Morphology, Feeding Rate and Larval Settlement Preference of the Corallivorous Nudibranch Phestilla subodiosa (Nudibranchia: Trinchesiidae) from Hong Kong

© 2022 Academia Sinica, Taiwan Open Access Morphology, Feeding Rate and Larval Settlement Preference of the Corallivorous Nudibranch Phestilla subodiosa (Nudibranchia: Trinchesiidae) from Hong Kong Sam King Fung Yiu1 and Jian-Wen Qiu1,* 1Department of Biology and Southern Marine Science and Engineering Guangdong Laboratory (Guangzhou), Hong Kong Baptist University, Hong Kong, China. *Correspondence: E-mail: [email protected] (Qiu). E-mail: [email protected] (Yiu) Received 3 May 2022 / Accepted 11 August 2022 / Published 25 November 2022 Communicated by Benny K.K. Chan We studied the morphology, host specificity, feeding rate and larval settlement preference of the corallivorous nudibranch Phestilla subodiosa collected from the field. These specimens collected from the scleractinian coral Monipora peltiformis in Hong Kong waters are morphologically different from the holotype and paratypes collected from an aquarium culture of Montipora spp., as they have diamondshaped swollen bulbs, and brown spots on cerata, as well as bulbs and the body immediately posterior to cerata. In experiments where P. subodiosa individuals were placed on the surface of several species of common scleractinian corals collected from Hong Kong waters, the nudibranchs were found to feed on M. peltformis at a rate of 0.05 cm2 individual-1 d-1; however, they were killed and eaten by other tested coral species (Pavnoa decussata, Porites lutea and Duncanopsammia peltata). When cultured in seawater conditioned with M. peltiformis, the veliger larvae required six days to become competent for settlement, and at day 9 could reach a maximum metamorphic rate of 31.1%. At competence, the veliger larvae could be induced to settle, indicating the presence of a larval settlement cue released by the host coral. Other coral species or their conditioned seawater did not induce settlement of the P. subodiosa larvae. Overall, our study expands the distribution record of P. subodiosa, adds this species to the list of corallivorous nudibranchs in Hong Kong waters, provides morphological features that were not included in the original description of this species, reveals the host specificity, and provides the feeding rate of this species. These results contribute to a better understanding of the diversity and potential impact of corallivorous nudibranchs in coral ecosystems. Key words: Coral health, Corallivory, Montipora, Phestilla, Predation, Subtropical Reef. BACKGROUND Phestilla Bergh, 1874 is a small group of nudibranchs that feed on scleractinian corals (Mehrotra et al. 2020) except P. chaetopterana, which lives inside the tube of an annelid (Ekimova et al. 2017). Absence of a cnidosac at the cerata tip is the key character that distinguishes Phestilla from other genera of Trinchesiidae (Rudman 1981). Currently there are 11 valid species of Phestilla (MolluscaBase 2022) including P. chaetopterana (Ekimova et al., 2017), P. fuscostriata Hu et al., 2020, P. goniophaga Hu et al., 2020, P. lugubris (Bergh, 1870), P. melanobrachia Bergh, 1874, P. minor Rudman, 1981, P. panamica Rudman, 1982, P. poritophages (Rudman, 1979), P. sibogae Bergh, 1905, P. subodiosa Wang et al., 2020, and P. viei, Mehrotra et Citation: Yiu SKF, Qiu JW. 2022. Morphology, feeding rate and larval settlement preference of the corallivorous nudibranch Phestilla subodiosa (Nudibranchia: Trinchesiidae) from Hong Kong. Zool Stud 61:59. doi:10.6620/ZS.2022.61-59. Zoological Studies 61:59 (2022) doi:10.6620/ZS.2022.61-59 1 © 2022 Academia Sinica, Taiwan al., 2020. Members of Phestilla have been widely found in the tropical and subtropical waters of the Indo-Pacific region (Faucci et al. 2007; Goodheart et al. 2017; Fritts-Penniman et al. 2020). Previous studies have reported the corallivory of Phestilla on seven genera of scleractinian corals: Dendrophyllia, Gardineroseris, Goniopora, Montipora, Pavona, Porites and Tubastraea (Gosliner et al. 2018). Four Phestilla spp. have been recorded in Hong Kong, with P. lugubris feeding on Porites spp., P. melanobrachia feeding on members of Dendrophylliidae, P. fuscostriata feeding on Pavona deccussata, and P. goniophaga feeding on Goniopora spp. (Morton and Morton 1983; Hu et al. 2020a b; Yiu et al. 2021). While P. melanobrachia and P. lugubris have been widely reported from the tropical Pacific (Harris 1968; Faucci et al. 2007), the other two species have been reported from only a few localities: P. goniophaga from Guam as Phestilla sp. 2 (RitsonWilliams et al. 2007 2009), and P. fuscostriata from Singapore (Chew 2021) and Indonesia (Ritson-Williams, personal communication). Previous studies on host specificity (RitsonWilliams et al. 2003 2007 2009) revealed that several Phestilla species including P. melanobrachia, P. minor, P. sibogae, P. goniophaga and one undescribed Porites eating species (Phestilla sp. 1 in Ritson-Williams et al. 2003) favoured a particular coral species or genus. Veliger larvae of these nudibranchs would settle or terminate the swimming phase of their life cycle in response to chemical signals of the particular coral host (Ritson-Williams et al. 2003 2007 2009). After settlement, some larvae also require chemical cues to induce metamorphosis – the physiological and morphological transformation from a larva to a juvenile (Ritson-Williams et al. 2007 2009). Very little is known about the larval development and ecology of Phestilla, except for P. sibogae (Hadfield 1977; Hadfield and Paul 2001), whose settlement is initiated in response to a water-soluble cue released from the host coral Porites compressa (Hadfield and Scheuer 1985; Hadfield and Pennington 1990). Phestilla sibogae can detect the inductive cue in the water column and stop swimming and settle on the coral surface (Hadfield and Koehl 2004). Given the lack of data on corallivorous nudibranchs in Hong Kong’s fringing coral communities (Yeung et al. 2021), we proposed a project to study their diversity, host specificity and potential impact to scleractinian corals (Hu et al. 2020a b; Yiu et al. 2021). While implementing this project in January 2021, we found a tiny nudibranch species feeding on Montipora peltiformis colonies in Bluff Island, Hong Kong. This was the first record of a corallivorus nudibranch associated with Montipora in the field. Our specimens look somewhat different in the gross shape and coloration pattern from the original description of P. subodiosa—the only currently known nudibranch that feeds on Montipora spp. (Wang et al. 2020). Phestila subodiosa was described based on the holotype and ten paratypes that were collected from Montipora fragments from an aquarium shop, as well as a paratype collected from Koh Tao, Thailand. The aims of the present study are therefore to determine the species identity of the Phestilla specimens collected from Montipora in Hong Kong waters by morphological and molecular analyses, and to provide information on the biology and ecology of this nudibranch, especially on the substrate preference of the adults and their feeding rate on the coral hosts, the larval development pattern, and their selectivity for settlement. MATERIALS AND METHODS Sample collection Samples of Phestilla were collected from Montiopora peltiformis fragments at Bluff Island (22°19'30.0"N 114°21'14.8"E) at 5 m depth by SCUBA diving in January 2021. The nudibranch samples were preserved in 95% ethanol for molecular and morphological analyses. All specimens examined in this study were deposited into the Marine Biodiversity Collections of the South China Sea (SCSMBC), Chinese Academy of Sciences, Guangzhou. DNA extraction, sequencing and analysis Genomic DNA was extracted from two specimens (SCSMBC030983-84) using the CTAB method (Stewart and Via 1993). DNA quantity was measured, and purity determined using a NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific, US). DNA quality was checked by 1% agarose gel electrophoresis. The products were processed and submitted to Novogene (Beijing, China) for commercial sequencing on an Illumina Novaseq 6000 platform to produce 10 GB paired-end sequencing data with an average read length of 150 bp. The sequences were assembled de novo using CLC v7. BLAST+ v2.2.26 was used to extract the scaffolds that matched the mitochondrial cytochrome c oxidase subunit I (COI), mitochondrial 16S rRNA subunit (16S rRNA) and nuclear Histone H3 (H3) genes from a query file containing the corresponding gene sequences of P. melanobrachia downloaded from NCBI’s GenBank. The sequences of these three genes in 35 nudibranch species belonging to 12 genera of page 2 of 12Zoological Studies 61:59 (2022) © 2022 Academia Sinica, Taiwan Trinchesiidae were downloaded from GenBank (Table S4) for determination of the phylogenetic position of the collected specimens. Alignments of the three genes were conducted separately and trimmed manually to 548 bp for COI, 435 bp for 16S rRNA and 327 bp for H3 using MEGA 7. Sequences were concatenated using SequenceMatrix v.1.7.8 (Vaidya et al. 2011) and then imported to the website version of IQ-Tree (http://iqtree.cibiv.univie.ac.at/; Nguyen et al. 