Revisionary study of the armored catfish Corydoras paleatus ( Jenyns , 1842 ) ( Siluriformes : Callichthyidae ) over 180 years after its discovery by Darwin , with description of a new species

The taxon known as Corydoras paleatus carries one of the most complex taxonomic histories among Corydoradinae catfishes. A comprehensive review of specimens attributed to that species from several localities was carried out, allowing the clear recognition of C. paleatus and also of a new species previously misidentified as C. paleatus, described herein. Corydoras paleatus can be distinguished from its congeners by presenting the following unique combination of features: perpendicularly directed serrations along entire posterior margin of the pectoral spine; three large black blotches along midline of flank; hyaline or black pectoral fin; and transversal black bars on caudal-fin lobes. Corydoras paleatus is known from the lower rio Paraná basin, coastal rivers from Southern Brazil and Uruguay and rio Uruguai basin. The new species can be distinguished from its congeners by having the following unique combination of features: three large black blotches along midline of flank; three nasal pores; mesethmoid short; infraorbital 2 not in contact with compound pterotic; striated black spots from the snout tip to nuchal plate region; markedly rounded snout; and anterior portion of dorsal fin spotted. The new species is known from the upper rio Uruguai basin, from the rivers Canoas, do Peixe and Pelotas.

The taxon known as Corydoras paleatus was originally described as Callichthys paleatus by Jenyns in 1842, based on material collected by Charles Darwin, during his circumnavigation voyage on board the ship HMS Beagle, in the years 1831 to 1836.As recorded by Jenyns (1842: 114), the five specimens of Callichthys paleatus, which were brought to England by Darwin, had lost their attached labels that contained corresponding numbers in Darwin's manuscripts with noted localities.The species description 1879 (p.32: a nomem novum for the same C. punctatus recorded by Valenciennes in Cuvier & Valenciennes (1840), and d' Orbigny (1847), but no description or definition was provided and no specimen traced).In addition, both authors restricted the type-locality of C. paletaus to "Argentina -Buenos Aires, Río Paraná at San Pedro, 33º43'S 59º45 'W,220 km NW of Buenos Aires" (p. 204).They also recorded that only three of the five syntypes reported by Jenyns (1842: 114) were found and designated a lectotype (BMNH 1917.7.14:18).Isbrücker (2001), in a brief catalogue of genera and species in Corydoradinae, validated several species.As criteria, he considered distances from type-localities and information from living forms.In this way, Isbrücker (2001: 229, 230) listed C. maculatus, C. marmoratus and C. microcephalus as valid, although doubtful, species.The author considered only Silurus quadricostatus Larrañaga, 1923 and Silurus septemradiatus Larrañaga, 1923, although also doubtful, in the synonymy of C. paleatus.Though not mentioned, Isbrücker (2001) apparently followed the work of Devicenzi (1925) to propose such synonymy.In that work, Devicenzi (1925) provided comments for all the fish species cited by Larrañaga (1923), and assigned the correspondence between Silurus quadricostatus and S. septemradiatus with C. paleatus (see Devicenzi, 1925: 307).Reis (2003: 301) followed Nijssen & Isbrücker (1980) in the synonymy of Corydoras paleatus, plus Larrañaga (1923) species.Knaack (2007a) analyzed the type-series of C. paleatus and concluded that the specimen stated as BMNH 1917.7.14.18 was not the same specimen designated as lectotype by Nijssen & Isbücker (1980) and declared that the true lectotype is lost, designating a neotype (BMNH 1917.7.14.19).This probably occurred because Knaack confused the labels of the lots, since the specimen presented as a possible lectotype of C. paleatus in fact corresponds to a specimen of the type-series of Corydoras micracanthus Regan, 1912(see Knaack, 2007a: 28).
One year later, Knaack (2008) mentioned: "researches in the internet led to pictures of a fish, which was named as lectotype (of Callichthys paleatus).This fish is clearly a member of the series of the syntypes and the fish, which was designated as lectotype by Nijssen & Isbrücker (1980)", considering invalid the designation of the neotype, since the supposed lost lectotype had been rediscovered.The author also provided morphometric and meristic data of the lectotype and the two paralectotypes.Despite comments and justifications presented by Knaack (2007a;2008), the specimen designated as the lectotype of Corydoras paleatus by Nijssen & Isbrücker (1980) was probably never lost, being only a possible mistake committed by Knaack (2007a), which may have confused the labels.Knaack (2008) did not mention any connection between the specimen presented by him as the wrongly stated C. paleatus lectotype (see Knaack, 2007a: 28) and C. micracanthus.
