Molecular Phylogeny and Morphological Distinctions of Two Popular Bivalves, Ctenoides scaber and Ctenoides mitis
One of the most well-known species in the bivalve family Limidae (d’Orbigny, 1846) is the brightly colored Ctenoides scaber (Born, 1778), commonly known as the rough file clam or flame scallop. Distinguishing this bivalve from its close relative, C. mitis (Lamarck, 1807), can be difficult using only morphological features and has led to much taxonomic confusion throughout the literature. In this study, morphological characters were compared to a molecular phylogeny constructed using three genes (COI, 28S, and H3) in order to differentiate C. scaber and C. mitis. The phylogeny recovered two well-supported clades that differ significantly in shell rib numbers, but not tentacle colors. The two species were then placed in a larger phylogenetic context of the Limidae family, which revealed the need for further systematic revision across genera. As these bivalves are popular in aquaria, cannot be tank-raised, and have been overcollected in the past, proper species identification is important for assessing sustainable collection practices.
The rapid increase of home aquaria owners (~1.5–2 billion people) and the lucrative global trade, valued at up to $330 million annually , have led to the overcollecting of many marine organisms [2, 3]. More than 500 species of marine invertebrates at numbers between 9 and 10 million are traded as ornamental species annually . Although there have been some advances in breeding ornamental marine species , the majority are still being collected from the wild. The impacts of ornamental fisheries on invertebrate species with regard to their reproductive age, growth rate, population density/distribution, and population connectivity are largely unknown .
One of the obvious challenges for evaluating the impact of ornamental fisheries is to correctly document the numbers of species collected from various localities , especially given that cryptic species in marine ecosystems are common , and taxonomy in many marine invertebrate groups awaits systematic revision . Currently, a large number of marine invertebrates have not been evaluated for IUCN (International Union for Conservation of Nature) status, despite their popularity in the aquarium trade.
One popular invertebrate group in the ornamental fisheries trade is the bivalve family Limidae, colloquially known as file shells or file clams. They are well-known to SCUBA divers  and shell collectors, and several charismatic species are routinely collected for the aquarium trade . Numerous studies have been conducted on their behavior [11–16], morphology [17–19], habitat [20–22], reproduction [23, 24], physiology , and vision [26–31]. However, systematics studies have only been done on a small number of species in this family [10, 19, 32–35]. It is also noted that conchological traits alone are not always sufficient to establish clear boundaries between taxa .
One of the most well-known species in Limidae is Ctenoides scaber (Born, 1778), which in the past has been protected with imposed daily bag limits in the state of Florida due to overcollection . It is known as the “flame scallop” due to its bright red coloration. According to a single retailer (Blue Zoo Aquatics, Hawthorne, CA, USA), around 10 C. scaber are sold each month (pers. comm.). Multiplying this average by the numerous retailers and wholesalers worldwide may warrant unsustainable collection levels. Furthermore, collection of these bivalves is difficult, as they inhabit rocky areas or crevices in corals  and substantial damage can be done to the local habitat when they are harvested. Aquaculture of the species is not possible because spawning cannot be induced in the laboratory, and larvae from wild-collected specimens did not survive the past 36 hours .
Documentation and identification of Ctenoides scaber in aquarium trade appear to be highly inconsistent and inaccurate. For example, the Aquariumtradedata.org database  uses shipment declarations and commercial invoices from the United States Fish and Wildlife Service to track aquarium trade data. The database listed 4,820 and 9,670 C. scaber being imported to the US in 2008 and 2011, respectively. But in 2009, only 22 individuals were documented. In addition, the documented countries of origin are the Philippines and Indonesia, even though the true C. scaber is a Caribbean species.