2015) for Maximum Likelihood tree reconstruction with 1,500 ultrafast bootstrap pseudoreplicates (Hoang et al. 2017). ModelTest (Kalyaanamoorthy et al. 2017) incorporated in IQ-Tree was applied for each partition of the concatenated sequences, which detected TVM+F+I+G4 as the best model for COI and 16S rRNA and TIM2e+G4 for H3 based on Bayesian Information Criterion. MrBayes v.3.2.7a (Ronquist and Huelsenbeck 2003) was used to perform the Bayesian Inference analysis with four Metropolis-coupled Markov Chain Monte Carlo applied to 10 million generations, sampled at every 1,000 generations with a 25% burnin. Since the model detected for the concatenated dataset by ModelTest was not available in MrBayes, it was substituted by GTR+I+G—the closest overparameterized model (Huelsenbeck and Rannala 2004). The phylogenetic trees were visualized and edited using FigTree v1.4.4. Pairwise p-distances for the respective COI, 16S rRNA and H3 genes were estimated using MEGA 7 separately using the bootstrap method with 10,000 pseudoreplicates for variance estimation. Rates among sites were gamma distributed with invariant sites (G+I) and the gamma parameter was set to four. Morphological analysis Specimens and their egg masses were photographed using an Olympus OM-D EM1markII with a M. Zuiko Digital ED 60mm f2.8 Macro lens. Morphological characteristics were examined under a Motic SMZ-171 stereomicroscope (Motic, China). Four specimens were dissected to extract the buccal mass and reproductive system. Cerata were also collected from the specimens and the inner contents were examined. The buccal masses were dissolved in 20% diluted bleach for 30 minutes at room temperature to remove connective tissues and muscles. The jaws and the radula were examined and photographed under a Motic BA210 compound microscope. The reproductive system was observed under the stereomicroscope and line-drawing was made to show the key structures. Natural history The egg masses laid on the underside fragments of M. peltiformis that were kept in a laboratory at Hong Kong Baptist University were taken and kept in artificial seawater at 30 psu to observe the early development. When the embryos developed into swimming veliger larvae, the transparent membrane enclosing the egg masses were gently broken using forceps to allow the larvae to escape. The larvae, collected using pipettes, were maintained in a 1-L plastic container with ~25% of the wall replaced with 50 µm mesh placed in a 10-L glass tank with filtered seawater and aeration. A 12.5 cm2 fragment of M. peltiformis was put in that beaker to encourage larval settlement. The water temperature was kept at 24 ± 0.5°C. The status of the larvae was monitored daily. Rate of nudibranch consumption of host coral A colony of M. peltiformis without P. subodiosa was cut into fragments of 5 cm × 2.5 cm. The coral fragments were kept in aquaria until the wounds healed. Four recovered coral fragments were used to determine the feeding rate of the nudibranch in separate 10 L aerated tanks. Twenty nudibranchs from infested M. peltiformis colonies were transferred to the surface of each of the test fragments. Photographs were taken from both sides of the fragments daily for four days. At the end of the experiment, the total consumed area of the fragments were measured using CPCe 4.0, and the feeding rates (i.e., area consumed per day) were calculated. Tests using non-host corals Three species of common corals in Hong Kong (Duncanopsammia peltata, Pavona decussata and Porites lutea) were used. For each species, a colony was cut into several fragments of a standard size (5 cm × 5 cm). One nudibranch was placed on the surface of one fragment of each of the three coral species in separate 10 L aerated tanks. In each species, there were four replicates. Observations on larval development and testing of larval settlement preference Membranes of newly laid egg masses were broken with needles, and the eggs were transferred into a small plastic tank (250 ml) with the wall replaced with 50 µm mesh. The beaker was placed in a bigger glass tank (10 L) filled with 33 psu artificial seawater and fitted with a recirculated filtration system. The development of the eggs was observed daily until they reached the veliger stage and hatched. Coral conditioned seawater was used to test page 3 of 12Zoological Studies 61:59 (2022) © 2022 Academia Sinica, Taiwan for the presence of chemical cues that are attractive for nudibranch larval metamorphosis. The host coral conditioned seawater was prepared by placing a 5 cm2 fragment of M. peltiformis into an aerated beaker containing 1 L