Revalidation and redescription of Steindachnerina nigrotaenia and redescription of S . insculpta ( Characiformes : Curimatidae )

Steindachnerina nigrotaenia is resurrected from the synonym of S. brevipinna and considered a valid species. The previous designation of the lectotype of S. nigrotaenia is considered invalid and a new lectotype is designated herein. Steindachnerina nigrotaenia and S. insculpta are redescribed based on type specimens and on additional material from the rio Paraguai and the upper rio Paraná basins, respectively. The two species can be separated by the number of scales of the lateral line and of the transverse series and by phylogenetic analyses of molecular data.


Introduction
The characiform family Curimatidae includes eight genera, two of which concentrate more than half of the species in the family, Cyphocharax Fowler, 1906, andSteindachnerina Fowler, 1906.Of the 24 valid species of Steindachnerina, four species occur in southern South America in the La Plata system: S. biornata (Braga & Azpelicueta, 1987), S. brevipinna (Eigenmann & Eigenmann, 1889), S. conspersa (Holmberg, 1891), andS. insculpta (Fernández-Yépez, 1948).Both S. biornata and S. conspersa appear at the base of the phylogeny of the genus with significant morphological differences when compared to the other two species, whereas S. brevipinna and S. insculpta has similar morphologies and are close related species (Vari, 1991;Melo et al., 2018).Vari (1991) provided a valuable progress in the systematics of the genus by analyzing many type specimens that enable him to complete the taxonomic revision and phylogeny of Steindachnerina.However, there is a putative conflict in the number of valid species in the rio Paraguai (two, S. conspersa and S. brevipinna) and the number of observed morphotypes in that region (three).The available name for the third species is Steindachnerina nigrotaenia (Boulenger, 1902), currently placed in synonym of S. brevipinna by Vari (1991).The history behind the types of Curimatus nigrotaenia explains the reasons of the mistaken synonymy.
Early in the 20 th century, the Director of the Museo Civico di Storia Naturale of Genova (MSNG) entrusted to Dr. George A. Boulenger of the British Museum (Natural History) (BMNH) a collection of fishes and reptiles collected by Dr. F. Silvestri in 1900 in Brazil, Paraguay and Argentina.One among several fish species described by Boulenger (1902) based on that material was Curimatus nigrotaenia Boulenger, 1902, a curimatid from the rio Coxipó, state of Mato Grosso, Brazil, upper Paraguai basin.The species was mainly characterized by the possession of 43 or 44 scales in the lateral line, 7.5 to 8.5 series of scales between the lateral line and dorsal-fin origin and six series of scales between the lateral line and pelvic-fin origin, supplemented by a dark, longitudinal stripe along midlateral surface of the body extending onto middle rays of the caudal fin.Boulenger did not mention any dark spot on the basal portion of the middle 2 e180076 [2] dorsal-fin rays, a distinct feature present in many species of curimatid fishes.These characters accurately discriminated C. nigrotaenia from other curimatids known at that time from the rio Paraguai basin.Silvestri's collection was partially returned to the MSNG, but, as reported by Boulenger (1902: 284), some of the specimens were retained at the BMNH.However, Boulenger did not indicate how many individuals of each species were retained at the BMNH versus returned to the MSNG.Vari (1991) redefined Steindachnerina, to incorporate various species previously assigned to Curimatus Oken, 1817 (= Curimata Bosc, 1817), including Curimatus nigrotaenia.After examining five specimens at the BMNH labeled as syntypes of C. nigrotaenia under catalog number BMNH 1902.2.10:30, Vari (1991:100) observed that the features of these specimens corresponded to those of Steindachnerina brevipinna described from Río de la Plata, Rosario, Argentina.The putative type specimens of C. nigrotaenia had 34 to 35 scales in the lateral line series from the supracleithrum to the hypural joint, 5.5 scales between the lateral line and dorsal-fin origin, 4.5 scales between the lateral line and anal-fin origin, a prominent midlateral dark stripe and a dark spot on the basal portion of the middle rays of the dorsal fin.As a result, Vari (1991:100) treated C. nigrotaenia as a junior synonym of S. brevipinna, designated the largest specimen as the lectotype of C. nigrotaenia, and transferred the other four putative syntypes to BMNH 1989.2.2:3-6 as paralectotypes.
However, the data presented by Boulenger (1902) in the original description disagreed substantially from the data collected by Vari (1991) when he examined the specimens.At that time, Vari (1991:100) considered it impossible to determine whether the original data published by Boulenger contained typographical errors or Boulenger's counts were incorrect.In addition, he assumed that Boulenger "failed to describe the prominent dark spot on the basal portion of the middle rays of the dorsal fin in the relatively small-sized syntypes" he examined.However, he did not discuss the possibility of a curatorial error at BMNH, resulting in the wrong specimens being placed in the syntype jar for Curimatus nigrotaenia.
After the revision of Vari (1991), one of us (HAB) suspected the occurrence of an error due to several morphological differences observed in the two morphotypes collected in the rio Paraguai.Despite Vari's conclusions, Britski (1996) maintained Steindachnerina nigrotaenia as a valid species in a brief communication in a regional meeting for biologists, and in the two volumes of the manual of fishes from the Pantanal Matogrossense (Britski et al., 1999;2007), suspecting that the specimens examined by Vari (1991) at the BMNH were not the syntypes of the species, but actually specimens of S. brevipinna incorrectly labeled as syntypes of S. nigrotaenia.Additionally, the ICZN (Case 3307) received in 2004 a proposal of conservation of usage of the name Curimatus nigrotaenia (= S. nigrotaenia) by designation of a new lectotype, proposed by Britski and Vari, without any resolution.
In this context, we present complete redescriptions of the two species taking into account that specimens of Steindachnerina nigrotaenia and S. insculpta were mixed in the redescription under the name of S. insculpta by Vari (1991), and that the description of Curimatus nigrotaenia by Boulenger (1902), the redescriptions by Britski et al. (1999;2007), as well as the original description of Cruxentina insculpta by Fernández-Yépez (1948) are very abridged and/or incomplete.In summary, the purpose of this paper is to (1) revalidate Steindachnerina nigrotaenia (Boulenger, 1902), (2) designate a new lectotype for the species, and (3) redescribe both S. nigrotaenia and S. insculpta based on new morphological and molecular data from both species.

