Morphological and molecular evidence of the Antarctic sleeper shark Somniosus antarcticus (Somniosidae) in northern Chile Authors: Angel Mancilla, Leandro Brizuela, Krishna Tapia, Carolina Vargas-Caro, Carlos Bustamante Published: 2026 ABSTRACT Sleeper sharks (family Somniosidae) are large, slow-growing deep-sea sharks that are rarely documented in the southeastern Pacific, and records from the Chilean waters have remained taxonomically uncertain. Here, we report a confirmed record of a large-bodied southern Somniosus from northern Chile based on a gravid female incidentally captured by a commercial deep-sea longline fishery off Antofagasta and a late-stage male embryo aborted during handling. The embryo measured 90.5 cm in total length and exhibited morphological and meristic characters consistent with large-bodied S. antarcticus, including a cylindrical body, short rounded snout, spineless dorsal fins, hook-like dermal denticles, and a 40-turn spiral valve. Additionally, a 639 bp fragment of the mitochondrial cytochrome C oxidase subunit I (COI) gene was obtained from both the female and embryo, and the two sequences were identical, consistent with maternal inheritance. Comparative analyses using public records placed the Chilean haplotype within the large-bodied Somniosus species complex. The Chilean haplotype was identical to an Antarctic record labelled S. pacificus and differed by two substitutions from a New Zealand record labelled S. antarcticus. However, the COI does not fully resolve species boundaries within the pacificus/antarcticus complex, as the divergence between nominal species remains low. These results provide the first well-documented record of the Antarctic sleeper shark in northern Chile and expand the documented occurrence of southern large-bodied Somniosus along the Chilean margin. More broadly, this record highlights the importance of integrating internal meristics, morphology, and molecular data when documenting rare deep-sea sharks in fishery bycatch. Introduction Sleeper sharks of the genus Somniosus Lesueur 1818 are large, slow-growing, deep-sea sharks characterized by the absence of dorsal fin spines, a heterocercal caudal fin with an elongated upper lobe, and a first dorsal fin originating posterior to the pectoral fin bases (Compagno, 1984; Ebert, Fowler & Compagno, 2021). Adults range from approximately 70 cm in the smallest species to >four m in large-bodied forms and inhabit continental and insular slopes, oceanic plateaus, and seamounts from upper slope depths to >2,000 m in polar, temperate, and some tropical regions (Yano, Stevens & Compagno, 2004; Ebert, Fowler & Compagno, 2021). Yano, Stevens & Compagno (2004) reviewed the systematics of Somniosus and recognized two subgenera based on body size, dermal denticles, spiral valve counts, and vertebral calcification: the large-bodied subgenus Somniosus, including S. microcephalus (North Atlantic and Arctic), S. pacificus (North Pacific and adjacent Arctic), and S. antarcticus (Southern Hemisphere); and the small-bodied subgenus Rhinoscymnus, including S. longus and S. rostratus. A sixth species, S. cheni, was later described in Taiwan within Rhinoscymnus, further underscoring the hidden diversity and taxonomic complexity of this group (Hsu, Lin & Joung, 2020). Historically, large sleeper sharks from the Southern Hemisphere were frequently referred to as S. microcephalus or S. pacificus, and the validity of S. antarcticus has been questioned (e.g., Bass, 1976; Compagno, 1984). Using extensive morphometric and meristic datasets, Yano, Stevens & Compagno (2004) argued that large-bodied southern Somniosus are diagnosable as S. antarcticus, with proposed differences relative to S. microcephalus and S. pacificus including dorsal-fin position and height, prebranchial length, spiral valve counts, dentition, and other internal features. Subsequent biological studies have reported that S. antarcticus can reach at least 4.5–6.0 m in total length and suggested very late maturation in females (≈4.3–4.5 m in females) and a birth size of ∼40 cm (Yano, Stevens & Compagno, 2007). However, the extent to which these nominal taxa represent discrete evolutionary lineages remains an open question. Mitochondrial markers have not fully resolved species boundaries within the large-bodied Somniosus complex in the past. Barcoding and mitogenomic studies often recover shallow divergence and overlapping haplotypes between S. pacificus and S. antarcticus, with some Southern Ocean specimens assigned to one species based on morphological clustering closely with the other species genetically (Benz et al., 2007; Murray