INTRODUCTION
Aglajidae is a diverse group of predominantly shallow-water cephalaspidean gastropods with an extensive degree of morphological and colour variation (Rudman, Reference Rudman1971, Reference Rudman1972a, Reference Rudmanb, Reference Rudmanc, Reference Rudman1974, Reference Rudman1978; Gosliner, Reference Gosliner1980, Reference Gosliner, Behrens and Valdés2008). The family contains seven recognized genera and approximately 80 species worldwide distributed in tropical, sub-tropical and temperate shores. They are mostly found in soft bottom habitats and alga tufts around rocky shores, coral reefs and seagrass meadows (Thompson, Reference Thompson1977; Martínez et al., Reference Martínez, Ballesteros, Ávila and Cimino1993; Nakano, Reference Nakano2004; Valdés et al., Reference Valdés, Behrens and DuPont2006; Gosliner et al., Reference Gosliner, Behrens and Valdés2008; Camacho-García et al., Reference Camacho-García, Ornelas-Gatdula, Gosliner and Valdés2013; Costello et al., Reference Costello, Bouchet, Boxshall, Fauchald, Gordon, Hoeksema, Poore, van Soest, Sto, Walter, Vanhoorne, Decock and Appeltans2013; Bouchet, Reference Bouchet2014; Malaquias, Reference Malaquias2014).
A recent phylogenetic hypothesis of the Aglajidae confirmed the monophyly of the genera Aglaja, Melanochlamys, Nakamigawaia, Navanax, Odontoglaja and Philinopsis but suggested the paraphyly of Chelidonura which branched in three sub-clades (Camacho-García et al., Reference Camacho-García, Ornelas-Gatdula, Gosliner and Valdés2013). The general morphology and anatomy of several species in these three subclades are well known and they all depict similar body plans (Rudman, Reference Rudman1974; Gosliner, Reference Gosliner1980; Yonow, Reference Yonow1992, Reference Yonow1994; Ornelas-Gatdula et al., Reference Ornelas-Gatdula, Dupont and Valdés2012). As a consequence, the taxonomic status of Chelidonura is presently not clear and thus, for the purpose of this research, we adopted its traditional definition (sensu Burn & Thompson, Reference Burn, Thompson, Beesley, Ross and Wells1998).
In addition to these seven genera, there has been some debate about the validity and inclusion in the family Aglajidae of three other lineages, namely Noalda, Pseudophiline and Spinoaglaja. The genus Spinoaglaja was proposed for western Atlantic species with a spine-like extension on the anterior part of the shell (Ortea et al., Reference Ortea, Moro and Espinosa2007), but Camacho-García et al. (Reference Camacho-García, Ornelas-Gatdula, Gosliner and Valdés2013) have regarded it as a synonym of Philinopsis; the latter authors did not consider Noalda as part of the Aglajidae, but this remains to be tested in a molecular phylogenetic framework. Pseudophiline has morphological similarities with Philine (Gosliner, Reference Gosliner1980; Kitao & Habe, Reference Kitao and Habe1982) and the genus was recently ascribed to the family Philinidae based on the presence of philinid-like gizzard plates and radula (Chaban, Reference Chaban and Lutaenko2011). More recently, the new genus Migaya was proposed by Ortea et al. (Reference Ortea, Caballer, Moro and Espinosa2014) for the western Atlantic species Aglaja felis, but the validity of this genus remains to be tested in a phylogenetic framework.
Herbivory is considered the plesiomorphic feeding condition in the Cephalaspidea (Jensen, Reference Jensen1994; Mikkelsen, Reference Mikkelsen1996; Göbbeler & Klussmann-Kolb, Reference Göbbeler and Klussmann-Kolb2009; Malaquias et al., Reference Malaquias, Berecibar and Reid2009), and carnivory was suggested by Malaquias et al. (Reference Malaquias, Mackenzie-Dodds, Bouchet, Gosliner and Reid2009) to have arisen independently two or three times in different lineages. The latter authors have hypothesized that dietary specialization played a major role in the adaptive radiation of Cephalaspidea gastropods and that relations between prey structure, habitat and anatomy were important in the diversification within each lineage, allowing the development of more specific predator–prey interactions.
