Parauchenoglanis stiassnyae (Siluriformes: Auchenoglanididae): A new species of giraffe catfish from Mfimi-Lukenie basin, central Africa, Democratic Republic of Congo

REGULAR ARTICLE Parauchenoglanis stiassnyae (Siluriformes: Auchenoglanididae): A new species of giraffe catfish from Mfimi-Lukenie basin, central Africa, Democratic Republic of Congo Myriam Y. Modimo 1 | Maxwell J. Bernt 2 | Raoul J. C. Monsembula Iyaba 1 | José J. M. M. Mbimbi 1 | Tobit L. D. Liyandja 1,3,4 1 Department of Biology, University of Kinshasa, Kinshasa, Democratic Republic of the Congo 2 Department of Biology, Susquehanna University, Selinsgrove, Pennsylvania, USA 3 Department of Biological Sciences, University of Toronto Scarborough, Toronto, Canada 4 Department of Natural History, Royal Ontario Museum, Toronto, Canada Correspondence Tobit L. D. Liyandja, Department of Biological Sciences, University of Toronto Scarborough, Toronto, Canada. Email: [email protected]; [email protected]n.ca and tobit.liyandja@ unikin.ac.cd Funding information National Science Foundation Graduate Research Fellowship Program, Grant/Award Number: 1655227; Axelrod Research curatorship (MJLS) Abstract A new, distinctively short-bodied giraffe catfish of Parauchenoglanis is described from the Ndzaa River, a small left-bank tributary of the Mfimi-Lukenie basin in the Central basin of the Congo River in the Democratic Republic of the Congo. The new species can be distinguished from all congeners by having 29 or fewer (vs. 33 or more) total vertebrae. It can further be distinguished from all congeners, except Parauchenoglanis zebratus Sithole et al., 2023 and Parauchenoglanis ngamensis (Boulenger 1911), by having 13 or 14 (vs. 16 or more) pre-anal vertebrae. The species is endemic to the Mfimi River basin, where it has been collected mainly in blackwater tributaries. KEYWORDS Congo basin, CT scan, DNA barcoding, morphology, Ndzaa River, Parauchenoglanis 1|INTRODUCTION With only three genera currently recognized and 21 valid species (Fricke et al., 2024), the African catfish family Auchenoglanididae, recently separated from the family Claroteidae (Lundberg et al., 2007), is one of the less diverse African catfish families. For example, its most closely related family, Claroteidae (Lundberg et al., 2007; Schedel et al., 2022), has three times more species (Fricke et al., 2024). However, this lack of recorded diversity likely reflects taxonomic confusion, poorly defined species and genera (Geerinckx et al., 2013), and the existence of species complexes (Sithole et al., 2023). Among the three auchenoglanidid genera currently recognized (Geerinckx et al., 2013), Parauchenoglanis Boulenger, 1911 is the most diverse, with 10 recognized species (Ferraris 2007, Sithole et al., 2023, Fricke et al., 2024). Parauchenoglanis was erected by Boulenger (1911) to accommodate two Auchenoglanis species, Auchenoglanis guttatus and Auchenoglanis macrostoma, that he noted were significantly different from Auchenoglanis sensu stricto (Geerinckx et al., 2004). Jordan (1920)designatedPimelodus guttatus (Lönnberg1895)asthetypespeciesofParauchenoglanis, yetonlyaftertheworkofTeugelsetal.( 1991)wasParauchenoglanis clearly defined and delineated (Geerinckx et al., 2004). The taxonomic history of Parauchenoglanis encompasses 18 nominal species described across the African continent. A systematic revision of the genus conducted by Geerinckx et al. (2004) reduced the number of valid Parauchenoglanis species to nine. However, species delimitation and diagnostic characters for Parauchenoglanis species still remain poorly defined. In a recent study describing a new species, Parauchenoglanis zebratus, Sithole et al. (2023) highlighted cryptic diversity of Received: 25 March 2024 Revised: 5 July 2024 Accepted: 12 July 2024 DOI: 10.1111/jfb.15885 FISH This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. © 