artificial water with a salinity of 33 psu and kept at 24°C. After 72 h, the water was filtered through a 0.22 µm membrane and stored in a freezer at -20°C until use. Coral conditioned seawater prepared with a 5 cm2 fragment of Porites lutea (the same way M. peltiformis was prepared) was used as a non-host control. In addition, filtered artificial seawater without prior contact with coral was used as a seawater control. The coral conditioned water and control seawater were spiked with antibiotics (90 μg ml-1 penicillin G and 75 μg ml-1 streptomycin sulfate) to suppress bacterial growth (Hadfield and Scheuer 1985; Hadfield and Pennington 1990). Six-well plates (Thermo ScientificTM 145380, USA) were used to test for the presence of coral released cues. Six ml of host or non-host coral conditioned seawater, or control seawater was added into each well. 10 to 15 veliger larvae were then transferred into each well. Each of the three treatments contained three replicate wells. The well plates were incubated at 26°C, and the number of settled individuals was determined by counting the empty shells daily. A previous study indicated that the shell would detach after a Phestilla veliger larva metamorphose into a juvenile (Hadfield 1977). The metamorphic rate was expressed in percentage. The differences in metamorphic rate among the three treatments at different time points were compared using one-way analysis of variance (ANOVA) followed by the post hoc SNK tests. RESULTS Molecular Analysis Three gene sequences were obtained from each of the two specimens. Alignment and concatenation resulted in a dataset of 1419 bp (658 bp for COI, 433 bp for 16S rRNA and 328 bp for H3). Our phylogenetic analyses showed that the two specimens we collected from the field were P. subodiosa (Fig. 1). The p-distance among two specimens were identical, and exhibited very small genetic distances with other specimens of P. subodiosa, including two collected from Montipora spp. in an aquarium (0.4% for COI, 0% for 16S and 0% for H3) (Wang et al. 2020), and one from an unknown habitat in Jeju Island, South Korea (1.1% for COI, 0.3 % for 16S) (Cho et al. 2018). The clade that is most closely related to P. subodiosa consists of two species – P. viei and P. fuscostriata. The p-distance between P. subodiosa and these two species are much larger: 13.2–17.4% for COI, 12.0–12.8% for 16S and 4.1–4.5% for H3 (Tables S1–S3). Fig. 1. Phylogenetic trees constructed using concatenated COI/16S rRNA/H3 sequence. Left: Maximum Likelihood tree with bootstrap values > 50 shown in the node. Right: Bayesian Inference tree with posterior probability values > 0.7 shown in the node. page 4 of 12Zoological Studies 61:59 (2022) © 2022 Academia Sinica, Taiwan TAXONOMIC ACCOUNT Class Gastropoda Cuvier, 1795 Order Nudibranchia Cuvier, 1817 Family Trinchesiidae Nordsieck, 1972 Phestilla subodiosa Wang, Conti-Jerpe, Richards & Baker, 2020 (Figs. 2–4) Materials examined: SCSMBC030983-86, collected from M. peltiformis colonies at Bluff Island (22°19'30.0"N, 114°21'14.8"E), Eastern Hong Kong waters, at 2–6 m water depth. External morphology (Fig. 3): Mature live specimens are elongated, measuring up to 5 mm in length and 0.5 mm in width. Body is white, with patches of rusty pigments inside cerata and the tissue next to cerata. Both rhinophores and oral tentacles are short digitiform. One pair of light black eyes are located slightly posterior to the base of rhinophore. There are 5 to 6 rows of cerata, and each row comprise one to three pairs of cerata, with larger individuals having more pairs. Cerata of bigger individuals have swollen bulbs that are spherical. Cerata tip lacks a cnidosac. Internal morphology (Fig. 4): Inside the cerata and the body next to the cerata, there are many symbiotic dinoflagellate cells which give the rusty coloration of the nudibranch. Jaws are translucent and thin, around 0.3 mm in width in a 3 mm long individual. Radula are located inside the jaw, with a formula of 8 × 0.1.0. There are three to four primary denticles on each side of radula, and the denticles are of similar lengths. Reproductive system consists of a large female gland mass, a small penial gland, a vas deferens bridging penile gland and female gland, and a tubular ampulla on the opposite side of the penial gland. Egg mass and veliger larvae: Egg masses are oval shaped, roughly 1 mm × 0.5 mm in size, with a translucent membrane enclosing around 20 eggs. Each egg is light yellow in colour, roughly 0.2 mm in diameter (Fig. 5A). At 24°C, eggs developed into Fig. 2. Selected photographs showing Phestilla subodiosa in the field. A, several