et al., 2008; Christensen, 2022; Claassens et al., 2023). Recently, phylogenomic evidence has further highlighted the limited differentiation between S. pacificus and S. antarcticus, raising the possibility that these nominal species may represent a single lineage or an unresolved species complex (Timm et al., 2023). More broadly, limitations in species-level discrimination using COI have also been documented in other elasmobranch datasets, where low divergence among closely related taxa, ambiguity in reference sequences, and conflicts between morphology and barcoding can complicate species assignment, reinforcing the value of integrative approaches (Wong, Shivji & Hanner, 2009; Amini et al., 2025; Khudamrongsawat et al., 2025). Recent observations and range extensions documented by deep-sea cameras and ROVs have often been conservatively reported as Somniosus cf. pacificus or Somniosus sp. (Acero et al., 2018; Claassens et al., 2023), emphasizing the need for integrative morphological and molecular datasets explicitly linked to well-documented specimens and reproducible reference sequence assemblies. Records of Somniosus from the southeastern Pacific, particularly from the waters off Chile, are scarce and taxonomically uncertain. The first documented Chilean specimen was reported as Somniosus cf. pacificus off Valdivia (39°45′S), based on an adult female captured by a deep-sea fishery (Pequeño, Lamilla & Crovetto, 1991). A second record of S. pacificus was reported from San Antonio (33°35′S) in central Chile, based on a stranded male (Brito, 2004). Because these identifications predate the systematic revision of Yano, Stevens & Compagno (2004) and are not supported by genetic data, some Southern Hemisphere records attributed to S. pacificus may instead correspond to S. antarcticus or, more broadly, to a southern large-bodied Somniosus lineage within the pacificus/antarcticus complex (Yano, Stevens & Compagno, 2004; Christensen, 2022; Timm et al., 2023). The southern sleeper shark (S. antarcticus sensu Yano, Stevens & Compagno, 2004) is currently assessed as “Least Concern” on the IUCN Red List; however, population trends are largely unknown, and the species is considered naturally uncommon throughout much of its reported range (Finucci, 2018). It is primarily caught as bycatch in deep-sea longline and gillnet fisheries across its distribution range (Ebert & Mostarda, 2016; Ebert, Fowler & Compagno, 2021). In Chile, deep-sea fisheries for teleosts and invertebrates operate along the continental slope and could interact with sleeper sharks; however, verified records of these interactions and specimens suitable for detailed taxonomic appraisal are rare. Here, we report the occurrence of a large-bodied southern Somniosus in northern Chile, based on a gravid female captured off the coast of Antofagasta and a late-stage male embryo that was aborted during handling. We (1) provide detailed morphometric and meristic information for the embryo, including dentition and spiral valve counts, and (2) generate COI sequences from the female and embryo to assess the placement of the Chilean haplotype within a comparative public dataset of Somniosus species. However, because COI does not fully resolve species boundaries within the S. pacificus/antarcticus complex (Timm et al., 2023), we interpret the molecular evidence as supporting a lineage affinity rather than species delimitation. We discuss the implications of this record for documenting large-bodied Somniosus in the southeastern Pacific and for improving understanding of deep-sea shark bycatch and conservation in northern Chile. Map of collection site. Geographic location of Somniosus antarcticus specimens captured off Antofagasta, northern Chile. The black circle indicates the capture site on the continental slope at ∼22°152S, 70°452W. Materials & Methods Sampling and specimen preservation In August 2023, a large female sleeper shark was incidentally captured by a commercial deep-sea longline fishery operating off Antofagasta, northern Chile (22°15′S, 70°45′W) at a depth of approximately 1,000 m depth (Fig. 1). During handling while the female was being lifted onboard, a single small male embryo was expelled and recovered by the fishers (capture-induced abortion). However, owing to operational constraints and the large body size of the female, the specimen could not be retained completely. Only the caudal fin was retained and preserved for documentation and genetic analyses. The embryo was collected whole and fixed in formalin for taxonomic examination. An umbilical opening/scar was visible and used, together with the capture