Redescription of the Deepwater Freckled Stargazer, Xenocephalus egregius (Uranoscopidae)
The Freckled Stargazer, Xenocephalus egregius, is a member of the family Uranoscopidae and the only member of the genus Xenocephalus found in the western Atlantic Ocean. Xenocephalus egregius has a complicated taxonomic history because of confusion identifying specimens due to ontogenetic changes. In this paper, we redescribe X. egregius based on 60 specimens, detailing ontogenetic changes in anatomy and color to improve identification of this species. We find dramatic changes in color and body pattern as the species grows, changing from blue and green streaks to yellow gold with brown freckles. Changes in anatomy, such as the size of the dentary flange and preopercular flange, are also evident. Using museum collections and new specimens, we evaluated the species’ geographic and depth distribution. We expand the range of X. egregius to include the southern Gulf of Mexico off the Yucatan Peninsula to the northwestern Atlantic Ocean near New Jersey. We also found depth-related differences between specimens collected in the Gulf of Mexico and those collected in the northwestern Atlantic Ocean. Finally, we sequenced the full mitochondrial genome of a specimen of X. egregius and use the cytochrome oxidase subunit 1 (COI) barcode region to compare specimens of X. egregius across the newly expanded range, as well as to other members of the Uranoscopidae, to test the monophyly of the genus. Using COI data, we found that all specimens of X. egregius examined are identical, and comparisons to other uranoscopids resolve both the species and genus as monophyletic. Nearly all specimens examined were collected on fisheries surveys, and thus our study highlights that such surveys not only inform fisheries management but contribute to building natural history collections and studies of biodiversity.
XENOCEPHALUS (Teleostei: Uranoscopidae) includes five species, four that occur in the western Pacific Ocean, between Japan and New Zealand, and one that occurs in the western Atlantic Ocean (Kishimoto, 1989; Roberts et al., 2015). The name Xenocephalus may allude to a defining trait of the genus, a pair of converging bony flanges that project anteriorly from the dentary (Kishimoto, 1989; Pietsch, 1989; Vilasri, 2013). Other characters that diagnose the genus include a large process on the ectopterygoid, a reduced cleithral spine, and a blunt posteriorly directed spine on the opercle (Vilasri, 2013). Species of Xenocephalus tend to be found deeper than other uranoscopids, and while sometimes found in shallow waters (<100 m), most specimens occur between 200–400 m (Berry and Anderson, 1961; Kishimoto, 1989). Formerly, the genus was classified as Gnathagnus, described by Gill (1861), but was reclassified to Xenocephalus in 1994 by Springer and Bauchot. After reexamining the holotype of Xenocephalus armatus Kaup, 1858, which was initially placed in the Gadidae, Springer and Bauchot (1994) recognized that the species should not only be classified in the Uranoscopidae, but also had overlapping diagnostic features with the genus Gnathagnus. Thus, Xenocephalus Kaup, 1858 took priority over Gnathagnus Gill, 1861 and the five species were reclassified.
Only one species of Xenocephalus occurs in the Atlantic Ocean, X. egregius, also known as the Freckled Stargazer for the spotted pattern on the body of adults. This species can be distinguished from other species of Xenocephalus by a lower anal-fin-ray count, a lower anal-pterygiophore count, a laterally projecting flange on the ventral part of the preopercle that reduces with growth, and geographic distribution (Berry and Anderson, 1961; Kishimoto, 1989; Vilasri, 2013). Xenocephalus egregius has a complicated taxonomic history both because of changes in the generic classification described above and dramatic ontogenetic shifts. In 1905, Jordan and Thompson described Execestides egregius based on a 50 mm standard length (SL) specimen captured on a reef near Garden Key, Tortugas, Florida. Jordan and Thompson (1905) allied the new taxon with Kathetostoma in the Uranoscopidae but noted differences in the length of the preopercle and armoring of the head. Later, Longley and Hildebrand (1940) described Benthoscopus laticeps based on a 205 mm SL specimen collected south of Tortugas, Florida. They designated a new genus for their taxon because scales covered most of the specimen’s body and noted the development of prominent bony ridges on the dentary (= dentary flange; Pietsch, 1989). In 1946, Myers classified B. laticeps into the western Pacific genus Gnathagnus. In A List of Common and Scientific Names of Fishes from the United States and Canada (Bailey et al., 1960), the name Arioscopus [sic] egregius is used to refer to the species without reference. It is likely that Arioscopus is a misspelling of Ariscopus, a genus found in Japan that was recognized 26 years later as a synonym of Gnathagnus (Amaoka et al., 1986). Berry and Anderson (1961) recognized that E. egregius and G. laticeps were different ontogenetic stages of the same taxon and synonymized the species under G. egregius. Today, the species is classified as X. egregius.
