. Burmeister et al. 1999). Consequently,
species with small repertoires can be excellent models to
understand the interplay between variable and more
stereotyped signal features not only conveying species
identity but also mediating male–male competition and
mate choice.
Many teleost fish species rely on their calls to acquire
mates (e.g. Malavasi et al. 2003) and to keep intruders
away from their territories (Ladich and Myrberg 2006).
Teleost fish produce relatively simple sounds in small
acoustic repertoires (Amorim 2006) and in a few species
the vocal motor networks and the auditory systems have
been well studied (Bass and Mckibben 2003). Hence,
teleosts provide simple models to study the role of
acoustic signals in reproductive and agonistic decisions
in vertebrates.
Fish from the family Batrachoididae (toadfishes and
midshipmen) have emerged as one of the groups where
acoustic communication is best studied in this taxon (Bass
and Mckibben 2003). During the breeding season, male
batrachoidids produce advertisement calls (boatwhistles or
hums) from their nests that are important for regulating
male spacing and for mate attraction (Fish 1972; Brantley
and Bass 1994). Nesting males form clusters and vocalise
close together in a chorus (Bass 1996; Amorim et al. 2006).
The gulf toadfish (Opsanus beta) changes calling rate and
sound duration both during territorial intrusions and during
increased crepuscular chorus activity, thus depicting a
remarkable ability for teleost fishes to modulate calls
according to the social context (Thorson and Fine 2002;
Remage-Healey and Bass 2005). Further, a recent study
based on passive acoustics has shown that the boatwhistles
of the Lusitanian toadfish Halobatrachus didactylus are
highly stereotyped and show individual differences when
considering short time periods (5–10 min) (Amorim and
Vasconcelos 2008). Individuality in fish vocalisations is
unusual and has only been mentioned for batrachoidids and
fishes of the Mormyridae family (Crawford et al. 1997;
Amorim and Vasconcelos 2008; but see also Myrberg et al.
1993). Calling rate and acoustic features of batrachoidids’
vocalisations are also affected by environmental factors
such as water temperature and lunar cycles (Brantley and
Bass 1994; Maruska and Mensinger 2009). In summary,
previous studies suggest that the batrachoidid advertise-
ment call may carry different messages in one relatively
simple signal.
Male Lusitanian toadfish nest frequently in intertidal
estuarine areas and are thus faced with fluctuating environ-
mental parameters, such as water level and temperature,
which may directly affect acoustic communication. In fact,
temperature influences muscle contraction and thus may
affect sound production (Connaughton et al. 2000). Sound
propagation is highly influenced by tide, since not only the
sound attenuation increases significantly with lowering
water level (Fine and Lenhardt 1983; Mann 2006) but also
background noise often changes with tide. Because mate
attraction in this species relies on a single call type that is
broadcasted in a physically variable and acoustically
complex environment, it provides an excellent model to
study how environmental and social factors affect the
variability of this advertisement signals. Here, we test the
hypothesis that the advertisement call of the Lusitanian
toadfish shows enough inter-male variability and intra-male
plasticity to convey information of male identity and
motivation in chorusing aggregations. We registered vocal-
isations from groups of Lusitanian toadfish in an intertidal
area to examine individual differences (stereotypy) among
well-identified Lusitanian toadfish males in different time
spans, from minutes to 1 week. We examined how
environmental constraints (tide level), male calling rate
and social environment modulate the Lusitanian toadfish’s
advertisement call and discuss the possible role of signal
stereotypy and plasticity.
Material and methods
Study species
The Lusitanian toadfish, H. didactylus (Batrachoididae) is a
benthic marine fish that inhabits coastal areas and brackish
environments from the Gulf of Guinea to the Tagus estuary,
Portugal, appearing occasionally up to the Bay of Biscay
(Roux 1986). Breeding males build nests under rocks in
shallow water and attract females to spawn with long
advertisement calls (boatwhistles) from May to July,
forming conspicuous choruses (dos Santos et al. 2000;
Amorim et al. 2006). Females show low fecundity since
they lay only a few hundreds of large eggs in a single batch
on the roof of a nest (Modesto and Canário 2003; Costa
2004) whose survival is assured through male parental care.
