Parauchenoglanis chiumbeensis, Sithole & Vreven & Bragança Tobias Musschoot & Chakona, 2024

Sithole, Yonela, Vreven, Emmanuel J. W. M. N., Bragança Tobias Musschoot, Pedro H. N. & Chakona, Albert, 2024, Nine in one: integrative taxonomic evidence of hidden species diversity in the widespread Zambezi grunter, Parauchenoglanis ngamensis (Siluriformes: Auchenoglanididae), from southern and south-central Africa, Zoological Journal of the Linnean Society 202, pp. 1-33 : 27-31

Parauchenoglanis chiumbeensis sp. nov.

( Fig. 15; Table 3)

Parauchenoglanis ngamensis ‘chiumbe’, Sithole et al., 2023.

Common English name: Chiumbe grunter.

Common French name: Mâchoiron de la Chiumbe.

Holotope: RMCA _ Vert _P. 161724, 131.5 mm SL, Angola, Kasai sub-basin, River Mololo, affluent river Chiumbe (7°49 ʹ S, 21°5 ʹ E); collector Max Poll; 10 January 1963 .

Paratopes (N = 10): Angola: RMCA _Vert_P.161721–161723, three, 58.2–117.5 mm SL, collection details same as for the holotype . RMCA _ Vert _P.161728−161729, two, 53.5–95.7 mm SL, Kasai sub-basin, River Kaino (7°58 ʹ S, 21°7 ʹ E); collector Max Poll; 11 January 1963. RMCA _Vert_P.161730−161731, two, 93.8–134.0 mm SL , Angola, Kasai sub-basin, River Tchimenji (7°58 ʹ S, 21°7 ʹ E); collector Max Poll; 18 January 1963. SAIAB 246237 (ex. RMCA _Vert_P.161725–161727), three, 93.4– 142.9 mm SL, collection details same as for the holotype .

Diagnosis: Parauchenoglanis chiumbeensis is distinguished from P.ahli , P. altipinnis , P. balaoi , P. buetikoferi , P. longiceps , P. monkei , P. pantherinus , and P. punctatus by a broad humeral process (vs. pointed humeral process). It is further distinguished from P. altipinnis , P. balaoi , P. pantherinus , and P. punctatus by coarse skin (vs. smooth skin). Parauchenoglanis chiumbeensis is distinguished from P. zebratus by humeral process clearly visible through the skin and anterior margin of the pectoral-fin spine mostly smooth (vs. humeral process embedded under the skin and anterior margin of the pectoral-fin spine mostly serrated). Parauchenoglanis chiumbeensis is readily distinguished from P. stiassnoae by truncated caudal fin (vs. rounded), dorsolaterally positioned eyes (vs. dorsally positioned), and the presence of regularly spaced vertical rows of spots on the flank (vs. irregularly spaced blotches). Parauchenoglanis chiumbeensis is distinguished from P. ngamensis , P. lueleensis , P. poikilos , P. ernstswartzi , and P. megalasma by absent background spots [ Fig. 15A; vs. present in P. ngamensis ( Fig. 5A), P. lueleensis ( Fig. 10A), and P. poikilos ( Fig. 11A) or vermiculated paưern in P. ernstswartzi ( Fig. 12A) or blotches in P. megalasma ( Fig. 13A)]. It is further differentiated from P. lueleensis , P. poikilos , P. ernstswartzi , and P. megalasma by relative long interdorsal–adipose distance, 6.3%– 12.7% SL (vs. narrow interdorsal–adipose distance, 3.7%–6.8% SL in P. lueleensis , 5.5%–5.8% SL in P. poikilos , 3.8%–5.0% SL in P. ernstswartzi , and 3.9%–4.8% SL in P.megalasma ). It is further differentiated from P. ernstswartzi and P. megalasma by vertical rows of spots [ Fig. 15A; vs. vertical rows of blotches in P. ernstswartzi ( Fig. 12A) and in P. megalasma ( Fig. 13A)]. Parauchenoglanis chiumbeensis is differentiated from P. luendaensis by faintly spoưed fins ( Fig. 15A; vs. unspoưed in P. luendaensis , Fig. 14A). It is further differentiated from P. poikilos by shorter prepectoral fin length, 26.7%–29.5% SL (vs. longer, 30.6%–31.4% SL in P. poikilos ). It is further differentiated from P. poikilos and P. megalasma by a shorter adipose fin, 26.0%–34.9% SL (vs. longer, 36.0%–36.6% SL in P. poikilos and 35.6%–36.9% SL in P. megalasma ). It is further differentiated from P. dolichorhinus by deeper body, 14.8%–21.3% SL (vs. shallow, 11.6%–14.0% SL in P. dolichorhinus ).