2024 The Author(s). Journal of Fish Biology published by John Wiley & Sons Ltd on behalf of Fisheries Society of the British Isles. J Fish Biol. 2024;105:1227–1239. wileyonlinelibrary.com/journal/jfb 1227 Parauchenoglanis in central and southern Africa, detecting several undescribed lineages. However, the latter study did not include specimens from the central Congo basin. In a recent study of fishes of the Mfimi River in the central Congo basin of the Democratic Republic of the Congo (DRC), two species of Auchenoglanididae were reported (Stiassny, Alter, Liyandja, et al., 2021): Parauchenoglanis punctatus (Boulenger 1902) and Notoglanidium macrostoma (Pellegrin 1990). One of the three specimens of “N. macrostoma,”including in the type series herein described, was collected in the Ndzaa River (Figure 1), a small left-bank tributary of the Mfimi River, itself a tributary of the Kasai drainage in the DRC. Stiassny, Alter, Liyandja, et al.'s (2021) assignments of these specimens to N. macrostoma were based on examination of external morphology of a few juvenile specimens. Subsequent expeditions to the Ndzaa River, between August and October 2021, June and July 2022, and August and September 2023, allowed the collection of several additional specimens of varying sizes. A more in-depth morphological and osteological study of these specimens coupled with molecular analyses indicated that they represent an undescribed lineage of Parauchenoglanis. The objective of this study is to provide a formal description of this new species of giraffe catfish from the Ndzaa River. The Ndzaa is a left-bank tributary entering the Mfimi, at 290 m a.s.l., near the settlement of Kutu (Mai-Ndombe Province, DRC) at the outflow of Lake Mai-Ndombe, upstream of which the river is named the Lukenie. The Ndzaa (also known as Ndjua or Ndjuw River) is a tea-colored stream meandering through dense riparian forest surrounded by open grasslands and savannah. The Ndzaa originates in southern Kutu territory, near the city of Semendwa, at 395 m a.s.l., and its catchment drains an area of 370 km 2 dominated by shrub savannahs and humid forests. The area drained by the Ndzaa River is under a humid tropical climate characterized by two main seasons: a longer wet season (from September to mid-May) and a shorter dry season (from mid-May to late August). The longer wet season is interrupted by a very short dry period from mid-January to mid-February (Bolanzowu et al., 2019). The pH in the Ndzaa River is generally acidic (pH 4.1–5.3) and differs from the characteristically humic and brownblack waters of the central basin of the Congo. The water conductivity is low (10–50 μS/cm), and the river is low in dissolved solids (TDS: 10–70 mg/L). 2|MATERIALS AND METHODS 2.1 |Ethics statement Type specimens were collected and euthanized in accordance with guidelines for the use of fishes in research (Jenkins et al., 2014) and ethical considerations for field research (Bennett et al., 2016). The collection and exportation of these fishes were conducted with permission of the Congolese Ministère de l'Agriculture, Secrètariat General à l'Agriculture, Pêche et Elevage, Direction des Pêches (permits 037/DP/SG/AGRIPEL/2016, 03/DP/SG/PEL/2018, and 23/DP/SG/PEL/2021, all on file at American Museum of Natural History [AMNH]). 2.2 |Molecular data collection and analyses We used the Qiagen Gentra Puregene Tissue Kit and manufacturer's protocols to extract total genomic DNA from four individuals of three Parauchenoglanis species from the Ndzaa River: P. punctatus, P. cf. punctatus_L3 (referred to as Parauchenoglanis monkei [Keilback FIGURE 1 Map indicating the location of the Ndzaa River and distribution of Parauchenoglanis stiassnyae sp. nov. in the MfimiLukenie basin. Colored areas represent different ecoregions following Abell et al. (2008). 