Cercamia Randall & Smith, 1988

Fraser, Thomas H., Bogorodsky, Sergey V., Mal, Ahmad O. & Alpermann, Tilman J., 2021, Review of the cardinalfishes of the genus Cercamia (Percomorpha: Apogonidae) of the Red Sea and Indian Ocean with descriptions of three new species, Zootaxa 5039 (3), pp. 363-394 : 367-369

Cercamia Randall & Smith, 1988

Cercamia Randall & Smith, 1988: 7 , type species Cercamia cladara Randall & Smith, 1988 , by original designation. Apogonoides Bleeker, 1849 , nomen oblitum— Fraser & Bogorodsky 2019: 579.

Diagnosis. Apogonid fishes with 9 + 15 vertebrae; no supraneurals; neural spines an anterior pterygiophores 0/0/0/1/2/1/; two epineurals articulate respectively with first two vertebrae, none on ribs 3–7, first rib on third vertebra, last rib associated with ninth vertebra; two hypurals, 1+2+3+4 fused to terminal centrum as deeply divided plate, free split-like hypural 5; parhypural closely appressed (or fused?) to hypural plate; no uroneurals; 3 epurals, third variably cartilaginous; terminal centrum with short urostyle and without sheath; vertebral neural foramina present near base of neural spines, no inferior foramina near base of haemal spines; dorsal fins VI + I,9 or 10, anterior pterygiophore with 1 supernumerary spine; anal fin II,11–14, anterior pterygiophore with 1 supernumerary spine, first anal pterygiophore straight; pectoral-fin rays 9–11, usually 10, upper most ray short, upper most second ray unbranched, middle third to sixth rays branched, lower eighth ray unbranched, ninth and tenth rays short; caudal fin forked, with 9+8 principal rays, 8+7 branched rays; scales deciduous, cycloid or weakly ctenoid; pored, pitted or notched lateral-line scales absent; one small spine at angle of preopercular ridge, 2–5 small spines near angle of preopercular edge; 1 small posttemporal spine; cleithrum narrow; scapula elongate; lower part of coracoid with a posterior nub and curved rod to cleithrum, coracoid with narrow anterodorsal flange reaching cleithrum creating a large gap; one postcleithrum with dorsal tip nearly level with base of supracleithrum thinning below coracoid, curved, rod-like, extending along more than half of the lower abdomen, ligament from postcleithrum to a small spine on external dorsal wing of basipterygium; supramaxilla absent; palatine teeth absent; vomer toothless or with 2 small teeth; gill rakers present on first and second arches; interarcual cartilage long; first pharyngobranchial variably separate cartilage or continuous long cartilage from first epibranchial; free neuromasts widespread on head, body and caudal fin; cephalic canals with a few large pores and a coronal commissure; in life, the species with blackish peritoneum (may appear silvery externally), narrow whitish band (swimbladder) as a boundary at the upper portion of the peritoneum to near vent, blackish above the swimbladder and below the vertebral column, body otherwise translucent with or without patterns of small reddish marks or absent, fins translucent, with or without reddish marks, head mostly red grading to reddish spots or a red and whitish stripe, stellate melanophores present or absent behind the eye or on chin, iris yellowish.

