Introduction

In recent decades much attention has been given to the symbiotic relationship between reef corals and their symbiotic algae (zooxanthellae), which became particularly apparent with the occurrence of coral bleaching (e.g. Hoeksema 1991a; Brown 1997; Sampayo et al. 2008; Suggett and Smith 2011; Hoeksema and Matthews 2011). Because of the increasing emphasis on reef corals as autotrophs, it almost seemed that their other role as heterotrophs (Goreau et al. 1971; Porter 1974, 1976; Bak et al. 1998; Houlbrèque and Ferrier-Pagès 2009; Tremblay et al. 2011) became less noticed.

Many observations regarding food intake by reef corals resulted from experiments that focused on their feeding mechanism (Boschma 1925; Sorokin 1981; Clayton and Lasker 1982; Sebens and Johnson 1991; Sebens et al. 1996, 1998; Coles 1997; Ferrier-Pages et al. 2003). In comparison, only a few studies focused on their specific prey, which predominantly consists of small demersal and planktonic animals like amphipods, copepods, nematodes, nemerteans, nereids, polychaetes, and jellyfish, as found in their gut contents (Boschma 1924; Porter 1974; Lewis and Price 1975; Johnson and Sebens 1993). Furthermore, it is assumed that prey is predominantly caught by corals that are active at night (Houlbrèque and Ferrier-Pagès 2009).

Monostomatous mushroom corals (Scleractinia: Fungiidae) are iconic for having large polyps with a single, large mouth. They have been used in various classic studies on feeding mechanisms (Duerden 1906; Boschma 1924, 1926; Yonge 1930; Abe 1938; Stephens 1962; Schuhmacher 1979). Polystomatous species are usually larger owing to their additional (secondary) mouths, which are either smaller or equal in size compared to the primary mouth (Hoeksema 1991b; Gittenberger et al. 2011).

During a recent biodiversity survey on the coral reefs of Kota Kinabalu, we observed several monostomatous and polystomatous mushroom corals preying on salps (Thaliacea: Salpida: Salpidae). To our knowledge, the feeding of corals on salps has been reported only once before, which was based on a single salp found in the gut contents of a colony of Montastraea cavernosa (Linnaeus, 1776) (see Porter 1974).

Materials and methods

A faunistic study of mushroom corals was performed in the period 16–28 July 2011 on the coral reefs of Kota Kinabalu, the capital of Sabah, Malaysia (5° 57'–6° 5'N, 115° 59'–116° 5'E). Thirty dives, each approximately 1 h in duration, were made using SCUBA. The roving diver technique was employed (see e.g., Hoeksema and Koh 2009), in which species incidence data were recorded at each reef over the whole depth range where corals occurred, from the reef flat to the reef base, but not deeper than 30 m. At 3–18 m depth, several mushroom corals had their mouths wide open. Closer examination revealed that these corals had caught transparent salps. We also encountered slow-swimming salps in the water column. An inventory was made of which recorded mushroom coral species appeared to prey on salps. An additional remark is given on a non-mushroom coral with a salp in its mouth.

Results

All but one of the observed salp-predating corals belong to the mushroom coral family Fungiidae (Hoeksema 1989; Gittenberger et al. 2011). Nine of the 34 recorded mushroom coral species were observed to prey on salps (Table 1). Specimens of Cycloseris costulata, C. fragilis, Danafungia scruposa, Fungia fungites, Pleuractis moluccensis, and P. paumotensis had transparent salps (ca. 2 cm) or their remnants stuck inside their wide-open mouths (Fig. 1a, d–g). An individual of Heliofungia actiniformis had a salp of ca. 6 cm captured by its long tentacles (Fig. 1h). Two salps that had landed on top of D. scruposa corals were transported by tentacles from the coral margin towards the opening mouth, which was slightly hindered by some wave action. The salps hardly moved by themselves and did not attempt to escape. Polystomatous corals of Halomitra pileus and Herpolitha limax had salps only in their largest mouths (Fig 1b, c). Apart from mushroom corals, the only other salp-consuming coral observed was a specimen of Euphyllia paraancora Veron, 1990 (Fig. 2). Some mushroom corals appeared to ingest their prey by showing barely visible salp remnants inside their wide-open mouth (Fig. 3).