Sound recording and analysis
We deployed 60 artificial concrete shelters every 1.5 m in
rows along the shoreline in an intertidal area of the Tagus
estuary (Portugal, Montijo, Air-Force Base 6; 38°42′N, 8°
58′W). The shelters had a hemicylinder shape capped at one
end (internal dimensions, 50 cm long, 30 cm wide and
20 cm high) and were readily occupied by toadfish in the
breeding season. These nests were only exposed to air
during spring low tides. The water level in the nesting area
varied between 0 m and 2.8 m. Three groups of six to eight
males (n=22) that spontaneously occupied these artificial
708 Behav Ecol Sociobiol (2011) 65:707–716
Author's personal copy
concrete nests were recorded over a period of 8 days in
June/July 2006 and 2007, during the peak of the reproduc-
tive season. Subject males had a mean total length of
42.9 cm (range, 37.9–47.7 cm) and a mean eviscerated
weight of 1,207 g (857–1,612 g). Each male was recorded
for an average of 35 h (11–56 h). Nests with the subject fish
were placed 1.5 m apart in two rows and were at least 15 m
apart from other nests that could be occupied by other
males. Nests’entrances were closed with a plastic mesh that
allowed prey items to enter but prevented males from
abandoning the nest during recordings. The plastic mesh
did not affect acoustic signals and allowed possible visual
interactions. One hydrophone (High Tech 94 SSQ hydro-
phone, sensitivity −165 dB re 1 V/μPa, frequency response
within ±1 dB from 30 Hz to 6 kHz) was placed at about
10 cm from the entrance of each subject male’s nest and
about 10 cm from the substrate. Simultaneous multi-channel
recordings were made to a laptop connected to USB audio
capture devices (Edirol UA25, Roland; 16 bit, 6 kHz
acquisition rate per channel) controlled by Adobe Audition
2.0 (Adobe Systems Inc., 2005). Recorded sounds could be
attributed to each male due to the high acoustic attenuation
observed in the simultaneous multi-channel recordings
between neighbouring males. Water temperature was mea-
sured every 3 h during recording periods and averaged 23°C
(range, 19.5–28°C). All subject fish experienced similar
water temperature variability during recordings.
Boatwhistles have been described in detail in Amorim
and Vasconcelos (2008). This sound has a variable duration
from a few hundred milliseconds up to over a second and is
composed of three different segments or phases character-
ised by different durations, pulse periods, relative amplitude
and dominant frequencies (Amorim and Vasconcelos 2008).
The tonal phase (P2) of the boatwhistle is the longest and
the most characteristic of boatwhistles in the Lusitanian
toadfish and in other batrachoidids (Thorson and Fine
2002; Amorim and Vasconcelos 2008). We analysed boat-
whistles for total sound duration (milliseconds, measured
from the start of the first pulse to the end of the last pulse),
pulse period of the tonal segment P2 (milliseconds, average
peak to peak interval of six consecutive pulses in the
middle of P2), dominant frequency of P2 (hertz, the
frequency with maximum energy in P2), dominant frequen-
cy modulation (the ratio of the dominant frequencies of the
initial and the tonal phases) and amplitude modulation (the
ratio of the mean amplitude of the initial and the tonal
phases). These acoustic parameters are depicted in Fig. 1.A
previous study has shown that these are the most important
acoustic parameters to discriminate among individuals
(Amorim and Vasconcelos 2008). Sound analysis was
carried out with Adobe Audition 2.0 and Raven 1.2.1 for
Windows (Bioacoustics Research Program, Cornell Labo-
ratory of Ornithology, Ithaca, NY, USA).
Note that since the hydrophones were just 10 cm away
from the calling males, any detected effects of tide level on
acoustic parameters were not related to water level
transmission loss dependence (see for example Fine and
Lenhardt 1983; Mann 2006).
Statistical analysis
We calculated the mean and standard deviation (SD) for
each of the five boatwhistle acoustic variables emitted by
13 males in two time frames: short, 10 boatwhistles per
male emitted over a time period of no longer than 10 min;
long, mean of 41 (range, 11–95) boatwhistles per male
produced during up to 8 days. We subsequently computed
the overall means and SDs for each acoustic variable using
the previously calculated mean values for each male. We
determined the within-male variability for the five acoustic
variables by calculating the within-male coefficient of
variance (CV
w
=SD/mean) for each male and subsequently
computed the mean for all males. We also determined the
between-male coefficient of variation (CV
b
) by dividing the
overall SD by the respective overall mean. The ratio CV
b
/
CV
w
was then calculated to obtain a measure of relative
between-male variability for each boatwhistle feature. When
this ratio is larger than one, it suggests that the acoustic
parameter is more variable between individuals relative to
its variability within individuals and could be used as a cue
for individual discrimination (Christie et al. 2004). Kruskal–
Wallis tests were computed to compare differences among
males for each acoustic feature in the two time frames.
Discriminant function analysis (DFA) was carried out as a
multivariate tool to determine if males could be discriminated
based on their sounds considering these five acoustic variables
and to verify which acoustic features better distinguish males.
DFA also gives a measure of discrimination accuracy by
revealing the percentage of sounds correctly assigned to each
individual (Mundry and Sommer 2007). DFA were performed
both on the short time frame (ten boatwhistles per male
emitted within 10 min by 14 males) and on the longer time
frame (ten boatwhistles per male emitted during up to 8 days
by the same males, n=13) data sets. Data were standardised
[(xi−mean)/SD] to remove differences of magnitude observed
between individuals for a given acoustic parameter. We
confirmed that DFA assumptions were met with the inspection
of residual plots (predicted vs residuals values, normal
probability plots), by performing Levene’s and the multivar-
iate Box Mtests for homogeneity of variances/covariances
and by checking tolerance levels to assess possible multi-
collinearity among variables. In addition, to validate the
models obtained, a cross-validation method (‘leave-one-out’)
was carried out (Mundry and Sommer 2007). In this method
each sound is classified by the discriminant functions derived
by the n−1 remaining sounds.
