Compsanthias, Gill, 2024

Gill, Anthony C., 2024, Compsanthias, new genus of Pacific Anthiadidae (Teleostei), Zootaxa 5463 (3), pp. 360-374 : 361-371

publication ID

https://doi.org/ 10.11646/zootaxa.5463.3.3

publication LSID

lsid:zoobank.org:pub:065492E6-456D-4245-85F0-2CC61B843628

DOI

https://doi.org/10.5281/zenodo.11619375

persistent identifier

https://treatment.plazi.org/id/E97A87D2-FFC2-FFB5-85C4-FE54FB98F310

treatment provided by

Plazi

scientific name

Compsanthias
status

gen. nov.

Compsanthias new genus

Compsanthias new genus,

type species Anthias (Pseudanthias) ventralis Randall, 1979 View in CoL .

Diagnosis. The following combination of characters is unique among anthiadids: dorsal-fin rays X,16–18, usually X,17; anal-fin rays III,9–10, usually III,9; pectoral-fin rays 14–16, usually 15, most rays branched; branched caudal-fin rays 7+6; tubed scales in lateral series 39–46.

Compsanthias has several unusual characters that are possibly synapomorphic for the included species but are either too poorly studied in other anthiadids or are otherwise difficult to evaluate, including the following.

Enlarged infraorbital 3. Although infraorbital 3 (IO3) is relatively large in other anthiadids, and second only in size to IO1, it is proportionally larger in Compsanthias than in other examined taxa (around 1.5 times length of IO4 + IO5, versus roughly equal to or less than length of IO4 + IO5; Fig. 1 View FIGURE 1 ). However, infraorbital morphology has not been adequately surveyed in the family. Katayama (1959) surveyed and illustrated several genera, which he grouped based on the relative development of suborbital shelves on bones in the infraorbital series. Comparison of Katayama’s illustrations with my material shows some disagreement in the relative sizes of the bones, suggesting a lack of accuracy in that aspect of his illustrations. More extensive survey of this character is required.

Modally III,9 anal-fin rays. Compsanthias is unusual among anthiadids in having a strongly modal count of III,9 anal-fin rays rather than the usual III,7–8. Although this appears to be apomorphic, and a possible synapomorphy for the species included in the genus, Caesioperca and two species of Tosanoides also have III,9 rays. This character is therefore discussed below as a potential synapomorphy of these taxa.

Oblique stripe from upper part of eye to upper part of head. Compsanthias species have a variously developed red to purple stripe extending from the middle of the upper orbital rim to the upper part of the head at about the vertical through the preopercle edge. In some individuals the stripe may be very short or reduced to a few isolated spots. Although the stripe may be a synapomorphy of the species included in the genus, similar markings are found in some species of Pyronotanthias and arguably in Serranocirrhitus Watanabe, 1949 .

Description. General body shape as in Fig. 2A View FIGURE 2 ; greatest body depth 32–38% SL; head length 30–35% SL; orbit diameter 8–13% SL; predorsal length 32–35% SL; preanal length 58–64% SL; prepelvic length 31–35% SL; dorsal-fin base length 55–64% SL; anal-fin base length 20–24% SL; caudal peduncle depth 13–15% SL.

Scales moderately large, tubed scales in lateral series 39–46; circumpeduncular scales 20–23; auxiliary scales sometimes present on nape and head; head fully scaled except for throat, branchiostegal membranes, lips, extreme front of snout and broad area around and below nostrils; dorsal and anal fins broadly scaled basally; scales with peripheral cteni only and no truncated basal cteni ( Roberts 1993).