All the aforementioned facts contribute to cause uncertainty about Corydoras paleatus identity.Nevertheless, specimens indiscriminately attributed to C. paleatus have been used in many different kinds of studies, like behavior, physiology, fish pathology, among others (e.g.Pruzsinszky & Ladich, 1998;Kaatz & Lobel, 1999;Fanta et al., 2003;Cazenave et al., 2005Cazenave et al., , 2006;;Pesce et al., 2008), which makes it one of the most useful corydoradine catfish in Experimental Biology.Therefore, it is important to precisely assign its limits and definition.After the analysis of several specimens from various localities, including lectotype and paralectotypes of C. paleatus and C. marmoratus, and the lectotype of C. microcephalus, allied to information provided in C. paleatus original description, it was possible to recognize the populations corresponding to C. paleatus and determine which of them refer to the species described by Jenyns (1842).Additionally, the analysis of the gathered material revealed a new species previously misidentified as C. paleatus, which is described herein.A redescription of C. paleatus is also provided to allow its clear definition and correct identification, and put a conclusive denouement about the taxonomic status of those specimens collected by Darwin around 180 years ago.
The synonymic list comprised only the works from which the voucher specimens could be examined or those that provided high quality photos, allowing then the clear recognition of Corydoras paleatus.Jenyns, 1842(In Jenyns, 1840-42): 113-114. Corydoras marmoratus Steindachner, 1879a: 26-28, pl. V fig. 1 (partim).Corydoras paleatus Eigenmann & Eigenmann, 1890: 471. -Almirón et al., 2008: 124. -Casciotta et al., 2005: 77, 167, 232, fig. 88. -Almirón et al., 2015: 29, 150, 151.Corydoras longipinnis Knaack, 2007: 36;39, figs. 5, 6 and 7;43, fig. 12 (partim)   Diagnosis.Corydoras paleatus can be distinguished from its congeners, with exception of C. armatus (Günther, 1868) and C. microcephalus, by the presence of perpendicularly directed serrations along entire posterior margin of the pectoral spines (vs.serrations directed towards pectoralspine origin; serrations directed towards pectoral-spine tip; perpendicularly directed serrations, if present, bifid or restricted to proximal region of pectoral spine); from C. armatus by the absence of contact between infraorbital 2 and compound pterotic (vs.presence); from C. microcephalus by the presence of a longitudinal series of three large black blotches along midline of flank (vs.four or five midline blotches in C. microcephalus).Additionally, C. paleatus can be distinguished from C. cohui Myers & Weitzman, 1954, C. diphyes Axenrot & Kullander, 2003, C. flaveolus, C. froehlichi, C. gryphus Tencatt, Britto & Pavanelli, 2014, C. habrosus, C. lacrimostigmata Tencatt, Britto & Pavanelli, 2014, C. longipinnis Knaack, 2007and C. lymnades Tencatt, Vera-Alcaraz, Britto & Pavanelli, 2013 by having the anterior region of the dorsal fin black (vs.with black spots or bars); from C. carlae Nijssen & Isbrücker, 1983, C. garbei Ihering, 1911, C. difluviatilis Britto & Castro, 2002 by the presence of hyaline or black pectoral fin (vs.with black spots or bars); from C. reynoldsi, C. tukano Britto & Lima, 2003 and C. weitzmani by the absence of a conspicuous coloration pattern in the region of the eye (vs.presence of a transversal black stripe across the eye); from C. gladysae, C. micracanthus, C. petracinii by having dorsal spine relatively long, with adpressed distal tip slightly surpassing the base of the last branched dorsal-ray (vs.conspicuously short, with adpressed distal tip not reaching the base of the last branched dorsal-ray; from C. ehrhardti by the presence of transversal black bars on caudal-fin lobes (vs.caudal-fin lobes hyaline or brownish); from C. steindachneri by having a fold in the middle portion of the lower lip (vs.absence).1. Head compressed with slightly convex dorsal profile; triangular in dorsal view.Snout slightly pointed.Head profile convex from snout to anterior nares; roughly straight from this point to tip of posterior process of parieto-supraoccipital.Profile slightly convex along dorsal-fin base.Postdorsal-fin body profile concave to adipose-fin spine; markedly concave from this point to caudal-fin base.Ventral profile of body nearly straight from isthmus to pectoral girdle; slightly convex from this point until pelvic girdle.Profile nearly straight from pelvic girdle to base of first anal-fin ray; slightly concave until caudal-fin base.Body roughly elliptical in cross section at pectoral girdle, gradually becoming more compressed toward caudal fin.