Behav Ecol Sociobiol (2011) 65:707–716 709
Author's personal copy
We also tested whether environmental and social factors
and male calling rate had a significant effect in boatwhistle
acoustic variables in a long time frame with multi-way
analysis of covariance (ANCOVA). In this analysis, we
used sounds registered in the longest possible time span per
individual (up to 8 days) and during different tide levels
and social environments. We considered an average of 34
sounds per male (range=10–92) for 16 males. As data was
unbalanced ANCOVA were based on sum of squares III.
Tide level was included as a factor with three levels: 1=full
tide, 2=ebb tide and 3=low tide. Rising tide was not
considered because of the reduced number of males calling
during this tide level, which was significantly lower than at
high and ebb tide, and lower, although not significantly,
than low tide (Kruskal–Wallis test, n=190, H= 26.11, p<
0.001; Fig. 2). Social environment was another factor with
two levels: 1=calling alone and 2 = calling in a chorus of at
least two males. We included a third factor with two levels
that represented male’s calling rate (1=low rate <9 BWmin
−1
and 2= high rate≥9 BWmin
−1
). Nine number of boatwhistles
per minute (BWmin
−1
) was considered the cutpoint since the
average calling rate for all fish was of 8.8 BWmin
−1
(n=576).
This factor was included because exploratory analysis showed
calling rate affected call features. We also considered calling
rate as a dependent variable. In this case only tide and social
factors were included in the models. Water temperature was
included as a continuous variable (covariate) because it can
influence acoustic parameters (Amorim et al. 2006). When the
effect of this covariate was not significant, multi-way ANOVA
were carried out instead. The final models complied with
normality and homogeneity of variance assumptions. We also
confirmed the absence of multicollinearity between the
predictors (variance inflation factors were always smaller than
5 and tolerance levels larger than 0.1; Montgomery et al.
2006) since water temperature tended to be on average 2°C
higher at low tide than in other tide levels.
Non-parametric statistical tests were carried out when
parametric assumptions were not met. Non-parametric
statistics and ANCOVA tests were carried out with
Statistica (9, Statsoft Inc., USA) and DFA were performed
using SPSS (16.0, SPSS Inc., USA) for Windows.
Results
Stereotypy: short time frame
All five acoustic parameters showed significant differences
between individuals (Table 1). There was a strong stereo-
typy in most acoustic parameters measured, and only the
Flood Full Ebb Low
Tide level
0
1
2
3
No. of calling males
ac
b
bc
c
Fig. 2 Variation of the number of calling males (max= 8) with tide
levels. Dots and error bars are means and standard errors. Different
letters indicate pairwise significant differences given by Dunn tests,
i.e. factor levels with the same letter show no significant differences
Time (s)
0.3 0.5 0.7 0.90.1
0.2
0.4
0.6
0.8
1
0
Frequency (kHz)
DF -60
-40
-20
0
Relative amplitude (dB)
0.5 1 1.5 20
Frequenc
y
(kHz)
PP 10 ms
C
0.3 0.5 0.7 0.90.1
0
-10
-20
10
20
Relative amplitude
A
BD
Sound duration
DF
Fig. 1 Oscillogram (a), sono-
gram (b) and power spectrum
(d) of a boatwhistle. Sound
duration (thick continuous line),
the initial phase (P1, fine con-
tinuous line) and the tonal phase
(P2, fine dashed line) of the
boatwhistle are depicted in the
oscillogram. Dominant frequen-
cy (DF) of the tonal phase is
shown in the sonogram and in
the power spectrum. Detail of
the boatwhistle tonal phase
waveform depicting the pulse
period (PP)(c)
710 Behav Ecol Sociobiol (2011) 65:707–716
Author's personal copy
frequency modulation showed a within-male CV
w
larger
than 0.11 (Table 1). Consistently, with the Kruskal–Wallis
results, all five acoustic variables were more variable
between than within males as all CV
b
/CV
w
ratios were
larger than one (Table 1).
A discriminant function analysis (DFA, n=140, Wilks’
lambda=0.0004, DF = 65, 580, p< 0.001) assigned boat-
whistles to the correct male with an average success of 86%
(range, 30–100%). The first two discriminant functions
explained 74% of data variability with P2 pulse period
weighing most heavily in explaining variation in the first
function and sound duration and amplitude modulation in
the second function (Table 2). After cross-validation the
correct classification assigned by the DFA model was
similar (80%).
Taken together, these data reveal that male Lusitanian
toadfish show individual differences in the properties of
boatwhistle in a time frame of a few minutes.
Variability: long time frame
Considering a period of time as long as possible for each
male (up to 8 days), significant differences between males
were kept for all five acoustic parameters (Table 1);
however, only the pulse period of P2 showed a within-
male CV smaller than 0.1 and all the CV
b
/CV
w
ratios were
smaller than one showing that when a long time frame is
considered all five acoustic variables become more variable
within than between males. A discriminant function
analysis ran on this long-term data set (DFA, n=130,
Wilks’lambda= 0.09, DF=60, 532, p< 0.001) showed that
the average classification success decreased to 45% (range,
0–90%) with the first two discriminant functions explaining
72% of data variability. The percentage of correct classifi-
cation after cross-validation (leave-one-out procedure) was
35%. In conclusion, significant differences among male
calls in a long time scale were still found, although to a
lesser extent than in a short time scale.