Table 1 Records of mushroom coral species (n = 34) and those predating on salps present on Kota Kinabalu reefs indicated by number of sites (total 30)
Fig. 1
figure 1

Mushroom corals of various species feeding on transparent salps at Kota Kinabalu, Sabah: a Danafungia scruposa (one salp in mouth and one beside), b Herpolitha limax (two mouths sharing one salp), c Halomitra pileus, d Cycloseris costulata, e Pleuractis paumotensis, f P. moluccensis, g Fungia fungites, h Heliofungia actiniformis. Scale bars 1 cm

Fig. 2
figure 2

Specimen of Euphyllia paraancora with a captured salp at Kota Kinabalu, Sabah. Scale bar 1 cm

Fig. 3
figure 3

Specimen of Fungia fungites with a partly ingested salp at Kota Kinabalu, Sabah. Scale bar 1 cm

The salps most probably belong to the subfamily Salpinae (R.W.M. Van Soest, personal communication); for salp taxonomy and phylogeny, see Godeaux (1998), Van Soest (1998) and Govindarajan et al. (2011).

Discussion

Although it is known that many species of corals can be active heterotrophs, ingesting organisms ranging from bacteria to mesozooplankton, there is very little information on what animals are eaten by corals (Houlbrèque and Ferrier-Pagès 2009). It was recently discovered that individuals of the monostomatous fungiid Danafungia scruposa are able to prey on large jellyfish (diameter up to 12 cm) in the Red Sea (Alamaru et al. 2009). In an earlier anecdotal account based on an aquarium experiment, it was reported that the mushroom coral Heliofungia actiniformis is able to use its long tentacles to predate on 1.5 cm long damselfish (Sisson 1973). Because little is known about the diet of corals and other anthozoans (see e.g., Van der Meij and Reijnen 2011), it is important that field observations concerning this topic are reported.

It is also relevant to note that some commensal animals are able to live in between the tentacles of mushroom corals without being eaten, such as particular species of fish and shrimp (Bos 2011; Hoeksema and Fransen 2011; Hoeksema et al. 2011). It is unclear whether they are immune to the coral venom and therefore escape predation.

Until recently, gelatinous zooplankton, like salps, ctenophores and pelagic cnidarians, were considered ‘trophic dead ends’ in food webs, i.e. zooplanktivores that seemed to lack obvious top predators themselves (Mianzan et al. 2001). However, various animals are known to eat salps, such as sea lions (Childerhouse et al. 2001), albatrosses (James and Stahl 2000), turtles (Van Nierop and Den Hartog 1984; Hatase et al. 2002; Eckert 2006; Dodge et al. 2011), fish (Lyle and Smith 1997; Morato et al. 2000; Mianzan et al 2001), and krill (Kawaguchi and Takahashi 1996).

To our knowledge, the present report is the first record dealing with corals in the process of capturing and eating salps, although Caribbean corals of Agaricia agaricites (Linnaeus, 1758) have also been observed to ingest planktonic tunicates (R.P.M. Bak, personal communication). With regard to the different growth forms of mushroom corals, the present observations suggest that a large surface area may facilitate catching food, while big mouths enable feeding on large prey when available. Both traits are extra advantageous when combined, like in most polystomatous fungiids (Hoeksema 1991b; Gittenberger et al. 2011). In cases where mushroom corals form dense aggregations (e.g. Hoeksema 2004; Hoeksema and Matthews 2011), salps may not easily escape capture. However, if the aggregations consist of regenerated mushroom coral fragments (Hoeksema and Gittenberger 2010; Hoeksema and Waheed 2011), only a few of them possess large primary mouths that may be used to ingest large prey.

Although it is advantageous for corals to have a large mouth if large prey is available, it is not clear whether they are as efficient when small prey is more abundant than large prey. In this instance, many small mouths might be more ideal because a particular polyp (or mouth) size may indicate a specific size spectrum of prey (Tsounis et al. 2010). This is beneficial for various mushroom coral species that have secondary small mouths in addition to a large primary mouth (Hoeksema 1991b; Gittenberger et al. 2011). Prey behaviour and environmental factors may interfere with the capture success of corals regardless of their polyp size (Sebens et al. 1996; Palardy et al. 2005). The prey intake by some mushroom corals was only slightly delayed by minor wave action. Stronger water movement at shallow depths may increase the probability of transporting the large salps away from their predators.