While X. egregius has been included in various guidebooks, we lack a complete understanding of the ontogenetic changes, geographic distribution, and depth range of the species. Although some ontogenetic changes in coloration and size of the preopercular flange were noted by Berry and Anderson (1961), ontogenetic series have not been well described. This is particularly relevant to the identification of X. egregius because guidebooks that include the species, such as McEachran and Fechhelm (2005) and Kells and Carpenter (2011), illustrated the juvenile, while others, such as Carpenter (2002) and Robins and Ray (1986), illustrated the adult; none describe ontogenetic changes. Berry and Anderson (1961) described the range of the species from northern Georgia to southern Texas based on specimens collected during fishery surveys and deposited in museum collections. Later, Smith-Vaniz et al. (1999: 303) reported a specimen from Bermuda, but doubted “a resident population … occurs in Bermuda.” In 2003, Moore et al. reported a northern range extension based on a single specimen collected in Block Canyon, east of New Jersey. Kells and Carpenter (2011) do not include the Moore et al. (2003) northern record, describing the range as Georgia to Florida, northwestern and southern Gulf of Mexico, and possibly Bermuda. Kells et al. (2022) did not include the species despite covering fishes from southern Texas to Yucatán, Bahamas, Bermuda, and the Caribbean Sea. These guides also suggest that juveniles and adults live at different depths, based on the observations of Berry and Anderson (1961), but this hypothesis has not been tested for statistical significance.
Using specimens of X. egregius in natural history museums and records from fisheries surveys by the National Oceanic and Atmospheric Administration (NOAA) and predecessor agencies, we can improve our understanding of the ontogeny and geographic and depth distribution, and provide clarification and synthesis of information in previous works for this species. In this study, we redescribe X. egregius with detailed descriptions of ontogenetic changes in color and morphology, highlighting important characteristics for the identification of this species. We also expand the geographic range and depth distribution of the species and test for a relationship between collection depth and specimen size as proposed by Berry and Anderson (1961). Finally, we generate a complete mitochondrial genome for X. egregius based on a vouchered specimen (USNM 466090). Combining our data with previously generated sequences of cytochrome oxidase subunit 1 (COI), we compare specimens of X. egregius sampled from Cape Hatteras, North Carolina, USA, western Florida, USA, and western Tabasco, Mexico, and generate a phylogeny to test the monophyly of the species and genus.
MATERIALS AND METHODS
Abbreviations and terminology.—
Museum collection codes follow Sabaj (2020). Terminology for morphology follows Pietsch (1989).
Specimens.—
Specimens of X. egregius were examined from Block Canyon (northern New Jersey, USA) to the Yucatán Peninsula, Mexico. For the northern range extension of North Carolina to New Jersey, eight new specimens of X. egregius were collected on NOAA Ship Henry B. Bigelow during NOAA Northeast Fisheries Science Center (NEFSC) Ecosystems Surveys Branch Bottom Trawl Survey from Cape Hatteras to the Gulf of Maine (Grosslein, 1969; Azarovitz, 1981; Despres-Patanjo et al., 1988; Link et al., 2008; Politis et al., 2014; Galbraith et al., 2022). Two of these specimens were deposited at the Museum of Comparative Zoology (MCZ), Harvard University (MCZ 172404 and MCZ 172411), and six at the National Museum of Natural History (NMNH), Smithsonian Institution (USNM 466014, USNM 466090, USNM 477753, USNM 477754, USNM 477755, USNM 477756). The field identifications of these eight NEFSC Bottom Trawl Survey specimens were variable. MCZ 172404 and USNM 466090 were identified only to “Uranoscopidae,” USNM 477756 was identified to “Astroscopus,” and USNM 477754 and USNM 477755 were identified as “Astroscopus guttatus.” Only USNM 466014 and MCZ 172411 were identified correctly as X. egregius. We also reviewed records of X. egregius in the Global Biodiversity Information Facility (https://www.gbif.org/) for northern specimens outside the Gulf of Mexico and identified two additional specimens, including one collected by a fishing vessel in 1993 (USNM 328522) and one collected by a fisheries observer on a commercial trawling vessel in 2006 (MCZ 166028). For the southern range expansion, six specimens of X. egregius were collected off the coast of Mexico by the B/O Justo Sierra and deposited at the Universidad Nacional Autónoma de México Fish Collection (CNPE-IBUNAM). One specimen was collected from northern Veracruz in 2002 (CNPE-IBUNAM 13160), four from northern Tamaulipas in 2008 (CNPE-IBUNAM 17970), and one from western Tabasco in 2009 (CNPE-IBUNAM 16521). Photographs of these specimens were sent to us by Dr. Eloísa Torres Hernández to confirm the identification and collect measurements. We also reviewed a single specimen of “X. egregius” collected from Puerto Rico at the California Academy of Sciences (CAS); however, we reidentified the specimen (CAS 61045) as Kathetostoma cubana and thus it is not further included in our analyses. We examined an additional 44 specimens of X. egregius from the original range in the northern Gulf of Mexico (72–307 mm SL), including the type specimen of B. laticeps (USNM 108879) and photographs of the type specimen of E. egregius (CAS 8411). We also confirmed the identification of the specimen of X. egregius collected in Bermuda (Bermuda Aquarium, Museum and Zoo [BAMZ] 1998-169-008) from photographs published by Smith-Vaniz et al. (1999). Comparative material for the genus was based on eight specimens representing three of the four other species of Xenocephalus (X. armatus, X. australiensis, and X. elongatus). Material for X. cribratus was unavailable to us, and the description here is based on data from Kishimoto (1989). Additional comparative material included representatives of all valid genera of the Uranoscopidae with a focus on species occurring in the Americas.