Compounding the unknown collection rates of Ctenoides scaber is the difficulty in differentiating it from the presumably sister species C. mitis. The classical distinction between the two species was that C. scaber had red pallial tentacles and C. mitis had white pallial tentacles . However, this was based only on populations from the Florida Keys, and variation in tentacle color outside the Florida Keys required further scrutiny . Mikkelsen and Bieler  found that the number of radial ribs (C. scaber 28–78, mean 55, and C. mitis 59–149, mean 89) differed between the two species, but they were not able to confirm any of the other previously cited differences, including shell size, shape, opacity, or distribution. In addition, several questions remain unanswered, including how much of tentacle color variations are due to phenotypic plasticity, and whether the two species are actually reproductively isolated.
Therefore, the goal of this study was to assess the phylogenetic status of Ctenoides scaber and C. mitis and to investigate their diagnostic morphological characteristics. Multigene molecular phylogenies were constructed for the two species as well as the Limidae family. Museum and aquaria specimens were examined to determine tentacle coloration and shell rib counts. Our results will aid proper identification of the two species for future collection management and provide a better understanding of the overall phylogeny of Limidae bivalves.
Museum specimens were loaned from the University of Florida Museum of Natural History (see Supplementary Table 1 for catalogue numbers). Species identifications were given based on museum labels (Ctenoides scaber: ; C. mitis: ). Additional “flame scallops” (presumably C. scaber, ) were purchased through Aqua Imports (Boulder, CO, USA), which were collected in Haiti () and the Florida Keys (). The outgroups Ctenoides ales () and Limaria sp. () were also purchased through Aqua Imports, which were collected in the Philippines. All store-bought specimens were deposited in the University of Colorado Museum of Natural History (Supplementary Table 1).
2.2. Molecular Data
Genomic DNA was extracted from the mantle edge (c. 25 mg) of each specimen using the E.N.Z.A. Mollusc DNA Kit (Qiagen) according to the manufacturer’s protocol. The mitochondrial gene COI, the nuclear ribosomal gene 28S, and the nuclear histone protein-encoding gene H3 were amplified for each specimen. Primers used to amplify the genes are listed in Table 1. The GoTaq® Hot Start Green Master Mix was used at concentrations following the manufacturers’ instructions. The Polymerase Chain Reaction (PCR) cycling parameters for COI were as follows: initial denaturation of 94°C for 1 min and then 35 cycles of 94°C for 30 s, 55°C for 30 s, and 72°C for 30 s, with a final extension of 72°C for 6 min. For 28S, an initial denaturation of 94°C for 4 min was used, followed by 36 cycles of 94°C for 40 s, 53°C for 40 s, and 72°C for 1 min and a final extension of 72°C for 10 min. For H3, an initial denaturation of 94°C for 1 min was used, followed by 35 cycles of 94°C for 30 s, 49°C for 30 s, and 72°C for 30 s with a final extension of 72°C for 6 min. PCR products were visualized using agarose gel electrophoresis (1% agarose). PCR purification and sequencing were conducted by Quintara Biosciences (Berkeley, CA, USA). All sequences were submitted to Genbank and their accession numbers are listed in Supplementary Table 1.
2.3. Phylogenetic Analysis
Sequences of Ctenoides scaber, C. mitis, C. ales, and Limaria sp. were assembled using CodonCode Aligner© software 7.0.1 (Centerville, MA, USA) and aligned using MUSCLE . Gene substitution models were selected using Partitionfinder 2  based on Akaike Information Criterion. The GTR+G model was selected for COI, the GTR+G model was selected for 28S, and the GTR model was selected for H3. Maximum-likelihood (ML) and Bayesian phylogenetic reconstructions were conducted for the three genes individually as well as the concatenated dataset. The ML analyses were conducted using the RAxML BlackBox online server with 100 bootstraps [40, 41]. Bayesian analyses were conducted using MrBayes . For each dataset, two runs were conducted simultaneously and trees were sampled every 10 iterations. Bayesian searches were run for 500,000 iterations for the COI, 28S, and H3 datasets. For the concatenated dataset, the analysis was run for 1,000,000 iterations. Convergence was ensured for each Bayesian run. For all datasets, a 50% majority rule consensus tree was generated after a 10% burnin and visualized in FigTree v1. 3.1  (Figure 1).