Description: Morphometric and meristic data are given in Table 3. Body elongated. Dorsal body profile gently rising from snout tip to origin of dorsal fin, and straight from the dorsal-fin origin to caudal-fin base. Body depth highest at origin of dorsal fin. Ventral body profile slightly convex from lower jaw to end of caudal-fin base. Caudal peduncle laterally compressed. Anus and urogenital opening positioned about halfway between pelvic- and anal-fin origins. Adipose fin longer than anal-fin base, originating anteriorly to anal-fin origin, between pelvic base and anal-fin origin, ending posteriorly to end of anal-fin base. Dorsal and pectoral fins with strong spines. Dorsal-fin origin anterior to pelvic-fin origin. Posterior tip of pectoral-fin rays does not reach pelvic fin. Posterior tip of pelvic-fin rays does not reach anal fin. Posterior margin of the pectoral-fin spine well serrated, and anterior margin mostly smooth. Caudal fin truncated.

Head moderately depressed. Snout bluntly triangular, on dorsal view ( Fig. 15B). Mouth subterminal. Lips fleshy. Eyes small and situated dorsolaterally. Ŋree pairs of barbels, base thick and tips pointed. External mandibular longest, reaching the distal tip of the pectoral-fin spine. Inner mandibular barbel shortest, reaching middle of the eye. Maxillary barbel reaching posterior edge of the eye. Posterior nostril slits positioned about halfway between tip of snout and eye.

Colouration in alcohol: Body brown dorsally and laterally, and light brown ventrally. Body with black spots smaller than the eye forming five to seven vertical rows; however, spots surrounded by a shade on one small specimen (53.5 mm SL). No additional spots in between the vertical rows, and this at all sizes. Head brown and unspoưed; however, spoưed on only two small specimens (53.5–58.2 mm SL). All fins with faint spots. Distal tip of fin rays with no markings. Usually, two black spots present above gill opening and one spot on caudal-fin base.

Distribution: Parauchenoglanis chiumbeensis is known from the Mololo and Tchimenji rivers, less bank affluents of the Chiumbe River.

Etomologo: Parauchenoglanis chiumbeensis is named asser the Chiumbe River, Kasai sub-basin, Angola, from which this species was collected. Ŋe suffix ‘- ensis ’, meaning ‘lives in’, has been added.

Identification key to Parauchenoglanis species

1a. Humeral process broadly triangular............................................................................................................................................................2

1b. Humeral process pointed........................................................................................................................................................................... 11

2a. Humeral process embedded under the skin; anterior margin of the pectoral-fin spine mostly serrated..................... P. zebratus

2b. Humeral process clearly visible through the skin; anterior margin of the pectoral-fin spine mostly smooth .............................3

3a. Rounded caudal fin ( Fig. 4A).......................................................................................................................................................................4

3b. Truncated caudal fin ( Fig. 4B) .....................................................................................................................................................................7

4a. Long barbels, external mandibular barbel reaching beyond distal tip of pectoral spine; total vertebrae count <30 ................... ......................................................................................................................................................................................................... P. stiassnoae

4b. Short barbels, external mandibular barbel never reaching beyond the distal tip of pectoral; total vertebrae count ≥32 ........... ............................................................................................................................................................................................................................5