adult individuals of Phestilla subodiosa feeding on a fragment of Montipora peltiformis. White arrows indicate the nudibranchs, yellow arrows indicate their egg masses. B, One juvenile and one adult individuals of Phestilla subodiosa feeding on a fragment of Montipora peltiformis. White arrows indicate the nudibranchs, yellow arrows indicate their egg masses. C, A juvenile individuals of Phestilla subodiosa feeding on a fragment of Montipora peltiformis. White arrows indicate the nudibranchs. Scale bars: A– C = 1 mm. page 5 of 12Zoological Studies 61:59 (2022) © 2022 Academia Sinica, Taiwan embryos (Fig. 5B) and then veliger larvae (Fig. 5C) in 2 days without feeding. Veliger larvae have a pair of black eyes and a well-developed swimming velum. After 8–10 days, the veliger larvae lost the velum and shell, and metamorphosed into more elongated juveniles that do not yet have tentacle or cerata (Fig. 5D). Feeding rate on host coral: Twenty individuals of P. subodiosa consumed 0.57 ± 0.24 cm2, 1.48 ± 0.95 cm2, 2.70 ± 1.48 cm2 and 3.80 ± 2.03 cm2 of M. peltiformis fragment when measured after 24 h, 48 h, 72 h, and 96 h of exposure, respectively (Fig. 6). The mean feeding rate was 0.95 ± 0.51 cm2 per day. Non-host coral test: All nudibranchs transferred to the surfaces of the other three coral species were found dead within one hour, but the method of nudibranch killing appeared to differ. Pavona deccussta killed the nudibranch by extruding the mesenterial filaments entangling the prey (Fig. 7A). Porites lutea killed the prey using the tentacles outside the body, then slowing digested the prey after it became a slurry (Fig. 7B). Duncanopsammia peltata killed the prey after secreting mucus to trap the nudibranch (Fig. 7C, D). Fig. 3. Selected photographs showing Phestilla subodiosa. and its coral host. A, several individuals of Phestilla subodiosa. feeding on a fragment of Montipora peltiformis, with a clear feeding scar along the lower edge. B, Dorsal view of SCSMBC030984. C–D, Dorsal view and ventral view of SCSMBC030985, respectively; E, Dorsal view of SCSMBC030986. Scale bars: A = 5 mm; B–E = 1 mm. page 6 of 12Zoological Studies 61:59 (2022) © 2022 Academia Sinica, Taiwan Fig. 4. Internal morphology of Phestilla subodiosa. A, Close-up of a section of the dorsum and two cerata; note that there is no cnidosac at the tip. B, dinoflagellates inside the cerata and the body (indicated by white arrows indicated). C, Jaw. D, Radula. E, Drawing of reproductive system; fm: Female gland; am: ampulla; pg: penile gland; vd: vas deferens. Scale bars: A = 500 µm; B = 250 µm; C = 150 µm; D = 25 µm; E = 120 µm. Fig. 5. Early development in Phestilla subodiosa. A, Newly laid egg mass attached to denuded coral skeleton. B, Developing embryos in the egg mass. C, A veliger. D, a post-metamorphosis juvenile. Scale bars: A–B = 500 µm; C = 200 µm; D = 300 µm. page 7 of 12Zoological Studies 61:59 (2022) © 2022 Academia Sinica, Taiwan Larval metamorphosis: There were no or very few metamorphosed larvae in the first five days, and the metamorphic rates were not significantly different among the three treatments (Fig. 8, Tables S5–S6). Starting from day 6, there were significant differences among the three treatments, with the metamorphic rate in the M. peltiformis conditioned seawater being significantly higher than in the other two treatments. By the end of the experiment at day 10, the metamorphic rate in the M. peltiformis conditioned seawater reached 31.11%, compared to only 11.11% in the P. lutea conditioned seawater, and 6.06% in the control seawater. DISCUSSION Based on phylogenetic and morphological analyses, we provide evidence that the Phestilla samples we collected from M. peltiformis colonies from the field in Hong Kong belong to P. subiodiosa. The DNA sequences of our specimens are very closely related to the sequences from the type specimens of P. subiodiosa, and such differences should be considered as intraspecific. Another nudibranch collected from Jeju Island, South Korea (Cho et al. 2018) also has very small genetic distances with the types and our samples, which also indicates that they are conspecific. Although Montipora millepora has been reported from Jeju Fig. 6. Feeding rate of Phestilla subodiosa on Montipora peltiformis. Each datum represents mean ± SD of four replicates. Fig. 7. Photographs showing the individuals of Phestilla subodiosa were killed by non-host coral species. A, Pavona deccussata. B, Porites lutea. C– D: Duncanopsammia peltata. page 8 of 12Zoological Studies 61:59 (2022) © 2022 Academia Sinica, Taiwan (Sugihara et al. 2014), the coral host of