Dorsal-fin rays X,16–18, usually X,17; dorsal-fin origin above posterior end of operculum; primary shaft of first dorsal pterygiophore angled slightly posterodorsally; ADPF S/S/3/1+1/1/1/1/1/1 or S/S+ Sv /3/1+1/1/1/1/1/1; distal radials of spinous dorsal pterygiophores closely applied to their serially associated middle + proximal radials, with prominent median process extending posteriorly between basal notch in serially associated spine; dorsal pterygiophores in interneural spaces 9–13 1/1/1+1/1+1/1, 1/1/1+1/1/1+1, 1/1/1+1/1+1/1+1 or 1/1+1/1/1+1/1+1; terminal dorsal pterygiophore in interneural space 18–19; anal-fin rays III,9–10, usually III,9; proximal tip of first anal pterygiophore either positioned about midway between vertebra 10 and 11, or abuts haemal spine of vertebra 11 (caudal vertebra 1); terminal anal pterygiophore in interhaemal space 6–7; pectoral-fin rays 14–16, usually 15, upper 2–3 and lower 1–6 rays unbranched, remaining rays branched near their distal tips (all rays unbranched in specimens smaller than about 25 mm SL); no serrations on longest pectoral-fin rays, but with low ridges extending dorsally and ventrally; pelvic-fin rays I,5, all segmented rays branched; caudal-fin rays 7–10+9+8+7–9=31–35; branched caudal-fin rays 7+6 ( Figs. 2B View FIGURE 2 , 3–4 View FIGURE 3 View FIGURE 4 ).

Vertebrae 10+16; no parapophyses on first caudal vertebra; epineurals present on vertebrae 1 through 12–13; ribs present on vertebrae 3 through 10; uppermost principal ray on hypural 5, lowermost principal ray between parhypural and pu2 haemal spine; no procurrent spur on upper procurrent ray in lower lobe ( Johnson 1975); penultimate procurrent ray in lower lobe not foreshortened ( Johnson 1975); parhypural and hypurals 1 through 5 present, autogenous; parhypural relatively broad proximally with well-developed hypurapophysis; pu2 and pu3 haemal spines autogenous; pu2 neural spine short; three epurals, decreasing in size posteriorly; single uroneural (posterior uroneural absent); autogenous cartilages present between distal tips of posterior epural and hypural 5, at distal tip of hypural 5, between distal tips of parhypural and pu2 haemal spine, at distal tip of pu2 haemal spine, between distal tips of pu2 and pu3 haemal spines, and anterior to pu3 haemal spine ( Figs. 2B View FIGURE 2 , 3 View FIGURE 3 ).

Anterior part of upper lip of males not hypertrophied; mouth large, oblique, posterior margin of maxilla reaching to point ranging from beneath middle of eye to posterior edge of eye; mouth terminal, lower jaw projecting slightly when mouth closed; supramaxilla absent; premaxilla with curved canine laterally at front of jaw, followed by broadly spaced series of slightly smaller and slenderer canine teeth on sides of jaw, posterior teeth pointing forward; inside these, premaxilla with band of small conical teeth, about 4 or 5 rows wide anteriorly reducing to single row posteriorly, with several nearest symphysis enlarged and directed medially; dentary with pair of canines anteriorly, which project anterolaterally, behind these about two or three rows of small conical teeth in band across front of dentary, those nearest symphysis enlarged and directed medially; anterior third of dentary with enlarged backward curving canine teeth, followed on posterior two thirds of jaw by series of inward (anterior few) to anteriorly curving (remaining teeth) canine teeth; vomer with chevron of small teeth, 1–2 rows wide; palatines with narrow band of small teeth, 2 rows wide; tongue, ectopterygoid and endopterygoid edentate ( Fig. 5 View FIGURE5 ).

Opercle with three flat spines, uppermost often indistinct and hidden by scales and skin, lowermost below junction with subopercle, middle (primary) spine closer to lowermost spine than to uppermost; preopercle finely serrated at angle and on posterior, vertical edge; interopercle and subopercle with weakly to well-developed serrations ( Fig. 5 View FIGURE5 ).

Anterior naris in short membranous tube, with weak to moderately large triangular flap dorsoposteriorly; posterior naris roughly semicircular, with at most weakly raised rim; six infraorbital (IO) bones, forming continuous tube for laterosensory canal, IO6 (dermosphenotic) tightly bound to skull; all infraorbital bones with open groove laterally, except IO1 (“lachrymal”), where groove veers ventrally on middle part of part of bone and ends near anterior part of bone, with separate tubular opening anteriorly; IO1 largest, accounting for about 40% of infraorbital canal system, IO3 second largest bone in series, about 75% length of IO1 (excluding suborbital shelf), IO2 and IO4 much smaller, subequal in size, IO5 and IO6 very short ( Fig. 1B View FIGURE 1 ); suborbital shelves present on IO2–5, that of IO3 broadest and extending from about middle of IO2 to middle of IO5, shelves on IO2 and IO4–5 much narrower; posterior rim of orbit without papillae (fleshy projections between laterosensory pores associated with frontal and infraorbitals).