1228 MODIMO ET AL. FISH 10958649, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jfb.15885 by Capes, Wiley Online Library on [29/08/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 1910] by Monsembula Iyaba et al., 2013 and as P. punctatus_L3 by Sithole et al., 2023), and the new species herein described. We amplified a 652-bp portion of the cytochrome coxidase subunit 1 (COI) which was sequenced on a Sanger sequencing platform following the methods of Lowenstein et al. (2011). All sequences have been deposited in GenBank (Table 1). We obtained 28 additional sequences (22 Parauchenoglanis, three Auchenoglanis [out-group], and three Notoglanidium [out-group]) from the Barcode of Life Data System (http://www.barcodinglife.org) and GenBank (www.ncbi.nlm.nih.gov/ genbank). We aligned COI sequences, trimming extremities, and calculated the percentage divergence between sampled taxa using Geneious Prime 2024.0 (https://www.geneious.com). We conducted exploratory phylogenetic analyses using maximum likelihood (ML) as implemented in IQ-TREE (Nguyen et al., 2015). 2.3 |Morphological data collection and analyses Institutional abbreviations follow Sabaj (2022), AMCC stands for the Ambrose Monel Cryo Collection of the AMNH. A total of 55 specimens of Parauchenoglanis were examined, including six Parauchenoglanis balayi Sauvage 1879, four Parauchenoglanis guttatus, four Parauchenoglanis longiceps Boulenger 1913, five Parauchenoglanis pantherinus Pellegrin 1929, eight P. punctatus,10P. cf. punctatus_L3, 14 specimens of Parauchenoglanis stiassnyae sp. nov. (see description), and four specimens of Parauchenoglanis ubangensis Boulenger 1902. Six meristic counts were taken on each specimen. Fin rays were counted under a stereomicroscope and verified using X-ray images. Vertebrae were counted, excluding the terminal preural centrum and Weberian vertebrae, using both X-ray and micro-computed tomography (μCT) scans (for some specimens) (Figure 2). Forty-five standard morphometric measurements following Geerinckx et al. (2004) were taken on each specimen. These measurements were taken pointto-point using a TRESNA SC30 digital caliper with an accuracy of 0.01 mm. Meristic and morphometric data were analysed separately using principal component analysis (PCA) as implemented in the R package FactoMineR (Lê et al., 2008) and the programme PAST 4.12b (Hammer et al., 2001). Morphometric data were analysed as logtransformed proportions of standard length (SL) to account for body size differences between species, except for fin lengths that were TABLE 1 GenBank and Barcode of Life Data System (BOLD) accession numbers, tissue codes, and catalog numbers for COI sequences of Parauchenoglanis specimens utilized in this study. Taxon Catalog number Tissue code COI Publication Parauchenoglanis balayi –A5-36-96 MK074561 Sonet et al., 2018 Parauchenoglanis cf. pantherinus –A7-31-739 MK074565 Sonet et al., 2018 P. cf. pantherinus –A7-31-738 MK074566 Sonet et al., 2018 P. cf. pantherinus –A7-31-602 MK074567 Sonet et al., 2018 Parauchenoglanis cf. punctatus AMNH 278142 AMCC 284824 PP461583 This study P. cf. punctatus –A7-31-764 MK074562 Sonet et al., 2018 P. cf. punctatus –A7-31-754 MK074563 Sonet et al., 2018 P. cf. punctatus –A7-31-765 MK074564 Sonet et al., 2018 P. cf. punctatus –A9-29-3533 KT192843 Decru et al., 2016 P. cf. punctatus –A9-29-3577 KT192848 Decru et al., 2016 P. cf. punctatus AMNH 250764 AMCC 256659 HM418200 Unpublished P. cf. punctatus AMNH 250860 AMCC 256672 HM418201 Unpublished P. cf. punctatus –A9-29-3232 KT192787 Decru et al., 2016 P. cf. punctatus –A9-29-3240 KT192789 Decru et al., 2016 Parauchenoglanis monkei –T391 HG803488 Peart et al., 2014 P. monkei –T568 HG803492 Peart et al., 2014 P. punctatus –B4-16-1805-J KX186050 Unpublished P. punctatus –B4-16-1813-J KX186045 Unpublished P. punctatus AMNH 278134 AMCC 284822 PP461584 This study Parauchenoglanis