New species of the genus Mesocletodes Sars, 1909 from the deep Gulf of California (Copepoda, Harpacticoida)

Abstract Investigations on the effects of the oxygen minimum zone on the distribution, abundance, and diversity of deep-sea benthic and pelagic fauna of the Gulf of California and Eastern Tropical Pacific has received attention recently. However, very little is known about the diversity of deep-sea benthic harpacticoids from this region, and only three species, Ancorabolus hendrickxi Gómez & Conroy-Dalton, 2002, Ceratonotus elongatus Gómez & Díaz, 2017 and Dendropsyllus californiensis Gómez & Díaz, 2017, have been described so far. The genus Mesocletodes Sars, 1909 is one of the most common and abundant genera of deep-sea harpacticoids. This genus has been traditionally subdivided into two species groups, the abyssicola and the inermis groups, based on the presence/absence of a dorsal cuticular process on the cephalothorax and anal somite, but some species have been reported to deviate from this scheme. As a result of their investigations, other researchers proposed the monophyly of the abyssicola group, and suggested the probable monophyly of two other species-groups. In this paper, the descriptions of three new species of the genus Mesocletodes from the deep sea of the Gulf of California are presented with some notes on their relationships with other species. Some comments on the monophyly of the genus are given.


Introduction
Great effort has been deployed since the late 1980's to study the biodiversity of the deep sea of the Gulf of California and Eastern Tropical Pacific (a complete list of contributions is available upon request). Extensive oceanographic campaigns (Talud I-XIII cruises) on board the research vessel "El Puma" of the Universidad Nacional Autónoma de México (UNAM) have been carried out from 1989 to 2009 in the frame of several research projects to better understand the effects of the oxygen minimum zone on the distribution, abundance and diversity of deep-sea benthic and pelagic fauna (crustaceans, molluscs, echinoderms, fishes and polychaetes among others) of the Gulf of California and Eastern Tropical Pacific (Hendrickx 2012). One of the components studied during Talud IV-Talud XIII cruises was the meiofauna, with especial attention to benthic harpacticoid copepods. Preliminary personal observations revealed a high diversity of harpacticoids. However, only three species of the family Ancorabolidae, Ancorabolus hendrickxi Gómez & Conroy-Dalton, 2002, Ceratonotus elongatus Gómez & Díaz, 2017 and Dendropsyllus californiensis Gómez & Díaz, 2017 have been described so far from the deep sea of the Gulf of California.
The genus Mesocletodes Sars, 1909 is one of the most common and abundant genera of the family Argestidae in deep-sea samples (Menzel 2011). Traditionally, two speciesgroups are recognized within the genus, and Bodin (1968Bodin ( , 1997 proposed the subdivision of the genus Mesocletodes into the abyssicola group and the inermis group, based on the presence/absence of a simple dorsal cuticular process on the cephalothorax and anal somite, but pointed out that that such subdivision has no taxonomic value. However, with the discovery of some new species, it was found that some deviate from Bodin's (1968Bodin's ( , 1997 scheme. For example, Por (1986: 95) suggested that M. opoteros Por, 1986, described without a dorsal process on the cephalothorax but with a dorsal process on the anal somite, could well belong to a different species-group within the genus, and suggested that Bodin's (1968Bodin's ( , 1997 division of this genus into species-groups could change with the discovery of new species. A similar condition has been observed for M. fladensis Wells, 1965and M. angolaensis Menzel & George, 2009. However, Menzel and George (2009 did not follow Bodin's (1968Bodin's ( , 1997 and Por's (1986) views. Instead, they suggested the monophyly of the abyssicola group based on the elongation of the caudal rami, and on the presence of a dorsal process on the cephalothorax and on the anal somite, and pooled all the species of Mesocletodes, with a dorsal cuticular process either on the cephalothorax or on the anal somite, or both, and with long or short caudal rami, in the abyssicola group, arguing that the deviation of Bodin's (1968Bodin's ( , 1997 scheme (presence of a dorsal process on the cephalothorax or on the anal somite only), and the shorter caudal rami observed in some species could eventually be regarded as secondary reductions (Menzel and George 2009: 253). Additionally, Menzel and George (2009) suggested the monophyly of, at least, two other groups of species, viz. those in which the males lack mouth parts (Menzel and George 2009: 253-254), and those with bifid dorsal processes on the P3-P5-bearing somites, and on the posterior half of the female double genital-somite (Menzel and George 2009: 254).
In this paper three new species of Mesocletodes from the deep sea of the Gulf of California are proposed. Additionally, some comments on the monophyly of Mesocletodes are provided.