this nudibranch remains unknown as this was not reported (Cho et al. 2018). Morphologically, our specimens are similar to most of the type specimens in being associated with Montipora and being small in body size. Nevertheless, the type specimens used for morphological description were collected from Montipora spp. from an aquarium whose origin could not be determined (Wang et al. 2020), while our specimens were collected in the field from M. peltiformis in Hong Kong. One of the paratypes was collected from Koh Tao, Thailand, but there is no specific morphological description of this specimen in Wang et al. (2020). Our fully grown specimens differed substantially from the described types in that the ceratal bulbs were diamond-shaped and there were brown spots on cerata, and their ceratal bulbs and body were immediately posterior to cerata. In contrast, the type specimens have spherical swollen ceratal bulbs, and light brownish speckles on cerata only (Wang et al. 2020). Moreover, the radula formula and number of denticles on each side of a tooth row are slightly different between our specimens and the types. Our feeding experiments provide the first consumption rate of P. subodiosa (~0.05 cm2 individual-1 day-1 = 3.2 polyps individual-1 day-1) on M. peltiformis. P. sibogae which can reach a maximum size of more than 30 mm has been reported to consume up to 6.4 cm2 Porites tissues per day (Haramaty 1991). Phestilla goniophaga (former name: P. sp. 2), which is similar in size with P. sibogae, also has a high consumption rate of up to 30 polyps per day when feeding on Porites (Ritson-Williams et al. 2003). Phestilla sp. 1, whose maximum length is 5 mm, has a low consumption rate of < 3 Porites polyps individual-1 day-1 (Ritson-Williams et al. 2003). Therefore, due to its small body size, P. subodiosa has a low consumption rate. This result indicates that P. subodiosa should not be able to cause substantial tissue loss in its host in the field. In fact, our field observation of M. peltiformis colonies did not find obvious wounds caused by nudibranchs, although we found occasional tissue damage of this coral by the corallivorous snails Drupella spp. Nevertheless, under laboratory conditions, P. subodiosa outbreaks have been observed, causing whole colony damage to M. peltiformis, which supports the hypothesis that the populations of such nudibranchs may be controlled by predators in the field (Gochfeld and Aeby 1997). Nevertheless, P. melanobrachia and P. goniophaga are much bigger than P. subodiosa and their feeding scars on their respective coral hosts are often easy to see in Hong Kong waters (Yiu et al. 2021; Hu et al. 2020b). Therefore the potential of a corallivorous nudibranch to cause substantial tissue damage to its coral hosts depends largely on its body size. Furthermore, our nonhost coral tests indicated that P. subodiosa has a high host specificity, and will not be able to survive on nonMonipora corals. This result is consistent with that of other studies showing high host specificity of Phestilla species (Ritson-Williams et al. 2003; Faucci et al. 2007). Our larval development experiments indicate that a chemical cue released from host coral M. peltiformis is an inducer to the larval metamorphosis of P. subodiosa. The larvae take 6 days to reach metamorphic competence. In other species of Phestilla, both shorter (3 days in P. minor 5 days in P. sibogae and P. goniophaga) and longer (> 12 days in P. melanobrachia) periods to reach the metamorphic competence have been reported (Ritson-Williams et al. 2003). Since the veliger larvae of Phestilla develop in the water column, the ability to remain competent to undergo metamorphosis after six days indicates that the veliger larvae of P. subodiosa can travel with the current to places tens of kilometres from its natal locality. Nevertheless, field populations of P. subodiosa have only been reported from Thailand (Wang et al. 2020), South Korea (Cho et al. 2018) and Hong Kong, and it is not clear how populations of this species are connected by ocean currents. Based on COI sequences showing substantial (i.e., 1.1%) divergence between the Hong Kong and Jeju specimens, we can conclude the there is no contemporary gene flow between these two populations. CONCLUSIONS The Phestilla specimens collected from Bluff Island were found to be P. subodiosa based on Fig. 8. The metamorphic rate of larvae response to chemical cue released by host and non-host coral species. Three replicates of each treatment were scored for cumulative metamorphosis. Error bars represent ± 1 SD. page 9 of 12Zoological Studies 61:59 (2022)