stiassnyae sp. nov. AMNH 278138 AMCC 284847 PP461585 This study P. stiassnyae sp. nov. AMNH 278139 AMCC 284823 PP461586 This study Parauchenoglanis sp. –B9-17-4663 KT193142 Decru et al., 2016 Parauchenoglanis sp. –SAIAB ES08 B148 SAFW418-08 Unpublished Parauchenoglanis sp. –SAIAB ES07 F 041 SAFW237-08 Unpublished Parauchenoglanis sp. –SAIAB ES08 B 308 SAFW563-09 Unpublished Parauchenoglanis sp. –SAIAB ES08 B310 SAFW565-09 Unpublished Abbreviation: AMCC, Ambrose Monel Cryo Collection; AMNH, American Museum of Natural History. MODIMO ET AL.1229 FISH 10958649, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jfb.15885 by Capes, Wiley Online Library on [29/08/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License excluded due to fin damage. Invariant meristic counts (simple and branched dorsal-fin rays) were removed from subsequent analyses. Four specimens of the new taxon and several comparative specimens of other Parauchenoglanis species were scanned using μCT at the AMNH Microscopy and Imaging Facility. Scans were made using a GE Phoenix vjtomejx using a 240-kV Nano Tube (General Electric, Fairfield, CT, USA), with resolution ranging from 10.6 to 24.3 μm. Beam energy was 120 kV and 166 mA. Scans were reconstructed using Phoenix datosjx (General Electric, Wunstorf, Germany) and were rendered and edited using VGStudio Max 3.3.4 (Volume Graphics, Heidelberg, Germany). 3|RESULTS 3.1 |Barcoding and phylogeny After trimming, we obtained a final alignment of 641 bp, for 32 specimens, including 179 distinct patterns, 162 parsimony-informative, 20 singletons, and 459 constant sites. Both AIC and Bayesian Information Criterion (BIC) selected the transition model with unequal base frequencies, empirical codon frequencies counted from the data, and discrete gamma with four categories (TIM +F+G4) as the best-fit substitution model for the alignment. The percent divergence calculation estimated that P. stiassnyae sp. nov. COI sequences diverge by more than 8.8% from all other sampled specimens except one from the Itimbiri River (KT193142), from which P. stiassnyae sp. nov. differs by only 2% (Table 2). Phylogenetic analyses retrieved the specimen from the Itimbiri River (KT193142) as the sister to P. stiassnyae sp. nov., which together form the sister to P. balayi from the KouilouNiari River system in the Lower Guinean ichthyofaunal province (Figure 3). Further investigation is needed to establish if the Itimbiri River specimens and P. stiassnyae sp. nov. are conspecifics or not. We did not have access to the specimens from Itimbiri and could not morphologically verify their identification. 3.2 |Meristics After removal of two invariant counts (unbranched and branched dorsal-fin rays), a PCA was performed on the five remaining meristic counts (caudal-fin rays, anal-fin rays, pre-anal vertebrae, caudal vertebrae, and total vertebrae) for all 55 specimens of the eight examined Parauchenoglanis species. We found that 98.5% of variation is explained by the first two principal components, with PC1 accounting for 94.9% and PC2 3.6% of variation. Differences in total, pre-anal, and caudal vertebrae counts contributed most to PC1 factor loadings, with the number of total and pre-anal vertebrae having the highest influence, whereas differences in the number of caudal vertebrae and anal-fin rays contributed most to PC2 loadings. A plot of PC1 against PC2 divided the eight examined Parauchenoglanis species into three groups: P. stiassnyae sp. nov. was the only member of the first group with the fewest total (28–29) and pre-anal (13–14) vertebrae; the second group, with intermediate total vertebral counts (33–35, mostly 33–34), contains P. longiceps and P. pantherinus; the third group, with the highest vertebral count (35–38, most with more than 35) includes P. balayi, P. guttatus,P. punctatus,P. cf. punctatus_L3, and P. ubangensis (Figure 4). 