(a) COI + 28S +H3
To assess the phylogenetic status of Ctenoides scaber and C. mitis in a larger comparative phylogenetic framework, Genbank sequences of the same three genes from all other available Limidae bivalves were downloaded (Supplementary Table 1). Sequences from two representative C. scaber and C. mitis individuals (based on the Ctenoides phylogeny result), as well as C. ales and two Limaria sp., were aligned with the Genbank sequences to create a concatenated dataset. ML and Bayesian phylogenies were constructed using this dataset based on methods described above (Figure 2). Mytilus edulis and Pecten maximus were used as outgroups.
2.4. Morphological Analysis
Specimens were photographed using a Canon EOS Rebel® T5 Digital SLR camera with a Canon EFS F/2.8 60 mm Macro Lens. Variations in tentacle coloration were recorded for live specimens or museum vouchers that contained photographs (Figure 3). Shell height and rib number were analyzed using ImageJ version 1.51n . The shell height of each individual was measured as the average value (mm) of the left and right valves from the umbo to the furthest distal point. The shell rib number of each individual was recorded as the average of the left and right valves, including the auricles, and was counted near the growth edge on the main body of the shell. Shell height versus shell rib number was plotted for all individuals based on phylogroups and tentacle colors. The Mann-Whitney test was used to determine whether the ratios were significantly different between groups.
3.1. Phylogenetic Analysis
For the Ctenoides only analyses, both the concatenated tree (Figure 1(a)) and the COI gene tree (Figure 1(b)) recovered two strongly supported clades. One clade contained individuals with the classic Ctenoides mitis coloration (pure white tentacles) and the other contained specimens with the classic C. scaber coloration (pure red tentacles). Therefore, the two groups were tentatively defined as the C. mitis and C. scaber clades, respectively (Figure 1). However, it is worth noting that both clades also contained individuals without the typical coloration of their corresponding species (Figure 2, discussed in the next section). For the museum specimens, one preidentified C. scaber was placed in the C. mitis clade. For the aquarium store “C. scaber” specimens, 14 were placed in the C. scaber clade while the other 4 were placed in the C. mitis clade.
Both 28S and H3 gene trees supported the monophyly of the C. mitis clade but could not resolve the relationships between the C. mitis clade and the C. scaber individuals (Figures 1(c) and 1(d)), possibly due to the slower rate of evolution of the two genes.
The family level phylogeny for Limidae (Figure 2) showed strong support for the genera Limaria, Lima, Acesta, and Ctenoides. However, internal relationships within these genera were occasionally poorly resolved, especially in Acesta. The C. mitis and C. scaber clades were recovered as sister groups to one another.
3.2. Morphological Analysis
In the Ctenoides scaber clade, four variations in tentacle coloration were observed: solid red tentacles (, Figure 3(a)), red tentacles with white stripes (, Figure 3(b)), white tentacles with red stripes (, Figure 3(c)), and solid white tentacles (, Figure 3(d)). The red coloration in C. scaber varied from dark orange to red but is recorded throughout as “red” for simplicity. In the Ctenoides mitis clade, three variations in tentacle coloration were observed: solid white tentacles (, Figure 3(e)), light orange tentacles (, Figure 3(f)), and white tentacles with red bands (banded), which may be a coloration observed only in juveniles (, Figure 3(g)).
Tentacle coloration patterns of museum and aquarium store specimens based on locality and phylogenetic position are summarized in Table 2. Based on the two classical morphologies (red tentacles for C. scaber and white tentacles for C. mitis), of the 10 Florida Keys specimens, 8 fell into their expected clade. Exceptions were two C. mitis, a light orange individual and a banded individual (likely a juvenile phenotype). Of the 12 Caribbean specimens, only two fell into the expected clades (one C. scaber and one C. mitis). The other specimens had tentacle color variations that were not typical of either group (Table 2). The atypical colorations did not show any obvious phylogenetic signals.