Xenocephalus egregius.—
CAS 8411 (50 mm SL), CNPE-IBUNAM 13160 (315 mm SL), CNPE-IBUNAM 16521 (303 mm SL), CNPE-IBUNAM 17970 (93, 95, 162, 193 mm SL), FSBC 023885 (239 mm SL), MCZ 166028 (157 mm SL), MCZ 172404 (96 mm SL), MCZ 172411 (335 mm SL), USNM 108879 (205 mm SL), USNM 157903 (205 mm SL), USNM 158393 (295 mm SL), USNM 158623 (125, 260, 307 mm SL), USNM 159651 (110 mm SL), USNM 184964 (115 mm SL), USNM 186209 (75 mm SL), USNM 186211 (84, 141 mm SL), USNM 186212 (94 mm SL), USNM 186213 (80 mm SL), USNM 186214 (165 mm SL), 186215 (131, 235 mm SL), USNM 186216 (250 mm SL), USNM 186217 (200 mm SL), USNM 186218 (210 mm SL), USNM 186219 (245, 245 mm SL), USNM 187900 (133, 217 mm SL), USNM 188012 (255, 275 mm SL), USNM 188226 (250, 250 mm SL), USNM 188231 (190, 255 mm SL), USNM 188281 (240, 275 mm SL), USNM 268438 (86 mm SL), USNM 268440 (200 mm SL), USNM 286441 (72 mm SL), USNM 268443 (163 mm SL), USNM 268445 (79 mm SL), USNM 268447 (233 mm SL), USNM 328522 (239 mm SL), USNM 375528 (113, 235 mm SL), USNM 378464 (235 mm SL), USNM 406516 (137 mm SL), USNM 466014 (265 mm SL), USNM 466090 (99 mm SL), USNM 477753 (183 mm SL), USNM 477754 (255 mm SL), USNM 477755 (318 mm SL), USNM 477756 (290 mm SL).
Comparative materials.—
Astroscopus guttatus: USNM 433105 (234 mm SL); A. sexspinosus: USNM 272712 (250 mm SL); A. y-graecum: USNM 159669 (187 mm SL); A. zephyreus: USNM 82721 (248 mm SL); Genyagnus monopterygius: USNM 1177084 (205 mm SL); Ichthyscopus pollicaris: USNM 296633 (177 mm SL); Kathetostoma albigutta: USNM 454757 (160 mm SL); K. averruncus: USNM 421232 (120 mm SL); K. cubana: USNM 374785 (79, 176 mm SL); Pleuroscopus pseudodorsalis: SIO 97-66 (246 mm SL), UWFC 21181 (541 mm SL); Uranoscopus bicinctus: USNM 437713 (204 mm SL); U. turbisquamatus: NMNZ P.027134 (240 mm SL); Xenocephalus armatus: USNM 176814 (280 mm SL), USNM 177051 (138, 200, 250, 264 mm SL); X. australiensis: USNM 175015 (182 mm SL); X. elongatus: USNM 296639 (244 mm SL).
Ontogenetic analysis.—
To compare changes in morphology of X. egregius at different sizes, we compared physical specimens, photographs, radiographs, and micro-computed tomography (μCT) scans of museum specimens. Radiographs were taken using a DURASCAN 1417 digital x-ray system at NMNH. Specimens of two X. egregius (USNM 328522 and USNM 466090) were scanned using a GE Phoenix v|tome| × M 240/180 kV Dual Tube μCT at NMNH to compare changes in the skeletal morphology. Descriptions of fresh coloration were based on two specimens (USNM 466014 and USNM 466090) photographed just after collection onboard NOAA Ship Henry B. Bigelow in September 2022.
Standard length was measured from the anterior margin of the dentary flange to the posterior margin of the hypural plate. Total length was measured from the anterior margin of the dentary flange to the posterior margin of the caudal-fin rays. Preopercular-flange length was measured with calipers from the proximal origin to the distal point of the flange along the anterior margin (i.e., greatest length of the flange). Dentary-flange length was measured with calipers from the proximal origin to the distal point of the flange along the antero-dorsal margin. Predorsal length was measured from the anterior margin of the dentary flange to the origin of the dorsal fin.
Distribution analyses.—
We used 60 specimens of X. egregius and the published record from Smith-Vaniz et al. (1999) to evaluate the geographic range of the species. We also found a trawl bycatch record from Derek P. Jones in Nova Scotia, Canada, in April 2015, in the form of a video of a specimen misidentified as an Atlantic Stargazer (Uranoscopus scaber) posted to the YouTube channel “D P Jones,”: https://youtu.be/9g4kN6hnkKI?si=gRh20n_EDBCfqukV. We were unable to verify the collection location for this specimen but note the possibility that X. egregius may be found farther north.