Materials and methods
Sediment samples for meiofaunal analyses were taken in August 2000 in the Southern Gulf of California from Carmen basin to off Nayarit State, and in February 2007 in the Southern Trough of Guaymas Basin, during Talud IV and Talud X cruises, respectively, on board the research vessel "El Puma" of the Universidad Nacional Autónoma de México (UNAM). Sediment samples were collected at depths ranging from about 520 m to 2120 m during Talud IV cruise using a multiple sediment corer equipped with six cores of 30 cm in length and sampling surface of 3.9 cm 2 , and from about 379 m to 1902 m during Talud X cruise using a box corer from which triplicate sub-samples were taken with 69 cm 2 cores of 20 cm in length. The upper 3 cm layer of sediment was preserved in 70% alcohol, sieved through 500 and 38 µm sieves to separate macro-and meiofauna, and stained with Rose Bengal. Meiofauna was sorted at a magnification of 40× using an Olympus SZX12 stereomicroscope equipped with DF PLAPO 1× objective and WHS10× eyepieces, and harpacticoid copepods were stored separately in 1 ml vials with 70% ethanol. Illustrations and figures were made from whole individuals and its dissected parts using a Leica DMLB microscope equipped with L PLAN 10× eyepieces, N PLAN 100× oil immersion objective, and drawing tube. The dissected parts were mounted on separate slides using lactophenol as mounting medium. Huys and Boxshall (1991) and Menzel and George (2009) were followed for general terminology. Huys (1996)  The type material was deposited in the Copepoda collection of the Instituto de Ciencias del Mar y Limnología, Unidad Académica Mazatlán (ICML-EMUCOP). The map showing the sampling locations where the new species were found were prepared with GeoMapApp (http://www.geomapapp.org/) and the Global Multi-Resolution Topography (GMRT) default basemap (Ryan et al. 2009). double-somite without cuticular process dorsally ( Fig. 2A, E), with few lateral spinules close to posterior margin (Fig. 2E), ventrally with medial genital field (Fig. 3B) and with some spinules close to lateral margins, second half with bifid cuticular process dorsally ( Fig. 2A, E), with lateral spinules along posterior margin, serrated posterior margin between pair of ventral sensilla less pronounced (Figs 2E,3B). Dorsal and lateral surface of fourth urosomite with short row of small spinules, with few sensilla, posterior margin serrated ( Fig. 2A, E), ventrally with posterior serrated margin slightly less  (Ryan et al. 2009).
Caudal rami slender, exceedingly elongated, 17 times as long as wide in lateral view, as long as P4-bearing somite and entire urosome combined, gently curved upwards from lateral view, covered with small spinules ( Fig. 2A, F); with seven elements as follows: seta I and II in distal part of first third of ramus, the former ventral to and smaller than the latter (Fig. 2F); seta III situated subdistally on dorsal surface (Fig. 2G); seta IV and VI small, subequal in length, arising at outer and inner distal corners, respectively; seta V longest; dorsal seta VII tri-articulated, situated on anterior part of second third of ramus.
Antennule (Fig. 4A, B) octa-segmented; all segments with small slender spinules; first segment without armature, with one medial and one distal row of strong spinules; second segment with strong protrusion bearing one strong element (arrowed in Fig. 4B), and two strong spinules; third segment elongated, as long as fourth-seventh segments combined; eighth segment smallest. Armature formula as follows: Antenna (Fig. 4C). Basis elongated, covered with small spinules. Exopod uni-segmented, with two apical setae. Endopod bi-segmented; first segment with strong inner spinules, covered with smaller spinules, shorter than basis; second segment covered with small spinules, inner margin with stronger spinules and with two thin subdistal spines with STE, and six apical elements (one inner strong spinulose spine, one spinulose spine, two geniculate spinulose elements, and two outer elements fused basally of which innermost longer and with STE).
Mandible (Fig. 5A). Mandibular gnathobase with some surface spinules, with two distal blunt teeth, and with broad grinding face. Palp uni-segmented (exopod and endopod fused to basis), with one basal, one exopodal, and one subdistal and three apical endopodal setae.
