Exploring the diversity of the deep sea—four new species of the amphipod genus Oedicerina described using morphological and molecular methods

Collections of the amphipod genus Oedicerina were obtained during six expeditions devoted to the study of deep-sea environments of the Pacific Ocean. The material revealed four species new to science. Two species ( Oedicerina henrici sp. nov. and sp. nov. ) were found at abyssal depths of the central eastern Pacific in the Clarion-Clipperton Zone; one species ( sp. nov. ) ( Oedicerina claudei sp. nov.) was recovered in the Sea of Okhotsk (north-west Pacific), and one ( Oedicerina lesci sp. nov.) in the abyss adjacent to the Kuril-Kamchatka Trench (KKT). The four new species differ from each other and known species by the shapes of the rostrum, coxae 1 and 4, basis of pereopod 7, armatures of pereonite 7, pleonites and urosomites. An identification key for all known species is provided. The study of the cytochrome c oxidase subunit I gene of the four new species and Oedicerina ingolfi collected in the North Atlantic confirmed their genetic distinction. However, small intraspecific variation within each of the studied species was observed. In the case of the new species occurring across the KKT, the same haplotype was found on both sides of the trench, providing evidence that the trench does not constitute an insurmountable barrier for population connectivity. None of the species have so far been found on both sides of the Pacific.


INTRODUCTION
Deep-sea exploration for mineral resources of the seafloor has increased in the last decade and several research programs have been conducted to study this largest and least-explored ecosystem on Earth.
Among the areas that have gained particular scientific and economic interest is the Clarion-Clipperton Zone (CCZ) in the central Pacific due to the presence of polymetallic nodule fields (Glover et al., 2016;Janssen et al., 2019;Christodoulou et al., 2020). Another deepsea region that was recently extensively sampled covers a large area of the north-west (NW) Pacific, namely the Sea of Japan, the Sea of Okhotsk and the Kuril-Kamchatka Trench (KKT) with the abyssal plain adjacent to it (Malyutina & Brandt, 2013;Brandt & Malyutina, 2015;Malyutina et al., 2018;Brandt et al., 2020). The sampling gear used nowadays during scientific cruises, the Brenke-type epibenthic sledge in particular (Brandt & Barthel, 1995;Brenke, 2005), enables the collection of the small-sized fraction of deep-sea fauna that was often neglected during previous expeditions. Moreover, change in practice of sample fixation and subsequent storage provide material available for molecular examination. Such an approach revealed an unexpectedly high diversity of deep-sea fauna including recognition of a number of species new to science, for both macrobenthic (Polychaeta, Ophiuroidea, Isopoda and Amphipoda) and meiobenthic (harpacticoid Copepoda) groups (Glover et al., 2002;Janssen et al., 2015;Jażdżewska & Mamos, 2019;Brix et al., 2020;Christodoulou et al., 2020;Khodami et al., 2020).
Currently, scientists are putting effort into formally describing these new taxa (e.g. Bober et al., 2018b;Bonifácio & Menot, 2019;Renz et al., 2019;Dong et al., 2021;Kaiser et al., 2021). Another example includes 29 species new to science from various taxa described in a single volume of Progress in Oceanography, summarizing results from NW Pacific deep-sea exploration . However, because describing species is a time-consuming process, many new species remain as morphologically identified Operational Taxonomical Units (OTU) or Molecular Operational Taxonomic Units (MOTU), and only given temporary names or codes. Such an approach allows for preliminary assessment of biodiversity, but it is important to stress that only species with names are recognized by the scientific community and society and only named species can become a subject of conservation (Delić et al., 2017;Britz et al., 2020). Moreover, only the species with described morphology can be compared with historical collections or recent material unavailable for molecular studies (Dupérré, 2020). The description of new species provides a baseline for further ecological or biogeography studies and as such it is a crucial and the only sustainable step in recognition of species and their service for ecosystem functioning.
The Amphipoda belong to the brooding peracarid crustaceans (Malacostraca) and form an abundant component of the deep-sea benthos. As an example, in the NW Pacific, deep-sea Amphipoda may constitute c. 7% of the total faunal abundance and are outnumbered by the Annelida, Copepoda and Isopoda. Similar values of abundance in that area were recorded for the Bivalvia and Ophiuroidea . Within peracarids, deep-sea Amphipoda and Isopoda are the two dominant orders jointly constituting from 50% to almost 90% of the abundance and 60-80% of recognized species (Frutos et al., 2017;Brandt et al., 2019). However, amphipod diversity and abundance is known to be high in the bathyal (40-60% of species, 25-50% of abundance) and usually decreases towards abyssal and hadal depths (Frutos et al., 2017;Brandt et al., 2019). In the South Polar Front, deep-sea Amphipoda are less abundant than Isopoda and probably also less species rich . More than 400 species of deep-sea benthic amphipods (recorded below 2000 m) are currently known, but it does not reflect the actual deep-sea amphipod species richness. For example, from only three deep-sea Antarctic expeditions, approximately 500 species new to science still await to be described (Jażdżewska, 2015).
The family Oedicerotidae is represented by 47 known genera and 246 species (Horton et al., 2020). This diverse family, including primarily infaunal species, constitutes an abundant component of benthic amphipod communities at all latitudes and depths (Weisshappel & Svavarsson, 1998;Jażdżewska, 2015;Brix et al., 2018b;Vause et al., 2019). Among oedicerotid genera, Oedicerina Stephensen, 1931 appears to be a typical deep-sea taxon. Its shallowest known record comes from a trawl conducted between 200 and 500 m in depth (Ledoyer, 1986), while all other records are between 470 m (Coleman & Thurston, 2014) and 4050 m (Hendrycks & Conlan, 2003). Five species are described in Oedicerina to date: Oedicerina ingolfi Stephensen, 1931, Oedicerina megalopoda Ledoyer, 1986, Oedicerina denticulata Hendrycks & Conlan, 2003, Oedicerina loerzae Coleman & Thurston, 2014 and Oedicerina vaderi Coleman & Thurston, 2014. The study of the material of Oedicerina obtained during six deep-sea expeditions to the central east and NW Pacific revealed four species new to science that are described here in detail. Additionally, an analysis of the mitochondrial cytochrome c oxidase subunit I gene (COI) was conducted in order to provide barcodes for the new species that together with scientific descriptions will be useful to unravel the ranges of species distributions and their biogeography in the abyss.

