Variability in the sonic muscles of the Lusitanian toadfish (Halobatrachus didactylus): acoustic signals may reflect individual quality

Abstract

Animal vocalizations are good examples of signals that have been shaped by sexual selection and often contribute to resolve contests or the choice of mates. We relate the mass of the sound-producing muscles of a highly vocal fish species, the Lusitanian toadfish (Halobatrachus didactylus (Bloch and Schneider, 1801)), with the senders physical features, such as body size, and reproductive and body condition. In this species, both sexes are known to emit sounds during agonistic interactions and males rely on their mate attraction vocalizations to reproduce. Sonic muscles were highly variable among males (CV = 40%) and females (CV = 33%) and showed sexual dimorphism. Regression analysis showed that variability in the sonic muscles was best explained by total length and fish condition in males and females. Liver mass in both genders, and the mass of the testes accessory glands, also explained sonic muscle variability. These variables explained 96% and 91% of the sonic muscle mass variability in males and females, respectively. As in teleost fishes sonic muscle mass correlates to particular sound acoustic features, we propose that in the Lusitanian toadfish sounds can inform the receiver about the senders quality, such as body size and condition, which are critical information in contests and mate choice.

Full text

Variability in the sonic muscles of the Lusitanian
toadfish (Halobatrachus didactylus): acoustic
signals may reflect individual quality
M.C.P. Amorim, R.O. Vasconcelos, and B. Parreira
Abstract: Animal vocalizations are good examples of signals that have been shaped by sexual selection and often contribute
to resolve contests or the choice of mates. We relate the mass of the sound-producing muscles of a highly vocal fish species,
the Lusitanian toadfish (Halobatrachus didactylus (Bloch and Schneider, 1801)), with the sender’s physical features, such as
body size, and reproductive and body condition. In this species, both sexes are known to emit sounds during agonistic inter-
actions and males rely on their mate attraction vocalizations to reproduce. Sonic muscles were highly variable among males
(CV = 40%) and females (CV = 33%) and showed sexual dimorphism. Regression analysis showed that variability in the
sonic muscles was best explained by total length and fish condition in males and females. Liver mass in both genders, and
the mass of the testes accessory glands, also explained sonic muscle variability. These variables explained 96% and 91% of
the sonic muscle mass variability in males and females, respectively. As in teleost fishes sonic muscle mass correlates to
particular sound acoustic features, we propose that in the Lusitanian toadfish sounds can inform the receiver about the send-
er’s quality, such as body size and condition, which are critical information in contests and mate choice.
Re
´
sume
´
: Les vocalisations animales sont de bons exemples de signaux qui ont e
´
te
´
fac¸onne
´
s par la se
´
lection sexuelle et
qui servent souvent a
`
de
´
terminer l’issue des joutes ou le choix de partenaires. Nous mettons en relation la masse des
muscles producteurs de sons d’un poisson a
`
vocalisations fre
´
quentes, le crapaud lusitanien (Halobatrachus didactylus
(Bloch et Schneider, 1801)), avec les caracte
´
ristiques physiques de l’e
´
metteur, telles que la taille du corps et les conditions
reproductive et corporelle. Chez cette espe
`
ce, les deux sexes sont reconnus pour e
´
mettre des sons durant les interactions
agressives et les ma
ˆ
les de
´
pendent des vocalises d’attraction de leur partenaire pour la reproduction. Les muscles du son
sont tre
`
s variables chez les ma
ˆ
les (CV = 40 %) et les femelles (CV = 33 %) et affichent un dimorphisme sexuel. Une ana-
lyse de re
´
gression montre que la variabilite
´
des muscles du son s’explique le mieux par la longueur totale et la condition
des poissons ma
ˆ
les et femelles. La masse du foie des deux sexes et la masse des glandes accessoires des testicules sont
aussi des variables explicatives de la variabilite
´
des muscles du son. Ces variables expliquent respectivement 96 % et
91 % de la variabilite
´
de la masse des muscles du son chez les ma
ˆ
les et les femelles. Comme chez les te
´
le
´
oste
´
ens la masse
des muscles du son est en corre
´
lation avec certaines caracte
´
ristiques acoustiques particulie
`
res, nous croyons que chez le
crapaud lusitanien, les sons peuvent renseigner l’auditeur sur la qualite
´
de l’e
´
metteur, en particulier sur la taille du corps
et sur la condition corporelle, qui sont des informations essentielles dans les joutes et le choix des partenaires.
[Traduit par la Re
´
daction]
Introduction
Exaggerated sexual secondary male traits have evolved
under sexual selective pressure through male–male competi-
tion, mate choice, or both (Andersson 1994). Animal vocal-
izations are good examples of such traits (Andersson 1994;
Bradbury and Vehrencamp 1998), and empirical evidence
has shown that acoustic signals may influence the outcome
of male contests or be subject to female preference in vari-
ous taxa (e.g., Davies and Halliday 1978; Hasselquist et al.
1996; Ma
´
rquez et al. 2008).
In order for communication to be adaptive, signals should
convey honest information. Signals are expected to be reli-
able if they are costly, and costly signals are likely to impose
even more constraints to animals in poor condition (Zahavi
1975; Grafen 1990). Alternatively, honest signals can be rel-
atively cost-free if signallers and receivers share a common
interest (in this case any signaller can do it), or if physical
or physiological constraints determine the quality of the sig-
nal (indices of quality) (reviewed in Maynard Smith and
Harper 2003). Vocal displays by ectothermic vertebrates are
thought to be one of the most energetically costly activities
(Taigen and Wells 1985; Prestwich 1994; but see Amorim
et al. 2002), and costs are determined mostly by duration,
amplitude, and rate of calling (Prestwich 1994). Hence, these
features can potentially be honest indicators of the sender’s
quality. Further, some acoustic features of vocal displays are
dependent on male’s characteristics, such as size, being a
reliable predictor of fighting ability or mating success (e.g.,
Bee et al. 1999; Ma
´
rquez et al. 2008).
Received 15 January 2009. Accepted 29 May 2009. Published
on the NRC Research Press Web site at cjz.nrc.ca on 30 July
2009.
M.C.P. Amorim.
1
Unidade de Investigac¸a
˜
o em Eco-Etologia,
Instituto Superior de Psicologia Aplicada, Rua Jardim do
Tabaco 34, 1149-041 Lisboa, Portugal.
R.O. Vasconcelos. Departamento de Biologia Animal e Centro
de Biologia Ambiental, Faculdade de Cie
ˆ
ncias da Universidade
de Lisboa, Bloco C2 Campo Grande, 1749-016 Lisboa, Portugal.
B. Parreira. Instituto Gulbenkian de Cie
ˆ
ncia, Rua da Quinta
Grande, n86, 2780-156 Oeiras, Portugal.
1
Corresponding author (e-mail: amorim@ispa.pt).
718
Can. J. Zool. 87: 718–725 (2009) doi:10.1139/Z09-067 Published by NRC Research Press

Many species of teleost fish use acoustic signals during
male–male competition and mate attraction (Ladich 2004;
Amorim 2006), and are thereby expected to be subject to
sexual selection pressure (Andersson 1994). Although fish
sounds do not seem energetically expensive (Amorim et al.
2002), calling activity seems limited by physiological con-
straints such as fatigue resistance (Mitchell et al. 2008),
making a high rate of sound production physiologically
challenging. Some acoustic parameters are also intimately
related to increased fish size, such as lower sound dominant
frequency, higher sound amplitude, and increased pulse du-
ration observed in larger fish (Myrberg et al. 1993; Con-
naughton et al. 2000). This evidence further suggests that
calling rate and particular acoustic parameters can honestly
signal the sender’s quality in male contests and in courtship.
A major mechanism of sound production in fish is the
rhythmical vibration of the swim bladder by the action of
specialized rapid sonic muscles (Ladich and Fine 2006). In
many species, sonic muscles show sexual dimorphism and
hypertrophy during the mating season, with males showing
heavier sonic muscles, with higher number of muscle fibres,
and differences in the fine structure of muscle fibres (Fine et
al. 1990; Brantley et al. 1993a; Connaughton et al. 2000;
Modesto and Cana
´
rio 2003a). Seasonal hypertrophy and
sexual dimorphism of sonic muscles are mediated by andro-
gens (Fine and Pennypacker 1986; Brantley et al. 1993b;
Connaughton et al. 2000), which also modulate courtship
behaviour (Knapp et al. 1999). Differences in the mass of
sonic muscles, and concomitant morphological changes,
thus seem to parallel the increase in vocal output by males
during the breeding season (Amorim et al. 2006).
In this study we examine the possibility that fish acoustic
signals can indicate the sender’s quality by relating the mass
of the sound-producing muscle of a highly vocal fish species
with its physical features. In a first step we investigated if
certain external features, such as fin size and mouth width,
that are associated with agonistic displays (Vasconcelos and
Ladich 2008) show sexual dimorphism. This part of the
study was carried out under the premise that these could po-
tentially be sexually selected traits and could give informa-
tion on the quality of an individual (e.g., Engen and Folstad
1999). Secondly we explored the relation between sonic
muscle mass and several traits, such as any dimorphic exter-
nal feature, body size, reproductive status, and body condi-
tion. We use the Lusitanian toadfish (Halobatrachus
didactylus (Bloch and Schneider, 1801)) (Batrachoididae) as
a model because they are versatile and prolific sound pro-
ducers, and males show a pronounced increase in the sonic
muscles mass and sonic activity during the breeding season
(Modesto and Cana
´
rio 2003a; Amorim et al. 2006; Amorim
et al. 2008). Indeed, batrachoidids, including the Lusitanian
toadfish, have been models for studies of acoustic communi-
cation, as they show prolonged bouts of vocal activity, males
nest in shallow water and are relatively easy to access, react
to playback experiments, and have been subject to a large
body of neurobiological studies (Cohen and Winn 1967;
Barimo and Fine 1998; Bass and McKibben 2003; Modesto
and Cana
´
rio 2003a, 2003b; Remage-Healey and Bass 2005).
Furthermore, batrachoidids present intra- and inter-sexual di-
morphism in the brain, sonic muscle, and vocal behaviour,
with territorial type I males showing reproductive singing
behaviour. whereas type II males (sneakers) and females
only produce agonistic calls (Bass and McKibben 2003).
Material and methods
Study species
During the breeding season (May–July), male Lusitanian
toadfish emit advertisement calls (boatwhistles) to attract fe-
males to the nests that they defend in estuarine shallow
waters (dos Santos et al. 2000; Amorim et al. 2006). Males
mate with several females, and care for the fertilized eggs
attached to the nest’s ceiling until the young are free-swim-
ming (dos Santos et al. 2000). Besides the boatwhistle, three
other sound types are commonly produced by nesting males:
grunt trains, long grunt trains, and double croaks, as well as
other less frequent sound emissions such as croaks and
mixed croak–grunt calls (Amorim et al. 2008). Similar to
other batrachoidids, this species presents sexual polymor-
phism with two male morphotypes that differ in morphomet-
ric and endocrine characteristics, as well as in vocal
behaviour. Nest-guarding males (type I) differ from sneak-
ing males (type II) by having smaller testes (sevenfold),
larger accessory glands (threefold; the accessory glands are
part of the male reproductive apparatus, secrete mucosub-
stances, and are connected to the spermatic duct), and higher
(sixfold) 11-ketotestosterone levels (Modesto and Cana
´
rio
2003a, 2003b). Females and sneaker males are only known
to emit grunt trains and females show lighter sonic muscles
than males, with type II males presenting intermediate sonic
muscle mass to females and type I males (Modesto and Can-
a
´
rio 2003a). Sonic muscles of type I males, but not of type
II males or of females, experience hypertrophy during the
breeding season (Modesto and Cana
´
rio 2003a), mirroring an
increase in vocal activity (Amorim et al. 2006).
Specimen and sample collection
Fish samples were collected by trawling, angling, and hand
capture by local fishermen during the months of June, August,
and September in 2003 and from April to September in 2004
from Tagus estuary, areas of Montijo (38842’N, 8858’W) and
Barreiro (38839’N, 9804’W). Specimens were kept frozen in
the laboratory until measured. This sample included both re-
productive (40.5% of males and 76.5% of females) and non-
reproductive specimens, since the breeding season typically
lasts from May to July (Modesto and Cana
´
rio 2003a).
To investigate the existence of sexual dimorphism in ex-
ternal morphological traits involved in agonistic displays,
we took the following measurements: mouth width (MW,
maximum width of the lower lip); pectoral fin length (PL,
base of the pectoral fin to the tip of its largest ray); ventral
fin length (VL, base of the ventral fin to the tip of its largest
ray); and dorsal fin length (DL, length of the largest fin ray
of the first dorsal fin). All mouth and fin measurements
were made to the nearest millimetre with callipers.
Gonad (M
G
; Fig. 1A), accessory glands (in males, M
AG
;
Fig. 1A), and liver (M
L
) mass were tallied to the nearest
milligram. Both sonic muscles, which are embedded in the
sides of the swim bladder (Figs. 1A, 1B), were gently cut
from the swim-bladder wall with a pair of fine dissection
scissors and were also weighed (M
SM
) to the nearest milli-
gram. We also obtained total length (TL), measured to the
Amorim et al. 719
Published by NRC Research Press

nearest millimetre, and eviscerated body mass (M
E
), meas-
ured to the nearest gram.
Males (n = 79) used in this study were 33.04 ± 5.9 cm TL
(mean ± SD; range 17.4–44.5 cm TL) and weighed 629 ±
303.1 g (range 84–1421 g) in eviscerated mass, whereas fe-
males (n = 34) were 27.02 ± 3.5 cm TL (range 18.7–
33.8 cm TL) and weighed 304 ± 109.0 g (range 107–547 g)
in eviscerated mass. All males were likely type I males.
Type II males were not considered in this analyses because
they were captured in very small quantities (n = 2).
Statistical analysis
We ran analysis of covariance (ANCOVA) to explore the
existence of differences between sexes for fins (PF, VF, and
DF) and mouth width (MW) variables, controlling for body
size (TL). Initial analyses included an interaction term, which
was subsequently removed because it was not significant in
all cases. We log
10
-transformed the dependent variables (fin
length and mouth width) and TL to meet the assumptions of
the models. Kolmogorov–Smirnov tests confirmed that the
assumption of normality was met in all analyses.
To quantify sonic muscle mass variability we calculated
the mean ± SD (range), as well as the coefficient of variations
(CV = (SD/mean)  100), for males and females. Sexual di-
morphism in sonic muscle mass was tested with ANCOVA
with logM
SM
as the response variable, sex as the factor, and
logM
E
as the covariate. As above, the interaction term was
not included in the final model because it was not significant.
We fitted a multiple regression model with a stepwise pro-
cedure to explain variation in sonic muscle mass (dependent
variable). We considered VL and DL in the initial model, as
they were sexually dimorphic (see results). We included TL
in the model as a metric of body size. We also considered
the mass of gonads, accessory glands (in males), and liver
as independent variables. We controlled for the influence of
body size on M
G
by using residuals of the simple linear re-
gression of M
G
on M
E
(RM
G
) in the multiple regression
model. We considered the eviscerated mass (M
E
) to represent
body mass because it is independent of M
G
, M
AG
, and M
L
.
Likewise, we controlled for the influence of size on varia-
tions of M
AG
(males only) and M
L
by regressing these varia-
bles on M
E
(RM
AG
and RM
L
, respectively). Similarly, we
used the residuals of the regression of M
E
on TL (COND) as
a metric of condition. Positive residuals indicate that males
are heavier than predicted and have good body condition,
whereas negative residuals represent animals with poor con-
dition. We log
10
-transformed TL and mass data both in the
simple and in the multiple linear regressions to meet the as-
sumptions of normality and to linearize allometric relation-
ships. Separate models were fitted for males and females.
All model assumptions were met for both male and female
models. All model residuals were normally distributed.
Additional residual analysis was performed using Durbin–
Watson statistics (males = 2.08 and females = 1.83), resid-
uals autocorrelation plots, and multicolinearity tests between
all used variables.
All statistical analyses were performed using R version
2.8.0 (R Foundation for Statistical Computing, Vienna,
Austria) and SPSS version 16.0 for Windows (SPSS Inc.,
Chicago, Illinois, USA).
Results
External morphological sexual dimorphism
Mouth width and pectoral fin length did not differ be-
tween males and females (ANCOVA; MW: F
[1,110]
= 2.10,
P > 0.05; PF: F
[1,110]
= 2.59, P > 0.05), but the ventral and
the dorsal fins were longer in females, controlling for body
length (ANCOVA; VF: F
[1,110]
= 14.32, P < 0.001; DF:
F
[1,76]
= 5.74, P < 0.05) (Fig. 2). The covariate TL had a
significant effect on mouth width and fin length, which in-
creased with body length (ANCOVA; MW: F
[1,110]
=
834.65; PF: F
[1,110]
= 494.76; VF: F
[1,110]
= 315.28; DF:
F
[1,76]
= 106.24, P < 0.001; Fig. 2).
Sonic muscle variability
Sonic muscle mass showed considerable variation among
males (11.09 ± 4.43 g, range 1.68–20.93 g) and among fe-
males (5.26 ± 1.75 g, range 1.74–8.71 g). Coefficient of var-
iation for this parameter was 39.9% in males and 33.2% in
females. Moreover, sonic muscle mass showed significant
dimorphism between sexes (Fig. 3), controlling for body
mass (ANCOVA; sex: F
[1,110]
= 21.37, P < 0.001; M
E
:
F
[1,110]
= 1549.19, P < 0.001).
Fig. 1. (A) Dissected type I male Lusitanian toadfish (Halobatrachus didactylus) showing the swim bladder (SB), the gonads (G), and the
accessory glands (AG). (B) Dorsal view of one lobe of the swim bladder depicting the embedded sonic muscle (SM).
720 Can. J. Zool. Vol. 87, 2009
Published by NRC Research Press

The best multiple regression model fitted with a stepwise
procedure (males: F
[4,74]
= 447.4, P < 0.001, r
2
= 0.96; fe-
males: F
[3,30]
= 105.6, P < 0.001, r
2
= 0.91) included TL,
condition, and liver mass (RM
L
) as explanatory variables in
both male and female models (Table 1). The residual acces-
sory gland mass (RM
AG
) was also included as a significant
independent variable in the final model for males. Body
length was the first variable included in the models and ex-
plained most of the variance of sonic muscle mass both for
males (88.7%) and females (81.9%) (Fig. 4A). Condition
(COND) explained an additional 6% of its variance in males
and 7.1% in females (Fig. 4B). The remaining variability ex-
plained by the final models was accounted by accessory
gland and liver mass in males (1.4%; Figs. 4C, 4D) and by
the liver mass (2.4%) in females (Fig. 4C).
Discussion
Sexual dimorphism in external morphological traits and
in sonic muscle mass
We examined the existence of sexual dimorphism in ex-
ternal morphological traits and in sonic muscle mass, which
are associated with visual and acoustic signals used during
social interactions. In the breeding season, male Lusitanian
toadfish defend territories centred in the nest by displaying
erected fins (dorsals and pectorals), the mouth wide open,
and the body raised on the pelvic fins (Vasconcelos and La-
dich 2008). Territorial defence also includes chasing the in-
truder, biting, and mouth locking (M.C.P. Amorim and R.O.
Vasconcelos, personal observation). Because the reproduc-
tive success of males depend on their ability to hold good
territories, it is plausible to hypothesize that sexual dimor-
phism in fin and mouth size could have evolved through
sexual selection favouring males with traits that are advanta-
geous in agonistic displays (Andersson 1994), such as found
in other teleosts (e.g., Oliveira and Almada 1995). Likewise,
acoustic displays seem to play a major role in both agonistic
and courtship contexts in the Lusitanian toadfish (dos Santos
et al. 2000; Amorim et al. 2006), and males with heavier
sonic muscles should also be favoured. We found differen-
ces between gender in dorsal and ventral fins but not in the
pectoral fin or in mouth width. Curiously, females had lon-
ger dorsal and ventral fins than males, contrary to the ex-
pected if these fins would have an important role in the
outcome of agonistic interactions. There is, however, a mod-
erate inter- and intra-sexual size dimorphism in body size
with only type I males being found at larger sizes (Modesto
and Cana
´
rio 2003b; Fig. 2 in this study), suggesting that a
large body size is an advantage for nesting males during ter-
ritorial defence. We also observed sexual dimorphism in the
sonic muscles, with males having significantly heavier sonic
muscles than females at a given length (approximately 25%
for mean TL of 30 cm), in agreement with the findings of
Modesto and Cana
´
rio (2003a). Consistently, external sexual
Fig. 2. Relation between fin length (VF, ventral fin; DF, dorsal fin;
PF, pectoral fin) and total length (TL), and between mouth width
(MW) and TL, in male (
*
) and female (*) Lusitanian toadfish
(Halobatrachus didactylus).
Fig. 3. Relation between sonic muscle mass (M
SM
) and body evis-
cerated mass (M
E
) in male (
*
) and female (*) Lusitanian toadfish
(Halobatrachus didactylus).
Table 1. Results of the multiple regression analyses of sonic muscle mass (log M
SM
)
on total length (log TL), condition (COND), residual accessory gland mass (RM
AG
;
males only), and residual liver mass (RM
L
) for male and female Lusitanian toadfish
(Halobatrachus didactylus).
Gender Coefficients Estimate SE tP
Males (n = 79) Intercept 6.76 0.22 –30.72 <0.001
log TL 2.60 0.06 40.75 <0.001
COND 0.12 0.01 9.21 <0.001
RM
AG
0.06 0.01 4.59 <0.001
RM
L
0.03 0.01 2.25 <0.05
Females (n = 34) Intercept –6.47 0.48 –13.59 <0.001
log TL 2.45 0.14 16.97 <0.001
COND 0.08 0.02 4.14 <0.001
RM
L
0.06 0.02 2.88 <0.01
Amorim et al. 721
Published by NRC Research Press

dimorphism in other batrachoidids is restricted to differences
in body size and to the shape of the urogenital papilla
(Brantley and Bass 1994), but differences in sonic muscle
mass between sexes can amount to 600% in plainfin mid-
shipman (Porichthys notatus Girard, 1854), owing to the dif-
ferent acoustic activity shown by different gender and male
Fig. 4. Relation between the sonic muscle mass (M
SM
) and the independent variables used in the regression model for male (
*
) and female
(*) Lusitanian toadfish (Halobatrachus didactylus): (A) total length (TL), (B) condition (COND, the residuals of eviscerated body mass on
total length), (C) residuals of liver mass on eviscerated body mass (RM
L
), and (D) residuals of accessory gland mass on eviscerated body
mass (RW
AG,
for males only). In (B), (C), and (D), the y axis represents the residuals from M
SM
regressed on TL, i.e., sonic muscle mass
with the effect of TL removed.
722 Can. J. Zool. Vol. 87, 2009
Published by NRC Research Press

morphotypes (Brantley and Bass 1994). This evidence sug-
gests that in the Lusitanian toadfish and in other batrachoi-
dids, acoustic signals and body size may reveal information
about individual quality during contests and mate choice. In
many species, body size and sexual secondary male traits
such as acoustic signals can directly affect the outcome of
male–male contests and mating success (e.g., Davies and
Halliday 1978; Castellano et al. 2000).
The sexual dimorphism in sonic muscle mass found in this
study is consistent with the findings of Modesto and Cana
´
rio
(2003a), which reported that swim-bladder mass (swim blad-
der plus embedded sonic muscles) shows sexual polymor-
phism, i.e., it is larger in type I males, intermediate in type II
males, and smaller in females. Sexual dimorphism in the
swim bladder, sonic muscle fibres, and neural circuitry of
sound production is typical among batrachoidids (Fine et al.
1984, 1990; Modesto and Cana
´
rio 2003b; see review in Bass
and McKibben 2003) and suggests that acoustic communica-
tion plays a prevalent role over other channels of communi-
cation and is essential for reproduction in the Lusitanian
toadfish as in other batrachoidids (Bass and McKibben
2003). Curiously, we found that the CV for sonic muscle
mass in males was only 7% higher than in females. Consider-
ing that our samples included nonspawners and that only
males experience sonic muscle hypertrophy associated with
the breeding season (Modesto and Cana
´
rio 2003a), it was
expected that females would show less variability than males
in the mass of sonic muscles. The lack of a sharper difference
between genders could reflect the relatively low sonic muscle
sexual dimorphism and also the possibility that females have
a higher vocal activity than traditionally described.
Traits affecting sonic muscle variability
This study showed that there is considerable variation in
sonic muscle mass both in males (CV = 40%) and in fe-
males (CV = 33%). Multiple regression analysis revealed
that body length and condition were good indicators of sonic
muscle mass in both gender. These results are consistent
with those of Modesto and Cana
´
rio (2003a), which showed
that swim-bladder mass increases with body size (eviscer-
ated body mass) in mature and immature specimens of this
species. Accordingly, swim bladder and associated sonic
muscles show continuous growth in other batrachoidids
(Fine et al. 1990; Brantley et al. 1993a). Sonic muscle mass
of spawning cod has also been reported to be positively as-
sociated not only with body size and condition but also with
fertilization potential (Rowe and Hutchings 2004), suggest-
ing that acoustic features associated with sonic muscle mass
also reveal information about individual quality in this vocal
species. Likewise, in other taxa, acoustic cues such as sound
frequency or acoustic repertoire size are related to body
mass and (or) condition (e.g., Davies and Halliday 1978;
Clutton-Brock and Albon 1979; Mager et al. 2007). For ex-
ample, male common loons (Gavia immer (Bru
¨
nnich, 1764))
in better condition and of larger body mass produce lower
frequency sounds (yodels) during territorial defence (Mager
et al. 2007). Mager and colleagues have shown with a play-
back experiment that lower frequency calls elicit stronger re-
actions from receivers, suggesting that dominant frequency
of the yodel may honestly communicate fighting ability
(Mager et al. 2007).
