Proposal of Marbefia , gen . n . and Inermiphonte , gen . n . , including updated keys to the species of Pseudonychocamptus Lang , 1944 and Paralaophonte Lang , 1948 ( Copepoda , Harpacticoida , Laophontidae )

Both sexes of the enigmatic laophontid species Pseudonychocamptus carthyi Hamond, 1968 (Copepoda, Harpacticoida) are redescribed in detail based on type and other material from Norfolk, England. Th e species exhibits marked diff erences with other Pseudonychocamptus species and is consequently fi xed as the type of a new genus Marbefi a, gen. n., being defi ned by the following autapomorphies: (1) P2–P4 enp-2 (except P4 enp-2 in ♂) infl ated, with two parallel rows of fl imsy setular extensions; (2) P4 with sexually dimorphic setation on enp-2 (outer distal seta reduced in ♀, very long in ♂); and (3) antennule ♀ with spinules along posterior margins of segments 1–6. Laophonte danversae Hamond, 1969, “Laophontid male, ?gen., ?sp.” sensu Hamond (1969) [renamed as Inermiphonte hamondi, sp. n.] and Laophonte ?drachi Médioni & Soyer, 1966 are collectively placed in a new genus Inermiphonte, gen. n. which shares as sistergroup relationship with Harrietella T. Scott, 1906. Marbefi a, gen. n. is considered sister to a clade comprising the genera Pseudonychocamptus Lang, 1944, Pilifera Noodt, 1952, Inermiphonte, gen. n. and Harrietella. Pseudonychocamptus kolarovi Apostolov, 2008 is transferred to Paralaophonte whereas Chislenko’s (1967) record of P. koreni Sars, 1908a is considered doubtful. Willey’s (1935) variety “fi ssirostris” of Paralaophonte brevirostris (Claus, 1863) is rejected pending a thorough analysis of the variability in the P. brevirostris complex of species. P. congenera mediterranea Lang, 1948 appears to be more closely related to P. lacerdai Jakobi, 1953 than to its nominotypical subspecies P. congenera congenera (Sars, 1908b) whereas ZooKeys 23: 1–38 (2009) doi: 10.3897/zookeys.23.168 www.pensoftonline.net/zookeys Copyright Rony Huys, Wonchoel Lee. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Launched to accelerate biodiversity research A peer-reviewed open-access journal


Introduction
In one of his papers dealing with the Laophontidae from Norfolk, Hamond (1968) described a new species in the genus Pseudonychocamptus Lang, 1944 for a single female found in washings of intertidal colonies of the hydroid Hartlaubella gelatinosa (Pallas, 1766) (as Laomedea gelatinosa) collected under the Hunstanton pier.Pseudonychocamptus carthyi Hamond, 1968 has been recorded in the literature only once since its original description and the male has remained unknown.Th e only other record is that by Holmes and Minchin (2000) who collected a single female from a Serpula vermicularis L. reef at 4 m depth near the ferry pier on the south side of Killary Harbour, Co. Galway, Ireland.
In their review of the genus Paronychocamptus Lang, 1948, Lee andHuys (1999) identifi ed a lineage within the subfamily Laophontinae, consisting of the genera Pseudonychocamptus, Pilifera Noodt, 1952, Weddellaophonte Willen, 1996 and Heteronychocamptus Lee & Huys, 1999.Th is "PWPH-clade" is primarily characterized by a distinctive type of swimming leg sexual dimorphism.In all four genera, females possess a seta on the proximal endopod segment of P3 and P4 but males consistently lack this armature element (Table 1). Lee and Huys (1999) also recognized a close relationship between Pilifera and Pseudonychocamptus but remarked that P. carthyi, by virtue of its primitive swimming leg armature, probably occupies a position close to the ancestor of the PWPH-clade.Pending the discovery of the unknown male, they refrained from attributing the species to a new genus.In this paper we have re-examined additional material of P. carthyi from Norfolk, which had been deposited in the collections of the Natural History Museum by Dr Richard Hamond.Th ese specimens, representing both sexes, form the basis for a redescription of the female and the fi rst description of the male, providing morphological evidence for its assignment to a new genus Marbefi a.

Material and methods
Before dissection, the habitus was drawn from whole specimens temporarily mounted in lactophenol.Specimens were dissected in lactic acid and the dissected parts were mount- Comparison between Harrietella and laophontid genera/species displaying sexual dimorphism in setation of P3-P4 enp-1 (exp = exopod; enp = endopod; benp = baseoendopod; A1 = number of antennulary segments; P3 ♂ = number of endopodal segments); losses in males indicated in boldface; species here reassigned to Inermiphonte gen.n. marked by *. a Hamond (1969) noted variability in the armature of P4 endopod ♀ (1.021 or 0.020 or 1.020); examination of additional material showed 1.021 to be the normal condition (as fi gured by Hamond 1969: his Fig 14).b sensu Hamond (1969: 9) = Inermiphonte hamondi, sp.n. c Note that Médioni andSoyer's (1966: 1057) armature formula is given in corrected format here to account for the small inner distal seta on P2-P4 exp-3 (fi gured by the authors for P3 but omitted in their formula; overlooked in P4; P2 was not fi gured).Wells (2007: 436, 458) had already amended the setal counts for P2-P3.0.02 [3][4] ed in lactophenol mounting medium.Broken glass fi bres were added to prevent the animal and appendages from being compressed by the coverslip and to facilitate rotation and manipulation, allowing observation from all angles.Preparations were subsequently sealed with Glyceel or transparent nail varnish.All drawings were prepared using a camera lucida on a Zeiss Axioskop diff erential interference contrast microscope.Total body length was measured from the anterior margin of the rostrum to the posterior margin of the caudal rami.Measurements were made with an ocular micrometer.Scale bars in illustrations are in μm.Th e descriptive terminology is adopted from Huys et al. (1996).
P6 forming well developed operculum with 2 small setae in ♀; asymmetrical in ♂ (with dextral or sinistral confi guration), with outer distal corner produced into conical process bearing 2 well developed elements.
Etymology.Th e name of the new genus honours the advances in our understanding of marine ecosystem functioning stimulated by the MarBEF Network of Excellence in Marine Biodiversity and Ecosystem Functioning.Gender: feminine.(Hamond, 1968), comb.n.
Material examined.
Description.Female.Body (Figure 1A, B).Total body length 696-738 μm (n=2; mean 717 μm).Largest width at posterior margin of cephalic shield (174 μm).Urosome gradually tapering posteriorly (Figure 1A).Rostrum triangular (Figure 2C), with pair of sensilla near apex; midventral tube-pore in subapical position; completely defi ned at base.Cephalothorax with posterior margin fringed with small spinules; pleural areas well developed and rounded with lobate posterolateral angles; entire surface covered with tiny spinules as illustrated in Figure 1A, B; sensilla and few pores present as illustrated in Figure 1A, B. Pedigerous somites covered with minute spinules.All prosomites without defi ned hyaline frills; posterior margins fringed with small spinules.Body constricted between individual somites.Urosome (Figures 1A, B; 2A) 5-segmented, comprising P5-bearing somite, genital double-somite and 3 free abdominal somites.Abdominal half of genital double-somite and second abdominal somite with lateral lobate extensions.All urosomites with surface ornamentation consisting of small spinules dorsally and laterally; ventral surface largely naked except for few rows of spinules and paired tube-pores; ventral hind margin with large spinules laterally and small spinules medially (Figure 2A).Hyaline frills of urosomites not distinct.Genital double-somite (Figures 1A; 2A) with transverse, surface ridge dorsally and laterally, indicating original segmentation; completely fused ventrally.Genital fi eld (Figure 2G) with medium-sized copulatory pore (arrowed in Figure 2G) located in median depression; gonopores fused medially forming single genital slit covered on either side by operculum derived from sixth leg; P6 with small protuberance bearing 2 bare setae; small triangular process located on medial side adjacent to protuberance.Anal somite (Figure 2B) with spinular ornamentation and pair of dorsal tube-pores; anal operculum well developed, bearing row of minute spinules and fl anked by pair of sensilla; anal opening bordered by well developed frill bearing long setular extensions.
Caudal rami (Figure 2A, B).Short, cylindrical, as long as wide; each ramus with 7 setae: seta I subventral, bare and shortest; setae II and III bare; setae IV and V bipinnate, fused basally, and with predesigned fracture planes [seta V presumably longest but incomplete in all specimens examined including holotype, cf.Hamond (1968: Fig. 3)]; seta VI bipinnate; seta VII tri-articulate at base.Each ramus with spinules on outer dorsal surface; additional spinular ornamentation present along inner and outer margins and around ventral and dorsal hind margins; long tube-pore present near ventral hind margin.
Antenna (Figure 3A).Th ree-segmented, comprising coxa, allobasis and free 1-segmented endopod.Coxa small, with 2 rows of spinules.Allobasis not elongate; without distinct surface suture marking original segmentation but with membranous insert around base of exopod; with one abexopodal bipinnate seta in distal half.Exopod small, twice longer than width, with 4 well developed pinnate setae (2 laterally, 2 apically), distal lateral and inner distal ones with fi ner spinules; one row of coarse spinules on anterior surface.Endopod shorter than allobasis; lateral armature arising in distal half, consisting of one bare seta fl anked by 2 strong pinnate spines; apical armature consisting of 2 strong spines and 3 geniculate setae (one geniculate seta fused basally to short seta).Endopod with 2 rows of long spinules laterally and 2 transverse hyaline frills subapically.
Labrum.Well developed; spinular ornamentation on anterior surface as in Fig- ure 2E.
Paragnaths (Figure 2F).Strongly developed paired lobes with medially directed hair-like setules, separated by medial lobe covered by dense pattern of short setules.
Maxillule (Figure 3C).Praecoxa with few long spinules around outer margin; arthrite strongly developed, with a tube-seta on anterior surface and 8 spines/setae around distal margin; one transverse row of long spinules on posterior surface, and one row of small spinules along inner margin of arthrite.Coxa with cylindrical endite bearing a naked seta and a curved, pinnate spine; with spinular row on anterior surface.Basis with cylindrical endite bearing 2 naked setae and a pinnate spine; with several spinular rows on anterior surface.Endopod completely incorporated into basis, forming cluster of 3 plumose setae; exopod 1-segmented, with 2 plumose setae and several rows of fi ne spinules on surface.
Maxilla (Figure 3D).Syncoxa with 3 endites; with a row of long spinules along outer margin, several rows of spinules on posterior surface, and several spinules along inner margin; praecoxal endite small and cylindrical, with one bare seta; both coxal endites with a strong pinnate spine and 2 pinnate setae with subapical tubular extension; proximal coxal endite with one row of spinules anteriorly.Allobasis drawn out into strong, slightly curved, distally pinnate claw; accessory armature consisting of 2 pinnate and 2 naked setae.Endopod represented by one bipinnate and 3 naked setae, surrounded by spinules at base.
Maxilliped (Figure 3E).Syncoxa elongate, with 2 plumose setae and two rows of spinules.Basis with one row of spinules along outer margin, palmar margin completely bare.Endopod drawn out into long claw; with one short accessory seta anteriorly, and one tube-pore posteriorly.
Fifth pair of legs (Figure 6E).Baseoendopods not medially fused, with distinct medial extension.Exopod and baseoendopod discrete, each with pattern of spinules as fi gured; spinular pattern on exopod denser than on baseoendopod.Baseoendopod forming long, outer setophore bearing basal seta and rows of spinules; with 3 tubepores on anterior surface; endopodal lobe just extending to proximal area of exopod, with 2 apical and 3 lateral bipinnate setae, proximal and inner middle ones minutely pinnate.Exopod elongate, about 3 times longer than wide; with one naked terminal seta, one bipinnate inner seta, and one bare and 3 pinnate setae along outer margin; terminal seta arising from short cylindrical process.
P4 (Figure 4C).Exopod as in ♀.Endopod relatively shorter than in ♀.Enp-1 with several spinules along inner margin (arrowed in Figure 4C); inner seta absent; outer lateral margin without setular extensions found in ♀.Enp-2 twice longer than enp-1; both apical setae about equally long and longer than in ♀.
Fifth pair of legs (Figure 6D).Baseoendopods fused medially, with spinular patch medially.Baseoendopod with large setophore bearing outer basal seta; endopodal lobe vestigial bearing one pinnate and one bare seta; with one tube-pore along medial margin, and one tube-pore near articulation with exopod; with few coarse spinules at base of setophore.Exopod about twice as long as maximum width; with one inner, one apical and 2 outer setae; all setae pinnate except for distal outer one being smooth; anterior surface with several large spinules.
Sixth pair of legs (Figure 6D).Asymmetrical; represented on both sides by small plate (fused to ventral wall of supporting somite on one side; articulating at base and covering gonopore on other side; dextral and sinistral confi gurations observed); outer distal corner produced into cylindrical process bearing several spinules, one bipinnate inner and one naked outer seta.
Discussion.Hamond's (1968) original description of the female of Pseudonychocamptus carthyi is detailed by contemporary standards.Our redescription revealed some minor observational errors: (1) the entire body is covered by fi ne spinules instead "… a few long limp hairs, widely scattered here and there"; (2) the third antennulary segment has 7 setae instead of 5; (3) the distal lateral and inner distal setae of the antennary exopod have fi ne spinules instead of membranous laminae; (4) the antennary endopod has an additional subdistal spine and the outermost geniculate seta is fused at the base to a minute seta; (5) the basis and coxal endite of the maxillule have an ad-ditional seta; (6) the maxilla has a unisetose praecoxal endite and the endopod 4 setae instead of 2; (7) the P1 displays distinct spinule rows along the outer margin of the praecoxa and coxa, and enp-2 has a minute accessory seta at the base of the claw; (8) the outer spines of P2-P3 exp-2 and -3 are bipinnate instead of smooth; and (9) the caudal ramus has 7 setae (seta I was overlooked).
Th e discovery of the male of P. carthyi provides conclusive evidence not only for the inclusion of this species in Lee and Huys' (1999) PWPH-clade (grouping the genera Pilifera, Pseudonychocamptus, Weddellaophonte and Heteronychocamptus) but also for its exclusion from the genus Pseudonychocamptus.Th e sexual dimorphism in swimming leg armature (the inner seta of P3-P4 enp-1 being absent in the male) unequivocally links this species to the PWPH-clade (Table 1).Th e sexually dimorphic P4 endopod, being 2-segmented in the female but reduced to a minute bisetose segment in the male, is the most signifi cant apomorphy of the genus Pseudonychocamptus, however, no such reduction is displayed by the male of P. carthyi.Th is species also displays the most primitive armature on P2-P5 within the PWPH-clade (Table 1), lending additional support to a diff erent generic assignment.Th e only other genus exhibiting two inner setae on P3-P4 exp-3 is Weddellaophonte, however, in P. carthyi two additional setae are found on P3 enp-2 (formula 321).Pseudonychocamptus carthyi is here fi xed as the type of a new genus, Marbefi a, gen.n., which can be defi ned by the following autapomorphies: (1) P2-P4 enp-2 (except P4 enp-2 in ♂) infl ated, with weakly chitinized outer margin bearing two parallel rows of fl imsy setular extensions, possibly surrounding a glandular structure; a similar but non-homologous structure is found on the P4 endopod of species of Corbulaseta Huys &Lee, 2000 (Huys andLee 2000;Gómez and Boyko 2006); (2) P4 with sexually dimorphic setation on enp-2 (outer distal seta reduced in ♀, very long in ♂); and (3) antennule ♀ with spinules along posterior margins of segments 1-6.Fiers (1992: 213) hinted at a relationship between Pseudonychocamptus and the monotypic genus Harrietella T. Scott, 1906 (type: Laophonte simulans T. Scott, 1894a) but provided no supporting evidence for this claim.Presumably his assertion was adopted from Lang (1948Lang ( : 1449) ) who recognized a similar relationship based on the shared more or less dorsoventrally fl attened body and the 2-segmented P3 endopod and 1-segmented P4 endopod in the male, in addition to the overall similarity in the maxilliped, P1 and male P3 exopod.Most Pseudonychocamptus species, however, do not have a dorsoventrally depressed body and the general habitus is considerably diff erent from that of Harrietella simulans (T.Scott 1894a; Sars 1920b).In Pseudonychocamptus the P4 endopod is suppressed during development in the male but not in the female, resulting in diff erent segmentation and setation patterns, whereas in Harrietella both sexes display a 1-segmented endopod and no sexual dimorphism in setation is discernible (e.g.Vervoort 1950;Hicks 1989).Th e 2-segmented P3 endopod in the male, displaying a straight spinous outgrowth on the distal segment, is indicative of a relationship with Pseudonychocamptus, Pilifera, Inermiphonte, gen.n. (see below) and Marbefi a, gen.n., however, Harrietella does not display the sexual dimorphism (on P3-P4 enp-1) that defi nes Lee and Huys' (1999) PWPH-clade.Although this absence is conceivably secondary as a result of the loss of the inner seta on P3 enp-1 in the female, confi rmation of its membership of this clade will probably have to await the arrival of molecular sequence data.Harrietella is here tentatively regarded as the sistergroup of Inermiphonte, gen.n. (see below).
Both Sars (1908a) and Wilson (1932) described Laophonte proxima on the basis of females only (the latter author from a freshwater locality!)whereas Klie (1929) provided the fi rst illustrations of the male, including the P5 which he fi gured with one seta on the endopodal lobe.Lang (1965) distinguished the closely related Pseudonychocamptus proximus and P. paraproximus on the basis of morphometric diff erences in the distal segment of the P4 exopod and P5 baseoendopod in the female and the P5 exopod in the male.An additional diff erentiating character used in his key referred to the number of setae on the male P5 baseoendopod.Hamond (1968) and Ceccherelli (1988) followed Lang's judgement, however Mielke (1975), in his redescription of the male of P. proximus, pointed out that the latter has two setae on the P5 baseoendopod (as in all other congeners), rendering the distinctiveness of P. paraproximus doubtful.Although we have followed Bodin (1997) and Wells (2007) in considering the latter a species of uncertain status (here ranked as species inquirenda), we have nevertheless included it in the key below, based on the interspecifi c diff erences displayed in the length/width ratio of the male P5 exopod (Table 2).
Recently, Apostolov (2008) added a new species, P. kolarovi, based on two males collected from the Kavala beach (Greece) in the Aegean Sea.Th e author claimed that the species occupied an isolated position in the genus on account of the structure of the caudal rami and P1-P5.Th e 2-segmented P4 endopod (with one inner and two distal setae), the presence of 5 setae on the P5 exopod and the sexually dimorphic distal inner spine on the P2 endopod clearly exclude P. kolarovi from the genus Pseudonychocamptus and particularly the latter character unequivocally points to a relationship with the genera Paralaophonte Lang, 1944 andLoureirophonte Jakobi, 1953.Th e species is here formally placed in the genus Paralaophonte as Paralaophonte kolarovi, comb.n. (see below).Th e six valid species currently recognized in the genus, and the problematic species P. paraproximus, can be differentiated by the key below and the character states tabulated in  Although his description of L. danversae was based solely on females from West Runton the author also illustrated a single male from the same locality under the nondescript name "Laophontid male, ?gen., ?sp.".Hamond (1969) suggested that the latter might have been the unknown male of L. danversae, were it not for the marked diff erences observed in the rostrum, caudal rami and the armature on the P2-P4 exopods.In an addendum to his paper he also remarked that "Laophontid male, ?gen., ?sp." and the male of Laophonte ?drachiMédioni & Soyer, 1966 shared the same type of sexual dimorphism on the P3 endopod and that the latter species was obviously closely related to L. danversae.As pointed out by Hicks (1982), Bodin (1971) clearly misinterpreted Hamond's remarks when he suggested conspecifi city between L. danversae and L. ?drachi.Th is misconception was perpetuated in the literature by Wells (1976) who considered Hamond's "Laophontid male, ?gen., ?sp." the true male of L. danversae and subsequently by Bodin (1979) who regarded both female and male L. danversae as junior synonyms of L. ?drachi.Th e issue of the unknown male of L. danversae was fi nally resolved by Hicks (1982) who provided its fi rst description based on material from Robin Hood's Bay (England).Hicks demonstrated that Hamond's (1969) "Laophontid male, ?gen., ?sp." and the males of L. danversae and L. ?drachi were signifi cantly diff erent in the swimming legs, leg 5 and caudal rami, corroborating their distinct specifi c identity but his comparative analysis did not, however, contribute anything new to elucidating the relationships of this species group within the Laophontidae.Hamond (1969) had previously suggested that L. danversae might best be classifi ed as an "anomalous species" of Laophonte, which does not fi t into any of the species groups proposed by Lang (1948).Médioni and Soyer (1966) tentatively assigned L. ?drachi to the inopinata-group of Laophonte and noted some similarities with L. ?platychelipusoidesNoodt, 1958(currently placed in Coullia Hamond, 1973;cf. Hamond 1973;Huys 2009a).Lee and Huys (1999) instead recognized a relationship between L. danversae and the laophontid genera that display a similar type of setal sexual dimorphism on the endopods of P3-P4 (Table 1) but neither formally included the species in the PWPH-clade or removed it from its fl oating status in the genus Laophonte.Laophonte danversae is here fi xed as the type species of a new genus Inermiphonte, gen.n. which also includes Hamond's (1969) "Laophontid male, ?gen., ?sp.".Inermiphonte is provisionally regarded as the sistergroup of Harrietella based on the following synapomorphies: (1) P2-P4 exp-3 ♀/♂ without inner setae, (2) P4 exp-2 without inner seta (exp-2 not expressed in Harrietella but inner margin of distal exopod segment without inner setae), (3) P3 enp-1 ♀ without inner seta, (4) P3 enp-2 ♀ with 0-1 seta, and (5) P4 enp-2 ♀ without inner setae.
Diagnosis.Laophontidae.Body linear or dorsoventrally depressed.Posterior margins of abdominal somites usually with spinule rows dorsally and ventrally.Rostrum delimited at base; broadly triangular, prominent.Genital double-somite ♀ with lateral incisions and ventrolateral internal chitinous ribs marking original segmentation.Pleural extensions of ♀ abdominal somites moderately developed.Caudal ramus rectangular or conical, longer than wide; inner margin with spinules; with 7 setae; setae IV and V well developed, pinnate; seta VI reduced, setiform.Anal operculum fi nely spinulose or smooth.
P6 forming well developed operculum with 2 small setae in ♀; asymmetrical in ♂ (with dextral or sinistral confi guration), with outer distal corner produced into process bearing 2 well developed elements.
Following Hicks' (1982) comparative analysis Hamond's (1969) "Laophontid male, ?gen., ?sp." is here assigned to a new species Inermiphonte hamondi, sp.n., which can be diff erentiated from the type species by the characters listed in Table 1 and Hicks ' (1982: 306) Table II.Laophonte ?drachi shows some important diff erences with the other two species such as the 6-segmented antennule in the female, the 2-segmented P1 exopod, the reduced inner distal setae on the female P2-P4 exp-3 (a character typical for male Inermiphonte), and the presence of three strong spines on the male P4 endopod.L. ?drachi resembles I. danversae in the presence of only one inner seta on P2 enp-2, no inner setae on ♂ P3 enp-2 and only three outer spines on P4 exp-3; it is similar to I. hamondi in the presence of only four elements on the male P5 exopod.It is here placed as species incertae sedis in Inermiphonte as I. drachi, comb.n.
Inermiphonte danversae is thus far known only from two localities in England.Hamond (1969) recorded three females from the shore at West Runton (Norfolk) and Hicks (1980Hicks ( , 1982) ) reported the species from a range of littoral algae in Robin Hood's Bay (North Yorkshire) where it attained highest densities on Corallina offi cinalis L. and Cladophora rupestris (L.) Kutz.Inermiphonte hamondi and I. drachi are known only from their respective type localities, the former from the intertidal zone at West Runton (Hamond 1969) where it may co-exist with I. danversae, the latter from colonies of the bryozoan Schismopora armata (Hincks, 1860) near Racou (Roussillon, France) (Médioni and Soyer, 1966).
According to the latest checklist (Wells 2007), the genus Paralaophonte currently includes 33 valid species and one species incertae sedis (Laophonte lamellipes Nicholls, 1944).Various authors (e.g.Willey 1935;Nicholls 1945;Yeatman 1970;Hamond 1973;Wells and Rao 1987;Wells 2007) have commented on the diffi culties in separating the two most widely distributed species, P. brevirostris (Claus, 1863) and P. congenera (Sars, 1908b), and some of them have casted doubt on the validity of this separation.Both species have also been reported to display a great deal of variability in the rostrum, antennule, leg 5 and caudal rami (e.g.Wells and Rao 1987) but it is unclear to what degree this pertains to genuine intra-or interpopulation variability or whether this is a refl ection of allopatric or sympatric closely related species (as hinted at by Mielke 1981).Th e further subdivision of both species in subspecies has also led to taxonomic confusion.For example, Willey's (1935) variety "fi ssirostris" of P. brevirostris was upgraded to subspecies level by Lang (1965) (and accepted by most subsequent authorities: e.g.Wells et al. 1982;Wells 2007) but allegedly diff ers from the nominotypical subspecies only in the shape of the rostrum.Given the reported variability in this structure in other species it seems impractical to maintain Willey's (1935) subspecies at present and hence they are here formally declared invalid.Similarly, P. congenera mediterranea Lang, 1948 appears to be more closely related to P. lacerdai Jakobi, 1953 than to its nominotypical subspecies P. congenera congenera.In fact both taxa can only be separated by subtle diff erences in the caudal ramus, length of endopodal setae in the female P5 and shape of the endopodal margin in the male P5.Likewise, Paralaophonte pacifi ca galapagoensis Mielke, 1981 is more similar to P. brevirostris than to its nominotypical subspecies P. pacifi ca pacifi ca Lang, 1965 (see also Wells 2007); the fi rst two taxa diff er primarily in details of the female antennule, P4 endopod and the male P3 exopod.Mielke's (1981) subspecies is here upgraded to full species rank since it diff ers suffi ciently from Lang's (1965) population to warrant such status: (a) second antennulary segment with blunt protuberance instead of spinous projection; (b) P1 enp-1 10 times instead 8 times as long as wide; (c) P2-P3 exopods ♀ distinctly shorter; (d) P3 exp-3 ♂ inner distinctly longer instead of shorter than outer distal spine; (e) baseoendopod P5 ♂ with strongly developed seta instead of tiny reduced seta; and (f) caudal ramus 1.3 times as long as maximum width instead of quadratic.Brady's (1910) original description of P. gracilipes (as Laophonte gracilipes) from the Kerguelen Islands (Grande Terre, Baie de l'Observatoire) is fragmentary and lacking in detail.Lang (1936) redescribed the species based on material from the Falklands and Pallares (1968) did the same using material from Puerta Deseado (Santa Cruz, Argentina).However, both redescriptions diff er signifi cantly in the number of inner setae on P3-P4 exp-3, the morphology of the female P5, the size of P1 exopod (relative to the endopod) and the length of the apophysis on the male P3, strongly suggesting that the Falklandian and Argentinian populations belong to diff erent species.Wells et al. (1982) and Wells (2007) adopted Lang's (1936) swimming leg armature pattern in their respective identifi cation keys, however, both incorrectly stated that the male P5 endopodal lobe lacked setae.Although Lang (1936) showed an endopodal seta, their error presumably stemmed from a misinterpretation of Brady's (1910) original fi gure of the male P5 which also shows the endopodal seta but not the typical outer basal one.Although the supporting evidence is admittedly slim (morphometric diff erences in P1 exopod and P2 endopod ♂) we have elected to attribute the Falklandian material to P. gracilipes and to propose a new species, P. pallaresae, sp.n., for the specimens illustrated by Pallares (1968).It should be noted that Pallares fi gured the female antennule as 8-segmented but stated in the text that it consists of only 7 segments (which is conceivably the correct number).Paralaophonte obscura Vervoort, 1962 from New Caledonia appears to be most closely related to P. pallaresae, sp.n. but diff ers from the latter in the female by the presence of a small conical protuberance on the second antennulary segment and the semicircular P5 exopod in which the inner distal seta is only slightly longer than proximal inner one, and in the male by the presence of only 4 elements on the P5 exopod.Paralaophonte pallaresae, sp.n. can be characterized by the following diff erential diagnosis: Paralaophonte.Abdominal somites without a middorsal spinous process.Caudal rami cylindrical, about 1.5 times as long as maximum width; with seta V elongate.Antennule ♀ 7-segmented; segment 2 without horn, spinous process or protuberance.P1 exopod 3-segmented.P3-P4 exp-3 with 1 inner seta.P3 enp-2 ♀ with 5 elements.P3 endopod ♂ 3-segmented with apophysis on enp-2 and 4 elements on enp-3; apophysis distinctly longer than endopodal segments combined.P4 enp-2 with 4 elements.P5 exopod ♀ longer than wide, inner distal seta twice the length of proximal inner one; baeoendopod with 4 elements.P5 exopod ♂ with 5 elements; longest seta at most 2.5 times the length of segment; endopodal margin with 1 long seta.Body length: 590-670 μm (♀), 500-510 μm (♂).
Th e original material collected by Dr Rosa Pallares is no longer available for re-examination (Dr S. Ménu-Marque, Universidad de Buenos Aires, pers.commn).In accordance with ICZN Arts 16.4 and 72.5.6 the male specimen illustrated by Pallares (1968: 87) in her plate XXXII (Figs 1,5,9,12,15) is here fi xed as the holotype of P. pallaresae, sp.n.Type locality: Puerto Deseado, Santa Cruz Province (Argentina), plankton.Jakobi (1953) proposed the genus Loureirophonte for two new species from Southern Brazil, L. catharinensis (type by original designation) and L. paranaensis and claimed a close relationship with the inopinata-group of Laophonte.Both Vervoort (1964) -who favoured a relationship with the inornata-group of Laophonte -and Lang (1965) questioned the accuracy of Jakobi's illustrations and generic diagnosis and doubted the validity of Loureirophonte until Mielke (1981) confi rmed its separate identity by providing a detailed description of a third species, L. isabelensis Mielke, 1981.Th is course of action was corroborated by Fiers (1993) in his revision of the genus which saw the addition of fi ve new species and another two which were transferred from other genera (Laophonte caesarea Por, 1964; Paralaophonte subterranea Lang, 1965).With the recent addition of L. psammophila Mielke, 2001 andL. minutum Gómez &Boyko, 2006 the genus now includes 12 species (Mielke 2001;Gómez and Boyko 2006) (Table 3) although it is known that other as yet undescribed species occur in the Galapagos (Mielke 1981).Fiers (1993) provided a key to species which was recently updated by Gómez and Boyko (2006).Fiers was also the fi rst to claim a sistergroup relationship between Paralaophonte and Loureirophonte based on the sexual dimorphism of the P2 endopod, the presence of 5 setae on the P5 exopod of both sexes, and the presence of 4 setae and one seta on the P5 endopodal lobe in females and males, respectively.He also listed a suite of characters that serves to distinguish Loureirophonte from its sistertaxon: (a) P2 enp-2 ♀ with 1 distal and 2 inner setae (0.210); (b) bulbous appearance of the sexually dimorphic distal inner seta of P2 enp-2 (with a lamellar structure in the distal third); and (c) 1-segmented P4 endopod.Characters (a) and (c) are autapomorphies of Loureirophonte (although Fiers (1993) did report specimens of L. mediterranea that displayed a 2-segmented P4 endopod) supporting the monophyly of the genus.Both Paralaophonte and Loureirophonte display a homologous sexual dimorphism on the P2 endopod, involving the modifi cation of the distal inner seta on enp-2.In the former genus the seta has attained a further derived state of modifi cation (as described under character (b) above), raising the suspicion that Loureirophonte is merely nested within Paralaophonte.Indeed, at present there is not a single apomorphy supporting the monophyly of Paralaophonte, potentially rendering the latter paraphyletic exclusive of Loureirophonte.In addition, the dichotomy between the two genera that was traditionally based on diff erent types of P2 endopod sexual dimorphism appears to be false since it is neither jointly exhaustive nor mutually exclusive.Th e evolutionary transition between both genera is best demonstrated by the modifi cation in male Paralaophonte macera Sars, 1908b andP. asellopsiformis Lang, 1965 which is virtually identical to the Loureirophonte condition and, similarly, the state of the inner distal seta in male L. furcata Fiers, 1993 which resembles the Paralaophonte condition (Sars 1908b;Fiers 1993).Although the presence of such intermediate taxa casts further doubt on the validity of the basal split between Loureirophonte and Paralophonte we have refrained from synonyzing these genera pending a thorough phylogenetic analysis at species level.
Five species have been added to the genus Paralaophonte since Wells' (2007) checklist: P. lamellipes (Nicholls, 1944), comb. n., P. galapagoensis Mielke, 1981, grad. n Wells and Rao (1987) also recorded specimens with a 7-segmented antennule in the female and without a spur on the second segment; Hamond (1972) also observed the 7-segmented condition while Petkovski (1964) recorded a female with a 6-segmented antennule without a spur on the second segment.
According to Wells and Rao (1987) the shape and size of the rostrum and the cephalothorax remains as the only reliable criterion to distinguish P. brevirostris from P. congenera.Pending a re-examination of topotype material, Willey's (1935) variety fi ssirostris, which was upgraded to subspecifi c level by Lang (1965), is no longer recognized as valid.
b Sars' (1908b) original description of the male shows 2 setae on the P5 baseoendopod; re-examination of the type material revealed that the short inner element represents a tube-pore.Both Hamond (1972Hamond ( , 1973) ) and Wells and Rao (1987) pointed out the diffi culties in discriminating P. congenera and P. brevirostris.Hamond's (1969) male shows a spur on the second segment of the antennule [also present in the North Carolina material examined by Hamond (1973)] and a distinctly pitted cephalothorax; Sars (1908b) did not observe these characters in the Norwegian material.Yoo and Lee's (1995) record from the Yellow Sea displaying only 4 setae on P5 exopod ♂, appears to represent a diff erent species; Wells (2007) pointed out that their illustration of the P4 in reality refers to the P2.c Monard's (1935) report of an 8-segmented antennule in this species is considered unlikely (cf. Lee and Huys 1999).Th e species was originally described from Salammbô (Tunisia) and has not been recorded again until recently when Nurul Huda and Zaleha (2005) found it in Peninsular Malaysia; their record (and that of Zaleha et al. 2006) which gives only the armature formula of P2-P4 requires confi rmation.Goddard's (2006) claim that his material from Robinson Crusoe Island (Juan Fernández Archipelago, Chile) may be attributed to P. octavia cannot be substantiated because it displays only one inner seta on P3-P4 exp-3; the only other known species which displays the same reduced formula is P. zimmeri but this species has a 6-segmented antennule in the female and a small spinous process on its second segment [8-segmented (?) and without such process in Goddard's material].
g Kornev and Chertoprud (2008) confi rmed the presence of a small spur on the second antennulary segment of the female [overlooked in Sars' (1908b) original description and Chislenko's (1967) redescription]; their swimming leg armature given on p. 311 contradicts the illustrations of P2 and P4.
h Although Hamond (1973) showed no hesitation in placing P. sculpta in the gracilipes-group of Paralaophonte, its generic assignment remains doubtful because of the absence of males.
i Chislenko (1977) did not fi gure the male P3 or P4, suggesting there is no sexual dimorphism in these legs; if this proves to be correct P. innae is suffi ciently diff erent from P. perplexa to warrant distinct species status, however, females of these two species may be extremely diffi cult to separate.j Chislenko's (1967)  exp-3 ♂ with only 1 inner seta instead of 2; and (6) P5 exopod ♂ and its middle element distinctly longer.Unfortunately, Kornev and Chertoprud's (2008) report on P. karmensis from the White Sea did not resolve the issue but instead added to the confusion surrounding it.Th ey confi rmed the presence of the antennulary spur but presented a radically divergent armature formula for P1 (exopod 3-segmented), P3 (exp: 0.1.123;enp: 0.121) and P4 (exp: 0.0.123) and considered the number of setae on the P5 exopod ♀ variable (4 or 5).
l Apostolov (2008) shows 2 inner and 2 distal setae on P3 enp-2 ♂ (formula 220); the 2-segmented condition and absence of an apophysis suggest that the outer spine is lacking in the female (as in e.g.P. innae); the number of inner setae expressed in P. kolarovi females may be 2 or 3 (in the latter case one seta is typically lost in male Paralaophonte).
m Nicholls (1944) stated that the P1 exp is 2-segmented, the distal two segments being fused with only 3 terminal setae (in reality the distal segment bears 5 elements); Huys and Lee (2000: 81) confi rmed the true nature of the modifi ed male P2 endopod and suggested a relationship with Paralaophonte; Wells (2007: 64) placed it as species incertae sedis in this genus; it is here formally transferred to Paralaophonte as Paralaophonte lamellipes (Nicholls, 1944) comb.nov. Lee and Huys (1999: 324) also confi rmed that the female antennule is 7-segmented instead of 8-segmented.
n Th e modifi ed inner distal seta resembles more the condition in Paralaophonte (Fiers 1993: Fig. 2c).
p Fiers (1993) doubted the presence of an outer seta on P3 enp-2 and P4 endopod; these "elements" are here reinterpreted as ornamental spinules.
r Th e inner setae of P2 enp-2 are displaced towards the apical margin of the segment creating the impression that there are 1 inner and 2 distal setae instead of 2 inner and 1 distal setae; taking into account the observed variability Lang (1965) erroneously tabulated the setal formula as 0.(0-1)20; Fiers' (1993) formula [0.
P. harpagone Gheerardyn, Fiers, Vincx & De Troch, 2006, P. kolarovi (Apostolov, 2008), comb. n. and P. pallaresae, sp. n. Although Wells (2007) listed Laophonte royi Jakubisiak, 1932 only as a species incertae sedis in the Laophontidae it is here included as a valid species of Paralophonte, being very closely related to P. majae Petkovski, 1964.Th e 39 species currently recognized as valid in the genus Paralaophonte can be identifi ed with the key below.