Laogarra, Kottelat, 2025

Kottelat, Maurice, 2025, Laogarra heokheei, a new genus and species of ' garra' from north-eastern Laos (Teleostei: Cyprinidae), Zootaxa 5666 (3), pp. 389-407 : 396-399

Laogarra and the genera of Semilabeoina

It is noteworthy that the elements of the disc are not homologous in the different garras. This was to be expected as some garras are presently placed in the subtribe Garraina and others in the subtribe Semilabeoina . Complete developmental series are missing for almost all species, but Kottelat (2020: 166, figs. 15–16) showed, however, that in Garra (subtribe Garraina ) the disc forms from tissues of the central gular area, while in Ageneiogarra (subtribe Semilabeoina ) it forms by the median fusion of tissues from the sides of the gular area, being notched in juveniles. The single available juvenile of L. heokheei suggests that the situation in Laogarra could be identical to that in Ageneiogarra ( Fig. 2c ). In some adult specimens of Ageneiogarra the posterior edge of the labrum is slightly concave. A short notch is also present in Supradiscus and Vinagarra (which too, are in Semilabeoina ), which suggests the hypothesis that all garras in the Semilabeoina share a similar ontogeny.

The shape of the mouth, lips, disc and gular area shows a much greater diversity in Semilabeoina than in Garraina . Contrary to the situation in the Garraina of Southwest Asia, authors have recognised several molecular lineages and morphological structures supporting the recognition of different genera among semilabeonine garras, but not always in unambiguous ways, in part because of the problems in naming the different features of the disc, lips and gular area. The naming issue was compounded by the use (at least in English) of a complex terminology inimical to convenient comprehension. Worse, different schools have used different terminologies (e.g., Zhang, 2005; Zhang et al., 2002: 209; Zhang & Zhou, 2012: 18; Nebeshwar & Vishwanath, 2017; Kullander & Fang, 2004: 258). Further, part of this terminology was anatomically incorrect, starting with the denomination of the disc, which is neither mental, nor mandibular, nor oral, but mainly composed of tissues from the gular region (as postulated as early as Hora, 1921b; Rauther, 1928; Minzenmay, 1933). Kottelat (2020) attempted to establish some standardization by using a much simpler terminology, by replacing the confusing multi-word names of the components (e.g., anteromedian fold, anterolateral lobe, lateroposterior flap, central callous pad, etc.) by a single name for each structure. As mentioned above, he also showed that in at least two lineages the ontogeny of the disc differs: the elements are apparently not all homologous between and within Garraina and Semilabeoina .

Recently, Li et al. (2024) described differences in the components of the disc, but still using the 'old' complex terminology, resulting in a text difficult to follow and understand. Besides, the text in their table 1 cannot be deciphered because of the microscopic font used by the journal, in addition to which it is not possible to extract information from their figure 4, in which the symbols cannot be distinguished, and it is not possible to discern differences between colours. Nevertheless, their conclusions seem to be justified and I largely follow them here, but adopting the terminology of Kottelat (2020).

As recognized by Li et al. (2024), the Southeast and East Asian garras belong to 10 genera. Eight belong to the subtribe Semilabeoina : Ageneiogarra (known from middle and upper Salween and Mekong drainages, northern Vietnam and southern China), Discocheilus (Pearl River drainage), Discogobio (Red, Pearl and Yangtze drainages), Placocheilus (two species P. caudofasciatus and P. robustus , from the upper Black and Red River drainages), Pseudoplacocheilus (a single species, P. cryptonemus , in the upper Salween drainage), Sinigarra (a single species, S. napoensis , from the Pearl River drainage), Supradiscus (two species in the upper Red and Pearl River drainages) and Vinigarra (four nominal species in the Black River drainage of China and Vietnam). The remaining two genera belong to the subtribe Garraina : Garra (all river drainages from East Asia to Africa), and Ceratogarra (Mekong, Chao Phraya and Mae Khlong drainages, Malay Peninsula).

Some structures of the disc are possibly only analogous, not homologous. See below for comments on Supradiscus, Discocheilus , Discogobio . A number of other genera have modified lower lip and gular structures but have not been treated as garras. Unfortunately, no material is available for examination of the genera Cophecheilus , Hongshuia , Parasinilabeo and Pseudogyrinocheilus . Published illustrations (e.g. Zhang, Yue & Chen, in Yue, 2000; Gan et al., 2017) suggest that, although they have no disc as such, they could have oral and gular structures with similarities to structures in the same position or location in garras. Detailed descriptions of the morphology of the disc, as well as its ontogeny, are needed for most species of semilabeonine garras.