5a. Head and fins with numerous black spots at all sizes (22.8–204.2 mm SL); body of medium- to large-sized specimens (>50.0 mm SL) with numerous spots, with some of them forming five to seven (median six) vertical rows and with other black spots situated in between those rows ( Fig. 5)............................................................................................................ P. ngamensis

5b. Head and fins with few black spots or spots entirely absent at all sizes (32.1–223.0 mm SL); body of medium- to large-sized specimens (> 100 mm SL) with black spots forming only five to six (median six, rarely seven) vertical rows and with no other spots situated in between these rows..........................................................................................................................................................6

6a. Head moderately depressed (head depth 48.5%–54.0% HL); deeper body depth, 15.5%–21.4% SL ( Fig. 11) ......................... ........................................................................................................................................................................................................ P. patersoni

6b. Head highly depressed (head depth 34.6%–40.0% HL); shallower body depth, 11.6%–14.0% SL ( Fig. 8) ................................ ................................................................................................................................................................................................ P. dolichorhinus

7a. Snout round or partly round; body with vertical rows of black blotches; head with black blotches.............................................8

7b. Snout bluntly triangular; body with black spots; head sometimes also with black spots................................................................9

8a. Vertical rows of black blotches extending onto adipose fin; blotches along the lateral line smaller than the eye size; eyes situated dorsally, i.e. high on the head, towards its upper surface ( Fig. 12) ..................................................................... P. ernstswartzi

8b. Vertical rows of black blotches not extending onto adipose fin; blotches along the lateral line larger than the eye size; eyes clearly situated dorsolaterally ( Fig. 13).............................................................................................................................. P. megalasma

9a. Head and fins with black spots; black spots present in between vertical rows of spots; external mandibular barbel not reaching the distal tip of pectoral-fin spine, when adpressed along the body ................................................................................. 10

9b. Head and fins without black spots or with faint black spots only; absence of black spots in between vertical rows of spots; external mandibular barbel reaching the distal tip of pectoral-fin spine, when adpressed along the body ............................... 11

10a. Mandibular barbels unspoưed; only one or two black spots in between vertical rows of black spots; long predorsal length, 39.4%–42.0% SL ( Fig. 10)........................................................................................................................................................ P. lueleensis

10b. Mandibular barbels spoưed; more than two black spots in between vertical rows of black spots; short predorsal length, 38.4%–39.2% SL ( Fig. 11)........................................................................................................................................................... P. poikilos

11a. Fins unspoưed; high adipose fin height ( Fig. 14)............................................................................................................ P. luendaensis

11b. Fins with faint black spots; low adipose fin height ( Fig. 15) ...................................................................................... P. chiumbeensis

12a. Anterior margin of the pectoral-fin spine completely serrated........................................................................................................... 13

12b. Anterior margin of the pectoral-fin spine only partly and slightly serrated, or smoothly serrated ............................................. 15

13a. Caudal peduncle relatively long as high ............................................................................................................................... P. buetikoferi

13b. Caudal peduncle shorter than high.......................................................................................................................................................... 14

14a. Barbels long, with external mandibular barbel reaching beyond distal tip of pectoral-fin spine; colour paưern with 6–10 (rarely 11) vertical rows of spots .............................................................................................................................................. P. punctatus

14b. Barbels short, with external mandibular barbel not reaching beyond distal tip of pectoral spine; colour paưern with five to seven bands....................................................................................................................................................................................... P. monkei

15a. Preorbital head length greater than maximum head depth (measured at level of supraoccipital process); numerous small black spots on both head and fins............................................................................................................................................................. 16

15b. Preorbital head length less than maximum head depth; colour paưern of body and fins variable, but not as above.............. 17

16a. Interorbital distance 28%–30% HL; spots on head and flanks of equal size ..................................................................... P. longiceps

16b. Interorbital distance 22%–25% HL; spots on head smaller than those on flanks...................................................... P. pantherinus

17a. Adipose and dorsal fins high (adipose fin height, 5%–8% SL; dorsal fin height, 19%–30% SL); colour paưern on flank faint, one horizontal row of two to six large spots, being as large as or larger than the eye diameter, visible on the level of lateral line .......................................................................................................................................................................................................... P. altipinnis

