Labeobarbus nzadinkisi, Vreven & Musschoot & Decru & Lunkayilakio & Obiero & Cerwenka & Schliewen, 2019

Vreven, Emmanuel J. W. M. N., Musschoot, Tobias, Decru, Eva, Lunkayilakio, Soleil Wamuini, Obiero, Kevin, Cerwenka, Alexander F. & Schliewen, Ulrich K., 2019, The complex origins of mouth polymorphism in the Labeobarbus (Cypriniformes: Cyprinidae) of the Inkisi River basin (Lower Congo, DRC, Africa): insights from an integrative approach, Zoological Journal of the Linnean Society 186, pp. 414-482 : 451-457

LABEOBARBUS NZADINKISI SP. NOV.

( FIGS 8F, 24; TABLES 1, 9, 10)

Varicorhinus latirostris non Boulenger, 1910 (Wamuini Lunkayilakio, 2010: 137)

Varicorhinus cf. latirostris non Boulenger, 1910 ( Wamuini Lunkayilakio et al., 2010: 85; table 1)

Holotype: MRAC A7-009 -P-1180, Inkisi River basin, Bongolo River , Kinsende village, Lower Congo ( DRC) (5°23’10.7’ S-15°15’25.2 ’E) Coll.S.Wamuini Lunkayilakio, 30 August 2006 (205.2 mm SL) (DNA tag n°721).

Paratypes: MRAC A7-009 -P-1361, Inkisi River , Kinsende village, Lower Congo ( DRC) (5°23’10.0’ S-15°15’17.1 ’E) Coll. S. Wamuini Lunkayilakio, 31 August 2006 (182.7 mm SL) (DNA tag n° 729) – MRAC A7-009 -P-1377, Inkisi River basin, N’Soni River , Yanama village, Lower Congo ( DRC) (5°24’08.2’ S-15°10’16.6 ’E) Coll. S. Wamuini Lunkayilakio, 17 July 2006 (101.2 mm SL) (DNA tag n°861) – MRAC A7-009 -P-1386, Inkisi River basin, Luidi River , Kinua-Nsudi village, Lower Congo ( DRC) (5°28’22.8’ S-15°13’11.1 ’E) Coll. S. Wamuini Lunkayilakio, 20 July 2006 (95.7 mm SL) (DNA tag n°877) – MRAC A9-014 -P-0381, Inkisi River basin, Vini River , Ngomina village , Mafukusa, Lower Congo ( DRC) (4°47’42.1’ S-14°56’06.0 ’E) Coll. S. Wamuini Lunkayilakio, 11/10/2008 (203.6 mm SL) (DNA tag n°1047) .

Differential diagnosis: Within the Congo basin L. nzadinkisi can be distinguished from L. altipinnis , L. ansorgii , L. batesii , L. brauni , L. cardozoi , L. caudovittatus , L. dartevellei , L. fasolt , L. habereri , L. humphri , L. iphthimostoma , L. iturii , L. jubbi , L. longidorsalis , L. longifilis , L. lufupensis , L. macroceps , L. macrolepidotus / cf. macrolepidotus , L. macrolepis , L. mawambi , L. mawambiensis , L. mirabilis , L. nanningsi , L. oxyrhynchus , L. paucisquamatus , L. stappersii , L. trachypterus , L. upembensis and L. wittei by its high number of lateral line scales, i.e. 35–41 (vs. less than 34); from L. leleupanus by its low number of lateral line scales, i.e. 35–41 (vs. 45–47); from L. tropidolepis and L. platyrhinus by its low number of scales between the lateral line and the dorsal and ventral midline, i.e. 4.5–5.5 and 5.5 (vs. 7.5–8.5 and 7.5– 9.5 in L. tropidolepis and 6.5–7.5 and 6.5–8.5 in L. platyrhinus ) and from the latter by its low number of circumpeduncular scales as well, i.e. 12–14 (vs. 16–18); from L. robertsi by the absence of papillae on the anterior edge of the lower jaw (vs. with numerous well identifiable papillae); from L. progenys by its non-prognathous lower jaw (vs. prognathous); from L. altianalis , L. gestetneri and L. somereni by its lack of both pairs of barbels (vs. two pair of well-developed barbels); and from L. pellegrini by its short prepelvic length, i.e. 46.5–48.5% SL (vs. 50.6% SL in L. pellegrini ); its short pelvic length, i.e. 17.9–21.0% SL (vs. 21.8% SL in L. pellegrini ); and its large eye, i.e. 29.1–34.6% HL (vs. 27.1% HL, in L. pellegrini ). Further, L. nzadinkisi can be distinguished from the other members of the Inkisi complex, L. nzadimalawu and the intermediate/hybrid specimens by the presence of a cornified Varicorhinus real cutting edge on the outer edge of the lower jaw in combination with (1) the absence of barbels and (2) poorly developed fleshy lips on the lateral side of the lower jaw [vs. never with a cutting edge but instead always with a free mental lobe in combination with (1) two pairs of well-developed barbels and (2) well-developed fleshy lips in L. nzadimalawu . Although a cornified Varicorhinus real cutting edge can be found in some specimens, this most often in combination with (1) at least a single pair of well-developed barbels and (2) well-developed fleshy lips in the hybrid specimens (for details see Table 1)]. In addition, L. nzadinkisi can be distinguished from L. nzadimalawu by its broad mouth width, 26.8–50.5% HL (vs. 16.1–26.5% HL); short head length, 20.1–22.1% SL (vs. 23.0– 26.4% SL); long dorsal-fin base length, 14.4–17.9% SL (vs. 12.1–16.0% SL); and short prepectoral distance, 20.0–22.1% SL (vs. 22.6–26.0% SL) ( Figs 13A, B, 14A, B). Finally, L. nzadinkisi can be distinguished from Acapoeta tanganicae by its low number of lateral line scales, i.e. 35–41 (vs. 57–67).