context, to classify the specimen as a late-stage (near-term) embryo. Sex was determined based on the presence of claspers. The embryo was catalogued as CHALLWA-UA-25 and deposited in the ichthyological collection of the Laboratorio de Biología Pesquera (CHALLWA), Universidad de Antofagasta. Morphological examination Morphological and meristic characters were examined following Yano, Stevens & Compagno (2004), with additional comparative information from Ebert & Mostarda (2016). A total of 64 morphometric measurements were obtained for the embryo using a measuring board and calipers, including natural total length (TL), head proportions, fin origins and dimensions, and trunk and caudal measurements. Measurements are reported as percentages of TL and interpreted in the context of published ranges for large-bodied Somniosus (Yano, Stevens & Compagno, 2004), recognizing the potential ontogenetic and allometric variation. External characters were documented, including overall body shape, coloration, relative dorsal fin size and position, pectoral and pelvic fin morphology, caudal peduncle shape, and caudal fin proportions. Because only the caudal fin was retained from the female, TL could not be measured directly. However, an approximate estimate of female TL was estimated using a proportional relationship between the upper caudal lobe length and TL (Yano, Stevens & Compagno, 2007). Dentition was examined from the jaws, and tooth counts were recorded for both the upper and lower jaws in the functional series, following Yano, Stevens & Compagno (2004). Dermal denticles were examined under a stereomicroscope using a skin sample taken from the lateral trunk below the first dorsal fin; crown shape and cusp morphology were described, noting that denticle morphology may vary with body region and ontogenetic stage. The body cavity of the embryo was opened to expose the internal organs. The spiral valve was removed, and the total number of turns was counted as a diagnostic characteristic; this value was compared with the published ranges for large-bodied Somniosus species (Yano, Stevens & Compagno, 2004; Ebert & Mostarda, 2016). Molecular analysis Genomic DNA was extracted from a small sample of embryo muscle collected prior to fixation and from frozen caudal fin tissue of the female using a standard phenol–chloroform protocol (Sambrook & Russell, 2001). A fragment of the mitochondrial cytochrome c oxidase subunit I (COI) gene was amplified by polymerase chain reaction (PCR) using universal fish barcoding primers, as described by Ward et al. (2005). Amplicons were purified and Sanger sequenced in both directions at the Austral-Omics Sequencing Facility (Valdivia, Chile) using standard protocols. Chromatograms were manually inspected for base-calling errors, and ambiguous sites were verified by comparing forward and reverse reads. Primer regions were trimmed, and a consensus sequence was assembled for each specimen. Female and embryo COI sequences were compared to document the mitochondrial haplotype identity and taxonomic affinity. The identity from the COI analysis was not used to infer kinship; shared haplotypes were reported as consistent with maternal inheritance, whereas the mother–embryo relationship was supported by field observations of capture-induced abortion during handling. Taxonomic identification was conducted using the NCBI BLAST tool against the GenBank nucleotide database. To contextualize the new sequences, a comparative COI dataset was assembled from GenBank by retaining Somniosus records explicitly linked to peer-reviewed publications and/or curated museum specimens. The final dataset included 24 sequences of Somniosus spp. and three somniosid sequences representing other genera used as outgroups (Table S1). Sequences were aligned using MUSCLE in Geneious Prime (version 2025.1.1), trimmed to a final length of 639 bp, and used to estimate pairwise genetic distances using the Kimura two-parameter (K2P) model in MEGA v7 (Kumar, Stecher & Tamura, 2016). A median-joining (MJ) haplotype network (Bandelt, Forster & Röhl, 1999) was constructed using NETWORK v10.2.0.0, with epsilon set to zero. Nodes represent haplotypes (node size proportional to haplotype frequency), branches represent mutational steps, and inferred median vectors are represented by small nodes. For clustering visualization, a distance-based neighbor-joining (NJ) tree was reconstructed using Geneious Prime from the K2P distance matrix, with node support assessed using 1,000 bootstrap replicates. The new COI sequences were deposited in GenBank under accession numbers PX559966 –PX559967. Ethics and permits