17b. Adipose and dorsal fins shallower (adipose fin height, 3%–7% SL; dorsal fin height, 15%–25% SL); colour paưern of body and fins variable but not as above............................................................................................................................................................. 18

19a. Barbels long, with external mandibular barbel reaching distal tip of pectoral spine (minimum one-third HL); whole body covered with spots of equal size.......................................................................................................................................................... P.ahli

19b. Barbels short, with external mandibular barbel never reaching beyond distal tip of pectoral spine (shorter than one-third HL); whole body covered with spots of different sizes, arranged in horizontal and/or vertical rows............................. P. balaoi

DISCUSSION

A recent alpha-taxonomic study on Parauchenoglanis species diversity taking an integrative approach including both genetics (COI barcoding) and morphology, i.e. colour paưern, meristic, and morphometric characters, documented the presence of hidden diversity in three of the widely distributed species, namely P. balaoi , P. punctatus , and P. ngamensis ( Sithole et al. 2023) . In the present study, a more in depth alpha-taxonomic revision was undertaken of P. ngamensis , a species which has long been considered to be widespread and found in four sub-basins of sub-Saharan Africa covering part of three IPs ( Skelton 2001, Geerinckx et al. 2004, Marshall 2011). Ŋis revealed, besides P. ngamensis itself, the existence of eight new species for science. Interestingly, six of the new species herein described are found in the Kasai sub-basin ( middle Congo Basin) and two in the Kwanza Basin. Ŋe number of Parauchenoglanis species in the Kasai sub-basin has thus increased from one to seven, the six new from the present study, and an additional recently described species, P. stiassnoae , from the Mfimi River, an affluent of the Kasai ( Modimo et al. 2024). Most interestingly, P. ngamensis , which was previously thought to be broadly distributed, was found to be confined to the Okavango and upper Zambezi sub-basins. In agreement with Sithole et al. (2023), in this study three clades were recovered within P. punctatus . However, any decision and interpretation on the taxonomy of P. punctatus in the present study is considered premature, requiring additional specimens to be sequenced and examined morphologically. Ŋus, the scope of the present study is limited to understanding the species diversity within the P. ngamensis group.

Ŋus, the total number of valid species recognised within the genus Parauchenoglanis doubled, from nine, as reported by Geerinckx et al. (2004), to 19, including the recently described P. zebratus ( Sithole et al. 2023) , P. stiassnoae ( Modimo et al. 2024) , and the eight new species described in the present paper. Ŋe discovery of new species, many of them occurring even in the same sub-basin, adds to the increasing evidence that many widely distributed species harbour undocumented species diversity. Ŋis has already been recovered for other African catfish genera, such as Amphilius Günther 1864 ( Ŋomson and Page 2010, Ŋomson et al. 2015, Mazungula and Chakona 2021) and Chiloglanis Peters 1868 ( Schmidt et al. 2016, 2017, Chakona et al. 2018).

Ŋe present study suggests that the P. ngamensis species group is not monophyletic, with three distinct clades including six of the eight new species herein described and P. ngamensis . However, considering the low support values of the deeper nodes in the phylogeny and mainly that it relied only on the mitochondrial gene COI, we believe that it is not possible to assume confidently that the group is not monophyletic. A future study including additional loci, preferably nuclear genes, will be able to test the monophyly of the P. ngamensis group. Ŋus, despite the suggested non-monophyletism, the group, which is easily diagnosable morphologically, is important and will probably facilitate future species discovery and descriptions. Ŋe nine species that belong to the P. ngamensis group share a broad humeral process, a rough skin, a rounded and truncated caudal fin, a short mandibular barbel, and the presence of five to seven vertical bars or rows of spots or blotches along the flank.