Generic comparative morphology & molecular phylogenetics. Some of the rich morphologic character variations include: principal caudal-fin rays are 9+8 with 8+7 branched for Lachneratus Fraser & Struhsaker, 1991 and Cercamia , 8+7 with 7+6 branched for Pseudamiops Smith, 1954 , 9+8 with 7+6, 7+7 or 8+7 branched for Gymnapogon Regan, 1905 and with 9+8 all branched for Paxton Baldwin & Johnson, 1999 ; low pectoral-fin ray counts of 9–11 for Lachneratus and Cercamia , 13–19 for Pseudamiops and Gymnapogon and 17–18 for Paxton . The first anal pterygiophore is straight for Lachneratus , Cercamia , Pseudamiops and Gymnapogon , curved in Paxton . A thin lower postcleithrum is present in Lachneratus (elongate), Cercamia (single, very elongate), Gymnapogon (single, long), Pseudamiops (short), and Paxton Baldwin & Johnson, 1999 (single, short). Species of Gymnapogon have single tooth or a small row to slightly enlarged teeth on the vomer, while Cercamia may variably have none, one or two small vomerine teeth, Lachneratus has 1–2 large teeth on the vomer, Pseudamiops has 1–2 large or a row of small vomerine teeth, and Paxton has 1–2 large vomerine teeth anteriorly and a semicircular patch of smaller teeth posteriorly. Gymnapogon is the only genus in the clade with a shelf on the third infraorbital. The first pharyngobranchial is variably a separate cartilage or a continuous long cartilage from the first epibranchial for Cercamia and contrasts with Gymnapogon ’s variable ossified tips or absence. The interarcual cartilage length is unknown for Lachneratus and Pseudamiops and is long in Cercamia , Gymnapogon and Paxton . The almost complete lack of reports on pored (except Yoshida et al. 2011) or pitted lateral scales in the literature for Cercamia perhaps coincides with the extremely deciduous, cycloid or ctenoid scales ( Allen 1987; Randall & Smith 1988; Aizawa & Senou 1991; Hayashi 1991; Allen et al. 2015; Sato et al. 2019, fig.7). The vertebral count 9+15 for all known species of Cercamia is unique among apogonids. Epineurals are reduced to two for Cercamia and Pseudamiops , three for Lachneratus , six for Gymnapogon and eight for Paxton . Two ossified larger epurals are present in Lachneratus , Pseudamiops and Gymnapogon . Cercamia has three epurals, the third may be cartilaginous. An appressed parhypural nearly fused to the hypural plate is just one step away from the fusion of these two elements in Gymnapogon . Cercamia and Lachneratus have deeply forked caudal fin. Species of Gymnapogon have forked, emarginate or rounded caudal fin while Pseudamiops and Paxton have rounded caudal fin. Species of Cercamia have two to five tiny spines at or near the angle of the preopercle while species of Gymnapogon have some variation from a strong spine at the angle of the preopercle to a shortened spine almost hidden in the flap and one widespread species complex (B-marked) usually have one additional tiny spine on the edge of the upper arm. Pseudamiops lacks spines, large or small on the preopercle while Lachneratus has one to three spines on the edge of the preopercle.

Cercamia like Gymnapogon , Pseudamiops and Paxton have widespread free neuromasts on the body ( Smith 1954; Ida & Moyer 1974; Sato et al. 2019). Lachneratus is lacking a full description of free neuromasts ( Fraser & Struhsaker 1991; Bergman 2004).All other members of the Apogoninae have a reduced coverage of free neuromasts on body scales, with variations on pored lateral-line scales, on lateral line pit scales on a few scales on the nape, and a few scales at the base of the caudal fin. All Apogoninae have caudal-fin rays with coverage variations of free neuromasts.