3.3 |Morphometrics A PCA was also performed on 42 morphometric measurements after removal of total length (TL), SL, and pectoral-spine length (due to spine damage) (Figure 5). The first six principal components accounted for 66.2% of total variation with PC1, PC2, and PC3 accounting for 25.1%, 12.7%, and 11.2% of variation, respectively. Differences in the interpectoral distance (5.6%), orbital diameter (4.8%), mouth width TABLE 2 Distance matrix indicating average percentage difference in partial cytochrome c oxidase subunit I (MT-COI) sequences among sampled Parauchenoglanis. Taxon 1 2 3 4 5 6 7 8 1. Parauchenoglanis balayi 2. Parauchenoglanis cf. punctatus L2 10.6 3. P. cf. punctatus L3 8.7 3.7 4. Parauchenoglanis cf. pantherinus 9.9 5.2 5.04 5. Parauchenoglanis monkei 10.9 11.3 10.5 10.5 6. P. punctatus 10.4 6.7 5.3 5.6 11 7. Parauchenoglanis sp. KT193142 9.7 10.1 8.5 9.1 11.1 10.4 8. Parauchenoglanis sp. Quanza 10.2 4.7 3.9 5.7 11 5.6 9.2 9. Parauchenoglanis stiassnyae 10.6 9.8 8.9 9.3 11.5 10.5 1.8 9.4 FIGURE 2 Micro-computed tomography (CT) scan the holotype of Parauchenoglanis stiassnyae sp. nov. (AMNH 278139), illustrating pre-anal and caudal vertebrae. 1230 MODIMO ET AL. FISH 10958649, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jfb.15885 by Capes, Wiley Online Library on [29/08/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License (4.6%), minimum caudal peduncle height (4.5%), pre-dorsal length (4.1%), and the head width (HW) (4%) contributed the most to PC1 factor loadings, whereas differences in interorbital distance (8.3%), preorbital head length (HL) (8.1%), HL (7.8%), anterior nostrils interdistance (7.6%), supraoccipital process–dorsal-fin interdistance (6.8%), and adipose-fin–caudal-fin interdistance (5.9%) contributed the most to PC2 factor loadings. Differences in pectoral-fin length (7.05%), orbital head height (6.1%), premaxillary toothplate width (5.9%), maximum caudal peduncle height (5.6%), prepectoral length (5.4%), and maximum body height (5.2%) contributed most to PC3 factor loadings. Overall, biplots of PC1 versus PC2 and PC1 versus PC3 divided species into three groups (Figure 5): the first group exclusively contained P. stiassnyae sp. nov., the second contained species from Lower Guinea (P. balayi,P. guttatus,P. longiceps, and P. pantherinus), whereas the third group contained all remaining species (P. punctatus,P. cf. punctatus_L3, and P.ubangensis). 3.4 |Taxonomic description P. stiassnyae, sp. nov., Zoobank acession numbers: urn:lsid:zoobank. org:act:60B7B236-14C5-497C-9E1A-AAF2DA952047 and urn:lsid: zoobank.org:pub:372F589C-0473-41F7-9873-FA30F1246992. N. macrostoma: Stiassny, Alter, Liyandja, et al., 2021 and Stiassny, Alter, Monsembula, & Liyandja, 2021. 3.4.1 | Holotype AMNH 278139 (AMCC 284823), 64.72 mm SL, main channel of the Ndzaa River over mud and plant debris, in forest habitat, 4.6 km upstream of the Ndzaa confluence with the Mfimi River, Kutu Territory, Mai-Ndombe Province, D. R. Congo, 0248017.300 S, 01811008.000 E, August 10, 2021, M. Y. Modimo. FIGURE 3 Maximum likelihood (ML) hypothesis of possible placement of Parauchenoglanis stiassnyae sp. nov. Bootstrap values are reported on/under branches. Sequences generated in this study are indicated in blue. L2 and L3 represent morphologically cryptic lineages of Parauchenoglanis punctatus as identified by Sithole et al. (2023). MODIMO ET AL.1231 FISH 10958649, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jfb.15885 by Capes, Wiley Online Library on [29/08/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 3.4.2 | Paratypes N=13. All Mai-Ndombe Province, D. R. Congo: AMNH 278138 (AMCC 284847), 60.22 mm SL, main channel of the Ndzaa River over mud and plant debris, in forest habitat, 6.9 km upstream of the Ndzaa River confluence with the Mfimi River, Kutu Territory, 0248055.800 S, 