Description. Morphometric data presented in Table
Eye rounded, located dorso-laterally on head; orbit delimited dorsally by frontal and sphenotic, ventrally by infraorbitals.Anterior and posterior nares close to each other, only separated by flap of skin.Anterior naris tubular.Posterior naris close to anterodorsal margin of orbit, separated from it by distance equal to diameter of naris.
Mouth small, subterminal, width nearly equal to bony orbit diameter.Maxillary barbel moderate in size, not reaching anteroventral limit of gill opening.Outer mental barbel slightly longer than maxillary barbel.Inner mental barbel fleshy, base slightly separated from its counterpart.Small rounded papillae covering entire surface of all barbels, upper and lower lips, and isthmus.
Mesethmoid short; anterior tip moderately developed, smaller than 50% of bone length (see Britto, 2003: 123, character 1, state 1; fig.1B), with very reduced lateral cornua; posterior portion widened, partially exposed and possessing minute odontodes.Nasal slender, curved laterally, anterior portion of inner margin laminar; mesial border contacting only frontal.Frontal elongated, narrow, width less than half entire length; anterior projection short, size smaller than nasal length.Frontal fontanel large, oval; posterior tip slightly entering anterior margin of parieto-supraoccipital.Parieto-supraoccipital wide, posterior process long and contacting nuchal plate; region of contact between posterior process and nuchal plate covered by thick layer skin.
Two laminar infraorbitals with minute odontodes; infraorbital 1 large, ventral laminar expansion moderately developed (Fig. 3a); anterior portion with large expansion; infraorbital 2 small, posteroventral margin contacting posterodorsal ridge of hyomandibula, dorsal tip slender, contacting only sphenotic (Fig. 3a).Posterodorsal ridge of hyomandibula close to its articulation with opercle oblong; exposed, slightly thickened and bearing small odontodes; dorsal ridge of hyomandibula between compound pterotic and opercle covered by thick layer of skin.Interopercle covered by a thin layer of skin, somewhat triangular, anterior projection well developed.Preopercle slender, elongated, minute odontodes sparse on external surface.Opercle compact, width slightly smaller than length; free margin angulated, with uncinate expansion in some specimens; without serrations and covered by small odontodes.Anteroventral portion of cleithrum exposed; Posterolateral portion of scapulocoracoid exposed; minute odontodes sparse on exposed areas.Vertebral count 21 (2), 22 (1); ribs 6 (3), first pair conspicuously large; complex vertebra compact in shape.Neural and haemal spines pointed.
Four branchiostegal rays decreasing in size posteriorly.Hypobranchial 2 somewhat triangular, tip ossified and directed towards anterior portion, posterior margin cartilaginous; ossified portion well developed, about twice size cartilaginous portion.Five ceratobranchials with expansions increasing posteriorly; ceratobranchial 1 with small process on anterior margin of mesial portion; ceratobranchial 3 notched on posterolateral margin; ceratobranchial 5 toothed on posterodorsal surface, 35 to 39 (3) teeth aligned in one row.Four epibranchials with similar size; epibranchial 2 slightly larger than others, with small pointed process on laminar expansion of posterior margin; epibranchial 3 with curved mesially uncinate process on laminar expansion of posterior margin.Two wide pharyngobranchials (3 and 4); pharyngobranchial 3 with large triangular process on posterior margin.Upper tooth plate oval; 37 to 45 (3) teeth aligned in two rows on posteroventral surface.