When describing Cercamia, Randall & Smith (1988) suggested possible generic relationships with Rhabdamia and Pseudamiops . Hayashi (1991) provided a more detailed description of Japanese specimens of Cercamia eremia including osteology. He discussed possible generic relationships and noted additional differences between Cercamia and Rhabdamia (scales, cephalic canal system and free neuromasts) and a similarity with Pseudamiops (postcleithrum). Fraser & Struhsaker (1991) suggested that their new genus Lachneratus shared some characters with Cercamia and listed some differences as did Hayashi (1991). Baldwin & Johnson (1999) discussed Cercamia with respect to Pseudamiops and the Pseudamiinae and noted the lack of characters in support for Cercamia as a member of the Pseudamiinae , excepting the absence of a basisphenoid and reduction in supraneurals. The latter two characters can be found elsewhere in members of the Apogoninae ( Fraser 1972) . Bergman (2004) partially described and illustrated the cephalic lateralis of C. cladara . Her review of other characters with the cephalic lateralis characters included in a cladistic analysis supported an unresolved clade of Cercamia and Lachneratus and excluded Rhabdamia and Verulux Fraser, 1972 . Also see Ida & Moyer (1974) for comparative illustrations of cephalic canals, pores and free neuromasts of Rhabdamia that support excluding a close relationship with Cercamia . Mabuchi et al. (2014) hypothesized that Pseudamiops (morphology only, no pored lateral-line scales, cycloid scales, free neuromasts on body and caudal fin, free neuromasts from the snout to nape in simple linear lines across the head, simple cephalic pores, translucent body) was a member of the Lachneratus clade rather than a member of the Pseudamiinae . However, Sato et al. (2019; 2021) presented a detailed description of the lateral-line systems for C. eremia , Pseudamia gelatinosa Smith, 1956 , Pseudamiops cf. springeri Gon, Bogorodsky & Mal, 2013 , and Gymnapogon japonicus Regan, 1905 . Sato et al. (2021) firmly placed Pseudamiops in the Pseudamiinae based on nerve characteristics and a molecular analysis rather than the Apogoninae . Analysis of gene sequences of C. eremia from Fiji (KUIT 4270) and Saipan (KUIT 5685) suggested possible different species despite using the name C. eremia for both sequenced gene fragments ( Thacker & Roje 2009). In their analysis Cercamia appears paraphyletic as the two specimens identified as members of Gymnapogon (both identified as “ G. urospilotus ”) were recovered within the joint clade as a sister to one of the two “ C. eremia ” specimens. However, information on morphological examination and the characters used to identify specimens as “ C. eremia ” and “ G. urospilotus ” are not provided in Thacker & Roje (2009).

Molecular phylogenetic analyses provided by Mabuchi et al. (2014), using two mitochondrial and two nuclear DNA fragments, found high support for an evolutionary clade including Lachneratus and Cercamia , which, in their analysis, is not obviously closely related to Rhabdamia and Verulux . Mabuchi et al. (2014) suggested that the species of Cercamia formed a sister relationship with two sequences of identified B-marked Gymnapogon . One of these vouchers (KUIT 4250) collected in Fiji is nested within Cercamia spp. and is named Cercamia sp. 3 herein. In addition, Mabuchi et al. (2014) included two molecularly distinct lineages, both identified as “ C. cladara ” ( Palau, ROM T04050 and Gambier Archipelago, USNM 399590), which emerged as another possible species identification issue. Recently, Allen et al. (2015) described Cercamia melanogaster from West Papua, eastern Indonesia. They suggested a sister relationship with C. cladara based on molecular inferences, but they did not include all available data from previous phylogenies. Their maximum likelihood tree for COI data includes two specimens identified as “ C. eremia ” that show an unusually high degree of divergence from the residual members of the genus and from each other, again prompting questions regarding the validity of the identification of the voucher specimens.

Baldwin & Johnson (1999) hypothesized that their new genus Paxton is a member of the Pseudamiinae with Gymnapogon as its sister genus. Baldwin & Johnson (1999) concluded that Cercamia was not a member of the Pseudamiinae . Bergman’s (2004) evidence and analysis agreed with the Gymnapogon — Paxton hypothesis. She noted that Pseudamiops , Gymnapogon and Paxton shared a single lachrymal pore while all other apogonids have two pores. Cercamia and Lachneratus were an unresolved clade for Bergman’s analyses. Mabuchi et al. (2014) recovered Pseudamia Bleeker, 1865 as the sole representative of the Pseudamiinae . Paxton (morphology only) was hypothesized to be the sole representative of Paxtoninae. Lachneratus and Cercamia formed a well-defined clade within the Apogoninae [One voucher of Gymnapogon reported in that molecular clade was misidentified and is a species of Cercamia ].

In conclusion, Cercamia is in a clade with Lachneratus and Gymnapogon , while Pseudamiops is in the clade with Pseudamia , and the relationships of Paxton remain puzzling.