Morphology. To resolve the curatorial issues involving
Steindachnerina nigrotaenia, we examined the type specimens deposited at the Natural History Museum of London (BMNH) and at the Museo di Storia Naturale of Genova (MSNG), and additional material from recent field collections of S. brevipinna, S. insculpta and S. nigrotaenia.Measurements were made with digital calipers to the nearest 0.1 mm.Counts and measurements were taken on the left side of specimens whenever possible.Lateral line scales were counted from the supracleithrum to caudal fin, including those overlying the hypural plate and extending to the base of median caudal-fin rays, which we also report separately.Counts of total vertebrae were usually taken from radiographs, the fused PU1+U1 was considered a single bone, and the vertebrae incorporated into the Weberian apparatus were counted as four elements.
Subunits of the head are presented as proportions of head length (HL).Head length itself and measurements of body parts are given as proportions of standard length (SL).Greatest body depth was taken at the dorsal-fin origin.In the counts of median-pectoral and pelvic fins, the unbranched rays are indicated by lower case roman numerals, and the branched-fin rays as Arabic numerals.Meristic data are given in the description with parentheses indicating the frequency of each count and asterisks indicating counts of holotype or lectotype specimens.

e180076[3]
Molecular analysis.Nineteen specimens of Steindachnerina were included in the molecular analysis: five specimens of S. brevipinna from rio Paraná, five specimens of S. insculpta from the main rio Paraná above the Itaipu dam and rio Tietê, eight specimens of S. nigrotaenia from the upper rio Paraguai, and one specimen of S. elegans from the rio São Francisco to root the tree.Sequences of ten out of nineteen specimens were obtained from Genbank (Tab.1).
Tab. 1. Taxon, voucher, locality information and Genbank accession numbers of specimens of Steindachnerina analyzed in the molecular approach.For the three newly sequenced individuals, genomic DNA was extracted from muscle tissues preserved in 95% ethanol with a DNeasy Tissue kit (Qiagen Inc.; http://www.qiagen.com)according to the manufacturer's instructions.Partial sequences of the mitochondrial gene cytochrome oxidase c subunit I were obtained by polymerase chain reaction (PCR) using the primers described by Melo et al. (2011).Total volume included 12.5 μl with 9.075 μl of double-distilled water, 1.25 μl 5x buffer, 0.375 μl MgCl2 (50mM), 0.25 μl dNTP mix, 0.25 μl of each primer at 10 μM, 0.05 μl Platinum Taq DNA polymerase enzyme (5 units/μl, Invitrogen; www.invitrogen.com) and 1.0 μl genomic DNA (10-50 ng).The PCR program consisted of an initial denaturation (4 min at 95º C) followed by 30 cycles of chain denaturation (30 s at 95ºC), primer hybridization (30-60 s at 52ºC), and nucleotide extension (45 s at 72ºC).Amplicons were visualized through 1% agarose gel, and used to perform the sequencing reaction using dye terminators (BigDye™ Terminator v 3.1 Cycle Sequencing Ready Reaction Kit, Applied Biosystems), and purified through ethanol precipitation.Samples were then loaded onto an automatic sequencer ABI 3130-Genetic Analyzer (Applied Biosystems) at the São Paulo State University, Botucatu, Brazil.