Maxilla (Fig. 5D). Syncoxa with outer and inner spinules, with smaller spinules close to allobasis; with two endites; proximal endite with one slender seta, distal endite with one strong spinulose element, one pinnate and one bare seta. Allobasis with longitudinal row of outer spinules, with one strong spinulose spine fused to allobasis, one slender seta and one spinulose spine. Endopod uni-segmented, very small, with two setae.  Maxilliped ( Fig. 5E) subchelate, strong. Syncoxa with inner long and outer small spinules as shown, with two setae, one of which strong and longer than syncoxa. Basis unarmed, with inner long and outer small spinules. Endopod uni-segmented, fused to strong spinulose claw. P1 (Fig. 6A). Coxa with outer and inner slender spinules, with stronger spinules medially. Basis seemingly without spinular ornamentation, with outer and inner spine, the former longer. Exopod tri-segmented; EXP1 as long as following segments combined; with spinules as depicted; EXP1 and EXP2 without inner armature; EXP3 with four elements, of which outer and apical elements with STE, inner seta slender and reduced. Endopod bi-segmented, small; ENP1 with few inner spinules and one inner seta; ENP2 with few inner spinules, two apical and one outer seta. P2-P4 (Figs 6B, 7A-B). Praecoxa as in P3 and P4, small (see Fig. 7A, B). Coxa with few strong spinules close to outer distal corner. Basis seemingly without spinules (but two small spinules were observed at the base of the basal setophore of P3); outer basal seta of P2 spiniform, without setophore (Fig. 6B), of P3 and P4 setiform, long, with well-developed setophore (Fig. 7A, B). Exopod tri-segmented; EXP1 and EXP3 elongated, subequal in length, EXP2 shortest; EXP1 without, EXP2 with inner seta; P2 EXP3 and P3 EXP3 with two outer spines, two apical elements, and two inner setae ( Fig. 6A-B), P4 EXP3 with two outer spines, two apical elements and one inner seta (Fig. 7B). Endopod bi-segmented; first segment with one inner seta; second segment of P2 with one inner and two apical setae (Fig. 6B), of P3 ( Fig. 7A) and P4 ( Fig. 7B) with one inner and one apical seta. P5 ( Fig. 8A-C) with some spinules on baseoendopodal setophore. Endopodal lobe poorly developed, with two setae (innermost lost during dissection), of which outermost very small (Fig. 8A, B). Exopod distinct, long, slender, 7.8 times as long as wide (maximum width measured at its base), with spinules as figured, with three outer, one apical and one inner seta, and one subdistal tube pore (arrowed in Fig. 8C).
Armature formula as follows:   Description of male. Unknown. Etymology. The specific epithet is derived from the Latin adjective brevis, meaning short, and the Latin noun seta, meaning hair, and refers to the reduced innermost seta of the female P5 EXP. The name is a noun in the nominative singular.
Remarks. Mesocletodes brevisetosus sp. n. seems to be more closely related to M. dorsiprocessus than to M. bicornis, both from the Angola Basin, by the combination of several characters: 1) serrated posterior margin of the cephalothorax and P2-bearing Diagnosis (based on the female only). Body subcylindrical. Cephalothorax with dorsal cuticular process curved posteriorly. Genital somite and third urosomite incompletely fused dorsolaterally. Anal somite quadrate from dorsal view, with simple dorsal cuticular process curved posteriorly. Caudal rami subcylindrical, 2.5 times as long as wide, with seven setae. Antennule octa-segmented, second segment with protrusion bearing a long seta, but not as pronounced as in other species of the genus. Antenna with basis and uni-segmented exopod. Gnathobase of mandible with grinding face, and tri-segmented palp. Maxillary syncoxa with two endites, proximal endite with one, distal endite with three elements; endopod uni-segmented, with two setae. Syncoxa of maxilliped with one seta. P4 ENP2 with four setae. Outer setae of the P5 EXP issuing close to each other.
Description of female. Body: total length 725 µm measured from anterior margin of rostrum to posterior margin of caudal rami, subcylindrical, tapering slightly posteriorly, without clear demarcation between prosome and urosome ( Fig. 9A, B). Rostrum fused to cephalothorax, with two sensilla. Cephalothorax 0.24 times as long as entire body length; ornamented with sensilla and spinular patches as shown; with dorsal cuticular process curved posteriorly, the latter as in Fig. 9C. P2-P5-bearing somites with sensilla and small spinules along posterior margin, with minute spinules laterally. Genital somite and third urosomite (genital double-somite) incompletely fused dorsolaterally (Fig. 9A, B) (i.e. posterior margin of genital somite indicated by suture with transverse row of spinules and few sensilla dorsolaterally), completely fused ven-  Caudal rami (Figs 9A, B, 10A, 11A, B) subcylindrical, slightly tapering posteriorly, nearly as long as anal somite and 2.5 times as long as wide; with seven setae as follows: setae I and II located midway lateral margin, the former ventral to the latter and shorter; seta III as long as seta II, arising close to outer distal corner; setae IV and V situated distally; seta VI smallest, arising at inner distal corner; dorsal seta VII bi-articulated, somewhat shorter than seta II.
