Pseudanthias hangapiko sp. nov. (Teleostei, Serranidae, Anthiadinae) is herein described from three specimens collected from a depth of 83 m in a mesophotic coral ecosystem off Hanga Piko, Rapa Nui (Easter Island), Chile. Pseudanthias hangapiko sp. nov. can be distinguished from its congeners in live coloration and by the following combination of characters: dorsal-fin rays X, 17; anal-fin rays III, 8; pectoral-fin rays 16 (left side of one specimen 17); vertebrae 10+16; scales relatively large, two scales above lateral-line to base of fifth dorsal spine, and 16–17 circumpeduncular scales; gill rakers 11+23; and a slender body, with greatest body depth 3.6 (3.4–3.8) in SL. The most similar DNA barcodes (mitochondrial COI gene) are from Pseudanthias ventralis Randall, 1979 and Pseudanthias hawaiiensis Randall, 1979, with 16.8% and 17.0% uncorrected divergence, respectively. This fish is one of four new species that were documented from a pair of technical dives at a single location in Rapa Nui, emphasizing the high number of undescribed species likely still unknown in mesophotic coral ecosystems, especially in geographically remote locations. Pseudanthias hangapiko sp. nov. adds to the Rapa Nui ichthyofauna, which hosts the second-highest level of endemism in both shallow and deep-water fishes.

Biodiversity, coral-reef twilight zone, ichthyology, island, mesophotic coral ecosystem, reef fish, taxonomy

The genus Pseudanthias Bleeker, 1871 (Teleostei, Serranidae, Anthiadinae) currently comprises more than sixty valid species found globally in temperate and tropical oceans (Allen and Erdmann 2012). Many members of the genus are protogynous hermaphrodites, and most can be easily identified by the sexually dichromatic coloration of living adult specimens (Kuiter 2004). They are conspicuous inhabitants of coral reef environments and mesophotic coral ecosystems (MCEs), occurring in large haremic aggregations and feeding on zooplankton (Allen and Erdmann 2012). The highest diversity of Pseudanthias species is recorded in the coral triangle, with nearly 30 described species occurring in Indonesia and the Philippines alone (Allen and Erdmann 2012). Moving eastward out of the coral triangle and into the South Pacific, diversity within the genus decreases, and the relative proportion of endemic species increases, following the general pattern observed in other groups (Randall 1998; Allen and Erdmann 2012). Most Pseudanthias species occur on MCEs (Rocha et al. 2018), where several new species have been recently described (Gill et al. 2017; Gill and Psomadakis 2018; Victor et al. 2020), a trend in discoveries that results from increased exploration of the deeper sections of coral reef habitats (Rocha et al. 2018; Pinheiro et al. 2019; Pimentel et al. 2020).

Through a partnership between the Pontificia Universidad Católica de Chile and the Hope for Reefs Initiative of the California Academy of Sciences, our team conducted surveys using technical diving and closed-circuit rebreathers (Hollis Prism 2) to depths of 110 m to identify and document the fish communities associated with MCEs at Rapa Nui. Here we describe a new species of Pseudanthias collected in a MCE off Hanga Piko, Rapa Nui (Easter Island), Chile in March 2017. This is the fourth species description resulting from this expedition to Rapa Nui, and the sixth member of the subfamily Anthiadinae to be found at the island. Caprodon longimanus Günther, 1859, Plectranthias parini Anderson & Randall, 1991, P. ahiahiata Shepherd, Phelps, Pinheiro, Pérez-Matus & Rocha, 2018, and Luzonichthys kiomeamea Shepherd, Pinheiro, Phelps, Pérez-Matus & Rocha, 2019 are the other described Anthiadinae known from Rapa Nui; a presumed Tosanoides has also been observed with remotely operated vehicles but has yet to be captured and examined (Easton et al. 2017). These species, along with Chromis mamatapara Shepherd, Pinheiro, Phelps, Easton, Pérez-Matus & Rocha, 2020, and the new anthias presented here, appear to be restricted to MCEs, as is the case for much of the ichthyological biodiversity being revealed by scientists exploring these ecosystems (Rocha et al. 2018; Pinheiro et al. 2019).

