Holophryxus, RICHARDSON, 1905 A

Huys, Rony, Savchenko, Alexandra S. & Kosobokova, Ksenia N., 2023, Discovery of a new species and host record of Holophryxus Richardson, 1905 (Isopoda: Dajidae) from the central Arctic: a model of enhanced descriptive standards for epicaridean isopods, Zoological Journal of the Linnean Society 198 (2), pp. 592-649 : 601-606

GENUS HOLOPHRYXUS RICHARDSON, 1905 A

Synonyms

Hypodajus Nierstrasz & Brender à Brandis, 1931

(type species: Hypodajus georgiensis Nierstrasz & Brender à Brandis, 1931; by monotypy);

Cryptonus Schultz, 1977

(type species: Cryptonus truncatus Schultz, 1977 ; by original designation);

Isophryxus Schultz, 1977

(type species: Isophryxus concavus Schultz, 1977 ; by original designation).

Taxonomic history

Holophryxus Richardson, 1905a is one of 22 valid genera currently recognized in the family Dajidae and, together with Notophryxus , the genus with the greatest number of species ( Boyko et al., 2022). Richardson (1905a: 220) proposed the genus for its type and only species, Ho. alaskensis , and based her description on three adult females collected in south-eastern Alaska between 269 and 640 m depth. No males were collected and the host was unknown. Richardson (1905b: 576) reproduced her original description under the incorrect subsequent spelling Ho. alascensis and mentioned the discovery of another specimen collected by the U.S. Bureau of Fisheries steamer Albatross in southern California. Subsequently, other specimens of Ho. alaskensis became available, allowing the final host to be determined and the male and larval stages to be described. Butler (1964) provided descriptions of both sexes and the late juvenile stage based on material that was largely found attached to Pasiphaea pacifica Rathbun, 1902 (Pasiphaeidae) in British Columbian waters. Coyle & Mueller (1981) made a significant contribution to the life cycle of the species by describing the epicaridium, microniscus, cryptoniscus and juvenile stages based on material from Prince William Sound in Alaska. The discovery on the same shrimp host, P. pacifica , of adult females containing unmetamorphosed cryptoniscid stage males in the brood pouch, enabled them to identify the cryptoniscid larvae from plankton samples taken in the same area and trace the development back to the epicaridium stage. Coyle & Mueller (1981) identified the calanoid copepod Paraeuchaeta elongata ( Esterly, 1913) as the intermediate host and noted that the epicaridium stage attaches itself to the copepodid V instar, mainly by the first pair of pereopods and typically to the lateral side of the cephalothorax with its head directed anteriorly. Their observations revealed that final metamorphosis from the juvenile to the adult stage involved a change in mandibular morphology from the sagitate type to a rasp-like organ which drills a hole in the host’s carapace, the formation of the incubatory plates on pereopods 1–5, the elongation of the area between pereopods 5 and 6, with loss of pereopod 6 and reduction of its coxal plate to a small bump, and the complete loss of the coxal plate of pereopod 7. It is noteworthy that Coyle & Mueller (1981) had overlooked Rustad’s (1935) study which had documented similar morphological changes associated with the final metamorphosis in Ho. richardi (see below).

In a subsequent study, Richardson (1908: 690, 692) added two new species to the genus, Ho. giardi from 4941 m depth east of Bering Island (Commander Islands), representing the deepest record for the genus Holophryxus so far, and Ho. californiensis from 512 m depth off the coast of California and at 274–512 m depth in south-eastern Alaska. The former species was based on one detached adult female and an immature female attached dorsally to the carapace of the benthesicymid prawn, Boreogennema borealis ( Rathbun, 1902) . Wasmer (1988: 24) re-examined the syntypes and confirmed the presence of the spinular/ setular projections around the semicircular posterior margin of the fifth incubatory plates which had not been mentioned or illustrated by Richardson (1908). The species has not been recorded again since its original description, and the male and larval stages remain undescribed. Holophryxus californiensis was described on the basis of an adult female (with an accompanying male—at that time the first one in the genus) from green mud [California —possibly the specimen mentioned by Richardson (1905b) and an adult female attached to a shrimp host, Pasiphaea pacifica (Alaska) ]. The male was described in some detail but the description of the female was deficient. In both Ho. giardi and Ho. californiensis the attached adult females showed a similar orientation on the host’s carapace, with the cephalon directed towards the abdomen of the host. No information was provided on the morphology of the cephalic appendages in either species. Butler (1964) re-examined the type material of both Ho. alaskensis and Ho. californiensis and confirmed their conspecificity based on the variation in body shape observed in his own material from the British Columbia coast. The author confirmed P. pacifica as the host, showing that specimens attached to the shrimp were elongate and symmetrical [like the attached holotype designated as Ho. californiensis by Richardson (1908: fig. 4), whereas those found free in the trawl catches were stout and irregular in body form (like those described as Ho. alaskensis by Richardson (1905a)]. Butler (1964) postulated that the stout body form of the female is associated with the attainment of sexual maturity and renders mature females more prone to accidental detachment from their host. Nagler et al. (2020) erroneously listed Ho. californensis [sic] as a valid species.