Lateral-line canal entering neurocranium through compound pterotic, splitting into two branches before entering sphenotic: pterotic branch with a single pore; preoperculomandibular branch conspicuously reduced, with a single pore opening close to postotic main canal.Sensory canal continuing through compound pterotic, entering sphenotic as temporal canal, which splits into two branches: one branch giving rise to infraorbital canal, other branch entering frontal through supraorbital canal.Supraorbital canal not branched, running through nasal bone.Epiphyseal pore opening at supraorbital main canal.Nasal canal with three openings, first on posterior edge, second on posterolateral portion and third on anterior edge.Infraorbital canal running through entire second infraorbital, extending to infraorbital 1 and opening into two pores.Preoperculomandibular branch giving rise to preoperculo-mandibular canal, which runs through entire preopercle with three openings, leading to pores 3, 4, and 5, respectively.Dorsal fin triangular, located just posterior to second dorsolateral body plate.Dorsal-fin rays II,8 (42), II,9 (3), posterior margin of dorsal-fin spine with five to nine serrations, only in distal portion of spine; serrations smoothly directed towards dorsal-spine tip; serrations absent in some specimens.Nuchal plate exposed with minute odontodes; spinelet short; spine relatively long, adpressed distal tip slightly surpassing base of last branched dorsal-ray; anterior margin with small odontodes.Pectoral fin triangular, its origin just posterior to gill opening.Pectoral-fin rays I,7 (14), I,8 (31); posterior margin of pectoral spine with 16-24 relatively well-developed serrations along its entire length; serrations disposed perpendicularly to spine (Fig. 4a).Pelvic fin oblong, located just below first ventrolateral body plate, and at vertical through first branched dorsal-fin ray.Pelvicfin rays i,5.Adipose fin roughly triangular, separated from base of last dorsal-fin ray by generally seven dorsolateral body plates.Anal fin triangular, located just posterior to 12 th ventrolateral body plate, and at vertical through anterior margin of adipose-fin spine.Anal-fin rays i,6 (42), i,7 (3).Caudal-fin rays i,12,i, generally three dorsal and ventral procurrent rays, respectively; bilobed, lobes with similar size.
Sexual dimorphism.In general, the presence of lanceolate genital papilla is useful to recognize Corydoradinae males (see Nijssen & Isbrücker, 1980;Britto, 2003), as can be observed on Corydoras paleatus as well.Additionally, dorsal-fin spine and odontodes are more developed in males; first and second dorsal-fin branched rays slightly surpass dorsal-spine tip.Females are generally larger than males.

Geographic distribution.
Corydoras paleatus is known from the lower rio Paraná basin in Argentina, coastal rivers from Southern Brazil, Rio Grande do Sul State, and Uruguay, rio de La Plata basin in Argentina and Uruguay, and rio Uruguay basin in Uruguay (Fig. 6).

Conservation status. Populations of Corydoras paleatus
are known from at least three relatively wide regions, comprising territories from Argentina, Brazil and Uruguay, and no imminent threat is suspected for the species as whole, therefore, according to the International Union for Conservation of Nature (IUCN) categories and criteria (IUCN Standards and Petitions Subcommittee, 2014), C. paleatus can be classified as Least Concern (LC).Remarks.In the review of Corydoras conducted by Nijssen & Isbrücker (1980), the authors restricted the typelocality of C. paleatus, however this was done without any justification.Despite this region has been part of the places visited by Darwin on his voyage (Darwin, 1839), there is no evidence for such a restriction of the type-locality.
Unfortunately, there are no concrete evidences that lead to the exact type-locality where Darwin may have collected C. paleatus, but it is more likely that this species has been collected in Uruguay and not in Argentina.Calviño (2007) rescued the type-locality of four species collected by Darwin during the voyage of the Beagle.The author analyzed potential locations where Darwin would have collected, finding the Laguna del Diario, a lake which supplies the city of Maldonado, where Darwin stayed for more than two months and collected several fish species.Jenyns (1842) mentions that "in a lake suddenly drained" (Darwin (1839: 134) mentioned "nearly drained") many fish samples were collected, which makes sense since the fish sampling tends to be more practical in low volume sites.Corydoras paleatus inhabits the Laguna Del Diario (Calviño pers.comm.), so it is possible that it may have been collected during the draining period.However, without the original labels, it is quite risky assign or even restricts the type-locality of C. paleatus.
Regarding the synonymic list of Corydoras paleatus proposed by Nijssen & Isbrücker (1980), some adjustments were proposed herein.The specimen described by Valenciennes (1840) was referred as Corydoras punctatus var.argentina in Steindachner (1879a), assigned as nomen nudum in the synonymy of C. marmoratus (see Nijssen & Isbrücker, 1980: 204).After that, C. maculatus was proposed as a nomen novum for Corydoras punctatus var.argentina in Steindachner (1879b).Despite the fact that this specimen could not be traced (Nijssen & Isbrücker, 1980: 204), d'Orbigny (1847) provided an illustration of it, in which it is possible to observe a slender and regular longitudinal black stripe just above midline of flank and no additional spots or blotches on the body, and also the entire distal border of dorsal fin black, being thus clearly different from the color pattern of C. paleatus.By this reason, both names should be removed from the synonymy of C. paleatus.