This study did not involve animal experiments or intentional harm. Specimens were obtained as bycatch from vessels targeting Patagonian toothfish (Dissostichus eleginoides) and processed in accordance with Chilean regulations. All work was conducted under special research permits granted by the Chilean Subsecretaría de Pesca y Acuicultura (E-PINV-2021-598). Results Morphological examination The male embryo measured 90.5 cm TL (Fig. 2). Despite some tissue deformation due to fixation, most external and internal characteristics remained intact and could be reliably assessed. The body was cylindrical and uniformly gray. The snout was short and rounded, the eyes were small, and the spiracles were immediately posterior to the eyes. Two spineless dorsal fins of similar size were present; the first dorsal fin originated well behind the pectoral fin bases, and the second dorsal fin originated over the posterior pelvic fin bases. The pectoral and pelvic fins were relatively small. The caudal peduncle lacked a lateral keel, and the caudal fin was asymmetrical, with a markedly longer dorsal than ventral lobe. The small male was recovered after being aborted by the female during handling, confirming female sexual maturity. An umbilical opening/scar was visible, consistent with a late-stage embryo. Near-term specimen of S. antarcticus. Left lateral view of a near-term male embryo of Somniosus antarcticus (CHALLWA-UA-25) specimen examined for morphometric analysis. Scale bar = 25 cm. Morphometric measurements expressed as percentages of TL are presented in Table 1. Values generally overlapped with the ranges reported for large-bodied southern Somniosus by Yano, Stevens & Compagno (2004), although some proportional characters are known to vary with ontogeny and among specimens. The upper caudal lobe of the adult female measured 47.5 cm (Fig. 3). Using published proportions for large-bodied southern Somniosus (Yano, Stevens & Compagno, 2007), female TL was estimated to fall between approximately 2.2 and 3.0 m. This estimate should be interpreted cautiously, given that it was derived from partial remains. 10.7717/peerj.21381/table-1Table 1Morphometry of S. antarcticus from Chilean waters.Morphometric and meristic measurements of the Chilean male embryo (CHALLWA-UA-25), compared with published ranges for large-bodied Somniosus species from Yano, Stevens & Compagno (2004). Morphometric values are expressed as percentages of total length (TL), except total length (cm) and meristic characters. Comparative values for S. microcephalus, S. pacificus, and S. antarcticus are shown as minimum–maximum ranges. Character CHALLWA-UA-25 S. microcephalus S. pacificus S. antarcticus Total length (cm)90.564.8–480.041.8–430.094.0–438.0 HEAD Snout tip to outer nostrils3.41.3–3.61.3–4.40.9–3.6Snout tip to eye8.25.0–8.75.5–9.04.3–8.9Snout tip to spiracle11.79.0–13.810.0–14.29.1–15.2Snout tip to mouth9.35.4–9.94.9–14.06.7–12.2Snout tip to 1st gill opening17.515.1–22.515.1–24.217.0–23.7Distance between inner nostril corners3.33.0–4.93.3–8.23.6–5.1Interorbital width9.98.7–14.28.2–13.16.3–12.5Mouth width8.17.9–10.96.5–11.37.2–13.9Eye horizontal diameter1.91.5–3.11.2–2.60.8–2.1Eye vertical diameter1.70.6–2.10.7–1.80.8–1.6 PREBRANCHIAL AND TRUNK Snout tip to 2nd gill opening18.917.7–22.324.220.8–25.9Snout tip to 3rd gill opening20.317.9–25.923.2–27.322.0–27.0Snout tip to 4th gill opening21.020.4–26.526.822.9–28.1Snout tip to 5th gill opening22.520.0–28.625.3–30.823.9–30.5Trunk width at pectoral origin10.4410.4–18.710.9–20.812.1–18.8Trunk height at pectoral origin5.06.4–20.02.1–20.59.9–21.7Snout tip to pectoral fin tip33.834.2–41.1 * 31.0–39.7Distance between preoral clefts9.725.1–8.16.3–13.25.4Eye to 1st gill opening10.19.0–12.95.8–13.410.6–10.9 PECTORAL FIN Base length4.24.5–8.25.1–9.24.8–7.4Anterior margin10.48.7–13.79.9–13.49.0–12.6Distal margin3.93.1–7.43.1–7.02.8–5.2Posterior margin2.74.2–8.54.0–8.24.0–8.4PELVIC FINOverall length4.57.6–11.07.9–10.58.0–10.6Base length3.64.7–7.74.1–7.44.9–7.5Anterior margin5.64.3–7.14.3–7.12.6–7.2Distal margin1.20.8–5.01.9–4.61.0–4.7Clasper length3.83.7–8.43.5–5.13.7–7.0Clasper length from pelvic axil0.50.8–2.41.00.7–1.9 FIRST DORSAL FIN Origin (AO)†39.717.2–36.628.8–42.927.1–40.6Origin (PO)†42.535.5–45.944.2–52.645.0–53.1Overall length (from AO)†19.212.7–32.316.0–29.412.5–30.6Overall length (from PO)†10.89.1–20.94.9–13.88.7–14.0Posterior margin5.23.7–6.84.4–6.63.8–6.9Height2.52.4–4.52.2–3.41.9–3.3 SECOND DORSAL FIN Overall length8.96.6–12.79.2–12.88.5–24.6Base length4.45.4–8.84.7–7.84.5–21.3Posterior margin5.75.5–7.75.6–8.04.4–7.3Height2.32.1–3.62.2–3.32.0–3.0 CAUDAL FIN