MATERIAL AND METHODS
The material examined consisted of 37 individuals sampled during six deep-sea expeditions (Table 1). Of these, seven individuals were collected from the central east Pacific (CCZ) and an additional 30 specimens from the NW Pacific. The norTh-wesT paCifiC sTudy area (NW PaCifiC) The area around the Sea of Okhotsk was surveyed during the SokhoBio expedition in 2015 using the R/V Akademik Lavrentyev (Malyutina et al., 2018). The KKT and its adjacent abyssal plain was explored with the R/V Sonne in 2012 and 2016 during the KuramBio I and II expeditions, respectively . Details of the oceanographic features of the studied area are available in Malyutina & Brandt (2013), Brandt & Malyutina (2015), Saeedi & Brandt (2020), and Brandt et al. (2020).

sample ColleCTion and proCessing
The samples used in this study were collected using two types of epibenthic sleds: a Brenke-type sled (Brandt & Barthel, 1995;Brenke, 2005) and a cameraequipped epibenthic sled [C-EBS (Brandt et al., 2013)]. The deployment protocol followed Brenke (2005). Upon recovery, samples were passed through 300 μm and either sorted out immediately and preserved in 80% ethanol kept at -20 °C, or immediately transferred into chilled (-20 °C) 96% ethanol. In the second case, the sorting by stereomicroscope was carried out after 48 h storage in a -20 °C freezer (Riehl et al., 2014). A series of photographic stacks were obtained, collecting overlapping optical sections throughout the whole preparation (Michels & Büntzow, 2010;Kamanli et al., 2017). All individuals, except for those whose posterior part of the body was broken, were measured (from the tip of the rostrum to the end of the telson) and chosen specimens were dissected and mounted on permanent slides using polyvinyl-lactophenol containing lignin pink. All slides were examined using a Nikon Eclipse Ci compound microscope equipped with a camera lucida. Pencil drawings from the microscope were used as the basis for line drawings. The drawings were inked with Adobe Illustrator CS6 following the recommendations of Coleman (2003Coleman ( , 2009).
The following terminology has been applied concerning setation and extensions of the cuticle (modified from d'Udekem d'Acoz, 2004): tooth-nonarticulated extension of the cuticle; seta-articulated slender extension (may be short or long, plumose, serrate, denticulate, cuspidate or smooth); setulevery small and delicate short seta; spine-articulated robust extension (usually short).
Apart from the standard measurements typically used for descriptions of amphipods, two additional ones were provided. One expresses the curvature of the rostrum (the angle between the dorsal margin of the head and the frontal margin of the rostrum- Fig.  1A-B), the second one measures the width to depth ratio of the posterior lobe of coxa 4 (Fig. 1C).
The registered type material is deposited in the Zoological Museum of Hamburg (CeNak), Germany (ZMH), in the Senckenberg Museum (Frankfurt, Germany) (SMF) and in the National Scientific Center of Marine Biology (Vladivostok, Russia) (MIMB). All the remaining material is kept in the scientific collection of the Department of Invertebrate Zoology and Hydrobiology, University of Lodz, Poland. The summary of all studied individuals is provided in the Supporting Information (Table S1).