Three individuals of the new species were collected with hand nets while diving on mixed-gas, closed-circuit rebreathers (Hollis Prism 2) around Rapa Nui, in March 2017. Specimens were collected and immediately transported to a field laboratory where they were photographed, tissue sampled, fixed in 10% formalin, and preserved in 75% ethanol. Measurements and x-radiographs were made at the California Academy of Sciences three and one-half years later. Counts were performed with the aid of a stereomicroscope. Morphological characters were measured to the nearest 0.01 mm with digital calipers following the conventions described in Anderson et al. (1990), Anderson and Heemstra (1980, 2012), and Gill et al. (2017). Morphometric data for the holotype and paratypes are presented in Table 1. Proportional measurements in the text are rounded to the nearest 0.1 mm. We measured standard length (SL) as the straight-line distance from the anteriormost point of the upper lip to the base of the caudal fin. Head length (HL) was measured from the anteriormost point of the upper lip to the posterior end of the opercular flap. Body depth (BD), was measured as the maximum depth from the origin of the spinous dorsal fin. Counts of principal caudal-fin rays follow Gill et al. (2016), and are presented in the form upper + lower, where the upper rays are those associated with hypurals 3–5, and the lower rays are those associated with hypurals 1–2 and the parhypural. Procurrent caudal-fin rays are those dorsal and ventral to the principal rays. Elongated filaments on the fin rays were not included in the measurements. Vertebral counts include the last vertebra fused to the hypural plate; vertebral counts are presented as precaudal + caudal, where the first caudal vertebra is the anteriormost vertebra bearing a haemal spine. Gill raker counts were made on the left side first arch, and are presented as upper (epibranchial) + lower (ceratobranchial) rakers on the anterior face of the first arch; the angle raker is included in the lower limb count. In the description, data are given first for the holotype, followed by a range of values for the paratypes, in parentheses, where variation was noted. Where counts are recorded bilaterally, both counts are given and separated by a vertical line (|). Where there is some degree of uncertainty, counts are followed by a question-mark (?). Comparisons with related species were based on literature accounts, particularly morphometric and meristic data from Pseudanthias ventralis Randall, 1979 and Pseudanthias hawaiiensis Randall, 1979, the more recently described Pseudanthias tequila Gill, Tea & Senou, 2017, Pseudanthias emma Gill & Psomadakis, 2018, Pseudanthias timanoa Victor, Teitelbaum & Randall, 2020, Allen and Erdman (2012), and Anderson’s (2018) checklist of the Anthiadinae.

Mitochondrial cytochrome c oxidase subunit I (COI) DNA was sequenced and analyzed for the new species. DNA extraction and PCR amplification of the COI were performed following protocols described by Arango et al. (2019), using BOLFishF1/BOLFishR1 primers. Alignments of DNA sequences were done using a standard Geneious global alignment with free end gaps and 65% similarity in the program Geneious Prime 2020.0.3 (Kearse et al. 2012). Because taxon sampling for Pseudanthias in public databases and our lab is not sufficient for a detailed phylogenetic analysis, and we only sequenced one mitochondrial DNA marker, we did not attempt a phylogenetic reconstruction. For our comparisons, genetic distances are uncorrected. COI sequences were compared to sequences of all other available Anthiadinae in GenBank, which include 29 species of Pseudanthias as well as representatives of the following genera: Anthias, Baldwinella, Choranthias, Hemanthias, Luzonichthys, Meganthias, Nemanthias, Odontanthias, Plectranthias, Sacura, Serranocirrhitus, and Tosanoides. The GenBank accession numbers for Pseudanthias hangapiko sp. nov. are MZ087699, MZ087670, and MZ087671. The holotype is deposited in the California Academy of Sciences ichthyological collection (CAS 247252) and the paratypes are deposited at the California Academy of Sciences ichthyological collection (CAS 247254) and the Smithsonian National Museum of Natural History (USNM 443821).