Antennule (Fig. 12A) octa-segmented; first segment with one proximal and one subdistal row of spinules; second and third segments with longitudinal short row of strong spinules; second segment somewhat globose and with protrusion bearing a long seta not as pronounced as in other species of the genus (indicated by an asterisk in Fig. 12A); third segment two times as long as wide; fourth segment with one, fifth segment without spinules; sixth segment with short transverse row of smaller spinules; Antenna (Fig. 12B). Coxa small, with few strong spinules. Basis with inner spinules. Exopod uni-segmented, with two setae. Endopod bi-segmented; first segment with strong inner spinules; second segment with some outer small spinules, inner margin with strong spinules and two thin lateral spines with STE, and six apical elements (one inner strong spinulose spine, two geniculate spinulose and one bare element, and two outer elements fused basally of which innermost longer and with STE).
Mandible (Fig. 12C, D) with robust coxa. Gnathobase with row of surface spinules, three distal single teeth as shown and several fused tooth-like elements forming a broad grinding face (Fig. 12C). Palp tri-segmented (Fig. 12D); basis with medial and distal small spinules, with one strong seta; exopod uni-segmented, small, with two setae of which innermost smaller; endopod quadrate, with six setae as shown.
Maxilla (Fig. 13D). Syncoxa with slender outer spinules, and with comparatively stronger spinules close to allobasis; with two endites; proximal endite with one slender seta, distal endite with one strong spinulose element and two slender smooth setae. Allobasis with few outer spinules, with one strong spinulose spine fused to allobasis, one slender seta and one spinulose spine. Endopod uni-segmented, very small; with two setae.
Maxilliped (Fig. 13E) subchelate, strong. Syncoxa with one spinulose strong seta slightly longer than basis; the latter with outer and inner spinules as shown. Endopod uni-segmented, fused to strong spinulose claw. P1 (Fig. 14A). Coxa with sets of outer, medial, and inner strong spinules, and with outer long slender spinules. Basis with strong spinules at base of outer and inner spine. Exopod tri-segmented; exopodal segments subequal in length, with spinules as depicted; EXP1 and EXP2 without inner armature; EXP3 with four elements, of which two outermost spines with STE, innermost slender and reduced. Endopod bi-segmented, not reaching tip of EXP1; ENP1 with few outer spinules and one inner seta, shorter than ENP2; the latter with few outer spinules and three elements.
P2-P4 (Figs 14B, 15A, B). Praecoxa as in P4 (see Fig. 15B), small, with transverse row of small spinules close to coxa. Coxa with small spinules close to basis, and with stronger outer spinules on anterior face and some medial strong spinules on posterior face. Basis with strong spinules at base of outer element, between rami, and at base of endopod, and with slender long inner spinules; basis of P2 ( Fig. 14B) with outer spiniform element, of P3 and P4 (Fig. 15A, B) with outer slender bare seta. Exopod tri-segmented; segments slender and elongate, ornamented as shown; EXP1 without, EXP2 with inner seta; P2 EXP3 and P3 EXP3 with two outer spines, two apical elements, and two inner setae (Figs 14B, 15A), P4 EXP3 ( Fig. 15B) with two outer spines, two apical elements and one inner seta. P5 (Fig. 10B) with some spinules on baseoendopodal setophore; with long outer basal seta. Endopodal lobe poorly developed, with three setae (innermost lost during dissection), of which outermost and medial elements close to each other, innermost separated from the former two elements. Exopod distinct, long, slender, 6.7 times as long as wide (maximum width at distal part), with outer and inner spinules as figured, with three outer and two apical setae, and one outer distal tube pore.
Armature formula as follows: Description of male. Unknown. Etymology. The specific epithet is derived from the Latin adjective simplex, meaning simple, and refers to the simple (not bifurcated) dorsal process on the anal somite. The name is an adjective in the nominative singular.
Remarks. Mesocletodes simplex sp. n. is attributed here to Bodin's (1968) abyssicola group. The dorsal process on the anal somite is simple in most species within this genus, but it is bifurcated in M. brevifurca, M. katharinae Soyer, 1964, M. meteorensis, M. monensis, and M. soyeri Bodin, 1968. The appendages of the species of this group exhibit an amalgam of conditions, most of which are shared with some species of the inermis group sensu Bodin (1968). The antenna of most species of Bodin's (1968) abyssicola group possesses a basis, but the condition of the antenna is inconclusive for M. bathybia Por, 1964a and M. brevifurca. Also, the antennary exopod of most species of this group is uni-segmented with two setae, but M. abyssicola seems to be unique among these species in that it is represented by one seta only (the condition of the antennary exopod of M. monensis is inconclusive, and the exopod of M. bathybia remains unknown). The mandibular palp of most species of this group is bi-segmented (exopod incorporated to basis, endopod uni-segmented), but uniramous in M. soyeri, M. bathybia and M. katharinae, and tri-segmented (with basis, uni-segmented exopod, and uni-segmented endopod) in M. simplex sp. n. The palp of the maxillule is  Smirnov, 1946 remains unknown). The P1 ENP is uni-segmented with three setae in M. abyssicola, M. robustus, and M. soyeri, but uni-segmented with one seta only in M. bathybia. A uni-segmented P1 ENP is present also in the species of Bodin's (1968) inermis group (e.g. M. makarovi Smirnov, 1946, M. guillei Soyer, 1964, M. inermis, M. trisetosa Schriever, 1983. The P1 ENP is bi-segmented with an armature complement of 0,3 in the first and second segment, respectively, in M. Dignosis (based on the male only). Body subcylindrical. Cephalothorax, free prosomites and urosomites, except for anal somite, with posterior margin serrated. Cephalothorax dorsoventrally flattened, without dorsal cuticular process. Anal somite quadrate, with dorsal cuticular process. Caudal rami 14 times as long as wide, with seven elements. Antennule octa-segmented, haplocer, second segment with strong protrusion bearing one strong seta and two strong spinules. Antenna with basis, without exopod. Mandibles, maxillules, maxillae and maxillipeds strongly atrophied, nontraceable. P2 and P3 ENP1 with one inner seta. P5 endopodal lobe with one seta.
Caudal rami (Figs 16A, B, 17B-F) slender, exceedingly elongated, 14 times as long as wide (maximum width measured at the base of ramus), as long as urosome, almost straight in lateral view (Figs 16B, 17D), covered with small spinules; with seven elements as follows: seta I and II issuing laterally in distal part of first third of ramus, the former ventral to and smaller than the latter (Fig. 17D, F); seta III situated subdistally on dorsal surface (Fig. 17B-E); seta IV reduced, fused basally to seta V (Fig. 17B, C, E), the latter longest; seta VI reduced, somewhat smaller than seta IV, arising at distal inner corner (Fig. 17B, C); dorsal seta VII bi-articulated, situated subdistally on posterior part of first half of ramus (Fig. 17B, D, F); with large outer pore distally (Fig. 17B-E).
Antennule (Fig. 18A) octa-segmented, haplocer; first segment without armature, with some medial and some distal spinules; second segment with strong protrusion bearing one strong element (arrowed in Fig. 18A), with two strong spinules; third and fourth segments smallest; fifth segment somewhat swollen, with aesthetasc fused basally to slender seta, with four spiniform elements, two of which modified; sixth segment with one normal and one modified element; sixth to eight segments elongated, subequal in length. Armature formula as follows: Antenna (Fig. 18B). Basis elongate, with few inner spinules on distal corner. Exopod absent. Endopod bi-segmented; first segment with strong inner spinules, as long as basis; second segment with inner strong spinules proximally and medially, with outer spinules on distal half of segment, laterally with one well-developed lateral spine with STE, and one very reduced element (the latter indicated in Fig. 18B, C), and with four apical elements (two spinulose spines, of which innermost smaller, and two geniculate elements).
Mandibles, maxillules, maxillae and maxillipeds strongly atrophied, non-traceable. P1 (Fig. 19A). Praecoxa small, with spinular row as shown. Coxa with small median and longer outer spinules. Basis with inner spinules, with outer and inner spines, the former somewhat longer. Exopod tri-segmented; EXP1 as long as EXP3, EXP2 shortest; with spinules as depicted; EXP1 and EXP2 without inner armature; EXP3 with four elements, of which outer and apical element with STE. Endopod bi-segmented, reaching proximal fourth of EXP3; ENP1 seemingly without spinular ornamentation, with one inner seta; ENP2 with few inner spinules, with one outer, one apical and one inner element.
Armature formula as follows: Figure 18. Mesocletodes unisetosus sp. n., male paratype. A antennule B antenna, reduced lateral spine arrowed C lateral armature of the antenna, reduced spine arrowed. Description of female. Unknown. Etymology. The specific epithet is derived from the Latin prefix ūni, meaning one, and the Latin noun seta, meaning hair, and refers to the presence of one seta only on the endopodal lobe of the male P5. The name is a noun in the nominative singular.  Huys, 1999) and Pseudotachidiidae (e.g. Paranannopus Huys, 2009;see Willen 2005). Menzel (2011) suggested that, regardless the developmental stage of the mouth parts (welldeveloped, strongly reduced or absent), the males of Mesocletodes become non-feeding during the last moult as an adaptation to the sparsely populated and oligotrophic deep-sea environments, and could represent a derived character. Also, Menzel (2011) noted that the sexual dimorphism in M. elmari (attributed to the inermis group), the only species for which both sexes have been described, is expressed, among other characters, in the antennule, P5, P6 and most interestingly, in the armature formula of the P1-P4 ENP2 (but the bi-segmented condition of the rami of the swimming legs is the same in both sexes), and that the mouth parts are present in both sexes regardless of whether the male is non-feeding or not. The males of M. fladensis, M. angolaensis, and M. unisetosus sp. n. are non-feeding unisetosus sp. n., seems to be more closely related to M. angolaensis than to M. fladensis. In addition to the characters above shared between these two species, they also share the presence of an inner seta on the P2 and P3 ENP1 (this seta is missing in M. fladensis) and the exceedingly elongated caudal rami more than 13 times as long as wide. Briefly, the male of the Mexican species, M. unisetosus sp. n., is unique and can be separated from the male of M. angolaensis, and M. fladensis, by the number of setae on the P5 endopodal lobe (two setae in M. angolaensis, and M. fladensis, but one seta only in the Mexican M. unisetosus sp. n.). Sars (1909: 290-291) created the genus Mesocletodes within the Cletodidae, and presented the diagnosis for the genus based solely on his redescription of the type species, M. irrasus (T. Scott & A. Scott, 1894) (Sars 1909: 291-292). Subsequently, Por (1986a) created and diagnosed the family Argestidae to reallocate, among other genera, the ge-nus Mesocletodes. For a general historical background on the genus Mesocletodes see also Menzel and George (2009) and Vakati et al. (2017). Sars (1921) described M. inermis Sars, 1921, andredescribed M. monensis (Thompson, 1893) and M. abyssicola (T. Scott & A Scott, 1901), and recognized the relationship between the latter two species (with a dorsal process on the cephalothorax and on the anal somite). He also noted the lack of dorsal processes on the cephalothorax and anal somite of M. inermis, but omitted any comment on the relationship of this species and M. irrasus. In his key to the species of Mesocletodes, Lang (1936) used the presence/absence of a curved dorsal process on the cephalothorax and on the anal somite to separate the species of Mesocletodes into two groups, M. irrasus and M. inermis without dorsal process on the cephalothorax and anal somite vs. M. brevifurca Lang, 1936, M. monensis and M. abyssicola with dorsal process on the cephalothorax and anal somite. Subsequent authors also recognized the relationships amongst those species with a dorsal process on the cephalothorax and on the anal somite (e.g. Soyer 1964, Por 1964a, and amongst those species without such processes (e.g. Por 1964b). It was Bodin (1968) who formally suggested that the species of Mesocletodes can be subdivided into two groups, the abyssicola group with a dorsal process on the cephalothorax and on the anal somite, and the inermis group without a dorsal process on the cephalothorax and anal somite, but noted that such subdivision has no taxonomic value. That Bodin (1968) included in his abyssicola group only those species with a dorsal process on the cephalothorax and on the anal somite is evident in his key to the species of Mesocletodes where he, for example, included M. fladensis Wells, 1965 with a dorsal process on the anal somite only, as part of his inermis group. Bodin's (1968) subdivision was accepted and used by subsequent authors (e.g. Coull 1973, Soyer 1975, Schriever 1983, 1985, Bodin 1997. Wells (1965) described M. fladensis from Fladen (in the Scottish sector of the North Sea) based on the male only. This species was described without dorsal process on the cephalothorax, but with a dorsal process on the anal somite, and without mouth parts (i.e. mandibles, maxillules, maxillae and maxillipeds strongly reduced and nontraceable). Later, in his report on five new species of Mesocletodes from the North Atlantic Ocean, Schriever (1985) described M. quadrispinosa Schriever, 1985, based on four females from the Iceland-Faroe Ridge. Mesocletodes quadrispinosa was described with a dorsal process on the cephalothorax, but without a dorsal process on the anal somite. Note that Menzel and George (2009: 252) and Menzel (2011: 47) diagnosed the genus Mesocletodes with four setae/spines on P1 EXP3, casting doubts about the relationships and position of M. quadrispinosa which was described with three setae only on P1 EXP3. Por (1986b) presented the description of M. opoteros based on five females from the Mozambique Channel between Mozambique and Madagascar. He described M. opoteros without a dorsal process on the cephalothorax, but with a dorsal process on the anal somite, and suggested that this species could well belong to a different speciesgroup within Mesocletodes, and noted that Bodin's (1968) division of the genus could change with the discovery of new species. However, upon re-inspection of the type material of M. opoteros, Menzel et al. (2011: 862) confirmed the presence of a cuticular dorsal process on the cephalothorax. Menzel and George (2009) presented the description of M. angolaensis from the Angola Basin (Southeastern Atlantic). The only male specimen of this species was described without dorsal process on the cephalothorax, but with a dorsal process on the anal somite, and without mouth parts, similar to what was reported for M. fladensis some years earlier. Note that Menzel and George (2009: 253, 254 table 6) erroneously commented on the lack of a dorsal process on the anal somite of M. fladensis (see the written description of the species in Wells (1965: 23-24, fig. 77)). They also presented the description of M. bicornis George, 2009 andM. dorsiprocessus Menzel &George, 2009, based on six and two females, respectively, from the Angola Basin (Southeastern Atlantic). They described the females of these two species with a dorsal process on the cephalothorax and anal somite, but also with small dorsal bifid cuticular processes on P3-P5-bearing somites and on the second half of the genital double-somite. In the same paper, Menzel and George (2009) presented the description of M. meteorensis Menzel & George, 2009 based on two females from the Angola Basin. This species was described with a dorsal process on the cephalothorax and anal somite, but contrary to the other species of the abyssicola group, with the caudal rami barely three times as long as wide. As a result of their investigations, Menzel and George (2009) gave an amended diagnosis of Mesocletodes for which they presented four synapomorphies/plesiomorphies, viz. 1) the presence of a strong protrusion with a strong, bipinnate seta pointing backwards on the second antennular segment/without such protrusion and corresponding seta normal, 2) the proximal outer spine of P1 EXP3 reduced/proximal outer spine well developed, 3) the presence of STE's on the spines of P1 EXP3/without STE, and 4) the mandibular gnathobase with a strong, grinding tooth/gnathobase of normal shape. They did not follow Bodin's (1968) and Por's (1986) views regarding the division of the genus, and pooled all the species of Mesocletodes with a dorsal cuticular process either on the cephalothorax or on the anal somite, or on both, and with long or short caudal rami, in the abyssicola group, arguing that the deviation of Bodin's (1968) scheme could eventually be regarded as secondary reductions (Menzel and George 2009: 253). Following this reasoning, they suggested the monophyly of the abyssicola group as defined by them, for which they proposed three synapomorphies/plesiomorphies: 1) presence of a dorsal cuticular process on the cephalothorax/cephalothorax without dorsal process, 2) presence of a dorsal cuticular process on the anal somite/anal somite without dorsal process, and 3) caudal rami remarkably elongated between seta III and VII/caudal rami not remarkably elongate.

Discussion
Additionally, Menzel and George (2009) suggested the probable monophyly of, at least, two other groups of species. On one hand, Menzel and George (2009: 254) suggested that the presence of bifid dorsal processes on the P3-P5-bearing somites and on the second half of the genital double-somite (not on the first urosomite as in Menzel and George 2009: 254) of M. bicornis and M. dorsiprocessus might be of high phylogenetic value to establish a monophylum. On the other hand, Menzel and George (2009: 253-254) suggested that the lack of mouth parts, as observed in M. fladensis and M. angolaensis, both of the abyssicola group sensu Menzel and George (2009), can be regarded as derived and therefore, might support a monophylum.
Some years later, Menzel (2011) presented a new and corrected diagnosis for Mesocletodes (in their generic diagnosis, Menzel and George (2009) omitted the maxillule, and they described the maxilliped of Mesocletodes as stenopodial), presented the sexually dimorphic modifications for the genus with M. elmari Menzel, 2011 (without dorsal process on the cephalothorax and anal somite, and with well-developed mouth parts in the male) as model of study, and relegated M. faroerensis Schriever, 1985and M. thieli Schriever, 1985(not M. thielei as in Menzel (2011) as incertae sedis within Argestidae due to the presence of an inner seta on P1 EXP2 in the former, and five setae on P1 EXP3 in the latter, and questioned the belonging of M. arenicola Noodt, 1952 to this family by the shape and armature of the caudal rami and armature complement of the P1 EXP3. This view is followed here and M. faroerensis, M. thieli and M. arenicola are relegated to species incertae sedis within Argestidae. The species presented herein matches the diagnosis of the genus Mesocletodes and exhibit the four synapomorphies for the genus.
