Molecular phylogenetics, a new classification, and a new genus of the Neotropical fish family Anostomidae (Teleostei: Characiformes)

Brian L. Sidlauskas^1,2,3 , Bruno F. Melo⁴, José L. O. Birindelli⁵, Michael D. Burns¹, Benjamin W. Frable^1,6, Kendra Hoekzema^1,7, Casey B. Dillman^2,8, Mark H. Sabaj⁹ and Claudio Oliveira¹⁰

PDF: EN XML: EN | Supplementary: S1 S2 S3 S4 | Cite this article

Abstract

Phylogenetic reconstruction of the Neotropical freshwater fish family Anostomidae using multilocus sequence data from 97 species supports the recognition of three subfamilies: Leporellinae with one genus, Anostominae with six genera, and Leporininae with nine genera. We reassign many Leporinus species to a redefined Hypomasticus and allocate Leporinus striatus to a new monotypic genus sister to Abramites and Megaleporinus. These taxonomic changes clarify that section of the anostomid tree and partially solve the longstanding non-monophyly of Leporinus. Though many relationships inferred herein match earlier morphological hypotheses, the subfamily Anostominae appears as the unambiguous sister to Leporellus,not Laemolyta, indicating that the superior mouths of Anostominae and Laemolyta evolved convergently. Several other clades have converged on subterminal or inferior mouths, including lineages within Hypomasticus, Leporellus, Leporinus, and Schizodon. This largest-ever phylogeny for Anostomidae will support further taxonomic research and provide a scaffold for morphological, biogeographic, and evolutionary studies in this transcontinental group of Neotropical fishes.

Keywords: Characoidei, Headstanders, Leporinus, Ostariophysi, South America.

A reconstrução filogenética da família Neotropical de peixes de água doce Anostomidae usando dados de sequência multilocus de 97 espécies apoia o reconhecimento de três subfamílias: Leporellinae com um gênero, Anostominae com seis gêneros e Leporininae com nove gêneros. Nós reatribuímos muitas espécies de Leporinus a um Hypomasticus redefinido e alocamos Leporinus striatus a um novo gênero monotípico irmão de Abramites e Megaleporinus. Essas mudanças taxonômicas esclarecem essa seção da árvore dos anostomídeos e resolvem parcialmente a não monofilia de longa data de Leporinus. Embora muitas relações inferidas aqui correspondam a hipóteses morfológicas anteriores, a subfamília Anostominae aparece como a irmã inequívoca de Leporellus, e não Laemolyta, indicando que as bocas superiores de Anostominae e Laemolyta evoluíram convergentemente. Vários outros clados convergiram em bocas subterminais ou inferiores, incluindo linhagens dentro de Hypomasticus, Leporellus, Leporinus e Schizodon. Esta maior filogenia de todos os tempos para Anostomidae dará suporte a mais pesquisas taxonômicas e fornecerá um arcabouço para estudos morfológicos, biogeográficos e evolutivos neste grupo transcontinental de peixes neotropicais.

Palavras-chave: América do Sul, Characoidei, Leporinus, Ostariophysi, Piaus.

Introduction

With 146 known species, Anostomidae Günther, 1864 represents the second largest family within the hyperdiverse fish order Characiformes and a major component of the Neotropical ichthyofauna (Sidlauskas, Birindelli, 2018; Toledo-Piza et al., 2024). The family achieves greatest diversity in the Amazon, Orinoco and Paraná-Paraguay systems and the northward draining river systems of the Guianas (Sidlauskas, Vari, 2012; Sidlauskas, Birindelli, 2018). Some anostomid species inhabit trans-Andean drainages from Peru to Panama, and others occur in isolated coastal drainages of eastern Brazil (Sidlauskas, Birindelli, 2018). Estimates for the family’s crown clade age are fairly recent, falling in the Eocene or Oligocene and ranging from 42 to 25 million years (Sidlauskas et al., 2021; Melo et al., 2022). The recency of those dates implies that the clade has generated many species in a brief timeframe, and at a rate surpassing that of most other characiform families (Melo et al., 2022). Given that scientists still describe about two new anostomid species per year (Sidlauskas, Birindelli, 2018; Toledo-Piza et al., 2024), the family’s true species diversity and rate of diversification may even exceed the current estimates.

Anostomids span substantial ecomorphological diversity (Sidlauskas, 2008; Sidlauskas, Vari, 2008; Fig. 1). They typically occupy omnivorous, invertivorous, or herbivorous niches, and have diversified oral jaws and dentition to accommodate their varied diets (Géry, 1961; Sidlauskas, 2007, 2008; Lofeu et al., 2021). Ichthyologists have long noted their remarkable variability in mouth position (Myers, 1950; Géry, 1972, 1977), with the family’s name recognizing the superior mouths (ano + stoma) possessed by some species, including Anostomus anostomus (Linnaeus, 1758). Other members of the family possess terminal, subterminal or inferior mouths (Géry, 1961; Sidlauskas, 2008), with some species shifting in jaw morphology and orientation markedly throughout ontogeny (Sidlauskas et al., 2007; Birindelli, Britski, 2009) and others demonstrating exceptional plasticity in the eventual adult form (Bonini-Campos et al., 2019; Lofeu et al., 2021). Members of the family possess single row of large teeth that have diversified into spade-shaped, crenulate, incisiform, peg-shaped and even clawlike forms (Géry, 1961; Sidlauskas, Vari, 2008). The family’s diversity in color pattern is similarly riotous, with most species displaying prominent bars, blotches, bands, stripes, and black or red spots that may serve a role in species signaling and mate recognition (Géry, 1977; Sidlauskas, Vari, 2008; Ramirez et al., 2020).

FIGURE 1| The diversity of anostomid fishes. First column from top: Anostomus ternetzi, MZUSP 97271, 55 mm SL, Brazil: Pará: Novo Progresso: rio Jamanxim; Synaptolaemus latofasciatus, AUM 43269, 79 mm SL, Venezuela: Amazonas: Río Orinoco; Sartor respectus, ANSP 198098, 115 mm SL, Brazil: Pará: rio Xingu; Gnathodolus bidens, LIA 1313, SL unknown, Brazil: Pará: Xadai, above São Félix do Xingu. Second column from top: Leporellus vittatus, ANSP 193001, 220 mm SL, Brazil: Pará: rio Iriri (Xingu drainage); Schizodon trivittatus, ANSP 194505, 238 mm SL, Brazil: Pará: purchased at market on tributary of rio Xingu; Rhytiodus microlepis, INPA 43220, SL unknown, Brazil: Pará: rio Xingu; Abramites hypselonotus, MZUEL 13996, 108 mm SL, Brazil: Pará: rio Xingu. Third column from top: Leporinus fasciatus, MZUEL 14698, 170 mm SL, Brazil: Amazonas: Rio Negro; Megaleporinus obtusidens, ANSP 203101, 225 mm SL, Uruguay: río Negro: río Uruguay, Hypomasticus torrenticola, ANSP 193014, 74 mm SL, Brazil: Pará: rio Xingu; Leporinus julii, ANSP 193054, 113 mm SL, Brazil: Pará: rio Xingu (Volta Grande). All photos taken of recently euthanized specimens by Mark Sabaj except Rhytiodus microlepis and Gnathodolus bidens by Leandro Sousa, Leporinus fasciatus by José Birindelli, and Abramites hypselonotus by Fernando Jerep. Images used with permission of the photographers.

Despite the family’s prominence, diversity, and biological interest, their phylogenetic relationships were almost entirely opaque until Winterbottom (1980) conducted a cladistic analysis on the subfamily Anostominae Günther, 1864, which contains the species with the most strongly upturned mouths. No further advances about intergeneric relationships occurred until Sidlauskas, Vari (2008) used a morphological dataset of 152 primarily osteological characters to frame a hypothesis of relationships and highlight the non-monophyly of the family’s largest genus Leporinus Agassiz, 1829. Bogan et al. (2012) expanded that dataset to include cranial characters of the fossil †L. scalabrinii (Ameghino, 1898) from the Miocene Ituzaingó Formation of Entre Ríos, Argentina. Dillman et al. (2016) formally combined that matrix with morphological data from Chilodontidae Eigenmann, 1903, Prochilodontidae Eigenmann, 1909 and Curimatidae Gill, 1858 in a supermatrix analysis. Due to the similarity in datasets, the results of these two studies return the same set of intergeneric relationship as did the original morphological phylogeny (Sidlauskas, Vari, 2008).

Though early molecular phylogenies of characiform fishes often included a handful of anostomid species (Calcagnotto et al., 2005; Oliveira et al., 2011; Melo et al., 2014), not until Ramirez et al. (2017b) did any team complete a thorough molecular study of the family. That study used two mitochondrial and three nuclear genes to reconstruct relationships among 44 species and break off a portion of Leporinus into a newly recognized genus of large-bodied species: Megaleporinus Ramirez, Birindelli & Galetti, 2017. The molecular study returned intergeneric relationships mostly like the morphological phylogeny (Sidlauskas, Vari, 2008) but with one fundamental disagreement: the morphological phylogeny reconstructed the distinctive subfamily Anostominae as sister to Laemolyta Cope, 1872 a genus whose members also possess superior mouths (Sidlauskas, Vari, 2008) particularly in young specimens. The molecular results separate those two clades substantially, suggesting independent evolution of upturned mouths (Ramirez et al., 2017b).

Other studies using molecular data have generally confirmed the intergeneric molecular framework or have expanded knowledge of interspecific relationships within subsets of Hypomasticus Borodin, 1929, Laemolyta, Leporinus, and Schizodon Agassiz, 1829 (Ramirez, Galetti Jr., 2015; Burns et al., 2017; Ramirez et al., 2017a, 2020; Birindelli et al., 2020b; Garavello et al., 2021). Two studies that deployed genome-scale data to reconstruct intrageneric relationships among the enclosing order Characiformes included several anostomid species, with results similar to those obtained with Sanger-sequenced data (Betancur-R. et al., 2019; Melo et al., 2022). Sidlauskas et al. (2021) combined molecular and morphological data in a total evidence framework and revealed a new genus, Insperanos Assega, Sidlauskas & Birindelli, 2021 that includes the only known living member of a lineage that diverged near the Eocene-Oligocene boundary and that demonstrates a unique combination of anatomical characteristics.

Though much has been learned from these disparate phenotypic and genetic datasets, the most densely sampled anostomid phylogenies to date (Ramirez et al., 2017b; Sidlauskas et al., 2021) have included no more than a third of the family’s known species richness. In addition, previous molecular phylogenies obtained weak support for several genus- and species-level relationships within Anostomidae (Ramirez et al., 2016, 2017b; Sidlauskas et al., 2021). Thus, until a new, more densely sampled phylogeny becomes available, comparative studies in Anostomidae will lack a crucial evolutionary scaffold. Herein, we use multilocus DNA sequence data to fill those gaps and present the most completely sampled phylogeny of Anostomidae to date.

Material and methods

Taxon sampling. The phylogenetic reconstruction herein spans 204 specimens representing 97 of the 146 known anostomid species (66.4% taxon coverage) and 16 outgroup species (Tab. 1). Museum codes for voucher specimens follow Sabaj (2020). Outgroup sequences come from previous studies (Melo et al., 2014, 2016, 2018). Most of the ingroup samples represent newly sequenced specimens, though we incorporated some anostomid sequences used previously as outgroups in studies of related families (Melo et al., 2014, 2016, 2018), the data from Burns et al.’s (2017) study of the Leporinus desmotes Fowler, 1914 species complex, and some sequences from Ramirez et al. (2017b) of taxa that were not otherwise represented in our dataset. One specimen of Leporinus brunneus Myers, 1950 (ANSP 192149) analyzed by Burns et al. (2017) was accidentally duplicated in the final matrix. All anostomid genera are included except for the recently described, highly divergent and monotypic Insperanos,which has proven difficult to amplify for most loci (Sidlauskas et al., 2021). The outgroup taxa include several species from the three families most closely related to Anostomidae: Chilodontidae (3 species), Curimatidae (10 species) and Prochilodontidae (2 species). We also included one species of Parodontidae Eigenmann, 1910 (Parodon nasus Kner, 1859) to root the phylogeny. The monophyly of a clade containing Anostomidae, Chilodontidae, Curimatidae and Prochilodontidae has been well-established on morphological (Vari, 1983; Dillman et al., 2016) and molecular grounds (Melo et al., 2014, 2022; Faircloth et al., 2020), and recent phylogenomic studies have recovered Parodontidae as the sister group to these four families (Betancur-R. et al., 2019; Melo et al., 2022).

TABLE 1 | Voucher specimens sequenced in this study.