Within the adjacent Lower Guinea ichthyofaunal province, L. nzadinkisi can be distinguished from L. axelrodi , L. batesii , L. brevispinis , L. cardozoi , L. caudovittatus , L. compiniei , L. habereri , L. fimbriatus , L. jaegeri , L. malacanthus , L. mariae , L. mbami , L. micronema , L. mungoensis , L. roylii , L. sandersi , L. semireticulatus , L. steindachneri , L. tornieri , L. versluysii and L. werneri by its higher number of lateral line scales, 35–41 vs. less than 34; from L. aspius , L. lucius and L. progenys by its non-prognathous lower jaw (vs. lower jaw clearly prognathous); and from L. rocadasi by its last unbranched dorsal-fin ray not being transformed into a well-developed spine, but instead being clearly segmented over approximately half its length, or 43.0–50.1% dorsal-fin height (vs. last unbranched dorsal-fin ray transformed into a spine, clearly segmented only at its most distal end). Finally, L. nzadinkisi can be distinguished from Sanagia velifera by its high number of lateral line scales, 35–41 (vs. 22–24).

Within the adjacent Quanza ichthyofaunal province, L. nzadinkisi can be distinguished from L. ansorgii , L. gulielmi , L. jubbi , L. nanningsi , L. rhinophorus , L. rosae and L. roylii by its high number of lateral line scales, 35–41 (vs. less than 34); from L. clarkeae , L. ensifer and L. varicostoma by the absence of papillae on the anterior edge of the lower jaw (vs. with well identifiable papillae); from L. lucius and L. progenys by its non-prognathous lower jaw (vs. lower jaw clearly prognathous); and from L. boulengeri [previously L. latirostris ( Boulenger, 1910) ], L. ensis , L. girardi , L. steindachneri , L. stenostoma and L. rocadasi by its last unbranched dorsal-fin ray not being transformed into a well-developed spine, but instead being clearly segmented over approximately half its length, or 43.0– 50.1% dorsal-fin height (vs. last unbranched dorsal-fin ray transformed into a spine, clearly segmented only at its most distal end).

Description: The holotype is illustrated in Figure 8F and Figure 24 for its mouth phenotype. Meristics and measurements are given in Tables 9 and 10, respectively. Labeobarbus nzadinkisi has a rather shallow body depth, low dorsal-fin height, and shallow and elongated caudal peduncle. A relatively small-sized Labeobarbus species with a maximum observed size of ± 205 mm SL.

Labeobarbus nzadinkisi has a typical Var.-mouth phenotype ( Table 1) characterized by the presence of a cornified cutting edge on the anterior edge of the lower jaw ( Fig. 8F). Anterior and posterior pair of barbels generally absent ( Fig. 8F), but a minute pair of posterior barbels might be present. Lower jaw always clearly shorter than upper jaw.

Tubercles ( Fig. 20) are present on all specimens (≥99.8 SL) except for the smallest examined one (95.7 mm SL) and, as such, are likely to be present in males and females (not dissected to identify sex). Specimens with tubercles were collected only in the major dry season and the beginning of the wet season (June and October) (see Wamuini Lunkayilakio, 2010). In these specimens, tubercles are present at least on the snout and often on the dorsum of the head too, although in a different degree. Large specimens (about ≥ 180.0 mm SL) with conical tubercles over mid-ventral and lateral sides of the snout and the lateral sides of head below the eye reaching posteriorly up to area of the preopercular/ opercular skin fold. A few larger tubercles are also present between nostril and eye. Further, numerous and tiny tubercles are scattered over dorsal regions of snout, head and nape, between the posterior edge of the head and the anterior edge of the predorsal area. Finally, remainder of body without tubercles. Tubercles are easiest observable in one of the largest specimens examined [ MRAC A9-014-P-0381: 203.6 mm SL (11/10/2006)].

