RESEARCH NOTE The sound of a snail: two cases of acoustic defence in gastropods

Sound production has evolved independently in many phyla over time (Senter, 2008). Many animals produce sounds for communication, either with congeners or in reaction to predators (for examples of the latter in invertebrates see Bura, Fleming & Yack, 2009; Olofsson, Jakobsson & Wiklund, 2012), but Vermeij (2010) noted that deliberate production of acoustic signals is entirely unknown in the phylum Mollusca. However, Braun (1887) was the first to report briefly about the soundproducing snail Cantareus apertus (Born, 1778). He wrote “Es ist allgemein bekannt [sic], dass namentlich die grösseren Helices dadurch erzeugen, dass die in Lufthöhle angesammelte Luft aus dem Athemloch herausgestossen wird; gewöhnlich wird eine geringe Menge Schleim in kleine Blasen dabei aufgetrieben, deren Platzen das Geräusch vermehrt” [It is well known that especially the larger helices generate noise as the air in the pulmonary cavity is expelled through the pneumostome; usually a small amount of mucus is extruded and the bursting of small bubbles increases the sound]. This species was also mentioned by Caziot (1914), together with references to sounds produced during locomotion by other helicids and some limacid slugs (Vlès, 1908, 1909; Jousseaume, 1909). However, C. apertus makes the noise when resting, its shell rocking back and forth (see video by Wenger, 2014); the sounds produced by this species are made in the context of antipredator defence behaviour, after the shell has been touched suddenly. In terrestrial gastropods retraction into the shell is an important defence mechanism, usually accomplished while the aperture is facing down on the locomotion surface. In addition to this passive mode of defence, secretions from the skin and mucus glands can deter predatory attacks, due to the distasteful and deterrent compounds they may contain (reviewed for marine molluscs by Derby, 2007; see Pakarinen, 1994, and Mair & Port, 2002, for terrestrial examples). The topic of sound production in (land) snails seems to have been largely neglected since the paper by Caziot (1914), except for the mention by Fechter & Falkner (1990: 244–245) of the common name ‘Grunzschnecke’ (grunting snail) for C. apertus and the remark by Vermeij (2010). Recently, G. Woehl and a colleague, while collecting amphibians at night in Brazil, recorded the sound of a snail by serendipity. The snail was found in leaf litter in Araucaria forest at Itaiópolis, Santa Catarina State, on 16 November 2013, where it had been attacked by a predator (presumably a mammal) shortly before. When they found the snail, it also secreted an orange mucus, possibly as a defensive mechanism. According to Woehl (personal communication) “the snail was moving and foaming”. In other words, it was alternately retracting after the shell was touched and then emerging (cf. movements shown by Wenger, 2014) and producing mucus. Moreover, the snail repeatedly emitted a sound when touched, of which one instance was recorded with a Sony voice recorder ICD-PX312. The original file was analysed using the audio software Ocenaudio v. 2 (OcenAudio Team); the extracted file of the snail sound is available in MP3-format in the Supplementary Material. The analysis revealed that the sound lasted 241 ms and consisted of two different pulses of, respectively, 52 and 58 ms, with a 131 ms interval between (Fig. 1). In the spectral view it can be seen that both pulses have a regular harmonic structure, with emphasis on the range 1500–10,000 Hz (Fig. 2). Further analysis suggests a fundamental at 1650 Hz and three harmonics at, respectively, 3300, 4970 and 6630 Hz in the first pulse (Fig. 3); the fundamental is a relatively high tone, comparable with G6. In the second pulse these tones are similar, although slightly higher (ca. 50–100 Hz). It should be noted that the tones are not even; the first pulse is slightly curved and in the second one there is a very brief compression after which the tones fade away (Fig. 2); in both pulses the minimum–maximum difference in the fundamental is ca. 400 Hz. Whether this composite sound was produced by means of compression of the pulmonary cavity or by other means remains unknown. A spectral analysis gives no clues about the underlying mechanical origin of the sound signal (Elemans, Muller & Heeck, 2008). It can be assumed, however, that the sound has to be interpreted either as a stress signal or a signal to a predator. Also of interest is the question of the taxonomic identity of the snail. Unfortunately, the snail was only photographed (Fig. 4) and not collected. Given the size (ca. 50 mm) and shape, there are only a few candidates in the known Brazilian malacofauna; these belong to the families Megalobulimidae and Amphibulimidae (both Stylommatophora). According to I. Agudo (personal communication) this is none of the localMegalobulimus species; he suggested instead a Plekocheilus species. This genus, belonging to the Amphibulimidae (sensu Breure & Romero, 2012), is represented by six species according to Simone (2006: 149–150). In addition the species classified by this author (Simone, 2006: 147) as Dryptus rhodocheilus (Reeve, 1848) belongs to Plekocheilus (Breure, 1979: 32; Breure & Ablett, 2011: 34). After carefully inspecting the photographs (Fig. 4), I tentatively identify the snail as Plekocheilus aff. rhodocheilus. This species was described from ‘Brazil’, without further specified location and has never been recorded since. The lectotype specimen is in the Natural History Museum, London (Breure & Ablett, 2011: fig. 21E–H). It may be noted that the type shell is damaged on the last whorl and at