Results
The analyses of morphological and molecular data strongly support the recognition of three species of the same clade in the La Plata basin: Steindachnerina brevipinna (rio Paraná and rio Paraguai), S. insculpta (upper rio Paraná) and S. nigrotaenia (upper rio Paraguai), supplemented by S. conspersa and S. biornata.The synonymization of S. nigrotaenia was caused by previous curatorial mistakes that guided Vari (1991) to recognize only one species.

e180076[4]
After finding the syntype of Curimatus nigrotaenia retained at the BMNH by Boulenger (1902) who assigned it the term "type", the following notes in literis were received from James Maclaine, curator of the British Museum of Natural History, explaining the erroneous changes that took place with materials in that fish collection: "A) BMNH 1902.2.10.30Vari (1991:87) about syntypes of Curimatus morawhannae.
Analysis of the three syntypes of Curimatus nigrotaenia deposited at the MSNG with the BMNH syntypes has confirmed that the material now in London was incorrectly curated.The still well-preserved syntypes at MSNG entirely fit Boulenger's description and differ from the BMNH "syntypes" studied by Vari (1991).Furthermore, we have successfully found the "type" of C. nigrotaenia, which was wrongly deposited in a jar labeled Curimatus morawhannae Eigenmann, 1912, under the number BMNH 1917.11.2:7 (see paragraph above).Data taken from this "type" agree entirely with the original description of C. nigrotaenia.The specimen of C. nigrotaenia registered under the number BMNH 1902.2.10.30by Boulenger also had already been cataloged in Günther's Catalogue of Fishes (museum catalog, vol.5, pp.288a) of the BMNH.That finding also revealed that Boulenger retained only one specimen of C. nigrotaenia (labeled as "type") at the BMNH, and not five syntypes as mentioned by Vari (1991).The number of syntypes sent back to MSNG was not specified by Boulenger, but by Tortonese (1961: 184), in his catalogue of types deposited at the MSNG mentioning three syntype specimens.Overall, the clearly morphological differences between Steindachnerina brevipinna and the syntypes of C. nigrotaenia at MSNG and BMNH illustrate that the synonymization was incorrect, and support the recognition of S. nigrotaenia as a valid species.Vari's (1991:100) designation of the lectotype of Curimatus nigrotaenia should be considered invalid because the lot from which he separated the specimen was incorrectly labeled as the syntype series of C. nigrotaenia.As specified in the ICZN (1999: Art.74.2), "If it is demonstrated that a specimen designated as a lectotype was not a syntype, it loses its status of lectotype".Günther's citation of BMNH 1902.2.10.30 as a "type" for C. nigrotaenia is similarly invalid because "the mere citation of 'Type' or equivalent expression, … or in an unpublished catalogue of a museum, or on a label, is not necessarily evidence that a specimen is or is fixed as any of the kinds of types referred in this Chapter."(ICZN, 1999: Art. 72.4.7).Consequently, we herein formally designate the specimen of C. nigrotaenia retained by Boulenger (1902) at the British Museum (BMNH 1902.2.10.30) as the new lectotype for Steindachnerina nigrotaenia, with the three syntypes at the Museo Civico di Storia Naturale of Genova (MSNG 14859) becoming paralectotypes.This is likely the same specimen considered by Günther when assembling his catalog of types, but definitely not the same specimen that Vari previously assigned to that catalog number.
How the "type" of Curimatus nigrotaenia was deposited in a jar labeled as C. morawhannae in the collection at the BMNH is an issue that possibly we will never know, but surely we cannot attribute this action to Boulenger when describing the species, because C. morawhannae was described in 1912 by Carl H. Eigenmann based on specimens from Guyana and, obviously, paratypes were sent to the British Museum after the description of the species (1912).So, this error should be credited to someone who worked with the type specimens of C. nigrotaenia and C. morawhannae afterwards, but before Vari's examination.

Molecular analysis.
Molecular results also support the recognition of Steindachnerina nigrotaenia.The matrix contained 675 bp with 48 polymorphic sites (7.1%), and 98.6% pairwise identity.Nucleotide frequencies corresponded to 22.9% adenine, 28.4% cytosine, 18.2% guanine, and 30.6% thymine/uracil.We did not find any saturation in the dataset.Tab. 1 includes voucher information, locality, and Genbank accession numbers.Both neighbor-joining analysis (not shown) and the maximum likelihood approach returned the same topology (Fig. 1).Results revealed a clear separation among samples of S. insculpta from the rio Paraná and the group S. brevipinna/S.nigrotaenia from rio Paraná and rio Paraguai, respectively.The eight specimens of S. nigrotaenia from the upper rio Paraguai (including one topotype from rio Coxipó) appear genetically closer to the five specimens of S. brevipinna from the rio Paraná.These results demonstrate that S. nigrotaenia clearly differs genetically from the morphologically similar S. insculpta, but much less from S. brevipinna which also occurs in the rio Paraguai.It appears that both S. brevipinna and S. nigrotaenia did accumulate morphological differences but not in the mitochondrial DNA.The sum of evidence allows us to redescribe S. nigrotaenia from the rio Paraguai basin, and S. insculpta from the upper rio Paraná basin.Steindachnerina nigrotaenia differs from S. dobula and S. varii by having a dark midlateral stripe extending from the rear of the opercle posteriorly onto middle rays of caudal fin (vs.dark midlateral stripe beginning posterior to a vertical line through the dorsal-fin origin, and continuing posteriorly to the middle rays of the caudal fin).Steindachnerina nigrotaenia differs from the very similar S. insculpta by the generally higher number of perforated scales along the lateral line series (43 to 48, 45 most frequent, vs. 40 to 45, 42 most frequent and 44 and 45 in only 1.3% of specimens examined for this feature, respectively; Fig. 3), by the total number of scale rows in the transverse series from the dorsal-fin origin to the pelvic-fin base (13 to 16, 14 most frequent, vs. 12 to 14, 12 most frequent and 13.5 and 14 in only 6.4% and 1.3% of specimens examined for this feature, respectively; Fig. 4).Steindachnerina nigrotaenia can also be discriminated from congeners by a combination of the following features: multiple lobulated fleshy processes on the roof of the oral cavity, absence of a wide, flattened prepelvic region of the body, possession of 43 to 48 perforated scales in the lateral line series, 13 to 16 scales in the transversal series, lack of a dark spot on the basal portions of the dorsal fin and the presence of a dark midlateral stripe along the lateral surface of the body.Dorsal-fin margin rounded; anteriormost rays three to 3.5 times the length of ultimate ray.Pectoral-fin margin acute, extending one-half to two-third the distance to origin of pelvic fin in smaller adults, barely beyond that point in largest specimens examined.Pelvic-fin margin acute; pelvic fin extending about one half the distance to origin of anal fin.Caudal fin forked.Adipose fin well developed.Anal fin emarginate, anteriormost branched rays 2.5 to three times the length of ultimate ray.

Description
Head profile pointed; upper jaw distinctly longer, mouth inferior; buccopharyngeal complex on roof of the oral cavity in adults consisting of multiple lobulated fleshy bodies; nostrils very close, anterior circular, posterior crescent-shaped with aperture closed by thin flap of skin separating nares; adipose eyelid present, more developed anteriorly, with broad, vertically ovoid opening over center of eye.