Sources of call variability: long time frame
Sound duration decreased significantly during low tide
(Table 3; Fig. 3a) and this parameter was also significantly
affected by the calling rate. The interaction between the
factors calling rate and social environment was significant
(Table 3) and sound duration significantly decreased in
males calling at a low rate and significantly increased in
Table 1 Within-male variability (CV
w
) and between-male variability (CV
b
) for the five acoustic variables analysed from the same 13 Lusitanian
toadfish males in a short (up to 10 min) and in a long time frame (up to 8 days)
Short time frame Long time frame
Acoustic variables Mean
a
(±SD) CV
w
CV
b
CV
b
/
CV
w
H
b
Mean
a
(±SD) CV
w
CV
b
CV
b
/
CV
w
H
c
Sound duration (ms) 723.1 (±161.9) 0.10 0.21 2.11 101.33 686.8 (±190.2) 0.22 0.21 0.95 147.47
Pulse period P2 (ms) 18.9 (±1.4) 0.02 0.07 2.92 109.9 18.5 (±1.7) 0.08 0.04 0.47 111.12
Dominant frequency P2
(Hz)
143.0 (±43.0) 0.11 0.27 2.54 95.08 127.6 (±46.6) 0.31 0.23 0.73 166.53
Frequency modulation 0.9 (±0.3) 0.18 0.21 1.20 63.11 0.8 (±0.4) 0.46 0.21 0.47 78.7
Amplitude modulation 0.7 (±0.1) 0.04 0.11 2.49 111.17 0.7 (±0.8) 0.31 0.23 0.77 175.54
Ten boatwhistles were analysed per male for the short time frame analysis, whereas an average of 41 sounds per male (range= 11–95) were
analysed from the long time frame data set. Note that only 13 individuals are considered since only these are in common in the short and the long
time frame analyses data sets. P2, middle tonal phase in the boatwhistle
a
Results computed for all sounds considered per analysis.
b
Results of Kruskal–Wallis tests (DF=12, n= 130) comparing differences between males for each acoustic feature
c
Results of Kruskal–Wallis tests (DF=12, n= 534) comparing differences between males for each acoustic feature. All comparisons are significant at
p<0.001
Table 2 Standardised canonical DFA coefficients, eigenvalues and
cumulative percentage of variance explained by the first two
discriminant functions of a DFA classifying 14 Lusitanian toadfish
males by their boatwhistles’acoustic characteristics
Discriminant functions
Discriminant variables First Second
Sound duration (ms) 0.05 0.57
a
Pulse period P2 (ms) 0.69
a
0.00
Dominant frequency P2 (Hz) −0.03 −0.22
Frequency modulation −0.25 0.07
Amplitude modulation 0.44 −0.55
Eigenvalue 10.53 9.21
Cumulative % of variance 39.6 74.3
Boatwhistles were emitted in a period of 10 min
a
Discriminant variable with the highest pooled within-groups correlations
with the standardised discriminant functions.
Behav Ecol Sociobiol (2011) 65:707–716 711
Author's personal copy
males calling at a high rate with the raise of social
complexity, i.e. in a chorus situation (Fig. 3b). There was
a difference in sound duration between males calling at
high and low rate only when the fish were singing in a
chorus (Fig. 3b).
P2 pulse period decreased significantly at low tide
(Table 3; Fig. 4). The interaction term between tide and
calling rate was significant. The decrease in the pulse
period with water level was more marked in males calling
at a high rate than at a low rate. Males calling at a high rate
had a shorter pulse period in their calls and significant
differences between call rate levels were observed at ebb
and low tide levels (Fig. 4a). The interaction between tide
level and social environment was also significant. Pulse
period decreased gradually with water level in males calling
alone, but in males calling in groups there was only an
abrupt decrease at low tide. There was, however, no
significant effect of social environment on pulse period as
differences for this parameter between males calling alone
or in a chorus were not significant in any tide level
(Fig. 4b).
P2 dominant frequency was only significantly affected by
calling rate (Table 3) and males calling at a high rate showed
significantly higher values for this acoustic parameter than
males calling at a low rate (Fig. 5).
Only the interaction term between tide level and calling
rate had a significant effect on frequency modulation and
males calling at a high rate showed higher frequency
modulations than less vocal males at ebb tide (Table 3;
Fig. 6). There was no significant effect of any variable or
interaction term in amplitude modulation (Table 3).
Social environment had an effect in the calling rate
(ANCOVA, social environment: F
1,541
=42.17, p<0.001;
covariate water temperature, F
1,541
=6.16, p=0.01). Data
inspection showed that males that call alone mostly call at
low rates, but males in a chorus call at all rates. There was a
significant effect of the interaction between social environ-
ment and tide level (ANCOVA, interaction: F
2,541
=11.55,
p<0.001, tide level: F
2,541
=0.97, p=0.38). Calling rate of
males calling alone increased at low tide and with the
exception of this tide level it was significantly lower than
the calling rate for males in a chorus situation (Fig. 7).