Parauchenoglanis ngamensis specimens originating from Lake Calundo on the upper Zambezi sub-basin were previously reported to be a possible hybrid between Kasai and Okavango – Zambezi populations ( Retzer 2008) or a putative new species ( Sithole et al. 2023) based on colour paưern. However, in the phylogenetic COI barcoding tree analysis, haplotypes of specimens with a colour paưern similar to that of the Lake Calundo specimens clustered within the clade of the specimens from Okavango and upper Zambezi (sub)basins. Hence, in this study, the Lake Calundo specimens are all considered to belong to P. ngamensis . Ŋe previous studies might have overlooked the colour paưern variation related to size within P. ngamensis specimens. Ŋerefore, it is concluded that P. ngamensis is distributed in both the Okavango and upper Zambezi (sub)basins, including Lake Calundo. Ŋis results in a more restricted distribution in comparison to what was previously reported for the species ( Skelton 2001, Geerinckx et al. 2004, Marshall 2011, Bruton et al. 2018). Ŋe Okavango and Zambezi (sub)basins were once connected and still connect in periods of high rainfalls via the Selinda Spillway ( Tweddle 2010, Musilová et al. 2013, Skelton 2019). Ŋis most probably allows interbasin dispersal, thus facilitating gene flow across populations from the two basins.

Similar distribution paưerns have been reported for other fish species, such as the distichodontids Nannocharax machadoi ( Poll 1967) , Nannocharax multifasciatus Boulenger 1923 , and Nannocharax dageti Jerep, Vari & Vreven 2014 ( Bragança et al. 2020), and the mormyrid Marcusenius altisambesi Kramer, Skelton, van der Bank & Wink 2007 . Ŋe inclusion of additional genetic markers and future phylogeographical study will be required to provide further insights into the possible mechanisms that allow for the maintenance of gene flow between the Zambezi and Okavango basins.

Intra- and interspecific colour paIJern variation Colour paưern differences were revealed to be extremely useful in distinguishing the species belonging to the P. ngamensis group. However, intraspecific size-related, i.e. ontogenetic, changes in colour paưern were also documented. Such changes were found in three of the nine species identified, i.e. in P. ngamensis and in both species identified from the Kwanza Basin, P. patersoni and P. dolichorhinus . Ŋese three species were observed to be genetically, at least based on COI barcoding evidence, closely related.

Ŋe colour paưern of small-sized specimens of these three species is mostly characterised by black bars (≤ 100 mm SL), whereas medium- to large-sized specimens are characterised by well-distinct vertical rows of black spots instead (> 100 mm SL). Such size-related colour paưern changes seem to be rather common in Parauchenoglanis species ( Seegers 2008). Ŋey have also been reported for P. monkei , a lower Guinea species ( Geerinckx et al. 2007, Seegers 2008), and recently documented for P. zebratus (upper Lualaba, upper Congo Basin) ( Sithole et al. 2023). Additionally, this has also reported in other catfish species. For instance, Ŋomson et al. (2015) reported different colour paưerns in Amphilius ( Amphiliidae ) between small- and large-sized specimens, in Amphilius jacksonii Boulenger 1912 , Amphilius pedunculus Ŋomson & Page 2015 , and Amphilius crassus Ŋomson & Hilber 2015 . Changes in colour paưern between juvenile and adult freshwater fishes are known to be a result of morphological changes that occur during ontogeny, which involve modifications in pigment concentration and in the density and morphology of chromatophores ( Leclercq et al. 2010, Nüsslein‐Volhard and Singh 2017). Additionally, these changes in colour paưern can be influenced by selective pressures, with some species evolving distinct colour paưerns to serve specific ecological or behavioural functions. For instance, these adaptations might be driven by the need for effective camouflage to avoid predators or to blend into their habitat, or they could play a role in signalling behaviours and reproductive states ( Maan and Sefc 2013). Although we have a foundational understanding of these processes, further research is needed to understand the drivers and significance of changes in colour paưern along the ontogeny in some species of Parauchenoglanis .

None of the six new species identified in the Kasai sub-basin, however, presents such size-related changes in colour paưern. Ŋis was evident in species for which a large number of small- and large-sized specimens are available, i.e. P. chiumbeensis , P. ernstswartzi , P. lueleensis , and P. luendaensis . However, further sampling for at least some of the species from the Kasai sub-basin, i.e. P. poikilos and P. megalasma , is needed because these are, to date, represented by only a few specimens of a similar size class ( Table 3).