Koehler (1911: 23) added a new species, Ho. richardi , from a plankton sample taken in the North Atlantic south of the Azores. A single immature female, without accompanying male, was collected at 0–2500 m depth, but the host was unknown. Koehler discussed similarities with the immature female of Ho. giardi , in particular the presence of traces of body segmentation in both forms. Stephensen (1912: 108, 109) recorded a detached female from the Davis Strait which he assigned to Ho. richardi based on the morphology of the fifth pair of incubatory plates. The specimen was about 2.5 times larger than Koehler’s (1911) holotype and lacked eggs or developing larvae in the marsupium, or an accompanying male. A second, larger, female specimen, found on the carapace of the prawn Eusergestes arcticus ( Krøyer, 1855) (Sergestidae) in a different locality in the Davis Strait, was tentatively assigned to ‘ Holophryxus sp. ( H. Richardi Koehler ?)’. Based on progressive changes in cephalon shape, incremental increases in body size and length/width ratio, and circumstantial evidence of host utilization, Stephensen (1912) believed that Koehler’s holotype and his own two Davis Strait females collectively formed a developmental series representing the same species. This claim was confirmed by subsequent re-examination of the material by Stephensen (1915: 24) who reported another detached female from the Celtic Sea. Individuals of Ho. richardi were found freely in plankton samples collected west and south of Iceland ( Hansen, 1916: 211), in western Norway ( Greve & Johannessen, 1981: 163) and various localities from east of Galicia to the Celtic Sea ( Jones & Smaldon, 1986; Table 2 ); at least in the latter two cases the likely host was Eusergestes arcticus because it was reported from the same station or trawl as the isopod. By far the most detailed account of the species is that by Rustad (1935) who studied 18 specimens at different stages of development, some of which he treated with a hot solution of sodium hydroxide and stained with pyrogallic acid (1,2,3-trihydroxybenzene) to reveal the detailed structure of the cephalic appendages. Rustad’s material was collected in the vicinity of Bergen, Norway, either as females attached to E. arcticus or as detached stages in plankton hauls that contained the host prawn. Although the quality of his study was exceptional by contemporary (and present-day) standards, it was not cited by any subsequent dajid authority for nearly half a century, presumably because it was published in the relatively obscure periodical Bergens Museums Årbok. His work was eventually rescued from oblivion when it was rediscovered in the late 1900s by Greve & Johannessen (1981). Based on marks left on the carapace of the host, Rustad (1935) concluded that the isopod moves repeatedly during its life cycle, with the mandibles perforating the carapace through to the epithelial and connective tissues underneath and the dactyli of the pereopods leaving shallower scars. His reinterpretation of the immature stages reported by Richardson (1908), Koehler (1911) and Stephensen (1912), in combination with personal observations of his Ho. richardi material, showed that the development of the juvenile female into the mature stage involved the loss of coxal plate 7 and pereopod 6 (with its coxal plate reduced to a small protuberance), the formation of incubatory plates 1–5, and the elongation of the body region between pereonites 5 and 6. Rustad (1935: 12) also suggested that the genus Holophryxus is possibly unique within the Dajidae in having the oostegites developed before the sixth pereopods have completely disappeared.