Discussion
Stereotypy vs. variability
There was strong stereotypy in the boatwhistles produced
by Lusitanian toadfish males when considering periods of
10 min consistent with the study of Amorim and
Vas c o n c e l o s ( 2008) based on unidentified fish. Males
differed significantly in all five acoustic parameters and
Table 3 Effects of tide level, social environment and male calling rate on boatwhistles’acoustic variables
Duration (ms) Pulse period P2 (ms) Dominant frequency P2 (Hz) Frequency modulation Amplitude modulation
Factors DF FP DF FP DF FP DF FP DF FP
Tide 2,535 6.01 0.003 2,536 56.42 <0.001 2,535 2.66 0.07 2,536 0.11 0.90 2,536 0.40 0.67
Social 1,535 0.12 0.73 1,536 0.25 0.73 1,535 0.01 0.92 1,536 0.10 0.75 1,536 0.05 0.82
Call rate 1,535 5.05 0.03 1,536 78.43 <0.001 1,535 7.61 0.006 1,536 3.63 0.06 1,536 0.49 0.49
T× S 2,535 0.04 0.96 2,536 4.28 0.01 2,535 0.72 0.49 2,536 0.57 0.57 2,536 0.08 0.93
T× CR 2,535 1.99 0.14 2,536 9.98 <0.001 2,535 0.68 0.51 2,536 7.08 <0.001 2,536 0.30 0.74
S× CR 1,535 4.52 0.03 1,536 2.00 0.16 1,535 0.00 0.94 1,536 0.10 0.76 1,536 0.00 0.97
T× S × CR 2,535 0.17 0.85 2,536 1.06 0.35 2,535 0.38 0.69 2,536 0.41 0.67 2,536 0.11 0.89
T°C 1,535 15.37 <0.001 –– – 1,535 19.05 <0.001 –– – – – –
Temperature was included as a covariate when its effect was significant
Ttide level, Ssocial environment, CR calling rate, T°C temperature
712 Behav Ecol Sociobiol (2011) 65:707–716
Author's personal copy
average correct classification of individuals based on their
calls was high (86%) as shown by DFA, with P2 pulse
period, sound duration and amplitude modulation being the
best parameters to discriminate among males. Individual
differences in signals are important to mediate social
interactions, namely between neighbouring territorial
males (Bradbury and Vehrencamp 1998). Individual
recognition is especially relevant when animals defend
long-term territories, such as breeding Lusitanian toadfish
males do, because territory holders can reduce aggression
towards familiar neighbours, which are less likely to
intrude their territories (‘dear enemy effect’, Temeles
1994). Although this effect was not yet identified in
toadfishes, examples exist in several taxa. For instance,
territorial male bullfrogs (Rana catesbiana) show individual
differences in their advertisement calls that mainly differ in
the fundamental frequency (Bee and Gerhardt 2001a).
Playback experiments have shown that this acoustic
parameter mediates neighbour–stranger recognition in this
species since they show less aggression in response to
familiar calls (Bee and Gerhardt 2001b).
When considering a longer time frame (up to a week)
there were still significant differences among Lusitanian
toadfish males for all five acoustic features, but these become
more variable within a male (Table 1). With the exception of
P2 pulse period, all other parameters showed high intra-male
variability over a week. This variability of the Lusitanian
toadfish advertisement calls in a longer time frame suggests
that boatwhistles are being modulated by either external
factors such as the physical or the social environment or by
Full Ebb Low
Tide level
15
16
17
18
19
20
21
P2 pulse period (ms)
Low
High
Calling rate
A
Full Ebb Low
Tide level
15
16
17
18
19
20
21
P2 pulse period (ms)
Social
B
Alone
Chorus
Fig. 4 Effect of tide level and male calling rate (low rate, filled circle
and high rate, open circle)(a) and of tide level and social environment
(alone, filled square and chorus, open square) on P2 pulse period (b).
Circles and squares are means and error bars are 95% confidence
intervals
Full Ebb Low
Tide level
550
600
650
700
750
800
Boatwhistle duration (ms)
A
Alone Chorus
Social environment
550
600
650
700
750
800
Boatwhistle duration (ms)
B
Low
High
Calling rate
Fig. 3 Effect of tide level on boatwhistle duration (a). Effect of social
environment (calling alone or in a chorus) and male calling rate (low
rate, filled circle and high rate, open circle) on sound duration (b).
Circles are means computed for the covariate water temperature mean
and error bars are 95% confidence intervals
Low High
Callin
g
rate
100
110
120
130
140
150
P2 dom. frequency (Hz)
Fig. 5 Effect of male calling rate on P2 dominant frequency. Dots are
means computed for the covariate water temperature mean and error
bars are 95% confidence intervals
Behav Ecol Sociobiol (2011) 65:707–716 713
Author's personal copy
internal factors such as the internal physiological state of the
male (Remage-Healey and Bass 2005).
Interestingly, the P2 pulse period was the least variable
acoustic feature both within and between males and kept
the CV
w
and CV
b
under 0.1, regardless of the time span
considered. Pulse period corresponds to the sonic muscle
contraction period (Skoglund 1961; Fine et al. 2001), which
is controlled by central vocal pattern generators in
batrachoidids, and is stereotyped at the species level (Bass
and McKibben 2003); nevertheless, P2 pulse period was the
parameter that contributed the most to discriminating
among individuals in a short time scale (DFA) suggesting
it presents fine differences among individuals regardless of
the time scale considered, which might be used for
individual recognition among neighbouring nesting males.
P2 pulse period could also be indicative of male quality. In a
recent study, Amorim et al. (2010) found that in the Lusitanian
toadfish the pulse period reflects male condition and males
with a higher body lipid content produced vocalisations with
shorter pulse periods. These authors suggested that males that
contract the sonic muscles at a very fast rate could reliably be
indicating to neighbouring males or females their better
quality (condition) with the ability to sustain sonic muscle
contraction close to their physiological limit. Consistently,
males of the nonpasserine bird Brown Skuas that produce
long difficult calls close to their performance limit are honestly
advertising a higher breeding success (Janicke et al. 2008).
Sources of variability
Tide level had a significant effect in boatwhistle duration and
P2 pulse period, which showed lower values at low tide.
Consequently, vocal conspicuity was reduced at low tide as
the number of calling males (see “Material and methods”)and
sound duration decreased. A decrease of calling effort at low
tide is expected because low frequency sounds attenuate very
rapidly in low water levels and the chance of being detected
by distant females is reduced (Fine and Lenhardt 1983;
Mann 2006). Nevertheless, males that were calling at a high
rate showed a pronounced increase of the fundamental
frequency (the inverse of the pulse period) with low tide thus
increasing the chances of sound detection at low water levels
(Mann 2006). To the best of our knowledge, the effect of tide
levels on fish vocalisations has not been studied before, but
Barimo and Fine (1998) have mentioned that oyster toadfish
increase calling on incoming tide, suggesting that as other
intertidal fish, batrachoidids may have endogenous activity
rhythms related to tides with lowest levels of activity at low
tide (Gibson 1982). Future work should address the stability
of call rate and characteristics in Lusitanian toadfish males
that inhabit deeper areas and are less subject to the harsh
fluctuating physical environment of the intertidal area.
Social environment affected calling rate in the Lusitanian
toadfish and chorusing males called at higher rates. Accord-
ingly, calling rate is influenced by the vocal behaviour of
nearby male conspecifics in other batrachoidids (Winn 1967;
Fish 1972; Remage-Healey and Bass 2005). For example,
Gulf toadfish males when experimentally placed in an active
calling environment started to vocalise within 48 h (Remage-
Healey and Bass 2005). Males in a chorus probably benefit
from increased mate attraction, reduced assessment costs or
reduced predation risks, although experimental evidence
supporting these hypotheses is scarce (Gerhardt and Huber
2002). We have also observed that high-calling-rate and
probably more motivated Lusitanian toadfish males pro-
duced longer boatwhistles than low-calling rate males, but
only when singing in a chorus. Burmeister and Wilczynski
(2001) examined the influence of androgens on calling
behaviour in the presence and the absence of social acoustic
signals in male green treefrogs (Hyla cinerea) and concluded
that the influence of androgens on the motivation to call
depended on the social stimuli. It is likely that also in the
Lusitanian toadfish, androgens have a differential effect in
boatwhistle acoustic features, namely duration, dependent on
social stimuli.
Full Ebb Low
Tide level
0,6
0,8
1,2
Frequency modulation
Calling rate
0.4
1
Low
High
Fig. 6 Effect of tidal level and male calling rate (low rate, filled circle
and high rate, open circle) on dominant frequency modulation. Circles
and error bars are means and 95% confidence intervals
Full Ebb Low
Tide level
0
2
4
6
8
10
12
14
Calling rate (BWmin
-1
)
Social
Alone
Chorus
Fig. 7 Effect of tidal level and social environment (alone, filled circle
and chorus, open circle) on calling rate. Circles are means computed
for the covariate water temperature mean and error bars are 95%
confidence intervals
714 Behav Ecol Sociobiol (2011) 65:707–716
Author's personal copy
Male calling rate also had a significant effect on acoustic
features of boatwhistles. High-calling rate males produced
on average longer boatwhistles with higher fundamental
frequencies. The higher calling rate and associated longer
sound durations observed in the Lusitanian toadfish could be
related to higher levels of circulating steroid hormones such
as 11-ketotestosterone (11KT), which is a teleost-specific
androgen (Remage-Healey and Bass 2006). Neurophysio-
logical experiments have shown that steroid hormones exert a
rapid and complex neuromodulatory effect in the activity of
the vocal pattern generator and in the vocal behaviour of
batrachoidids, modulating calling rate and call duration
(Remage-Healey and Bass 2003,2004,2006). Also, nesting
batrachoidid males have elevated 11 KT during vocal
advertisement compared to non-calling periods (Knapp et
al. 2001; Remage-Healey and Bass 2005). In addition,
batrachoidid males presented higher calling rate, call
duration and elevated levels in circulating 11KT after
experiencing a territorial intrusion simulated by playback
(Remage-Healey and Bass 2005). In this context, Lusitanian
toadfish males with higher calling rates and longer sounds
could reliably advertise high circulating androgen levels,
which also mediates aggressive behaviour (social status) and
reproduction (Remage-Healey and Bass 2006). Moreover,
calling rate reflects male condition (body lipid content) in
this species (Amorim et al. 2010), suggesting that calling rate
and associated call features could be advertising male quality
traits that are important for social status, territorial defence
and possibly parental ability (Andersson 1994). From the
female perspective, females that would choose males based
on their vocal behaviour could both gain direct benefits (such
as enhanced male parental care and better territories for the
development of their offspring) and indirect (genetic)
benefits (Andersson 1994).