More recently, Vakati et al. (2017) presented the description of two new species attributable to the inermis group, M. tetrasetosus Vakati, Thistle &Lee, 2017 andM. nudus Vakati, Thistle &Lee, 2017, based on one female and three males, respectively, from the San Diego Trough. Vakati et al. (2017) described the female of M. tetrasetosus without dorsal cuticular process on the cephalothorax and anal somite, and with mouth parts. On the other hand, they described the male of M. nudus without dorsal processes on the cephalothorax and anal somite, but also, as in the males of M. fladensis and M. angolaensis, which belong to the abyssicola group sensu Menzel and George (2009), with atrophied mouth parts. Vakati et al. (2017) presented an amended key to the species of Mesocletodes based on Schriever's (1985) key that, in turn, followed Bodin's (1968) scheme.
No unequivocal apomorphies have been detected so far to justify the monophyly of the family Argestidae, and its monophyletic status has not yet been demonstrated (George 2004(George , 2008(George , 2011. Corgosinho and Martínez Arbizu (2010) suggested that the shape and armature of the maxilla could shed some light on the monophyly of the family. The genus Mesocletodes has been diagnosed based on four synapomorphies, but the phylogenetic relationships within the genus are far from clear . At this point, the abyssicola group sensu Menzel and George (2009) is considered monophyletic based on the synapomorphic dorsal cuticular process on the cephalothorax and anal somite and the remarkable elongation of the caudal rami between seta III and VII, and the deviations from this scheme have been tentatively attributed to secondary reductions ). The synapomorphic condition of the dorsal process on the cephalothorax and anal somite seems to be well supported since they do not appear in the ground-pattern of Harpacticoida and Argestidae . Careful inspection of about 800 adult females of Mesocletodes with a dorsal process on the cephalothorax, but without an evident dorsal process on the anal somite, revealed the presence of a very small, inconspicu-ous, dorsal process on the anal somite (Menzel, pers. comm., in litt.). Interestingly, Soyer (1964: 604, fig. D) described the anal operculum of M. guillei with a complex ornamentation and with a "crête médiane se terminant par une courte dent…." (Soyer 1964: 604, fig. D). Whether this tooth is homologous to the dorsal process on the anal somite of other species of Mesocletodes, remains obscure. Menzel´s observations and the probable unique condition of the dorsal process on the anal somite of M. guillei, supports Menzel and George's (2009) view regarding the definition and monophyly of their abyssicola group. Regarding the length:width ratio of the caudal rami, Menzel (2011) hypothesized that the presence of extremely elongated caudal rami in some species of the inermis group, viz. M. elmari, and in the abyssicola group, could be due to convergence. On the other hand, the inermis group does not seem to be supported by any synapomorphy.
Given all the above, there seems to be another approach towards the monophyly of the genus Mesocletodes. It seems plausible that this genus could eventually be attributed to a new subfamily defined by the four synapomorphies currently known for the genus, viz. the presence of a strong protrusion with a strong, bipinnate seta pointing backwards on the second antennular segment, the proximal outer spine of P1 EXP3 reduced, the presence of STE's on the spines of P1 EXP3, and the mandibular gnathobase with a strong, grinding tooth, plus the presence of a dorsal cuticular process on the cephalothorax and anal somite.
Under this scheme, the presence of a dorsal process on the cephalothorax and anal somite, and the extreme elongation of the caudal rami between seta III and VII could be regarded as plesiomorphic within the subfamily, and the lack of such processes and the reduction of the caudal rami, as secondary apomorphic reductions. However, this requires more robust, in-depth analyses, accompanied by the diagnosis of this hypothesized subfamily, the re-diagnosis of the genus Mesocletodes, and the proposal of a new genus to include all the remaining species. These two genera could be composed as follows: Hypothetical genus. M. dorsiprocessus, M. bicornis, and M. brevisetosus sp. n.; defined by the synapomorphic bifid dorsal processes on P3-P5-bearing somites and posterior half of genital double-somite. The bifid dorsal process on the cephalothorax could be regarded as autapomorphic for M. bicornis. The bifid dorsal process on the anal somite would be regarded as plesiomorphic. The monophyly of this taxon was suggested earlier by Menzel and George (2009).