Aglajids show several unique evolutionary traits with potential relevance for feeding strategies and diversification, such as the reduction and posterior internalization of the shell, simplification of the digestive system (e.g. loss of hard mastication structures like the radula and gastric plates in the large majority of species), development and thickening of the buccal bulb and gizzard, secretion of deterrent chemicals, and cephalization of sensorial organs (Rudman, Reference Rudman1972a, Reference Rudmanb, Reference Rudmanc, Reference Rudman1978; Gosliner, Reference Gosliner1980; Sleeper et al., Reference Sleeper, Paul and Fenical1980; Leonard & Lukowiak, Reference Leonard and Lukowiak1984; Wägele & Klussmann-Kolb, Reference Wägele and Klussmann-Kolb2005; Cruz-Rivera, Reference Cruz-Rivera2011).
Several sensory structures in gastropods (e.g. eyes, anterior lateral folds, Hancock's organs, cephalic bristles, labial palps) are known to aid in tracking mucous trails (Kohn, Reference Kohn, Saleuddin and Wilbur1983) and for example, Paine (Reference Paine1963, Reference Paine1965) and Leonard & Lukowiak (Reference Leonard and Lukowiak1984) have demonstrated that active predation in Navanax involves mucous trail and chemoreception. The few empirical data available on the crawling speed of aglajids suggests that they are among the fastest moving sea slugs (Turner, Reference Turner1978).
The size of the buccal bulb and its ability of eversion are also important features in feeding: in Aglaja, Melanochlamys, Navanax and Philinopsis, the buccal bulb occupies almost half of the body length; while in Chelidonura, Nakamigawaia and Odontoglaja it is reduced to one-fifth of the body (Rudman, Reference Rudman1971). Two different forms of the buccal bulb prevail in Philinopsis: the typical bulbous shape (e.g. P. taronga, P. orientalis) that is also present in Aglaja, Melanochlamys and Navanax; and a tubular variation (e.g. P. depicta, P. pilsbryi) (Rudman, Reference Rudman1971, Reference Rudman1972a, Reference Rudmanb, Reference Rudmanc, Reference Rudman1974, Reference Rudman1978) (Figure 1A–C). The buccal bulb in Aglaja and Navanax can evert completely, whereas this ability is absent in Chelidonura, Nakamigawaia, Melanochlamys and Philinopsis (Rudman, Reference Rudman1971, Reference Rudman1974; Gosliner, Reference Gosliner1980, Reference Gosliner, Harrison and Kohn1994).
The genus Navanax feed upon other cephalaspideans including conspecifics, sacoglosans, anaspideans, nudibranchs, caenogastropods, polychaetes, crustaceans and small fish (Paine, Reference Paine1963, Reference Paine1965; Blair & Seapy, Reference Blair and Seapy1972; Gosliner, Reference Gosliner1980; Leonard & Lukowiak, Reference Leonard and Lukowiak1984; Pennings, Reference Pennings1990; Pennings et al., Reference Pennings, Natisch and Paul2001; Korb, Reference Korb2003); Philinopsis and Melanochlamys feed upon cephalaspideans and polychaetes (Rudman, Reference Rudman1972a, Reference Rudmanb, Göbbeler & Klussmann-Kolb, Reference Göbbeler and Klussmann-Kolb2009); Chelidonura upon flatworms (Gosliner, Reference Gosliner1987, Reference Gosliner, Harrison and Kohn1994; Yonow, Reference Yonow1992; Mangubhai, Reference Mangubhai2007); while Odontoglaja feeds on polychaetes and bivalves (Rudman, Reference Rudman1978; Wägele & Klussmann-Kolb, Reference Wägele and Klussmann-Kolb2005; Lobo-da-Cunha et al., Reference Lobo-da-Cunha, Ferreira, Coelho and Calado2009). No data are available on the diet of Aglaja and Nakamigawaia.
In this study we provide the first assessment of the dietary habits of Aglajidae sea slugs based on a comprehensive review of the literature and examination of gut contents of specimens representing the generic diversity of the family. We discuss our findings in relation to the distinctive anatomical, ecological and behavioural adaptations of these slugs.
MATERIALS AND METHODS
Ninety-two specimens belonging to 32 species of Aglajidae were dissected and their gut contents removed and examined (Table 1). Buccal bulb, oesophagus, intestine, and digestive gland were extracted and opened and the contents spread in Petri dishes filled with 70% ethanol and identified to the lowest possible taxonomic level using stereo, compound and scanning electron microscopy (SEM). Food items were mounted on SEM metallic stubs and coated with gold-palladium. Macrophotography was also used when convenient (Figures 2–4). In addition to gut content analyses, we revised the literature for records of Aglajidae food preferences in the wild (Table 2).
