Astyanax lacustris ( Lütken, 1875 ), Lutken, 1875

De Lucena, Carlos Alberto S. & Soares, Helena Gouvea, 2016, Review of species of the Astyanax bimaculatus “ caudal peduncle spot ” subgroup sensu Garutti & Langeani (Characiformes, Characidae) from the rio La Plata and rio São Francisco drainages and coastal systems of southern Brazil and Uruguay, Zootaxa 4072 (1), pp. 101-125 : 103-114

Astyanax lacustris ( Lütken, 1875)

( Fig. 1 ; Table 1 )

Tetragonopterus lacustris Lütken, 1875: 131 (type locality: Lagoa Santa [rio das Velhas, drainage of rio São Francisco]. Tetragonopterus jacuhiensis Cope, 1894: 88 (type locality: Rio Grande do Sul, restricted to the rio Jacuí, Rio Grande do Sul, by Fowler, 1906: 435). [Syn. nov.]

Astyanax (Poecilurichthys) bimaculatus paraguayensis Eigenmann, 1921: 256 (type locality: Paraguay [River] and upper Tocantins [River]). [Syn. nov.]

Astyanax bimaculatus asuncionensis Géry, 1972: 3 (name in substitution of Astyanax (Poecilurichthys) bimaculatusparaguayensis Eigenmann, 1921 , preoccupied by Bertoniolus paraguayensis Fowler, 1918 ). [Syn. nov.] Astyanax altiparanae Garutti & Britski, 2000: 67 (type locality: rio Grande, Volta Grande Dam, Miguelópolis, São Paulo, Brazil). [Syn. nov.]

Diagnosis. Astyanax lacustris is distinguished from other species of the genus Astyanax by presenting a horizontally oval, black humeral spot and two brown vertical bars situated in the humeral region, characteristics that include it in the Astyanax bimaculatus group. It is distinguished from A. argyromarginatus , A. clavitaeniatus , A. goyacensis , A. incaicus , A. novae , A. rupununi , A. saltor , A. siapae , A. unitaeniatus , and A. utiariti by the absence of a conspicuous midlateral black stripe extending above the lateral line, from the humeral spot or just behind it (from the second vertical bar), until the caudal peduncle, continuing along the median rays of the caudal fin; the midlateral stripe continually narrows forward (vs. midlateral black stripe present), and one black spot is present on the caudal peduncle (vs. spot absent). From the other species of the same group, A. lacustris differs from A. bimaculatus and A. validus by the absence of maxillary teeth (one tooth present on the left side of one of 237 examined specimens) (vs. presence). It further differs from A. validus by the absence of horizontal lines, forming a zigzag pattern on body (vs. presence); and from A. bimaculatus by presenting a markedly concave external surface of the second tooth of the internal series of the premaxilla, with misaligned cusps (vs. teeth having slightly concave surfaces with cusps almost aligned and on the same plane). Astyanax lacustris differs from A. abramis by a reduced number of perforated scales on lateral lines 30–39 (rarely 40 or 41) (vs. 42–48). See “Note on other species of the Astyanax bimaculatus subgroup caudal peduncle spot” below for A. maculisquamis , A. borealis , and A. vittatus , and “Remarks” on Astyanax lacustris for A.orbignyanus .

Description. Morphometric data summarized in Table 1 . Body compressed and high, with greatest body depth in around origin of pelvic fins. Dorsal profile of head straight or slightly concave from region above eye to supraoccipital spine. Dorsal body profile convex from tip of supraoccipital spine to dorsal-fin origin; straight from that point to adipose fin origin. Ventral profile of body convex from mandibular symphysis to anal fin origin and posterodorsally slanted along anal-fin base. Caudal peduncle deep, nearly straight on dorsal and ventral margins.

Snout rounded from margin of upper lip to vertical through anterior nostrils. Head somewhat pointed anteriorly in lateral profile. Mouth terminal, jaw isognathous. Maxilla extending posteriorly to vertical through anterior margin of orbit. Maxilla widened anteroposteriorly.

Two tooth rows in premaxilla; outer row with 4(140) or 5(3) tricuspid teeth, central cusp longer; inner row with five teeth, gradually decreasing in length from first to fifth; usually with five or six cusps on first tooth, five, six or seven cusps on second to fourth teeth and four or five cusps on fifth tooth; central cusp twice as long and broad as other cusps. Maxilla without teeth (one tricuspidate tooth present in the left side of a specimen—MCP 47854- of 237 examined).Four anteriormost dentary large teeth with five, six or seven cusps; remaining four to seven (9 c&s)(4–12 in Garutti &Britski, 2000) teeth smaller, uni- to tricuspid ( Fig.2 ).