Coloration in alcohol.
Overall coloration of specimens retaining guanine on scales silvery to silvery golden, darker on dorsal portions of head and body.Ground coloration of specimens lacking guanine on scales tan to yellow, darker dorsally, with irregular patch of dark pigmentation extending from rear of orbit across opercle; degree of intensity of dark pigmentation and extent of patch variable among individuals.Irregular, dark longitudinal stripe extending along lateral line from supracleithrum to base of middle caudal-fin rays.Stripe slightly wider posteriorly, continuous posteriorly with dusky stripe on middle caudal-fin rays.Stripe in adults continuous, about one scale wide; not as well developed in juveniles, consisting of discrete spots surrounding pores of lateral line.Caudal-fin with stripe more prominent on the anterior portion of rays.Anterior margin and distal portions of dorsal fin typically dusky in adults.Ventral lobe and dorsal rays of dorsal lobe of caudal fin dusky.All caudal-fin rays outlined by small chromatophores on membranes.Adipose fin dusky distally.Dorsal fin with dusky anterior region in many specimens, without any discrete spot on middle rays.Other fins hyaline.
Sexual dimorphism.Secondary sexual characters were not found.
Tab. 2. Morphometric data for Steindachnerina nigrotaenia and S. insculpta.N = number of specimens.SD = Standard deviation.Range of each species includes the lectotype of S. nigrotaenia (BMNH 1902.2.10.30)    this feature; Fig. 3), and by the total number of scale rows in the transverse series from the dorsal-fin origin to the pelvicfin base (12 to 14, 12.5 most frequent, 13.5 and 14 less frequent vs. 13 to 16, 14 most frequent; 13 in only 8.16% of specimens examined for this feature; Fig. 4).Steindachnerina insculpta can also be discriminated from its congeners by a combination of the following features: multiple lobulated fleshy processes on the roof of the oral cavity, absence of a wide, flattened prepelvic region of the body, 40 to 45 scales on the lateral line, 12 to 14 scales on the transversal series, lack of a dark spot on basal portions of the dorsal fin and the presence of a dark midlateral stripe along the body.Median keel barely discernable posterior to pelvic-fin origin.Dorsal-fin margin rounded; anteriormost rays three to three and half times length of ultimate ray.Pectoral-fin margin acute, extending one-half to two-third distance to origin of pelvic fin in smaller adults, barely beyond that point in the largest examined specimens.Pelvic-fin margin acute; pelvic fin extending about one-half distance to origin of anal fin.Caudal fin forked.Adipose fin well developed.Anal fin emarginate, anteriormost branched rays two and one-half to three times the length of ultimate ray.
Head pointed in profile; upper jaw distinctly longer, mouth inferior; buccopharyngeal complex on roof of oral cavity in adults consisting of multiple lobulated fleshy bodies; nostrils very close, anterior circular, posterior crescent-shaped with aperture closed by thin flap of skin separating nares; adipose eyelid present, more developed anteriorly, with broad, vertically ovoid opening over center of eye.

Coloration in alcohol.
Overall coloration of specimens retaining guanine on scales silvery to silvery golden, darker on dorsal portions of head and body.Ground coloration of specimens lacking guanine on scales tan to yellow, darker dorsally, with irregular patch of dark pigmentation extending from rear of orbit across opercle; degree of intensity of dark pigmentation and extent of patch variable among individuals.Irregular, dark longitudinal stripe extending along lateral line from supracleithrum to base of middle caudal-fin rays.Stripe slightly wider posteriorly, continuous posteriorly with dusky stripe on middle caudal-fin rays.Stripe in adults continuous, about one scale wide; not as well developed in juveniles, consisting of discrete spots surrounding pores of lateral line.Caudal-fin stripe more prominent on anterior portion of rays.Anterior margin and distal portions of dorsal fin typically dusky in adults.Ventral lobe and dorsal rays of dorsal lobe of caudal fin dusky.All caudal-fin rays outlined by small chromatophores on membranes.Adipose fin dusky distally.Dorsal fin with dusky anterior region in many specimens, without any discrete spot on middle rays.Other fins hyaline.
Sexual dimorphism.Secondary sexual characters were not found.