Food items were classified in ‘sessile’ and ‘vagile’ according to their mobility capacities (Menge et al., Reference Menge, Berlow, Blanchette, Navarrete and Yamada1994; Wägele, Reference Wägele2004; Madden et al., Reference Madden, Goodin, Allee, Finkbeiner and Bamford2008) and an estimate of food preference (vagile vs sessile) was inferred based on the total diversity of food items recognized during this study and from literature records presented in Table 2.
RESULTS
Food items were found in the gut of 11 out of the 32 species studied and in 24 of the 92 specimens dissected, corresponding to 26% of the specimens analysed (Table 1): one specimen of Aglaja and Nakamigawaia, three specimens of Navanax, four specimens of Chelidonura, Philinopsis and Odontoglaja, and seven of Melanochlamys.
Based on literature records and our own results, 70 different food items were recognized belonging to 20 major taxonomic groups, with vagile organisms accounting for 94% (=66 food items) of the diet composition. Carnivory is confirmed as the only feeding strategy in Aglajidae. The sessile organisms recognized in the gut of aglajids were bivalves, foraminiferans and sponge spicules (Table 2).
Foraminiferans were the only food item found in the gut contents of Aglaja and Nakamigawaia, whereas Navanax yielded the most diverse assemblage of food items, namely sponges, cephalaspidean gastropods, nudibranch gastropods, sacoglossan gastropods, caenogastropods, annelids, crustaceans and fish.
Field observations showed that Chelidonura inornata feed upon conspecifics and Navanax inermis was observed sucking in juveniles of Aplysia sp. (A. Zamora, personal observation).
DISCUSSION
With the exception of the studies by Paine (Reference Paine1963, Reference Paine1965) on the diet of the genus Navanax, knowledge about dietary preferences of aglajid slugs is based on sparse records included in general works about the diversity or morphological aspects of the Aglajidae (e.g. Tchang-Si, 1934; Marcus, Reference Marcus1961; Marcus & Marcus, Reference Marcus and Marcus1966; Blair & Seapy, Reference Blair and Seapy1972; Rudman, Reference Rudman1972a, Reference Rudmanb, Reference Rudman1978; Pennings, Reference Pennings1990; Pennings et al., Reference Pennings, Natisch and Paul2001; Padilla et al., Reference Padilla, Carballo and Camacho2010; Gosliner, Reference Gosliner2011; Camacho-García et al., Reference Camacho-García, Ornelas-Gatdula, Gosliner and Valdés2013; see Table 2). This study is the first comprehensive account dedicated to understand the trophic interactions of Aglajidae slugs as a whole.
The rather low percentage of slugs found with food remains in the gut (26%) may be partly explained by the fact that some aglajids regurgitate the hard parts of prey items after digestion when those seem to be above a certain threshold size. This behaviour was documented by Rudman (Reference Rudman1971, Reference Rudman1972a) for the species Philinopsis speciosa, which he observed regurgitating several empty shells of the gastropod Bulla ampulla after 2–3 h of capture. Aglajids do not have a crushing gizzard with plates and in some cases large shells are likely too difficult to be carried along the digestive tract and end up, therefore, being regurgitated. However, Paine (Reference Paine1963) and Pennings (Reference Pennings1990) have demonstrated that in the large sized-body aglajid species Navanax inermis (average adult size c. 40 mm; Leonard & Lukowiak, Reference Leonard and Lukowiak1984), complete shells of small ‘prosobranchs’ and ‘opisthobranchs’ (e.g. Aplysia, Bulla) and hard-parts of sea slugs (e.g. radulae, jaws, shells) can be defecated unaltered.
A striking result of this research is the recognition that aglajids feed nearly exclusively upon vagile prey (94% of food items; Table 2). Motile organisms secrete mucus to aid in crawling or as a protective mechanism (e.g. opisthobranchs, nematodes, platyhelminths, annelids, gastropods) (Brusca & Brusca, Reference Brusca and Brusca2003; Hickman et al., Reference Hickman, Roberts and Larson1993), leaving behind mucous trails that can be located and traced by aglajids using their sensorial organs (Paine, Reference Paine1965; Kohn, Reference Kohn, Saleuddin and Wilbur1983; Davies & Blackwell, Reference Davies and Blackwell2007; Terrence et al., Reference Terrence, Saltin, Davies, Johannesson, Stafford and Williams2013).