According to Li et al. (2024: [9]), Supradiscus includes two species known from upper tributaries of the Red River ( S. micropulvinus ; Zhou et al., 2005) and Pearl River drainages ( S. incisorbis ; Zheng et al., 2016). They are distinguished in having the "central callous pad [pulvinus] composed of multiple stacked parts", the "antero-median fold [torus] with a pocket opening backwards"; the posterior margin [labrum] of the "lower lip" [disc] with a median notch.

In the two examined specimens of Supradiscus micropulvinus (CMK 18394, 90.8 and 92.6 mm SL; Fig. 3 ) the organisation of the disc is in fact more complex. The posterior edge of the labrum has a median notch extending across about ¼ of its width. The torus is wide, with an anterior fold along its whole length (subtoral fold?). The torus is entirely covered by papillae, those along the posterior edge being more elongated than the others, and somewhat digitiform. The toral groove is very deep and extends slightly under the pulvinus. At both extremities of the groove, intercalated between the torus and the pulvinus, there are flaps of tissue connected to the torus, ending with a single or a few digitiform papillae, identical to those of the torus, with some appearing as if detached from the torus ( Fig. 3b ). The pulvinus is made up of an anterior element covered by low papillae and margined posteriorly by a medially interrupted groove. The area between this groove and the postpulvinal groove is smooth and appears continuous with the torus.

Discocheilus is distinguished from all semilabeonine garras by the wide torus, margined by a deep toral groove forming a pocket under the torus. It is also the only genus with the rostral cap not crenulated. The disc has some superficial similarity with that of Garra variabilis from Orontes, Tigris and Euphrates (see Kottelat, 2020: fig. 25).

Discogobio has a horseshoe-shaped torus that completely circles the pulvinus and is in direct contact with the labrum; all elements, including the pulvinus, are densely papillated ( Fig. 9a ). The torus and pulvinus are clearly narrower than the mouth gape in the species I examined ( D. tetrabarbatus , Fig. 9a; D . yunnanensis) and those figured in the literature (e.g. Zhang, Yue & Chen, in Yue, 2000).

Vinagarra too, has a horseshoe-shaped torus not in contact with the labrum ( Li et al., 2008: fig. 2a; indistinct on pictures in Endruweit, 2014: figs. 7–8). In addition, there is a notch in the labrum reaching forward to the pulvinus. Vinagarra is geographically quite close to Laogarra . It includes four nominal species, all described from the Black River watershed: V. findolabium from China with a usable description ( Li et al., 2008) and three others from Lai Chau Province in Vietnam: V. laichowensis (the type species), V. elongata and V. tamduongensis . Li et al. (2024: [11]) had some doubts whether there exist three species of the same genus in such a small area and I concur with their doubts.

Vinagarra was created by Nguyen & Bui (2009) to include V. laichowensis . The diagnosis of the new genus is somewhat obscure and the figures are uninformative: no detail of the disc can be discerned. The original description of V. laichowensis is hardly more informative (paper reproduced in Kottelat, 2001 and translated in Nguyen & Doan (2007); the original description did not include the figure, which was later printed in Nguyen & Doan (2007: 81, fig. 9) and in Nguyen & Ngo (2001: 543, fig. 277)). The description provides no information on the disc but the figure shows a notched labrum and this notch is mentioned in Mai (1978: 52, 54; translated in Kottelat, 2001: 80).

Li et al. (2008) described Garra findolabium . Based on photographs of G. laichowensis , they distinguished the two species by the absence of barbels in G. findolabium (vs. 2 pairs in G. laichowensis ), caudal fin with a dark patch extending to 50 % of the length of the fin (vs. no dark patch), "posterior edge of [...] disc with a median fissure and dividing it into two lamellae" (vs. "with a median notch and not divide [...] into two lamellae") and various small differences in morphometric and meristic characters). Subsequently, the two species were placed in Vinagarra ( Endruweit, 2014) . Li et al. (2024: [11]) treated V. laichowensis as congeneric with V. findolabium .

Nguyen & Bui (2010) described V. elongata and V. tamduongensis , some specimens of which were collected together at the same locality, and which Li et al. (2024: [11]) could not place in their system and even doubted the validity of. The pictures are uninformative (completely distorted), in addition to which the disc is not illustrated. Endruweit (2014) figured some types of the three nominal species of Vinagarra in Vietnam. He described the disc of V. laichowensis as "The mandibular possesses a horseshoe-shaped distal formation anteriorly and is laterally separated by a shallow depression from the central mental disk [...]. The posterior mandibular margin is deeply notched creating two lateroposterior flaps [...]." His figure 6 apparently shows the same disc construction as V. findolabium figured by Li et al. (2008: fig. 2a); however, Li et al. described a 'fissure' which is more appropriate to describe the condition illustrated in their figure of V. findolabium , than a notch as figured for V. laichowensis . Endruweit (2014: 148, figs. 7–8) discussed and figured a paratype of V. elongata and the holotype of V. tamduongensis , which he considered a valid species of Vinagarra . This should be confirmed by the examination of fresh, well preserved topotypes.

