Caridina sobrina Riek, 1953 stat. nov. (Figs. 1–3) Caridina indistincta sobrina Riek, 1953: 119, fig. 9.— Springthorpe & Lowry, 1994: 75.— Davie, 2002: 225.— Page et al., 2005: 139, 140. Caridina indistincta sp. C3 Page & Hughes, 2007a: 225. Type material. i) ovigerous female (most intact individual), pocl 3.5 mm, Rocky Creek, Fraser Island, leg. E.F. Riek, viii.1941; ii) 11 individuals (pocl 2.6–3.0 mm); iii) 19 ind. (pocl 2.8–3.5 mm); iv) 4 ind. (pocl 2.6–3.8 mm), dissected by A. Gurney (AM P13329). Lots i–iv separated, but in same jar; holotype designated by Riek (1953), not visibly separated nor separable in vial. Other material examined. 5 females (pocl 3.5–4.4 mm), Bowarrady Creek, Fraser Isl., 25.133°S, 153.165°E, leg. T.J. Page, J.H. Fawcett & B.D. Cook, 21.v.2003 (AM P107219); 4 females, 3 males (pocl 3.6–4.6 mm), Bowarrady Creek, Fraser Isl., 25.133°S, 153.165°E, leg. T.J. Page, J.H. Fawcett & B.D. Cook, 21.v.2003 (AM P107220); 1 ov. female, 4 females (pocl 3.5–4.1 mm), Rocky Creek, Fraser Isl., 25.473°S, 153.010°E, leg. T.J. Page, J.H. Fawcett & B.D. Cook, 22.v.2003 (AM P107221); 5 females, 1 male, 3 juveniles (pocl 2.9–3.8 mm), Rocky Creek, Fraser Isl., 25.473°S, 153.010°E, leg. T.J. Page, J.H. Fawcett & B.D. Cook, 22.v.2003 (AM P107222); female (pocl 4.3 mm; illustrated) Rocky Creek, Fraser Isl., 25.473°S, 153.010°E, leg. T.J. Page, J.H. Fawcett & B.D. Cook, 22.v.2003 (AM P107223); male (pocl 4.0 mm; illustrated) Rocky Creek, Fraser Isl., 25.473°S, 153.010°E, leg. T.J. Page, J.H. Fawcett & B.D. Cook, 22.v.2003 (AM P107224). Description. Rostrum usually reaching to distal margin of scaphocerite, occasionally slightly shorter or longer (Fig. 1A); upper margin straight, bearing 11–21 sutured, equidistant teeth, with additionally 0–2 placed behind orbit; rostrum deeper at distal one-third, gradually narrowing distally; ventral margin with 3–7 teeth, proximal quarter unarmed; tip simple. Lower orbital angle prominent, weakly triangular (Fig. 1A); antennal tooth separated from orbit, triangular, slightly longer than orbital angle; no other carapacial teeth; pterygostomial angle weakly quadrate (Fig. 1A). Eyes of usual form and shape for epigean Caridina (Fig. 1A); cornea well-pigmented, eye reaching to distolateral margin of stylocerite. Abdominal somites smooth, with rounded pleura; sixth somite twice as long as fifth; preanal carina with well-developed upright tooth (Fig. 4C). Telson broadest proximally, gradually tapering towards distal margin (Fig. 1D); bearing 4 pairs of dorsolateral spinules, off-set from lateral margin; distal margin with medial tooth (Fig. 1E), flanked by 4 pairs of setulose, spiniform setae, outer pair shortest, outer-most submedian pair longest. Antennular peduncle (Fig. 1B) with well-developed stylocerite, almost reaching to distomedial margin of basal segment; anterolateral lobe of basal segment well-developed, pointed; intermediate segment longer than distal segment; distal segment less than half length of basal segment. Scaphocerite (Fig. 1A, C) reaching to tip of rostrum, about 3 times as long as wide; outer margin straight, ending in well-developed tooth, falling short of distal margin of lamella. Mouthparts without special features, as illustrated (Fig. 2). First pereiopod robust, short (Fig. 3A), reaching to about tip of stylocerite; fingers short, blunt and stubby, slightly longer than palm; carpus weakly excavated distally, slightly shorter than chela; widening distally (prior to excavation) to almost twice proximal with; merus slightly longer than carpus; ischium shorter.Second pereiopod more slender than first (Fig. 3B), reaching to about distal margin of second article of antennular peduncle, occasionally shorter or longer; finger blunt, about 1.4–1.5 times as long as palm; carpus long, slender, barely distally excavated; about 1.4–1.5 times as long as chela, widening distally to almost twice proximal with; merus about 0.8–0.9 times as long as carpus, almost twice as long as ischium. Third pereiopod (Fig. 3C–D) reaching to around distal margin of antennular peduncle; dactylus with 7–9 spiniform setae on flexor margin (not counting unguis), gradually increasing in size, sub-terminal one largest, terminal spiniform seta about 1.2 as long as sub-terminal one; propodus about 5.0–5.2 times as long as dactylus, flexor margin armed with series of marginal spinules over its entire length; carpus about 0.60–0.62 times as along as propodus, flexor margin with well-developed subdistal spiniform seta and 3 or 4 spinules along flexor margin; merus elongated, about 1.23–1.25 times as long as propodus, armed with 3 spiniform setae on flexor margin; ischium short, armed with single spiniform seta on flexor margin. Fourth pereiopod similar