Lateral line scales of the examined specimen ( Fig. 21) with less sinuous and parallel striae towards their horizontal midlines but slightly radiating striae towards their upper and lower edge. Number of striae on the posterior edge of these scales between about 46–48.

The pharyngeal teeth number of the only dissected specimen is 2.3.4. (left) – 5.3.2. (right), with the first tooth of the inner row being absent on the left pharyngeal bone ( Fig. 22). Note that on both the left and right pharyngeal bone there are two additional smaller teeth present median of the inner row; these are not firmly attached to the bone but rather are loosely embedded in the surrounding soft tissue; one of them well-developed, the other one is more like a dome-like cusp.

Coloration: Live specimens silvery grey on lateral sides of head and body. Dorsal midline of head blackish. Dorsal midline of body, in front, along and behind the dorsal fin, with a yellowish-green silver hue. Proximal part of scales, i.e. scale pockets, darker in colour, especially for the scales above the lateral line. Fins of comparable blackish overall colour with their distal margins whitish translucent. Fin rays whitish. Ventral side of head and body yellowish-white.

In alcohol, small-size specimens (±≤ 130 mm SL) uniformly brownish, becoming darker towards the dorsal midline of head and body. Scale pockets, especially those above the lateral line, typically darker brown due to more numerous melanin dots. Anteriormost ventral part of snout and jaws, ventral side of head and belly yellowish light brown. Fins whitish translucent and even transparent with sparsely set minute dark brown or blackish melanin dots. Large-size specimens (±≥ 183 mm SL) uniformly dark brown, becoming even darker above the lateral line towards the dorsal midline of the head and body. No distinct scale pocket coloration at first sight, although the proximal part of the scales is dark brown while the overall distal border is translucent or even transparent. Anteriormost ventral part of snout and jaws, ventral side of head and belly yellowish light brown. Head generally somewhat lighter of colour compared to body. Fin tissues faintly brownish or even more blackish. Fin rays often more white-yellowish. Distal edge of fins whitish translucent and sometimes even transparent.

Distribution: Species endemic to the Inkisi River, Lower Congo River basin above the Zongo Falls (Wamuini, 2010) ( Fig. 23). Currently only known from the DRC part of the basin.

Etymology: The current name of the Inkisi River is derived from its local appellation ‘ Nzadi i nkisi ’ (i.e. the river of the ‘ nkisi ’) in Kikongo (Kintandu/Kindibu dialects), referring to the missionaries who threw the ‘ mi -nkisi ’ (‘fetish-es’) [‘fetish’ object containing a certain nkisi spirit – the oldest ancestor spirit, which has an entirely spiritual existence until it has been caught and incorporated in a man-made object, a ‘fetish’, which is referred to as a nkisi ( Jackobson-Widding, 1979: 131–132)] in the river in their effort to convert the local populations to Christianity. Species name referring to this new name of the river basin to which it appears endemic. In addition, by its reference to the nkisi -objects, indirectly referring to the enigmatic hybridization complex of which this species is a parental species. Species name to be treated as a noun in apposition ( ICZN, 1999: Articles 31.2.1. and 34.2.1), making its gender ending unchangeable.

Ecology: A small-scale ecological study of ten fishing stations on the Inkisi basin (Wamuini Lunkayilakio, 2010: tables 4.2, 4.3 4.5, 4.8; Wamuini Lunkayilakio et al., 2010: figs 1, 3) revealed that L. nzadinkisi , the Var.-mouth phenotype species, was most abundant in the stations characterized by grass banks, i.e. at Luidi, Bongo, Wungu and Nua, all affluents of the Inkisi. Instead, L. nzadimalawu , the Lab.-mouth phenotype species, was far less abundant at those stations. For more details on L. nzadimalawu , see above. Although preliminary, these data seem to reveal important differences in habitat preferences between both species. Unfortunately, no data are available for the intermediate-mouth phenotype specimens, i.e. hybrids, which were not included in these analyses.

Other specimens examined: MRAC A7-009 -P-1195, Inkisi River basin, Luidi River , Kinua-Nsudi village, Lower Congo ( DRC) (5°28’22.8’ S-15°13’11.1 ’E) Coll. S. Wamuini Lunkayilakio, 21 July 2006 (99.8 mm SL) (DNA tag n°878) – MRAC A9-014 -P-0382, Inkisi River basin, Vini River , Ngomina village , Mafukusa, Lower Congo ( DRC) (4°47’42.1’ S-14°56’06.0 ’E) Coll. S. Wamuini Lunkayilakio, 11/10/2008 (122.2 mm SL) (DNA tag n°1054) – MRAC A9-014 -P-0380, Inkisi River basin, Vini River , Kizinga Village , Kindawula, Lower Congo ( DRC) (4°46’58.6’ S-14°56’47.8 ’E) Coll. S. Wamuini Lunkayilakio, 11 October 2008 (129.9 mm SL) (DNA tag n°1040) .

INTERSPECIFIC HYBRIDS: