African Caridina (Crustacea: Decapoda: Caridea: Atyidae): redescriptions of C. africana Kingsley, 1882, C. togoensis Hilgendorf, 1893, C. natalensis Bouvier, 1925 and C. roubaudi Bouvier, 1925 with descriptions of 14 new species Author Jasmine Richard Author Paul F. Clark text Zootaxa 2009 2009-02-04 1995 1 75 journal article 29144 10.5281/zenodo.1455866 3f0364f8-ad91-4eff-932a-523ab28ee21d 1455866 Caridina africana Kingsley, 1882 ( Fig. 1 ) Caridina africana Kingsley 1882 : 127 , pl. 1, figs. 3, 3a; Sharp 1893 : 111; Ortmann 1894 : 402, 404; Bouvier 1905 : 74; Hart et al. 2001 : 111, fig. 6.4 A . Non Caridina africana .— De Man (in Weber et al . 1897 ): 170–174, pl. 15, fig. 2 a–f; Lenz 1912 : 5 = Caridina natalensis ( Bouvier, 1925 ) comb. nov. Non Caridina africana .— Gordon 1933 : 356–358, fig. 4; Barnard 1950 : 661, fig. 123, m, n; Holthuis 1951 : 17–18; Monod 1968 : 1350–1353 ; Magnetti, 1980 : 41–44, fig.1, 2; Hussein and Obuid-Allah 1990 : 71–85, figs. 1–9 = Caridina togoensis Hilgendorf, 1893 . Material examined. Syntype : ♀ South Africa sta. 412, Natal, Zululand, coll. S. A . Grant, USNM 84331, call acc. no. 2876. Description. Total length of syntype : 25 mm . Carapace length of syntype : 6.2 mm . Rostrum ( Fig. 1a, b ): equal to antennal scale. 4.5 mm in length. 0.74 × long as carapace. 13 teeth on dorsal margin leaving 0.4 of length distally unarmed. 3 post-orbital teeth. 14 teeth present on ventral margin leaving short distal end unarmed. Tip bifid. Formula (3) 13/14. Antennular peduncle ( Fig. 1b ): appendage damaged with third segment missing. Stylocerite 0.65 × length of basal segment. Anterolateral teeth of basal segment 0.25 × second segment. First pereiopod ( Fig. 1c ): dactylus equal to length of palm of propodus. Chela 1.8 × long as broad. Carpus 1.9 × long as broad. Anterior excavation deep. Third pereiopod ( Fig. 1d, e ): dactylus 3 × long as broad. 9 spines on dactylus, spine at anterior end absent due to damamge. Propodus 5 × long as dactylus and 13.5 × long as broad with 10 spines arranged along inner margin. Carpus 0.6 × long as propodus, with 4 spines on inner margin. Merus 1. 7 × longer than carpus length and with 3 large spines on inner margin. First female pleopod ( Fig. 1f ): endopod 0.67 × long as exopod. Sixth abdominal somite : 0.66 × long as carapace. Telson ( Fig. 1g , h ): narrow and tapering distally. 0.82 × long as sixth abdominal somite. 6 dorsal spines on one side and 5 on other side (including subterminal spine). Posterior margin converging to middle, ending in triangular median point, making posterior margin triangular. Other than one short spine, spines on posterior margin of telson missing, remnants suggest 1 pair of lateral spines and 2 pairs of short inner spines. FIGURE 1. Caridina africana Kingsley, 1882 , USNM 84331, Type ♀: a) Lateral view of rostrum; b) Anterior region of cephalothorax; c) First pereiopod; d) Third pereiopod; e) Dactylus of third pereiopod; f) First pleopod; g) Telson; h) Posterior margin of telson, i) Uropod diaeresis spinules, j) Preanal carina. Uropod ( Fig. 1i ): 14 uropod diaeresis spinules. Preanal carina ( Fig. 1j ): unarmed. Remarks. Kingsley (1882) described C. africana , from Zululand, South Africa . He figured the lateral view of the cephalothorax and the first pereiopod together with a brief description. Later, Sharp (1893) compiled a catalogue of the crustaceans in the ANSP, and listed C. africana Kingsley, 1882 ( No . 163 type ) under family Atyidae . He also indicated that more than ten specimens were present. Ortmann (1894) , while revising the Family Atyidae , provided a key to the species of Caridina , and he examined “many” ‘type’ specimens of C. africana from ANSP. He limited the distribution of C. africana to Zululand, South Africa and provided a useful description of the species, noting “Having examined the type of this species in the Museum of the Academy of Nat. Sci. Philadelphia, I can give the following details…….” A review of literature referring to C. africana for the present study revealed the necessity of examining the type material of species. Such an examination would confirm the identity of C. africana , corroborate the validity of a host of synonymies associated with this species, and verify the status of related taxa described as subspecies, varieties or forms. The USNM, made a single type specimen of C. africana available for examination. This was a female registered USNM 84331: Caridina africana Kingsley , Atyidae : Decapoda : Natal, Zululand, Sta; Grant; S. A ., There is no doubt that this specimen is part of the type series, and although the female is damaged, it is described and illustrated ( Fig. 1 ) here. Sharp (1893) and Ortmann (1894) confirmed the presence of a series of specimens at ANSP so the Academy was approached for more specimens. The Academy confirmed that there was an entry in the hand written Carcinological catalog, Number 163, with a red dot confirming type status with the following data, Name: Caridina africana Kingsley 1882 , Locality: Zulu Land, S. Africa, Specimen Count: + [= many], Collector: [blank], Donor: S. A . Grout, Notes: (Kingsley types ), [the "u" characters in Zulu and Grout are written with a breve (unicode 016D). However, this type material was not included in the list of Academy specimens listed by Spamer & Bogan (1992) . Unfortunenatly, a curator at the Academy confirmed, “It is possible that there are specimens of C. africana in our collections, but they are not where they should be. I know there is a record of a single wet lot under this name, but as I have explained previously two of us have spent a considerable time searching for it without success.” Consequently, the information derived from the single female, USNM 84331, together with the descriptions of Kingsley (1882) and Ortmann (1894) have been assimilated to provide description of Kingsley’s C. africana for the present study. De Man (in Weber et al. 1897 ), considered the details provided by Kingsley (1882) and Ortmann (1894) as ‘short’, and redescribed C. africana in detail from a male specimen captured in the River Umholti near Verulum, Natal. This specimen is extant in ZMA, ( Caridina africana, Umholti River by Verulum, Natal, coll. M. Weber, 1895, det. De Man 1897 , ZMA De 202881, 1 male ) and was examined for this present study ( Figs. 7 , 8 ). De Man’s single male is not considered to be C. africana , but is here referred to C. natalensis ( Bouvier, 1925 ) comb. nov. , see the remarks for C. natalensis . The following details of Kingsley’s (1882) description for C. africana are of interest: “Carapax smooth, compressed; suborbital spine alone present; rostrum nearly as long as the carapax; its apex slightly reflexed and armed above and below with minute teeth. Peduncle of antennulae about as long as the rostrum,…..” “Ambulatory feet very slender, short, the carpus of first pair about as long as the palm, of the second as long as the hand …..., Dactyli of the three posterior pairs of feet ….about one fifth of the propodal joints” “Telson narrow, the sides straight and but slightly converging to the truncate tip…” From Ortmann’s (1894) description of C. africana , the following morphological feature are considered relevant: “Rostrum as long as the stalk of the antennulae. Teeth on the dorsal margin of the rostrum 10–15, not con- tinued up to the tip, 1–2 teeth at the tip; teeth on the ventral margin are 7–12.” “Carpus of first pereiopod, twice as long as broad, a little shorter than hand, fingers equal to palm; Carpus of second pereiopod, four times long as broad, little longer than hand, finger about 1.5 times palm; Dactylus of fourth pereiopod about 1/5 times propodus; No fifth pereiopod was available in the preserved specimens”. From the present examination of the damaged female type specimen, the rostrum was reconstructed ( Fig. 1a, b ), the first pereiopod together with the third and fourth were available but detached from the specimen and the telson was intact. Consequently the description in this study is based on data obtained by re-examination of the only extant type specimen, and from the works of Kingsley (1882) and Ortmann (1894) . No descriptive information was available for the fifth pereiopod, first and second pleopods of male, and the number and size of the eggs. The descriptions of C. africana presented by Gordon (1933) , Barnard (1950) , Holthuis (1951) , Monod (1968) , Magnetti (1980) and Hussein and Obuid-Allah (1990) are all referred here to C. togoensis . As observed in the present study the C. africana of De Man (in Weber et al . 1897 ) is not C. africana Kingsley, 1882 ; see below in the remarks for C. natalensis . While describing C. togoensis from Bismarkberg, Togoland , Hilgendorf (1893a) stated that the dactylus of fifth pereiopod was longer in his new species than that of C. africana . However, he did not describe the spinulation of the posterior margin of the telson. Later, Hilgendorf, (1898) , described a new variety namely C. togoensis var. stuhlmanni from Bach bei, Undussuma. He made reference to a new description of C. africana by De Man (in Weber et al . 1897 ) of a specimen from Zululand, Natal. Hilgendorf (1898) then suggested that his C. togoensis and C. togoensis stuhlmanni were closer to the C. africana of De Man (in Weber et al . 1897 ) particularly in possessing pinnate bristles at the hind end of the telson. This conclusion of Hilgendorf (1898) probably resulted in the identification of the specimens with the characters of C . togoensis as C. africana . This present study reveals C. togoensis of Hilgendorf (1893a) differs from C. africana in having more teeth on the dorsal margin of rostrum that are arranged either up to the tip of the rostrum or with an unarmed distal dorsal margin, fewer teeth on ventral margin of rostrum, the tip of the rostrum always pointed, and in possessing a longer dactylus on the fifth pereiopod. Bouvier (1905 , 1913 ), in his key to the species of Caridina , provides distinguishing characters for C. togoensis . Bouvier (1905) dealt with C. africana and C. togoensis , on the basis of the proportion of the dactylus of the fifth pereiopod to the propodus, “sont plus courts que le 1/3” in C. africana Kingsley (1882) and “ont environ le 1/3” in C. togoensis Hilgendorf (1893a) . He also indicated that the rostral formula of C. africana was 10–15/4–12 and the “Rostre á peu près aussi long ou un peu plus long que pedoncule antennulaires, Carène dorsale inerme en avant ” and that of C. togoensis as 3+14/6 and “Rostre atteignant environ l’extremité des pedoncules antennulaires, Rostre garni de dents jusqu’au bout”. The key of Bouvier (1905) indicates the distinction between C. africana and C. togoensis . While describing C. africana forme typica , Bouvier (1925) referred to the number of teeth of C. africana as 10–15 on the dorsal margin leaving the distal margin unarmed and 7–12 teeth on the ventral margin (as rightly described by Kingsley, 1882 and Ortmann, 1894 ). However, Bouvier (1925) identified his one ovigerous female with a C. togoensis like arrangement of teeth on the rostrum (formula 3+19/9) as C. africana forme typica . Gordon (1933) who described C. africana from Lake Edward, probably Kampala and Arua , West Nile, recognised the differences in rostral formula between her C. africana and C. africana Kingsley, 1882 , and wrote, “The rostral formula is not quite the same as that given by Kingsley and Ortmann”. Caridina africana is to date reported only from the type locality, Zululand, South Africa . Specimens of C. africana described from other regions of Africa are not the species of Kingsley (1882) . They either belong to C. togoensis or are new species. The illustration of C. africana by Hart et a l . (2001) indicates that the arrangement of the teeth on the rostrum is similar to Kingsley’s (1882) . The reduced number of post-orbital teeth and upwardly turned rostrum suggested that these could be a different species. A sample of 12 males , six females and one juvenile collected by Rob Hart from Cabris Road Stream, Hay fields, Pietermatrizburg, Natal, were examined. Unfortunately the rostrum and telson of most specimens were damaged. However, the variation in the number of setobranchs suggested that the sample might contain more than one species. Though it was possible that the damaged sample may contain C. africana , the identification of the material could not be confirmed. The figures, morphometric features and taxonomic characters given by Bello-Olusoji et al . (2004) for C. africana from Erin-Ijesha water falls, Osun state, Nigeria provide no clues to the real identity of their specimens. Based on the locality of the collection site, the material of Bello-Olusoji et al . (2004) could belong to any of the species reported from Nigeria in the present study. It is however, certain that it is not the C. africana of Kingsley as described in this study. The apparent loss of most of the C. africana type series, which is reported to be ‘more than 10 specimens’, is tragic because a full description of this species including variation is now difficult. Extensive collections of C. africana Kingsley (1882) from the type locality, including males, ovigerous females and juveniles could provide more information on pereiopods, especially that of fifth pereiopod, first and second pleopods of male, egg size. However, material of Caridina in the collections of the SAM, Cape Town and the AM, Grahamstown, South Africa were examined. One lot from the SAM, SAM A 3133, containing two males, two ovigerous females, five females and one juvenile collected from Umbilo River pools, Pine Town, Natal, 10 miles from sea, appeared to be morphologically closer to the type specimen of C. africana examined here. These specimens are illustrated and described below. Description. Total length : 25–35 mm . Carapace length : 5–5.75 mm . Rostrum ( Fig. 2a, b, c ): straight, extending to peduncle or equal to antennal scale. 4.0– 4.9mm in length, 0.8–0.9 × long as carapace.11–20 teeth present on dorsal margin leaving 0.35–0.5 of length distally unarmed. 3–4 post-orbital teeth. 9–17 teeth on ventral margin arranged from proximal end to tip. Tip normally bifid, one specimen with acute tip. Formula (3–4) 11–20/9–17. Antennular peduncle ( Fig. 2d ): 0.6–0.7 × carapace. Stylocerite 0.6–0.8 × length of basal segment. Anterolateral teeth of basal segment 0.2–0.25 × second segment. 8–10 segments bearing aesthetascs. First pereiopod ( Fig. 3a ): dactylus 1.0–1.1 × palm of propodus. Chela 2.0–2.2 × long as broad. Carpus 1.7–1.9 × long as broad, with moderate excavation anteriorly. Second pereiopod ( Fig. 3b ): dactylus 1.1–1.3 × long as palm of propodus. Chela 2.5–2.7 × long as broad. Carpus 4.1–4.3 × long as broad. Third pereiopod ( Fig. 3c, d ): dactylus 2.7–3 × long as broad. Spines on dactylus varied from 9–11 (including terminal spines). Propodus 4.5–5 × long as dactylus and 11–13 × long as broad with 9–14 spines arranged along inner margin. Carpus 0.65–0.7 × long as propodus, with 1 large spine and 4–5 small spines on inner margin. Merus 1.5–1.7 × carpus length. Merus with 3 large spines along posterior margin. Ischium with one large spine. Fifth pereiopod ( Fig. 3e, f ): dactylus 3.75–4.2 × long as broad with 45–55 spines arranged in comb-like fashion on inner margin. Propodus 15–17 × long as broad and 4.1–5 × long as dactylus with 45–55 spines arranged along posterior margin. Carpus 0.5–0.6 × propodus length and with one large spine and 4–6 minute spines along inner margin. Merus 1.4–1.6 × carpus length, with 2 large spines along posterior margin. Setobranchs : 4 on all pereiopods. First male pleopod ( Fig. 3g , h ): endopod 0.45–0.55 × exopod length; with appendix interna. First female pleopod ( Fig. 3i ): endopod 0.5–0.65 × long as exopod. Eggs ( Fig. 3j ): 45–55, 1.0–1.1 × 0.65–0.75mm in size. Second male pleopod ( Fig. 3k, l ): appendix masculina 1.9–2.1 × appendix interna, 0.4 × endopod. Sixth abdominal som ite: 0.5–0.6 × long as carapace. Telson ( Fig. 3m , n ): 1.0–1.1 × long as sixth abdominal somite. 5 or 6 pairs of dorsal spines (including subterminal spines). Posterior margin triangular converging to median pointed process with 1 pair of lateral spines and two pairs of shorter inner spines. All spines sparsely plumose. FIGURE 2. Caridina africana Kingsley, 1882 , SAM A 3133. ♀ a) Entire; b) Anterior region of cephalothorax. ♀: c) Anterior region of cephalothorax; d) Antennular peduncle. FIGURE 3. Caridina africana Kingsley, 1882 , SAM A 3133, ♀: a) First pereiopod; b) Second pereiopod; c) Third pereiopod; d) Dactylus of third pereiopod; e) Fifth pereiopod; f) Dactylus of fifth pereiopod. ♂: g) First pleopod; h) Endopod of first pleopod. ♀ ovig.: i) First pleopod; j) Eggs. ♂: k) Second pleopod; l) Appendix masculina; m) Telson; n) Posterior margin of telson; o) Uropod diaeresis spinules; p) Preanal carina. FIGURE 4. Caridina nilotica (P. Roux, 1833), NHM 1922.11.22.36–55, ♀: a) Anterior region of cephalothorax; b) Telson; c) Posterior margin of telson. Caridina longirostris H. Milne Edwards, 1837 , MNHN Na 746, Type ♀: d) Anterior region of cephalothorax; e) Telson; f) Posterior margin of telson. Co-types NHM 1955.5.3.57–59, ♂; g) Endopod of first pleopod. MNHN Na 746 Type ♀: h) Preanal carina, NHM 1955.5.3.57–59, ♀: i) Anterior region of cephalothorax. Uropod ( Fig. 3o ): 11 or 12 diaeresis spinules. Preanal carina ( Fig. 3p ): unarmed. The SAM A 3133 material collected from Pine Town, Natal, seems to agree well with the observations made on the female type specimen and with the descriptions of Kingsley (1882) and Ortmann (1894) . In comparison with the type material, however, the following points should be considered. With respect