Stephensen (1912: 112) added Ho. acanthephyrae , from the Labrador Sea, based on a single adult female attached to a deep-water shrimp, with the head orientated towards the abdomen of the host. The type host was originally named as Acanthephyra purpurea A. Milne Edwards, 1881 but, on the advice of the Norwegian carcinologist Oscar Sund, the identification was later corrected to Acanthephyra multispina Coutière, 1905 ( Stephensen, 1913: 329) , which is now widely accepted as a junior subjective synonym of Acanthephyra pelagica ( Risso, 1816) (Acanthephyridae) . Although Stephensen (1912) recognized a certain resemblance with Ho. californiensis (= Ho. alaskensis ) and comparison with this species was hampered by Richardson’s (1908) incomplete description, differences in host utilization and type locality were considered sufficient evidence to assign distinct specific status to the Labrador Sea female. According to his text description and ventral view of the cephalon (his fig. 16), the antennules are 3-segmented and the antennae 5-segmented, but as Stephensen rightly assumed himself, the transverse folds in these appendages were misinterpreted as segment boundaries. His observation of a ‘gnathopod’ between the antenna and first pereopod is ambiguous, leaving open to interpretation which part of the cephalon he misidentified as a limb. Without dissection only two pairs of incubatory plates (probably IP1 and IP5) could be revealed and based on the marks on the carapace it was inferred that the parasite had changed position four to five times. In addition to observations of the female, Stephensen also provided detailed illustrations of epicaridium larvae obtained from the marsupium. No males were found. Gurjanova (1932) subsequently recorded the species at 2000 m depth off the Cape Farewell peninsula in southern Greenland. It is unclear whether her identification of A. purpurea as the host was correct or in reality referred to A. pelagica . Jones & Smaldon (1986) redescribed the female from specimens collected in the Celtic Sea (1500 m) and from the Biscay Abyssal Plain (1130–1800 m) and described the male for the first time. Schultz (1977), Wasmer (1988) and Alves-Júnior et al. (2019) added records from the South Atlantic, South Pacific and Southern Oceans making it the only species in the genus that occurs in both northern and southern hemispheres, being distributed from the Labrador Sea to the Amundsen Sea (see below).

Nierstrasz & Brender à Brandis (1931: 212) proposed the morphologically close genus Hypodajus for the newly described type and only species, Hyp. georgiensis , collected at 366 m depth in the Strait of Georgia in British Columbia, Canada. The single adult female (with accompanying male) was found on the ventral side of an unidentified pasiphaeid shrimp which in the vicinity of the type locality would almost certainly have been Pasiphaea pacifica (cf. Butler, 1964). Nierstrasz & Brender à Brandis provided no proper argumentation for the establishment of their new genus and the two generic keys they provided fail to shed light on the issue. In the first key the main character used to distinguish Hypodajus from Holophryxus relates to the segmentation of the body which is described in the former genus as ‘retained partially only in pereon and absent or nearly absent in pleon’ and in the latter as ‘absent in median part of pereon and completely absent or represented by small rudiments in pleon’. Although some faint transverse furrows are present in the anterior portion of the pereon in Hyp. georgiensis (Nierstrasz & Brender à Brandis, 1931: figs 106, 107), similar grooves can be found in some Holophryxus species. Such integumental folds do not reflect segmentation patterns and have no significance at generic or even species level. Their second key uses similarly dubious character states to separate both genera, casting further doubt on the validity of Hypodajus . Butler (1964: 975) pointed out that the number of incubatory plates remained the only difference between the descriptions of Hyp. georgiensis and Ho. alaskensis , but that this discrepancy was based on the failure to recognize the large ventral thoracic plates as the fifth pair of oostegites in the former. Hypodajus georgiensis was consequently relegated to a junior subjective synonym of Ho. alaskensis and Hypodajus as a junior synonym of Holophryxus ( Butler, 1964: 976) .

Shiino (1937: 188) described Ho. fusiformis based on numerous females (occasionally with males in the marsupium) found attached to the dorsal surface of the carapace of its host, Prehensilosergia prehensilis ( Spence Bate, 1881) (Sergestidae) , collected off the Pacific coast of Japan. Shiino claimed that both pairs of antennae and the maxillae were absent in the female and that the maxillipeds were divided into two parts. Reinterpretation of the cephalon in ventral aspect strongly suggests that the so-called bipartite maxillipeds in reality represent the antennules and adjacent antennae (compare Fig. 4A herein). According to Shiino (1937), Ho. fusiformis closely resembles Ho. giardi and Hyp. georgiensis (= Ho. alaskensis ), but differs from both species in the shape of the cephalic ridge (= ‘lunar-shaped lamella’ surrounding the frontal margin of the oral area), the presence of only four pairs of incubatory plates instead of five, and in having two pairs of small papillae at the posterior end of the pereon, either side of the hind margin of the fifth oostegites. Shiino (1937) also pointed out that Ho. fusiformis differs from Ho. giardi in the complete absence of the articulation between the pereon and pleon, and from Ho. alaskensis by the indistinct segmentation of the pereon in the female and by the shape of the pleon in the male abdomen. Holophryxus fusiformis remained unnoticed until Huang et al. (2018) reported it from Taiwan where it is commonly found attached to the Sakura shrimp, Lucensosergia lucens ( Hansen, 1922) (Sergestidae) . A single specimen was also recorded from the same host collected off Nagasaki, Japan, which led the authors to suggest that the parasite is dispersed from Taiwanese waters with the north-flowing Kuroshio current. Huang et al. (2018) redescribed both sexes in detail, rectifying some of the observational errors in Shiino’s (1937) description, and challenged the latter’s claim that only four incubatory plates are present in the Japanese material.