Dorsal-fin rays iii (9 c&s) (first very small),8(6), 9(168) or 10(33); second unbranched ray approximately half length of third ray. Distal margin of dorsal fin nearly straight to slightly convex. Dorsal fin origin slightly behind middle of SL. Origin of adipose fin at vertical through base of sixth, seventh or eighth last anal-fin rays. Anal-fin rays iv(6 c&s) or v(5 c&s), 21(5), 22(7), 23(23), 24(39), 25(39), 26(62), 27 (40), 28(36), 29(8), 30(4) or 31(1) (up to 34 in Garutti& Britski, 2000). Anal-fin profile smoothly concave. Anal fin origin posterior to vertical through base of last dorsal-fin ray. Pectoral-fin rays i,11(108), 12(42), 13(1), or 15(2). Pectoral-fin tip reaching or slightly surpassing pelvic-fin insertion; not reaching in specimens smaller than 11.5 mm SL. Pelvic-fin rays i,7. Pelvic-fin origin anterior to vertical through dorsal-fin origin. Pelvic-fin tip does not reach or barely reaches anal-fin origin. Caudal fin forked, lobes similar in size, with i+17+i principal rays, no scales in the lobes. Dorsal procurrent caudalfin rays 7(1), 9(1), 10(2), 11(2) or 12(2), and ventral 7(1), 8(1), 9(4) or 10(2) (8 c&s).

Lateral line complete, with 32(1), 33(26), 34(38), 35(60), 36(65), 37(41), 38(10), 39(9) or 40(2) scales (up to 41 in Garutti & Britski, 2000). Scale rows between dorsal-fin origin and lateral line 6(103), 7(121), 8(26) or9(2); scale rows between lateral line and pelvic-fin origin 4(1), 5(88), 6(115),7(46), 8(13) or 9(2); scale rows between lateral line and anal-fin origin 5(4), 6(36), 7(66), 8(73), 9(74) or 10(7). Scale rows around caudal peduncle 11(1), 12(8), 13(25), 14(65), 15(66),16(60), 17(15), 18(12) or 19(1).

Precaudal vertebrae 16(4), 17 (2) or 18(1); caudal vertebrae 18(5), 19(1) or 20(1); total vertebrae 34(2), 35(3) or 36(2). Supraneurals 4(5)or 5(2). Gillrakers at first arch: upper branch6(7),7(10), 8(47), 9(104), 10(75) or 11(5); lower branch 12(17), 13(94), 14(65), 15(50), 16(19), 17(4) or 18(1); total number 20(6), 21(29), 22(46), 23(71), 24(52), 25(33) or 26(10).

Color in alcohol. Dorsal and dorsolateral portions of head and body dark brown. A dark oval humeral spot horizontally placed, followed by two vertical bars tenuously darkened; one exceeds the oval spot and the other remains two or three scales from it. Dark spot on caudal peduncle conspicuous, extending over median rays of caudal fin; anteriorly, it can reach the vertical through the origin of adipose fin.Body with dark spots at the posterior region of scales forming 2 to 4 horizontal stripes above lateral line and 2 to 5 below ( Figs 3 A, B) (see Remarks). For color of living or freshly collected specimens see Desordi et al. (2011), Venere & Garutti (2011), Malabarba et al.(2013), Serra et al.(2014), and Graça & Pavanelli (2007).

Sexual dimorphism. Male specimenswith hooks on anal and ventral fins. Anal fin with hooks from last unbranched ray (can be absent) until the first five, or until the tenth branched ray; one pair per segment. Hooks are present on all branched rays of the pelvic fin. Gill fusion glands were not found macroscopically on the first gill arch. Téran et al. (2014) have found glandular cells but not fusion on the gill filaments in A. asuncionensis (= A. lacustris ).