Distribution paIJerns of the different species of the P.

ngamensis species group

Despite the existence of one widely distributed species, P. ngamensis , in the Okavango and the upper Zambezi (sub)basins, the distribution paưern of other species of the P. ngamensis group can be categorised in two main paưerns: sympatric and allopatric (Supporting Information, Table S3). Two species, i.e. P. patersoni and P. dolichorhinus , occur in sympatry in the Kwanza Basin. Ŋe current sympatric occurrence of both P. patersoni and P. dolichorhinus in the middle and the upper Kwanza Basin is probably aưributable to a past speciation event that occurred within the basin. In the phylogenetic analysis, P. dolichorhinus (= P. ngamensis sp. ‘kwanza 2’) seems to be the sister clade to P. patersoni (= P. ngamensis sp. ‘kwanza 1’) and P. ngamensis (= P. ngamensis sp. ‘okavango–zambezi’). It can be hypothesised that both Kwanza species originated in distinct subparts/sub-basins of the upper Kwanza Basin, with a subsequent downstream colonization.

Additionally, three species ( P. lueleensis , P. megalasma , and P. ernstswartzi ) show some degree of distribution overlap in the Luele River, Kasai sub-basin ( middle Congo Basin). Parauchenoglanis megalasma and P. ernstswartzi occur in sympatry with P. lueleensis , which has a slightly wider distribution. Interestingly, both P. megalasma (= P. ngamensis sp. ‘kasai 4’) and P. erntswartzi (= P. ngamensis sp. ‘kasai 3’) are well supported as sister species. Instead, P. lueleensis belongs to a highly distinct COI barcoding lineage, being the sister species of P. poikilos known from a neighbouring upstream affluent, the Lovuo River. Ŋe genetic COI barcoding data seem to suggest that there is no ongoing gene flow between these lineages, i.e. species. Further studies should investigate the (micro)habitat preferences and behaviour (food use) of these Parauchenoglanis species to understand the factors that probably contribute to and/or enable their coexistence in the Kasai and Kwanza Basin.

Some of the species described from the Kasai sub-basin ( middle Congo Basin) are distributed in parallel affluents. Ŋis is true for P. poikilos (= P. ngamensis sp. ‘kasai 2’) known to occur in the Lovua River, P. luendaensis (= P. sp. ‘luenda’) in the Luenda River, and P. chiumbeensis (= P. sp. ‘chiumbe’) in the Chiumbe River, respectively. Unfortunately, owing to the lack of DNA data for two of these species inhabiting the Kasai sub-basin, it is not possible to suggest the most likely biogeographical process explaining the current distribution of these Parauchenoglanis species in the Kasai sub-basin.

Allopatric speciation in freshwater fishes is influenced by vicariance events, such as changes in basin affiliation owing to river captures, or the presence of physical barriers, such as waterfalls ( Bernardi 2013, Musilová et al. 2013, Seehausen and Wagner 2014). During river captures, basin affiliation can be reshuffled, resulting in its fauna being transferred from one basin to another ( Burridge et al. 2007). Ŋe distribution of P. ngamensis s.l. could have been shaped by some past river capture events. For example, the upper Kasai River is considered to have been part of the upper Zambezi in the past until it was captured by the Congo River (Bell-Cross 1965, Skelton 2019: fig. 11.3). Ŋis most probably resulted in the passive transfer of some species from the former to the laưer basin.