Schultz (1977: 93) established the genus Isophryxus for its type and only species, I. concavus , from various midwater trawl catches in the South Pacific and Southern Oceans. Both sexes were described but the host was unknown for all seven females obtained because the isopods apparently fell off their crustacean hosts at the time of collection. Schultz differentiated Isophryxus from Holophryxus solely by the dorsal expression of the pereonites in the former. Wasmer (1988) re-examined the type material of I. concavus and concluded that it was conspecific with Ho. acanthephyrae , thus relegating the generic name Isophryxus to a junior synonym of Holophryxus . He dismissed the significance of the dorsal transverse folds in the pereon of I. concavus as possible evidence of an ancestral segmentation pattern, confirmed the presence of the previously overlooked spinular/ setular projections (‘crests’) around the semicircular posterior margin of the fifth incubatory plates, and provided circumstantial evidence to suggest that A. pelagica was the unknown host of Schultz’s (1977) material. Additional observational errors in the original description were corrected by Wasmer (1988), including the supposed absence of antennules and antennae in the females (mistakenly interpreted as ‘flattened mouthparts’ by Schultz) and the misnumbering of the lateral edges of the pereonites (coxal plates). Alves-Júnior et al. (2019) recently recorded a single female of Ho. acanthephyrae from a new deep-sea host, Acanthephyra acanthitelsonis Spence Bate, 1888 (Acanthephyridae) , in the benthopelagic zone off Rocas Atoll. Their concise description does not add anything significant to the morphology of the species. The authors also erroneously cited Holthuis (1947) as one of the records of Ho. acanthephyrae , but that work only deals with the synonymy of the type host, A. pelagica (cf. Jones & Smaldon, 1986: 304).

Schultz (1977: 108) proposed the genus Cryptonus for its type and only species, Cryptonus truncatus Schultz, 1977 , found in bathypelagic trawl samples from 18 localities in the South Pacific and Southern Oceans. The species was based on cryptoniscus larvae and placed in the Cryptoniscoidea but without assignment to any particular family. Recently, Williams et al. (2022: 240) transferred the genus to the Dajidae and relegated it to a junior subjective synonym of Holophryxus based on the close morphological similarity between the cryptoniscus larva of C. truncatus and that described by Coyle & Mueller (1981) for Ho. alaskensis . Accordingly, they renamed the former under the new combination Holophryxus truncatus ( Schultz, 1977) and speculated that it may represent the larva of one of the four adult-based Holophryxus species known from the Southern Hemisphere ( Table 2 ) although none of them have been recorded as far south.

Schultz (1978) added three new species from the South Pacific to Isophryxus , all of which were based on females (and accompanying males in the brood pouch) found in plankton hauls: Isophryxus polyandrous Schultz, 1977 , Isophryxus quadratahumerale Schultz, 1977 and Isophryxus septapodus Schultz, 1977 . Wasmer (1988: 27) transferred the first two species (the second one misspelled as I. quadratohumerale ) to Holophryxus based on the presence of weak anterior furrows suggesting segmentation of the pereon, the presence of a series of spinular/setular projections around the semicircular posterior margin of the fifth pair of incubatory plates, and the morphology of the males. Schultz (1978) amended the generic diagnosis of Isophryxus to accommodate the alleged presence of seven pereopods in the adult female of I. septapodus and stated that the posterior margin of the marsupium lacks spinular projections. However, Wasmer’s (1988: 27) re-examination of the gravid female holotype proved them to be present, thus demonstrating that it is a specimen of Holophryxus in which the fringing marsupial spines are present, but which has retained vestiges of the seventh pereopods and their corresponding coxal plates (represented by the posterior two pairs of processes on the posteroventral part of the pereon), as well as jointed sixth pereopods and the corresponding sixth coxal plates (represented by the more anterior two pairs of processes on the posteroventral part of the pereon). Wasmer (1988) suggested that there may be individual variation regarding the stage of development at which the sixth pereopods finally disappear in Ho. septapodus [as previously recorded for Ho. richardi by Rustad (1935: 16, 17, figs 3, 6)]. Recently, Boyko & Williams (2021b) added new records of Ho. alaskensis and Ho. acanthephyrae from the Pacific and Atlantic Oceans, respectively, and reviewed the synonymy and bibliography of both species.