Residuals of liver mass gave a minor (1%–2%) but signif-
icant contribution to explain sonic muscle mass variability in
both genders of Lusitanian toadfish. Both condition and hep-
atosomatic indices (analogous to the residuals of eviscerated
body mass and liver mass used in the present study) are
thought to represent good measures of condition in fish
(Chellappa et al. 1995) and show minimal values at the end
of the breeding season in the Lusitanian toadfish, probably
associated to gamete production in females and to the in-
creased metabolic needs of territorial defence and vocal ac-
tivity in males (Modesto and Cana
´
rio 2003b). Likewise the
residuals of accessory gland mass contributed to sonic
muscle mass in males. These glands are responsible for the
exocrine production of mucosubstances, which seems to be a
common feature of batrachoidids and of many other teleost
species, and are thought to embed sperm and create sperm
trails to reduce sperm dispersion (Barni et al. 2001). This
would increase the chances of fertilizing females eggs in
the nest by type I males, who have considerable larger ac-
cessory glands than type II males (Modesto and Cana
´
rio
2003b). Nesting males with larger accessory glands may
therefore gain higher chances of providing parental care to
a higher percentage of own offspring than males with
smaller accessory glands. Interestingly, the residuals of go-
nad mass did not enter in the final model, although it shows
a similar seasonal variation as sonic muscle and accessory
gland mass (Modesto and Cana
´
rio 2003a).
Acoustic signals have a relevant role in the mating system
of different taxa, allowing animals to convey information
about their quality as mates, competitors, or both (Ander-
sson 1994). Although fish are probably the largest group of
sound-producing vertebrate, having evolved an outstanding
variety of sonic organs, the functional role of their signals
remains largely unknown, especially when comparing with
the wealth of knowledge existent for other taxa (Ladich
2004; Ladich and Fine 2006). The present study shows that
the variability of the sonic muscle mass could indicate indi-
vidual quality in the Lusitanian toadfish, namely larger body
size and better condition (somatic and liver) in both males
and females and larger accessory glands in males. In batra-
choidids, larger males with heavier sonic muscles show
higher calling capabilities and emit sounds with higher am-
plitude (Fine et al. 2001; Vasconcelos and Ladich 2008).
Frequency of sound is imparted by sonic muscle contraction
rather than swim-bladder resonance, and it does not vary
with fish size in the oyster toadfish (Opsanus tau (L.,
1766)) (Fine et al. 2001), although grunt dominant fre-
quency decreases with fish size in the Lusitanian toadfish
(Vasconcelos and Ladich 2008). Females approaching a Lu-
sitanian toadfish chorus could thus potentially select a better
quality male based on acoustic cues, such as calling rate or
call amplitude. Likewise males could judge their opponents
based on grunt amplitude and dominant frequency. Future
work is needed to associate calling rate and acoustic charac-
teristics of sounds with sonic muscle mass and reproductive
success to further support our conclusions.
Acknowledgement
We thank Montijo local fishermen for providing the
specimens used in this study and Paulo Fonseca and Yorgos
Stratoudakis for providing valuable comments on the manu-
Amorim et al. 723
Published by NRC Research Press

script. This study was supported by Ministe
´
rio da Cie
ˆ
ncia,
Tecnologia e Ensino Superior (Portugal) with the project
PDCT/MAR/58071/2004 and the Fundac¸a
˜
o para a Cie
ˆ
ncia e
a Tecnologia grants (SFRH/BD/30491/2006 to R.O.V. and
SFRH/BPD/14570/2003 to M.C.P.A.).
References
Amorim, M.C.P. 2006. Diversity of sound production in fish. In
Communication in fishes. Edited by F. Ladich, S.P. Collin, P.
Moller, and B.G. Kapoor. Science Publishers, Enfield, N.H.
pp. 71–104.
Amorim, M.C.P., McCracken, M.L., and Fine, M.L. 2002. Meta-
bolic costs of sound production in the oyster toadfish, Opsanus
tau. Can. J. Zool. 80(5): 830–838. doi:10.1139/z02-054.
Amorim, M.C.P., Vasconcelos, R.O., Marques, J.F., and Almada,
F. 2006. Seasonal variation of sound production in the Lusita-
nian toadfish, Halobatrachus didactylus. J. Fish Biol. 69(6):
1892–1899. doi:10.1111/j.1095-8649.2006.01247.x.
Amorim, M.C.P., Simo
˜
es, J.M., and Fonseca, P.J. 2008. Acoustic
communication in the Lusitanian toadfish, Halobatrachus
didactylus: evidence for an unusual large vocal repertoire.