Ŋe high number (seven) of endemic species of Parauchenoglanis identified so far in the Kasai sub-basin is particularly intriguing, with all species with a distribution range limited to a single affluent of it. Ŋis might, however, be explained when looking at the hydrography of the Kasai sub-basin, which is mostly composed of numerous parallel, south–northoriented affluents ( Runge 2007: fig. 14.4). Ŋese originate on the Bié Plateau and flow further northwards, below the plateau, to converge with the Kasai River as less bank affluents of it, in the cuveưe central of the Middle Congo Basin ( Mbuebue et al. 2016). Furthermore, the existence of numerous waterfalls within this sub-basin has served for some fish species, at least, as a physical barrier to upstream migration ( Ŋieme et al. 2005, Rahel 2007, Schluter 2009, Kano et al. 2012, Mbimbi et al. 2021). Ŋus, preventing genetic exchange between populations, thereby leading to genetic isolation and potentially driving speciation ( Kano et al. 2012). Ŋerefore, this highly remarkable configuration seems ideal to propagate speciation in each of those parallel affluents and between the up- and downstream parts of these affluents. As such, this part of the Middle Congo Basin might contain an important amount of yet undocumented species diversity, not only within the genus Parauchenoglanis but also in many other fish genera occurring in the Kasai sub-basin. Ŋis phenomenon and its related species diversity might have remained largely invisible for several reasons. For example, there has been an overall poor exploration of the ichthyofauna of the Kasai sub-basin, resulting in a limited amount of specimens available from the different affluents that compose it.

Anthropogenic impacts and conservation/protection issues According to the last International Union for Conservation of Nature (IUCN) Red List evaluation, now more than a decade ago (2010), P. ngamensis is listed as least concern (LC), owing to its wide distribution (Marshall et al. 2010). In this study, however, it has been demonstrated to be a species group containing at least nine species, with each, except for the nominal species itself, having a much more restricted single sub-basin distribution. Parauchenoglanis ngamensis , however, remains the most widely distributed species in the species group, covering two adjacent, large (sub)basins, i.e. the Okavango and the upper Zambezi.

Anthropogenic activities, such as: (i) habitat destruction and alteration, e.g. through dam building for the generation of electricity for industrial purposes, such as mining; (ii) overfishing, e.g. owing to the increased fishing effort resulting from commercial and subsistence fisheries; (iii) the introduction of invasive species, e.g. the introduction of the Nile tilapia, Oreochromis niloticus (Linnaeus 1758) , for aquaculture; and (iv) pollution, e.g. the use of artificial fertilizers, have been observed in the Okavango and upper Zambezi basin catchments ( Ŋieme et al. 2005, Bruton et al. 2018). However, both the Okavango and upper Zambezi rivers are regarded as only lightly impacted by these factors. As a result, the majority of the fish species occurring in these basins have been classified as Least Concern (LC) ( Tweddle et al. 2009, Bruton et al. 2018). Ŋe conservation status of these newly described species needs to be evaluated following the IUCN Red List criteria, in light of the new evidence for the existence of taxa with more restricted distribution ranges.

Ŋe Kwanza Basin is threatened by dams built on the main river. For instance, by 2019 three hydroelectric dams were finalised, and building of more dams is anticipated ( Skelton 2019). Furthermore, an alien species, the Mozambique tilapia, Oreochromis mossambicus (Peters 1852) , has been introduced into the basin ( Skelton 2019). Ŋe Kasai Basin, which harbours the greatest species diversity of Parauchenoglanis , is currently threatened by intense diamond and copper mining activities, which are causing major habitat modifications ( Mbimbi et al. 2021). Ŋe area is also impacted by deforestation and overfishing, with the use of illegal, small mesh-size gill nets. In addition, all these activities are anticipated to intensify further following the demographic growth in the area ( Stiassny et al. 2011, Mbimbi et al. 2021). Despite this, the Kasai and Kwanza basins remain poorly studied. Ŋis has resulted in most species being assessed as Data Deficient (DD), Least Concern (LC), or Not Evaluated (NE) ( Skelton 2019). Ŋe discovery of hidden species diversity in both basins highlights the crucial need for further exploration in these basins, in order to document their freshwater fish species diversity further. As such, it is hoped that the present paper and the important species diversity described herein might be an incentive to take proper care of fish conservation/protection in both Southern Africa and the southern Congo sub-basins.

SU P P LE M E N TA RY DATA

Supplementary data is available at Zoological Journal of the Linnean Societo online.