Of the 60 specimens available, 54 had collection depth recorded. To test the relationship between SL and collection depth, we compared three sets of data: (1) all records combined (n = 54); (2) a separate analysis of specimens collected north of North Carolina (n = 10); and (3) a separate analysis for Gulf of Mexico specimens (n = 44). For each group, we report the range and average collection depth and calculate the Pearson correlation coefficient and the significance of the relationship. Finally, we tested if the month of collection was correlated with collection depth to assess possible seasonal movements. The juvenile Bermuda record from Smith-Vaniz et al. (1999: 303) was excluded from depth analyses because it was reported to be “swimming on surface,” likely not yet having settled out of the water column.
Mitochondrial genome generation and annotation.—
We extracted genomic DNA from a single specimen of X. egregius (USNM 466090). Extractions were made on an AutoGenPrep 965 (AutoGen, Holliston, MA, USA) following the manufacturer’s extraction protocol. The library was prepared using enzymatic shearing and other methods described in Hoban et al. (2022) and sent to Admera Health Biopharma Services for sequencing on an Illumina NovaSeq 6000 platform. Demultiplexed sequence data received in compressed FASTQ format were cleaned of adapter contamination and low-quality bases using fastp (Chen et al., 2018; Chen, 2023). Cleaned reads can be downloaded from GenBank (SRA SRR25465036 and BioProject PRJNA720393). We assembled the mitochondrial genome using the ‘map to reference’ function in Geneious version 11.1.5 (Kearse et al., 2012) with the settings described in Girard et al. (2022) and a reference mitogenome downloaded from GenBank (X. elongatus, OL944337). The assembled mitogenome was annotated using MitoAnnotator (Iwasaki et al., 2013; Sato et al., 2018; Zhu et al., 2023) and submitted to GenBank (PV166463).
Genetic comparisons and phylogenetic analysis.—
To identify if other samples of X. egregius had been uploaded to public repositories, we submitted the barcode region of COI from USNM 466090 to GenBank BLAST and the BOLD SYSTEMS Identification Engine “All Barcode Records on BOLD” v.4 dataset (Ratnasingham and Hebert, 2007). We identified two matching specimens, one from the northern Gulf of Mexico (FSBC 23885, off Florida, USA) and one from the southern Gulf of Mexico (CNPE-IBUNAM 16521, off western Tabasco, Mexico). We used the three samples of X. egregius to compare genetic similarity in COI across the species’ range. To compare X. egregius to other species in the Uranoscopidae, we used publicly available COI data for representatives of each genus that we could identify to species (Supplemental Table 1; see Data Accessibility). Sequences were aligned in Geneious Prime using MAFFT v7.490 (Katoh and Standley, 2013). IQ-TREE version 2.2.0 was used to partition the data (i.e., MFP; Chernomor et al., 2016; Kalyaanamoorthy et al., 2017; Minh et al., 2020), and we recovered an optimal partitioning scheme of three models for each of the three codon COI positions based on BIC: 1—TIMe+I+G4; 2—TN+F+G4; and 3—K3Pu+F+I. Ten tree searches were performed in IQ-TREE using the optimal partitioning scheme. We generated 1,000 bootstrap replicates for node support. Analyses were rooted on two species of the Ammodytidae, Ammodytes dubius (USNM 438477) and Hyperoplus immaculatus (ZMH 200179), selected because of the family’s hypothesized relationship with the Uranoscopidae (Pietsch, 1989; Betancur-R et al., 2017).
Xenocephalus egregius (Jordan and Thompson, 1905)
Freckled Stargazer (English)
Miracielo Pecoso (Spanish)
Figures 1–6, Supplemental Figure 1, Supplemental Tables 1–2
Execestides egregius Jordan and Thompson, 1905: 253, figs. 5–6 (holotype: CAS 8411; Garden Key, Dry Tortugas, Florida, U.S.A.); Longley and Hildebrand, 1941: 244 (description and identification key); Hildebrand, 1954: 318 (collection record and description; erroneously under family Brotulidae).
Benthoscopus laticeps Longley and Hildebrand, 1940: 264, figs. 20–21 (holotype: USNM 108879; Tortugas, Florida, U.S.A.); Longley and Hildebrand, 1941: 244 (description and identification key).
Gnathagnus laticeps. Myers, 1946: 42 (synonymy of genus).
Arioscopus egregius. Bailey et al., 1960: 42 (common name of species); Staiger, 1970: 61 (listed as found or collected).
Gnathagnus egregius. Berry and Anderson, 1961: 580 (detailed description); Herrema, 1974: 107 (regional fishes guide); Walls, 1975: 300 (description and regional fishes guide); Berry, 1978: 2 (regional fishes guide and identification key); Finucane et al., 1979: 81, 117 (listed as found or collected); Ditty et al., 1988: 815, table 1 (listed; erroneously under family Opistognathidae); Robins and Ray, 1986: 220, pl. 42 (regional fishes guide); Potts and Ramsey, 1987: 78 (regional fishes guide); Kishimoto, 1989: 313 (comparative morphology of genus); Pietsch, 1989: 295 (phylogenetic analysis of genus); Boschung, 1992: 158 (listed); McEachron et al., 1994: 125, table 1 (listed as found or collected during trawls); Hoese and Moore, 1998: 226 42 (regional fishes guide); Smith-Vaniz et al., 1999: 303 (collection record, description, and photograph); Carpenter, 2002: 1747 (regional fishes guide and identification key); Nelson et al., 2004: 161 (common name of species); McEachran and Fechhelm, 2005: 570 (regional fishes guide); Sulak et al., 2007: 88, table 3a (listed as found or collected); Rudders and DuPaul, 2009: 17, table 8 (listed as found or collected); Kells and Carpenter, 2011: 325 (description, habitat, biology, and illustration).
Xenocephalus egregius. Springer and Bauchot, 1994: 79 (synonymy of genus); Moore et al., 2003: 227 (northern range extension); Brodziak et al., 2004: A8, table A4 (listed); Page et al., 2013: 166 (names of fishes record); Smith-Vaniz and Collette, 2013: 180 (updated name from Smith-Vaniz et al., 1999); Vilasri, 2013: 91 (phylogenetic analysis of genus); Snyder and Burgess, 2016: 251 (listed); Ramírez et al., 2019 (listed); Page et al., 2023: 176 (Spanish common name); Martinez et al., 2025: table 7 (listed as found or collected).
Diagnosis.—
Dorsal-fin rays 12–14, anal-fin rays 16–17, pectoral-fin rays 21–23 (20–24 in Berry and Anderson [1961]), caudal-fin rays 5–6 + 5–6, vertebrae 28 (11 precaudal + 17 caudal). A species of Xenocephalus that has a preopercular flange that reduces with growth, the presence of a pterotic tubercle (Kishimoto, 1989; Pietsch, 1989; Vilasri, 2013), and a greatly reduced posterior nostril that resembles a lateral-line pore. Xenocephalus egregius can be differentiated from all other species of Xenocephalus examined, except X. armatus, based on the presence of a thin projecting cirrus of the anterior nostril (thick in X. australiensis and X. elongatus). Xenocephalus egregius can be further differentiated from X. australiensis and X. elongatus by a longer predorsal region (>60% of SL) and from X. australiensis by the presence of two vomerine tooth plates (X. australiensis has one; Kishimoto, 1989; Vilasri, 2013). Xenocephalus egregius can be differentiated from X. armatus and X. cribratus by having 28 vertebrae (X. armatus and X. cribratus have 27 vertebrae; Kishimoto, 1989). Xenocephalus egregius can additionally be distinguished from X. armatus by large, well-spaced sensory pore openings (X. armatus has clusters of small sensory pore openings).
Description.—
Dentary flange small, projecting straight anteriorly in specimens <150 mm SL, becoming more pronounced and curved toward midline anteriorly with growth (Figs. 1A, 2, 3). Body of small specimens (<75 mm SL) rectangular, becoming more elongate in larger specimens. Scales of smaller specimens (<100 mm SL) present but weak, embedded. Scales visible along trunk, between lateral line and dorsal fin by ∼150 mm SL. Head about half of body length in ∼50 mm SL specimens, becoming a third of body by 75 mm SL. Eyes become more dorsal and smaller relative to the head as the species grows. Large preopercular flange on ventral margin projects laterally, about half the width of the neurocranium, in holotype (50 mm SL); greatly reduced with growth to small nob in specimens >250 mm SL (Figs. 1–3). Opercle with large shelf spanning the dorsal margin reduced with size to small bump in specimens >275 mm SL (Fig. 2). Subopercle with large lateral knob projecting ventral to opercle, reducing with growth. Anguloarticular with anteriorly directed ventral spine and lateral shelf that forms the ventral-most edge; both reduce with size. Frontal with strong bony tubercles posterior to the eye, reducing with growth and lost around 200 mm SL (Fig. 2). Pterotic with tubercle that projects dorsolaterally and decreases in size with growth. Parietal with tubercle on posterior margin, reducing in size until barely visible in specimens around 200 mm SL.
Citation: Ichthyology & Herpetology 113, 3; 10.1643/i2024112
Citation: Ichthyology & Herpetology 113, 3; 10.1643/i2024112
Citation: Ichthyology & Herpetology 113, 3; 10.1643/i2024112
Color in fresh specimens.—
In smaller 99 mm SL specimen, dorsal head and body green-gold with small melanophores and steel blue streaks and vermiculations (Fig. 1B). Ventral surface white. Laterally, head has areas with dense green-brown pigment on pale white background, giving a speckled appearance. Speckled green-brown coloration extends to the oral valve of the dentary. Cheek has large unpigmented regions. Dorsal fin green-gold at base, becoming brown, then gray with broad white margin. Anal fin white with pigment medially. Caudal fin green-gold becoming dark before broad white band at distal margin. Pectoral fin white at base, becoming dark brown with a wide white margin. Pelvic fins white. Lateral line white and conspicuous. Eyes with golden ring with evenly placed dark regions; pigment distributed over much of tissue covering the eye.
In larger 265 mm SL specimen, dorsal head and body yellow-green-tan with darker brown patterning; pattern stops abruptly, becoming whitish-gray along mid-lateral surface from behind pectoral fin to base of caudal fin (Fig. 1C). Pattern on dorsal surface becomes more spotted and less streaked (Fig. 1C, D). Bones on dorsal and lateral surface of head become darker than surrounding yellow-green background color. Spotted pattern of body extends onto the oral valve of the dentary. Cheek fully pigmented, like body. Base of the dorsal, caudal, and pectoral fins dark brown and black with white margin. Anal fin white. Lateral line less conspicuous than in smaller specimens. Eyes gold with small brown spots on tissue covering eye.
Color in preserved specimens.—
Holotype (CAS 8411, 50 mm SL), color faded. Pale streaks along body. Pectoral, dorsal, anal, and caudal fins darker at the base and become white distally. Lateral line pale and conspicuous (Fig. 1A). Other preserved specimens brown to tan with darker brown markings related to body pattern in life. Body pattern changes gradually from fully streaked/vermiculated pattern to fully spotted/freckled pattern with growth (Figs. 1A, D, 2, 3). Streaks found in specimens between 50–150 mm SL, mix of both streaks and spots in specimens from 75–300 mm SL, only spots in specimens 200–335 mm SL (Fig. 3).
Size.—
To at least 335 mm SL (400 mm TL; MCZ 172411).
Distribution.—
Xenocephalus egregius occurs from northern New Jersey, USA, to the Yucatán Peninsula, Mexico, predominantly along the continental shelf (Fig. 4). The species may also occur east to Bermuda, but only one specimen has been collected (BAMZ 1998-169-008). It is possible that X. egregius also occurs farther north than we report in this study based on a video a fisherman posted to the YouTube channel “D P Jones” of a specimen of X. egregius collected during a trawl in Nova Scotia, Canada. This record is not confirmed and therefore represented on the map as a question mark (Fig. 4).
Citation: Ichthyology & Herpetology 113, 3; 10.1643/i2024112
Based on the collection depth of 54 specimens, the range was 146–493 m, with an average collection depth of 313 ± 99 m. The ten specimens from north of North Carolina had a collection depth range of 208–288 m, with an average collection depth of 232 ± 27 m. For the 44 specimens collected in the Gulf of Mexico, the collection depth ranged from 146–493 m, with an average of 331 ± 100 m.
When comparing SL to collection depth for all specimens, our Pearson correlation coefficient was 0.52 with a P-value of <0.0001 (Fig. 5). We found the northern records had the same Pearson correlation coefficient as the overall dataset (0.52), but the P-value was not significant (P > 0.05); in contrast, the Gulf of Mexico records had a Pearson correlation coefficient of 0.71 that was significant (P < 0.001).
Citation: Ichthyology & Herpetology 113, 3; 10.1643/i2024112
We did not find a relationship between month of collection and collection depth that would indicate seasonal movements (r = 0.26, P > 0.05).
Mitogenome.—
The complete mitogenome for X. egregius is 16,626 base pairs in length (Supplemental Fig. 1; see Data Accessibility). The base composition is 27.6% A, 25.6% T, 16.9% G, and 29.9% C. It encodes 37 mitochondrial loci (13 protein-coding loci, 22 tRNAs, and 2 rRNAs) and one non-coding control region (D-loop). Of these, 26 loci are on the majority strand (J strand), and the remaining nine are on the minority strand (N strand). All protein-coding genes start with the ATG codon except for ND2, which starts with ATA, and COI, which starts with GTG. The number and order of loci are the same as in other mitochondrial genomes for species of uranoscopids that are publicly available (e.g., Ichthyscopus pollicaris [NC_068019], Kathetostoma albigutta [NC_083034], Uranoscopus cognatus [NC_043875], and Xenocephalus elongatus [NC_068020]).
Phylogenetic analysis.—
Comparison of the barcode region of COI shows USNM 466090 is identical to the sequences of X. egregius from northern (FSBC 23885, GenBank PP390551) and southern (CNPE-IBUNAM 16521, BOLD CBPM126-11) Gulf of Mexico. Our phylogenetic analysis resulted in a single optimal tree (Fig. 6) and has a score of –5125.400. We recover all genera in the Uranoscopidae as monophyletic with high support values. Xenocephalus is recovered in a clade that includes Kathetostoma and Pleuroscopus. Xenocephalus egregius was recovered as sister to X. cribratus. While the bootstrap values are high for genus-level support, they are poor for the relationships between the genera.