Geographic variation. Variation of meristic characters showed ample overlapping after comparison between the drainages. Therefore, they were not useful in proposing diagnoses according to our criteria. This does not mean that we have not found statistically significant differences in the analysis undertaken. For example, Dunn´s method indicates significant differences in the number of total gill rakers and lower branch gill rakers between the populations of rio São Francisco and the other three populations from Paraguai, Paraná and SLPUR (Figs. 4A, B). The last count was significantly different between the rios Paraná and Paraguai and SLPUR (Fig. 4B). The number of gill rakers on the upper branch was significantly different between populations of the Paraná and Paraguai vs. the São Francisco and SLPUR (Fig. 4C). Significant differences were found in the number of branched anal fin rays between the São Francisco and Paraguai and SLPUR (Fig. 4D). The number of perforated scales of the lateral line and of scales around the caudal peduncle was significantly different between the rio São Francisco populations and those of the Paraná and Paraguai. The same occurred between the populations of SLPUR and the Paraguai (Figs. 4D, E). The number of scales around the caudal peduncle was significantly different between SLPUR and Paraná populations (Fig. 4F). Significant differences were found in the number of scales from the lateral line and pelvic fin origin between São Francisco and Paraguai and SLPUR; the same occurred between the Paraná and Paraguai and SLPUR (Fig. 5A). The scale count from lateral line to anal fin origin was significantly different between therios São Francisco, Paraguai and Paraná, and SLPUR; the same count was different between the rios Paraguai and Paraná (Fig. 5B). Significant differences were found in the number of scales from dorsal fin to lateral line between the rio Paraguai and the other three populations (Fig. 5C).

Morphometric data analyzed did not indicate diagnostic differences ( Tab. 1 ), and in an exploratory test PCA did not reveal discriminate populations at all (Fig. 5D).In PCA, the three first axes explained 96.26% variation of the data, but the scree plot between axes indicated that only the first two are informative. The axis PC1 corresponds to the axis of size with all variables showing positive loadings. Axis PC2, however, showed the population of the Paraná partially discriminated from populations of the rios São Francisco and Paraguai, and SLPUR. The most influential variables in this discrimination in the second axis (explaining 2.892% of variance) were diameter of the eye and interorbital width on the positive side, and pre-anal distance on the negative side. The drainages — the rios São Francisco, Paraguai and Sistema da Laguna dos Patos— did not show discrimination between them.

FIGURE 4. Astyanax lacustris populations by river drainages.(A) Number of total gill rakers on the first gill arch in. (B) Number of gill rakers on lower branch of first gill arch. (C) Number of gill rakers on upper branch of first gill arch. (D)Number of branched anal-fin rays. (E) Number of perfored scales of lateral line. (F) Number of scales around caudal peduncle. Tukey box plots: the middle line is the median and lateral lines of the box are 25th and 75th respectively.

FIGURE 5. Astyanax lacustris populations by river drainages (A) Number of scales between pelvic-fin and lateral line in. (B) Number of scales rows between anal-fin and lateral line. (C) Number of scales rows between dorsal-fin and lateral line. Tukey box plots: the lateral lines of the box are 25th and 75th respectively. (D) Projection of individual scores in the space of the first and second Principal Component axis for the populations of Astyanax lacustris .

In specimens, principally from the rio São Francisco drainage, anterior prolongation of caudal spot can be more tapered (contrary to truncated, a more common form in the other drainages), reaching the vertical through nearly half the distance between origin of adipose fin and end of dorsal-fin base ( Figs. 1 A, B). This prolongation, however, sometimes is difficult to visualize, both in its form and scope, suggesting that it may be related to fixation ( Fig. 6 A).

Distribution. Rio La Plata and rioSão Francisco drainages, Sistema da Laguna dos Patos and rio Tramandaí drainage, and according to Garutti & Langeani (2009),in rio Araguaia (Tocantins-Araguaia drainage) ( Fig. 7 A).

Remarks. Tetragonopterus lacustris was described by Lütken based on specimens captured in Santa Lagoon, rio São Francisco, and taken by Reinhardt to the Zoological Museum at the University of Copenhagen. Lütken did not mention the number of specimens examined, but according to M. Krag (pers. comm.), collection manager of this museum, there are 25 specimens catalogued in that museum. Other syntypes are in the Museum für Naturkunde, Berlin, Musée National d'Histoire Naturelle, Paris, and in the Staatliches Museum für Naturkunde, Stuttgart. Photos of all specimens were examined and confirmed as belonging to the A. bimaculatus subgroup with caudal peduncle spot (see type material examined)( Figs. 1 C, D).

