Psammonitocrella Huys, 2009

Karanovic, Tomislav & Hancock, Peter, 2009, On the diagnostic characters of the genus Stygonitocrella (Copepoda, Harpacticoida), with descriptions of seven new species from Australian subterranean waters 2324, Zootaxa 2324 (1), pp. 1-85 : 46-47

Genus Psammonitocrella Huys, 2009

Diagnosis emended. Small and very slender Ameiridae , with cylindrical habitus and no distinct demarcation between prosome and urosome. Integument weakly chitinized and without cuticular windows; hyaline fringe of all somites smooth. First pedigerous somite incorporated into cephalothorax. Prosome ornamented only with sensilla, urosome additionally ornamented with rows of small spinules. Genital somite in female free; genital field with single large copulatory pore, wide copulatory duct and two small semicircular seminal receptacles; single small genital aperture covered by fused reduced sixth legs, without armature or ornamentation. Anal operculum unornamented, wide and convex, not reaching to posterior end of anal somite. Caudal rami slender, cylindrical and slightly divergent, longer than anal somite; dorsal seta inserted at 4/5 and close to inner margin, less than half ramus length; proximal lateral seta arising somewhat dorsolaterally at middle; distal lateral seta arising at 4/5 and laterally; inner apical seta minute; principal apical setae without breaking plane, outer one as long as or shorter than ramus. Antennula long and slender, eight-segmented in female and ten-segmented and geniculate in male; without seta on short first segment. Antenna composed of coxa, basis, two-segmented endopod and one-segmented exopod; exopod armed with three setae. Mandibula with narrow cutting edge and two-segmented palp; basis unarmed, endopod with three to five apical setae. Maxillular endopod armed with two apical setae or with reduced armature. Maxilla with single endite on syncoxa; endopod fused to basis, represented by surface seta. Maxilliped three-segmented, unarmed. All swimming legs with three-segmented exopod. Endopod of first leg three-segmented; endopod of second and third swimming legs two-segmented or one-segmented, while endopod of fourth swimming leg reduced to small knob or completely absent. All exopodal segments of about same length; first exopodal segment of all legs without inner seta, second with; second exopodal segment of first leg without outer spine; third exopodal segment of first leg with three outer spines and no inner setae, that of second leg with one or two outer spines and no inner setae, third leg with one outer spine and no inner setae and fourth leg with one outer spine and one or no inner setae. First endopodal segment of first leg unarmed, large, reaching to posterior margin of second exopodal segment; penultimate endopodal segment of second and third legs (if present) unarmed, ultimate segment with one apical seta; basis of first leg in male with unmodified inner spine; no other sexual dimorphism in swimming legs. Fifth legs similar in both sexes, fused to somite, with or without recognizable endopodal lobe and with recognizable exopodal lobe; endopodal lobe (if present) armed with one element, while exopodal lobe bears three, two or only one seta.

Type species. Psammonitocrella boultoni Rouch, 1992 .

Other species. Psammonitocrella longifurcata Rouch, 1992 .

Remarks. The genus was described by Rouch (1992) to accommodate two interstitial species collected in the hyporheic zone of an intermittent desert stream in Arizona, USA. As mentioned in the Introduction section, Rouch did not provide the mandatory type fixation, and the generic name Psammonitocrella was unavailable until Huys (2009) fixed P. boultoni Rouch, 1992 as the type species, thereby making the name available under his authorship. Huys (2009) also treated both species as new combinations and cited author names in parentheses, just as Reid et al. (2003) did for the genus Stygonitocrella Reid, Hunt & Stanley, 2003 (see the Remarks section for this genus). Although neither Article 11.9.3 nor Article 51.3 of the ICZN specifically addresses these two cases, the generic names are not different from the original ones, since the author attribution does not form part of a name in zoological nomenclature (Article 51.1). Therefore, we think parentheses should not have been used and it was at least an overstatement to refer to such cases as new combinations.

The familial placement of Psammonitocrella was questioned by Martínez Arbizu and Moura (1994), who removed it from the Ameiridae Monard, 1927 and regarded it as a sister-group of the family Parastenocarididae Chappuis, 1940 . Lee & Huys (2002) proposed relocating it back to the Ameiridae , which has been accepted by subsequent workers (see Reid et al. 2003; Boxshall & Halsey 2004; Karanovic 2004a, 2006). The reason for all this confusion is the absence of sexual dimorphism on the inner basal spine of the first leg in this genus. The modification of this spine in male ameirids is one of the most important family autapomorphies, but the condition observed in Psammonitocrella is probably paedomorphic and it was also recorded in the completely unrelated Australian genus Archinitocrella Karanovic, 2006 . It apparently originated independently (as a convergence) in the two lineages and the main differences between them were explained by Karanovic (2006).

However, the genus Psammonitocrella remains well defined by many autapomorphic features, probably the most important one of them being the loss of the outer spine on the second exopodal segment of the first leg ( Table 2, character 18). Unlike all other ameirids studied here, this genus also lacks the ancestral inner apical seta on the third exopodal segment of the second, third and fourth legs (characters 31 & 33), as well as the outer middle spine on the third exopodal segment of the third and fourth legs (character 32). A number of reductions were also reported for the mouth appendages, antennula and fifth legs, which makes it very difficult to establish a close relationship with other freshwater ameirids.

As we mentioned above, reduction of the endopod of the fourth swimming leg to a small knob has originated convergently in Psammonitocrella , Kimberleynitocrella gen. nov. and Neonitocrella Lee & Huys, 2002 , while the first anennular segment without any armature is found in Psammonitocrella , Eduardonitocrella gen. nov. and Inermipes Lee & Huys, 2002 (see the Remarks sections for Kimberleynitocrella and Eduardonitocrella ). Caudal rami longer than the anal somite are only found in one other species studied here ( Stygonitocrella sequoyahi Reid, Hunt & Stanley, 2003 ), but this character seems to be of little phylogenetic importance outside the genus Psammonitocrella .

Several other important points one can learn from the two members of the genus Psammonitocrella are that characters of the fifth leg do not carry much phylogenetical importance (the variability of this character in P. longifurcata Rouch, 1992 is amazing) and the endopod of the second and third swimming legs can be reduced from a two-segmented state into a one-segmented state in a single event (at least in this group). However, the armature formula of the ultimate endopodal segment of the second to fourth legs is the same in both species (1.1.0) and is unique in Stygonitocrella s. l. That is why we decided to study the nature of the armature elements on the endopodal and exopodal segments, rather than using the segmentation of these rami as main generic characters, as was done in the past.

There are very few features to study in this group of highly reduced freshwater ameirids, where the number of homoplastic characters is expectedly very high. Different segmentation patterns between closely related species were also observed in the genera Stygonitocrella Reid, Hunt & Stanley, 2003 (as redefined above), Reidnitocrella gen. nov. and Megastygonitocrella gen. nov. One of the few plesiomorphic features of the genus Psammonitocrella is the inner seta on the second exopodal segment of the first leg (character 17), which shows that the ancestor of this genus could not have originated from a member of the genus Megastygonitocrella , but from some other, more primitive freshwater ameirid. The cladistic analysis presented in this paper shows that we should look for the ancestor of this North American genus somewhere in the genera Reidnitocrella or Stygonitocrella , the latter interestingly containing the only other North American representative of Stygonitocrella s. l.