Dorsal lobe length14.414.8–21.211.9–21.515.8–21.2Ventral lobe length10.68.6–16.68.2–15.810.8–15.2Dorsal tip to notch5.76.2–9.74.8–9.25.5–9.7Notch depth1.51.0–2.40.8–2.40.7–1.8 DISTANCES BETWEEN FINS Origin of pectoral to origin of pelvic fins30.735.0–46.930.8–39.131.0–46.1Base of pectoral to base of pelvic fins25.029.5–39.224.2–33.226.6–35.6Origin of pelvic to origin of caudal fin11.78.3–14.58.8–12.39.5–13.4Origin of 1st to origin of 2nd dorsal fin14.816.0–23.413.2–18.614.0–20.4Origin of 2nd dorsal to origin of caudal fin10.47.0–11.56.5–9.46.9–11.1DENTITION (tooth formula‡)Upper jaw18–0–19(17–24)–(0–1)–(17–24)(15–25)–(0–1)–(15–24)(18–24)–(0–1)–(18–24)Lower jaw25–0–27(22–28)–(0–1)–(22–28)(23–31)–(0–1)–(23–32)(24–29)–(0–1)–(24–29) MERISTICS Number of spiral valve turns4029–3432–3736–41 Notes. †AO, anterior origin of the first dorsal fin; PO, posterior origin of the first dorsal fin.‡Tooth counts are reported as tooth files in the functional series (left–symphysis–right; L–S–R); replacement rows were not counted.*Indicates that no comparative value was available in Yano, Stevens & Compagno (2004) for that character. Morphometry of S. antarcticus from Chilean waters. Morphometric and meristic measurements of the Chilean male embryo (CHALLWA-UA-25), compared with published ranges for large-bodied Somniosus species from Yano, Stevens & Compagno (2004). Morphometric values are expressed as percentages of total length (TL), except total length (cm) and meristic characters. Comparative values for S. microcephalus, S. pacificus, and S. antarcticus are shown as minimum–maximum ranges. Notes. AO, anterior origin of the first dorsal fin; PO, posterior origin of the first dorsal fin. Tooth counts are reported as tooth files in the functional series (left–symphysis–right; L–S–R); replacement rows were not counted. Indicates that no comparative value was available in Yano, Stevens & Compagno (2004) for that character. Caudal fin of female specimen of S. antarcticus. Lateral view of the caudal fin of an adult female Somniosus antarcticus captured off Antofagasta, showing the upper caudal lobe used to estimate the total length. Scale bar = 25 cm. The dentition was heterodont. The upper jaw comprised 18–0–19 tooth files (total 37) with small lanceolate cusps, whereas the lower jaw comprised 25–0–27 tooth files (total 52) with short, strongly oblique cusps and very high, narrow roots (Fig. 4A). No symphyseal tooth files were observed in either jaw. Dermal denticles exhibited erect, narrow crowns with hook-like cusps, giving the skin a rough, spiny texture (Fig. 4B). These features, together with the external morphology, are consistent with the genus Somniosus and not Rhinoscymnus. The spiral valve contained 40 turns, which falls within the range reported for S. antarcticus sensu Yano, Stevens & Compagno (2004). Therefore, our morphological and meristic evidence is most consistent with S. antarcticus under the diagnostic framework of Yano, Stevens & Compagno (2004). Details of dermal denticles and teeth of S. antarcticus. Diagnostic characteristics of Somniosus antarcticus. (A) Upper and lower jaws showing heterodont dentition, with lower teeth bearing short, strongly oblique cusps and high, narrow roots. (B) Dermal denticles with erect, narrow crowns, and hook-like cusps, giving the skin a rough spiny texture. Molecular analysis A 639 bp fragment of the COI gene was obtained from both the gravid female and late-stage embryo. The two Chilean sequences were identical across the trimmed alignment, indicating a shared mitochondrial haplotype that was consistent with maternal inheritance. BLAST searches against GenBank returned S. antarcticus as the closest match, with 99.7–100% similarity to sequences from New Zealand and Antarctic waters (p-distance = 0.000–0.003). No other Somniosus species matched the Chilean haplotype as closely. The Chilean haplotype differed by two substitutions (p-distance = 0.00313) from the New Zealand record labelled S. antarcticus (JN312346) and was identical to an Antarctic record labelled S. pacificus (JN641139), indicating haplotype sharing across nominal labels. The comparative dataset comprised 27 sequences spanning four Somniosus taxa (S. microcephalus, S. pacificus, S. antarcticus, and S. rostratus), and three somniosid outgroups representing other genera within the family. Collapsing of identical sequences yielded 14 unique haplotypes. The female and embryo shared the same Chilean haplotype, identified as H2, in the median-joining network (Fig. 5). The median-joining network (epsilon = 0) recovered a shallow haplotype structure within the large-bodied Somniosus