Additionally, based on the original description by Regan (1912) and on the analysis of the lectotype of C. microcephalus, it was possible to observe the presence of a longitudinal series of four to five large black blotches along midline of flank, clearly distinguishing this species from C. paleatus, which presents only three blotches.Some specimens of C. paleatus present fragmented midline blotches, which can also be observed in C. longipinnis (see Knaack, 2007b: 41, fig. 9), nevertheless, even this condition is clearly different from what is displayed by the lectotype of C. microcephalus (see Knaack, 2007a: 29, fig. 10).Analysis of some specimens from Argentina, Buenos Aires Province near Tres Arroyos, also revealed the presence of four to five large midline blotches.However, the material is relatively old and damaged making it difficult to accurately assign these specimens to C. microcephalus.Nevertheless, these specimens clearly fit more with the morphotype of C. microcephalus than of C. paleatus, and, by this reason, they were considered here as Corydoras cf.microcephalus.For now, the most reasonable decision is to consider C. microcephalus as valid and removing it from C. paleatus synonymy until further analysis be performed.Reis (2003: 301) presented a new synonymic list for Corydoras paleatus, which includes the same species from Nijssen & Isbrücker (1980) study plus Silurus quadricostatus and Silurus septemradiatus.Devicenzi (1925: 307), in his review of the fish species of Larrañaga from Uruguay, stated that the species described by Larrañaga (1923) were actually synonyms of C. paleatus.However, this statement was given without any concrete evidence, since it appears that Devicenzi (1925) only provided the original description by Larrañaga and added some comments.It is not completely clear but, apparently, he does not examine those specimens.It is also important to mention that the direct link between Larrañaga's species and C. paleatus is not that obvius since the brief descriptions fit for several other callichthyids and no specimens could be traced for examination, either for the Larrañaga's (1923) or the Devicenzi's (1925) works.Therefore, both species described by Larrañaga constitute species inquirendae following the definition of the Code (a species of doubtful identity needing further investigation; ICZN, 1999: 116) and also need to be removed from the synonymy of C. paleatus.
The inclusion of part of the material of Corydoras longipinnis in the synonymy of C. paleatus is justifiable due to the fact that the allotype (AI 222) clearly corresponds to C. paleatus.Examination of C. marmoratus type-series also indicated the presence of more than one species.The lectotype and some paralectotypes (Fig. 7) (NMW 46775-1, 46775-2, 46777-1 and 46777-2) clearly fit with the description of C. paleatus, and therefore, this species is kept in its synonymy.However, the paralectotype NMW 46776 corresponds to the description of C. longipinnis.Description.Morphometric data presented in Table 1.Head compressed with slightly convex dorsal profile; triangular in dorsal view.Snout roughly rounded.Head profile convex from snout to tip of posterior process of parieto-supraoccipital.Profile slightly convex along dorsalfin base.Postdorsal-fin body profile concave to adiposefin spine; markedly concave from this point to caudal-fin base.Ventral profile of body nearly straight from isthmus to pectoral girdle; slightly convex from this point until pelvic girdle.Profile nearly straight from pelvic girdle to base of first anal-fin ray; slightly concave until caudal-fin base.Body roughly elliptical in cross section at pectoral girdle, gradually becoming more compressed toward caudal fin.
Eye rounded, located dorso-laterally on head; orbit delimited dorsally by frontal and sphenotic, ventrally by infraorbitals.Anterior and posterior nares close to each other, only separated by flap of skin.Anterior naris tubular.Posterior naris close to anterodorsal margin of orbit, separated from it by distance equal to diameter of naris.Mouth small, subterminal, width nearly equal to bony orbit diameter.Maxillary barbel short, not reaching anteroventral limit of gill opening.Outer mental barbel slightly longer than maxillary barbel.Inner mental barbel fleshy, base slightly separated from its counterpart; insertion of barbel in middle of lower lip.Small rounded papillae covering entire surface of all barbels, upper and lower lips, and isthmus.
Mesethmoid short; anterior tip moderately developed, smaller than 50% of bone length (see Britto, 2003: 123, character 1, state 1; fig.1B), with small lateral cornua; posterior portion widened, partially exposed and possessing minute odontodes.Nasal slender, curved laterally; inner margin laminar, outer margin laminar in some specimens; mesial border contacting mesethmoid and frontal.Frontal elongated, narrow, width less than half entire length; anterior projection short, size smaller than nasal length.Frontal fontanel large, oval; posterior tip extension slightly entering anterior margin of parieto-supraoccipital.Parietosupraoccipital wide, posterior process long and contacting nuchal plate; region of contact between posterior process and nuchal plate covered by thick layer skin.