Shell rib analysis revealed that the distribution of shell rib numbers in the C. scaber clade () was significantly lower than that of the C. mitis clade (, Figure 4(a), Mann-Whitney Test, ). Rib numbers of individuals from the C. scaber clade ranged from 37 to 58 (mean 48.3) and rib numbers in the C. mitis clade ranged from 73 to 103 (mean 85.8). Red and striped tentacles corresponded to the low shell rib number group, while light orange and banded tentacles corresponded to the high shell rib group (Figure 4(b)). White tentacles corresponded to both the low and high shell rib groups. Figure 5 shows a representative shell from C. scaber (a) and C. mitis (b).
In summary, our phylogenetic analysis recovered two well-supported clades corresponding to Ctenoides scaber and C. mitis. Genetic variations within each clade are relatively low, but that could be the result of limited geographical sampling. Morphological analyses showed that tentacle color alone cannot be used to determine species, as colors vary inter- and intraspecifically (Figure 3). Shell rib counts gave statistically different distributions for the two species (Figure 4), but the two distributions have some degree of overlap.
Distinguishing Ctenoides scaber from C. mitis is challenging, as is evidenced by observed misidentification in museum specimens and by the hobbyist community. The two species have many characteristics in common; they share ecological similarities with similar maximum depths and overlapping geographical distributions, and they are often found in the same location, suggesting overlapping ecological niches . Internal anatomy, including mantle and eye structures, is also similar between the two species [17, 29, 30, 45].
Both C. scaber and C. mitis exhibit tentacle color variations. Mikkelsen and Bieler  placed high confidence in species identification using tentacle coloration in the Florida Keys. Our results confirmed the white and very light orange tentacle colorations in C. mitis, and the red coloration in C. scaber from the Florida Keys. However, outside of the Florida Keys, C. scaber exhibits many tentacle color variations without showing obvious genetic distinction from the red individuals (Figure 3).
As individuals of both species can possess solid white tentacles, tentacle color alone is not sufficient to determine species. This study confirmed that the C. scaber group has significantly fewer shell ribs than the C. mitis group. However, the overlap between the two species (between 59 and 78 ribs) means that no universal rule can be established to determine species based on shell ribs alone. Shell roughness is another characteristic which may be used to distinguish the two species, as the radial ribs of C. mitis are finer than C. scaber (Figure 5). However, shell roughness, size, and shape do vary to some extent across individuals, and, without instrumentation, species differentiation based on roughness could be challenging. Overall, the morphological diversity of shell ribs combined with the variety of tentacle coloration schemes stresses the importance of using molecular phylogenetics to verify species identification in future studies.
It is unclear whether the differential rib numbers between the two species has ecological significance. One possible hypothesis is that the high and low rib numbers are related to different predator defense mechanisms. The bright coloration (the result of carotenoids) in the tentacles of C. scaber  is hypothesized to be warning coloration for distasteful tissue. As C. mitis generally lacks brightly colored tentacles, they might have developed higher number of shell ribs to increase shell strength [47, 48]. However, this hypothesis does not explain the presence of white-colored C. scaber, so the effects of rib number and rib “roughness” on shell strength need to be examined further.
Based on the phylogenetic results, it is likely that C. scaber and C. mitis are recently diverged sister species. However, the speciation mechanism between the two has not been well studied. Ecological speciation has been demonstrated in numerous shallow-water marine invertebrates, with a divergence in life history being the most conspicuous difference between sibling species . In bivalves, these differences include bathymetry limits, unique dietary adaptations, or host/symbiont associations . Ctenoides scaber and C. mitis have overlapping geographical ranges and are often collected from the same substrate and at the same depth . This strong similarity in microhabitat makes it difficult to hypothesize what ecological factor(s) could result in divergent selection between the two species. One potential difference is the prevalence of whiter tentacles in C. mitis, which may be the result of fewer carotenoids in the diet. Future research is needed to determine if the two species have different dietary adaptations and how they utilize carotenoids.