The presence of Pseudanthias hangapiko sp. nov. in Rapa Nui extends the known geographic range of the genus Pseudanthias by nearly 2,000 km eastward in the South Pacific. Pseudanthias ventralis, the species with the smallest uncorrected genetic distance at the mtDNA COI gene, is also the one occurring nearest, with the south-easternmost edge of its range occurring at Pitcairn Island (Randall 1979). Although the type specimens for P. ventralis are from Pitcairn, it likely comprises a species flock with at least three distinct groupings based on coloration and geographical range (Kuiter 2004), and many phylogeographic studies have found evidence of strong biogeographic partitions in the Pacific resulting in genetically distinct conspecifics (Bowen et al. 2016). A closely related species, P. hawaiiensis, marks the north-easternmost edge of the genus’ distribution in Hawai’i (Randall 1979). Not surprisingly, diversity within the genus Pseudanthias decreases moving eastward out of the coral triangle: there are at least 30 species of Pseudanthias in the Coral Triangle (Allen and Erdmann 2012), however, aside from P. ventralis, only an additional nine species of Pseudanthias occur in French Polynesia (Bacchet et al. 2017). Only three species, all presumed to be endemic, occur in the Marquesas (Williams et al. 2013; Delrieu-Trottin et al. 2015). There are no Pseudanthias known from the Eastern Pacific, which is relatively well-explored even at depth (Cortés 2019). As such, Pseudanthias hangapiko sp. nov. likely marks the easternmost distribution for the genus.

Taxonomy within the anthiadine fishes is problematic. Several genera, including Pseudanthias, may be well-documented and easily recognized groups of reef fishes, but they are poorly diagnosed, likely polyphyletic, and greatly in need of revision (Gill et al. 2017). With more complete sampling of extant taxa using multiple molecular markers, several members of the genus Pseudanthias will likely be assigned to other genera (Gill et al. 2017). Additionally, the high genetic distance between P. hangapiko and its closest sequenced relatives (>16%) indicates that the placement of this new species within the genus Pseudanthias is uncertain. Nevertheless, creating new genera without a complete understanding of the phylogenetic and morphological variation within Pseudanthias today would likely only generate more taxonomic instability. Thus, the generic placement of P. hangapiko should be considered provisional.

Understanding remote island ecosystems is critical to advancing scientific knowledge of speciation and ecology (Pinheiro et al. 2017; Quimbayo et al. 2019). However, the challenging logistics of conducting mesophotic exploration in geographically isolated locations such as Rapa Nui has undoubtedly limited knowledge of the deep reef biodiversity occurring in such places. With recent advances in technology, such as remote operated vehicles (ROV), consumer-grade closed-circuit rebreathers (CCR), baited remote underwater video (BRUV) employing inexpensive, low-light digital cameras, and autonomous underwater vehicles (AUV), scientists now have multiple tools to use for exploring deeper habitats that were not available to us even a decade or two ago (Pimentel et al. 2020). These studies remain expensive and administratively complex, requiring both significant funding and effective collaboration in order to be successful. Our 2017 expedition to Rapa Nui resulted in descriptions of four new species, all of which were collected on a pair of dives at a single location near the island. This remarkable discovery rate demonstrates the magnitude of the biodiversity waiting to be unveiled in Rapa Nui and elsewhere through deep reef exploration and purposeful collaboration.

This work was funded by the generous support of donors to the California Academy of Sciences’ Hope for Reefs Initiative, and Fondecyt # 1151094 granted to APM. This research was conducted under permit 2231 from the Ministerio de Economía Fomento Y Turismo de Chile and exported under certificate 669081 from the National Fisheries Service of Chile (SERNAPESCA). We are grateful to colleagues who helped in the field, lab, and with discussions: M. Bell, C. Castillo, D. Catania, J. Fong, V. (Tuto) Garmendia, E. Hey, I. Hinojosa, K. Jewett, M. Lane, A. Mecho, J. McCosker, C. Rocha, and W. Teao. We thank A. Gill for reviewing the first draft of the manuscript and for his assistance with verifying osteological characters. Logistical and equipment support was provided by Orca Diving Center and Hollis Rebreathers. We would like to thank J. Fong for taking the radiograph presented here. This work was approved by the Institutional Animal Care and Use Committee of the California Academy of Sciences (CAS IACUC approval number 2016-01).