Astyanax paraguayensis was described by Fowler (1918) as type species of the new genus Bertoniolus . Géry (1972) synonymized Bertoniolus with Astyanax . To avoid a homonymy with Astyanax (Poecilurichthys) bimaculatus paraguayensis Eigenmann, 1921 , Géry proposed for this species the substitute name A. bimaculatus asuncionensis , maintaining as such the name A. paraguayensis for the Fowler species. Extensive series of specimens were analyzed from the rios La Plata and São Francisco drainages and SLP, as well as the syntypes ( Fig. 6 B). In light of this analysis, and considering that the only characteristic mentioned by Eigenmann to distinguish Astyanax bimaculatus paraguayensis from the others subspecies recognized by him ( lacustris , novae , vittatus , paraguayensis , and borealis ) was “rows of spots along the centers of the scales”, a pattern widely observed in the populations from the studied area ( Figs. 6 C), a synonymization is proposed for the name A. bimaculatus asuncionensis with A. lacustris , the oldest name (see details on coloration further below).

The diagnostic characters of A. altiparanae ( Fig. 8 A) presented by Garutti & Britski (2000), are: “33 to 41scales on the lateral line, 10 to 17 scales on the transversal line, 22 to 34 rays on the anal fin, lower portion of body (when divided by the horizontal line that passes through the oral cleft) greater than the upper portion, 4 to 12 (frequently 6 to 8) small teeth on the dentary; large dentary teeth with five, six, or seven cusps; latero-ventral and ventral region of the trunk silvery, absence of maxillary tooth and black lateral band”. While comparing the new species with A. lacustris , the authors affirm that it possesses “the same uniform coloration as altiparanae on the ventral flank, differing from it by other characters, among which are the approximately equal height between the upper and lower portions of the horizontal line that passes through the oral cleft (vs. lower portion higher than the upper in A. altiparanae ), the large teeth with five or six cusps on the dentary bone (vs. five, six, or seven cusps in A. altiparanae ), and the lesser number of small teeth on the dentary (3 to 6) (vs. 4–12, usually 6–8 teeth in A. altiparanae )”. The last two characters overlap between the two species, and therefore are not considered diagnostic.Regarding the number of small posterior teeth on the dentary, we found the total range of 4–7 with the following distribution: 5–7 in specimens of rio Paraná drainage ( MCP 20339 and MCP 37737); 4–6 in specimens of rio Paraguai drainage ( MCP 39750); 4–5 in specimens of SLP ( MCP 18666); 4–7 in specimens of rio Iguaçu drainage ( MCP 26610),and 4–5 in specimens of rio São Francisco drainage ( MCP 17155 and MCP 16997). The differences in height between the upper and lower portions of the body (considering a horizontal line that passes by the oral cleft, as per the authors) were variable in the material examined. For example, lot MCP 27862 (7, 70.0– 107.0 mm) of the rio Tietê, upper rio Paraná, in which specimens present weak dark spots on the scales that are almost imperceptible on the ventral part of the body, presents a pattern that according to Garutti & Britski (2000) is characteristic of the species A. lacustris and A. altiparanae . We have verified that the upper and lower portions of the body are approximately equal among all specimens, with a tendency for the upper part to be a bit larger. See also the syntypes of Tetragonopterurs lacustris ( Figs. 1 C, D), in which the height of the upper portion is smaller or approximately equal to the ventral portion.

Salgado (2013) indicates characters differentiating A. altiparanae and A. lacustris , which include shape of the distal margin of the anal fin, shape of the margin of the operculum, length of the second ramified ray of the anal fin, and number of cusps on the symphyseal tooth. None of these characters were diagnosed in this study.

Tetragonopterus jacuhiensis was described by Cope (1894) based on various specimens collected in “Rio Grande do Sul” [locality restricted by Fowler (1906) to the rio Jacuí, Sistema da Laguna dos Patos]. In characterizing the species, Cope mentions diameter of orbit (4 times in head length) and number of anal fin rays (28) as main diagnostic characters. Examination of the holotype revealed the orbit diameter occupying 33.2% of the head lenght (about 3 times in head length), a value within the limits of A. lacustris . The same occurred regarding the number of anal fin rays. Although Cope did not mention spots on the scales forming horizontal stripes, examination of the holotype ( Fig. 9 A) and some paratypes showed the presence of spots with weak intensity. The extensive material examined from the type locality, however, confirmed the presence of this character in this species, and highlighted the variability detailed above ( Fig. 9 B).

In reference to the coloration of Astyanax lacustris , although the concentration of chromatophores occurs in the same position on the scales, the intensity of spots and number of stripes may vary independently from the size of each specimen, or from its area of occurrence. Two color patterns are observed: (1) overall body light, silver on the lateroventral portion, with small dark spots on the posterior portion of some scales faded, forming 2 to 3 horizontal stripes above the lateral line, and 2, maximum 3, stripes below the lateral line ( Figs. 1 A; 6B, and 9B); and (2) body darker, especially in the lateroventral region, with conspicuous spots, small on the posterior portion of most scales, forming in conjunction 3 to 4 horizontal stripes above the lateral line and 4, rarely 5 stripes below the lateral line ( Figs.1 B, 6C, and 8B).