J.Mar.Biol.Assoc.U.K.88(05): 1069–1073. doi:10.1017/
S0025315408001677.
Andersson, M. 1994. Sexual selection. Princeton University Press,
Princeton, N.J.
Barimo, J.F., and Fine, M.L. 1998. Relationship of swim-bladder
shape to the directionality pattern of underwater sound in the
oyster toadfish. Can. J. Zool. 76(1): 134–143. doi:10.1139/cjz-
76-1-134.
Barni, A., Mazzoldi, C., and Rasotto, M.B. 2001. Reproductive ap-
paratus and male accessory structures in two batrachoid species
(Teleostei, Batrachoididae). J. Fish Biol. 58(6): 1557–1569.
doi:10.1111/j.1095-8649.2001.tb02312.x.
Bass, A.H., and McKibben, J.R. 2003. Neural mechanisms and
behaviors for acoustic communication in teleost fish. Prog.
Neurobiol. 69(1): 1–26. doi:10.1016/S0301-0082(03)00004-2.
PMID:12637170.
Bee, M.A., Perrill, S.A., and Owen, P.C. 1999. Size assessment in
simulated territorial encounters between male green frogs (Rana
clamitans). Behav. Ecol. Sociobiol. 45(3-4): 177–184. doi:10.
1007/s002650050551.
Bradbury, J.W., and Vehrencamp, S.L. 1998. Principles of animal
communication. Sinauer Associates, Inc., Sunderland, Mass.
Brantley, R.K., and Bass, A.H. 1994. Alternative male spawning
tactics and acoustic signals in the plainfin midshipman fish Por-
ichthys notatus Girard (Teleostei, Batrachoididae). Ethology, 96:
213–232.
Brantley, R.K., Tseng, J., and Bass, A.H. 1993a. The ontogeny of
inter- and intrasexual vocal muscle dimorphisms in a sound-
producing fish. Brain Behav. Evol. 42(6): 336–349. doi:10.1159/
000114170. PMID:8275300.
Brantley, R.K., Marchaterre, M.A., and Bass, A.H. 1993b. Andro-
gen effects on vocal muscle structure in a teleost fish with inter-
and intra-sexual dimorphism. J. Morphol. 216(3): 305–318.
doi:10.1002/jmor.1052160306. PMID:8315650.
Castellano, S., Rosso, A., Laoretti, F., Doglio, S., and Giacoma, C.
2000. Call intensity and female preferences in the European
green toad. Ethology, 106(12): 1129–1141. doi:10.1046/j.1439-
0310.2000.00639.x.
Chellappa, S., Huntingford, F.A., Strang, R.H.C., and Thomson,
R.Y. 1995. Condition factor and hepatosomatic index as esti-
mates of energy status in male three-spined stickleback. J. Fish
Biol. 47(5): 775–787. doi:10.1111/j.1095-8649.1995.tb06002.x.
Clutton-Brock, T.H., and Albon, S.D. 1979. The roaring of red deer
and the evolution of honest advertisement. Behaviour, 69(3):
145–170. doi:10.1163/156853979X00449.
Cohen, M.J., and Winn, H.E. 1967. Electrophysiological observa-
tions on hearing and sound production in the fish, Porichthys
notatus. J. Exp. Zool. 165(3): 355–369. doi:10.1002/jez.
1401650305. PMID:6076901.
Connaughton, M.A., Taylor, M.H., and Fine, M.L. 2000. Effects of
fish size and temperature on weakfish disturbance calls: implica-
tions for the mechanism of sound generation. J. Exp. Biol.
203(9): 1503–1512. PMID:10751166.
Davies, N.B., and Halliday, T.R. 1978. Deep croaks and fighting
assessment in toads Bufo bufo. Nature (London), 274(5672):
683–685. doi:10.1038/274683a0.
dos Santos, M.E., Modesto, T., Matos, R.J., Grober, M.S., Oliveira,
R.F., and Cana
´
rio, A. 2000. Sound production by the Lusitanian
toadfish, Halobatrachus didactylus. Bioacoustics, 10: 309–321.
Engen, F., and Folstad, I. 1999. Cod courtship song: a song at the
expense of dance? Can. J. Zool. 77(4): 542–550. doi:10.1139/
cjz-77-4-542.
Fine, M.L., and Pennypacker, K.R. 1986. Hormonal basis for sex-
ual dimorphism of the sound-producing apparatus of the oyster
toadfish. Exp. Neurol. 92(2): 289–298. doi:10.1016/0014-
4886(86)90081-6. PMID:3956662.
Fine, M.L., Economos, D., Radtke, R., and McClung, J.R. 1984.
Ontogeny and sexual dimorphism of the sonic motor nucleus in
the oyster toadfish. J. Comp. Neurol. 225(1): 105–110. doi:10.
1002/cne.902250111. PMID:6725634.
Fine, M.L., Burns, N.M., and Harris, T.M. 1990. Ontogeny and
sexual dimorphism of sonic muscle in the oyster toadfish. Can.
J. Zool. 68(7): 1374–1381. doi:10.1139/z90-205.
Fine, M.L., Malloy, K.L., King, C.B., Mitchell, S.L., and Cameron,
T.M. 2001. Movement and sound generation by the toadfish
swimbladder. J. Comp. Physiol. A Neuroethol. Sens. Neural
Behav. Physiol., 187(5): 371–379. doi:10.1007/s003590100209.