Generic diagnosis

The generic boundaries of Holophryxus have been blurred since the inception of the genus, being largely due to the choice of diagnostic characters employed by various authors. Richardson (1905a) differentiated Holophryxus from the seven dajid genera known at the time by a few selected key features (listed in parentheses): Dajus (no trace of segmentation, absence of pleopods), Aspidophryxus Sars, 1883 and Notophryxus (number of incubatory plates, no trace of segmentation, shape of oral area, etc.), Heterophryxus Sars, 1885 (position and morphology of last pair of pereopods), Branchiophryxus Caullery, 1897 (number of pereopods and incubatory plates), Zonophryxus (general body shape, form of pleon, complete absence of pleopods) and Prodajus (pleon unsegmented and not bifurcate). Nierstrasz & Brender à Brandis (1931) constructed two keys to the genera of the Dajidae and employed the following female character states in the identification of Holophryxus : (1) pereon and pleon with no or vestigial traces of segmentation, (2) posterior margin of pleon smooth (without spiniform processes), (3) presence of five, morphologically similar, pereopods, and (4) absence of uropods. Wasmer (1988: 24) identified the presence of spinular/setular extensions around the posterior margin of the fifth incubatory plates as a key diagnostic character of the genus, enabling him to assign the species previously described as members of Isophryxus ( Schultz, 1977, 1978) to Holophryxus . According to Williams & Boyko (2021: table 1), adult Holophryxus females can be differentiated from those of other dajid genera by the combined presence of a distinct cephalon, a pereon that displays rudimentary traces of segmentation and bears five pairs of pereopods and associated incubatory plates (oostegites), and an unsegmented pleon lacking pleopods and uropods. The first two character states are ambiguous because (1) the cephalon is fused to the pereon (with the first pair of coxal plates incorporated into the ventral cephalic ridge) and is only demarcated from it by dorsal and lateral constrictions, and (2) the surface folds or furrows sometimes present on the pereon do not reflect ancestral segmentation patterns. Williams & Boyko (2021: table 2) also compared the diagnostic characters of adult males among the 15 genera for which they are known and listed the following combination of character states as typical for Holophryxus : (1) antennule 1- to 3-segmented, (2) antenna longer, 1–3 segments?, (3) terminal pereonite fused to first pleonite, (4) seven pairs of pereopods, (5) pleon unsegmented, and (6) pleopods absent. Unfortunately, some of these states are scored incorrectly because they are based on previously published observational errors (see Discussion). For example, the antennule is unsegmented in all Holophryxus males known so far, the flagellum—when present—having been misinterpreted as one or two additional segments (e.g. Shiino, 1937; Jones & Smaldon, 1986; Huang et al., 2018). The antenna typically consists of a basal peduncle and an indistinctly segmented flagellate part while pereonite 7 is only occasionally (and only partly) fused to the pleon. The recently presented generic diagnosis by Boyko & Williams (2021b) omits characters related to the cephalic appendages and pereopods. Their statement that the ‘… anterior hood [is] square …’ in the adult female is also incorrect because it only applies to some species and is based on an ambiguous definition of the character.

Most generic diagnoses proposed in the Dajidae have traditionally been overly concise and incomplete, containing inadequate or misleading information, and characters whose states were wrongly assessed. Below we present an updated diagnosis of Holophryxus , combining both female and male character states, with the aim that it will serve as a model for future revisionary work on other dajid genera.

Dajidae . Females ectoparasitic on dendrobranchiate prawns (Sergestoidea, Penaeoidea) and caridean shrimps (Oplophoroidea, Pasiphaeoidea); typically attached dorsally to carapace with cephalon pointing towards pleon of host, occasionally to ventral surface of abdomen. Males living inside marsupial cavity of female. Eyes absent in both sexes.