Considering the results obtained in the present study and exposed above, and based on the pictures of the type material examined, we conclude that the names Tetragonopterus jacuhiensis Cope, 1894 , Astyanax asuncionensis Géry, 1972 , and A. altiparanae Garutti & Britski, 2000 , are junior synonyms of A. lacustris Lütken, 1875 ( ICNZ, 1999: art. 23).

continued.

Paraná São Francisco

Measurements Range Range Material examined. Types. Brazil: Minas Gerais state: Tetragonopterus lacustris , lagoa Santa, all pictures. ZMUC P241324, about 85.0 mm SL, ZMUC P241322-23 and P241325-28, 6, about 49.0– 92.5 mm. MNHN 9580, 1, about 80.0 mm. NMW 57540, 1, about 100 mm. ZMB 9200, 2, 98.0–99.0 mm SL. SMNS, 2, about 74.0–100.0 mm. São Paulostate: Astyanax altiparanae , rio Grande, paratypes, MZUSP 48180, 32 (10, 63.3–112.8 mm. Rio Grande do Sul state. Tetragonopterus jachuiensis, Jacuhy , holotype ,, ANSP 21912, 95.1 mm,same locality of holotype, paratypes, ANSP 21683-21701, 19 (7, specimens numbers 1 (88.5 mm), 2 (69.7 mmm)and 3( Astyanax (Poecilurichthys) bimaculatus paraguayensis , 67.2 mm), 4(63.9 mm) and 5 (52.7 mmm), 7(48.1 mm) and 14 (32.4 mm). Paraguay: rio Paraguay, syntype, CAS 38987, 9.3 mm, pict and CAS 38998, 70.0 mm, pict.