Species	Museum & Catalog Num.	Tissue	River basin	Coordinates	State	Country
Abramites hypselonotus	ANSP 178126	1712	Río Napo, Amazonas	3°29’10”S 73°6’24”W	Loreto	Peru
Abramites hypselonotus	AUM 53775	T08985	Río Tucupido, Orinoco	8°54’50.8”N 69°45’40”W	Portuguesa	Venezuela
Anostomoides atrianalis	AUM 53813	T09941	Río Caura, Orinoco	7°02’37.6"N 64°57’40.8"W	Bolívar	Venezuela
Anostomoides atrianalis	FMNH 123875	T45	Río Nanay, Amazonas	3°46’45.1"S 73°22’05.9"W	Loreto	Peru
Anostomoides atrianalis	UFRO-ICT 008702	3398 67001	Rio Aripuanã, Madeira	5°08’37.6"S 60°22’53.6"W	Amazonas	Brazil
Anostomus anostomus	USNM 402897	GY11-2-08	Cuyuni River, Essequibo	6°53’1.6"N 60°14’52.4"W	Mazaruni	Guyana
Anostomus anostomus	USNM 402905	GY11-3-52	Cuyuni River, Essequibo	6°48’12.3"N 60°5’34.3"W	Mazaruni	Guyana
Anostomus ternetzi	AUM 43551	P4696	Río Siapa, Casiquiare	1°26’51.0"N 65°38’55.0"W	Amazonas	Venezuela
Anostomus ternetzi	AUM 53596	T09714	Río Parucito, Orinoco	5°20’51.7"N 66°01’24.3"W	Amazonas	Venezuela
Anostomus ternetzi	LBP 4375	24146	Rio Branco	2°18’02.0"N 60°55’20.7"W	Roraima	Brazil
Anostomus ternetzi	MZUSP 97271	7163	Rio Jamanxim, Tapajós	8°11’04"S 055°21’28"W	Pará	Brazil
Brevidens striatus	LBP 3180	16871	Rio Paranapanema	23°19’56.3"S 48°38’28.4"W	São Paulo	Brazil
Brevidens striatus	LBP 14028	58363	Rio Cuiabá, Paraguai	17°49’37.8"S 57°22’53.4"W	Mato Grosso	Brazil
Caenotropus labyrinthicus	OS 18770	PE10-082	Río Nanay, Amazonas	3°45’06.0"S 73°17’14.0"W	Loreto	Peru
Caenotropus mestomorgmatos	ANSP 180516	T48	Río Nanay, Amazonas	3°46’45.0"S 73°22’06.0"W	Loreto	Peru
Chilodus punctatus	OS 18781	PE10-100	Río Nanay, Amazonas	3°45’06.0"S 73°17’14.0"W	Loreto	Peru
Curimata cyprinoides	MHNG 2705.006	SU07-217	Cuyuni River	2°24’22.1"N 56°55’36.5"W	Sipaliwini	Suriname
Curimatopsis macrolepis	ANSP 178188	1697	Río Nanay	3°45’23.0"S 73°17’28.0"W	Loreto	Peru
Curimatopsis myersi	LBP 14006	58310	Rio Paraguai	17°49’26.7"S 57°31’03.0"W	Mato Grosso	Brazil
Cyphocharax corumbae	LBP 17244	68909	Rio Corumbá, Paraná	17°42’03.4"S 48°33’18.4"W	Goiás	Brazil
Cyphocharax gilbert	LBP 8343	40130	Rio Mucuri	17°41’42.4"S 40°46’11.3"W	Minas Gerais	Brazil
Cyphocharax modestus	LBP 8362	40165	Rio Araras, Paraná	22°22’42.4"S 47°25’37.9"W	São Paulo	Brazil
Gnathodolus bidens	AUM 53991	T09892	Río Orinoco	4°55’04.0"N 67°49’58.6"W	Amazonas	Venezuela
Hypomasticus copelandi	LBP 8098	37533	Rio Mucuri	17°41’42.4"S 40°46’11.3"W	Minas Gerais	Brazil
Hypomasticus copelandi	LBP 8098	37536	Rio Mucuri	17°41’42.4"S 40°46’11.3"W	Minas Gerais	Brazil
Hypomasticus despaxi	ANSP 189010	6866	Lawa River, Marowijne	3°19’31.0"N 54°03’48.0"W	Sipaliwini	Suriname
Hypomasticus despaxi	ANSP 189010	6988	Lawa River, Marowijne	3°19’31.0"N 54°03’48.0"W	Sipaliwini	Suriname
Hypomasticus despaxi	MHNG 2716.080	SU08-196	Tapanahony River	3°21’57.6"N 55°25’55.6"W	Sipaliwini	Suriname
Hypomasticus despaxi	MHNG 2718.021	SU08-641	Paloemeu River, Marowijne	3°10’42.0"N 55°25’09.1"W	Sipaliwini	Suriname
Hypomasticus granti	MHNG 2718.023	SU08-653	Paloemeu River, Marowijne	3°10’42.0"N 55°25’09.1"W	Sipaliwini	Suriname
Hypomasticus lebaili	ANSP 189043	6864	Lawa River, Marowijne	3°19’31.0"N 54°03’48.0"W	Sipaliwini	Suriname
Hypomasticus lebaili	ANSP 189043	6873	Lawa River, Marowijne	3°19’31.0"N 54°03’48.0"W	Sipaliwini	Suriname
Hypomasticus lineomaculatus	LBP 5391	27083	Rio Iratapuru, Jari	0°33’51.0"S 52°34’45.0"W	Amapá	Brazil
Hypomasticus lineomaculatus	LBP 5391	27084	Rio Iratapuru, Jari	0°33’51.0"S 52°34’45.0"W	Amapá	Brazil
Hypomasticus megalepis	USNM 402921	GY11-4-33	Cuyuni River, Essequibo	6°47’49.6"N 59°59’33.9"W	Mazaruni	Guyana
Hypomasticus megalepis	USNM 402923	GY11-1-78	Cuyuni River, Essequibo	6°53’1.6"N 60°14’52.4"W	Mazaruni	Guyana
Hypomasticus melanostictus	LBP 5179	26615	Rio Jari	0°37’10.0"S 52°30’48.0"W	Pará	Brazil
Hypomasticus mormyrops	LBP 6448	29070	Rio Paraíba do Sul	23°22’26.2"S 46°03’10.6"W	São Paulo	Brazil
Hypomasticus mormyrops	LBP 8103	37539	Rio Paraíba do Sul	21°14’07.4"S 43°30’50.5"W	Minas Gerais	Brazil
Hypomasticus mormyrops	LBP 8103	37540	Rio Paraíba do Sul	21°14’07.4"S 43°30’50.5"W	Minas Gerais	Brazil
Hypomasticus sp.	LBP 5147	26274	Rio Tapajós	13°37’02.0"S 58°00’50.0"W	Mato Grosso	Brazil
Hypomasticus steindachneri	MCNIP 0379	L297	Rio Itacambiruçu, Jequitinhonha	16°36’24.0"S 42°49’46.0"W	Minas Gerais	Brazil
Hypomasticus thayeri	LBP 2380	16074	Rio Paraíba do Sul	22°00’00.0"S 41°20’00.0"W	Rio de Janeiro	Brazil
Hypomasticus thayeri	LBP 10727	49707	Rio Paraíba do Sul	22°04’07.8"S 41°54’36.2"W	Rio de Janeiro	Brazil
Hypomasticus thayeri	LBP 18618	49708	Rio Paraíba do Sul	22°04’07.8"S 41°54’36.2"W	Rio de Janeiro	Brazil
Hypomasticus torrenticola	ANSP 194410	t0810	Rio Xingu, Amazonas	3°16’41.6"S 52°02’18.4"W	Pará	Brazil
Hypomasticus torrenticola	ANSP 194751	t0519	Rio Xingu, Amazonas	3°21’56.4"S 51°43’36.5"W	Pará	Brazil
Laemolyta fernandezi	GEPEMA 5598	L130	Rio Araguaia	15°53’42.4"S 52°15’16.5"W	Goiás	Brazil
Laemolyta fernandezi	LBP 4003	23066	Rio Araguaia	11°40’09.0"S 50°51’00.3"W	Mato Grosso	Brazil
Laemolyta garmani	OS 18777	PE10-089	Río Nanay, Amazonas	3°45’06.0"S 73°17’14.0"W	Loreto	Peru
Laemolyta garmani	OS 18777	PE10-098	Río Nanay, Amazonas	3°45’06.0"S 73°17’14.0"W	Loreto	Peru
Laemolyta proxima	OS 18778	PE10-092	Río Nanay, Amazonas	3°45’06.0"S 73°17’14.0"W	Loreto	Peru
Laemolyta proxima	OS 18778	PE10-094	Río Nanay, Amazonas	3°45’06.0"S 73°17’14.0"W	Loreto	Peru
Laemolyta proxima	OS 18778	PE10-097	Río Nanay, Amazonas	3°45’06.0"S 73°17’14.0"W	Loreto	Peru
Laemolyta proxima	UFRO-ICT uncat	3310 or 67058	Rio Cautario, Madeira	~12°11’57.7"S 64°28’21.4"W	Rondônia	Brazil
Leporellus vittatus	AUM 54212	T09912 or T09913	Río Cataniapo, Orinoco	5°32’01.5"N 67°22’26.2"W	Amazonas	Venezuela
Leporellus vittatus	ANSP 182609	P6322	Río Nanay, Amazonas	3°42’49.0"S 73°16’43.0"W	Loreto	Peru
Leporellus vittatus	ANSP 182218	P6295	Río Amazonas	3°43’21.0"S 73°12’14.0"W	Loreto	Peru
Leporellus vittatus	LBP 1669	11255	Rio Paranapanema	22°56’21.2"S 50°15’08.3"W	Paraná	Brazil
Leporellus vittatus	LBP 10418	48957	Rio São Francisco	17°21’00.7"S 44°57’08.4"W	Minas Gerais	Brazil
Leporinus affinis	LBP 12699	43695	Rio Araguaia	13°19’S 50°37’W	Mato Grosso	Brazil
Leporinus agassizii	LBP 6857	32703	Rio Negro	0°49’00.0"S 62°49’00.0"W	Amazonas	Brazil
Leporinus agassizii	MHNG uncat.	SU07-065	Kwamalasamutu, Sipaliwini	2°20’53.9"N 56°49’45.8"W	Sipaliwini	Suriname
Leporinus agassizii	OS 18369	PE10-073	Río Nanay, Amazonas	3°45’06.0"S 73°18’58.5"W	Loreto	Peru
Leporinus agassizii	OS 18775	PE10-090	Río Nanay, Amazonas	3°45’06.0"S 73°17’14.0"W	Loreto	Peru
Leporinus agassizii	USNM 402629	GY11-5-31	Cuyuni River, Essequibo	6°41’30.7"N 59°34’38.0"W	Mazaruni	Guyana
Leporinus agassizii	USNM 403520	GY11-3-10	Cuyuni River, Essequibo	6°50’41.6"N 60°7’47.6"W	Mazaruni	Guyana
Leporinus altipinnis	LBP 4459	24381	Rio Negro	0°40’03.1"S 62°58’23.5"W	Amazonas	Brazil
Leporinus amazonicus	UFRO-ICT 004288	7046 67060	Igarapé Jatuarana II, Madeira	8°38’43.5"S 63°54’56.6"W	Rondônia	Brazil
Leporinus amblyrhynchus	LBP 3917	21828	Rio Paranapanema	23°08’01.1"S 49°40’34.4"W	São Paulo	Brazil
Leporinus amblyrhynchus	LBP 3917	21829	Rio Paranapanema	23°08’01.1"S 49°40’34.4"W	São Paulo	Brazil
Leporinus arimaspi	ANSP 180317	T32	Río Nanay, Amazonas	3°46’45.0"S 73°22’06.0"W	Loreto	Peru
Leporinus arimaspi	AUM 53659	T09705	Río Manapiare, Orinoco	5°20’13.7"N 66°03’05.3"W	Amazonas	Venezuela
Leporinus arimaspi	OS 18327	PE10-158	Río Nanay, Amazonas	3°46’51.5"S 73°21’50.0"W	Loreto	Peru
Leporinus bleheri	MZUSP 113988	L039	Rio Guaporé, Madeira	14°50’46.1"S 60°03’58.5"W	Mato Grosso	Brazil
Leporinus britskii	LBP 14158	59201	Rio Tapajós	4°33’45.4"S 56°15’36.9"W	Pará	Brazil
Leporinus britskii	LBP 14214	59375	Rio Tapajós	4°47’24.1"S 56°46’39.5"W	Pará	Brazil
Leporinus brunneus	AUM 54394	T09305	Río Ventuari, Orinoco	3°58’42.3"N 67°03’37.7"W	Amazonas	Venezuela
Leporinus brunneus	AUM 53512	T09239	Río Orinoco	4°23’03.1"N 67°46’29.0"W	Amazonas	Venezuela
Leporinus brunneus	ANSP 192149	7575	Río Orinoco	4°23’03.0"N 67°46’29.0"W	Amazonas	Venezuela
Leporinus cylindriformis	LBP 15940	65660	Rio Coluene, Xingu	13°29’41.8"S 53°04’57.7"W	Mato Grosso	Brazil
Leporinus desmotes	ANSP 179650	2068	Rupununi River, Essequibo	3°51’44.0"N 59°17’04.0"W	Upper Takutu-Upper Essequibo	Guyana
Leporinus ecuadorensis	ROM 93046	T13726	Río Quevedo, Guayas	1°01’20"S 79°27’50"W	Los Rios	Ecuador
Leporinus ecuadorensis	ROM 93046	T13727	Río Quevedo, Guayas	1°01’20"S 79°27’50"W	Los Rios	Ecuador
Leporinus enyae	AUM 43700	V5273 or V5274	Río Casiquiare	2°09’20.5"N 66°27’49.6"W	Amazonas	Venezuela
Leporinus fasciatus	ANSP 180321	NS61	Río Nanay, Amazonas	3°45’09.0"S 73°17’00.0"W	Loreto	Peru
Leporinus fasciatus	AUM 43300	P4315	Río Orinoco	3°06’01.0"N 66°27’46.0"W	Amazonas	Venezuela
Leporinus fasciatus	LBP 4248	22701	Rio Juruá	7°39’15.3"S 72°40’36.2"W	Acre	Brazil
Leporinus fasciatus	MHNG 2621.070	SU01-118	Nickerie River	4°45’00.2"N 56°52’34.0"W	Sipaliwini	Suriname
Leporinus fasciatus	MHNG 2717.030	SU08-342	Paloemeu, Marowijne	3°11’54.0"N 55°24’27.0"W	Sipaliwini	Suriname
Leporinus friderici	ANSP 189264	7015	Lawa River, Marowijne	3°19’31.0"N 54°03’48.0"W	Sipaliwini	Suriname
Leporinus friderici	MHNG 2718.02	SU08-655	Paloemeu, Marowijne	3°10’42.0"N 55°25’09.1"W	Sipaliwini	Suriname
Leporinus friderici	MZUSP 113983	L018	Rio Turvo, Paraná	20°25’24.0"S 49°15’16.6"W	São Paulo	Brazil
Leporinus geminis	LBP 1621	11676	Rio Araguaia	15°53’35.2"S 52°15’00.9"W	Goiás	Brazil
Leporinus jamesi	LBP 164	4061	Rio Acre, Purus	10°03’19.2"S 67°51’27.0"W	Acre	Brazil
Leporinus jatuncochi	OS 18325	PE10-164	Río Amazonas	~3°41’29.9"S 73°15’32.7"W	Loreto	Peru
Leporinus jatuncochi	OS 18325	PE10-165	Río Amazonas	~3°41’29.9"S 73°15’32.7"W	Loreto	Peru
Leporinus julii	ANSP 197405	t3105	Rio Iriri, Xingu	3°50’32.3"S 52°44’03.9"W	Pará	Brazil
Leporinus julii	INPA-ICT 047459	ANSP t11938	Volta Grande, Rio Xingu	3°21’54.0"S 51°43’59.0"W	Pará	Brazil
Leporinus julii	MZUSP 96386	7177	Rio Jamanxim, Tapajós	7°43’51.0"S 55°16’36.0"W	Pará	Brazil
Leporinus julii	MZUSP 97206	7275	Rio Iriri, Xingu	8°19’07.0"S 55°05’23.0"W	Pará	Brazil
Leporinus lacustris	LBP 2340	15941	Rio Tietê, Paraná	22°46’29.9"S 48°08’43.5"W	São Paulo	Brazil
Leporinus maculatus	ANSP 189041	6982	Lawa River, Marowijne	3°19’31.0"N 54°03’48.0"W	Sipaliwini	Suriname
Leporinus maculatus	FMNH 144982	7035	Lawa River, Marowijne	3°19’31.0"N 54°03’48.0"W	Sipaliwini	Suriname
Leporinus maculatus	MHNG 2724.062	SU08-343	Paloemeu River	3°11’54.0"N 55°24’27.0"W	Sipaliwini	Suriname
Leporinus melanopleura	MZUSP 111246	69556	Rio Una	15°05’52.4"S 39°21’10.4"W	Bahia	Brazil
Leporinus melanopleura	MZUSP 111246	69557	Rio Una	15°05’52.4"S 39°21’10.4"W	Bahia	Brazil
Leporinus melanopleurodes	MZUSP 111242	69554	Rio das Almas	13°36’27.8"S 39°08’38.8"W	Bahia	Brazil
Leporinus microphthalmus	LBP 2491	16370	Rio Montividiu, Paraná	17°26’25.8"S 51°10’26.7"W	Goiás	Brazil
Leporinus microphthalmus	LBP 6222	29303	Rio Grande, Paraná	21°21’03.1"S 46°29’33.2"W	Minas Gerais	Brazil
Leporinus multimaculatus	LBP 2455	16284	Rio Araguaia	15°40’57.7"S 52°13’24.8"W	Mato Grosso	Brazil
Leporinus multimaculatus	LBP 15911	65601	Rio Xingu	13°31’34.1"S 52°43’52.5"W	Mato Grosso	Brazil
Leporinus multimaculatus	LBP 15911	65602	Rio Xingu	13°31’34.1"S 52°43’52.5"W	Mato Grosso	Brazil
Leporinus nigrotaeniatus	USNM 403524	GY11-1-14	Cuyuni River, Essequibo	6°52’28.7"N 60°14’54.5"W	Mazaruni	Guyana
Leporinus nigrotaeniatus	USNM 403536	GY11-2-49	Cuyuni River, Essequibo	6°50’50.0"N 60°8’10.7"W	Mazaruni	Guyana
Leporinus octofasciatus	LBP 3939	19828	Rio Paranapanema	23°08’01.1"S 49°40’34.9"W	São Paulo	Brazil
Leporinus octomaculatus	LBP 13249	69386	Rio Tapajós	13°59’04.1"S 57°04’01.7"W	Mato Grosso	Brazil
Leporinus ortomaculatus	ANSP 191182	BO6132	Río Ventuari, Orinoco	4°17’53.3"N 66°17’58.6"W	Amazonas	Venezuela
Leporinus ortomaculatus	ANSP 191200	BO6143	Río Orinoco	3°52’55.9"N 67°00’49.0"W	Amazonas	Venezuela
Leporinus pachycheilus	UFRO uncat	2156 L479	Rio Riozinho, Madeira	–	Rondônia	Brazil
Leporinus parae	LBP 5487	26567	Rio Jari	0°51’16”S 52°32’30”W	Amapá	Brazil
Leporinus parae	OS 18726	PE10-022	Río Nanay, Amazonas	3°45’04.6"S 73°17’25.5"W	Loreto	Peru
Leporinus pearsoni	ANSP 180841	P4124	Río Tahuamanu, Madre de Dios	11°27’46.0"S 69°18’23.0"W	Madre de Dios	Peru
Leporinus piau	LBP 260	4162	Rio São Francisco	18°13’39.7"S 45°14’51.4"W	Minas Gerais	Brazil
Leporinus piau	LBP 340	4309	Rio São Francisco	19°35’21.7"S 45°18’00.4"W	Minas Gerais	Brazil
Leporinus reticulatus	LBP 13250	69390	Rio Tapajós	13°59’04.1"S 57°04’01.7"W	Mato Grosso	Brazil
Leporinus inexpectatus	LBP 3808	21936	Rio Paranapanema	23°01’26.2"S 48°49’32.6"W	São Paulo	Brazil
Leporinus inexpectatus	LBP 3808	21937	Rio Paranapanema	23°01’26.2"S 48°49’32.6"W	São Paulo	Brazil
Leporinus sp.	LBP 16116	66817	Rio Tapajós	4°33’09.7"S 56°17’59.6"W	Pará	Brazil
Leporinus taeniatus	LBP 261	4267	Rio São Francisco	18°13’39.7"S 45°14’51.4"W	Minas Gerais	Brazil
Leporinus taeniatus	LBP 329	4250	Rio São Francisco	19°52’39.2"S 45°26’04.6"W	Minas Gerais	Brazil
Leporinus tigrinus	GEPEMA 5514	L176	Rio Araguaia	15°53’42.4"S 52°15’16.5"W	Goiás	Brazil
Leporinus tristriatus	MZUSP 110987	6199	Rio Teles Pires, Tapajós	10°23’10.0"S 54°18’22.0"W	Mato Grosso	Brazil
Leporinus unitaeniatus	LBP 17190	68734	Rio Vermelho, Araguaia	15°10’23.2"S 51°09’27.1"W	Goiás	Brazil
Leporinus vanzoi	LBP 16476	67482	Rio Tapajós	4°38’58.9"S 56°17’28.2"W	Pará	Brazil
Leporinus villasboasorum	ANSP 195954	ANSP t10752	Rio Xingu, Amazonas	3°33’41.1"S 51°51’29.1"W	Pará	Brazil
Leporinus villasboasorum	INPA 40521	ANSP t8731	Rio Xingu, Amazonas	2°53’18.7"S 51°56’24.5"W	Pará	Brazil
Megaleporinus brinco	MZUSP 111257	69555	Rio Gongogi, Rio de Contas	14°21’16.5"S 39°46’23.9"W	Bahia	Brazil
Megaleporinus conirostris	–	L210	Rio Paraibuna	–	–	Brazil
Megaleporinus elongatus	MCNIP 0375	L300	Rio Itacambiruçu, Jequitinhonha	16°35’50.0"S 42°50’22.0"W	Minas Gerais	Brazil
Megaleporinus elongatus	–	L307	Rio Jequitinhonha	–	–	Brazil
Megaleporinus garmani	MCNIP 0021	L293	Rio Itacambiruçu, Jequitinhonha	16°35’16.0"S 42°48’53.0"W	Minas Gerais	Brazil
Megaleporinus macrocephalus	LBP 1422	19493	Rio Taquari, Paraguai	18°25’42.5"S 54°50’02.8"W	M. Grosso Sul	Brazil
Megaleporinus macrocephalus	LBP 1422	12505	Rio Taquari, Paraguai	18°25’42.5"S 54°50’02.8"W	M. Grosso Sul	Brazil
Megaleporinus cf. muyscorum	LBP 3030	19114	Río Orinoco	7°38’11.6"N 66°19’04.2"W	Bolívar	Venezuela
Megaleporinus muyscorum	IAvH-P8589	6606/ t4499	Río Magdalena	5°12’21.8"N 74°44’04.2"W	Tolima	Colombia
Megaleporinus obtusidens	LBP 3912	16872	Rio Paranapanema	23°20’00.0"S 48°34’00.0"W	São Paulo	Brazil
Megaleporinus piavussu	LBP 3303	19851	Rio Tietê	22°37’55.7"S 48°10’30.2"W	São Paulo	Brazil
Megaleporinus piavussu	LBP 3304	19848	Rio Tietê	22°37’55.7"S 48°10’30.2"W	São Paulo	Brazil
Megaleporinus reinhardti	LBP 259	4156	Rio São Francisco	18°13’39.7"S 45°14’51.4"W	Minas Gerais	Brazil
Megaleporinus reinhardti	LBP 259	4160	Rio São Francisco	18°13’39.7"S 45°14’51.4"W	Minas Gerais	Brazil
Megaleporinus cf. trifasciatus	LBP 12722	43631	Rio Araguaia	13°19’00.0"S 50°37’00.0"W	Mato Grosso	Brazil
Megaleporinus trifasciatus	OS 18311	PE10-108	Río Amazonas	~3°42’40.0"S 73°14’00.0"W	Loreto	Peru
Megaleporinus trifasciatus	OS 18311	PE10-109	Río Amazonas	~3°42’40.0"S 73°14’00.0"W	Loreto	Peru
Parodon nasus	LBP 1135	5635	Rio Tietê	22°52’21.1"S 48°22’26.7"W	São Paulo	Brazil
Petulanos brevior	ANSP 189141	6912	Lawa River, Marowijne	3°19’31.0"N 54°03’48.0"W	Sipaliwini	Suriname
Petulanos brevior	FMNH 144981	6879	Lawa River, Marowijne	3°19’31.0"N 54°03’48.0"W	Sipaliwini	Suriname
Petulanos brevior	MHNG 2718.041	SU08-440	Paloemeu River, Marowijne	3°10’42.0"N 55°25’09.1"W	Sipaliwini	Suriname
Petulanos intermedius	MZUSP 96778	6191	Rio Peixoto de Azevedo, Tapajós	10°17’14.0"S 54°50’57.0"W	Mato Grosso	Brazil
Petulanos intermedius	MZUSP 97330	7160	Rio Jamanxim, Tapajós	7°43’51.0"S 55°16’36.0"W	Pará	Brazil
Petulanos sp.	AUM 44717	G5207	Takutu River, Branco-Negro	3°28’13.6"N 59°48’35.8"W	Upper Takutu-Upper Essequibo	Guyana
Petulanos sp.	AUM 44717	G5208	Takutu River, Branco-Negro	3°28’13.6"N 59°48’35.8"W	Upper Takutu-Upper Essequibo	Guyana
Potamorhina altamazonica	OS 18312	PE10118	Río Nanay	3°43’05.8"S 73°12’46.1"W	Loreto	Peru
Prochilodus nigricans	OS 18792	PE10045	Río Itaya	3°47’23.5"S 73°14’58.5"W	Loreto	Peru
Psectrogaster amazonica	OS 18313	PE10113	Río Amazonas	3°43’05.8"S 73°12’46.1"W	Loreto	Peru
Pseudanos gracilis	ANSP 180320	T73	Río Nanay, Amazonas	3°52’21.0"S 73°32’43.0"W	Loreto	Peru
Pseudanos gracilis	FMNH 113530	T57	Río Nanay, Amazonas	3°52’21.0"S 73°32’43.1"W	Loreto	Peru
Pseudanos trimaculatus	AUM 48210	G07-746	Burro Burro River, Essequibo	4°10’57.7"N 59°03’49.5"W	Potaro-Siparuni	Guyana
Pseudanos trimaculatus	AUM 48210	G07-747	Burro Burro River, Essequibo	4°10’57.7"N 59°03’49.5"W	Potaro-Siparuni	Guyana
Pseudanos trimaculatus	OS 18795	PE10-057	Río Nanay, Amazonas	3°45’04.6"S 73°17’25.5"W	Loreto	Peru
Pseudanos trimaculatus	OS 18779	PE10-020	Río Nanay, Amazonas	3°45’04.6"S 73°17’25.5"W	Loreto	Peru
Pseudanos varii	ANSP 182659	V5301	Río Atabapo, Orinoco	3°40’54.0"N 67°25’48.0"W	Amazonas	Venezuela
Pseudanos winterbottomi	AUM 39292	V170	Río Manapiare, Orinoco	5°25’43.0"N 66°08’10.0"W	Amazonas	Venezuela
Pseudanos winterbottomi	AUM 39855	V120	Río Ventuari, Orinoco	5°25’43.0"N 66°08’10.0"W	Amazonas	Venezuela
Pseudocurimata troschelii	LBP 9371	43968	Río Tumbes	3°36’37.7"S 80°26’23.9"W	Tumbes	Peru
Rhytiodus argenteofuscus	ANSP 178129	1721	Río Napo, Amazonas	3°29’10.0"S 73°06’24.0"W	Loreto	Peru
Rhytiodus argenteofuscus	ANSP 180242	T44	Río Nanay, Amazonas	3°46’45.0"S 73°22’06.0"W	Loreto	Peru
Rhytiodus microlepis	ANSP 182603	P6326	Río Nanay, Amazonas	3°42’49.0"S 73°16’43.0"W	Loreto	Peru
Rhytiodus microlepis	LBP 4239	22737	Rio Juruá	7°09’49.6"S 73°43’29.7"W	Acre	Brazil
Rhytiodus microlepis	LBP 9781	53201	Río Amazonas	3°42’00.0"S 73°13’00.0"W	Loreto	Peru
Sartor respectus	LBP 15280	62058	Rio Xingu	13°31’34.1"S 52°43’52.5"W	Mato Grosso	Brazil
Schizodon borellii	LBP 12626	47096	Rio Cuiabá, Paraguai	17°49’37.8"S 57°22’53.4"W	M. Grosso Sul	Brazil
Schizodon borellii	LBP 12626	47097	Rio Cuiabá, Paraguai	17°49’37.8"S 57°22’53.4"W	M. Grosso Sul	Brazil
Schizodon fasciatus	FMNH 113522	T55	Río Itaya, Amazonas	3°46’18.8"S 73°14’16.1"W	Loreto	Peru
Schizodon fasciatus	OS 18310	PE10-104	Río Amazonas	~3°42’40.0"S 73°14’00.0"W	Loreto	Peru
Schizodon intermedius	LBP 3305	19845	Rio Tietê	22°37’55.7"S 48°10’30.2"W	São Paulo	Brazil
Schizodon knerii	LBP 11334	45534	Rio das Velhas, São Francisco	17°13’33.7"S 44°48’27.9"W	Minas Gerais	Brazil
Schizodon knerii	LBP 11334	45535	Rio das Velhas, São Francisco	17°13’33.7"S 44°48’27.9"W	Minas Gerais	Brazil
Schizodon nasutus	LBP 2162	15161	Rio Tietê	22°40’32.2"S 48°19’05.8"W	São Paulo	Brazil
Schizodon nasutus	LBP 2735	17482	Rio Paranapanema	23°20’02.3"S 48°38’27.9"W	São Paulo	Brazil
Schizodon nasutus	LBP 3708	21859	Rio Paraná	20°26’00.7"S 51°15’41.2"W	São Paulo	Brazil
Schizodon nasutus	LBP 3708	21860	Rio Paraná	20°26’00.7"S 51°15’41.2"W	São Paulo	Brazil
Schizodon scotorhabdotus	AUM 53654	T09707	Río Manapiare, Orinoco	5°20’13.7"N 66°03’05.3"W	Amazonas	Venezuela
Schizodon scotorhabdotus	AUM 54067	T09044	Río Apure, Orinoco	7°53’56.5"N 67°28’23.8"W	Apure	Venezuela
Schizodon scotorhabdotus	LBP 3046	19130	Río Orinoco	7°38’11.6"N 66°19’04.2"W	Bolívar	Venezuela
Schizodon trivittatus	LBP 12849	53463	Rio Jamanxim, Tapajós	4°45’19.2"S 56°26’16.4"W	Pará	Brazil
Schizodon vittatus	LBP 3994	23098	Rio Araguaia	11°40’09.0"S 50°51’00.3"W	Mato Grosso	Brazil
Schizodon vittatus	LBP 12752	41006	Rio Araguaia	13°18’37.3"S 50°36’47.6"W	Mato Grosso	Brazil
Schizodon vittatus	LBP 12752	41007	Rio Araguaia	13°18’37.3"S 50°36’47.6"W	Mato Grosso	Brazil
Semaprochilodus taeniurus	LBP 1691	12757	Rio Amazonas	3°04’38.0"S 59°49’32.0"W	Amazonas	Brazil
Steindachnerina elegans	LBP 8272	38329	Rio Verde Grande, São Francisco	15°19’24.2"S 43°39’52.5"W	Minas Gerais	Brazil
Synaptolaemus latofasciatus	ANSP 182230	V099	Río Ventuari, Orinoco	4°15’12.0"N 66°20’41.0"W	Amazonas	Venezuela
Synaptolaemus latofasciatus	AUM 54407	T09316	Río Ventuari, Orinoco	3°58’42.3"N 67°03’37.7"W	Amazonas	Venezuela
Synaptolaemus latofasciatus	MZUSP 97460	7230	Rio Jamanxim, Tapajós	7°03’51.0"S 55°26’28.0"W	Pará	Brazil