Citation: Ichthyology & Herpetology 113, 3; 10.1643/i2024112
DISCUSSION
Identification of Xenocephalus egregius.—
Identification of X. egregius has been problematic since the description of two separate species based on different ontogenetic stages. By comparing the morphology of specimens 50–335 mm SL, we found that the appearance of X. egregius changes continuously as it grows. Certain characters, such as the dentary flange, preopercular flange, and body pattern, vary greatly depending on the size of the specimen (Figs. 1–3). However, this variation is often underappreciated in species identification guides. Most identification guides to the western Atlantic and Gulf of Mexico that include X. egregius illustrate it with either the characteristics of a large specimen (e.g., large dentary flange, long body, no preopercular flange, spotted body pattern) or a small specimen (e.g., small dentary flange, short body, large preopercular flange, streaked/vermiculate body pattern) without additional information on the ontogenetic changes (e.g., Carpenter, 2002; McEachran and Fechhelm, 2005; Kells and Carpenter, 2011). One problem with this can be seen in guides such as Robins and Ray (1986), Hoese and Moore (1998), and Kells and Carpenter (2011) that use the presence of a preopercular flange for identification even though we show this character is greatly reduced or sometimes lost completely in X. egregius > 200 mm SL. There also appears to be confusion regarding the terminology for some anatomical structures. Hoese and Moore (1998: 226) describe X. egregius (then called G. egregius) as having “winglike projections from the lower edge of the opercular membrane.” This description is incorrect because the “winglike projections” are a lateral flange of the bone of the preopercle and positioned anterior to the opercular membrane (Figs. 1, 2). Changes in the body pattern and coloration of X. egregius are also not well discussed in the literature, likely due to the lack of documentation of fresh specimens. Only Smith-Vaniz et al. (1999) include a photograph of X. egregius, but this was published in black and white and of only one specimen (40 mm SL). In this study, we found that the body of the species can range from a dark green and blue streaked pattern in smaller specimens (99 mm SL) to a golden-brown with brown spots in larger specimens (265 mm SL; Figs. 1, 2). The only guide with a color illustration is Kells and Carpenter (2011); however, the illustration has the anatomy of a small specimen (head with many tubercles, large preopercular flange, streaked/vermiculate pattern, short body) but the coloration is tan and brown, more like the colors of a large individual or a preserved specimen. The variety of ways to describe and depict X. egregius is not surprising given the continuous ontogenetic changes the species undergoes (Figs. 1–3).
Maximum length of the species also confounds identification. Berry (1978) reported the maximum length of X. egregius as 330 mm TL, and this measurement has been used in guides since (e.g., Robins and Ray, 1986; Kells and Carpenter, 2011). Based on MCZ 172411, a specimen collected during the 2014 fall NEFSC Bottom Trawl Survey, we increase the known maximum size to 400 mm TL (335 mm SL). In addition, we found eight other specimens with a TL greater than 330 mm that includes specimens from across the Gulf of Mexico, as well as two other specimens collected in the western Atlantic, indicating the species grows larger than 330 mm TL throughout its geographic range.
There are some gaps in our understanding of the species morphology that this study was not able to address. Due to having fresh-coloration photographs of only two specimens, we can identify major color changes between 99 mm SL and 265 mm SL, but lack information on the gradient between those sizes and how this color might change as the species grows past 265 mm SL. Despite reviewing extensive larval collections at NMNH, the Nunnally Ichthyology Collection at the Virginia Institute of Marine Science, and specimen databases, we did not find any larval records of X. egregius, and thus the larval form remains to be described (Fritzsche, 1978; Richards, 2005).
Depth distribution.—
All specimens collected north of North Carolina, USA, were collected between 208–288 m, which was shallower than 68% of the Gulf of Mexico specimens (Fig. 5). The difference in collection depth may be because this species occurs shallower in the northern part of its distribution, perhaps following an isotherm (Colvocoresses and Musick, 1984), or the result of sampling bias because depths of >400 meters are not commonly sampled in the northern part of the range.
Specimen size was moderately correlated with collection depth when comparing all specimens (r = 0.52, P < 0.0001). When we separated the specimens based on collection locality, those collected north of North Carolina had no significant relationship between size and collection depth (r = 0.52, P > 0.05); instead, the result for all specimens is driven by the stronger relationship between size and collection depth in the Gulf of Mexico specimens (r = 0.71, P < 0.0001). While there is a strong relationship in the Gulf of Mexico, specimens ranging from 86–307 mm SL were collected at 350–400 m, indicating that smaller specimens can also occur at deeper depths (Fig. 5).
In comparison to other stargazers that overlap in geographic range, X. egregius is found further offshore and deeper along the continental slope from around 140–500 m (Figs. 4, 5). Astroscopus guttatus and A. y-graecum occur coastally in 0–100 m and Kathetostoma albigutta occurs in 40–300 m (Berry and Anderson, 1961; NEFSC, unpubl. data).
Monophyly of Xenocephalus.—
Identification trouble due to ontogenetic change has occurred not only in X. egregius, but also in X. elongatus, the Bluespotted Stargazer from Japan, where the adult was described as Uranoscopus elongatus by Temminck and Schlegel (1843) and the juvenile was described by Jordan and Snyder (1902) as Ariscopus iburius. Additionally, species of Xenocephalus all occupy similar depth ranges, and while sometimes found shallow (<100 m), most are found between 200–400 m (Kishimoto, 1989). The monophyly of Xenocephalus is well supported based on morphology and depth distribution (Kishimoto, 1989; Pietsch, 1989; Vilasri, 2013); our analyses of COI sequence data also resolves the genus as a clade (Fig. 6). A morphology-based cladogram comparing the relationships among species of Xenocephalus in Vilasri (2013) shows two sister groups, one containing X. armatus and X. egregius, and one containing X. australiensis and X. elongatus. In the COI tree generated in this study, we found X. armatus and X. elongatus are recovered in a grade leading to a clade of X. cribratus and X. egregius (Fig. 6). However, both our study and Vilasri (2013) have missing taxa; in Vilasri (2013) the tree is missing X. cribratus while our tree is missing X. australiensis. The evolutionary relationships within Xenocephalus have not been tested with genetic data prior to this study. Analyses of the Uranoscopidae that include genetic data for Xenocephalus have used a single species (Smith et al., 2006; Yu et al., 2023; Liu et al., 2024). While we compared COI, our sequencing and annotation of the full mitochondrial genome of X. egregius provides the foundation for future work to build phylogenies based on multiple loci.
Importance of fisheries surveys.—
Without data collected from fisheries surveys, this study would not have been possible. Most of the specimens examined from the Gulf of Mexico were collected by fisheries surveys conducted by Bureau of Commercial Fisheries (NOAA predecessor). These historical records, combined with our new specimens from the NOAA NEFSC Bottom Trawl Survey, highlight that independent fisheries surveys both inform fisheries management and contribute to building natural history collections and studies of biodiversity (e.g., Moore et al., 2003; Bemis et al., 2018; Galbraith et al., 2022). Our photographs, description of ontogenetic changes, and data on geographic and depth distribution of X. egregius will improve the identification of this species, including on fisheries surveys, and thus in turn increase our knowledge of this species’ biology.
DATA ACCESSIBILITY
Supplemental material is available at https://www.ichthyologyandherpetology.org/i2024112. Unless an alternative copyright or statement noting that a figure is reprinted from a previous source is noted in a figure caption, the published images and illustrations in this article are licensed by the American Society of Ichthyologists and Herpetologists for use if the use includes a citation to the original source (American Society of Ichthyologists and Herpetologists, the DOI of the Ichthyology & Herpetology article, and any individual image credits listed in the figure caption) in accordance with the Creative Commons Attribution CC BY License.
AI STATEMENT
The authors declare that no AI-assisted technologies were used in the design and generation of this article and its figures.
Lateral photographs of Xenocephalus egregius showing ontogenetic changes in body shape, color, and pattern. Scale bar is 20 mm. (A) CAS 8411, 50 mm SL, holotype, preserved. Photograph by Jon Fong, CAS Department of Ichthyology, copyright California Academy of Sciences (used with permission). (B) USNM 466090, 99 mm SL, freshly collected. (C) USNM 466014, 265 mm SL, freshly collected. (D) MCZ 172411, 335 mm SL frozen. Photograph by Andrew Williston, Museum of Comparative Zoology, Harvard University, copyright 2024 President and Fellows of Harvard College (used with permission).
Dorsal and anterior photographs of Xenocephalus egregius showing ontogenetic change in bony features. (A) USNM 186209, 75 mm SL. (B) USNM 268443, 163 mm SL. (C) USNM 375527, 275 mm SL. Abbreviations: DEN, dentary; DF, dentary flange; FROT, frontal tubercle; MAX, maxilla; OPS, opercular shelf; PAT, parietal tubercle; PEC, pectoral fin; PEL, pelvic fin; POF, preopercular flange; PTOT, pterotic tubercle.
Scatter plots of changes in morphology relative to growth of Xenocephalus egregius. Top plot shows changes in length of dentary flange across SL. Middle plot shows changes in length of preopercular flange across SL. Bottom plot shows changes in body pattern across SL. Body pattern was separated into three categories: Sp, spotted; Mix, spotted and streaked; and St, streaked. Icons on the right indicate examples of the morphological states.
Distribution of Xenocephalus egregius based on examined specimens (Supplemental Table 2; see Data Accessibility). Green symbols indicate records within the original range given in Berry and Anderson (1961). Blue striped symbols indicate records since Berry and Anderson (1961) that fall outside of the original range. Triangular symbol indicates both type localities because holotypes of Benthoscopus laticeps and Execestides egregius were both collected in Tortugas Florida. The plus sign indicates a single specimen collected by Smith-Vaniz et al. (1999) in Bermuda. The question mark indicates a potential record from a fishing vessel that we were unable to examine.
Standard length vs. collection depth of Xenocephalus egregius based on examined specimens (Supplemental Table 2; see Data Accessibility). Coloration and shape of symbols match Figure 4. Of the holotypes, only the adult (Benthoscopus laticeps) had an associated collection depth record.
Maximum likelihood COI tree for the Uranoscopidae. Tree rooted on two species of the Ammodytidae (Ammodytes dubius and Hyperoplus immaculatus). Bootstrap values less than 50 not shown. For X. egregius: NC, North Carolina; FL, Florida; and MX, Mexico. See Data Accessibility for tree file.
Contributor Notes
Associate Editor: T. Grande.