Non-types. Brazil: Sistema da laguna dos Patos: Rio Grande do Sul state: MCP 18666, 47 (2, 47.6– 65.0mmc&s) arroio on the road Candelária to Chachoeira. MCP 37927 (1, 96, 9 mm) rio das Antas. MCP 33613, 3 (1, 89, 8 mm), rio das Antas. MCP 33584, 4 (1, 54, 8 mm), arroio Jararaca. MCP 33587, 7 (1, 98,1mm), arroio Trajano, rio das Antas. MCP 21459, 12 (6, 68,1–104,9 mm), Lageado Fortaleza. MCP 38805 (1, 112,0 mm), rio das Antas. MCP 37919 (1, 99, 2 mm), rio das Antas. MCP 45787 (1, 66, 8 mm, arroio Corrientes. MCP 8879 (1, 75,9mm), rio Camaquã, at reservoir. MCP 21292, 14 (1, 64,0 mm), Lageado do Gringo ca 2km da UHE Dona Francisca. MCP 9440, 153 (3, 47,7–52,5 mm), pools of rio Vacacaí. MCP 37892 (1, 58,0 mm) rio das Antas. MCP 33585, 3 (1, 83, 2 mm), rio das Antas. MCP 8498 (1, 81, 2 mm), rio dos Sinos. MCP 8971, 2 (1, 81, 6 mm), rio dos Sinos. MCP 21219, 5 (2, 72,2–74,3 mm), arroio on the road to UHE Dona Francisca. MCP 21259, 10 (2, 60,9–71,7 mm), arroio da Gringa. MCP 21306 (2, 46.2–49.5 mm), arroio Corupá. MCP 21225, 23 (1, 47, 1 mm), rio Jacuí. MCP 23847, 17 (1, 68, 4 mm), arroio Faxinal. MCP 13711, 5 (1, 91, 2 mm), rio Jacuí. MCP 37883, 5 (1, 76.0 mm), rapids of Caninana. MCP 9157, 7 (1, 67, 7 mm), beach at Tapes. MCP 17320, 13 (1 50,6 mm), arroio Martins. MCP 33641, 2 (1 72,5 mm CP), rio Taquari, near do mouth of rio Guaporé. MCP 33616 (2, 70,2–80,0 mm), rio Taquari. MCP 21517, 7 (1, 61.2 mm), tributary of Ingaí. MCP 20050 (1, 91, 4 mm), arroio Feitoria. Rio Uruguai drainage: Rio Grande do Sul state: MCP 26807, 26 (9, 50.4–107.5 mm), rio Ibicuí Mirim. MCP 27596 5 (1, 71, 9 mm),arroio Cambará. MCP 14190 (1,62,8 mm), rio Ibicui da Armada. MCP 19222, 12 (1, 68, 6 mm), rio Alegre. MCP 13184 (1, 81, 6 mm), rio Comandaí. MCP 13263, 3 (1, 94, 6 mm), mouth of rio Ijuí-Mirim. MCP 21611, 15 (12, 53,1–92,1 mm) rio Uruguai. MCP 28903 (1, 115, 3 mm), rio Ibicui-Mirim. MCP 22870, 9 (1, 94, 6 mm), rio Piraizinho. MCP 23144, 3 (1, 76, 7 mm), rio Inhacunda. MCP 13225, 7 (2, 96,9–124,1mm), rio Uruguai. MCP 6610, 6 (2, 74,8–86,9 mm), arroio Lageado. MCP 12715, 103 (2, 57,8–103,5 mm), arroio Passo do Alto. MCP 12660, 21 (2, 89,6–120,6 mm), rio Uruguai. MCP 13049, 3 (2, 95,6–144,8 mm), rio Uruguai. MCP 13273, 4, 71.3– 79.2 mm, mouth of rio Ijuí. MCP 27607, 26, 49.3–102.9 mm. Arroio Caraí-Passo. MCP 13089, 3, 113.4– 138.2 mm, rio Comandaí. MCP 6282, 25, 51.7–77.5 mm, arroio na estrada Barra do Guarita/Pinheirinho. Santa Catarina state: MCP 18989, 1, 120.0 mm, rio Uruguai em Itá MCP 18444, 2, 113.4– 113.6 mm, rio Uruguai. MCP 12039, 8 (1, 62,2mm), rio Canoas. MCP 18582, 4 (2, 86,6–89,9 mm), rio Jacutinga. MCP 18839, 6 (1, 131.6 mm) rio Uruguai. MCP 20868, 7 (1, 117,0 mm), arroio da Taipa. Rio Tramandaí drainage: Rio Grande do Sul state: MCP 25322, 4, 72.2–92.6 mm, rio Três Forquilhas. MCP 10090, 4, 68.3–86.5 mm, rio Maquiné near mouth with Lagoa dos Quadros. MCP 28277, 5, 51.5–54.5 mm, rio Maquiné. Rio Paraná drainage: São Paulostate: LBP 3502, 36 (15, 55,4– 104.3 mm), rio Tietê. LBP 3588, 5 (1, 52, 8 mm), rio Tietê LBP 3908, 11.4–35.2 mm, rio Paranapanema. LBP 10682, 5, 26.6–50.5 mm, rio Uberaba. LBP 9754, 15, 34.6–49.3 mm. NUP 11967, 7 (2, 93.9– 95.4 mm), córrego Toledinho. DZSJRP 14082, 128 (2, 70.7–77.1 mm), córrego at bridge near Canavial. DZSJRP 12654, 15, (1, 