01811019.000 E, August 8, 2021, M. Y. Modimo; AMNH 274612 (AMCC 258190), 70.83 mm SL, main channel of the Mfimi River around the Nioki Port, in grass habitat, Kutu Territory, 0243025.400 S, 01741038.500 E, July 20, 2018, Fishermen; AMNH 269908 (two specimens), 66.24–66.28 mm SL, Lebéé River (tributary of Mfimi River) near Kilako village in about 16.7 km downstream of the town of Kutu, over mud, in grass habitat, Kutu Territory, 0248030.9200 S, 01801034.9000 E, August 8, 2015, R. Monsembula et al.; AMNH 278137 (1, CT-scanned), 73.9 mm SL, main channel of the Ndzaa River over mud and vegetal debris, in forest habitat, in about 7.3 km upstream of the Ndzaa River confluence with the Lukeni River, Kutu Territory, 0249029.400 S, 01811031.700 E, August 7, 2021, M. Y. Modimo; AMNH 278164 (1), 71.7 mm SL, main channel of the Ndzaa River over mud and vegetal debris, in forest habitat, in about 5.5 km upstream of the Ndzaa confluence with the Lukeni River, Kutu Territory, 0248035.200 S, 01811009.300 E, July 25, 2022, R. Monsembula; AMNH 278165 (2), 68.1–70.4 mm SL, in a tributary of Ndzaa River over mud and vegetal debris, in forest habitat, in about 6.3 km upstream of the Ndzaa confluence with the Lukeni River, Kutu Territory, 0254058.500 S, 01810059.900 E, July 24, 2022, R. Monsembula; AUM 86509 (1), 49.8 mm SL, collected with the holotype, M. Y. Modimo, August 10, 2021; MRAC 2024.008.P.0001 (1), 62.55 mm SL, Tshe River (tributary of Mfimi River) in about 5.9 km upstream of Nioki, over mud and vegetal debris, in grass habitat, Kutu Territory, 0244042.0100 S, 01744033.8700 E, August 2015, R. Monsembula et al.; ROM 112355 (2); 61.8–70.8 mm SL, same location as AMNH 278165, July 30, 2023, R. Monsembula; ZSM 48482 (1), 67.5 mm SL, main channel of the Ndzaa River over mud and vegetal debris, in forest habitat, in about 26.2 km upstream of the Ndzaa River confluence with the Lukeni River, Kutu Territory, 0258025.0800 S, 01807055.2600 E, July 25, 2018, fishermen. 3.4.3 | Additional non-type material AMNH 278167 (12); 38.2–57.4 mm SL, collected with ROM 112355; AMNH 278140 (1), 44.3 mm SL, main channel of the Ndzaa River over mud and plant debris, in forest habitat, 42.2 km upstream of the Ndzaa River confluence with the Lukeni River, Kutu Territory, Mai-Ndombe Province, D. R. Congo, 0306010.200 S, 01805014.100 E, M. Y. Modimo, August 13, 2021; AMNH 278169 (1), 34.1 mm SL, collected with the holotype, August 10, 2021, M. Y. Modimo. 3.4.4 | Diagnosis P. stiassnyae is distinguished from all congeners by having 28–29 vertebrae (vs. 33 or more). P. stiassnyae is also distinguished from all FIGURE 4 Principal component analysis (PCA) biplot of PC1 against PC2 for an analysis of five meristic counts for 55 specimens of sampled Parauchenoglanis taxa. Parauchenoglanis stiassnyae sp. nov.: ;Parauchenoglanis longiceps: ; Parauchenoglanis pantherinus: ; Parauchenoglanis guttatus:;Parauchenoglanis ubangensis: ; Parauchenoglanis cf. punctatus_L3: ; P. punctatus: ; Parauchenoglanis balayi: . 1232 MODIMO ET AL. FISH 10958649, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jfb.15885 by Capes, Wiley Online Library on [29/08/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License congeners by the possession of 13–14 pre-anal vertebrae (vs. 15 or more) except for Parauchenoglanis zebratus (14–17) and Parauchenoglanis ngamensis (13, holotype). The new species can further be distinguished from P. cf. punctatus_L3, P.balayi,P.longiceps,P.pantherinus, P.punctatus, and P.ubangensis by a narrower supraoccipital process– nuchal plate interdistance (1.4%–2.9% vs. >3% HL); from P. cf. punctatus_L3, P. guttatus,P. longiceps,P. pantherinus, and P. punctatus by a wider orbital HW (64.7%–76.2% vs. 54.9%–63.9% HL); from P. guttatus,P. longiceps, and P.ubangensis by a wider mouth (37.8%–50.8% vs. 25.9%–35.7% HL); from P. guttatus,P. punctatus,P. ubangensis, and