PMID:11529481.
Grafen, A. 1990. Sexual selection unhandicapped by the Fisher
process. J. Theor. Biol. 144(4): 473–516. doi:10.1016/S0022-
5193(05)80087-6. PMID:2402152.
Hasselquist, D.S., Bensch, S., and von Schantz, T. 1996. Correla-
tion between male song repertoire, extra-pair paternity and off-
spring survival in the great reed warbler. Nature (London),
381(6579): 229–232. doi:10.1038/381229a0.
Knapp, R., Wingfield, J.C., and Bass, A.H. 1999. Steroid hormones
and paternal care in the plainfin midshipman fish (Porichthys
notatus). Horm. Behav. 35 (1): 81–89. doi:10.1006/hbeh.1998.
1499. PMID:10049606.
Ladich, F. 2004. Sound production and acoustic communication. In
The senses of fish: adaptations for the reception of natural sti-
muli. Edited by G. von der Emde, J. Mogdans, and B.G. Kapoor.
Narosa Publishing House, New Delhi, India. pp. 210–230.
Ladich, F., and Fine, M.L. 2006. Sound-generating mechanisms in
fishes: a unique diversity in vertebrates. In Communication in
fishes. Edited by F. Ladich, S.P. Collin, P. Moller, and B.G. Ka-
poor. Science Publishers, Enfield, N.H. pp. 1–43.
Mager, J.N., III, Walcott, C., and Piper, W.H. 2007. Male common
loons, Gavia immer, communicate body mass and condition
through dominant frequencies of territorial yodels. Anim. Behav.
73(4): 683–690. doi:10.1016/j.anbehav.2006.10.009.
Ma
´
rquez, R., Bosch, J., and Eekhout, X. 2008. Intensity of female
preference quantified through playback setpoints: call frequency
versus call rate in midwife toads. Anim. Behav. 75(1): 159–166.
doi:10.1016/j.anbehav.2007.05.003.
724
Can. J. Zool. Vol. 87, 2009
Published by NRC Research Press

Maynard Smith, J., and Harper, D. 2003. Animal signals. Oxford
University Press, Oxford.
Mitchell, S., Poland, J., and Fine, M.L. 2008. Does muscle fatigue
limit advertisement calling in the oyster toadfish Opsanus tau?
Anim. Behav. 76(3): 1011–1016. doi:10.1016/j.anbehav.2008.
03.024.
Modesto, T., and Cana
´
rio, A.V.M. 2003a. Hormonal control of
swimbladder sonic muscle dimorphism in the Lusitanian toad-
fish Halobatrachus didactylus. J. Exp. Biol. 206(19): 3467–
3477. doi:10.1242/jeb.00581. PMID:12939377.
Modesto, T., and Cana
´
rio, A.V.M. 2003b. Morphometric changes
and sex steroid levels during the annual reproductive cycle of
the Lusitanian toadfish, Halobatrachus didactylus. Gen. Comp.
Endocrinol. 131(3): 220–231. doi:10.1016/S0016-6480(03)
00027-3. PMID:12714003.
Myrberg, A., Ha, S., and Shamblott, M. 1993. The sounds of bico-
lor damselfish (Pomacentrus partitus): predictors of body size
and a spectral basis for individual recognition and assessment.
J. Acoust. Soc. Am. 94(6): 3067–3070. doi:10.1121/1.407267.
Oliveira, R.F., and Almada, V.C. 1995. Sexual dimorphism and al-
lometry of external morphology in Oreochromis mossambicus.
J. Fish Biol. 46: 1055–1064.
Prestwich, K.N. 1994. Energy and constraints to acoustic communi-
cation in insects and anurans. Am. Zool. 94: 625–643.
Remage-Healey, L., and Bass, A.H. 2005. Rapid elevations in both
steroid hormones and vocal signaling during playback challenge:
a field experiment in Gulf toadfish. Horm. Behav. 47(3): 297–
305. doi:10.1016/j.yhbeh.2004.11.017. PMID:15708758.
Rowe, S., and Hutchings, J.A. 2004. The function of sound produc-
tion by Atlantic cod as inferred from patterns of variation in
drumming muscle mass. Can. J. Zool. 82(9): 1391–1398.
doi:10.1139/z04-119.
Taigen, T.L., and Wells, K.D. 1985. Energetics of vocalization by
an anuran amphibian (Hyla versicolor). J. Comp. Physiol. B
Biochem. Syst. Environ. Physiol. 155: 163–170.
Vasconcelos, R.O., and Ladich, F. 2008. Development of vocaliza-
tion, auditory sensitivity and acoustic communication in the Lu-
sitanian toadfish Halobatrachus didactylus. J. Exp. Biol. 211(4):
502–509. doi:10.1242/jeb.008474. PMID:18245626.
Zahavi, A. 1975. Mate selection — a selection for a handicap. J.
Theor. Biol. 53(1): 205–214. doi:10.1016/0022-5193(75)90111-
3. PMID:1195756.
Amorim et al. 725
Published by NRC Research Press