82.6 mm) córrego do Tapeirão. DZSJRP 14362, 4 (1, 86.6 mm), córrego Água do Silveira. MCP 14429, 11 (4, 84.4–97.9 mm), rio Pardo. MCP 14869, 5 (1, 60.8 mm), lagoon at farm Santa Lúcia. MCP 27862, 18 (8, 78.0– 104.3 mm) rio Piracicaba. MCP 44583 (1, 94.6 mm) rio Paranapanema. MCP 27863, 5 (3, 52.9–59.7 mm), rio Preto. Paraná: NUP 11103, 13 (5, 49.0– 71.8 mm), riacho Caracu. NUP 7950 (3, 55.3–62.7 mm), rio Paracai. MCP 14869, 5 (1, 54.0 mm), lagoon at farm Santa Lúcia. Minas Gerais state: MCP 27871 (1, 8.8 mm), rio Quebra-Anzol. MCP 34675 (2, 70.1–81.7 mm), Ribeirão Coelhos. MCP 27867 (1, 66.9 mm), córrego Bagagem. MCP 27864 (1, 61.4 mm), córrego on the road Davinópolis/Douradoquara. MCP 27861 (1, 69.4 mm), Córrego Grande. MCP 27865 (1, 66.3 mm), ribeirão da Estiva. MCP 27869 (1, 53, 1 mm), Córrego na estrada Argenita/ Pratinha. MCP 34682 (1, 49.5 mm), Rio Jacaré. Goiás: MCP 42947 (1, 117.0 mm), Ribeirão João Leite. MCP 42987 (1, 75.9 mm CP), Córrego das Pedras. MCP 34256, 2 (1, 70.6 mm), rio Meia-Ponte. MCP 20339, 56 (12, 61.4–71.6 mm, 2, 59.3–60.8 mm c&s) rio Corumbá. MCP 27870 (2, 60,7–61,1 mm), rio Paranaíba. MCP 42977 (1, 56.7 mm) Córrego Barreiro. Rio Iguaçu drainage: Paraná state: MCP 25603 18 (12, 54.1–68.4 mm) rio Iguaçú. MCP 25610, 64 (4, 59.5–73.3 mm, 2, 58.4–63.4 mm c&s), rio Lageado Grande. MCP 25609, 35 (6, 76.2–92.1 mm) rio Capanema. NUP 4342, 26 (2, 61.8–63.9 mm) Salto Santiago dam. NUP 1313, 19 (2, 61.2–72.9 mm), Caxias dam. NUP 1623 100 (2, 50.7–57.6 mm), reservatório Caxias. NUP 4126, 8 (2, 40.3– 40.2 mm), Santo Osóriodam. NUP 7368, 6 (4, 54,6–85,9 mm) córrego Brasília. Rio Paraguai drainage: MatoGrosso state: CPUFMT 118, 9 (7, 50.7–77,0 mm), flooded area on roadside betweenSanto Antonio do Leverger and Barão de Melgaço. CPUMFT 951, 1, 60.5 mm, rio Itiquira. MCP 48430, 2, 38.0–39.0mm, rio Sangradouro. MCP 32464 (2, 50.9–59.7 mm), córrego Piraputanga. MCP 48421 (2, 53.7–74.2 mm), córregoon the road Barra dos Bugres – Cáceres. MCP 39279, 16 (4, 57.4–66.1 mm), córrego on the transpantaneira road. MCP 48429, 1, 49.8 mm, ribeirão Figueira. MCP 39590, 34 (8, 50.3–72.1 mm),córrego on the road BR-070. MCP 48428, 1, 59.8 mm, córregoon the road MT-124. MCP 37737, 226, 37.7–69.4mm, 42, 52.3–68.5 mm c&s),ribeirão Figueira. MCP 8338, 5 (4, 71.9–84.8 mm), rio Pixaim. MCP 39665, 2, 56.0–65.0 mm SL, tributary of rio Caeté, drainage of rio Jauru. Rio SãoFrancisco drainage: Minas Gerais state: MCP 16935, 21 (1, 73.7 mm), rio Poções. MCP 17154, 150 (5, 51.7–60.7 mm), rio Verde Pequeno. MCP 16997, 28 (10, 57.8–62.8 mm), stream near Montes Claros. MCP 16930, 14 (4, 48.7–57.5 mm), rio Peru-Açu. MCP 17155, 44, 30.4–81.6 mm, stream near Januária. MCP 16970, 2, 49.9–68.7 mm, córrego at São João das Missões. MCP 27810, 6 (2, 56.9–71.1 mm), stream near Claro de Minas. MCP 34307, 5, 24.6–38.4 mm, córrego Olhos d´água. MCP 33748, 5, 33.1–37.1 mm, rio Peru-Açu. MCP 27807, 2, 19.1–52.3 mm, rio Pirapitinga. MCP 44767 (1, 82.3 mm), Ribeirão Neco. Bahia sate: MCP 16863, 18, 43.4–56.6 mm, rio Tatu. MCP 16924, 8 (2, 68.3–72.6 mm), Riacho Santana. MCP 17098, 21 (1, 47.0 mm), rio da Olaria. Pernambuco state: MCP 31014, 75 (4, 55.5–56.0 mm), rio Pageu. MCP 31192, 714 (1, 60.0 mm), stream on the road BR-316. Alagoas state: MCP 31471, 105, 37.5–62.7mm, tributary of rio Capiá.