Discussion
Populations of Steindachnerina from the rio Paraguai basin and from the upper rio Paraná basin were considered by Vari (1991:71) as belonging to a single species (S. insculpta) but studies carried out on these populations have allowed us to conclude that they belong to two distinct species: S. insculpta (upper rio Paraná basin) and S. nigrotaenia (rio Paraguai basin), supplemented by the very distinct S. brevipinna (rio Paraguai basin) (Fig. 7).
Although the first two species are very similar in many morphological aspects, they might be diagnosed from each other firstly on the basis of the number of scales in the lateral line and number of scales in the transverse line, and they can be further diagnosed by S. brevipinna by the lack of the dark spot on basal portions of the dorsal fin, and by modal differences in the number of scales of the lateral line and of the transverse line (Figs.3-4).Furthermore, a molecular analysis strongly supports the recognition of the three species.Vari (1991) interpreted the presence of Steindachnerina insculpta (= S. nigrotaenia) populations in the rio Paraguai basin as a result of a possible introduction of specimens originated from the upper rio Paraná.This interpretation was first a consequence of the supposition that S. nigrotaenia was a synonym of S. brevipinna, and secondarily because Vari (1991) had only a small number of specimens from the rio Cuiabá collected at that time.That small number of lots concentrated in a small area of the rio Paraguai basin and geographically distant from populations of the upper rio Paraná led him to hypothesize a human-mediated introduction to explain the disjoint distribution of what he considered to be populations of the same species (S. insculpta).Obviously, the present findings disprove his hypothesis, showing that he was working with two distinct species, S. insculpta and S. nigrotaenia.
Several pieces of evidence indicate a gap separating the distributions of Steindachnerina nigrotaenia (upper rio Paraguai and lower rio Paraná) from that of S. insculpta (upper rio Paraná, upriver Itaipu).Inventories in the rio Paraná below Itaipu reported neither S. insculpta nor S. nigrotaenia, and indicated, instead, the presence of S. brevipinna, S. biornata and S. conspersa (e.g., López et al., 2003;2005).In addition, many species of the middle rio Paraná were incorporated to the ichthyofauna of the upper rio Paraná after the construction of the Itaipu dam (Agostinho et al., 1992(Agostinho et al., , 1997;;Júlio Júnior et al., 2009), including S. brevipinna that that is genetically distinct of the sympatric S. insculpta (Oliveira et al., 2002), but no evidence of S. nigrotaenia in that section has been reported recently (Graça, Pavanelli, 2007;Ota et al., 2018).
Specimens identified as Curimata nasa (Steindachner, 1882) by Géry et al. (1987:425) from Paraguay were considered by Vari (1991:71) to belong to S. insculpta (sensu Vari, 1991).However, the five specimens mentioned by Géry et al. (1987) are from different localities: two of them collected in the rio Paraguai (possibly S. nigrotaenia), and three other specimens in the rio Paraná basin (S. insculpta).Nevertheless, any definitive decision on the status of these specimens without a direct examination is premature.More recently, Koerber et al. (2017:16) in the checklist of fishes from Paraguay cited S. brevipinna, S. conspersa and S. insculpta, the latter being S. nigrotaenia according to the concept herein established; they also erroneously refer to the type locality of S. insculpta as being in Venezuela, disregarding that Britski (1969:201), andVari (1991:72) had already clarified this question.Aguilera et al. (2018) also reported the first occurrence of S. insculpta in Argentina, which, based on their description, the species in question also represents S. nigrotaenia.

e180076[13]
After the revision of Steindachnerina (Vari, 1991), two other species were described in the genus and another species is validated herein, then accumulating 25 species of Steindachnerina.Another species, Cyphocharax corumbae (Pavanelli & Britski, 1999) was originally described in Steindachnerina but recently transferred to Cyphocharax (Melo et al., 2018) This study resolves a long-standing question in the systematics of Steindachnerina and adds one more species to the ichthyofauna of the rio Paraguai.Further research might investigate the phylogenetic placement of S. nigrotaenia within the recently proposed S. dobula clade (sensu Melo et al., 2018) and also compare morphological features of the three related species occurring in the La Plata system.

Fig. 1 .
Fig. 1.Best maximum likelihood tree based on partial sequences of the mitochondrial gene cytochrome oxidase c subunit I among samples of Steindachnerina nigrotaenia, S. insculpta and S. brevipinna.Numbers near nodes represent bootstrap values of maximum likelihood/neighbor-joining analyses.

Fig. 5 .
Fig. 5. Distribution map of Steindachnerina brevipinna in both rio Paraná and rio Paraguai (green), S. nigrotaenia in the rio Paraguai (red), and S. insculpta in the upper rio Paraná (yellow).Pentagons indicate type localities.Some symbols represent more than one collecting locality.
Description.Morphometric data of Steindachnerina insculpta are summarized in Tab. 2. Body relatively elongate, somewhat compressed.Dorsal profile of head very slightly convex anteriorly, straight from above orbit to rear of head.Dorsal profile of body straight or very slightly convex e180076[11]from rear of head to origin of dorsal fin; straight at base of dorsal fin; straight from base of last dorsal-fin ray to caudal peduncle.Dorsal surface of body transversely rounded anteriorly, with indistinct median keel immediately anterior to dorsal fin, smoothly rounded transversely posterior to dorsal fin.Ventral profile of body gently curved from tip of lower jaw to caudal peduncle.Prepelvic region obtusely flattened, with median series of scales proximate to pelvicfin origin, median series less regularly arranged anteriorly.