Among the diverse anuran behavioural repertoire, intraspecific communication is the most studied aspect, specifically during the breeding season (Wells, 1977; 2007). During this activity, acoustic communication can be used for territorial defence, attraction of reproductive partners, and predator avoidance (Wilczynski and Chu, 2001; Wells, 2007; Köhler et al., 2017). Communication is therefore vital for survivorship and directly related to individual fitness. However, acoustic communication is not used by mute species. Mute anurans are those that do not have advertisement calls (Wells, 1977; Emerson and Inger, 1992). For these species, it is likely that alternative forms of communication, such as visual, tactile and chemical, are used in reproductive and aggressive intraspecific interactions (e.g., Hödl and Amézquita, 2001; Lee and Waldman, 2002; Barros and Feio, 2011; Lipinski, Caldart and Iop, 2012).

Visual signalling may represent a complementary communication channel for species that breed in sites with high background noise (Hödl and Amézquita, 2001), since this environmental pressure could be related to the loss of advertisement call (Emerson and Inger, 1992). Indeed, visual communication is observed in unrelated species that breeds in environments with high background abiotic noise, such as Atelopus limosus (Bufonidae), Crossodactylus gaudichaudii (Hylodidae), Litoria nannotis (Hylidae), Staurois parvus (Ranidae), and Taudactylus eungellensis (Myobatrachidae) (Hödl and Amézquita, 2001).

The family Hylodidae includes three genera and 47 species (Frost, 2017) that inhabit noisy streams in Brazilian Atlantic Forest (Haddad et al., 2013). The species of the genera Crossodactylus and Hylodes, except for H. vanzolinii, present acoustic communication. Several cases of visual or bimodal (acoustic and visual communication concomitantly) communication, in reproductive or agonistic contexts, have also been reported for both genera (Weygoldt and Carvalho-e-Silva, 1992; Haddad and Giaretta, 1999; Hödl and Amézquita, 2001; Narvaes and Rodrigues, 2005; Caldart, Iop and Cechin, 2014; de Sá, Zina and Haddad, 2016). In addition, philopatry is commonly observed in hylodid species, generally related to territoriality (Haddad and Giaretta, 1999; Nascimento, Pombal Jr. and Haddad, 2001; de Sá et al., 2015). On the other hand, all species of the genus Megaelosia are considered mute, since there is no information on the emission of advertisement calls (Lutz, 1930; Weygoldt and Carvalho-e-Silva, 1992; Giaretta, Bokermann and Haddad, 1993; Pombal Jr., Prado and Canedo, 2003) and there is a species, M. goeldii, which does not have vocal slits or vocal sacs (Giaretta, Bokermann and Haddad, 1993). To date, there have been no records of alternative communication strategies in Megaelosia species. Therefore, we investigated a species of the genus Megaelosia, M. apuana, to evaluate whether alternative (i.e., visual) communication is used by males to compensate for the lack of vocalization.

We conducted the study at Caparaó National Park (PNC), a conservation unit located in the states of Minas Gerais and Espírito Santo, southeastern Brazil. Sampling occurred at Caparaó river, Vale-Verde, municipality of Alto Caparaó, state of Minas Gerais (20°25′10″ S, 41°50′47″ W; about 1200 m a.s.l.). Fieldwork occurred between October 2015 and January 2017 and consisted of seven expeditions with a mean interval of two months between visits. Sampling occurred more frequently in the summer (period of the species highest activity). In total we conducted 56 days of sampling and 448 person-hours of active search (table S1). According to Pombal, Prado and Canedo (2003), M. apuana has strictly twilight/nocturnal activity. However, diurnal observations were also employed to evaluate whether the species exhibit daytime activity. The study was approved by Ethics Committee of the University of Campinas (CEUA/Unicamp#4396-1) and by Instituto Chico Mendes de Conservação da Biodiversidade (ICMBio#52352-2).

We filmed focal individuals of M. apuana with a GoPro Hero 3+ Black Edition and a Sony DCR-SR47 coupled with red light illumination. Cameras were positioned in front of focal individuals and set to record for two uninterrupted hours. To verify the lack of vocalization in the species, an autonomous acoustic recorder Song Meter SM3 Acoustic (Wildlife Acoustics) was installed in the river’s margin. Audio recordings were performed with a sampling rate of 24 kHz and 16-bit resolution. Audio files were screened in Raven Pro 1.4 (Cornell Lab of Ornithology, Bioacoustics Research Program). In total, 40 hours of video and 54 hours of audio were recorded. Files were deposited in the video collection of the Fonoteca Neotropical Jacques Vielliard, Zoology Museum “prof. Adão José Cardoso”, Unicamp, Campinas, São Paulo (Aggressive behaviour and visual signalling file catalogue number FNJV-V 402).

To explore whether M. apuana present site fidelity we walked the total extension of the study area (approximately 300 m along the Caparaó river) daily, visiting all the spots with the presence of an adult individual, and verifying if the same individual remains in the same microhabitat for consecutive days. Finally, two M. apuana juvenile males (SVL 6.10 and 6.47 cm) were handled in the field to determine the ability of the species to emit distress calls. The test consisted of holding the individuals by the hind limbs and shaking them, as described by Toledo and Haddad (2009).

