Hypostomus khimaera Tencatt, Zawadzki & Fröehlich, 2014 (Siluriformes, Loricariidae) in the upper rio Paraná basin, Brazil: first record and comments on its occurrence

. Hypostomus khimaera Tencatt, Zawadzki & Fröehlich, 2014 is a species native to the rio Paraguay basin, but it is recorded here from the upper rio Paraná basin for the first time. Our new record was made in the córrego Mimoso, a tributary of the rio Anhanduí, approximately 180 km from the closest previously reported locality. We are unsure whether the occurrence of this species in the upper rio Paraná basin is a consequence of natural phenomena, such as headwater connection during floods or stream capture, or an artificial introduction. Together with the invasive H. cochliodon Kner, 1854, H. khimaera is the second species of the H. cochliodon group recorded in the upper rio Paraná basin. Here, we provide an updated diagnosis of H. khimaera and highlight the differences in opercle exposition as the easiest way to differenti‑ ate it from the syntopic and morphologically similar H. cochliodon .


INTRODUCTION
Widely distributed and currently comprising more than 150 valid species, Hypostomus Lacépède, 1803 is one of the most species-rich genera of Neotropical freshwater fishes (Silva et al. 2016;Queiroz et al. 2020;Fricke et al. 2024).The Hypostomus cochliodon group (sensu Armbruster 2003) currently harbors 22 species, which can be distinguished from the remaining congeners by having teeth with mesial cusp ranging from shovel-shaped or spatulate to spoon-shaped (except for young H. khimaera, which can present villiform teeth; Tencatt et al. 2014), loss of a notch between the metapterygoid and the hyomandibula, and strongly angled dentaries, with up to about 80° between dentaries (Armbruster 2003(Armbruster , 2004;;Tencatt et al. 2014;Oliveira et al. 2020).
In this study, we present the first record of H. khimaera in the upper rio Paraná basin.This species was originally described from tributaries of the upper rio Paraguay and until now it had not been recorded outside this river basin.We discuss the occurrence of this species in the upper rio Paraná basin and provide diagnostic characters to distinguish it from other H. cochliodon group species.

METHODS
One individual of Hypostomus khimaera was collected on 3 October 2023 in the córrego Mimoso, near the mouth on rio Anhanduí, tributary of the rio Pardo, upper rio Paraná basin, Ribas do Rio Pardo municipality, Mato Grosso do Sul state, Brazil.The sampling was done as part of an inventory for a new conservation unit in the region of the Pardo and Anhanduí rivers.The specimen was collected in an artisanal trawl measuring 1.5 × 1 m.It was euthanized by eugenol overdose in accordance with Lucena et al. (2013).After death, the specimen was fixed in a 4% formaldehyde solution, where it was kept for 48 h and then transferred to 70% ethanol.This voucher specimen is deposited in the Coleção Ictiológica de Três Lagoas (CITL).The collection was authorized by the Sistema Nacional de Informação da Biodiversidade of Brazil (SISBio permit 81153/6 -2023) Measurements were made with digital calipers having a precision of 0.1 mm.Measurements follow Boeseman (1968), with the modifications adopted by Tencatt et al. (2014).To access previously recorded distributional data on H. khimaera, we searched for lots with geographic coordinates in collection databases hosted on the SpeciesLink website (CRIA 2024).The resulting data were plotted in a map using QGIS v. 3.30.2.Identification.Hypostomus khimaera can be distinguished from congeners, except the species in the H. cochliodon group, by the absence of a notch between metapterygoid and hyomandibula and by the strongly angled dentaries, with up to about 80° between dentaries (vs.more obtusely angled dentaries, with clearly more than 80° between dentaries) (Tencatt et al. 2014).It can be distinguished from other species of the Hypostomus cochliodon group, except H. basilisko and H. soniae Hollanda Carvalho & Weber, 2005 by the presence of a dark stripe along the midline on flank (vs.absence of midline stripe).Hypostomus khimaera differs from H. basilisko and H. soniae by the presence of black spots on the body   and/or fins and moderately developed keels and odontodes (vs.black spots absent from body and fin and strongly developed keels and numerous well-developed odontodes).Hypostomus khimaera is additionally distinguished from H. cochliodon by the greater number of teeth (12-27 vs. 7-9), by having an externalized opercle with the exposed region easily visible (vs.almost entirely internalized opercle with exposed region not easily visible), weakly developed papillae on both upper and lower jaws, not forming series (vs.papillae well developed, forming one series on the inner face of both upper and lower jaw), moderately developed keels, with moderately developed odontodes (vs.weakly developed keels and odontodes) (Tencatt et al. 2014;Figures 3, 4).The specimen collected in the córrego Mimoso, upper rio Paraná basin, presents the combination of characters described for H. khimaera: acutely angled dentary (less than 80°), a dark stripe along flank midline, black spots on body and fins, 14 teeth on premaxilla, and an externalized opercle with exposed region easily visible.