to the SAM material, the adult is larger ( 25–35mm ), the rostrum is straight, extending up to the antennular peduncle or equal to the scale. Three or four post-orbital teeth are present. Teeth on the dorsal margin arranged proximally leaving 0.35–0.50 unarmed. The tip of the rostrum is bifid (one specimen had an acute tip). The teeth on the ventral margin are arranged from the proximal end up to the tip with a rostral formula of (3–4) 11–20/9–17. This is similar (though the number of teeth is more) to the rostral formula given by Ortmann (1894) as 10–15 (not continued to the tip, one or two at the tip) 7–12 and as observed in the type specimen (3) 13/14. The measurements of pereiopods are close to the type specimen, particularly third pereiopod with propodus 4.5–5.0 × the dactylus and fifth pereiopod with propodus 3.75–4.20 × dactylus. The telson has a triangular posterior margin with a median point. The uropod diaeresis has 11 or 12 spinules ( 14 in the type specimen) and preanal carina without spinulation. The spines on the posterior margin of the telson are damaged on the type specimen except for one spine, but there are traces of 1 pair of lateral spines and two pairs of inner spines. Further, there is no evidence from the type as to the length of the lateral spines. In the SAM A 3133 specimens, the telsons possess a triangular posterior margin with a median process, and 3 or 4 pairs of spines that are either equal in length or with the inner spines shorter than the laterals. Considering the similarity in the arrangement of the teeth on the rostrum of C. africana , C. nilotica ( P. Roux, 1833 ) and C. longirostris H. Milne Edwards, 1837 , specimens of C. nilotica were examined from Hermilik canal, Egypt (NHM 1922.11.22.36–55) and the type specimens of C. longirostris from Algerie , Riv. De la Macta, Pred’ de Oran (MNHN Na 746: 2♂, 10♀ , cotypes NHM 1907.1.7.26–27, 2♀ ; NHM 1955.5.3.57–59 1♂, 2♀ ). Caridina africana differs from C. nilotica in having a straight rostrum that is equal to antennular peduncle or reaching as far as the end of the antennal scale, 3 or 4 post-orbital teeth, teeth on dorsal margin within lesser range being 11–20 having a triangular posterior margin of the telson with a median process, always having 10 or more uropod spinules (10–14) and in having 45– 55 eggs of 1.0–1.1 × 0.65–0.75mm in size. Caridina nilotica ( Fig. 4a ) has an upturned rostrum that is longer than the antennal scale (rarely equal to scale), rostral formula (1–3) 10–28/10–28 mostly (1–2) 14–24/13–19, 1–3 pairs (usually 1 pair) of short and stout spines on the posterior margin of the telson ( Fig. 4b, c ) which is almost flat and not protruding, uropod diaeresis spinules are 7–12 and 85– 140 eggs of 0.5–0.85 × 0.35–0.55mm in size. Caridina longirostris could be differentiated by smaller size of adult, 12–20mm . The rostrum ( Fig. 4d ) equal in length to antennal scale or longer with a rostral formula (1–2) 14–20/13–16 and turned upwards. The stylocerite is 0.8–0.9 × first segment with telson ( Fig. 4f ) tapering towards posterior margin that ends sharply triangular with 1 pair of longer lateral spines and 2 pairs of shorter spines. The endopod of first male pleopod ( Fig. 4g ) does not possess an appendix interna, but the anterior margin of the endopod with a cleft. Preanal carina with a spine ( Fig. 4h ). Caridina africana can be differentiated from C. longirostris by its larger adult size (total length 25–35mm ), 3 or 4 post-orbital teeth, lesser range of dorsal and ventral teeth 10–15/7–12 as observed by Ortmann (1894) , (3) 13/14 as in the type specimen, the female examined in the present study and (3–4) 11–20/ 9–17 as in SAM A 3133, stylocerite 0.6–0.8 times as long as the first segment of the antennular peduncle, broader posterior margin of telson with shorter lateral spines and a plain preanal carina. While selection of a lectotype for Caridina africana Kingsley, 1882 might seem desireable, it presents a number of problems for the authors. The specimen borrowed from the USNM appears to be the only extant sample of the type series, but it is damaged. Furthermore although the complete type series has been reported as missing by the ANSP curators, this lot may simply be misplaced within their collections, or perhaps in another institution. Therefore, as this is the first study to report the type series as missing, selecting a lectotype appears to be premature at this time.