Pectoral girdle consisting of cleithrum, two extrascapulae, posttemporal, supracleithrum, dorsal and ventral postcleithra (= postcleithra 2 and 3, respectively, of Gottfried 1989), scapula, coracoid, four proximal radials and unossified distal radials at bases of pectoral-fin rays; laterosensory canal exits pectoral girdle on medial surface of supracleithrum; posterior exposed edge of posttemporal smooth or with indistinct serrations; uppermost two proximal radials articulating with scapula, next with both scapula and coracoid, and lowermost with coracoid (= proximal radial formula 2-1-1); scapula with well-developed foramen; ventral tip of coracoid in narrow pocket between medial lamina and posteroventral process of cleithrum; posteroventral process of cleithrum well-developed ( Fig. 6 View FIGURE 6 ).

Paired pharyngobranchials (pb) 1 through 4 present, pb4 cartilaginous; tooth plates present on pb2 through pb4, tooth plate on pb4 small and autogenous; paired epibranchials (eb) 1 through 4 present; uncinate process on eb1 fan-shaped, broadly rimmed with cartilage, and directed posterodorsally; no tooth plate on eb2; relatively large toothplate on eb3 ( Fig. 7 View FIGURE 7 ); paired ceratobranchials (cb) 1 through 5 present; tooth plate on cb5; paired hypobranchials (hb) 1 through 3 present, none with toothplates; median basibranchials (bb) 1 through 4 present, bb4 as cartilage, none with toothplates; small accessory cartilages at eb1–cb1 and eb2–cb2 junctions; gill rakers elongate on anterior/lateral face of first arch, 6–9 on eb1, uppermost at junction with pb1, and 21–25 on cb1 and hb1; shorter club-like rakers or rudiments present on posterior face of eb1, cb1 and hb1, on both anterior and posterior faces of eb2–3, cb2–4 and hb2, and on anterior face of hb3, eb4 and cb5; seven branchiostegals; ceratohyals closely applied but not sutured; anterior three branchiostegals medially on narrow anterior portion of anterior ceratohyal (corresponding to Group II branchiostegals of Doosey 2013), next two laterally on broad portion of anterior ceratohyal, final two laterally on posterior ceratohyal (posterior four corresponding to Group I branchiostegals of Doosey 2013).

Composition. Compsanthias hawaiiensis ( Randall, 1979) , C. ventralis ( Randall, 1979) .

Etymolology. From the Greek κομψός, meaning elegant or pretty, and anthias, alluding to the graceful form and striking coloration of the included species. Gender is masculine.

Comparisons. Compsanthias is relatively distinctive, and unlikely to be confused with other anthiadid genera. As noted below, the high number of segmented anal-fin rays distinguishes it from all other anthiadids except Caesioperca and two species each of Tosanoides and Luzonichthys . Compsanthias is readily distinguished from Caesioperca in having fewer supraneural bones (two, sometimes with the vestige of a third bone versus three well-developed bones), fewer segmented dorsal-fin rays (16–18 versus 19–22), fewer lateral-line scales (39–46 versus 47–65), and fewer branched caudal-fin rays (7+6 versus 8+7), and in lacking transformed (truncated basal) cteni on scales (present in Caesioperca ). It differs from Tosanoides in having most pectoral-fin rays branched (versus all unbranched in Tosanoides ) and more lateral-line scales (39–46 versus 29–36), and from Luzonichthys in having fewer lateral-line scales (39–46 versus 51–78), fewer pectoral-fin rays (14–16, usually 15 versus 17–23) and an undivided dorsal fin (versus notched or divided), and in lacking papillae on the posterior orbital rim and serrations on the longest few pectoral-fin rays (versus present in Luzonichthys ).

In describing Pseudanthias hangapiko, Shepherd et al. (2021) noted that COI sequences for their new species were most like those of Compsanthias species (as Pseudanthias hawaiiensis and P. ventralis ), although with substantial difference (uncorrected divergence of 17.0% and 16.8%, respectively). Although available morphological information suggests that P. hangapiko is not correctly placed in Pseudanthias (as also suggested by COI sequences), the species is too poorly known to justify inclusion in Compsanthias or any other anthiadid genus. As noted by Shepherd et al., it differs from Compsanthias in having fewer anal-fin rays (III,8 versus III,9–10) and a slenderer body form (greatest body depth 26–29% SL versus 32–38% SL).