Alternatively, the speciation could have been allopatric rather than ecological, with geographical isolation occurring following secondary contact. Although absolute geographical barriers are fewer in the sea, abiotic factors such as ocean currents or temperature can result in population isolation . More information regarding each species’ historical geographical range would be necessary to examine this hypothesis. It should be noted that the sampling in this study did not cover the entire geographical range of both species, so it is possible that additional genetically distinct clades or morphotypes may exist.
Although our family-level phylogeny based on limited taxon sampling was able to resolve phylogenetic relationships between Ctenoides scaber and C. mitis, it demonstrates the need for a more comprehensive phylogenetic reconstruction. To our knowledge, phylogenetic relationships among/within the nine extent Limidae genera are not well established. This is partially because taxon coverage in past studies are poor and partially because taxonomical classification of Limidae is largely based on shell morphologies, some of which may not reflect true phylogenetic signals. For example, a recent study shows that the genera Limatula and Antarctolima are actually nested within Ctenoides . We have also found that taxonomical inconsistencies are common in Genbank entries of Limidae species.
Limidae species exhibit many morphological and behavioral traits that are quite unique for bivalves. For example, a diverse spectrum of visual abilities can be found in the family, ranging from eyeless (Limaria) to pit eyes (Lima) to closed-lens eyes (Ctenoides) [29–31]. Members of the family also show distinct defense mechanisms, including nest-building behavior , swimming, tentacle autotomy [14, 18, 20, 21], and, potentially, defensive chemicals . The evolution of these traits cannot be addressed without a comprehensive phylogeny.
Finally, conservation concerns must be addressed. Ctenoides scaber is highly collected for the aquarium trade, but estimates of their abundance are lacking. How many of the collected file clams are actually C. scaber and not C. mitis is unknown, as is the abundance of both of the species throughout the western Atlantic. Considering the popularity of these organisms with shell collectors, SCUBA divers, and within the aquarium trade, a closer look at their conservation status is warranted.
The authors complied with all ethical standards in conducting this research.
Conflicts of Interest
The authors have no conflicts of interest.
The authors thank Spencer Hagerman and Brianna Maldonado for assistance with DNA extraction and PCR. The authors thank Gustav Paulay, John Slapcinsky, and Heather Robeson for assistance with specimen loan and deposits, and Aqua Imports for assistance with specimen acquisition. This work is funded and supported by the University of Colorado Museum of Natural History.
Supplemental Table 1. Specimen information and Genbank accession numbers of all individuals used in the phylogenetic analyses. Sequences newly generated in this study are highlighted in grey. Where necessary, species names were updated to the currently accepted taxonomy. UF = University of Florida; UCM = University of Colorado Museum of Natural History. (Supplementary Materials)
C. Wabnitz, M. Taylor, E. Green, and T. Razak, “From Ocean To Aquarium: The Global Trade In,” in Marine Ornamental Species Advanced Aquarist | Aquarist Magazine and Blog. UNEP, World Conservation Monitoring Centre, Cambridge, UK, 2003.View at: Google Scholar
P. Humann and N. DeLoach, Reef Creature Identification; Florida, Caribbean, Bahamas, New World Publications, Jacksonville, FL, USA, 3rd edition, 2013.
T. Okutani, “A new discovery of streak-like flashing luminescence in Ctenoides ales (Finley, 1927) (Bivalvia: Limidae),” Venus Japanese J Malacol, vol. 53, pp. 57–59, 1994.View at: Google Scholar
S. Okubo, T. Kurokawa, T. Suzuki et al., “On the mechanism of streak-like flash on the mantle of Ctenoides ales (Bivalvia : Limidae),” The Japanese Journal of Malacology Venus, vol. 56, pp. 259–269, 1997.View at: Google Scholar
B. Morton, “A comparison of lip structure and function correlated with other aspects of the functional morphology of lima lima, limaria (Platilimaria) fragilis, and limaria (Platilimaria) hongkongensis sp.nov. (Bivalvia: Limacea),” Canadian Journal of Zoology, vol. 57, no. 4, pp. 728–742, 1979.View at: Publisher Site | Google Scholar
C. Trigg, D. Harries, A. Lyndon, and C. G. Moore, “Community composition and diversity of two Limaria hians (Mollusca: Limacea) beds on the west coast of Scotland,” Journal of the Marine Biological Association of the United Kingdom, vol. 91, no. 7, pp. 1403–1412, 2011.View at: Publisher Site | Google Scholar
L. F. Dougherty, “Distribution, habitat, and ecology of the “disco” clam, Ctenoides ales,” in Oysters and Clams: Cultivation, Habitat Threats and Ecological Impact, J. L. Romalde, Ed., pp. 1–14, 2016.View at: Google Scholar
C. J. Lodeiros and J. H. Himmelman, “Reproductive cycle of the bivalve Lima scabra (Pterioida: Limidae) and its association with environmental conditions,” Revista de Biología Tropical, vol. 47, no. 3, pp. 411–418, 1999.View at: Google Scholar
A. K. Dukeman, N. J. Blake, and W. S. Arnold, “The reproductive cycle of the flame scallop, Ctenoides scaber (Born 1778), from the lower Florida keys and its relationship with environmental conditions,” Journal of Shellfish Research, vol. 24, no. 2, pp. 341–351, 2005.View at: Publisher Site | Google Scholar
P. J. Stephens, “Mechanical and Chemical Sensitivity at the Mantle Edge of the File Clam Lima scabra (Born),” Marine Behaviour and Physiology, vol. 5, no. 2, pp. 79–90, 1978.View at: Google Scholar
A. Bell and G. J. Mpitsos, “Morphology of the eye of the flame fringe clam, Lima scabra,” Biological Bulletin, pp. 414-415, 1968.View at: Google Scholar
G. Mpitsos, “Physiology of vision in the molusk Lima scabra,” J Neurophysiol, pp. 36–371, 1973.View at: Google Scholar
L. F. Dougherty, R. R. Dubielzig, C. S. Schobert, L. B. Teixiera, J. Li et al., “Do you see what I see? Optical morphology and visual capability of ‘disco’ clams (Ctenoides ales),” Biology Open, vol. 6, pp. 648–653, 2017.View at: Google Scholar
T. J. Page and K. Linse, “More evidence of speciation and dispersal across the Antarctic Polar front through molecular systematics of Southern Ocean Limatula (Bivalvia: Limidae),” Polar Biology, vol. 25, no. 11, pp. 818–826, 2002.View at: Google Scholar
C. Oliveira and R. S. Absalão, “The genera Limatula and Limea (Mollusca, Pelecypoda, Limidae) from deep waters off Brazil,” Zootaxa, no. 1940, pp. 48–58, 2008.View at: Google Scholar
J.-M. Gagnon, E. Kenchington, A. Port, L. J. Anstey, and F. J. Murillo, “Morphological and genetic variation in North Atlantic giant file clams, Acestaspp. (Bivalvia: Limidae), with description of a new cryptic species in the northwest Atlantic,” Zootaxa, vol. 4007, no. 2, pp. 151–180, 2015.View at: Publisher Site | Google Scholar
R. Lanfear, P. B. Frandsen, A. M. Wright, T. Senfeld, and B. Calcott, “PartitionFinder 2: new methods for selecting partitioned models of evolution formolecular and morphological phylogenetic analyses,” Molecular Biology and Evolution, vol. 34, no. 3, pp. 772-773, 2016.View at: Publisher Site | Google Scholar
A. Rambaut, FigTree version 1.3.1, Computer program, 2009, http://tree.bio.ed.ac.uk/software/figtree/.
A. Lin and L. Pompa, “Carotenoids of the rea clam Lima scabra,” Boletín del Inst Ocean la Univ Oriente, vol. 16, pp. 83–86, 1977.View at: Google Scholar
J. E. N. Veron, Corals in space and time: the biogeography and evolution of the Scleractinia, Comstock/Cornell University Press, Ithaca, NY, USA, 1995.