FEMALE: Body unsegmented, without functional articulation between cephalon, pereon and pleon; surface folds or furrows sometimes present but not reflecting ancestral segmentation patterns. Cephalon relatively small, typically vaulted, narrower than pereon and demarcated from it by constrictions dorsally and bilaterally; anterior margin of dorsal protuberance rounded or emarginate, occasionally forming broad hood-like extension with straight anterior margin. Ventral cephalic ridge strongly developed, ranging in shape from slightly arched to distinctly crescentiform, embracing oral area; frontal part with or without paired pit organs; coxal plates of first pair of pereopods integrated into posterolateral corners of ridge, represented by lobate extensions or pointed processes. Antennules and antennae unsegmented, flattened and lobate appendages surrounding oral area. Oral opening bordered anteriorly by bilobate labrum and posteriorly by medially incised labium. Mandible slender, with cluster of densely arranged denticles forming abrasive surface around tip of gnathobase. Maxillipeds apparently present in some species; lamellar and originating from sternal plate. Median sternal plate with or without paired posteriorly directed appendices.

Pereon with five pairs of prehensile pereopods; coxal plates of pereopods 2–4 (and often 5) represented by rounded or acute projections posterior to cephalic ridge and lateral to bases of limbs. Each pereopod 5-segmented, consisting of basis, ischiomerus (fused ischium and merus), carpus, propodus and dactylus; carpus and propodus with sensory elements along medial margin; dactylus recurved and claw-like, with setae and sensory elements. Five pairs of incubatory plates (oostegites) arising from basal parts of pereopods; only first and fifth pairs visible, covering other three pairs; first pair occasionally bilobate; fifth pair largest, extending to posterior end of pereon and forming largest part of marsupial cavity; plates of fifth pair overlapping medially, each with series of spinular/setular projections around semicircular posterior margin.

Pleon unsegmented, without traces of original segmentation; representing at most one fifth of total body length; often demarcated from remainder of body by bilateral constriction; without pleopods. Anus subterminal, represented by small longitudinal slit ventrally. Uropods not expressed.

MALE: Much smaller than female. Body elongate, curved in lateral aspect; comprising cephalon, segmented pereon and unsegmented pleon. Cephalon large, with rounded or slightly bilobed anterior margin; completely (forming cephalothorax) or incompletely fused to pereonite 1; lateral margins typically indented, indicating boundary between cephalon and first pereonite; anterior and lateral margins with numerous sensory papillae. Antennule unsegmented, flattened; with or without flagellum. Antenna elongate, comprising (in)completely separated basal peduncle and indistinctly segmented flagellate part. Oral cone with well developed labrum and labium. Mandible slenderer and with fewer denticles on gnathobase than in female. Maxillipeds absent.

Pereon comprising six free pereonites (2–7), each with well developed, rounded pleurotergites; occasionally pereonite 7 partly fused to pleon. All seven pairs of pereopods prehensile, slenderer than in female; segmentation and presence of sensory elements as in female.

Pleon unsegmented, conical; except for papillae, without external traces of original segmentation; proportional length variable; without pleopods; with sensory elements and spinular combs around anal area. Anus subterminal, represented by small longitudinal slit ventrally. Uropods not expressed.

Type species: Holophryxus alaskensis Richardson, 1905a (by monotypy).

Other species: Holophryxus acanthephyrae ; Ho. citriformis ; Ho. fusiformis ; Ho. giardi ; Ho. polyandrus ( Schultz, 1978) ; Ho. quadratahumerale ( Schultz, 1978) ; Ho. richardi .

Species inquirendae: Ho. septapodus ( Schultz, 1978) ; Ho. truncatus ( Schultz, 1977) .

Figures

Figure 4. Holophryxus citriformis sp. nov. Holotype, adult female: A, anterior part of body, showing cephalic and thoracic regions with associated limbs, ventral view; left pereopods and incubatory plate 1 omitted to show position of incubatory plates 2–4 (A1 = antennule, A2 = antenna, CP 1–4 = coxal plates 1–4, CR = cephalic ridge, IP1–5 = incubatory plates 1–5, Md = mandible, P5 = pereopod 5, SP = sternal plate); B, posterior part of body, showing spinous margins of incubatory plates (IP5) and position of anal opening (indicated by arrow).