DISCUSSION
Hypostomus khimaera was described by Tencatt et al. (2014) from the córrego Salobo (the type locality), a tributary of the rio Paraguay in Mato Grosso, and was considered to be endemic to the rio Paraguay basin.
It occurs in a variety of habitats, such as rapids and marginal habitats with sandy substrate, and is most associated with small streams, although occasionally found in the main channel of larger rivers (Tencatt et al. 2022).Its previously known distribution extends between the upper rio Paraguay and rio Cuiabá basins to the north and the rio Aquidauana, to the south.However, it is mostly confined to the upper portions of the rio Paraguay basin, in the plateau area surrounding the Pantanal (Tencatt et al. 2014: 594, fig.9), which is demonstrated by the 85 lots of H. khimaera in the SpeciesLink database (CRIA 2024) (Figure 2).In the Alternatively, our new record of H. khimaera is from a tributary of the lower rio Anhanduí, which is more than 180 km away from the divide between the Pardo and Aquidauana river basins, and it is closer to the rio Paraná.This seems to weaken hypothesis that this occurrence is due to natural causes.Our second hypothesis to explain the origin of H. khimaera in the upper rio Paraná basin is that it is the result of a biological invasion in the Itaipu reservoir.The major invasion route into the upper rio Paraná basin is through the Itaipu Reservoir, whose filling in 1982 caused the permanent submersion of the Sete Quedas Falls, which suppressed the ancient biogeographic barrier separated the upper and lower basins of the rio Paraná.Even though this filling took place only once, the opening of the Canal da Piracema in 2002 provided a continuing pathway for fish dispersal between the two ecoregions (Júlio Junior et al. 2009).The Itaipu Reservoir and the Canal da Piracema together enabled a massive biological invasion event in the upper rio Paraná basin, which included the arrival H. cochliodon (Júlio Junior et al. 2009;Ota et al. 2018).
However, by suggesting that the presence of H. khimaera in the upper rio Paraná basin is the result of a biological invasion, we must make two important assumptions.First, we assume the species exists in the lower rio Paraná basin, and second, we assume that Nupélia researchers did not detect this species in the Itaipu Reservoir and on the upper rio Paraná floodplain despite decades of regular monitoring of fish in the region.The presence of H. khimaera in the lower rio Paraná basin has not been confirmed to date, even though there are no physical barriers between the Paraguay and lower Paraná river basins.The non-detection of this species in the lower rio Paraná basin may be a consequence of its absence in the ecoregion, sampling gaps, incorrect identifications, and/or possibly low abundance as a response to unknown ecological conditions.
Finally, there is the third possibility that the occurrence of H. khimaera in the upper rio Paraná basin is the result of intentional releases by humans, since Hypostomus cochliodon group species are appreciated by hobby aquarists (Novák et al. 2022).Other fish species are thought to have been introduced in the upper rio Paraná basin in this way (Garcia et al. 2018), which lends support to this hypothesis.However plausible, with only a single specimen collected from an area that is distant from urban centers and with no evidence pointing to a specific person or group who might be responsible for introducing this species in the upper Paraná basin, this third hypothesis lacks support for now.
Finding the hypothesis that fits the data best will demand further fieldwork in the rio Pardo basin, sampling tissues for DNA analysis, and thoroughly reevaluating previously collected specimens.Specimens of H. khimaera from other localities in the ecoregion may be available in fish collections, perhaps misidentified as H. cochliodon, and their eventual discovery could reveal clues to the pathway through which H. khimaera has expanded its geographic range.Our first hypothesis would be corroborated by the discovery of additional specimens from the upper reaches of the rio Pardo basin and which share haplotypes with specimens from the rio Aquidauana basin.Our second hypothesis would be corroborated by the confirmation that H. khimaera is native to the lower rio Paraná basin, as well as by the finding of specimens collected in the Itaipu Reservoir or in the upper rio Paraná Floodplain within a time frame that makes possible the dispersion of this species to rio Pardo basin.If neither of the first two hypotheses finds support with new data, our third hypothesis will perhaps be the most likely.

Figure 2 .
Figure 2. Distribution of Hypostomus khi maera.Black circles = previously known records in the SpeciesLink database (CRIA 2024); yellow star = new record in upper rio Paraná basin.

Figure 4 .
Figure 4.Additional diagnostic characters of Hypostomus khimaera collected in the upper rio Paraná basin (CITL 1015, 75.9 mm SL). A. Moderately developed keels (white arrows), with moderately developed odontodes (dorsal view).B. Mouth and teeth, with papillae not forming series in the inner face of upper and lower jaws, showing the detail of the teeth crowns (outlined in black).