Multilocus sequencing. We extracted DNA from muscle or fin samples corresponding to vouchered specimens deposited in museum and university collections (Tab. 1). Tissue samples were preserved in 95% ethanol or a saturated DMSO/NaCl solution. Extractions employed a DNeasy Tissue kit (Qiagen Inc.) following manufacturer’s instructions or a NaCl extraction protocol adapted from Lopera-Barrero et al. (2008). Following other phylogenetic studies with characiforms (e.g., Oliveira et al., 2011; Abe et al., 2014; Melo et al., 2018, and many others), we amplified partial sequences of the mitochondrial genes 16S rRNA (16S, 520 bp), cytochrome c oxidase subunit 1 (COI, 699 bp) and cytochrome B (Cytb, 987 bp) using one round of polymerase chain reaction (PCR). Additionally, we obtained sequences of the nuclear myosin heavy chain 6 gene (Myh6, 702 bp), recombination activating gene 1 (Rag1, 1374 bp), and recombination activating gene 2 (Rag2, 1020 bp) through nested-PCR following Oliveira et al. (2011). Primers used in this study are listed in Tab. 2. We did 12.5 μl reactions containing 9.075 μl of double-distilled water, 1.25 μl 5x reaction buffer, 0.375 MgCl₂, 0.25 μl dNTP mix at 8 mM, 0.25 μl of each primer at 10 μM (primers in Tab. 2), 0.05 μl Platinum Taq DNA polymerase enzyme (Invitrogen; www.invitrogen.com) and 1.0 μl genomic DNA (10–50 ng). The PCR consisted of an initial denaturation (4 min at 95 ºC) followed by 28–30 cycles of chain denaturation (30 s at 95 ºC), primer hybridization (30–60 s at 52–54 ºC), and nucleotide extension (30–60 s at 72 ºC). After visualization of the fragments on a 1% agarose gel, we sequenced using dye terminators (BigDye™ Terminator v. 3.1 Cycle Sequencing Ready Reaction Kit, Applied Biosystems) purified again through ethanol precipitation. We sequenced the samples on an ABI 3130-Genetic Analyzer (Applied Biosystems) at either Arizona State University (Tucson, Arizona, USA) or the Universidade Estadual Paulista (Botucatu, São Paulo, Brazil).

TABLE 2 | Locus information and primer specifications.

Locus	Length	Primer	Reference
16s	520	16Sa-L / 16Sb-H	Palumbi (1996)
COI	699	L6252-Asn / H7271-COXI	Melo et al. (2011)
Cytb	987	LNF / H08R2	Oliveira et al. (2011)
Myh6	702	F329 / A3R1 A3F2 / A3R2	Li et al. (2007)
Rag1	1,374	Rag1CF1 / Rag1CR1 Rag1CF2 / Rag2CR2	Oliveira et al. (2011)
Rag2	1,020	164F / Rag2-R6 176F / Rag2Ri	Lovejoy, Collette (2001); Abe et al. (2013)

Quality control and error checking. After sequencing, for nuclear and mitochondrial genes alike, we assembled and edited consensus sequences in Geneious v. 6.1.8 (Kearse et al., 2012) and applied IUPAC ambiguity codes where we detected uncertain nucleotide identity. Anomalously short or low-quality sequences were discarded. We aligned the consensus sequences of each gene using the MUSCLE algorithm (Edgar, 2004) in Geneious v. 6.1.8, trimmed sequences to a uniform length, and inspected for misalignments by eye. At this stage, we also constructed single-gene trees using neighbor joining or maximum likelihood to detect errors such as cross contamination or swapped tubes. We concluded that contamination had occurred when individuals of different nominal genera displayed identical sequences for a given locus, and all such contaminated sequences were removed from downstream analyses. A few unambiguous instances of swapped tubes became apparent when two distantly related taxa (usually in different genera) traded places on a single gene tree and, in such cases, we corrected the sample assignments and proceeded with analysis.

The presence of loops in 16S rRNA caused substantial alignment ambiguity and low support for many nodes in that gene tree in preliminary trials. To address that problem, we excised the hypervariable loop regions, generated a reduced 16S submatrix, and concatenated the six genes into a matrix. While moderate levels of missing data do not represent a major concern for phylogenetic inference (Kearney, 2002; Wiens, 2006), large amounts of missing data can destabilize phylogenetic analysis by introducing rogue taxa (Sanderson, Shaffer, 2002; Roure et al., 2012). Taxa sequenced for only one or two loci are also more susceptible to misplacement due to laboratory errors (e.g., sample contamination or swapped tubes). To limit these potential problems, we removed any individuals that had amplified for only one or two of the six genes. GenBank accession numbers for all sequences used in the final analysis is in S1.

Partitioning and phylogenetic analysis. We divided the concatenated matrix into sixteen possible partitions: the 16S stems, and each codon position for each of the five coding genes. Then, we used a greedy search in PartitionFinder 1.1.1 (Lanfear et al., 2012) to determine the best partitioning scheme and to select the best-fit model for each partition. That search employed the raxml modelset, linked branch lengths and selected the final schema using the corrected Akaike Information Criterion (AICc). This process identified an optimal subdivision into sixteen partitions using seven models (Tab. 3).

TABLE 3 | Gene partitions and their best AICc models of nucleotide evolution for MrBayes.

Gene	Position	Subset	# sites	Best AICc model
16S	1–520	1	520	GTR+I+G
COI 1^st position	521–1219\3	2	233	GTR+I+G
COI 2^nd position	522–1219\3	3	233	HKY+I+G
COI 3^rd position	523–1219\3	4	233	GTR+I+G
Cytb 1^st position	1220–2206\3	5	329	SYM+I+G
Cytb 2^nd position	1221–2206\3	6	329	GTR+I+G
Cytb 3^rd position	1222–2206\3	7	329	GTR+I+G
Myh6 1^st position	2207–2908\3	8	234	GTR+I+G
Myh6 2^nd position	2208–2908\3	9	234	GTR+I
Myh6 3^rd position	2209–2908\3	10	234	HKY+G
Rag1 1^st position	2910–4282\3	11	458	GTR+I+G
Rag1 2^nd position	2911–4282\3	12	458	GTR+I+G
Rag1 3^rd position	2912–4282\3	13	458	SYM+G
Rag2 1^st position	4283–5302\3	14	340	HKY+I+G
Rag2 2^nd position	4284–5302\3	15	340	HKY+I+G
Rag2 3^rd position	4285–5302\3	16	340	K80+G

We based our primary phylogenetic reconstruction on a maximum likelihood (ML) analysis performed in RAxML HPC2 on XSEDE (Stamatakis, 2006) as implemented on the CIPRES Scientific Gateway v. 3.3 (Miller et al., 2010). This analysis employed the seven partitions identified above and a random starting tree with other parameters left at default values. One thousand bootstrap pseudoreplicates tested the support for each node in the most likely topology. We also conducted a parallel Bayesian inference with two independent runs of 10 million generations each saving trees at every 1,000th generation in MrBayes v. 3.2.2 (Ronquist et al., 2012) on XSEDE via the CIPRES webserver (Miller et al., 2010). The first 1,000 trees in each run were discarded as 10% burn-in. Convergence and stationarity were inspected and verified by checking ESS values (ESS>200) and trace distributions in Tracer v. 1.7.1 (Rambaut et al., 2018). The 9,001 trees obtained from one of the runs were then read in TreeAnnotator v. 2.6.2 (distributed as part of BEAST; Drummond, Rambaut, 2007) and the maximum clade credibility tree was summarized in FigTree v. 1.4.4 (Rambaut, 2018).

Results

The final matrix with 204 individuals representing 97 anostomid species (66.4% species richness) and 16 outgroups contains 5,302 bp of which 2,201 sites were variable, 1,797 were parsimony informative and 404 represent autapomorphies. The supplementary materials contain this full data matrix (S2). Stop codons, deletions and insertions were absent in all coding sequences. The base composition was 25.8% adenine, 26.2% cytosine, 23.5% guanine, 24.6% thymine and 15.9% missing data. No saturation in transition or transversions were detected. The restriction of the dataset to individuals sequenced for at least three genes led to the exclusion of Insperanos, but the phylogenetic position, evolutionary history, and timing of divergence of this distinctive lineage were recently addressed (Sidlauskas et al., 2021).

The narrative below recognizes a new genus containing the continentally distributed Leporinus striatus Kner, 1858. Our analysis (Figs. 2, S3, S4) and other works using morphology and/or molecules (Sidlauskas, Vari, 2008; Ramirez et al., 2016, 2017b; Mirande, 2019; Birindelli et al., 2020b; Sidlauskas et al., 2021) all place this species as more closely related to Abramites Fowler, 1906 and Megaleporinus than to any other living species of Leporinus. Thus, transferring it helps to alleviate the non-monophyly of Leporinus. It also possesses distinctive dentition that affords a simple generic diagnosis (see below). We describe it formally here to properly contextualize the use of the new name throughout the remainder of this contribution.

FIGURE 2| Lateral photographs and distribution of Brevidens striatus. White arrow in inset image points to the miniscule fourth dentary tooth from which the genus takes its name. Top photograph: ROM uncatalogued, SL unknown, Colombia: Río Atrato, photo by Nathan Lujan. Bottom photograph: MZUEL 7929, 105 mm SL, Brazil: rio Paraguay, photo by José Birindelli. Image reproduced from Birindelli, Britski (2013) with permission of the authors and journal.

Brevidens Birindelli, Sidlauskas & Melo, new genus

urn:lsid:zoobank.org:act:7AF3926D-074D-41D9-8406-28D124AB554D

Type-species. Leporinus striatus Kner, 1858:79, herein designated.

Diagnosis. Brevidens differs from all other members of Anostomidae by possessing a fourth dentary tooth that is distinctly smaller than the anterior three teeth and separated from those teeth by a diastema (Fig. 2). Brevidens is further diagnosed by the following non-exclusive morphological features: four dark longitudinal stripes on the body; three premaxillary teeth; 16 scale rows around the caudal peduncle; a red spot on the ventral portion of the upper lip in life; subterminal mouth, its cleft longitudinally aligned with the ventral border of the eye in 60 mm SL or larger specimens.

Comparisons. Externally, Brevidens striatus can be quickly distinguished from all anostomids, except Anostomus anostomus, A. ternetzi Fernández-Yépez, 1949, Hypomasticus arcus (Eigenmann, 1912), H. despaxi (Puyo, 1943), H. tepui (Birindelli, Britski & Provenzano, 2019), Leporinus sexstriatus Britski & Garavello, 1980, L. tristriatus Birindelli & Britski, 2013,and Petulanos brevior (Géry, 1961) (new combination, see below), by having more than two dark longitudinal stripes on the body. Brevidens striatus is distinguished from all species of Anostominae by having a subterminal mouth (vs. superior), three premaxillary teeth (vs. four), and unicuspid incisiform teeth (vs. multicuspid crenate teeth). Brevidens is distinguished from Hypomasticus arcus, H. despaxi, and H. tepui by having three teeth on the premaxilla (vs. four), the dark midlateral stripe on the body continuous with a dark stripe on the head (vs. discontinuous), and the ventralmost stripe on the body not continuing anteriorly onto the head (vs. continuing). Brevidens is distinguished from L. tristriatus and L. sexstriatus by having four dark stripes on the body (vs. three or six), 16 scale rows around the caudal peduncle (vs. 12), a red spot on the ventral portion of the upper lip in life (vs. absent) and a subterminal mouth with its cleft longitudinally aligned with the ventral border of the eye in 60 mm SL or longer specimens (vs. a subinferior mouth with a cleft aligned with the ventral border of the infraorbitals).

Etymology. From the Latin brevis, meaning short, plus the Latin dens, meaning tooth, in reference to the abbreviated fourth dentary tooth that diagnoses the genus. Gender masculine.

Geographical distribution. Brevidens has a large geographical distribution including cis- and trans-Andean rivers. On the eastern side of the Andes, it occurs in the Uruguay, Paraná and Paraguay basins of Argentina, Bolivia, Brazil, and Paraguay, in the western Amazon basin of Brazil, Bolivia, Colombia, Ecuador, and Peru, and in the Orinoco basin of Colombia and Venezuela. Its trans-Andean distribution includes the Atrato, Magdalena, and Sinú rivers of Colombia (Birindelli, Britski, 2013).

Diversity. Birindelli, Britski (2013) found no obvious external morphological characteristics that would subdivide Brevidens striatus into multiple species. Nevertheless, its widespread geographical distribution includes areas that have been isolated from adjacent drainages for millions of years, such as the upper La Plata, Amazon, Orinoco, and the Magdalena basins (Lundberg et al., 1998; Albert, Reis, 2011; Aguilera et al., 2013). That occurrence in so many different river systems suggests that the diversity of Brevidens may be underestimated.

Monophyly, relationships and taxonomy of subfamilial lineages. Maximum likelihood (Figs. 3, S1) and Bayesian reconstructions (Fig. S2) agreed unequivocally about the monophyly of Anostomidae, its three major subclades, and the relationships among them. Within Anostomidae, the initial split divides Leporellus Lütken, 1975 and the genera of Anostominae (sensu Winterbottom, 1980) from the remainder of the family (Fig. 3). Each of these lineages received full statistical support (100% bootstrap; 1.0 posterior probability) and have been recovered consistently in prior phylogenetic studies incorporating molecular data (Ramirez et al., 2017b; Betancur-R. et al., 2019; Sidlauskas et al., 2021; Melo et al., 2022). As such, we elevate the three major lineages to subfamilial status, using names established in prior literature. We refer Leporellus to Leporellinae Eigenmann, 1910 (Fig. 3), Anostomus, Gnathodolus, Petulanos, Pseudanos, Sartor and Synaptolaemus to Anostominae Günther, 1864, and Abramites, Anostomoides, Brevidens, Hypomasticus, Insperanos, Laemolyta, Leporinus, Megaleporinus, Rhytiodus, and Schizodon to Leporininae Eigenmann, 1912 (Fig. 3). The inclusion of Insperanos in Leporininae is based on the results of Sidlauskas et al. (2021) which placed Insperanos as the sister lineage to the remainder of that subfamily with 86% posterior probability.

FIGURE 3| Intrageneric relationships among anostomid fishes, based on the maximum likelihood reconstruction and using the revised taxonomy proposed herein. Color coding highlights membership in subfamilies Leporellinae (green), Anostominae (red) and Leporininae (blue). Diagrams illustrating body shape and color pattern by José Birindelli and Peter van der Sleen, used with permission.

Though the family group name for Anostominae comes from Anostomatina of Günther, 1864, the first usage of Anostominae (with that spelling) appears to be that of Myers (1950). His subfamilial concept matches that of the whole family Anostomidae as treated herein, as does Günther’s (1864) Anostomatina. Our concept of subfamily Anostominae more closely matches that of Winterbottom (1980).

Though our classification for Anostomidae uses Linnean-rank subfamilies, we also provide definitions that adhere to PhyloCode (Queiroz, 2006; Queiroz, Cantino, 2020; Laurin, 2023), a rank-free system of classification that has gained substantial current traction and that uses explicit patterns of common ancestry and apomorphy to define taxa. In so doing, we erect a framework compatible with the two major classifications of actinopterygian fishes currently being debated by the ichthyological community (Betancur-R. et al., 2013; Near, Thacker, 2024)

Anostomidae Günther, 1864

Anostomatina Günther, 1864:279. Type-genus: Anostomus Scopoli, 1777. —Günther, 1864:279. Originally as Anostomatina, and thus implying stem Anostomat–. —Gill, 1896:209 corrected the stem to Anostom–.

Pithecocharacinae Fowler, 1906:319. Type-genus: Pithecocharax Fowler, 1906. Pithecocharacinae Fowler, 1906 is a junior objective synonym of Anostomatina Günther, 1864 because Pithecocharax Fowler, 1906 is an objective junior synonym of Anostomus Scopoli, 1777.

Type-genus. Anostomus Scopoli, 1777.

Phylogenetic definition. The crown cladeoriginating in the most recent common ancestor of Anostomus anostomus, Leporellus vittatus (Valenciennes, 1850) and Leporinus fasciatus (Bloch, 1794) (Figs. 2, S1, S2).