Figures

FIGURE 1. (A) Astyanax lacustris, MCP 16935, 73.7 mm SL, rio Poções, rio São Francisco drainage. (B) Astyanax lacustris, MCP 16930, 55.4 mm SL, rio Peru-açu, rio São Francisco drainage. (C) Tetragonopterus lacustris, ZMUC P 241324, 87.5 mm SL, syntype, rio das Velhas at Lagoa Santa. Copyright by the Zoological Museum of University of Copenhagen, all rights reserved. (D) Tetragonopterus lacustris, NMW 57540, syntype, circa 100 mm SL, rio das Velhas at Lagoa Santa. Copyright by the Naturhistorisches Museum Wien, all rights reserved.

FIGURE 2. Astyanax lacustris, MCP 20339, 60.8 mm SL, Goiás, rio Corumbá; jaws and dentition, lateral view, left side.

FIGURE 3. (A) Astyanax altiparanae, parátipo, MZUSP 48180, 108.6 mm SL. Rio Grande, upper rio Paraná drainage. (B) Astyanax lacustris, MCP 37737, 66.9 mm SL, ribeirão Figueira, rio Paraguai drainage. Lateral view of the body showing the position of the concentration of chromatophores.

FIGURE 6. (A) Astyanax lacustris, MCP 15706, 97.8 mm SL, unnamed stream, rio Paraguai drainage. (B) Astyanax asuncionensis, CAS 38987, syntype, 90.3 mm SL, Asunción. Copyright by the Academy of Natural Sciences, California, all rights reserved. (C) Astyanax lacustris, MCP 37737, 66.9 mm SL, ribeirão Figueira, rio Paraguai drainage.

FIGURE 7. Geographic distribution of the species (A) Astyanax lacustris, red stars (Type-localities: black stars = Tetragonopterus jacuhiensis; green stars = Tetragonopterus lacustris; purple stars = Astyanax (Poecilurichthys) bimaculatusparaguayensis; blue stars = Astyanax altiparanae). (B) Astyanax abramis, blue stars (Type localities: red stars: Bertoniolus paraguayensis; green stars = Tetragonopterusabramis). Some symbols represent more than one locality.

FIGURE 8. (A) Astyanax altiparanae, paratype, MZUSP 48180, 108.6 mm SL. rio Grande, upper rio Paraná drainage. (B) Astyanax lacustris, DZSJRP 14362, 70.8 mm SL, rio Paranapanema, upper rio Paraná drainage.

FIGURE 9. (A) Tetragonopterus jacuhiensis. ASNP 21912, holotype, 95.1 mm SL, rio Jacuí. Copyright by the Academy of Natural Sciences, Philadelphia, all rights reserved. (B) Astyanax lacustris, MCP 13711, 90.0 mm SL, rio Jacuí, Sistema da Laguna dos Patos.

Tables

TABLE 1. Morphometric data of Astyanax lacustris according to the drainages studied. N = number of specimens measured; SD = standard deviation.

	SLPUR				Paraguai
Measurements	Range				Range
N	Min Max	Mean	SD	N	Min Max	Mean	SD
Standard length (mm) 41	46.2 112.0	77.1	-	39	47.43 84.85	62.42	-
Percent of Standard Length
Body depth 73	37.3 47.1	42.0	2.242	39	41.0 53.0	45.3	3.13
Predorsal length 73	47.9 53.9	51.1	1.347	39	49.1 56.0	53.0	1.27
Prepectoral length 73	24.8 30.4	27.2	1.212	39	26.0 30.0	27.8	0.90
Preanal length 73	59.9 68.9	65.3	1.697	39	63.3 73.1	68.2	2.44
Prepelvic length 73	41.0 51.1	47.1	1.782	39	47.1 52.6	49.5	1.25
Dorsal-fin length 72	25.2 31.2	28.2	1.534	38	23.7 32.1	28.9	1.52
Pectoral-fin length 41	18.9 28.3	22.6	1.540	39	18.1 26.4	23.3	1.35
Pelvic-fin length 41	14.1 21.5	18.4	1.543	39	13.1 23.4	17.9	1.56
Anal-fin base length 73	27.2 34.6	31.6	1.644	38	27.6 35.4	31.6	2.73
Caudal peduncle length 73	10.0 16.1	13.3	1.306	39	10.0 14.9	12.1	1.21
Caudal peduncle depth 73	10.7 13.0	12.1	0.652	39	10.6 13.0	12.0	0.62
Head length 73	23.2 28.3	25.3	1.023	39	25.8 30.3	27.7	1.03
Percent of Head Length
Snout length 73	25.0 35.4	29.8	2.265	39	22.0 36.0	27.0	2.98
Interorbital width 73	35.4 41.8	38.6	2.682	39	27.8 37.6	34.5	2.07
Horizontal orbit iameter 73	28.3 44.7	34.2	3.831	39	28.8 43.0	36.4	2.82
Upper jaw length 73	36.0 49.6	41.5	2.323	39	32.0 44.8	38.2	2.33