to third, reaching to around distal margin of first article of antennular peduncle, occasionally noticeably shorter; dactylus with 8 or 9 spiniform setae (including unguis) on flexor margin. Fifth pereiopod, shorter (Fig. 3E–F), proportions similar to other pereiopods; dactylus with 20–29 spiniform seta (including unguis) on flexor margin, merus with 2 spiniform setae along flexor margin, ischium unarmed. Third maxilliped and first four pereiopods with mastigobranchs; setobranch present on all pereiopods (reduced on fifth). Male first pleopod (Fig. 4D) with endopod (Fig. 4E) about 0.3 times as long as exopod, conical in overall shape. Second pleopod (Fig. 4F) with endopod shorter than exopod; appendix masculina (Fig. 4G) reaching to about 0.6 of endopod length, with series of long, spinulose setae; appendix interna shorter, about 0.80 of appendix masculina length, distally furnished with numerous cincinnuli. Single ovigerous female examined with 9 eggs, non-eyed, approximately 0.6 × 1.1 mm in diameter. Uropodal protopod (Fig. 4A) with lateral lobe ending in prominent tooth over base of exopod; exopod with straight lateral margin, ending in distinct tooth (Fig. 4B); diaeresis with 10–12 regularly spaced, short spiniform setae. Distribution. Only known from Fraser Island (Alligator Creek, Bowarrady Creek, Coongul Creek, Lake Birrabeen, Rocky Creek, Wanggoolba Creek, Woralie Creek,), Queensland (see Page & Hughes 2007a: fig. 1). Genbank COI haplotype accession numbers: DQ656429 (Bowarrady Creek, AM P107219), DQ656432 (Bowarrady Creek, AM P107220), AY795019 (Rocky Creek, AM P107221), AY795020 (Rocky Creek, AM P107222). Remarks. Chenoweth & Hughes (2003) reported on the genetic divergence of three unnamed species of Caridina from coastal drainages in Queensland, which they referred to as the C. indistincta complex. Both Caridina indistincta sp. A and C. indistincta sp. B were only found on the mainland, and C. indistincta sp. C exclusively encountered on Moreton Island. The COI divergence between the three “species” was high, with C. indistincta sp. C having a nucleotide divergence of 15.85 (± 2.04) and 13.23 (± 1.79) to C. indistincta sp. A and C. indistincta sp. B, respectively; and 9.03 (± 1.39) between C. indistincta sp. A and C. indistincta sp. B. Page & Hughes (2007b) further explored the geographic genetic structure within C. indistincta sp. C, and recovered a deep split between three lineages: C. indistincta sp. C1, found on Moreton Island and along the western side of North Stradbroke Island; C. indistincta sp. C2, found along the eastern side of North Stradbroke Island and a few mainland locations and C. indistincta sp. C3, restricted to Fraser Island (type locality of C. indistincta sobrina). The COI divergence of the Fraser Island population (sp. C3) was found to be high, 9.6% (± 0.1) and 5.3% (±0.1) to sp. C1 and sp. C2, respectively. The sequenced material of Caridina indistincta sp. C 3 in Page & Hughes (2007b) and herein restudied (see non-type material) corresponds perfectly with the type series of C. indistincta sobrina Riek, 1953, which is herein redescribed and elevated to species level, based on the genetic divergences reported in Chenoweth & Hughes (2003) to the mainland “species” A and B. Although it remains somewhat unclear, one of these two lineages (A, B) should logically correspond to C. indistincta Calman, 1962; with a cursory examination of the type material by S. Choy indicating it indeed to be taxon C. indistincta sp. B (T. Page, pers. comm.). Whilst it may seem premature to redescribe a subspecies, elevate it to species status and partially resolve a species complex, without first considering the nominal subspecies at its core, the imprecise type locality (“St George’s District) and the complex nature of Queensland Caridina interrelationships is currently impeding progress on this. As it seems possible to resolve (at least in part) the taxonomy of the Caridina populations occurring on the sand islands, it appears best to proceed piecemeal. Riek (1953) designated a holotype for C. indistincta sobrina in the paper describing the taxon, although it was already noted by Springthorpe & Lowry (1994) that this specimen was not separated by Riek from the rest of the type series and the description is too vague to allow recognition of the specimen. From the 13 other, named species of Caridina previously reported to occur in Queensland (Davie 2002; Choy et al. 2019; de Mazancourt et al. 2018, 2020), the dentate rostrum of C. sobrina easily allows the species to be distinguished from C. confusa, C. malanda, C. spinula, C. typus and C. zebra, all of which have short, dorsally unarmed rostra. The presence of exopods on the first pereiopods in C. wilkinsi immediately separates that species from C. sobrina, in which these are lacking; whilst C. gracilipes can be easily distinguished by its very long, upturned rostrum (cf. Riek 1953; de Mazancourt et al. 2018). Caridina thermophila, only known from hot water springs in the Great Artesian Basin (Choy 2020) is superficially somewhat similar to C. sobrina, but differs in the higher number of dorsal teeth on the rostrum, 20–24 (vs. 11–21 in C. sobrina) albeit overlapping, as well as its smaller size and smaller egg size (0.85 × 0.53 vs. 1.05 × 0.6), the higher number of spiniform seta on flexor margin of the dactylus of the fifth pereiopod (45–49 vs. 19–28) and the higher number of diaeresis spines (15–17 vs. 10–12). There are probably further differences evident between both species, but only the short, insufficiently illustrated account of Riek (1953) and a few comments by Choy (2020) are available for comparison. The recently described C. pisuku, can be easily distinguished from C. sobrina by the distal, unarmed portion of the rostrum and the shape of the male first pleopod; as well as egg size (0.36–0.47 × 0.28–0.29 in C. pisuku vs. 0.6 × 1.1 in C. sobrina) (see de Mazancourt et al. 2020). Caridina sobrina can be differentiated from C. serratirostris by the angular pterygostomial margin, the lower number of postorbital, dorsal rostral teeth (0–2 in C. sobrina vs. 7–10 in C. serratirostris), the higher number of dorsal rostral teeth overall (22–26 in C. serratirostris vs. 11–21 in C. sobrina), the lower number of spiniform setae on the dactylus of the fifth pereiopod (19–28 vs. 10–15, respectively), the number of diaeresis spines (10–12 vs. 15–17 respectively), as well as egg size (0.30–0.38 × 0.18–0.23 in C. serratirostris vs. 0.6 × 1.1 in C. sobrina) (based on de Mazancourt et al. 2020). No detailed morphological descriptions are available for specimens from Queensland, identified as C. celebensis or C. nilotica, nor is it clear whether the local populations actually belong to those species, other species from species complexes like C. nilotica or indeed whether novel taxa are involved. A morphological comparison with those forms is thus left out of this study, although it is emphasised that in the genetic analysis of Page et al. (2007a), these taxa are unrelated. Perhaps not surprisingly, morphologically (as well as genetically) C. sobrina is closest to C. indistincta and C. pagei sp. nov. Based on the description of C. indistincta by Calman (1926), both species can be separated by the number of spiniform setae on the meri of ambulatory pereiopods 3–5 (2, 2, 1 in C. indistincta vs. 3, 3, 2 in C. sobrina), the number of dorsal teeth on the rostrum (20–32 in C. indistincta vs. 11–21 in C. sobrina), the general shape of the rostrum, which is more gracile, and the stylocerite only reaching the middle of the first antennal segment in C. indistincta rather than exceeding the middle in C. sobrina. The type description of Calman (1926) was, however, only based on eight specimens and more variation may be evident when larger sample sizes are available. Indeed, Riek (1953) mentions the number of spiniform setae on the meri of the ambulatory pereiopods to be 3, 3, 2 for C. indistincta at species level and does not mention a difference in this number for his new subspecies, C. indistincta sobrina, potentially negating this, the outlined difference between the species. Riek (1953) further mentions the number of spinules on the dactylus of the fifth pereiopod to be 20–32 for C. sobrina in contrast to 37–43 mentioned for the species (perhaps alluding to the nominal subspecies?). Additionally, Page et al. (2005) in their key mention differences in rostrum shape, as being lance-like for sp. A–B (the latter being the presumed true C. indistincta, see above) versus dagger-like for sp. C (i.e., C. sobrina and C. pagei sp. nov.), as well as differences in egg size and number (less than 1.5 mm in diameter and more than 50 in number in sp. A–B vs. larger than 1.5 mm and less than 50 in sp. C). Although a fuller comparison must await a detailed redescription of C. indistincta, it is evident that both species are morphologically very similar and it is acknowledged that identification may be difficult without genetic data, but such is the nature of Caridina taxonomy. For morphological differences from C. pagei sp. nov., see diagnosis of that species below.

Published as part of Christodoulou, Magdalini & Grave, Sammy De, 2023, Redescription of Caridina sobrina Riek, 1953 stat. nov. and description of a new species of Caridina from south-eastern Queensland (Decapoda, Atyidae), pp. 582-594 in Zootaxa 5353 (6) on pages 583-589, DOI: 10.11646/zootaxa.5353.6.5, http://zenodo.org/record/10027165