In conclusion, our study has shown that the advertisement
calls of the Lusitanian toadfish differ among individuals even
when considering several days, but high stereotypy is found
only in P2 pulse period when considering longer periods of
time. Differences in this parameter could promote individual
recognition and also advertise male quality if only some males
can perform close to the physiological limit. Our study has also
produced evidence to support the acoustic plasticity hypoth-
esis since calling rate seems to be one of the major factors
influencing boatwhistle characteristics and could be informa-
tive about male motivation and physiological condition (i.e.
male quality). The social acoustic environment, on the other
hand, seems to mainly influence calling rate (vocal facilita-
tion). We suggest that absolute differences in calling rate and
in call characteristics (mainly P2 pulse period, dominant
frequency and sound duration), together with the differential
way in which tide modulates boatwhistles according to male’s
motivation level may be used in male–male assessment and in
mate choice by females.
Acknowledgements We would like to thank Air Force Base No. 6
of Montijo (Portugal) for allowing this study in their military
establishment. This research was funded by the Science and
Technology Foundation, Portugal (project PDCT/MAR/58071/2004,
pluriannual programmes UI&D 331/94 and UI&D 329, grants SFRH/
BPD/14570/2003 and SFRH/BPD/41489/2007).
References
Amorim MCP (2006) Diversity of sound production in fish. In:
Ladich F, Collin SP, Moller P, Kapoor BG (eds) Communication
in fishes. Science Publishers, Enfield, pp 71–105
Amorim MCP, Vasconcelos RO (2008) Variability in the mating
calls of the Lusitanian toadfish Halobatrachus didactylus:
potential cues for individual recognition. J Fish Biol 72:1355–
1368
Amorim MCP, Vasconcelos RO, Marques JF, Almada F (2006)
Seasonal variation of sound production in the Lusitanian
toadfish, Halobatrachus didactylus. J Fish Biol 69:1892–1899
Amorim MCP, Simões JM, Mendonça N, Bandarra NM, Almada VC,
Fonseca PJ (2010) Lusitanian toadfish song reflects male quality.
J Exp Biol 213:2997–3004
Andersson M (1994) Sexual Selection. Princeton University Press,
Princeton
Barimo JF, Fine ML (1998) The relationship of swimbladder shape to
the directionality pattern of underwater sound in the oyster
toadfish. Can J Zool 76:134–143
Bass AH (1996) Shaping brain sexuality. Am Sci 84:352–363
Bass AH, McKibben JR (2003) Neural mechanisms and behaviors for
acoustic communication in teleost fish. Progr Neurobiol 69:1–26.
doi:10.1016/S0301-0082(03)00004-2
Bee MA, Gerhardt HC (2001a) Neighbour-stranger discrimination by
territorial male bullfrogs (Rana catesbeiana): I. Acoustic basis.
Anim Behav 62:1129–1140
Bee MA, Gerhardt HC (2001b) Neighbour-stranger discrimination by
territorial male bullfrogs (Rana catesbeiana): II. Perceptual basis.
Anim Behav 62:1141–1150
Bee MA, Kovich CE, Blackwell KJ, Gerhardt HC (2001) Individual
variation in advertisement calls of territorial male green frogs,
Rana clamitans: implications for individual discrimination.
Ethology 107:65–84. doi:10.1046/j.1439-0310.2001.00640.x
Bradbury JW, Vehrencamp SL (1998) Principles of animal commu-
nication. Sinauer Associates, Sunderland
Brantley RK, Bass AH (1994) Alternative male spawning tactics and
acoustic signals in the plainfin midshipman fish, Porichthys
notatus (Teleostei, Batrachoididae). Ethology 96:213–232
Burmeister SS, Wilczynski W (2001) Social context influences
androgenic effects on calling in the green treefrog (Hyla cinerea).
Horm Behav 40:550–558. doi:10.1006/hbeh.2001.1723
Burmeister S, Wilczynski W, Ryan MJ (1999) Temporal call changes
and prior experience affect graded signalling in the cricket frog.
Anim Behav 57:611–618
Christie PJ, Mennill DJ, Ratcliffe LM (2004) Chickadee song
structure is individually distinctive over long broadcast distances.