Two laminar infraorbitals with minute odontodes; infraorbital 1 large, ventral laminar expansion moderately developed (Fig. 3b); anterior portion with moderately developed expansion; infraorbital 2 small, posteroventral margin contacting posterodorsal ridge of hyomandibula, dorsal tip slender, contacting only sphenotic (Fig. 3b).Posterodorsal ridge of hyomandibula close to its articulation with opercle oblong; exposed, slightly thickened and bearing small odontodes; dorsal ridge of hyomandibula between compound pterotic and opercle covered by a thick layer of skin.Interopercle covered by thick layer of skin, somewhat triangular, anterior projection well developed.Preopercle slender, elongated, minute odontodes sparse on external surface.Opercle dorso-ventrally elongated, width equal or smaller than half of its length; free margin rounded, not angulated; few specimens with smooth uncinate expansion; covered by small odontodes.Anteroventral portion of cleithrum exposed; Posterolateral portion of scapulocoracoid exposed; minute odontodes sparse on exposed areas.Vertebral count 22 (3); ribs 6 (3), first pair conspicuously large; complex vertebra compact in shape.Neural and haemal spines with distal portion expanded.
Four branchiostegal rays decreasing in size posteriorly.Hypobranchial 2 somewhat triangular, tip ossified and directed towards anterior portion, posterior margin cartilaginous; ossified portion well developed, about twice size cartilaginous portion.Five ceratobranchials with expansions increasing posteriorly; ceratobranchial 1 with small process on anterior margin of mesial portion; ceratobranchial 3 notched on posterolateral margin; ceratobranchial 5 toothed on posterodorsal surface, 30 to 37 (3) teeth aligned in one row.Four epibranchials with similar size; epibranchial 2 slightly larger than others, with small pointed process on laminar expansion of posterior margin; epibranchial 3 with curved mesially uncinate process on laminar expansion of posterior margin.Two wide pharyngobranchials (3 and 4), pharyngobranchial 3 with large triangular process on posterior margin.Upper tooth plate oval; 38 to 42 (3) teeth aligned in two rows on posteroventral surface.
Lateral-line canal entering neurocranium through compound pterotic, splitting into two branches before entering sphenotic: pterotic branch with a single pore; preoperculomandibular branch conspicuously reduced, with a single pore opening close to postotic main canal.Sensory canal continuing through compound pterotic, entering sphenotic as temporal canal, which splits into two branches: one branch giving rise to infraorbital canal, other branch entering frontal through supraorbital canal.Supraorbital canal not branched, running through nasal bone.Epiphyseal pore opening at supraorbital main canal.Nasal canal with three openings, first on posterior edge, second on posterolateral portion and third on anterior edge.Infraorbital canal running through entire second infraorbital, extending to infraorbital 1 and opening into two pores.Preoperculomandibular branch giving rise to preoperculo-mandibular canal, which runs through entire preopercle with three openings, leading to pores 3, 4, and 5, respectively.Dorsal fin roughly triangular, located just posterior to second dorsolateral body plate.Dorsal-fin rays II,7 (1), II,8* (31), posterior margin of dorsal-fin spine with 17 to 18 serrations along entire posterior margin of spine; serrations directed towards dorsal-spine tip; nuchal plate exposed with minute odontodes; spinelet short; spine relatively long, adpressed distal tip slightly surpassing base of last branched dorsal-ray, anterior margin with small odontodes.Pectoral fin triangular, its origin just posterior to gill opening.Pectoral-fin rays I,8; posterior margin of pectoral spine with 20-22 well developed serrations along its entire length; serrations directed towards pectoral-spine tip; presence of bifid perpendicularly directed serrations on proximal region of spine (Fig. 4b).Pelvic fin oblong or rounded, located just below first ventrolateral body plate, and at vertical through first branched dorsal-fin ray.Pelvicfin rays i,5.Adipose fin roughly triangular, separated from base of last dorsal-fin ray by generally six dorsolateral body plates.Anal fin triangular, located just posterior to 12 th ventrolateral body plate, and at vertical through anterior margin of adipose-fin spine.Anal-fin rays i,6.Caudal-fin rays i,12,i, generally four dorsal and ventral procurrent rays; bilobed, lobes with similar size.