P. zebratus by a wider premaxillary toothplate (12.9%– 18.6% vs. 6.6%–12.5% HL); from P.guttatus,P. longiceps, P. pantherinus, and P. zebratus by a wider head (HW: 70.1%–81.1% vs. 58.9%–69.3% HL); from P.balayi and P.pantherinus by a shorter dorsal-fin spine (10.8%–16% vs. 16.1%–18.8% SL); from P.guttatus and P.pantherinus by a smaller orbital diameter (9.5%–14.2% vs. 14.4%–16.9% HL) and a wider interpectoral distance (16.7%– 21.4% vs. 15.3%–16.6% SL); from P.balayi,P. ngamensis (holotype), and P.ubangensis by a shorter adipose-fin–caudal-fin interdistance (2.7%–5.2% vs. 6.2%–10.5% SL); and from P.balayi by a longer head (HL: 31.3%–35% vs. 28.1%–30.6% SL) and a narrower interorbital (IOD: 19.5%–27.1% vs. 27.3%–28% HL). 3.4.5 | Description Based on the holotype and 13 paratypes. General appearance as in Figures 2, 6, and 7(complete skeleton, external morphology, skull and FIGURE 5 Principal component analysis biplots for an analysis of 42 morphometric measurements for 55 specimens of sampled Parauchenoglanis taxa: (a) PC1 against PC2 and (b) PC1 against PC3. MODIMO ET AL.1233 FISH 10958649, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jfb.15885 by Capes, Wiley Online Library on [29/08/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License pectoral girdle, respectively), proportional measurements in Table 3, and meristic counts in Table 4. Small-bodied species (see Figure 6), maximum observed size 73.9 mm SL (88.3 mm TL; paratype: AMNH 278137), elongate (SL 4.4–6.4 times maximum body height [MxBH]), somewhat dorsoventrally depressed (MxBH 82.4%–109.5% interpectoral distance [IPcD]). Head depressed (head height [HH] 43.5%– 58.7% HL, orbital head height [OHH] 28.9%–43% HL, and OHH 38.5%–66.1% orbital head width [OHW]) with very depressed snout (snout height [SnH] 18.1%–26.9% HL) and large terminal mouth (37.8%–50.8% HL). Upper jaw prominent with moderately wide premaxillary toothplate (its width 12.9%–18.6% of the HL) comprising two tooth patches. Branchiostegal rays eight or nine. Maxillary and internal mandibular barbels shorter than HL (their length 51.2%– 97.8% HL and 37.2%–77.6% HL, respectively), not surpassing opercular opening. External mandibular barbel generally longer than head (their length 95.4%–147% HL), surpassing tip of adpressed pectoralfin spine in several specimens. Eye small but not reduced (orbit diameter 9.5%–14.2% HL) and dorsally positioned. Pectoral fin short (length 15.9%–23.1% SL), with seven or eight soft rays, inserted underneath head. Pectoral-fin spine also short (its length 12.9%–18.2% SL) with serrations on both sides. Pelvic fin short with six (one unbranched, five branched) soft rays and generally inserted just posterior to body midline (pre-anal length [PAnL]: 60.3%–72.6% SL). Dorsal fin short (its length 13.3%–18% SL) with two hard and seven soft rays. Dorsal-fin spine also short (10.8%–16% SL), slightly serrated anteriorly, lacking posterior serrations. Caudal fin rounded with 15 or 16 principal rays, including two unbranched. Dorsal surface of neurocranium ornamented with tubercles and with single oblong anterior fontanel. Nuchal plates similarly ornamented on dorsal surface, forming a narrow, anteriorly pointed triangle that surrounds anterior base of dorsal-fin spine (Figure 7a). Anterior bifurcation of mesethmoid narrow, not exceeding posterior mesethmoid width. Infraorbital series present as four narrow cylindrical ossifications. Orbit bordered dorsally by frontal, anteriorly by lateral ethmoid, posteriorly by sphenotic, and ventrally by third and fourth infraorbitals (see Figure 7a). Urohyal trifurcate posteriorly (see Figure 7b). Cleithrum with concave anterior margin and prominent FIGURE 6 Photographs of preserved (a) holotype (AMNH 278139 in lateral view) and (b–d) paratype (AMNH 278165, 68.1 mm standard length [SL], respectively, in dorsal, lateral, and