Behav 141:101–124
Connaughton MA, Taylor MH, Fine ML (2000) Effects of fish size
and temperature on weakfish disturbance calls: implications
for the mechanism of sound generation. J Exp Biol 203:1503–
1512
Costa JL (2004) The biology of the Lusitanian toadfish, Haloba-
trachus didactylus (Bloch and Schneider, 1801) and its role in the
structuring and functioning of the biological communities;
special reference to the Mira estuary population. Dissertation,
University of Lisbon
Behav Ecol Sociobiol (2011) 65:707–716 715
Author's personal copy
Crawford JD, Cook AP, Heberlein AS (1997) Bioacoustic behavior of
African fishes (Mormyridae): potential cues for species and
individual recognition in Pollimyrus.JAcoustSocAm
102:1200–1212
dos Santos M, Modesto T, Matos RJ, Grober MS, Oliveira RF,
Canario A (2000) Sound production by the Lusitanian toadfish,
Halobatrachus didactylus. Bioacoustics 10:309–321
Fine ML, Lenhardt ML (1983) Shallow-water propagation of the
toadfish mating call. Comp Biochem Physiol A 76:225–231
Fine ML, Malloy KL, King C, Mitchell SL, Cameron TM (2001)
Movement and sound generation by the toadfish swimbladder. J
Comp Physiol A 187:371–379
Fish JF (1972) The effect of sound playback on the toadfish. In: Winn
HE, Olla B (eds) Behavior of marine animals, vol 2. Plenum
Press, New York, pp 386–434
Gerhardt HC, Huber F (2002) Acoustic communication in insects and
anurans: common problems and diverse solutions. The Univer-
sity of Chicago Press, Chicago and London
Gibson RN (1982) Recent studies on the biology of intertidal fishes.
Oceanogr Mar Biol Annu Rev 20:363–414
Janicke T, Hahn S, Ritz MS, Peter H-U (2008) Vocal performance
reflects individual quality in a nonpasserine. Anim Behav 75:91–
98
Knapp R, Marchaterre MM, Bass AH (2001) Relationship between
courtship behavior and steroid hormone levels in parental male
plainfin midshipman fish. Horm Behav 39:335, Abstract
Ladich F, Myrberg AA (2006) Agonistic behaviour and acoustic
communication. In: Ladich F, Collin SP, Moller P, Kapoor BG
(eds) Communication in fishes, vol 1. Science Publishers,
Enfield, pp 121–148
Malavasi S, Torricelli P, Lugli M, Pranovi F, Mainardi D (2003) Male
courtship sounds in a teleost with alternative reproductive tactics,
the grass goby, Zosterisessor ophiocephalus. Environ Biol Fish
66:231–236
Mann DA (2006) Propagation of fish sounds. In: Ladich F, Collin SP,
Moller P, Kapoor BG (eds) Communication in fishes, vol 1.
Science Publishers, Enfield, pp 107–120
Maruska KP, Mensinger AF (2009) Acoustic characteristics and
variations in grunt vocalizations in the oyster toadfish Opsanus
tau. Environ Biol Fish 84:325–337
Modesto T, Canário AVM (2003) Morphometric changes and sex
steroid levels during the annual reproductive cycle of the
Lusitanian toadfish, Halobatrachus didactylus.GenComp
Endocrinol 131:220–231
Montgomery DC, Peck EA, Vining GG (2006) Introduction to linear
regression analysis, 4th edn. Wiley, New York
Mundry R, Sommer C (2007) Discriminant function analysis with non
independent data: consequences and an alternative. Anim Behav
74:965–976
Myrberg AA, Ha SJ, Shamblott H (1993) The sounds of bicolor
damselfish (Pomacentrus partitus): predictors of body size and a
spectral basis for individual recognition and assessment. J Acoust
Soc Am 94:3067–3070
Remage-Healey L, Bass AH (2003) Steroid hormones exert rapid, sex-
specific changes in vocal communication signals in gulf toadfish,
Opsanus beta. Horm Behav 44:72
Remage-Healey L, Bass AH (2004) Rapid, hierarchical modulation of
vocal patterning by steroid hormones. J Neurosci 24:5892–5900
Remage-Healey L, Bass AH (2005) Simultaneous, rapid, elevations in
steroid hormones and vocal signalling during playback challenge:
a field experiment in Gulf toadfish. Horm Behav 47:297–305
Remage-Healey L, Bass AH (2006) A rapid, neuromodulatory role for
steroid hormones in the control of reproductive behavior. Brain
Res 1126:27–35
Roux C (1986) Batrachoididae. In: Whitehead PJP, Bauchot ML,
Hureau JC, Nielsen J, Tortonese E (eds) Fishes of the
Northeastern Atlantic and the Mediterranean, vol III. Unesco,
Paris, pp 1360–1361
Skoglund CR (1961) Functional analysis of swimbladder muscles
engaged in sound production of the toadfish. J Biophys Biochem
Cytol 10(Suppl):187–200
Temeles EJ (1994) The role of neighbours in territorial systems: when
are they ‘dear enemies’? Anim Behav 47:339–350
Thorson RF, Fine ML (2002) Crepuscular changes in emission rate
and parameters of the boatwhistle advertisement call of the gulf
toadfish, Opsanus beta. Environ Biol Fish 63:321–331.
doi:10.1023/A:1014334425821
Winn HE (1967) Vocal facilitation and the biological significance of
toadfish sounds. In: Tavolga WN (ed) Marine Bio-Acoustics, vol
2. Pergamon Press, Oxford, pp 283–304
716 Behav Ecol Sociobiol (2011) 65:707–716
Author's personal copy