moleCular invesTigaTion
Eighteen individuals representing each identified species (from one to ten individuals per species) were chosen for cytochrome c oxidase subunit I gene (COI) analysis. Additionally, five individuals of O. ingolfi collected in North Atlantic during the IceAGE 1 and 2 expeditions (Brix et al., 2018b) were used in our molecular study. The total genomic DNA was extracted from one pleopod (if the posterior part of the body was missing the last present leg was used). The DNA extraction of individuals from the Central Pacific and from the North Atlantic was performed using 100 μL InstaGene Matrix (BIO-RAD). Digestion was carried out at 56 °C for 40 min. The extraction of individuals collected from the NW Pacific was carried out using a mixture of 150 μL pure H 2 O with 0.015 g Chelex (Sigma-Aldrich Co.) and 10 μL proteinase K. The digestion at 55 °C lasted for 6 h.
T h e D N A b a r c o d i n g f r a g m e n t o f C O I (658 bp) was amplified using the degenerate LCO1490-JJ (CHACWAAYCATAAAGATATYGG) a n d H C O 2 1 9 8 -J J ( AWA C T T C V G G R T G V C CAAARAATCA) primer pair (Astrin & Stüben, 2008). In the case of the CCZ and the North Atlantic specimens, polymerase chain reaction was performed with AccuStart II PCR SuperMix (Quantabio), whereas for the NW Pacific specimens DreamTaq Green PCR Mastermix (Thermo Scientific) was used. In both cases the reaction conditions followed Hou et al. (2007). Sequences were obtained by Macrogen Inc. (the Netherlands) on an Applied Biosystems 3730xl capillary sequencer. One-way (forward) sequencing was the standard procedure for all samples, but in addition, at least one individual of each species (preferably the holotype) was sequenced in both directions. As a result, each species received at least one sequence of the barcode fragment of the full length. Electropherograms were viewed in Geneious 10.1.2 and primer sequences and ambiguous positions were trimmed. Sequences were initially blasted using default parameters on NCBI BLASTn and translated into amino acid sequences to confirm that no stop codons were present. All sequences were deposited in GenBank with the accession numbers: MN346926 and MW377925-MW377946. Relevant voucher information, taxonomic classifications and sequences are deposited in the data set "DS-OEDICERI" in the Barcode of Life Data System (BOLD) (dx.doi.org/10.5883/DS-OEDICERI) (www. boldsystems.org) (Ratnasingham & Hebert, 2007).
The sequences were subjected to the Barcode Index Number (BIN) System (Ratnasingham & Hebert, 2013) in BOLD. It compares newly submitted sequences with the sequences already available. They are clustered according to their molecular divergence using distance-based algorithms (single linkage clustering followed by Markov clustering) that aim at finding discontinuities between Operational Taxonomic Units (OTU). Each OTU receives a unique and specific code (BIN), either already available or new if submitted sequences do not cluster with already-known BINs.
Intraspecific variation: Due to the bad condition of the individuals not much can be said about sexual or size-dependent dimorphism within the studied species. The only observed difference is the smaller size of the posterodorsal tooth on pleonite 3 in the immature male.
Molecular identification: Following the definition given by Pleijel et al. (2008), the sequence of the holotype male of O. henrici (ZMH K-60658, GenBank accession number MW377935) is designed as a hologenophore of all obtained sequences. The sequences of the paratype and an additional individual of the species are deposited in GenBank with the following accession numbers: MW377932, MW377937. The species has also received a Barcode Index Number from BOLD: AEB1524 (dx.doi. org/10.5883/BOLD:AEB1524).   Oedicerina teresae jażdżewska, sp. nov.