Diversity. Anostomidae includes approximately 150 valid species (Toledo-Piza et al., 2024) allocated among 17 nominal genera: Abramites, Anostomoides Pellegrin, 1909, Anostomus, Brevidens, Gnathodolus Myers, 1927, Hypomasticus, Insperanos, Laemolyta, Leporellus, Leporinus, Megaleporinus, Petulanos Sidlauskas & Vari, 2008, Pseudanos Winterbottom, 1980, Rhytiodus Kner, 1858, Sartor Myers & Carvalho, 1959, Schizodon,and Synaptolaemus Myers & Fernández-Yépez, 1950.

Anostominae Günther, 1864

Type-genus. Anostomus Scopoli, 1777.

Phylogenetic definition. The crown cladeoriginating in the most recent common ancestor of Anostomus anostomus, Pseudanos trimaculatus (Kner, 1858) and Gnathodolus bidens Myers, 1927.

Diversity. Anostominae includes 16 species allocated among Anostomus, Gnathodolus, Petulanos, Pseudanos, Sartor and Synaptolaemus.

Leporellinae Eigenmann, 1910

Type-genus. Leporellus Lütken, 1875.

Phylogenetic definition. The most inclusive crown clade that contains the common ancestor of Leporellus, but not of Anostomus, Insperanos or Hypomasticus.

Diversity. Leporellinae includes a single genus, Leporellus.

LeporellusLütken, 1875

Phylogenetic definition. Leporellus is herein defined as the clade within Anostomidae for which scales covering the caudal-fin rays and three or more dark stripes on the caudal fin, as inherited by Leporellus vittatus, are apomorphies.

Diversity. Though Toledo-Piza et al. (2024) considered the genus to encompass just two valid species (Leporellus pictus and L. vittatus) data herein suggest higher intrageneric diversity (Fig. 4).

FIGURE 4| Relationships among species of Leporellinae and Anostominae, based on the maximum likelihood reconstruction. Gray shading on inset phylogeny indicates the region detailed. Colored circles represent ranges of bootstrap values. Diagrams illustrating body shape and color pattern by José Birindelli and Peter van der Sleen, used with permission,

Leporininae Eigenmann, 1912

Type-genus. Leporinus Agassiz, 1829.

Phylogenetic definition. The crown cladeoriginating in the most recent common ancestor of Leporinus fasciatus, Hypomasticus mormyrops (Steindachner, 1875), and Insperanos nattereri (Steindachner, 1876). This definition is based on the combined molecular and morphological phylogenetic hypothesis of Sidlauskas et al. (2021)

Diversity. Leporininae includes 128 species allocated among the genera Abramites, Brevidens, Anostomoides, Hypomasticus, Insperanos, Laemolyta, Leporinus, Megaleporinus, Rhytiodus, and Schizodon.

Taxonomic changes to alleviate non-monophyly. With two thirds of the known species in Anostomidae now placed in a phylogenetic context, we can reassign several species to correct some obvious cases of genus-level paraphyly or polyphyly (Tab. 4). For example, because three analyzed specimens of Anostomus brevior Géry, 1961 are placed with high statistical support (96% bootstrap; 1.0 posterior probability) as sister to a clade containing both sequenced species of Petulanos and not with the remainder of Anostomus (Figs. 4, S1, S2) we formally transfer Anostomus brevior to an expanded concept of Petulanos. Similar logic justifies the transfer of Leporinus striatus to the new genus Brevidens.

TABLE 4 | Taxonomic changes proposed in this study. Nominal species are arranged alphabetically.

Nominal species	Valid name prior to study	New combination assigned herein
Anostomus brevior Géry, 1961	Anostomus brevior	Petulanos brevior
Leporinus arcus Eigenmann, 1912	Leporinus arcus	Hypomasticus arcus
Leporinus gomesi Garavello & Santos, 1981	Leporinus gomesi	Hypomasticus gomesi
Leporinus granti Eigenmann, 1912	Leporinus granti	Hypomasticus granti
Leporinus julii Santos, Jégu & Lima, 1996	Hypomasticus julii	Leporinus julii
Leporinus lebaili Géry & Planquette, 1983	Leporinus lebaili	Hypomasticus lebaili
Leporinus melanostictus Norman, 1926	Leporinus melanostictus	Hypomasticus melanostictus
Leporinus nijsseni Garavello, 1990	Leporinus nijsseni	Hypomasticus nijsseni
Leporinus pachycheilus Britski, 1976	Hypomasticus pachycheilus	Leporinus pachycheilus
Leporinus santosi Britski & Birindelli, 2013	Leporinus santosi	Hypomasticus santosi
Leporinus striatus Kner, 1858	Leporinus striatus	Brevidens striatus
Leporinus tepui Birindelli, Britski & Provenzano, 2019	Leporinus tepui	Hypomasticus tepui
Leporinus torrenticola Birindelli, Teixeira & Britski, 2016	Leporinus torrenticola	Hypomasticus torrenticola

The remaining taxonomic changes involve the composition of Hypomasticus, originally conceived as a subgenus of Leporinus possessing subterminal or inferior mouths (Borodin, 1929; Géry, 1960). Results here and elsewhere have reconstructed many species of Hypomasticus

as belonging to a distinct lineage originating early in the history of the family (Sidlauskas, Vari, 2008; Ramirez et al., 2017b; Birindelli et al., 2020b; Sidlauskas et al., 2021). However, the composition of that clade is broader than originally envisioned, and molecular results here (Fig. 5) and elsewhere (Birindelli et al., 2020b; Sidlauskas et al., 2021) universally agree that it contains several species with terminal mouths that have been traditionally assigned to Leporinus. Based on their inclusion in this clade in molecular analysis, we transfer Leporinus granti Eigenmann, 1912, L. lebaili Géry & Planquette, 1983, L. melanostictus Norman, 1926, and L. torrenticola Birindelli, Teixeira & Britski, 2016to Hypomasticus. Though not included in our molecular sampling, we also transfer Leporinus arcus, L. gomesi Garavello & Santos, 1981, L. nijsseni Garavello, 1990 and L. santosi Britski & Birindelli, 2013 to Hypomasticus based on their morphological similarity to Leporinus granti (now H. granti) in aspects of coloration, body shape, dentition and squamation (Britski, Birindelli, 2013; Birindelli et al., 2019).

FIGURE 5| Relationships among species of Hypomasticus, Leporinus ecuadorensis and L. brunneus, based on the maximum likelihood reconstruction. Gray shading on inset phylogeny indicates the region detailed. Colored circles represent ranges of bootstrap values. Boxed numeral 1 indicates the most recent common ancestor of the L. ecuadorensis clade as discussed herein. Note that the species listed herein as Hypomasticus sp. appears as Leporinus sp. 1 in the data matrix and supplementary figures, which use the species combinations as they existed prior to this work. Specimen ANSP 192149 (Leporinus brunneus) appears twice because it was accidentally duplicated in the final matrix. Diagrams illustrating body shape and color pattern by José Birindelli and Peter van der Sleen, used with permission.

On the other hand, results herein (Fig. 6) agree with several other studies (Ramirez et al., 2017b; Mirande, 2019; Birindelli et al., 2020b; Sidlauskas et al., 2021) in concluding that Hypomasticus julii (Santos, Jégu & Lima, 1996) and H. pachycheilus (Britski, 1976) are distantly related to the remainder of Hypomasticus. Though Sidlauskas, Vari (2008) transferred those species to Hypomasticus on account of their strongly inferior mouth position, results here and elsewhere have clarified this as a morphological convergence. As such, we transfer those two species back to their original placement in Leporinus.

FIGURE 6| Relationships among most species assigned to the non-monophyletic genus Leporinus, based on the maximum likelihood reconstruction. Gray shading on inset phylogeny indicates the region detailed. Colored circles represent ranges of bootstrap values. Asterisks indicate clades that differ between the maximum likelihood and Bayesian reconstructions. Boxed numerals indicate the most recent common ancestors of subclades of Leporinus discussed herein: 2) L. melanopleura clade, 3) L. pachycheilus clade, 4) L. friderici clade, 5) L. fasciatus clade. Diagrams illustrating body shape and color pattern by José Birindelli and Peter van der Sleen, used with permission.

In total, these changes begin to alleviate the non-monophyly of Leporinus. While it would be possible to elevate other clades to generic status, we refrain from doing so because of the large number of species in Leporinus that still await inclusion in molecular phylogenies, and our generally low confidence in predicting their phylogenetic placement. Some of the subclades recognized herein include species that differ substantially in patterns and body shapes [e.g., the Leporinus ecuadorensis Eigenmann & Henn, 1916, L. melanopleura Günther, 1864 and L. friderici (Bloch, 1794) clades] and many of the missing species have not been included in any modern studies of morphology or genetics. Further changes to the genus-level taxonomy of Leporinus surely await, but reassignments beyond those reported here are premature.

Genus-level relationships. Leporellus is the sisterclade to Anostominae. Within Anostominae, we reconstruct a basal split between a clade containing Gnathodolus and Sartor on one hand, and the members of Anostomus, Petulanos, Pseudanos and Synaptolaemus on the other. Synaptolaemus appears as sister to the remaining three genera, and Anostomus (as amended above) is sister to Pseudanos plus Petulanos (Fig. 3).

Even with the taxonomic changes summarized above, the large genus Leporinus is non-monophyletic in its current composition. As such we divide it into several clades for the purposes of discussion (Fig. 3). The basal split within subfamily Leporininae separates Hypomasticus from the remaining taxa, noting that Hypomasticus as conceived herein includes some former species of Leporinus (Tab. 4). The next split within the family divides the Leporinus ecuadorensis clade from all other members of Leporininae. A clade containing Anostomoides, Laemolyta, Rhytiodus,and Schizodon appears as sister to a clade containing Abramites, Brevidens, Megaleporinus, and all remaining members of Leporinus. Rhytiodus and Schizodon are sister taxa, with Laemolyta appearing as the closest genus to that pair and Anostomoides sister to the clade containing all three. Abramites and Megaleporinus are sister taxa, with Brevidens as the most closely related lineage to that pair. The Leporinus jamesi clade appears as sister to the clade containing Brevidens, Abramites and Megaleporinus. The Leporinus fasciatus clade contains the type-species for the genus and shares a sister relationship with the Leporinus friderici clade. The Leporinus pachycheilus clade is sister to that pair. The Leporinus melanopleura clade is sister to the clade containing Abramites, Megaleporinus, Brevidens, and the Leporinus fasciatus, L. friderici, L. jamesi Garman, 1929 and L. pachycheilus clades.

Due to the large number of species in Anostomidae, we prefer to initially report species-level relationships within each of its subclades visually rather than textually, as in Fig. 4, which details Anostominae and Leporellinae. Such figures are based on the maximum likelihood results and include bootstrap values reported categorically. Asterisks mark the few instances in which a clade appears in the maximum likelihood reconstruction, but not the Bayesian. The full trees with exact support values resulting from likelihood and Bayesian analyses appear in Figs. S1 and S2. A detailed comparison of these relationships to previous hypotheses appears below.

Discussion

Leporellinae. Results reveal geographic structure that suggests the presence of multiple species within the current concept of Leporellus vittatus, a result also obtained with comparative cytogenetics (Aguilar, Galetti Jr., 2008). The data suggest three lineages: one in the upper Amazon, one in the Orinoco, and a third in the non-Amazonian rivers of southern and eastern Brazil (Fig. 4). The distribution of the genus is, however, substantially more widespread and includes the lower Amazon, the Essequibo, the Paraná-Paraguay system, the São Francisco basin and various Amazonian tributaries (Géry, 1977; Géry et al., 1987; Santos, Jégu, 1989; Sidlauskas, Vari, 2012). Denser molecular sampling and morphological investigation will be needed to determine whether these molecular divisions indicate deep population structure or species-level diversification.

Anostominae. The long hypothesized sister relationship between Sartor and Gnathodolus based on external (Myers, Carvalho, 1959) and internal morphology (Winterbottom, 1980; Sidlauskas, Vari, 2008) is again returned here. Though the molecular study of Burns, Sidlauskas (2019) separated these genera (nesting Sartor deeply within Leporinus), thatplacement likely results from a swapped tube or contamination because no other study has ever suggested a placement for Sartor outside of Anostominae.

We obtained a novel placement of Sartor+Gnathodolus with respect to previous morphological studies, which strongly supported the monophyly of a clade uniting Sartor, Gnathodolus,and Synaptolaemus and reconstructed Pseudanos as sister to the remaining anostomine genera (Winterbottom, 1980; Sidlauskas, dos Santos, 2005; Sidlauskas, Vari, 2008; Dillman et al., 2016; Sidlauskas et al., 2021). The result herein accords partially with recent phylogenomic results, which reconstructed Synaptolaemus as sister to a clade containing members of Anostomus, Petulanos and Pseudanos (Betancur-R. et al., 2019; Melo et al., 2022). However, the fact that Sartor and Gnathodolus do not cluster with Synaptolaemus is surprising, given the distinctive similarities in dentition, jaw structure and mouth position that those genera share (Sidlauskas, Vari, 2008).

Ichthyologists have long recognized these three genera as the most anatomically and ecologically divergent anostomids (Myers, Carvalho, 1959; Winterbottom, 1980; dos Santos, Jégu, 1987), each of which possesses numerous autapomorphies. Gnathodolus bidens, for example, has a backwards-facing superior mouth, lower dentition reduced to a single clawlike tooth on each dentary, and a suspensorium that appears to glide forward and back along grooves in the vomer (Sidlauskas, Vari, 2008). In morphological reconstructions, characters of the infraorbital and opercular series unite these three genera, and additional characters of the infraorbitals, anterior neurocranium, dentition, maxilla, quadrate, opercle, urohyal and other bones support their placement within a clade also including Petulanos and Anostomus (Sidlauskas, Vari, 2008). Therefore, the morphological support for a clade uniting these three genera is robust, as is the support for the clade’s origin from an ancestor resembling the less specialized members of Anostominae.

The molecular results herein imply a very different evolutionary scenario, in which the most recent common ancestor (MRCA) of Gnathodolus, Sartor and Synaptolaemus is also the MRCA of all anostomines, and with Synaptolaemus more closely related to Anostomus, Petulanos and Pseudanos than to Gnathodolus and Sartor. This topology implies that at least some of the unusual morphologies of Gnathodolus, Sartor and Synaptolaemus are ancestral for Anostominae, and that the more generalized morphology of Anostomus, Petulanos and Pseudanos represents reversal. This scenario would be remarkable if true but should be treated with some skepticism due to the low statistical support for Synaptolaemus sister to Anostomus+Pseudanos+Petulanos (70/0.79; Figs. S1, S2), and the large molecular branch lengths subtending Gnathodolus and Sartor, which are the longest of all lineages within the family. Those branch lengths indicate high rates of molecular evolution and the possibility of sequence saturation, which might produce long branch attraction at the base of Anostominae and affect the rooting of the subtree. If the anostomine subtree were instead rooted along the lineage subtending the Anostomus+Petulanos+Pseudanos clade, that slight adjustment would lead to monophyly of Gnathodolus+Sartor+Synaptolaemus and accord much more closely with the morphological data. And indeed, the total evidence analysis of Sidlauskas et al. (2021) places the Gnathodolus+Sartor+Synaptolaemus clade as sister to Anostomus, with Petulanos sister to the remainder of the subfamily. Though this arrangement seems drastically different from the molecular-only result reported herein at first glance, it simply represents yet another possible rooting of the anostomine subtree. The unrooted network for that subfamily is identical in the two studies. Future research will be needed to determine the true position of that root and to reconstruct the evolutionary sequence leading to the exceptional morphologies of Sartor and Gnathodolus.

Results herein strongly support a close relationship between Anostomus brevior (now Petulanos brevior) and two species of Petulanos, despite the former possessing a color pattern strongly resembling Anostomus anostomus. The osteology of P. brevior has not been examined in detail, and future research should investigate whether it possesses the distinctive triangular lamina of the symplectic that has been considered a clear synapomorphy of Petulanos (Sidlauskas, Vari, 2008). With or without that lamina, the species clearly belongs to a lineage outside of Anostomus sensu stricto, and its new placement within Petulanos should stabilize its taxonomy. The specimens labelled as Petulanos sp. from the Takutu drainage of Guyana may represent P. plicatus (Eigenmann, 1912), a species known from the Essequibo drainage, a basin known to share fish species with the nearby Takutu which flows into the Branco, a tributary of the Negro in the Amazon basin. The color pattern of the Takutu Petulanos, however, more closely resembles P. spiloclistron from rivers of Suriname.

Phylogenetic results within Pseudanos confirm the closer relationships among P. gracilis (Kner, 1858), P. varii Birindelli, Lima & Britski, 2012 and P. winterbottomi Sidlauskas & Santos, 2005 than between those species and P. trimaculatus (Winterbottom, 1980; Sidlauskas, Santos, 2005; Birindelli et al., 2012). Pseudanos varii and P. winterbottomi are scarcely divergent in the molecular dataset, despite their obvious difference in coloration (the former has a lateral stripe and the latter has lateral spots), the number of branchiostegal rays and ecological specialization for white or black waters (Sidlauskas, Santos, 2005; Birindelli et al., 2012). These species occupy adjacent drainages and were once considered to be conspecific (Winterbottom, 1980); their close genetic similarity may breathe new life into that old hypothesis. Conversely, these may represent the product of recent speciation along ecological lines, in which case insufficient time may have elapsed for neutral loci to have differentiated. The four samples of P. trimaculatus, two from Peru and two from Guyana, were recovered as closely related, and indeed almost identical at the sequenced loci. Winterbottom (1980) previously considered the Guyanese population to represent a distinct species (P. irinae Winterbottom, 1980) due to the presence of dark spots on the anterior portion of the body scales, which he considered absent in Amazonian specimens. Examining a larger sample, Birindelli et al. (2012) concluded that the dark spots were variably present in specimens from the Amazon and synonymized P. irinae with P. trimaculatus. Our results corroborate their conclusion.

Leporininae. The relationships within Leporininae match earlier reconstructions in broad strokes (Fig. 3), confirming the position of Hypomasticus as sister to the remainder of the subfamily, the close relationship between Anostomoides, Laemolyta, Rhytiodus and Schizodon, the paraphyly of Leporinus, and the existence of a clade containing Abramites, Brevidens, and Megaleporinus. Sidlauskas, Vari (2008) reconstructed many of these intrageneric relationships on osteological grounds, though their phylogeny placed the anostomines within the clade containing Anostomoides, Laemolyta, Rhytiodus and Schizodon, likely due to convergence in mouth position and coloration between Laemolyta, Anostomus and Pseudanos. Later molecular reconstructions or total-evidence approaches confirmed the early divergence of Hypomasticus, the non-monophyly of Leporinus and the monophyly of a group containing Anostomoides, Laemolyta, Rhytiodus and Schizodon,but relocated Anostominae to an earlier diverging position (Ramirez et al., 2016, 2017b; Betancur-R. et al., 2019; Burns, Sidlauskas, 2019; Mirande, 2019; Sidlauskas et al., 2021; Melo et al., 2022). Ramirez et al. (2017b) recognized the close affinity of several large bodied species of Leporinus with Abramites and Brevidens (as L. striatus), and they erected the genus Megaleporinus to accommodate these large, distinctive and commercially important species. The newly reconstructed relationships within each major subclade of Leporininae herein provide the most detailed picture and densest taxonomic sampling of any phylogeny to date. They also reveal that many nominal “groups” within Leporinus based on color pattern, tooth formulae, mouth position or body shape (Géry, 1977; Garavello, 1979; Sidlauskas et al., 2011) are not clades, but rather result from frequent convergence in those character systems.

Hypomasticus. We reconstruct a monophyletic Hypomasticus as the sister clade to the remainder of Leporininae (Figs. 2, 5) with strong support for its monophyly (98% bootstrap, posterior probability of 0.99). As obtained herein, Hypomasticus includes species previously assigned to this genus [H. megalepis (Günther, 1863), H. lineomaculatus Birindelli, Peixoto, Wosiacki & Britski, 2013, H. despaxi, H. steindachneri (Eigenmann, 1907) H. copelandii (Steindachner, 1875), H. mormyrops and H. thayeri (Borodin, 1929)] plus several formerly in Leporinus (Fig. 5). Species newly transferred to Hypomasticus based on these molecular results include H. torrenticola, H. granti, H. lebaili and H. melanostictus (Tab. 4). An undescribed species from the Juruena basin of Brazil with a distinctive color pattern also belongs to this clade (Fig. 5). Birindelli et al. (2020b) foreshadowed this result when they obtained L. granti and L. nijsseni nested within a paraphyletic Hypomasticus (albeit with weak support for the paraphyly). Results herein agree with Birindelli et al. (2020b) in concluding that Hypomasticus contains some species traditionally considered to be part of Leporinus. As such, the characteristics that Sidlauskas, Vari (2008) used to define Hypomasticus are likely homoplastic, particularly given that they all involve the shape and orientation of the bones of the oral jaws or anterior neurocranium, and are thus likely prone to covary with mouth position. While some of the species in the clade possess the inferior mouth from which the genus takes its name, others have the mouth in a subterminal (e.g., H. copelandii) or terminal position (e.g., H. granti). Importantly, two species previously assigned to Hypomasticus because of their distinctively inferior mouths fall outside of this clade and lie within the broad paraphyletic grade of Leporinus (Fig. 6), a result previously obtained in other molecular analyses (Ramirez, Galetti Jr., 2015; Ramirez et al., 2017b; Mirande, 2019; Birindelli et al., 2020b; Sidlauskas et al., 2021). As such, we transfer Hypomasticus julii and H. pachycheilus back to Leporinus (Tab. 4).