Placogobio too(also described from the Red River drainage in Vietnam)seems to be a garra( Nguyen & Ngo, 2001: 555). The poor photograph in the original description of the type species ( P. nahangensis ) seems to show a notched labrum. There is no postpulvinal groove. Two species were included, P. nahangensis and P. bacmeensis . Placogobio nahangensis reportedly has 46–48 lateral-line scales (including scales on caudal fin) while P. bacmeensis has 37– 38, which excludes them from possibly being synonyms of Laogarra heokheei . Garroides (sometimes misspelled Garraoides), also described from northern Vietnam, with two included species ( G. binhluensis , G. banthangenis ) is another genus from the Red River drainage ( Nguyen & Vu, 2014). I cannot extract usable information from the original description and the poor figures, except that the general body shape appears unlike that of Laogarra , with a deeper body and a higher cheek; elements of the disc cannot be discerned.

Li et al. (2024) recognised two species in Placocheilus . They diagnosed the genus, among others, by having the torus not connected to the labellum, by the absence of a groove between the pulvinus and the labrum, and by a colour pattern including a black longitudinal stripe on the dorsal fin in addition to a black semicircular band on each lobe of the caudal fin. Two species are now placed in Placocheilus : the type species P. caudofasciatus from the Black River drainage, and P. robustus from the Red River drainage.

Placocheilus bibarbatus and P. imbarbatus , described together from the same locality in the Black River drainage, are poorly described and with uninformative illustrations ( Nguyen & Ngo, 2001: 552–555). Li et al. (2024: [10]) could not even decide their generic allocation. It is noteworthy, however, that this locality is close to those of Vinigarra tamduongensis and V. elongata .

Pseudoplacocheilus includes a single species, Ps. cryptonemus , from the middle Salween drainage. It was originally placed in Placocheilus but moved to a distinct genus by Li et al. (2024). Garra dulongensis , from the upper Irrawaddy drainage, was also originally placed in Placocheilus and transferred to Garra by Kottelat (2020).

Ageneiogarra presently includes about six valid species from northern Indochina and South China. The genus was revalidated by Yang et al. (2012: 370) analysis. These are large garras, relatively slender, with a wide depressed head, without large tubercles on the snout, and without barbels. They have 43–50 + 2–3 lateral line scales; the snout has neither a transverse lobe nor conspicuous tubercles (though minute horny projections may be present); the anus is positioned very close to the pelvic-fin base (about at mid-length of the posterior ray); the torus is separated from the pulvinus by a deep groove extending laterally and posteriorly along the sides of the pulvinus as a narrow strip of tissue, separated from the labrum by a deep groove ( Fig. 9b ; Kottelat, 2020: fig. 17). The narrow strip from the torus along the pulvinus is also observed in Laogarra .

Sinigarra is distinguished by the deeply notched labrum, the absence of the torus and the absence of papillae on the whole disc. It has been compared with other garras but it is not clear how closely it is related to them. I have not been able to examine material, but the published illustrations of the type species, S. napoensis ( Zhang & Zhou, 2012: fig. 3; Gan et al., 2017: 206; Li et al., 2024: fig. 1a), seem to show the pulvinus and labrum as thicker than in the other garras, in addition to which the homology of these elements needs further examination (for example the description of an ontogenetic series to eliminate the hypothesis that the disc is possibly a modified lower lip rather than gular tissue).

Figures

FIGURE 2. Laogarra heokheei; mouth and disc; a, MHNG 2797.028, holotype, 62.7 mm SL (arrows: upper lip); b, CMK 28985, 65.5 mm SL (arrow: postpulvinal groove); c, CMK 28985, 23.0 mm SL.

FIGURE 3. Supradiscus micropulvinus, CMK 18394, 92.6 mm SL; a, mouth and gular region; b, close-up of corner of mouth.

FIGURE 9. Mouth and gular region of: a, Discogobio tetrabarbatus, CMK 21917, 100.4 mm SL; b, Ageneiogarra theunensis, CMK 22502, 154 mm SL. Yellow arrow: lateral extension of torus; black arrow: anterior fold of torus. From Kottelat, 2020: 162.