(figs 7-12)
Z o o b a n k r e g i s t r a t i o n : u r n : l s i d : z o o b a n k . org:act:8B11C501-328E-4F33-AA78-F2229D4847A1.
Intraspecific variation: No distinct differences were observed between the holotype and the mature female collected. The difference between adult individuals and the juveniles is expressed by the number of articles of flagella of antenna 1 and antenna 2 which is smaller in the latter.
Molecular identification: Following the definition given by Pleijel et al. (2008)  Etymology: The species is named for Krzysztof Leszek (Latin Lescus) Jażdżewski, the first author's brother.

Intraspecific variation:
The development of posterior teeth on pleonites 1-3 varies with size. In juveniles (3.1-7.0 mm) the teeth on pleonites 1-2 are weakly developed; however, the upright tooth on pleonite 3 is conspicuous. On the contrary, in larger individuals the teeth on pleonites 1-2 are distinct, whereas the upright tooth on pleonite 3 is weak. Urosomite 1 in some males is posteriorly slightly protruded forming a small hump (absent in females). Large individuals (both males and females) have urosomite 3 produced into a small subacute tooth over the telson. Etymology: The species is named for Dr. Claude De Broyer, a great friend and one of the first author's scientific mentors and renowned specialist in amphipod taxonomy, diversity and ecology.

Sexual dimorphism:
No sexual or size-dependent variation observed as the individual is unique.
Molecular identification: Following the definition given by Pleijel et al. (2008), the sequence of the holotype juvenile of O. claudei (SMF-56781, GenBank accession number MW377945) is designed as a hologenophore of all obtained sequences. The species has received also a Barcode Index Number from BOLD: AEA4699 (dx.doi. org/10.5883/BOLD:AEA4699).

moleCular invesTigaTion
Each of the morphologically recognized species received a unique Barcode Index Number. Across all species, the intraspecific diversity calculated on haplotypes is low, ranging from 0.002 (O. lesci) to 0.005 (O. henrici) for both K2P and p-distance. Each of the species is represented by three haplotypes (Table  2; Fig. 24B). An exception is O. claudei, as only one individual of this taxon was collected. The distances between the studied taxa varies from 0.059 to 0.238 of p-distance and from 0.061 to 0.289 of K2P (Table 3) (Fig. 24A).
The haplotype networks show a star-like topology (Fig. 24B)  With the description of four new species, the number of known Oedicerina species is almost doubled. Coleman & Thurston (2014) indicated high similarity of the species within this genus because only a few characters were used for species recognition (mainly ornamentation of pleonites and urosomites). This was noted also in the present study; our new data document that there are only minute differences in the mouthparts observed between species, i.e. in the setation of the mandibular palp, the shape of article 3 of the maxilliped palp (Table 4). In O. claudei and O. teresae, the number of setal teeth on the inner plate of maxilla 1 was eight (nine in all other species). However, this difference may derive from the fact that The distances were calculated using the P-distance method. Triangles indicate the relative number of individuals studied (height) and sequence divergence (width). The numbers in front of the nodes indicate bootstrap support (1000 replicates, only values higher than 50% are presented). Note that this tree does not represent a reconstruction of evolutionary history of the presented taxa. B, median joining network of the identified haplotypes. Each line represents a mutation between sequences. Colours denote sampling stations.
in both cases the studied individuals were not fully developed adults. Nevertheless, a number of characters differentiating species were observed. Apart from the already mentioned differences in the mouthparts, further differences are observed in the shape of the rostrum and its curvature, in the shape and setation of coxae 1-4, the length of dactylus of both gnathopods and pereopods 3-4, the shape of the basis of pereopod 7 as well as the shape of the lateral lobes of the telson. Additionally, some differences were observed in the ornamentation (size and shape of dorsal teeth) of pereon segment 7, the pleonites and urosomite 1.
Of the four species described here, two i.e. O. henrici and O. claudei, belong to taxa with a strongly deflexed rostrum. They share this feature with O. denticulata, O. megalopoda and O. vaderi. From all these species O. henrici can be separated by the first coxal plate that is distinctly wider than deep whereas in the remaining species both dimensions of coxa 1 are similar. Additionally, O. henrici differs from O. megalopoda in the shape of maxilliped article 3 that is strongly produced in the latter and unproduced in the former. The differences are also observed in the shape of the basis of pereopod 7. In both O. henrici and in O. vaderi  Table 1. it is slightly tapering distally; however, the posterior margin is slightly crenulated in the newly described species, while it is smooth in the latter. In O. claudei and in O. denticulata the shape of the basis is more ovate, but the first species possesses a posterodistal lobe nearly as long as the ischium, which is absent in O. denticulata. O. henrici is characterized by the presence of dorsal teeth on pereonite 7 and pleonites 1-3, that are absent in O. vaderi. In O. denticulata they are developed on pleonites 2-3, while in O. claudei a small tooth is observed on pleonite 3 only.
The group of the remaining four species shares the curved but not strongly deflexed rostrum. Within them only O. teresae possesses dorsal teeth on pereonite 7 and pleonites 1-3. Both O. ingolfi and O. lesci have a smooth dorsal surface in pereonite 7 and pleonites 1-3 each possess a single tooth. In O. loerzae only pleonites 1 and 2 are toothed. O. teresae differs from the other species also by the shape of coxa 1 that has the anteroventral corner subacute whereas it is bluntly rounded in the remaining species. O. lesci and O. ingolfi share many morphological characters. They can be separated by the shape of coxa 4 that is the widest at 2/3 of its length in O. lesci whereas in the other species it is widest close to the anteroventral corner. Additionally, the shape of the basis of pereopod 7 allows to distinguish the two species. It is posteriorly weakly sinuous in O. ingolfi, whereas in O. lesci it is straight. These two species differ also in the setation of coxal plates 1-3 which is distinctly denser in O. ingolfi.

moleCular sTudy and biogeographiCal remarks
The studied taxa presented a low level of intraspecific variability, as three haplotypes were recorded for each of the studied species. It has to be stressed that in the case of O. henrici and O. teresae, a low number of individuals was collected and the sampling covered only a small part of the potential species range. Further studies are required to assess more precisely the level of intraspecific variation in these taxa.
By contrast, O. lesci was represented by almost 30 individuals collected on both sides of the KKT (Fig.  25C). This distribution was confirmed by molecular results that distinguished three closely related haplotypes. In the same area, a comparatively low intraspecific diversity was observed among representatives of two families grouping moderately mobile Isopoda (Bober et al., 2018b(Bober et al., , 2019. In O. lesci the dominant haplotype appeared to be shared by individuals coming from stations situated on both sides of the trench as far as 700 km from each other (Fig. 24B). A recent molecular study of Amphipoda from the abyss adjacent to the KKT has also confirmed that this physiographic feature does not constitute a complete barrier for the gene flow of some molecularly Oedicerina denticulata Hendrycks & Conlan, 2003 Oedicerina henrici sp. nov. Stephensen, 1931 Head rostrum (shape)

Oedicerina ingolfi
Moderately deflexed, the angle between the dorsal head margin and rostrum margin a little more than 90 °S trongly deflexed, the angle between the dorsal head margin and rostrum margin c. 90 °S trongly deflexed, the angle between the dorsal head margin and rostrum margin 90 ° or less Not deflexed, wide angle between the head dorsal margin and rostrum margin

Head rostrum (length)
As long as 1 st article of peduncle A1 As long as 1 st article of peduncle A1  (Jażdżewska & Mamos, 2019). The dispersal of benthic amphipods that are brooders depends only on adults whose swimming abilities differ between species. The identification of the taxa in the above cited research was preliminary thereby prohibiting a final conclusion regarding the influence of lifestyle on the genetic connectivity of these species. Nevertheless, within the group of ten molecularly defined taxonomic units that appeared to be present on both sides of the KKT as many as nine belonged to the taxa of higher mobility-the Eusiridae, Lysianassoidea, Pardaliscidae, Synopiidae and Vemanidae. However, one species of that group was identified as a representative of the family

Species/ character
Oedicerina claudei sp. nov. Hendrycks & Conlan, 2003 Oedicerina henrici sp. nov. Stephensen, 1931 Urosomite  1a. Rostrum strongly deflexed, the angle between the head dorsal margin and rostrum margin c. 90 ° (Fig. 1A) . Phoxocephalidae grouping benthic dwellers (Brix et al., 2018b;Jażdżewska & Mamos, 2019). These results support previous assumptions that abyssal species display wide geographic ranges and that underwater physical barriers have no or only moderate influence on genetic connectivity (Zardus et al., 2006;Brix et al., 2011Brix et al., , 2015Etter et al., 2011). A study of the isopod family Haploniscidae in the NW Pacific based exclusively on morphological identification also revealed some species occurring on both sides of the KKT (Johanssen et al., 2019). However, recent molecular studies based on isopods indicate that the lifestyle of the studied group may influence the geographic range of species Brix et al., 2018aBrix et al., , 2020Riehl et al., 2018). Good swimming abilities may promote genetic exchange between populations living across the trench, as observed for Rhachotropis saskia . The family Oedicerotidae, to which the newly described species belong, includes benthic infaunal taxa regarded as permanent burrowers (De Broyer et al., 2003;Brix et al., 2018b). As a result they are considered to be moderately mobile and their population connectivity seems to be restricted leading to higher genetic structure. However, an example of interesting behaviour potentially explaining higher than expected gene exchange of this moderately mobile group of amphipods was noted in some shallow-water oedicerotids as they were observed to migrate into the water column for reproduction (Brix et al., 2018b).

Oedicerina ingolfi
Nothing is known about the mating behaviour of deepsea species from this family; however, Hendrycks & Conlan (2003) reported O. denticulata from samples collected at least 50 m above the seafloor, supporting the assumption that despite having an ability to dig in the sediment, at least some of the species in this genus may also occur in the water column. Based on our results O. lesci has its occurrence restricted to abyssal depths (4681-5419 m), while the KKT extends down to c. 9500 m depth (Dreutter et al., 2020). A study of the bathymetric distribution of 28 MOTUs of Amphipoda from the KKT and adjacent abyssal plain revealed only four MOTUs being present both in the abyss and the hadal zone (Jażdżewska & Mamos, 2019). The identification of the collection coming from the deepest stations in the KKT has not been finished yet, however, to date there is no evidence of the presence of O. lesci at hadal depths (Jażdżewska A, pers. obs.). Consequently, it may be assumed that the dispersal of this species along and across the trench depends on its swimming over the bottom, possibly also taking advantage of the near-bottom currents flowing in that area (Mitsuzawa & Holloway, 1998) instead of crawling on the sediment surface. The species of the genus Oedicerina were found in all three oceans in both the Northern and Southern Hemispheres (Fig. 25A) (Stephensen, 1931;Ledoyer, 1986;Hendrycks & Conlan, 2003;Coleman & Thurston, 2014). Two species, O. ingolfi and O. vaderi, occur in the North Atlantic, O. megalopoda in the Indian Ocean and the remaining six species in the Pacific. O. loerzae was described from the bathyal of the Chatham Rise and it constitutes the southernmost record of the genus. The NW Pacific is inhabited by two newly described species: O. lesci and O. claudei, while in the eastern Pacific three species have been found so far: O. denticulata, O. henrici and O. teresae. A typical description of deep-sea fauna includes its rarity and patchy distribution (Kaiser et al., 2007). Many of the species described from the deep sea have never been sampled anywhere else and are known only from the original type localities, often based on single individuals. According to the definitions used in Kaiser et al. (2007), all species of the genus Oedicerina,

Oedicerina lesci sp. nov.
Oedicerina loerzae Coleman & Thurston, 2014 Oedicerina megalopoda Ledoyer, 1986 Oedicerina teresae sp. nov. apart from O. ingolfi and O. lesci, may be treated as rare taxa, whereas the cited two show patchiness of distribution (in the case of these species, > 75% of all known individuals were collected at two stations). The analysis of bathymetric ranges of species within Oedicerina indicates two species inhabiting the upper bathyal (200-530 m), two species preferably occurring at middle and deep bathyal depths (1802-3200 m), and four species that are typical of the abyss (4050-5419 m) (Table 4). O. claudei was recorded at 3307 m in the Sea of Okhotsk, which at depths > 3000 m, is defined as an abyssal plain . At the time of this publication, the deepest record of the genus was 4050 m observed for O. denticulata in the north-east Pacific (Hendrycks & Conlan, 2003). Three out of four species presented here were collected in even deeper waters, with the deepest station at c. 5420 m, where this genus was observed in the abyss adjacent to the KKT.