Color in alcohol.Ground color of body yellowish, with top of head black.Presence of striated black spots from tip of snout to nuchal plate region.Opercle with yellowish border; anterior portion dark brown.Dorsal series of four black blotches, first on anterior portion of dorsalfin base, second on posterior portion of dorsal-fin base, third on adipose-fin base and fourth on caudal-fin base.Dorsal portion of dorsolateral plates generally with dark chromatophores scarcely scattered.Midline of flank with longitudinal series of three large black blotches; anterior and middle blotches conspicuously rectangular; posterior ellipsoid.Dorsal fin generally with transversal black bar on its middle portion.Pectoral fin with diffuse spots; hyaline in some specimens.Pelvic fin generally hyaline; with black spots in few specimens.Adipose fin hyaline; with black area on distal portion of membrane in some specimens.Anal fin with transversal black bar; hyaline in some specimens.Middle portion of caudal-fin base with small black dot.Caudal fin with generally three transversal black bars; hyaline portion of caudal fin with same width as bars; transversal bars conspicuously larger in width.
Sexual dimorphism.Male specimens of Corydoras froehlichi also present lanceolate genital papilla (see Nijssen & Isbrücker, 1980;Britto, 2003).As C. paleatus, males of the new species also present dorsal-fin spine and odontodes more developed than females, and first and second dorsal-fin branched rays slightly surpassing dorsal-spine tip.Females are generally larger than males.In addition, presence of elongated pelvic fin in males (Fig. 9a); females with shorter and rounded pelvic fin (Fig. 9b).The rounded pelvic fin of the females is used to make a ventral pouch to hold the eggs during spawning (H.Evers, pers.comm.).

Distribution.
Corydoras froehlichi is known from the upper rio Uruguai basin, from the rivers Canoas, do Peixe and Pelotas (Fig. 6).
Etymology.Corydoras froehlichi is named in honor and memory of Dr. Otávio Froehlich (1958-2015), UFMS (Universidade Federal de Mato Grosso do Sul), great teacher, mentor and dear friend, for generously sharing his knowledge with several colleagues, besides contributing to the development of LFCT as researcher and person.

Discussion
After the original description of C. paleatus, various changes occurred on its taxonomy.Explanations for such a set of changes could be related to misconceptions about its diagnosis and description, and also the absence of accurate information about the type-locality.Additionally, the geographic distribution of the populations erroneously identified as C. paleatus is relatively wide, occurring in almost entire Southern region of South America, which made the taxonomic problems related to this species even worse.
Even with the unsolved question regarding C. paleatus type-locality, Jenyns (1842: 113)  Despite the presence of multiples cases of convergent color pattern in Corydoras (see Alexandrou et al., 2011), some phylogenetic hypotheses (e.g.Alexandrou et al., 2011;Vera-Alcaraz, 2013)   paleatus and most of its close-related species seems to be shared, it is quite possible that C. froehlichi could also be close related to the aforementioned species.
Corydoras longipinnis is the species most often misidentified as C. paleatus.Knaack (2007b) considered specimens with the same color pattern of C. paleatus as females of C. longipinnis, and, by this reason, specimens with black pectoral fins from the rio Dulce drainage (Fig. 10) are frequently misidentified as females of C. longipinnis.Additionally, data presented by Knaack (2007b) do not fit with the type-material analyzed herein, such as the number of dorsolateral body plates, stated ranging from 20 to 21 in the description, and from 22 to 23 here.Therefore, the morphotypes from the upper rio Paraná, rio Uruguai and coastal rivers of southern Brazil and Uruguay basins correspond to C. longipinnis.However, the scarcity of available material of C. longipinnis topotypes made it difficult to clearly identify this species and to conclude that those morphotypes are certainly from that species.With the exception of the holotype, allotype, and five paratypes, the remaining type material is in the private collection of Knaack, who recently passed away (Zarske, 2013), and therefore the location of this material, about 60 specimens, is currently uncertain.In addition to the holotype and allotype, four topotypes could be examined here.
In the original description of C. longipinnis, Knaack (2007b) mentioned that one of the main diagnostic features of the species is the elongation of the first and second branched dorsal-fin rays in males of C. longipinnis.Some specimens from the rio Uruguai, rio Tibagi, coastal rivers of southern Brazil and Uruguay and mainly rio Iguaçu basins (see Baumgartner et al., 2012: 116), all misidentified as C. paleatus, also present this feature, but it occurred variably and, apparently, is not related to the size of the specimen, since large males may lack this elongation.Despite the similarity in the morphology and color pattern with the holotype, the diagnostic morphological features pointed by Knaack (2007b) overlap with the examined populations and many other similar species, like C. carlae, C. ehrhardti and even C. paleatus.By this reason, these populations were treated here as C. longipinnis until more material of this species, especially from the lower rio Paraná, can be examined for a possible future redescription and improvement of its geographical distribution.