We observed an aggressive interaction between two adult M. apuana males during a territorial dispute. These males were found at 2015 h on 14 Jan 2017 (18.6°C air temperature; 70% of relative humidity) in a lentic pool (∼1.5 m in diameter) in the Caparaó river margin. When our observation began, the males were already engaged in a physical combat, but immediately disengaged from each other after our initial approach. The larger individual (male L), moved to the pool margin, while the smaller male (approximately 80% of male L’s size; hereafter male S) hid under a rock in the pool margin. After 14 minutes, male S returned to the margin pool. Male L quickly approached the pool’s centre, approximately 50 cm from male S, and began displaying visual signals: male L expanded and retracted both vocal sacs concomitantly, for four times, with mean interval of 7.66 ± 2.56 s (5-10 s) between displays. Male L then approached male S and intensified the signalling, expanding and retracting the vocal sacs in a higher rate for six times, with a mean interval of 3.16 ± 0.41 s (3-4 s). The lateral vocal sacs have white dense pigmentation, contrasting with the predominant dark greyish green colour of the frog dorsum (fig. 1). It was not possible to observe whether male S also emitted visual signals in response. After signalling, male L approached male S and the physical combat begun. The males bit each other’s mouths (fig. 2) and positioned their front limbs on the opponent’s shoulders. They remained in this posture for 79 minutes. During this period male S remained motionless, while male L alternated between being motionless and engaging in fast lateral shakes. When they released one another, male S swam to the pool margin, while male L swam to the centre of the pool, climbed on a rock, and remained there until male S retreated.

No sound was emitted during the entire aggressive interaction, including the visual communication and the physical combat. We also did not record any type of vocalization for M. apuana in the recordings obtained either from the video cameras or from the autonomous recorders installed in river margins. Finally, no distress calls were emitted due to handling and shaking the two juvenile males.

View Full Size

Visual signalling of Megaelosia apuana, recorded at Caparaó National Park, municipality of Alto Caparaó, state of Minas Gerais, Brazil; air temperature: 21.8°C. Adult male with both vocal sacs retracted (A) and expanded (B).

Citation: Amphibia-Reptilia 39, 2 (2018) ; 10.1163/15685381-20181000

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Aggressive behaviour of Megaelosia apuana. Details of the mouths of both males interweaved.

Citation: Amphibia-Reptilia 39, 2 (2018) ; 10.1163/15685381-20181000

• Download Figure
• Download figure as PowerPoint slide

Megaelosia apuana adult males showed site fidelity, using the same territories of activity and rest. We observed 11 males using the same sites for consecutive days (five of them were reported using the same sites for 11 consecutive days).

Our study reinforces the suggestions that individuals of the genus Megaelosia are mute (Lutz, 1930; Giaretta, Bokermann and Haddad, 1993; Giaretta and Aguiar Jr., 1998). The only report of sound emission in this genus was based on three adult females of M. boticariana while being handled in the field (Giaretta and Aguiar Jr., 1998). Therefore, these calls are likely distress or release calls (Köhler et al., 2017), also observed in Incilius periglenes, Leiopelma archeyi and L. pakeka, others anurans that are mute (Jacobson and Vanderberg, 1991; Bell, 2010).

While it is possible that M. apuna make ultrasonic calls such as Huia cavitympanum and Odorrana tormota (Narins et al., 2004; Arch, Grafe and Narins, 2008), it is unlikely because they do not have the morphological specialization in the ears necessary to receive ultrasound (Augusto-Alves and Toledo, persn. observation).

Many diurnal anurans or those that breed in environments with high background noise also emit visual signals. Thus, it has been commonly suggested that high background noise and luminosity are related to the evolution of visual signalling (e.g., Haddad and Giaretta, 1999; Hödl and Amézquita, 2001; Narvaes and Rodrigues, 2005; Hirschmann and Hödl, 2006). On the other hand, visual communication among nocturnal anurans is also widely reported, but in a bimodal context. This is the first case reported of a nocturnal unimodal visual display. The throat display without sound associated is reported also for another hylodid species, Hylodes japi (de Sá et al., 2016), and a Tanzanian phrynobatrachid frog, Phrynobatrachus krefftii (Hirschmann and Hödl, 2006). Both are diurnal, inhabit streams, and are not mute. The visual display in H. japi is associated with territoriality, female attraction, and courtship (de Sá et al., 2016). In P. krefftii this behaviour is just associated with territoriality (Hirschmann and Hödl, 2006).

Anurans are the vertebrates with the highest visual capacity in low-light environments (Kelber, Yovanovich and Olsson, 2016), allowing nocturnal visual communication. Despite their outstanding nocturnal sight, nocturnal visual signalling would be enhanced by the exhibition of contrasting colouration. Thus, the dense white colouration in the lateral vocal sacs of M. apuana, contrasted with the dark dorsal pattern (fig. 1), may facilitate the observed nocturnal visual communication. Vocal sacs are used in visual signalling in some Hylodes species as well (Haddad and Giaretta, 1999; Hartmann et al., 2005; de Sá, Zina and Haddad, 2016). In those species, vocal sacs are transparent and reflect sunlight during the day (fig. S1).

Our observations of the same individuals using the very same spots in the river for several days, and also returning to that place shortly after being disturbed, serves as evidence of philopatry (site fidelity) in M. apuana. We hypothesize the use of visual signalling in males of M. apuana as an advertising signal for territory defence against intruding conspecific males (see also Haddad and Giaretta, 1999; Nascimento, Pombal Jr. and Haddad 2001; Caldart, Iop and Cechin, 2014; de Sá, Zina and Haddad, 2016). These territories provide the resident male profitable feeding area (Alves and Toledo, 2017) and, probably, retreat and egg laying sites.

Concluding, we provided further information corroborating that M. apuana is mute and described for the first-time visual signalling and aggressive behaviour for the genus Megaelosia. Furthermore, we believe that additional field observations are necessary to test if throat display in the genus Megaelosia is also used for female attraction.