Relationships. The phylogenetic position of Compsanthias is unresolved. The following is a summary of characters that may have bearing on relationships to other anthiadids. The first five of these characters were recently reviewed by Pogonoski & Gill (2021), who provide more detailed discussion of the characters and their taxonomic distribution. Character 6 is readily observable, and a standard character in species descriptions. In contrast, the remaining characters are only readily observable in cleared and stained specimens, and thus suffer from inadequate survey. I include them here in part to bring them to the attention of other researchers.

1. Number of branched caudal-fin rays. Plesiomorphically, anthiadids have 8+7 branched caudal-fin rays. Compsanthias is relatively derived in having only 7+6 branched rays. However, reduction to 7+6 or fewer branched rays is characteristic of the vast majority of anthiadid genera.

2. Presence or absence of transformed (truncated basal) cteni. Anthiadids plesiomorphically have transformed (truncated basal) cteni on their scales ( Roberts 1993), although these are often confined to the head and anterior body. Compsanthias is relatively derived in only having peripheral cteni on the scales. As for the previous character, however, the derived state is widespread among anthiadid genera.

3. Presence or absence of trisegmental pterygiophores in dorsal and anal fins. Pterygiophores in the soft dorsal and anal fins may be trisegmental, composed of separate proximal, middle and distal radials, or bisegmental, where the proximal and middle radials are undifferentiated. The former condition is plesiomorphic in anthiadids. Compsanthias has only bisegmental pterygiophores.However, like the previous two characters, the derived condition is widespread among anthiadids.

4. Presence or absence of parapophyses on first caudal vertebra. In most anthiadids, the posteriormost pair of parapophyses are borne on the terminal precaudal vertebra. In other anthiadids, the posteriormost pair of parapophyses are borne on the first caudal vertebra, which also bears a haemal spine. The former condition is plesiomorphic and found in Compsanthias (Gill 2022: fig. 7; Fig. 2B View FIGURE 2 ). As noted by Gill (2022), the apomorphic condition supports monophyly of a clade consisting of Mirolabrichthys , Nemanthias , Pyronotanthias , Luzonichthys , Franzia , Rabaulichthys Allen 1984 , Anatolanthias Anderson, Parin & Randall, 1990 , Tosana , Hemanthias Steindachner, 1875 , Choranthias Anderson & Heemstra, 2012 and Pseudanthias (as currently defined, excluding P. hangapiko ).

5. Number of supraneural bones. Plesiomorphically, anthiadids have three well-developed supraneurals, with one in the preneural space and two in the first interneural space. Compsanthias is relatively derived in either lacking the posterior supraneural or having it present only as a tiny vestige ( Figs. 2B View FIGURE 2 , 4 View FIGURE 4 ). Reduction or loss of one of more supraneurals occurs widely in anthiadid genera, and varies among species in Pseudanthias , Tosanoides and Plectranthias . Meganthias Randall & Heemstra, 2006, Odontanthias and Sacura Jordan & Richardson, 1910 also have two supraneurals, but with an apomorphic arrangement, with the two bones closely spaced such that it is often difficult to determine whether they insert within the preneural or first interneural spaces (see Gill & Russell 2019). The two supraneurals of Compsanthias are more widely spaced, typical of other taxa with two supraneurals.

6. Number of anal-fin rays. The vast majority of anthiadid species have modal counts of either III,7 or III,8 anal-fin rays. Compsanthias is apomorphic in having a strongly modal count of III,9 anal-fin rays. Aside from occasional specimens of some species, a count of III,9 is otherwise known from species of Caesioperca Castelnau, 1872 and two species of Tosanoides ( T. aphrodite Pinheiro, Rocha & Rocha, 2018 and T. fucinus ( Randall & Ralston,1985) ; the remaining species of Tosanoides have III,8 anal-fin rays). This character therefore challenges monophyly of the genus Tosanoides , as well as characters and preliminary molecular studies that suggest only a distant relationship between Compsanthias and Caesioperca . The alternative is that the relatively high number of anal-fin rays is independently derived, which would thus support monophyly of Compsanthias (see above Diagnosis). Two species of Luzonichthys Herre, 1936 also have nine segmented anal-fin rays, L. microlepis ( Smith, 1955) and L. williamsi Randall & McCosker, 1992 . However, in these species the high count is in part due to the transformation of the posterior (third) spine to a segmented ray (giving a count of II,9), and is thus not homologous with the III,9 condition.