Within Hypomasticus as construed herein, three major subclades exist, all with strong statistical support (Fig. 5). Interestingly, two of the clades occur in the Amazon, but are not sister to each other, implying that the phylogenetic relationships within the genus reflect a complicated biogeographic history and a possible dispersal from the Amazon to the river systems of coastal Brazil (Fig. 7). The first subclade contains Hypomasticus torrenticola, H. megalepis, H. lineomaculatus and H. despaxi. This set of species demonstrates marked color pattern variation, with the two species possessing numerous widely spaced spots, one possessing closely spaced spots that align to form fragmented longitudinal stripes and the last possessing four complete longitudinal stripes (Fig. 7 illustrates two of these patterns). The reconstruction is congruent with Birindelli et al.’s (2013b) proposal of a sister-group relationship between H. lineomaculatus and H. despaxi. On the other hand, the placement of Hypomasticus torrenticola as closely related to H. megalepis isunexpected, given that these species have markedly different mouth positions and that Birindelli et al. (2016) proposed their similarity of coloration (Figs. 1, 7) as potentially convergent. In the same contribution, those authors described Leporinus multimaculatus Birindelli, Teixeira & Britski, 2016, another anostomid with a similar coloration that, according to our results is more closely related to Leporinus tristriatus and L. britskii Feitosa, Santos & Birindelli, 2011. These results contribute to a growing consensus that coloration, mouth position and tooth formulae evolve rapidly and converge frequently within Anostomidae.

FIGURE 7| Geographic range, diversity of body shape and color variation within the expanded concept of Hypomasticus. Inset phylogeny and map demonstrates that the clade of species inhabiting river systems of Brazil’s southeastern coast (lower box) is nested within a paraphyletic assemblage of species inhabiting the Amazon and river systems draining the Guiana Shield. Photographed live specimens include: H. despaxi, ANSP 207676, 62.7 mm SL, Mutura River, Oyapock basin, Brazil; H. lebaili, ANSP 189043, 50.5 mm SL, Lawa River, Marowijne basin, Suriname; H. megalepis, MZUEL 10200, 98.6 mm SL, Uatumã River, Brazil; H. melanostictus, MZUSP 103242, 225 mm SL, Jari River, Brazil; H. mormyrops, MZUEL 8022, 125.0 mm SL, Paraíba do Sul River, Brazil; H. nijsseni, MZUSP uncat. Mutura River, Oyapock basin, Brazil; H. steindachneri, MZUEL 17996, 185.0 mm SL, Pardo River, Brazil. Photographs of H. megalepis, H. melanostictus, H. mormyrops and H. steindachneri by José Birindelli, those of H. despaxi, H. lebaili and H. nijsseni by Mark Sabaj.

The two other major subclades within Hypomasticus together form the sister group to the H. megalepis clade. The first of these contains H. copelandii, H. mormyrops, H. steindachneri and H. thayeri, a collection of species that share three prominent dark blotches along the lateral line scale row (Fig. 7), 12 scale rows around the caudal peduncle and four teeth on each premaxilla and dentary (Birindelli et al., 2020b). Though not included in our analysis, H. santanai Birindelli & Melo, 2020 also belongs to this subclade of Hypomasticus (Birindelli et al., 2020b). A phylogeographic study detected at least two clear genetic lineages within H. copelandii: one in the Paraíba do Sul and another in the Jucuruçu, Mucuri and Doce rivers (Mendes et al., 2022). If H. copelandii represents a complex of closely related species, then this may be the most species-rich subclade within Hypomasticus. The members of this clade also vary substantially in mouth position with H. thayeri, H. mormyrops and H. santanai possessing subterminal or inferior mouths, and H. steindachneri and H. copelandii possessing terminal mouths. Intriguingly, the species with terminal mouths do not appear to be the closest relatives of one another; rather the H. copelandii complex is sister clade to H. thayeri. Like the results described above for the subclade containing H. megalepis, these results suggest that mouth position in Anostomidae is more evolutionary labile than Sidlauskas (2008) previously hypothesized.

The third major subclade within Hypomasticus includes three species: H. granti, H. lebaili, H. melanostictus and the undescribed species from the Juruena. Morphological similarity suggests that several other unsampled species of Leporinus likely belong to the same subclade of Hypomasticus. For example, ichthyologists have long recognized the similarity between Leporinus granti, L nijsseni and L. gomesi (Garavello, 1990; Planquette et al., 1996; Sidlauskas et al., 2011). More recently, Britski, Birindelli (2013) described L. santosi, and considered it closely related to the aforementioned species due to similarities in coloration, dentition, and squamation. Birindelli et al. (2019) later expanded this group to include L. arcus, L. melanostictus,and L. tepui, and diagnosed the clade as having small epidermal dark spots on the anterior portion of the scales of the lateral body surface. These spots are red in live specimens, as in H. nijsseni (Fig. 7), but fade in alcohol. Members of the clade also have terminal mouths with four premaxillary teeth, four or five dentary teeth, relatively few scales in the lateral line (usually fewer than 36), and a dark spot immediately posterior to the sixth infraorbital.

Though molecular data were available for only some of these species, specialists on anostomid taxonomy have consistently recognized their close morphological similarity for decades. Given the clear molecular signal herein for the placement of Leporinus granti and L. lebaili within an expanded concept of Hypomasticus, we transfer all species of the Leporinus granti group into Hypomasticus and consider this solution preferable to leaving several probable members of the group stranded within a non-monophyletic concept of Leporinus. These results call attention to the need to redefine and diagnose Hypomasticus on morphological grounds, since several of the characters that Sidlauskas, Vari (2008) listed as synapomorphic for the genera are clearly homoplastic, and because the genus as recognized herein is not limited to the subterminal and inferior-mouthed species.

Leporinusecuadorensis clade. The basal split within Leporininae exclusive of Insperanos and Hypomasticus separates Leporinus ecuadorensis and L. brunneus Myers, 1950 from the rest. The pairing of those two Leporinus is somewhat unexpected, given that L. ecuadorensis is a relatively deep bodied, trans-Andean species that displays three midlateral blotches, while the cis-Andean L. brunneus has a relatively elongate body and variable color pattern. Leporinus brunneus occurs in the Orinoco system in Venezuela, where it often displays a black lateral stripe (Chernoff et al., 1991), and in several tributaries of the rio Amazonas in Brazil, where it typically lacks any trace of dark blotches or stripes (Ohara et al., 2017). Though this sister-relationship obtains from multiple specimens, the bootstrap support of 81% is relatively low, and Mirande (2019) suggested an alternative placement for L. brunneus as sister to Hypomasticus. We refrain from any nomenclature changes until further research tests the validity of this apparent clade.

The remainder of the radiation divides into two well-supported sister clades. One of these contains Anostomoides, Laemolyta, Rhytiodus and Schizodon, and the other contains Abramites, Brevidens, Megaleporinus, and the remaining species of Leporinus (Fig. 3). Five distinct clades of Leporinus emerge within this larger clade, collectively forming a grade. The non-monophyly of Leporinus is well-known (Sidlauskas, Vari, 2008; Ramirez et al., 2017a,b; Sidlauskas et al., 2021), but some of the relationships among species recovered herein are novel. We discuss each of these five clades within Leporinus below.

Leporinusmelanopleura clade. The basal split subtends the Leporinus melanopleura clade, which includes four species that differ in mouth position, but all possess a prominent midlateral dark stripe that increases in width at the body’s midpoint: L. microphthalmus Garavello, 1989, L. melanopleura, L. amblyrhynchus Garavello & Britski, 1987,and L. melanopleurodes Birindelli, Britski & Garavello, 2013. Birindelli et al. (2013) described L. melanopleurodes and hypothesized a close relationship to L. melanopleura, because both species share a broad (two scale rows in depth) longitudinal stripe running from the snout tip to the caudal fin and are endemic to coastal systems of eastern Brazil. Mirande (2019) also recovered a close relationship between Leporinus microphthalmus and L. amblyrhynchus, but ours is the first study including the four species. We reconstruct L. microphthalmus as sister to the other three, and note that its coloration differs markedly from the other three in that its midlateral stripe is relatively faint and forms from deep-lying dermal pigments (Birindelli, Britski, 2009). While those authors considered L. microphthalmus (from the upper Paraná drainage) to resemble L. marcgravii Lütken, 1875 (São Francisco basin) closely in external morphology, we lacked tissues of L. marcgravii to test whether the closely morphological similarity results from convergence, a sister relationship, or synonymy. Though Birindelli, Britski (2009) also considered L. microphthalmus to resemble L. guttatus Birindelli & Britski, 2009, L. octomaculatus Britski & Garavello, 1993, and L. reticulatus Britski & Garavello, 1993, results herein place the latter two within the L. friderici clade. Tissue samples were not available for L. guttatus,nor for Leporinus amae Godoy, 1980 from southern Brazil, Uruguay and Argentina, which shares with L. melanopleura and L. melanopleurodes a similar body shape, subterminal mouth, lateral dark stripe and lack of dark bars on the dorsum (Almirón et al., 2013).

Leporinus melanopleura, L. melanopleurodes, L. amblyrhynchus and L. amae superficially resemble many other congeners with dark lateral stripes such as L. taeniatus Lütken, 1875, L. taeniofaciatus Britski, 1997, L. geminis Garavello & Santos, 2009, L. unitaeniatus Garavello & Santos, 2009, L. vanzoi Britski & Garavello, 2005, L. microphysus Birindelli & Britski, 2013, L. parvulus Birindelli, Britski & Lima, 2013,and L. sidlauskasi Britski & Birindelli, 2019 (Feitosa et al., 2011; Birindelli et al., 2013a; Britski, Birindelli, 2019). However, those species sort into at least two lineages distantly related to the L. melanopleura clade (Fig. 6). Lateral stripes have evolved several times other within Leporinus and Anostomidae, and species with lateral stripes occur in Anostomus, Brevidens, Hypomasticus, Laemolyta, Leporellus, Pseudanos, Rhytiodus and Schizodon.

Leporinuspachycheilus clade. The next major grouping includes Leporinus pachycheilus and L. julii, a pair of distinctive species with inferior mouths, numerous dark lateral blotches, elongate bodies, small swim bladders (Birindelli, Britski, 2013), and a series of small red epidermal spots linearly arranged posterior to the operculum (Fig. 1). Early authors considered these species as closely related to those previously assigned to subgenus Hypomasticus within Leporinus (Britski, 1976; Santos et al., 1996). When Sidlauskas, Vari (2008) elevated Hypomasticus to generic status, they also transferred these species to that genus (Sidlauskas, Vari, 2008; Toledo-Piza et al., 2024). Our result confirms several other reports that these species do not share a close relation with other Hypomasticus (Ramirez et al., 2017b; Betancur-R. et al., 2019; Mirande, 2019; Birindelli et al., 2020b; Sidlauskas et al., 2021). We allocate them back to Leporinus, based on the compelling evidence that they provide another instance of convergent evolution of inferior jaws within Anostomidae like that occurring in Schizodon nasutus Kner, 1858 (Sidlauskas, Vari, 2008). In all these cases, the transition to an inferior mouth is accompanied by several osteological changes, most prominently the development of a wide lateral trough in the ascending process of the anguloarticular through which the A1 component of the adductor mandibulae muscle runs (character #56 of Sidlauskas,Vari, 2008). Future studies should investigate the functional implications of the repeated evolution of this morphology.

Leporinusfriderici clade. The Leporinus friderici clade (Fig. 6) includes a diverse assemblage of Leporinus with terminal to slightly subterminal mouths, but widely varying color patterns such as vertical bars, longitudinal stripes, and lateral blotches. Though the clade received a bootstrap of just 66%, its posterior probability is high (0.99), and both of its subclades received the maximum possible statistical support (Fig. 6). One subclade includes L. maculatus Müller & Troschel, 1844 and L. ortomaculatus Garavello, 2000.Alternating wide and thin vertical dark bars make the former one of the most distinctive Leporinus (Géry et al., 1988), while the latter displays small lateral blotches in addition to the three large midlateral blotches (Garavello, 2000). No other molecular study has included these species, and this is the first suggestion of a close relationship between them. Sidlauskas, Vari (2008) included both in their osteological study, where they listed L. maculatus as its junior synonym L. pellegrinii Steindachner, 1910 (Toledo-Piza et al., 2024) but those authors were not able to resolve their relationships within a paraphyletic Leporinus.

The other half of the Leporinus friderici clade includes Leporinus lacustris Amaral Campos, 1945, L. parae Eigenmann, 1907, L. taeniatus, L. octofasciatus Steindachner, 1915, L. agassizii Steindachner, 1876, L. piau Fowler, 1941, and Leporinus inexpectatus Britski, Garavello, Oliveira & Birindelli, 2024. Like L. friderici, many of these species display three prominent midlateral blotches or a partial dark lateral stripe connecting those three spots as in Leporinus agassizii. Though not included in our sequencing effort, Leporinus punctatus Garavello, 2000 from the Orinoco has color pattern almost identical to L. agassizii, appears to replace that species in that drainage (Garavello, 2000), and may be synonymous (Sidlauskas, Vari, 2012). Thus, we hypothesize that L. punctatus also belongs to the Leporinus friderici clade.

Other members of this clade differ substantially in color patterns, such L. taeniatus (with a complete lateral stripe and dark bars on the dorsum) and L. octofasciatus (with eight vertical bars formed by deep dermal pigmentation). Steindachner (1915, 1917) described that latter species from coastal rivers of Santa Catarina but it has never again been collected there. That apparent error in the type-locality led several authors (Kner, 1859; Ringuelet, Arámburu, 1961) to erroneously identify specimens of L. octofasciatus from elsewhere as L. fasciatus until Britski, Garavello (1978) redescribed the species using specimens from upper Paraná basin where it is widespread (Dagosta et al., 2024). Despite the superficial similarity in coloration, L. octofasciatus and L. fasciatus do not appear to share a close relationship, and closer examination of the barred pattern reveals substantial structural differences. The bars in Leporinus fasciatus and closely related taxa such as L. desmotes include shallow epidermal pigments and are more intense than the bars in L. octofasciatus (Burns et al., 2017), which are formed entirely from dermal pigments. In addition, L. fasciatus and closely related species have a dark vertical bar on the snout, whereas L. octofasciatus has a longitudinal or diagonal dark stripe (Britski, Garavello, 1978). Those differences in pigmentation reinforce the inference that L. octofasciatus is a distant relative of L. fasciatus despite their superficially similar and apparently convergent coloration.

We note that the composition of this clade largely matches that treated by Silva-Santos et al. (2018) and recovered as monophyletic. Our results agree with theirs in recognizing distinct clades of individuals displaying Leporinus friderici and Leporinus agassizii-like phenotypes and support their hypothesis that the Leporinus friderici group sensu lato includes geographically distinct lineages representing undescribed species. Accordingly, this portion of the Leporinus phylogeny merits intensive taxonomic effort.

Leporinusfasciatus clade.The fourth major clade within Leporinus appears in the overview phylogeny (Fig. 3) as the Leporinus fasciatus clade and includes the type-species of Leporinus, L. novemfasciatus Spix & Agassiz, 1829, objective synonym of L. fasciatus (Bloch, 1794). This well-supported clade (98% bootstrap, posterior probability of 1.0) includes all Leporinus species with vertical bars except for L. octofasciatus (Fig. 6). Thus, the existence of the clade confirms the long-assumed relationships between L. affinis Günther, 1864, L. altipinnis Borodin, 1929, L. bleheri Géry, 1999, L. desmotes, L. enyae Burns, Chatfield, Birindelli & Sidlauskas, 2017, L. fasciatus, L. jatuncochi Ovchynnyk, 1971, L. pearsoni Fowler, 1940, L. tigrinus Borodin, 1929, and L. villasboasorum Burns, Chatfield, Birindelli & Sidlauskas, 2017 (Günther, 1864; Borodin, 1929; Mahnert et al., 1997; Géry, 1999; Sidlauskas, Vari, 2008; Britski, Birindelli, 2016; Burns et al., 2017). Though not sequenced, L. yophorus Eigenmann, 1922 from the Orinoco, which is morphologically similar to the Amazonian L. pearsoni (Fowler, 1940; Géry, 1999), likely belongs to this clade as well. Interestingly, all these barred species form a paraphyletic grade within the Leporinus fasciatus clade as construed herein, which also includes many species with blotched or striped color patterns. Two earlier studies using similar data (Ramirez et al., 2017b; Mirande, 2019) foreshadowed this result when they recovered a blotched species (L. octomaculatus) and a striped species (L. unitaeniatus) together as the sister to a clade including four barred species (L. bleheri, L. fasciatus, L. tigrinus and L. affinis), to the exclusion of L. desmotes and L. jatuncochi.

Even so, the number of sequenced species that nest within the L. fasciatus clade without displaying barred coloration is higher than expected. In addition to L. octomaculatus and L. unitaeniatus,these include L. multimaculatus, L. britskii, L. tristriatus, L. reticulatus, L. cylindriformis Borodin, 1929, L. nigrotaeniatus (Jardine, 1841), L. vanzoi and L. geminis. In addition to revealing additional instances of rapid color pattern evolution within the family, this pattern of relationships challenges Sidlauskas, Vari’s (2008) interpretation of the shared possession of nine branched pelvic-fin rays (their character #111) as an unreversed synapomorphy linking Leporinus fasciatus, L. desmotes, L. jatuncochi and L. falcipinnis Mahnert, Géry & Muller, 1997 (=L. altipinnis). Though anostomids are otherwise remarkably conservative in fin-ray counts (Sidlauskas, Vari, 2008), our phylogeny implies either multiple gains or multiple losses of the ninth pelvic ray, and perhaps a curious link between the possession of the extra fin ray and a color pattern involving many vertical bars.

Considering the pattern of relationships within the Leporinus fasciatus clade in detail, we recover a clade containing L. pearsoni, L. villasboasorum, L. enyae, L. jatuncochi and L. desmotes as sister to the remainder. This pattern of relationships matches that recovered by Burns et al. (2017), except for a shallow trichotomy containing two specimens of L. jatuncochi and one of L. desmotes, where the earlier study obtained a monophyletic L. jatuncochi,but with weak support. Though these species are morphologically and genetically similar and possibly synonymous, an unusual circumpeduncular scale polymorphism and disjunct geographical distribution in L. desmotes complicates the situation (Burns et al., 2017).

We recover Leporinus altipinnis as sister to the remainder of the Leporinus fasciatus clade, but with low support. In a more taxonomically restricted analysis, Burns et al. (2017) recovered this species as forming a clade with L. fasciatus and L. affinis, and our results do not strongly reject that more intuitive placement. The three species have very similar color patterns and meristic counts (Britski, Birindelli, 2016), and L. altipinnis was described originally as a subspecies of L. fasciatus (Borodin, 1929). While L. altipinnis is distinguishable from L. fasciatus and L. affinis on the basis of its falcate dorsal fin, deeply forked caudal fin, greater number of dark bars in large adults (Mahnert et al., 1997; Britski, Birindelli, 2016) and marked genetic divergence (Burns et al., 2017), the similarities among that trio are nevertheless striking.