ventral views). Scale bar: 1 cm. FIGURE 7 Computed tomography scan of the neurocranium of Parauchenoglanis stiassnyae, AMNH 269908, paratype, 68.28 mm standard length (SL), in (a) dorsal view with left side of suspensorium removed and (b) ventral view. ang, anguloarticular; apal, autopalatine; br, branchiostegal rays; ch-a, anterior ceratohyal; ch-p, posterior ceratohyal; cl, cleithrum; cp, cleithral process; cor, coracoid; den, dentary; fr, frontal; hh-v, ventral hypohyal; hm, hyomandibula; ioc, infraorbital canals; iop, interopercle; leth, lateral ethmoid; meth, mesethmoid; mpt, metapterygoid; mx, maxilla; nas, nasal; nu, nuchal plates; op, opercle; pmx, premaxilla; pop, preopercle; pp4, parapophysis of 4th vertebra; pp5, parapophysis of 5th vertebra; pop, preopercle; pt, pterotic; pto, posttemporal; q, quadrate; scl, supracleithrum; soc, supraoccipital; sph, sphenotic; uh, urohyal. 1234 MODIMO ET AL. FISH 10958649, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jfb.15885 by Capes, Wiley Online Library on [29/08/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License TABLE 3 Morphometric data for the holotype and 13 paratypes of Parauchenoglanis stiassnyae. Variables Holotype Holotype +paratypes Minimum Maximum Mean ± SD Total length (mm) 78.53 62.25 91.14 80.9 ± 7.6 Standard length (mm) 64.72 49.83 73.87 66.1 ± 6.2 Head length (mm) 22.57 16.89 25.14 22.14 ± 2.3 In percentage of standard length (SL) Pre-anal length (PAnL) 65.6 60.3 72.6 64 ± 3 Pre-pelvic length (PPvL) 54.9 50.7 56.4 53.8 ± 1.8 Pre-dorsal length (PDL) 42.7 38.7 43.1 41.2 ± 1.2 Prepectoral length (PPcL) 30.8 25.5 33.5 30.2 ± 1.8 Adipose-fin length (AdFL) 31.4 30 42.1 35.7 ± 3.4 Interdorsal–adipose distance (IDAdD) 5.6 1.8 9.8 5.1 ± 2.4 Dorsal-fin base length (DFBL) 17 13.7 20 16.6 ± 1.9 Dorsal-fin length (DFL) 15.2 13.3 18 15.7 ± 1.4 Dorsal-fin spine length (DSL) 14.5 10.8 16 14 ± 1.5 Pectoral-fin length (PcFL) 18.4 15.9 23.1 19.3 ± 1.9 Pectoral-fin spine length (PcSL) 18.1 12.9 18.2 16.3 ± 1.6 Pelvic-fin length (PvFL) 17.9 15.7 21.1 17.6 ± 1.5 Anal-fin length (AFL) 14.1 12.3 18.1 15 ± 1.6 Interpectoral distance (IPcD) 20 16.7 21.4 19.4 ± 1.4 Interpelvic distance (IPvD) 7.5 5.9 7.8 7 ± 0.6 Maximum body height (MxBH) 21.2 15.7 22.7 19.4 ± 2 Pelvic body height (PvBH) 18.2 14.5 20.7 17.4 ± 2 Minimum caudal peduncle height (MnCPH) 15 12.4 15.7 14.3 ± 0.9 Maximum caudal peduncle height (MxCPH) 15.6 13.2 17.4 14.9 ± 1.1 Adipose-fin–caudal-fin interdistance (AdCID) 4 2.7 5.1 4 ± 0.8 Anal-fin–caudal-fin interdistance (AnCID) 14.2 13 16.6 14.5 ± 1.1 Adipose-fin height (AdFH) 7.5 4.7 9.6 6.6 ± 1.3 Head length (HL) 34.9 31.3 35 33.5 ± 1.3 In percentage of head length (HL) Postorbital head length (POL) 41.6 41.3 59.2 45.5 ± 4.3 Preorbital head length (PrOL) 52.1 47.8 52.1 49.9 ± 1.6 Head width (HW) 74.1 70.1 81.1 74.9 ± 3.4 Orbital head width (OHW) 70 64.7 76.2 69.6 ± 3.6 Head height (HH) 55.5 43.5 58.7 54.3 ± 4.4 Orbital head height (OHH) 40.5 29 43.2 38.4 ± 3.6 Snout height (SnH) 26.8 18.1 26.9 23.1 ± 2.3 Maxillary barbel length (MxBL) 68.9 51.2 97.8 69.4 ± 12.4 External mandibular barbel length (EMdBL) 107.5 95.4 147 115.8 ± 14.9 Internal mandibular barbel length (IMdBL) 46.5 37.2 77.6 56.3 ± 10.7 Mandibular barbels interdistance (MdBID) 12.3 8.4 17.4 13.5 ± 2.6 Interorbital distance (IOD) 24.1 19.5 27.1 23.3 ± 2.2 Anterior nostrils interdistance (ANID) 24 15.7 24.6 20.7 ± 3 Posterior nostril interdistance (PNID) 12.9 10.1 14.5 12.3 ± 1.4 Supraoccipital process–nuchal plate interdistance (SPNPID) 2.2 1.4 2.9 2.1 ± 0.4 Supraoccipital process–dorsal-fin interdistance (SPDFID) 21 14.9 21 17.4 ± 1.8 Prehyoid length (PHL) 20.3 18.7 23.4 21.2 ± 1.3 (Continues) MODIMO ET AL.1235 FISH 10958649, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jfb.15885 by Capes, Wiley Online Library on [29/08/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License