7. Presence or absence of interarcual cartilage. Plesiomorphically, anthiadids have a well-developed interarcual cartilage (IAC) between pharyngobranchial 2 and the uncinate process on epibranchial 1 ( Baldwin 1990: fig. 20; Gill 2022: fig. 11A,B). Compsanthias is relatively derived in lacking an interarcual cartilage ( Fig. 7 View FIGURE 7 ). Baldwin (1990) noted that a well-developed IAC is present in species currently assigned to Plectranthias , Caprodon and Hemanthias , a reduced IAC is found in Anthias Bloch, 1792 , Baldwinella Anderson & Heemstra, 2012 , Choranthias and Pronotogrammus Gill, 1863 , whereas Luzonichthys , Nemanthias and Pseudanthias lack an IAC. I find the distinction between reduced and well-developed IAC somewhat arbitrary, as specimens of Anthias examined by me appear to have a relatively well-developed IAC (Gill 2022: fig. 11B). Other anthiadid taxa with a well-developed IAC include Caesioperca , Trachypoma Günther, 1859 , Hypoplectrodes Gill, 1862 and Lepidoperca Regan, 1914 . In Tosanoides IAC is present as a tiny nubbin (Gill 2022: fig. 11C). I confirm Baldwin’s reported absence of IAC in Luzonichthys , Nemanthias and Pseudanthias , and note that it is also absent in Franzia , Mirolabrichthys , Rabaulichthys , Pyronotanthias , Serranocirrhitus and Tosana (see Gill 2022: figs. 5–6, 11D). Although the absence of IAC is a potential synapomorphy uniting these taxa with Compsanthias , the character needs more extensive review.

8. Shape of uncinate process on epibranchial 1. Anthiadids plesiomorphically have a more-or-less rodlike uncinate process on epibranchial 1 ( Baldwin 1990: figs. 20; Gill 2022: fig. 11A,B). In Compsanthias the uncinate process is distinctively fan-shaped, broadly rimmed with cartilage, and inflected slightly dorsally ( Fig. 7 View FIGURE 7 ). Similar derived morphology is found in Franzia , Luzonichthys , Mirolabrichthys , Nemanthias , Pseudanthias , Pyronotanthias , Rabaulichthys , Serranocirrhitus , Tosana and Tosanoides (Gill 2022: figs. 5, 11C,D). As with the previous character, this potential synapomorphy needs more extensive review.

9. Posterior extent of gill-raker rudiments on ceratobranchials. Plesiomorphically in anthiadids the posteriormost gill-rakers are present (as rudiments) on the anterior face of cb4. Compsanthias is relatively derived in having a series of small gill-raker rudiments on the anterior face of cb5, as well as on the posterior face of cb4. Similar morphology is found in Anthias , Franzia , Luzonichthys , Mirolabrichthys , Nemanthias , Pseudanthias , Pyronotanthias , Rabaulichthys , Tosana and Tosanoides . Serranocirrhitus has gill-raker rudiments on cb5 but differs from the other taxa in lacking rudiments on the posterior face of cb4. More extensive survey of this character is required.

10. Presence or absence of cartilage anterior to preural 3 haemal spine. Johnson (1983) considered the presence of a cartilage anterior to the preural 3 haemal spine (termed third preural radial cartilage by Johnson and inter-haemal cartilage of fourth preurual centrum or CIHPU4 by Fujita1989) as plesiomorphic in perciform fishes. Johnson proposed the absence of the cartilage in serranid fishes as a synapomorphy of the family (in which he included anthiadids as a subfamily), although noted it also absent in several other families (e.g., see Johnson 1984: tab. 120). Uniquely among examined anthiadids, Compsanthias , Serranocirrhitus and Tosanoides have a small cartilage in the position of CIHPU4 ( Fig. 3 View FIGURE 3 ). However, more detailed survey is required to not only determine the taxonomic distribution of this character, but also whether the cartilage represents a novel structure (and thus a synapomorphy of taxa that possess it) or plesiomorphic presence of CIHPU4 (suggesting monophyly of anthiadids that lack the cartilage).