Our results show a well-supported clade including L. affinis, L. bleheri, L. fasciatus, L. tigrinus,and an undescribed species from the Tapajós River (Fig. 6). The close relationships of the first three species has long been assumed (Borodin, 1929; Burns et al., 2017; Boaretto et al., 2024) but in his original description of L. bleheri, Géry (1999) hypothesized a closer relationship of that species with L. desmotes, L. pearsoni and L. yophorus, and mentioned L. tigrinus only in passing. Though the node separating the lineage from the Tapajós (LBP 16116) from L. affinis is shallow, the putatively new species has a distinctive pattern of dark spots at the centers of its scales unlike L. affinis and the other members of the Leporinus fasciatus clade. We also note additional structure separating an individual of L. fasciatus from the Marowijne drainage of Suriname from a cluster of individuals ranging widely from the upper Amazon of Peru to the Nickerie drainage of Suriname. It is likely that one of these clusters represents an additional undescribed species. Given that the type-locality of L. fasciatus is simply “Suriname” (Bloch, 1794), a detailed revision will be needed to determine to which lineage the holotype likely belongs. A complicating factor is L. novemfasciatus, the type-species of Leporinus. This species was described from somewhere in the lower Amazon basin (Spix, Agassiz, 1829–31) and is considered a synonym of L. fasciatus (Mahnert et al., 1997; Garavello, Britski, 2003). If the Marowijne specimen of L. fasciatus in our analysis (MHNG 2717.03) represents the true L. fasciatus, L. novemfasciatus may be a valid species applicable to one or more individuals in our wide-ranging clade of L. fasciatus.

Three clades of non-barred species exist within the Leporinus fasciatus clade (Fig. 6). The first includes L. britskii, L. multimaculatus and L. tristriatus,all of which possess subterminal or subinferior mouth positions. Leporinus britskii and L. tristriatus both possess a prominent dark midlateral stripe and additional dark pigmentation dorsal and ventral to that stripe. In L. britskii that pigmentation most commonly forms a lateral series of dark blotches. Those blotches are slightly separated in specimens from the Jari River and completely fused into a midlateral stripe in specimens from the Tapajós basin (Feitosa et al., 2011). In L. tristriatus the blotches typically unite to form three lateral stripes with irregular margins (Birindelli, Britski, 2013). Pigmentation in L. multimaculatus more closely resembles that of Hypomasticus megalepis, H. torrenticola or Leporinus pachycheilus, with a series of three large midlateral dark blotches surrounded by smaller blotches scattered across the lateral surface of the body. As Birindelli et al. (2016) hypothesized, this color pattern has evolved many times in Anostomidae. Interestingly, specimens of L. multimaculatus collected in the adjacent Tapajós and Xingu drainages sort into two clades that are not each other’s closest relatives (Fig. 6), suggesting that another undescribed species may await discovery within the present concept of that species.

The next subclade includes L. reticulatus and L. octomaculatus, sympatric species that share almost identical meristic counts but differ markedly in coloration and adult head shape (Britski, Garavello, 1993; Birindelli, Britski, 2009). Leporinus reticulatus displays numerous dark blotches on the side of the body that vary widely in size, placement, and shape, creating an overall reticulate pattern similar to many species of Characidium Reinhardt, 1867(Birindelli, Britski, 2009). This species also displays a remarkable positive allometry of the snout otherwise known only in Megaleporinus elongatus (Valenciennes, 1850) (Birindelli, Britski, 2009), and L. amblyrhynchus (Britski et al., 2012). Leporinus octomaculatus possesses eight roughly circular dark blotches along the lateral line scale row, with additional series of smaller dark blotches distributed dorsally and ventrally; it does not display marked snout allometry. Though we obtained weak support for this relationship (bootstrap of 38, posterior probability of 0.9), Ito et al. (2023) also recovered the pair as closely related in a COI gene tree, though not as sister-taxa. Based on similarity of COI sequences, the recently described L. oliveirai Ito, Souza-Shibatta, Venturieri & Birindelli, 2023 likely also belongs to this clade (Ito et al., 2023).

The third non-barred subclade includes L. cylindriformis, L. nigrotaeniatus, L. vanzoi, L. geminis and L. unitaeniatus. Though not sequenced herein, L. sidlauskasi is morphologically similar to L. vanzoi, L. geminis and L. unitaeniatus and differs primarily in meristic counts, the absence of a fourth premaxillary tooth, and nuances of coloration (Britski, Birindelli, 2019). It diverges only slightly from L. vanzoi at the COI locus (Ito et al., 2023) and likely belongs to this clade. All of these species possess fusiform bodies and are among the most slender members of Leporinus, hence the name of L. cylindriformis, one of the earliest described species (Borodin, 1929). However, not all elongate Leporinus belong to this clade, and the admittedly artificial “L. cylindriformis group” treated by Sidlauskas et al. (2011) turns out to include several species placed elsewhere in our phylogeny. Within the subclade containing L. cylindriformis, only that species possesses lateral blotches. The remaining species in this subclade possess a partial (L. nigrotaeniatus) or complete (L. vanzoi, L. geminis, L. unitaeniatus and L. sidlauskasi) dark stripe along the lateral-line scale row. Based on modest COI sequence similarity, one more striped species (Leporinus bistriatus Britski, 1997) likely belongs to the Leporinus fasciatus clade sensu lato (Ito et al., 2023). However, the substantial barcode divergence between that species and all others leaves ambiguity in the exact phylogenetic placement.

Leporinusjamesi clade. We refer Leporinus jamesi, L. amazonicus Santos & Zuanon, 2008 and L. arimaspi Burns, Frable & Sidlauskas, 2014 to the final major subclade within Leporinus as conceived herein. All three species in the Leporinus jamesi clade possess a prominent dark midlateral blotch ventral to the dorsal fin and bars on the dorsum, with other pigmentation limited to a fainter blotch along the lateral line scale row dorsal to the anal fin origin or on the caudal peduncle (Burns et al., 2014; Garavello et al., 2014). Leporinus jamesi and L. amazonicus are morphologically similar, possessing higher lateral-line scale counts (42 or more scales) and notably tapered bodies with the greatest depth at the dorsal fin origin or slightly before, though L. amazonicus grows to much larger sizes (Garavello et al., 2014). Leporinus arimaspi from the Orinoco is distinct from those two species but bears strong morphological similarity to two unsampled species: Leporinus niceforoi Fowler, 1942 from the upper Amazon of Colombia (and possibly Ecuador and Peru) and L. aripuanaensis Garavello & Santos, 1981 from the Aripuanã River, a right-bank tributary of the Madeira River in the Amazon basin (Burns et al., 2014). Those two species possess the intense midlateral blotch ventral to the dorsal fin along with other intense blotches, either located dorsal to the anal-fin origin and on the caudal peduncle (L. niceforoi) or just on the caudal peduncle (L. aripuanaensis) (Fowler, 1943; Garavello, Santos, 1992). These may well belong to the L. jamesi subclade as construed herein.

Abramites, Brevidens, and Megaleporinus. The next clade unites Abramites, Brevidens and Megaleporinus (Fig. 8). Based on possession of a distinctive ZZ/ZW sex chromosome system, several authors hypothesized a close relationship among the species originally known as Leporinus obtusidens Valenciennes, 1837, L. macrocephalus Garavello & Britski, 1988, and L. conirostris Steindacher, 1875, all of which are now in Megaleporinus (Galetti Jr. et al., 1981; Galetti Jr., Foresti, 1986; Galetti Jr. et al., 1991; Molina et al., 1998; Venere et al., 2004). Sidlauskas, Vari (2008) recognized the close affinity between Abramites and L. striatus (now in Brevidens) on osteological grounds, with Bogan et al. (2012) later placing the fossil †L. scalabrinii as sister to Abramites and that pair sister to L. striatus. In that analysis, the clearest synapomorphy uniting these three taxa was the presence of three teeth on each premaxilla, unlike the four possessed by most anostomids. Ramirez et al. (2016) found strong support for this clade using molecular data, a result foreshadowed a year earlier in an analysis that omitted Abramites but included the other taxa in question (Avelino et al., 2015). Ramirez et al. (2017b) then described the genus Megaleporinus to contain the large bodied species of Leporinus with the distinctive dental formula and ZW sex determination system. Subsequent analyses have supported the monophyly of Megaleporinus and its affinity with Abramites and Brevidens (as Leporinus striatus) (Mirande, 2019; Birindelli et al., 2020b; Sidlauskas et al., 2021; Melo et al., 2022).

FIGURE 8| Relationships among species of Abramites, Brevidens, Megaleporinus and members of the Leporinus jamesi clade, based on the maximum likelihood reconstruction. Gray shading on inset phylogeny indicates the region detailed. Colored circles represent ranges of bootstrap values. Boxed numeral 6 indicates the most recent common ancestor of the Leporinus jamesi clade. Diagrams illustrating body shape and color pattern by José Birindelli and Peter van der Sleen, used with permission.

The strength of the morphological and molecular support for this relationship led us to remove Brevidens striatus from the non-monophyletic concept of Leporinus. While it would have been possible to expand Abramites to include this entire clade, this would obscure the morphological and ecological divergence of the two species currently assigned to Abramites. These herbivorous species possess distinctive diamond-shaped bodies (Fig. 1) and have been suggested to represent a lineage with one of the highest rates of character evolution in the family (Sidlauskas, Vari, 2008). As detailed in the diagnosis above, Brevidens also demonstrates substantial anatomical distinction, including a unique morphology of the dentary teeth.

The erection of Brevidens mandates re-evaluation of the fossil taxon †Leporinus scalabrinii,which is now the only species assigned to Leporinus that falls within the clade containing Megaleporinus, Brevidens and Abramites. Though Bogan et al. (2012) placed this taxon as sister to Abramites, Sidlauskas et al. (2021) obtained a position sister to Megaleporinus [a genus that was not yet recognized at the time of Bogan et al.’s (2012) writing]. To help relieve the non-monophyly of Leporinus, †L. scalabrinii should clearly be transferred to either Abramites or Megaleporinus. We prefer to defer that decision until a formal re-examination of the fossil can compare it to a much broader range of taxa within Megaleporinus and infer an expanded osteological topology for that section of Anostomidae.

Within Megaleporinus itself, we reconstruct nearly the same set of relationships as did two earlier studies (Ramirez et al., 2017b; Birindelli et al., 2020a). Megaleporinus muyscorum (Steindachner, 1900) from the trans-Andean drainages of Colombia is sister to the remainder of the family. The remainder of the taxa divide among two major subclades. The first of these contains Megaleporinus trifasciatus (Steindachner, 1876) Megaleporinus macrocephalus, and two enigmatic lineages tentatively identified as M. cf. muyscorum and M. cf. trifasciatus that are geographically separated from the named species [see also Ramirez et al. (2017a)]. Megaleporinus cf. muyscorum (LBP 3030) is from the Orinoco drainage of Venezuela, while the specimen of Megaleporinus cf. trifasciatus originated in the Araguaia drainage of Brazil, far from the middle and upper Amazon regions from which the species and its junior synonym Leporinus wolfei Fowler, 1939 (Steindachner, 1876; Fowler, 1939) were described. Our specimens representing the true M. trifasciatus originate from the Nanay River of Peru.

The second major subclade includes M. conirostris as sister to the remaining species, which divide amongst a clade containing M. brinco (Birindelli, Britski & Garavello, 2013), M. garmani (Borodin, 1929) and M. reinhardti (Lütken, 1875) on one hand, and a clade containing M. piavussu (Britski, Birindelli & Garavello, 2012), M. obtusidens, and M. elongatus on the other.The first of these clades includes three morphologically distinctive species. Megaleporinus brinco possesses a subterminal mouth unlike most members of the genus, a unique color pattern consisting of notable red pigmentation above the pectoral-fin base, and dark pigmentation on the dorsal and ventral margins of the scales forming a series of wavy longitudinal lines (Birindelli et al., 2013). Like many other anostomid species, M. brinco has three dark midlateral blotches increasing in size posteriorly, with the first blotch sometimes absent. Birindelli et al. (2013) considered this coloration to be somewhat like that of M. garmani and M. conirostris, which possess a single dark blotch on the caudal peduncle. Megaleporinus garmani has a completely inferior mouth, which has led other authors to consider it closely related to Hypomasticus mormyrops (Borodin, 1929), Leporinus pachycheilus (Britski, 1976) or both (Sidlauskas, Vari, 2008). Unlike the other two species in this small clade, Megaleporinus reinhardti has a terminal mouth and a distinctive dark longitudinal line connecting the midlateral dark blotches that is like L. agassizii, L. moralesi Fowler, 1942, and L. nigrotaeniatus. Our results differ slightly from those of Ramirez et al. (2017b) in reconstructing M. brinco and M. garmani as sister species, albeit with weak support. The earlier study obtained a sister relationship between M. garmani and M. reinhardti.

Megaleporinus piavussu, M. obtusidens, and M. elongatus form a clade (Fig. 8). These morphologically similar species share the presence of three dark midlateral blotches and faint vertical bars formed of deep-lying dermal pigments (Fig. 1). They are particularly known for their large body size, value to artisanal fisheries, and potamodromous behavior (Agostinho et al., 2004). The recently described species M. gaiero Birindelli, Britski & Ramirez, 2020 with a somewhat more subterminal mouth and an irregular longitudinal stripe connecting the lateral blotches, likely belongs to this clade. Several relationships in this portion of the phylogeny receive low bootstrap support here (Fig. 8) and in earlier studies (Ramirez et al., 2017b, 2020; Birindelli et al., 2020a), indicating the need for future research to clarify the precise relationships among the species of this economically important genus.

Anostomoides, Laemolyta, Rhytiodus and Schizodon. The final major clade within Leporininae unites Anostomoides, Laemolyta, Rhytiodus,and Schizodon (Figs. 3, 9). The union of those four genera emerged from Sidlauskas, Vari’s (2008) osteological study and has been consistently obtained thereafter (Ramirez et al., 2017a; Burns, Sidlauskas, 2019; Mirande, 2019; Sidlauskas et al., 2021; Melo et al., 2022), or sometimes to the exclusion of Anostomoides (Ramirez et al., 2016, 2017b; Mirande, 2019).

FIGURE 9| Relationships among species of Anostomoides, Laemolyta, Rhytiodus and Schizodon, based on the maximum likelihood reconstruction. Gray shading on inset phylogeny indicates the region detailed. Colored circles represent ranges of bootstrap values. Diagrams illustrating body shape and color pattern by José Birindelli and Peter van der Sleen, used with permission.

Despite the widespread agreement about the existence of the clade containing Anostomoides, Laemolyta, Schizodon and Rhytiodus, studies have differed substantially in their placement of that clade within the backdrop of a paraphyletic Leporinus. We reconstruct it as sister to the majority of Leporinus plus Abramites, Brevidens and Megaleporinus, similar to the placement obtained by Mirande (2019) in a total evidence analysis. Other analyses using morphological data have inferred a more nested position (Sidlauskas, Vari, 2008; Sidlauskas et al., 2021). Using multilocus data, Ramirez et al. (2017b) placed it sister to a clade containing the Leporinus fasciatus, L. friderici and L. pachycheilus clades as conceived herein, but with very low statistical support. Phylogenomic studies tend to place this taxon as sister to Abramites+Brevidens+Megaleporinus, albeit with limited taxon sampling in the studies completed to date (Betancur-R. et al., 2019; Melo et al., 2022).

Studies tend to agree much more on the genus-level relationships. Sidlauskas et al. (2021) evaluated the phylogenetic placement of the monotypic Anostomoides on morphological and molecular grounds and obtained that genus as sister to Laemolyta plus Schizodon and Rhytiodus (Fig. 9). The placement of Laemolyta as sister to a clade containing Schizodon and Rhytiodus is also broadly congruent with several other molecular (Ramirez et al., 2016, 2017b, 2020; Burns, Sidlauskas, 2019; Melo et al., 2022) and total evidence studies (Mirande, 2019; Sidlauskas et al., 2021). It is consistent with morphology-only studies (Sidlauskas, Vari, 2008; Dillman et al., 2016) other than for their inclusion of the subfamily Anostominae within the clade also containing those three genera. Schizodon has been previously recognized as the sister lineage to Rhytiodus based on osteological characters such as the hypertrophy of the pharyngeal tooth plates (Sidlauskas, Vari, 2008); our molecular results confirm that relationship.

We obtained Laemolyta garmani (Borodin, 1931) as sister to a clade containing L. fernandezi Myers, 1950, and L. proxima (Garman, 1890). These relationships concord with the results of the only other studies to sample this genus densely (Ramirez, Galetti Jr., 2015; Ramirez et al., 2020).

Our results confirm the monophyly of the distinctive genus Rhytiodus, the members of which possess highly elongate bodies (Fig. 1), a unique form of multicuspidate dentition, and herbivorous dietary niches (Santos 1980, 1981). We obtained strong support for the presence of two species within the genus: the large-scaled Rhytiodus argenteofuscus Kner, 1858 and R. microlepis Kner, 1858, which possesses the smallest scales of any anostomid. At the time of this writing another small-scaled species in the genus (Rhytiodus lauzannei Géry, 1987) is considered valid (Toledo-Piza et al., 2024), with that species putatively distinguished from R. microlepis by body depth in the pre-pelvic region (Géry, 1987). In our sampling, the deep bodied voucher specimen LBP 4239 fits the morphometric description of R. lauzannei while the very slender LBP 9781 and ANSP 182603 fit the concept of L. microlepis. However, the three samples diverge less than 0.3% across the six-gene dataset. Ramirez et al. (2020) also obtained very minor genetic differences between these nominal species using a distinct dataset. This evidence suggests that R. lauzannei is a junior synonym of R. microlepis, with the apparent morphometric differences arising from ecomorphology, sexual dimorphism, or even the fullness of the stomach. A test of that conjecture would require specimens from the type-locality of R. lauzannei in the Mamoré system of Bolivia, which we currently lack. For the time being, we assign all three voucher specimens to R. microlepis.

Within Schizodon, we obtain two well supported subclades, one containing S. knerii (Steindachner, 1875) and S. nasutus (herein called the Schizodon nasutus clade) and one containing S. borellii (Boulenger, 1900), S. fasciatus Spix & Agassiz, 1829& S. intermedius Garavello & Britski, 1990, S. scotorhabdotus Sidlauskas, Garavello & Jellen, 2007, S. vittatus (Valenciennes, 1850) and the recently described S. trivittatus Garavello, Ramirez, Oliveira, Britski, Birindelli & Galetti, 2021 (herein called the Schizodon fasciatus clade). In the larger subclade, we reconstruct S. scotorhabdotus as sister to the remaining five species, and S. trivittatus and S. vittatus as sister to S. fasciatus plus S. intermedius and S. borellii. These relationships are broadly congruent with results reported by Ramirez et al. (2020), who also noted that the species in the former clade possess non-descript color patterns consisting primarily of an elongate dark mark on the caudal peduncle, while those in the second clade possess a series of vertically elongate blotches, a dark midlateral stripe, or both. However, the reconstructions differ in the placement of S. trivittatus and S. vittatus. Ramirez et al. (2020) and Garavello et al. (2021) obtained the couplet of S. trivittatus (listed as S. aff. vittatus in the earlier paper)and S. vittatus as sister to the similarly patterned S. fasciatus, while we obtain the arrangement described above. Both studies yielded low statistical support in this section of the phylogeny, suggesting that either topology is plausible and that both studies may be attempting to infer the sequence of a relatively rapid series of speciation events from limited data.

Final comments and future directions. With approximately two thirds of the known anostomid species now included, this phylogeny represents the most detailed picture of relationships among one of the largest families of characoid fishes. It reveals the major patterns in the evolutionary history of the radiation, fills in many fine-scale relationships and sets the stage for additional studies of taxonomy, systematics, biogeography, and evolutionary diversification. For example, these densely sampled relationships will support further investigations about the tempo and mode of morphological evolution, convergence, and the rate of speciation. All three processes appear to operate differently within Anostomidae in comparison to other characiform clades (Sidlauskas, Vari, 2008; Lofeu et al., 2021; Melo et al., 2022) but rate variation within the family remains completely untested. Such studies are now within reach.

The relationships that we report herein differ in some important ways from earlier studies. Certainly, this rich dataset allows us to see further and more deeply than Sidlauskas, Vari (2008) could using osteological information from a much smaller sample of species and with no prior information about relationships other than the taxonomically limited study of Winterbottom (1980). However, comparisons of historical and recent reconstructions can fall into a trap of focusing too much on what earlier studies got wrong, rather than celebrating what they got right. Relationships first postulated using osteology and confirmed herein with molecular data include the early divergence of Leporellinae, the monophyly of Anostominae, the non-monophyly of Leporinus, the reality of Hypomasticus and its early divergence, the close affinity of Schizodon, Rhytiodus, Laemolyta and Anostomoides, the close relationship between Brevidens and Abramites, and the validity of Pseudanos and Petulanos.