complex and highlighted a single haplotype shared between nominal S. pacificus and S. antarcticus (the Chilean haplotype), whereas S. rostratus formed a distinct haplotype group separated from the large-bodied complex by substantially larger mutational distances (Fig. 5). A median-joining (MJ) haplotype network based on a 639-bp fragment of the mitochondrial cytochrome c oxidase subunit I (COI) gene for Sominosus spp. Circle size is proportional to haplotype frequency. Colors indicate the nominal species label, as reported in GenBank. Hatch marks on branches represent mutational steps; small filled nodes represent inferred median vectors. The genetic distance summaries were consistent with the haplotype network (Table 2; Table S2). Within-group divergence was greatest in S. microcephalus, whereas nominal S. pacificus and S. antarcticus showed comparatively low internal variations. The divergence between S. pacificus and S. antarcticus also remained low, and the nearest-neighbor distance between these nominal labels was zero, reflecting haplotype sharing between the Chilean specimens and an Antarctic record labelled as S. pacificus. In contrast, S. rostratus and the somniosid outgroups were more divergent from the large-bodied Somniosus complex, supporting the separation of Rhinoscymnus and other somniosid lineages at this locus. A distance-based neighbor-joining reconstruction provided a complementary clustering visualization of these patterns (Fig. 6), recovering the same broad groupings and a shallow structure within the large-bodied Somniosus complex. Within this complex, two closely related clusters were recovered: one composed mainly of sequences labelled S. pacificus and another containing the Chilean haplotype together with sequences labelled S. antarcticus. However, one Antarctic sequence labelled S. pacificus clustered with the latter group, so nominal S. pacificus and S. antarcticus were only partially separated in the COI dataset. This pattern is consistent with the haplotype network and distance summaries, which showed low inter-label divergence and haplotype sharing among southern records. 10.7717/peerj.21381/table-2Table 2COI genetic distances for Somniosus spp.Within-group COI genetic distances and nearest-neighbor distances (NND) for Somniosus taxa included in the comparative dataset. Distances were calculated from a 639 bp alignment and are reported as mean ± SD, with minimum–maximum values in parentheses. NND represents the minimum intergroup distance between any sequence in the focal taxon/lineage and any sequence outside that taxon/lineage. Taxon/lineage n sequences Pairwise comparisons p -distance, mean ± SD (min–max) K2P distance, mean ± SD (min–max) NND (p-distance) NND (K2P) S. microcephalus 8280.0050 ± 0.0042 (0.0000–0.0110)0.0051 ± 0.0042 (0.0000–0.0110)0.00310.0031 S. pacificus 7210.0016 ± 0.0014 (0.0000–0.0047)0.0016 ± 0.0014 (0.0000–0.0047)0.00000.0000 S. antarcticus 330.0021 ± 0.0018 (0.0000–0.0031)0.0021 ± 0.0018 (0.0000–0.0031)0.00000.0000Southern complex (S. pacificus + S. antarcticus)10450.0021 ± 0.0016 (0.0000–0.0063)0.0021 ± 0.0016 (0.0000–0.0063)0.00310.0031 COI genetic distances for Somniosus spp. Within-group COI genetic distances and nearest-neighbor distances (NND) for Somniosus taxa included in the comparative dataset. Distances were calculated from a 639 bp alignment and are reported as mean ± SD, with minimum–maximum values in parentheses. NND represents the minimum intergroup distance between any sequence in the focal taxon/lineage and any sequence outside that taxon/lineage. Distance-based neighbor-joining clustering visualization for Somniosus species complex derived from Kimura two-parameter (K2P) distances for the COI alignment used in haplotype analyses. Bootstrap support (1,000 pseudoreplicates) at branch nodes is shown (with an *) when >80%. Tip labels include GenBank accession numbers and nominal species assignments, as reported in GenBank (Table S1). Chilean specimens are indicated in bold. The three topologies were rooted using representative somniosid genera outside Somniosus as outgroups (Family Somniosidae). Discussion Our integrative morphological and molecular evidence documents the occurrence of a large-bodied Somniosus off northern Chile and provides a curated record from a region where sleeper shark records have remained scarce and are uncertain taxonomically. Previous records of large-bodied Somniosus from Chilean waters included specimens reported from approximately 33°S and 40°S (Pequeño, Lamilla & Crovetto, 1991; Brito, 2004); however, these records predated the revision by Yano, Stevens & Compagno (2004) and lacked both genetic data and internal meristic information, which are now regarded as important for species identification. In this context, the present record strengthens the evidence that the large-bodied Somniosus occurring in Chilean waters belongs to the Southern Hemisphere lineage, traditionally referred to as S. antarcticus sensu Yano, Stevens & Compagno (2004). A particularly important aspect of this record is the reproductive context. The small male was not captured independently but was aborted during handling, while the gravid female was being brought onboard. The direct observation of capture-induced abortion confirmed the sexual maturity of the female, representing a significant contribution to the reproductive biology of this genus. The estimated size of the gravid female appears small relative to previously published maturity estimates for the genus (approximately 300 to 400 cm TL, Finucci, 2018); however, this comparison should be interpreted cautiously because the estimate was derived from partial remains rather than from a complete specimen. The Chilean embryo exhibits a combination of characters consistent with Somniosus and not Rhinoscymnus. These include hook-shaped dermal denticles, dentition, and external morphology compatible with a large-bodied sleeper shark, and a spiral valve count that clearly exceeds the range reported for small-bodied somniosids. Within the large-bodied Somniosus complex, the spiral valve count provides the strongest meristic support for consistency with S. antarcticus, under the framework of Yano, Stevens & Compagno (2004). Although several morphometric proportions overlap among large-bodied nominal species and may vary with ontogeny, our morphological and meristic evidence is most consistent with S. antarcticus, as currently understood under the diagnostic framework proposed by Yano, Stevens & Compagno (2004) and Yano, Stevens & Compagno (2007). Our molecular results are compatible with this interpretation, but they also reinforce that COI does not fully resolve species boundaries within the pacificus/antarcticus complex (Timm et al., 2023). In the haplotype network, the Chilean haplotype was identical to an Antarctic sequence labelled S. pacificus and differed by only two substitutions from a New Zealand sequence labelled S. antarcticus. Accordingly, the divergence between nominal S. pacificus and S. antarcticus remained low, and the nearest neighbor distance between these taxa was zero. These patterns are consistent with previous mitochondrial studies showing shallow divergence and haplotype sharing among large-bodied Somniosus species in the Southern Hemisphere and adjacent regions (Benz et al., 2007; Murray et al., 2008; Christensen, 2022; Claassens et al., 2023). The NJ reconstruction recovered two closely related clusters within the large-bodied Somniosus complex: one composed mainly of sequences labelled S. pacificus and another including the Chilean haplotype together with sequences labelled S. antarcticus. However, one Antarctic sequence labelled S. pacificus clustered with the latter group, indicating that nominal S. pacificus and S. antarcticus were not clearly separated by COI. This pattern is consistent with the network and distance analyses and supports the view that COI is informative for broad lineage placement but does not fully resolve species boundaries within this species complex (Murray et al., 2008; Christensen, 2022; Claassens et al., 2023). In the present case, the Antarctic record labelled S. pacificus deserves particular caution because its taxonomic identity was assigned before the revised morphological framework of Yano, Stevens & Compagno (2004). This may represent a case of taxonomic uncertainty in the public record rather than true evidence of a well-differentiated Southern Hemisphere S. pacificus. Therefore, we interpret the Chilean specimens as belonging to the large-bodied Southern Hemisphere lineage that is most consistent with S. antarcticus sensu Yano, Stevens & Compagno (2004), while acknowledging that the precise boundary between S. antarcticus and S. pacificus remains unresolved with mitochondrial data alone. This interpretation is consistent with that of recent studies. Christensen (2022), based on complete mitogenomes, found no clear mitochondrial break between S. pacificus and S. antarcticus, and Timm et al. (2023) further questioned whether these nominal taxa represent distinct evolutionary lineages when evaluated using genomic data. Taken together, these studies suggest that the pacificus/antarcticus complex may reflect incomplete lineage sorting, shallow mitochondrial divergence, or an incompletely resolved taxonomic boundary. Under these circumstances, a pragmatic approach is warranted: large-bodied Southern Hemisphere Somniosus can be identified operationally as S. antarcticus sensu Yano, Stevens & Compagno (2004) based on morphology and internal meristics, while mitochondrial data are used primarily to document lineage affinity and to flag cases of taxonomic overlap or uncertainty. Future studies integrating nuclear markers or genome-scale datasets with detailed morphological examination of vouchered specimens from the Southern Hemisphere, North Pacific, and Atlantic/Arctic regions are essential to determine whether S. antarcticus and S. pacificus represent distinct biological species or geographically structured components of a single, broader lineage. The Chilean specimens reported herein provide important data points from a previously undersampled region of the southeastern Pacific. The Antofagasta record supports the occurrence of a large-bodied Somniosus consistent with S. antarcticus along the continental slope of northern Chile at approximately 22°S in deep waters associated with the Peru–Chile margin. This finding helps fill the geographic gap between Southern Ocean records and previous reports of large-bodied Somniosus spp. from central Chile (Pequeño, Lamilla & Crovetto, 1991; Brito, 2004; Yano, Stevens & Compagno, 2007; Ebert, Fowler & Compagno, 2021). More broadly, it supports the growing view that large-bodied Somniosus lineages occupy a wider latitudinal range in deep waters than previously documented, including temperate and, at least, occasionally subtropical environments (Acero et al., 2018; Claassens et al., 2023). The reproductive context of this record is noteworthy. Capture-induced abortion of a late-stage embryo indicates that gravid females occur on the northern Chilean slope and suggests that these habitats may form part of the reproductive range of the southern large-bodied lineage. However, a single event is insufficient to infer the presence of discrete pupping or nursery areas. Repeated records of gravid females, embryos, or neonates are required before northern Chile can be interpreted as a recurrent reproductive habitat. Deep-sea chondrichthyans typically share life-history traits that make them highly vulnerable to fishery-induced mortality, including slow growth, delayed maturity, and low fecundity (Finucci et al., 2022). Large-bodied Somniosus sharks are expected to conform to this pattern, and even infrequent capture may have disproportionate demographic consequences (Yano, Stevens & Compagno, 2007; Finucci, 2018). Our record provides evidence that deep-sea longline fisheries operating off northern Chile interact with large-bodied Somniosus, even though such captures appear to be uncommon. This pattern is consistent with observations from the same artisanal fleet targeting Patagonian toothfish (Dissostichus eleginoides) in austral waters, where sleeper sharks reported as Somniosus pacificus, and possibly corresponding to S. antarcticus sensu Yano, Stevens & Compagno (2004), occurred as bycatch and, despite being represented by only two individuals, contributed nearly 40% of the total deep-sea chondrichthyan biomass recorded in the study area (Reyes & Torres-Florez, 2009). Bycatch monitoring programs that systematically document elasmobranch catches, record sex and length whenever possible, secure tissue samples such as fin clips, and retain diagnostic materials (e.g., jaws and spiral valves) across both northern and southern fleets are essential to improve occurrence data and distinguish Somniosus species. Whenever possible, live individuals should be released into the wild, in line with the current recommendations of the National Fishery Authority for industrial longline and gillnet fisheries (Bustamante, Acuña & Vargas-Caro, 2024). On a broader scale, taxonomic uncertainty should not prevent fishery monitoring or conservation actions. Until genome-scale analyses clarify the status of S. antarcticus and S. pacificus, fishery assessments may need to track these taxa pragmatically as a large-bodied Somniosus complex while still recording morphological evidence that supports operational identifications. The morphometric, meristic, and COI data presented here contribute to this framework and provide a useful regional reference for future studies in the southeastern Pacific. Supplemental Information Supplementary Tables Comparative context of morphometric and meristic data Novel genetic sequences from Somniosus antarcticus The raw sequences of the COI gene from specimens collected in northern Chile.