11. Position of ventral tip of coracoid. In Compsanthias , the ventral tip of the coracoid inserts in a narrow pocket between the medial lamina and posteroventral process of the cleithrum near the base of the posteroventral process ( Fig. 6 View FIGURE 6 ). Comparison with other taxa suggests this relatively anterior placement of the coracoid tip is plesiomorphic among anthiadids. In Franzia , Luzonichthys , Mirolabrichthys , Nemanthias , Pseudanthias , Pyronotanthias , Rabaulichthys and Tosana , the coracoid tip is positioned farther back, solely on the posteroventral process, sometimes near the distal tip of the process. However, the distinction between the two conditions is subtle, and not always clearly visible on radiographs. Further survey is required.

Remarks. Randall (1979) originally described C. hawaiiensis (type locality Moku Manu, Oahu, Hawaiian Islands) as a subspecies of C. ventralis (type locality off Gannet Ridge, Pitcairn Island). This largely reflects the close similarity of the two, which differ mostly in live coloration. The latter species is currently regarded as widely distributed in the West and central Pacific. However, it exhibits pronounced geographic variation in coloration and therefore is likely to be a complex of species ( Kuiter 2004; Tea 2015; Rowlett 2016). This possibility is under study separately.

C.C. Baldwin (pers. comm., 2024) drew my attention to observations she had made as a graduate student on a 14.2 mm SL larval specimen of C. hawaiiensis from Hawaii. I can confirm her identification based on her recorded fin-ray counts (dorsal-fin rays X,17; anal-fin rays III,9; pectoral-fin rays 15). She considered spination features of the specimen to be very distinctive, and further noted the important potential of larval spination features in anthiadid systematics, particularly at the supraspecific level (e.g., see Kendall 1984; Baldwin 1990). The following is a summary of spination characteristics recorded by her from the specimen: frontals and parietals very rugose (almost spinelike); suproccipital with small ridge (almost a thickening of the rugosity); supraorbital ridge of frontal with many spines and closely spaced, well-developed ridges; infraorbitals serrate; anguloarticular with spines; posttemporal with very large spine, bearing serrations and well-developed ridges; opercle with large serrate spines; preopercle with a heavily serrated and ridged very large spine at angle, with one tiny and one large serrate spines ventrally and six serrate smaller spines dorsally; interopercle with relatively small serrate spine dorsal to large preopercular spine. Other features of the specimen: no pigment on trunk; a few spots on dorsal-fin membranes between second and third spines, and third and fourth spines; premaxilla with a series of spots; pelvic fins with pigment; ventral preopercular spine with spots; scales with a single row of cteni.

I asked B.C. Mundy if he was aware of any larval specimens matching Baldwin’s description. His search of collections revealed a single 10.5 mm SL specimen of C. hawaiiensis in the reference collection of the Pacific Islands Fisheries Science Center, Honolulu (PIFSC). I have reproduced his photograph of the specimen here ( Fig. 8 View FIGURE 8 ).

Material examined. Compsanthias hawaiiensis, AMS I. 18720-001 (1W,X, paratype, 49.0 mm SL), BPBM 10560 View Materials (1X, holotype, 76.9 mm SL), USNM 215285 View Materials (1W,X, paratype, 42.0 mm SL); C. ventralis, AMS I. 45300- 254 (1CS, 36.5 mm SL), AMS I.49427-001 (11W, 24.5–43.7 mm SL), AMS I.49577-002 (1W, 29.4 mm SL); AMS I.49577-003 (1W, 35.6 mm SL), BPBM 16883 View Materials (1X, holotype, 51.5 mm SL), USNM 215289 View Materials (1X, paratype, 29.8 mm SL), USNM 400605 View Materials (1X, 52.0 mm SL), USNM 422955 View Materials (1X, 58.4 mm SL).

Kingdom

Animalia

Phylum

Chordata

Order

Perciformes

Family

Anthiadidae

Loc

Compsanthias

Gill, Anthony C. 2024
2024
Loc

Compsanthias

Gill 2024
2024
GBIF Dataset (for parent article) Darwin Core Archive (for parent article) View in SIBiLS Plain XML RDF