The two greatest disagreements between the morphological and molecular trees involve putative clades united by character states that co-vary with mouth position, and thus with trophic ecology. For example, all the character states that Sidlauskas, Vari (2008) postulated as synapomorphies of Hypomasticus involve the orientation or form of oral jaw bones or the anterior bones of the neurocranium that articulate with the oral jaws. The characters that those authors proposed to unite the anostomine genera with Laemolyta (their clade 21) or with the broader clade also including Schizodon and Rhytiodus (their clade 14) involve primarily aspects of dentition, the form of the dentary, the orientation of the premaxilla, or elements of coloration. In both cases, the character suite includes morphologies with clear trophic significance and that clearly evolve rapidly and differ substantially even among closely related species throughout Anostomidae. At least some anostomid species also exhibit marked developmental plasticity in mouth position and a clear functional link between mouth position and foraging mode (Bonini-Campos et al., 2019; Lofeu et al., 2021, 2024), implying the inherent capacity for these features to evolve quickly and for distant lineages to adapt equivalently to similar niches. Anatomical convergence can of course mislead osteological phylogenetics (Zou, Zhang, 2016; van den Ende et al., 2023). As such, we should perhaps be unsurprised that the topology resulting from morphology alone tend to group species with similar trophic ecologies, such as in the original concept of Hypomasticus. The integration of molecular and morphological information has revealed an even richer biological story in which distantly related lineages within the family have converged on every orientation of the mouth, from superior to inferior.

The next phase of phylogenetic studies in Anostomidae should expand molecular and morphological data collection throughout the approximately fifty species that have not yet been analyzed. For example, the four species transferred to Hypomasticus herein based on phenotypic similarity to sequenced species need that placement verified on molecular grounds. Data collection for the missing species of Leporinus holds similar importance, as the final breakup of that artificial genus will depend upon complete or nearly complete sampling of the known species. While we have hypothesized the probable placement of many unsampled taxa in the discussion above, there are some for which we hesitate to even speculate. For example, Leporinus moralesi is a rare and enigmatic species with bicuspid dentition unlike that possessed by other anostomids, and which has at times been considered to belong to a distinct subgenus (Géry, 1977). Its placement within the diffuse concept of Leporinus is undetermined. Leporinus microphysus strongly resembles L. taeniatus and L. unitaeniatus in coloration (Birindelli, Britski, 2013), but those species are widely separated phylogenetically. Is it closely related to one of those two species, or does its unusually small swimbladder (Birindelli, Britski, 2013) hint at a close relationship with L. julii and L. pachycheilus, which also possess reduced bladders and inhabit benthic niches in rapid currents (Birindelli, Britski, 2013)?

Some aspects of the deeper structure of the anostomid phylogeny still require additional investigation. These include determining the proper rooting of the anostomine subtree, testing the reality of the Leporinus ecuadorensis clade, exploring the placement of the clade containing Anostomoides, Laemolyta, Rhytiodus and Schizodon, and verifying the status of Insperanos as sister to the remainder of Leporininae. Other lineages would benefit from species-level revisions, most notably Leporellus and the Leporinus friderici complex, but by no means limited to these. All such investigations would benefit from a total evidence approach combining and comparing phylogenomic data with expanded anatomical datasets (Keating et al., 2023).

Despite the many gaps remaining in our knowledge, or perhaps because of them, this is an exciting and dynamic time for the systematics of Anostomidae. In the past fifteen years, our knowledge of their relationships has gone from a near tabula rasa to a comprehensive picture of the relationships uniting a hundred species. These animals differ remarkably and converge repeatedly in their dentition, mouth position, diet, ecology, and coloration, and we are just beginning to unravel the causes of that incredible diversification. With this phylogeny and its future iterations as our guide, the next fifteen years of systematic study promise to resolve the remaining questions about evolutionary relationships, reveal many more new species, and determine why this family spins out species so quickly, in so many forms and colors most beautiful.

Acknowledgments

We dedicate this effort to our mentor, collaborator, and friend Richard Vari (1949–2016), who began this project with us more than a decade ago and sadly passed away long before its conclusion. We also thank the numerous fieldworkers who collected specimens and tissues and the collection managers and curators who supplied those materials to our study. These include Mariangeles Arce H. (ANSP), Jonathan Armbruster and David Werneke (AUM), Leo Smith, Caleb McMahan, Kevin Swagel and Susan Mochel (FMNH), Lucia Rapp Py-Daniel and Renildo Ribeiro de Oliveira (INPA), Renato Devidé and Ricardo Teixeira (LBP), Tiago Pessali (MCNIP), Sonia Fisch-Muller and Raphäel Covain (MHNG), Mario de Pinna and Michel Gianeti (MZUSP), Hernán Lopez Fernández (formerly at ROM), Christian Cramer (UFRO), Dianne Pitassy, Jeff Williams, Kris Murphy, Lisa Palmer and Jeff Clayton (USNM),. We thank Fernando Jerep, Nathan Lujan, Leandro Sousa and Peter Van der Sleen for permission to reproduce their photographs and drawings. NSF Grant DEB–1257898 to BLS supported the effort of BLS, MDB, KH, BF and CBD and funded much of the sequencing. BFM received grants from FAPESP #11/08374–1 and #13/16436–2 and the AMNH Axelrod Research Curatorship. JLOB received a research grant from CNPq (process 308846/2023–0), and funds from Taxonline (Fundação Arauacária via NAPI program). CO received grants from FAPESP # 2020/13433–6, CNPq # 306054/2006–0 and #441128/2020–3, and Pro-Reitoria de Pesquisa da Universidade Estadual Paulista Júlio de Mesquita Filho (Prope-UNESP). Fieldwork was supported in part by All Catfish Species Inventory (NSF DEB–0315963, Senior Personnel MHS), the iXingu Project (NSF DEB–1257813, PI MHS) and the Smithsonian Institution’s Biodiversity of the Guiana Shield program (PI BLS). We also gratefully acknowledge a 2023–2024 US-Brazil Fulbright Fellowship that allowed BLS to visit JLOB in Brazil and complete this contribution.

References

Abe KT, Mariguela TC, Avelino GS, Castro RMC, Oliveira C. Multilocus molecular phylogeny of Gasteropelecidae (Ostariophysi: Characiformes) reveals the existence of an unsuspected diversity. Mol Phylogenet Evol. 2013;69(3):1209–14. https://doi.org/10.1016/j.ympev.2013.07.005

Abe KT, Mariguela TC, Avelino GS, Foresti F, Oliveira C. Systematic and historical biogeography of the Bryconidae (Ostariophysi: Characiformes) suggesting a new rearrangement of its genera and an old origin of Mesoamerican ichthyofauna. BMC Evol Biol. 2014;14:152. https://doi.org/10.1186/1471-2148-14-152

Agostinho AA, Gomes LC, Suzuki HI, Júlio Jr. HF. Migratory fishes of the upper Paraná river basin. In: Carolsfeld J, Harvey B, Ross C, Baer A, editors. Migratory fishes of South America. Victoria: World Fisheries Trust; 2004. p.19–98.

Aguilar CT, Galetti Jr. PM. Chromosome mapping of 5S rRNA genes differentiates Brazilian populations of Leporellus vittatus (Anostomidae, Characiformes). Genet Mol Biol. 2008; 31:188–94. https://doi.org/10.1590/S1415-47572008000200004

Aguilera O, Lundberg J, Birindelli J, Sabaj Pérez M, Jaramillo C, Sánchez-Villagra MR. Palaeontological evidence for the last temporal occurrence of the ancient western Amazonian River outflow into the Caribbean. PLoS ONE. 2013; 8(9):e76202. https://doi.org/10.1371/journal.pone.0076202

Albert JS, Reis RE, editors. Historical biogeography of Neotropical freshwater fishes. Berkeley, Los Angeles and London: University of California Press; 2011.

Almirón AE, Casciotta JR, Bruno MC, Piálek L, Doubnerová K, Říčan O. First record and supplementary description of Leporinus amae Godoy, 1980 (Characiformes: Anostomidae) from freshwaters in Argentina. Check List. 2013; 9(4):820–22. https://doi.org/10.15560/9.4.520

Avelino GS, Britski HA, Foresti F, Oliveira C. Molecular identification of Leporinus from the south portion of South America. DNA Barcodes. 2015; 3(1):98–109.

Betancur-R. R, Arcila D, Vari RP, Hughes LC, Oliveira C, Sabaj MH et al. Phylogenomic incongruence, hypothesis testing, and taxonomic sampling: the monophyly of characiform fishes. Evolution. 2019; 73(2):329–45. https://doi.org/10.1111/evo.13649

Betancur-R. R, Broughton R, Wiley E, Carpenter K, López J, Li C et al. The tree of life and a new classification of bony fishes. PLOS Currents TOL. 2013; 5:1–33. https://doi.org/10.1371/currents.tol.53ba26640df0ccaee75bb165c8c26288

Birindelli JLO, Britski HA. New species of the genus Leporinus Agassiz (Characiformes: Anostomidae) from the rio Curuá, rio Xingu basin, Serra do Cachimbo, Brazil, with comments on Leporinus reticulatus. Neotrop Ichthyol. 2009; 7(1):1–10. https://doi.org/10.1590/S1679-62252009000100001

Birindelli JLO, Britski HA. Two new species of Leporinus (Characiformes: Anostomidae) from the Brazilian Amazon, and redescription of Leporinus striatus Kner 1858. J Fish Biol. 2013; 83(5):1128–60. https://doi.org/10.1111/jfb.12206

Birindelli JLO, Britski HA, Garavello JC. Two new species of Leporinus Agassiz (Characiformes: Anostomidae) from eastern basins of Brazil, and redescription of L. melanopleura Günther. Neotrop Ichthyol. 2013a; 11(1):9–23. https://doi.org/10.1590/S1679-62252013000100002

Birindelli JLO, Britski HA, Provenzano F. New species of Leporinus (Characiformes: Anostomidae) from the highlands of the Guiana Shield in Venezuela. Neotrop Ichthyol. 2019; 17(2):e190022. https://doi.org/10.1590/1982-0224-20190022

Birindelli JLO, Britski HA, Ramirez JL. A new endangered species of Megaleporinus (Characiformes: Anostomidae) from the Rio de Contas basin, eastern Brazil. J Fish Biol. 2020a; 96(6):1349–59. https://doi.org/10.1111/jfb.14299

Birindelli JLO, Lima FC, Britski HA. New species of Pseudanos Winterbottom, 1980 (Characiformes: Anostomidae), with notes on the taxonomy of P. gracilis and P. trimaculatus. Zootaxa. 2012; 3425(1):55–68. https://doi.org/10.11646/zootaxa.3425.1.4

Birindelli JLO, Peixoto LAW, Wosiacki WB, Britski HA. New species of Hypomasticus Borodin, 1929 (Characiformes: Anostomidae) from tributaries of the lower rio Amazonas, Brazil. Copeia. 2013b; 2013(3):464–69. https://doi.org/10.1643/CI-12-148

Birindelli JLO, Melo BF, Ribeiro-Silva LR, Diniz D, Oliveira C. A new species of Hypomasticus from eastern Brazil based on morphological and molecular data (Characiformes, Anostomidae). Copeia. 2020b; 108(2):416–25. https://doi.org/10.1643/CI-19-335

Birindelli JLO, Teixeira TF, Britski HA. Two new species of Leporinus Agassiz, 1929 (Characiformes: Anostomidae) from tributaries of the lower Amazon basin in Brazil. Zootaxa. 2016; 4178(1):97–115. https://doi.org/10.11646/zootaxa.4178.1.4

Bloch ME. Naturgeschichte der ausländischen Fische. Berlin. 1794; 8:1–174. https://doi.org/10.5962/bhl.title.63303

Boaretto MP, Souza-Shibatta L, Birindelli JLO. A systematic assessment of Leporinus tigrinus (Characiformes: Anostomidae) using morphological and molecular data. Zootaxa. 2024; 5432(4):535–54. https://doi.org/10.11646/zootaxa.5432.4.4

Bogan S, Sidlauskas B, Vari RP, Agnolin F. Arrhinolemur scalabrinii Ameghino, 1898, of the late Miocene – a taxonomic journey from the Mammalia to the Anostomidae (Ostariophysi: Characiformes). Neotrop Ichthyol. 2012; 10(3):555–60. https://doi.org/10.1590/S1679-62252012000300008

Bonini-Campos B, Lofeu L, Brandt R, Kohlsdorf T. Different developmental environments reveal multitrait plastic responses in South American Anostomidae fish. J Exp Zool Part B Mol Dev Evol. 2019; 332(7):238–44. https://doi.org/10.1002/jez.b.22905

Borodin NA. Notes on some species and subspecies of the genus Leporinus Spix. Mem Mus Comp Zool Harvard Coll. 1929; 50(3):269–90.

Britski HA. Sobre uma nova espécie Leporinus da Amazônia. Acta Amaz. 1976; 6(4):87–89. https://doi.org/10.1590/1809-43921976064s087

Britski HA, Birindelli JLO. A new species of Leporinus Agassiz, 1829 (Characiformes: Anostomidae) from the rio Tocantins, Brazil. Neotrop Ichthyol. 2013; 11(1):25–32. https://doi.org/10.1590/S1679-62252013000100003

Britski HA, Birindelli JLO. Redescription of Leporinus altipinnis, a senior synonym of L. falcipinnis, and comments on L. holostictus (Characiformes: Anostomidae). Ichthyol Explor Freshw. 2016; 27(1):25–40.

Britski HA, Birindelli JL. Description of a new species of Leporinus (Characiformes: Anostomidae) from the Rio Tapajós basin, Brazil. Zootaxa. 2019; 4603(1):183–91. https://doi.org/10.11646/zootaxa.4603.1.10

Britski HA, Birindelli JLO, Garavello JC. A new species of Leporinus Agassiz, 1829 from the upper Rio Paraná basin (Characiformes, Anostomidae) with redescription of L. elongatus Valenciennes, 1850 and L. obtusidens (Valenciennes, 1837). Pap Avulsos Zool. 2012; 52(37):441–75. https://doi.org/10.1590/S0031-10492012021700001

Britski HA, Garavello JC. Sobre Leporinus octofasciatus Steindachner da bacia do Paraná (Pisces, Anostomidae). Pap Avulsos Zool. 1978; 31(16):237–50. https://doi.org/10.11606/0031-1049.1978.31.p237-250

Britski HA, Garavello JC. Descrição de duas espécies novas de Leporinus da bacia do Tapajós (Pisces, Characiformes). Comun Mus Cienc PUCRS Sér Zool. 1993; 6:29–40.

Burns MD, Chatfield M, Birindelli JLO, Sidlauskas BL. Systematic assessment of the Leporinus desmotes species complex, with a description of two new species. Neotrop Ichthyol. 2017; 15(2):e160166. https://doi.org/10.1590/1982-0224-20160166

Burns MD, Frable BW, Sidlauskas BL. A new species of Leporinus (Characiformes: Anostomidae), from the Orinoco Basin, Venezuela. Copeia. 2014; 2014(2):206–14. https://doi.org/10.1643/CI-13-071

Burns MD, Sidlauskas BL. Ancient and contingent body shape diversification in a hyperdiverse continental fish radiation. Evolution. 2019; 73(3):569–87. https://doi.org/10.1111/evo.13658

Calcagnotto D, Schaefer SA, DeSalle R. Relationships among characiform fishes inferred from analysis of nuclear and mitochondrial gene sequences. Mol Phylogenet Evol. 2005; 36(1):135–53. https://doi.org/10.1016/j.ympev.2005.01.004

Chernoff B, Machado-Allison A, Saul WG. Morphology, variation and biogeography of Leporinus brunneus (Pisces: Characiformes: Anostomidae). Ichthyol Explor Freshw. 1991; 1(4):295–306.

Dagosta F, Monção MS, Nagamatsu BA, Pavanelli CS, Carvalho FR, Lima FCT et al. Fishes of the upper rio Paraná basin: diversity, biogeography and conservation. Neotrop Ichthyol. 2024; 22(1):e230066. https://doi.org/10.1590/1982-0224-2023-0066

Dillman CB, Sidlauskas BL, Vari RP. A morphological supermatrix-based phylogeny for the Neotropical fish superfamily Anostomoidea (Ostariophysi: Characiformes): phylogeny, missing data and homoplasy. Cladistics. 2016; 32(3):276–96. https://doi.org/10.1111/cla.12127

Drummond A, Rambaut A. BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evol Biol. 2007; 7(1):214. https://doi.org/10.1186/1471-2148-7-214

Edgar RC. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 2004; 32(5):1792–97. https://doi.org/10.1093/nar/gkh340

van den Ende C, Puttick MN, Urrutia AO, Wills MA. Why should we compare morphological and molecular disparity? Methods Ecol Evol. 2023; 14(9):2390–410. https://doi.org/10.1111/2041-210X.14166

Faircloth BC, Alda F, Hoekzema K, Burns MD, Oliveira C, Albert JS et al. A target enrichment bait set for studying relationships among ostariophysan fishes. Copeia. 2020; 108(1):47–60. https://doi.org/10.1643/CG-18-139

Feitosa FdS, Santos GM, Birindelli JLO. Leporinus britskii: a new anostomid species from the Tapajós and Jari drainages, Brazil (Characiformes: Anostomidae). Zootaxa. 2011; 3120(1):55–62. https://doi.org/10.11646/zootaxa.3120.1.4

Fowler HW. A collection of fishes obtained by Mr. William C. Morrow in the Ucayali River basin, Peru. Proc Acad Nat Sci Phila. 1939; 91:219–89. https://www.jstor.org/stable/4064291

Fowler HW. Zoological results of the second Bolivian expedition for the Academy of Natural Sciences of Philadelphia, 1936–1937. Part I. The Fishes. Proc Acad Nat Sci Phila. 1940; 92:43–103. https://www.jstor.org/stable/4064303

Fowler HW. A collection of fishes from Colombia, obtained chiefly by Brother Nicéforo Maria. Proc Acad Nat Sci Phila. 1943; 95:223–66. https://www.jstor.org/stable/4064347

Galetti Jr. PM, Foresti F. Evolution of the ZZ/ZW system in Leporinus (Pisces, Anostomidae): Role of constitutive heterochromatin. Cytogenet Cell Genet. 1986; 43(1–2):43–46. http://hdl.handle.net/11449/219118

Galetti Jr. PM, Foresti F, Bertollo LAC, Moreira Filho O. Heteromorphic sex chromosomes in three species of the genus Leporinus (Pisces, Anostomidae). Cytogenet Cell Genet. 1981; 29(3):138–42. https://doi.org/10.1159/000131562

Galetti Jr. PM, Mestriner CA, Venere PC, Foresti F. Heterochromatin and karyotype reorganization in fish of the family Anostomidae (Characiformes). Cytogenet Cell Genet. 1991; 56(2):116–21. https://doi.org/10.1159/000133063

Garavello JC. Revisão taxonômica do gênero Leporinus Spix. 1829 (Ostariophysi. Anostomidae). [PhD Thesis]. São Paulo, Brazil: Universidade de São Paulo; 1979.

Garavello JC. A new species of the anostomid genus Leporinus Spix from Suriname, with redescriptions of two related species (Pisces, Characiformes, Anostomidae). Bull Zool Mus Univ Amst. 1990; 12(11):161–70. https://repository.naturalis.nl/pub/505366

Garavello JC. Two new species of Leporinus Spix with a review of the blotched species of the Rio Orinoco system and rescription of Leporinus muyscorum Steindachner (Characiformes: Anostomidae). Proc Acad Nat Sci Phila. 2000; 150:193–201. https://www.jstor.org/stable/4065070

Garavello JC, Britski HA. Family Anostomidae. In: Reis RE, Kullander SO, Ferraris Jr. CJ, editors. Check list of the freshwater fishes of South and Central America. Porto Alegre: Edipucrs; 2003. p.71–84.

Garavello JC, Britski HA, Birindelli JLO. Redescription of Leporinus jamesi (Characiformes: Anostomidae), a poorly known species of Leporinus from the lowlands of the central Amazon, Brazil. Neotrop Ichthyol. 2014; 12(2):317–26. https://doi.org/10.1590/1982-0224-20130114

Garavello JC, Santos GM. Leporinus trimaculatus, a new species from Amazonia, Brazil, and redescription of the sympatric Leporinus aripuanaensis (Pisces, Characiformes, Anostomidae). Bull Zool Mus Univ Amst. 1992; 13(12):110–17. https://repository.naturalis.nl/pub/505439

Garavello JC, Ramirez JL, Oliveira AK, Britski HA, Birindelli JLO, Galetti Jr. PM. Integrative taxonomy reveals a new species of Neotropical headstanding fish in genus Schizodon (Characiformes: Anostomidae). Neotrop Ichthyol. 2021; 19(4):e210016. https://doi.org/10.1590/1982-0224-2021-0016

Géry J. Contributions a l’étude des poissons characoïdes (No. 7). Validité de Leporinus despaxi Puyo et du sous-genre Hypomasticus Borodin. Bull Mus Natl Hist Nat. 1960; 32(3):222–29.