Fig. 6 .
Fig. 6.Map showing the geographical distribution of Corydoras paleatus (stars) and Corydoras froehlichi (squares).The red star represents the Laguna del Diario, Maldonado, Uruguay, one of the plausible places that may be the type-locality of C. paleatus.The red square represents the type-locality of C. froehlichi, rio Pelotas, Rio Grande do Sul, Brazil.Each symbol may represent more than one locality.
Conservation status.Despite Corydoras froehlichi is known only from the rio Pelotas and two of its tributaries, Canoas and do Peixe rivers, it is probably widespread in other tributaries to the rio Pelotas.Additionally, no imminent threat is suspected, therefore, according to the International Union for Conservation of Nature (IUCN) categories and criteria (IUCN Standards and Petitions Subcommittee, 2014), C. froehlichi can be classified as Least Concern (LC).
mentioned that the color pattern of pectoral, pelvic and anal fins of C. paleatus were "almost wholly dusky", which was essential for its clear recognition.With the exception of C. microcephalus, the color patterns of the aforementioned fins, plus dorsal fin, can be very useful to distinguish C. paleatus from its most similar congeners, such as C. carlae, C. ehrhardti, C. froehlichi and C. longipinnis (see Diagnosis).On the other hand, C. froehlichi can be promptly distinguished from its most similar congeners, C. carlae, C. ehrhardti, C. longipinnis, C. microcephalus and C. paleatus, by the presence of striated black spots from the snout tip to nuchal plate region, forming a somewhat marbled pattern in some specimens (vs.spots, when present, rounded, not forming marbled pattern) and thicker transversal black bars on caudal fin (vs.bars, when present, slender).
suggest that many species sharing similar color pattern with C. paleatus may constitute a monophyletic grouping.In Alexandrou et al. (2011) phylogenetic study, C. paleatus appeared in a relatively small clade comprised by C. albolineatus, C. diphyes, C. ehrhardti, C. flaveolus, C. longipinnis, C. nattereri, C. potaroensis, C. reynoldsi and C. tukano.Vera-Alcaraz (2013) presented a comprehensive phylogenetic hypothesis based on morphological and molecular data, proposing a 'C.paleatus clade', which includes C. cochui, C. diphyes, C. ehrhardti, C. flaveolus, C. longipinnis, C. nattereri and C. tukano.Among the aforementioned species, only C. albolineatus, C. nattereri and C. potaroensis does not share similar color pattern with C. paleatus.Tencatt et al. (2013) also found a similar result with the inclusion of C. lymnades in Britto's (2003) data matrix, finding it as a sister-group of C. flaveolus, and both close related to C. paleatus.Since the general color pattern displayed by C.
Color in alcohol.Ground color of body grayish yellow, with top of head black.Dark spots, when present, generally restricted to snout tip; until anterior margin of orbit in some specimens.Opercle with yellowish border; anterior portion black or dark brown.Dorsal series of four black blotches, first on anterior portion of dorsal-fin base, second on posterior portion of dorsal-fin base, third on adipose-fin base and fourth on caudal-fin base.Dorsal portion of dorsolateral body plates generally with black area; forming longitudinal dorsal stripe in some specimens.Midline of flank with longitudinal series of three large conspicuous black blotches; anterior and middle blotches generally rectangular, and posterior generally ellipsoid; blotches conspicuously elongated in some specimens; coalescent, forming thick and irregular longitudinal stripe in few specimens; fragmented blotches in some specimens.Presence of conspicuously smaller irregular black patches between midline blotches in few specimens.Ventral portion of ventrolateral body plates black in some individuals.First and second branched rays of dorsal fin, including membranes, black or dark brown, and the remaining rays mottled.Pectoral fin black with hyaline distal margin; few specimens with entirely hyaline pectoral fin.Pelvic fin black on its anterior half; entirely hyaline in some specimens.Adipose-fin membrane black or with dark brown area on distal portion.Anal fin blotched; almost entire black in some specimens.Middle portion of caudal-fin base with small black dot.Caudal fin with three to five transversal black bars; hyaline portion of caudal fin with smaller or equal width as bars.