The genera Navanax and Philinopsis include relatively large animals (adult size over 10 mm in most cases) with a massive buccal bulb that occupies about half of the body cavity (Rudman, Reference Rudman1972a, Reference Rudman1974) (Figure 1). These slugs are active crawlers and can feed upon larger prey (e.g. fish, bulloid gastropods, polychaetes, flatworms, crustaceans, ctenophores and other sea slugs; Table 3) by a rapid, partial or complete eversion of the buccal bulb. Paine (Reference Paine1963) documented cannibalism in Navanax (N. inermis) but only between animals of dissimilar size and when those attempted to mate. The diet of N. inermis is by far the best known among aglajids as a result of the long-term and dedicated studies by Paine (Reference Paine1963, Reference Paine1965; see Table 2). These studies seem to indicate that the genus Navanax is the most generalist among the Aglajidae, but of course this can be the result of the uneven amount of data collected for this genus when compared with the others. The presence of sponge spicules found by us in the gut of Navanax is likely the result of random ingestion.
On the other hand, Chelidonura, Melanochlamys and Odontoglaja species are on average smaller slugs (adult size less than 10 mm in most cases; exceptions are common in Chelidonura) that have comparatively a reduced and non-eversible buccal bulb (the latter is partially eversible in Odontoglaja) (Rudman, Reference Rudman1972b, Reference Rudman1974; Figure 1; Table 3). Chelidonura seems to have a preference for epifaunal organisms (e.g. flatworms, slugs, shelled gastropods), whereas Melanochlamys feed predominantly upon infaunal prey such as polychaetes, nemerteans, nematodes and kinorhynchs (Table 3).
Odontoglaja, the only confirmed genus with radula (Gosliner et al., Reference Gosliner, Behrens and Valdés2008; Figure 1; referred to a possible Chelidonura with a vestigial radula), which is well developed with strong bicuspid lateral teeth (Gosliner et al., Reference Gosliner, Behrens and Valdés2008; Figure 1), has apparently a preference for organisms with thicker dermis such as crustaceans, ?holothurians, and polychaetes (Hickman et al., Reference Hickman, Roberts and Larson1993; Table 3).
Aglaja and Nakamigawaia are genera for which nothing was previously known about their diet. Most species have an average adult size over 10 mm (Rudman, Reference Rudman1972c; Baba, Reference Baba1985), but based on our results it is not possible to ascertain where the sole presence of foraminiferans found in the gut reflects a dietary preference or results from accidental ingestion. Because of average size, anatomical configuration of the digestive tract, and crawling capacities of these snails (Rudman, Reference Rudman1972c; Baba, Reference Baba1985; Gosliner et al., Reference Gosliner, Behrens and Valdés2008; Figure 1; Tables 2 & 3), our expectation was to find a diet composed by several motile invertebrates.
Malaquias et al. (Reference Malaquias, Berecibar and Reid2009) mapped the diet of most lineages of cephaslaspids onto a molecular phylogeny of the group and have suggested that dietary specialization played a major role in the adaptive radiation of these gastropods. Our results reinforced the view that Aglajidae slugs are the only active hunter group of cephalaspids and the only one to be specialized on motile prey (Malaquias et al., Reference Malaquias, Berecibar and Reid2009; Göbbeler & Klussmann-Kolb, Reference Göbbeler and Klussmann-Kolb2009).
ACKNOWLEDGEMENTS
We are indebted to A. Cosgrove-Wilke (Western Australian Museum), E. Naranjo-García (Colección Nacional de Moluscos, UNAM-México), J. Chuk (Museum Victoria, Australia), N. Anthes (University of Tübingen), K. Jensen and O. Tendal (Zoologisk Museum, Copenhagen), T. Nickens (Smithsonian Institution, National Museum of Natural History), and M. Caballer (Muséum national d¹Histoire naturelle, Paris) for providing specimens for this study. We also thank J. L. Cervera (University of Cádiz), N. Budaeva (University Museum of Bergen, Norway) and P. Valentich-Scott (Santa Barbara Museum of Natural History) for helping with identification of some food items. We are grateful to E. Erichsen (University of Bergen) for his support with the electron microscopy sessions. M. Caballer made valuable comments on this paper.
FINANCIAL SUPPORT
This work was funded through a doctoral grant given to the first author by the Consejo Nacional de Ciencia y Tecnología (CONACYT-México), fellowship BAZS/188890/2010. Additionally, this research benefited from specimens gathered through visits of the second author to European natural history museums funded by the SYNTHESYS Project, http://www.synthesys.info/, which is financed by the European Community Research Infrastructure Action under the FP7 ‘Capacities’ Program.