Géry J. Contributions à l’étude des poissons Characoïdes 13, structures et évolution des Anostominae. Bull Aquat Biol. 1961; 2(19):93–112.

Géry J. Notes sur quelques anostomidae (Pisces, Characoidei) du bassin Amazonien. Vie Milieu. 1972; 23(fasc .1, sér. C.):143–75. https://hal.sorbonne-universite.fr/hal-02982032v1

Géry J. Characoids of the world. Neptune City: T.F.H. Publications; 1977.

Géry J. Description d’une nouvelle espèce de poisson anostomidé (Ostariophysi, Characoidei) du Rio Mamoré, Bolivie: Rhytiodus lauzannei sp. n. Cybium. 1987; 11(4):365–73. https://doi.org/10.26028/cybium/1987-114-003

Géry J, Mahnert V, Dlouhy C. Poissons characoides non characidae du Paraguay (Pisces, Ostariophysi). Rev Suisse Zool. 1987; 94(2):357–464. https://doi.org/10.5962/bhl.part.79720

Géry J, Planquette P, Le Bail P-Y. Nomenclature des éspèces du groupe Leporinus maculatus et formes affines des Guyanes (Pisces, Characoidei, Anostomidae). Rev Suisse Zool. 1988; 95(3):699–713. https://doi.org/10.5962/bhl.part.81929

Géry J. A new anostomid species, Leporinus bleheri n.sp., from the Rio Guaporé-Iténez basin, with comments on some related species (Teleostei: Ostariophysi: Characiformes). Aqua. 1999; 3(3):105–12.

Günther A. Catalogue of the fishes in the British Museum, vol. 5. Catalogue of the Physostomi, containing the families Siluridae, Characinidae, Haplochitonidae, Sternoptychidae, Scopelidae, Stomiatidae in the collection of the British Museum. London: British Museum (Natural History), Department of Zoology; 1864. https://doi.org/10.5962/bhl.title.8809

Ito ISB, Souza-Shibatta L, Venturieri M, Birindelli JLO. A new species of Leporinus (Characiformes: Anostomidae) from the Serra do Cachimbo, Pará State, Brazil. Neotrop Ichthyol. 2023; 21(3):e230058. https://doi.org/10.1590/1982-0224-2023-0058

Jenkins J, Bart Jr. H, Bowker J, Bowser P, MacMillan J, Nickum J et al. Guidelines for the use of fishes in research. Bethesda, Maryland: American Fisheries Society; 2014.

Kearney M. Fragmentary taxa, missing data, and ambiguity: mistaken assumptions and conclusions. Syst Biol. 2002; 51(2):369–81. https://doi.org/10.1080/10635150252899824

Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S et al. Geneious basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics. 2012; 28(12):1647–49. https://doi.org/10.1093/bioinformatics/bts199

Keating JN, Garwood RJ, Sansom RS. Phylogenetic congruence, conflict and consilience between molecular and morphological data. BMC Ecol Evol. 2023; 23:30. https://doi.org/10.1186/s12862-023-02131-z

Kner R. Zur familie der characinen. III. Folge der ichthyologischen beiträge. Denkschr Kaiserl Akad Wiss Wien. 1859; 17:137–82. https://biostor.org/reference/184570

Lanfear R, Calcott B, Ho SYW, Guindon S. PartitionFinder: combined selection of partitioning schemes and substitution models for phylogenetic analyses. Mol Biol Evol. 2012; 29(6):1695–701. https://doi.org/10.1093/molbev/mss020

Laurin M. The PhyloCode: the logical outcome of millennia of evolution of biological nomenclature? Zool Scr. 2023; 52(6):543–55. https://doi.org/10.1111/zsc.12625

Li CH, Orti G, Zhang G, Lu GQ. A practical approach to phylogenomics: the phylogeny of ray-finned fish (Actinopterygii) as a case study. BMC Evol Biol. 2007; 7:44. https://doi.org/10.1186/1471-2148-7-44

Lofeu L, Anelli V, Straker LC, Kohlsdorf T. Developmental plasticity reveals hidden fish phenotypes and enables morphospace diversification. Evolution. 2021; 75(5):1170–88. https://doi.org/10.1111/evo.14221

Lofeu L, Montefeltro F, Simon MN, Kohlsdorf T. Functional modularity and mechanical stress shape plastic responses during fish development. Evolution. 2024; 78(9):1568–82. https://doi.org/10.1093/evolut/qpae086

Lopera-Barrero NM, Povh JA, Ribeiro RP, Gomes PC, Jacometo CB, Lopes TS. Comparison of DNA extraction protocols of fish fin and larvae samples: modified salt (NaCl) extraction. Cienc Investig Agrar. 2008; 35(1):65–74.

Lovejoy NR, Collette BB. Phylogenetic relationships of New World needlefishes (Teleostei: Belonidae) and the biogeography of transitions between marine and freshwater habitats. Copeia. 2001; 2001(2):324–38. https://doi.org/10.1643/0045-8511(2001)001[0324:PRONWN]2.0.CO;2

Lundberg JG, Marshall LG, Guerrero J, Horton B, Malabarba MCSL, Wesselingh F. The stage for Neotropical fish diversification: a history of tropical South American rivers. In: Malabarba LR, Reis RE, Vari RP, Lucena ZMS, Lucena CA, editors. Phylogeny and classification of Neotropical fishes. Porto Alegre: Edipucrs; 1998. p.13–48.

Mahnert V, Géry J, Muller S. Leporinus falcipinnis n. sp., a new species from the lower rio Tapajos basin, Para, Brazil (Pisces, Characiformes, Anostomidae). Rev Suisse Zool. 1997; 104(4):837–44. https://doi.org/10.5962/bhl.part.80025

Melo BF, Benine RC, Mariguela TC, Oliveira C. A new species of Tetragonopterus Cuvier, 1816 (Characiformes: Characidae: Tetragonopterinae) from the rio Jari, Amapa, northern Brazil. Neotrop Ichthyol. 2011; 9(1):49–56. https://doi.org/10.1590/S1679-62252011000100002

Melo BF, Sidlauskas BL, Hoekzema K, Frable BW, Vari RP, Oliveira C. Molecular phylogenetics of the Neotropical fish family Prochilodontidae (Teleostei: Characiformes). Mol Phylogenet Evol. 2016; 102:189–201. http://dx.doi.org/10.1016/j.ympev.2016.05.037

Melo BF, Sidlauskas BL, Hoekzema K, Vari RP, Dillman CB, Oliveira C. Molecular phylogenetics of Neotropical detritivorous fishes of the family Curimatidae (Teleostei: Characiformes). Mol Phylogenet Evol. 2018; 127:800–12. https://doi.org/10.1016/j.ympev.2018.06.027

Melo BF, Sidlauskas BL, Hoekzema K, Vari RP, Oliveira C. The first molecular phylogeny of Chilodontidae (Teleostei: Ostariophysi: Characiformes) reveals cryptic biodiversity and taxonomic uncertainty. Mol Phylogenet Evol. 2014; 70:286–95. https://doi.org/10.1016/j.ympev.2013.09.025

Melo BF, Sidlauskas BL, Near TJ, Roxo FF, Ghezelayagh A, Ochoa LE et al. Accelerated diversification explains the exceptional species richness of tropical characoid fishes. Syst Biol. 2022; 71(1):78–92. https://doi.org/10.1093/sysbio/syab040

Mendes IS, Melo BF, Damasceno JS, Teixeira DF, Carvalho DC. Phylogeography of Hypomasticus copelandii (Teleostei, Anostomidae) reveals distinct genetic lineages along Atlantic coastal drainages of eastern Brazil. Diversity-Basel. 2022; 14(1):29. https://doi.org/10.3390/d14010029

Metcalfe J, Craig J. Ethical justification for the use and treatment of fishes in research: an update. J Fish Biol. 2011; 78(2):393–94. https://doi.org/10.1111/j.1095-8649.2010.02900.x

Miller MA, Pfeiffer W, Schwartz T, editors. Creating the CIPRES Science Gateway for inference of large phylogenetic trees. 2010 Gateway Computing Environments Workshop (GCE), New Orleans, LA, USA, 2010. https://doi.org/10.1109/GCE.2010.5676129

Mirande JM. Morphology, molecules and the phylogeny of Characidae (Teleostei, Characiformes). Cladistics. 2019; 35(3):282–300. https://doi.org/10.1111/cla.12345

Molina WF, Schmid M, Galetti Jr. PM. Heterochromatin and sex chromosomes in the Neotropical fish genus Leporinus (Characiformes, Anostomidae). Cytobios. 1998; 94:141–49.

Myers GS. Studies on South American fresh-water fishes. II. The genera of anostomine characids. Stanford Ichthyol Bull. 1950; 3(4):184–98.

Myers GS, Carvalho AL. A remarkable new genus of anostomin characid fishes from the upper Rio Xingú in central Brazil. Copeia. 1959; 1959(2):148–52. https://doi.org/10.2307/1440067

Near TJ, Thacker CE. Phylogenetic classification of living and fossil ray-finned fishes. Bull Peabody Mus Nat Hist. 2024; 65(1):3–302. https://doi.org/10.3374/014.065.0101

Ohara WM, Lima FC, Salvador G, Andrade M. Peixes do rio Teles Pires: diversidade e guia de identificação. Goiânia: Editora Amazonas; 2017.

Oliveira C, Avelino GS, Abe KT, Mariguela TC, Benine RC, Ortí G et al. Phylogenetic relationships within the speciose family Characidae (Teleostei: Ostariophysi: Characiformes) based on multilocus analysis and extensive ingroup sampling. BMC Evol Biol. 2011; 11:275. https://doi.org/10.1186/1471-2148-11-275

Palumbi SR. Nucleic acids II: the polymerase chain reaction. In: Hillis DM, Moritz C, Mable B, editors. Molecular Systematics. Massachusetts: Sinauer Associates Inc.; 1996. p.205–47.

Planquette P, Keith P, Le Bail PY. Atlas des poissons d’eau douce de Guyane (tome 1). Paris: IEGB – M.N.H.N., INRA, CSP, Min. Env.; 1996.

Queiroz K. The PhyloCode and the distinction between taxonomy and nomenclature. Syst Biol. 2006; 55(1):160–62. https://doi.org/10.1080/10635150500431221

Queiroz K, Cantino P. International code of phylogenetic nomenclature (PhyloCode). 1st ed. Boca Raton: CRC Press; 2020. https://doi.org/10.1201/9780429446320

Rambaut A. FigTree, a graphical viewer of phylogenetic trees (Version 1.4.4); 2018. Available from: https://github.com/rambaut/figtree/releases

Rambaut A, Drummond AJ, Xie D, Baele G, Suchard MA. Posterior summarization in Bayesian phylogenetics using Tracer 1.7. Syst Biol. 2018; 67(5):901–04. https://doi.org/10.1093/sysbio/syy032

Ramirez JL, Galetti Jr. PM. DNA barcode and evolutionary relationship within Laemolyta Cope 1872 (Characiformes: Anostomidae) through molecular analyses. Mol Phylogenet Evol. 2015; 93:77–82. http://dx.doi.org/10.1016/j.ympev.2015.07.021

Ramirez JL, Carvalho-Costa LF, Venere PC, Carvalho DC, Troy WP, Galetti Jr. PM. Testing monophyly of the freshwater fish (Characiformes, Anostomidae) through molecular analysis. J Fish Biol. 2016; 88(3):1204–14. https://doi.org/10.1111/jfb.12906

Ramirez JL, Birindelli JL, Carvalho DC, Affonso PRAM, Venere PC, Ortega H et al. Revealing hidden diversity of the underestimated Neotropical ichthyofauna: DNA barcoding in the recently described genus Megaleporinus (Characiformes: Anostomidae). Front Genet. 2017a; 8. https://doi.org/10.3389/fgene.2017.00149

Ramirez JL, Birindelli JLO, Galetti Jr. PM. A new genus of Anostomidae (Ostariophysi: Characiformes): diversity, phylogeny and biogeography based on cytogenetic, molecular and morphological data. Mol Phylogenet Evol. 2017b; 107:308–23. https://doi.org/10.1016/j.ympev.2016.11.012

Ramirez JL, Santos CA, Machado CB, Oliveira AK, Garavello JC, Britski HA et al. Molecular phylogeny and species delimitation of the genus Schizodon (Characiformes, Anostomidae). Mol Phylogenet Evol. 2020; 153:106959. https://doi.org/10.1016/j.ympev.2020.106959

Ringuelet RA, Arámburu RH. Peces argentinos de agua dulce: claves de reconocimiento y caracterización de familias y subfamilias, con glosario explicativo. Agro. 1961; 3(7):1–98. http://sedici.unlp.edu.ar/handle/10915/130483

Ronquist F, Teslenko M, van der Mark P, Ayres DL, Darling A, Hohna S et al. MrBayes 3.2: efficient Bayesian phylogenetic inference and model choice across a large model space. Syst Biol. 2012; 61(3):539–42. https://doi.org/10.1093/sysbio/sys029

Roure B, Baurain D, Philippe H. Impact of missing data on phylogenies inferred from empirical phylogenomic data sets. Mol Biol Evol. 2012; 30(1):197–214. https://doi.org/10.1093/molbev/mss208

Sabaj MH. Codes for natural history collections in ichthyology and herpetology. Copeia. 2020; 108(3):593–669. https://doi.org/10.1643/ASIHCODONS2020

Sanderson MJ, Shaffer HB. Troubleshooting molecular phylogenetic analyses. Annu Rev Ecol Syst. 2002; 33(1):49–72. https://doi.org/10.1146/annurev.ecolsys.33.010802.150509

Santos GM. Aspectos de sistemática e morfologia de Schizodon fasciatus Agassiz 1829, Rhytiodus microlepis Kner 1859 e Rhytiodus argenteofuscus Kner, 1829 (Osteichthyes, Characoidei, Anostomidae) do Lago Janauacá-Amazonas. Acta Amaz. 1980; 10(3):635–49. https://doi.org/10.1590/1809-43921980103635

Santos GM. Estudo da alimentaçao e hábitos alimentares de Schizodon fasciatus Agassiz, 1829, Rhytiodus microlepis Kner, 1859, e R. argenteofuscus Kner, 1859 do Lago Janauacá-AM (Osteichthyes, Characoidei, Anostomidae). Acta Amaz. 1981; 11(2):267–83. https://doi.org/10.1590/1809-43921981112267

Santos GM, Jégu M. Novas ocorrências de Gnathodolus bidens, Synaptolaemus cingulatus e descrição de duas espécies novas de Sartor (Characiformes, Anostomidae). Amazoniana. 1987; 10(2):181–96. https://repositorio.inpa.gov.br/handle/1/39875

Santos GM, Jégu M. Inventário taxonômico e redescrição das espécies da anostomideos (Characiformes, Anostomidae) do baixo rio Tocantins, PA, Brasil. Acta Amaz. 1989; 19:159–213. https://doi.org/10.1590/1809-43921989191213

Santos GM, Jégu M, Lima AC. Novas ocorrências de Leporinus pachycheilus Britski, 1976 e descrição de uma espécie nova do mesmo grupo na Amazônia Brasileira (Osteichthys, Anostomidae). Acta Amaz. 1996; 26(4):265–80. https://doi.org/10.1590/1809-43921996264280

Sidlauskas B. Continuous and arrested morphological diversification in sister clades of characiform fishes: a phylomorphospace approach. Evolution. 2008; 62(12):3135–56. https://doi.org/10.1111/j.1558-5646.2008.00519.x

Sidlauskas BL. Testing for unequal rates of morphological diversification in the absence of a detailed phylogeny: a case study from characiform fishes. Evolution. 2007; 61:299–316. https://doi.org/10.1111/j.1558-5646.2007.00022.x

Sidlauskas BL, Assega FM, Melo BF, Oliveira C, Birindelli JLO. Total evidence phylogenetic analysis reveals polyphyly of Anostomoides and uncovers an unexpectedly ancient genus of Anostomidae fishes (Characiformes). Zool J Linn Soc. 2021;1 94(2):626–69. https://doi.org/10.1093/zoolinnean/zlab016

Sidlauskas BL, Birindelli JLO. Family Anostomidae – Toothed headstanders. In: Van der Sleen P, Albert JS, editors. Field guide to fishes of the Amazon, Orinoco and Guianas. Princeton: Princeton University Press.; 2018. p.82–89.

Sidlauskas BL, Garavello JC, Jellen J. A new Schizodon (Characiformes: Anostomidae) from the Río Orinoco system, with a redescription of S. isognathus from the Rio Paraguay system. Copeia. 2007; 2007(3):711–25. https://doi.org/10.1643/0045-8511(2007)2007[711:ANSCAF]2.0.CO;2

Sidlauskas BL, Santos GM. Pseudanos winterbottomi, a new anostomine species (Teleostei: Characiformes: Anostomidae) from Venezuela and Brazil, and comments on its phylogenetic relationships. Copeia. 2005; 2005(1):109–23. https://doi.org/10.1643/CI-04-100R2

Sidlauskas BL, Vari RP. Phylogenetic relationships within the South American fish family Anostomidae (Teleostei, Ostariophysi, Characiformes). Zool J Linn Soc. 2008; 154:70–210. https://doi.org/10.1111/j.1096-3642.2008.00407.x

Sidlauskas BL, Mol JH, Vari RP. Dealing with allometry in linear and geometric morphometrics: a taxonomic case study in the Leporinus cylindriformis group (Characiformes: Anostomidae) with description of a new species from Suriname. Zool J Linn Soc. 2011; 162(1):103–30. https://doi.org/10.1111/j.1096-3642.2010.00677.x

Sidlauskas BL, Vari RP. Diversity and distribution of anostomoid fishes (Teleostei: Characiformes) throughout the Guianas. Cybium. 2012; 36(1):71–103. https://doi.org/10.26028/cybium/2012-361-008

Silva-Santos R, Ramirez JL, Galetti Jr. PM, Freitas PD. Molecular evidences of a hidden complex scenario in Leporinus cf. friderici. Front Genet. 2018; 9:47. https://doi.org/10.3389/fgene.2018.00047

Spix JB, Agassiz L. Selecta genera et species piscium quos in itinere per Brasiliam annis MDCCCXVII-MDCCCXX jussu et auspiciis Maximiliani Josephi I…. collegit et pingendos curavit Dr J. B. de Spix…. Monachii. Munich: Typis C. Wolf; 1829-31. 438 + 95 plates. https://doi.org/10.5962/bhl.title.157964

Stamatakis A. RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics. 2006; 22(21):2688–90. https://doi.org/10.1093/bioinformatics/btl446

Steindachner F. Ichthyologische beiträge (V). Sitz-Ber K Akad Wiss, Math-Naturwiss Kl. 1876; 74:49–240. https://biostor.org/reference/163875

Steindachner F. Beiträge zur kenntniss der flussfische Südamericas. V. Denkschr Kaiserl Akad Wiss Wien. 1915; 93:15–106.

Steindachner F. Beiträge zur kenntnis der flußfiche Sudamerikas V. Denkschr Kaiserl Akad Wiss Wien. 1917; 93:15–106. https://biostor.org/reference/181980

Toledo-Piza M, Baena E, Dagosta FCP, Menezes NA, Ândrade M, Benine RC et al. Checklist of the species of the order Characiformes (Teleostei: Ostariophysi). Neotrop Ichthyol. 2024; 22(1):e230086. https://doi.org/10.1590/1982-0224-2023-0086

Vari RP. Phylogenetic relationships of the families Curimatidae, Prochilodontidae, Anostomidae, and Chilodontidae (Pisces: Characiformes). Smithson Contrib Zool. 1983; 378:1–60. https://doi.org/10.5479/si.00810282.378

Venere PC, Ferreira IA, Martins C, Galetti Jr. PM. A novel ZZ/ZW sex chromosome system for the genus Leporinus (Pisces, Anostomidae, Characiformes). Genetica. 2004; 121(1):75–80. https://doi.org/10.1023/b:gene.0000019936.40970.7a

Wiens JJ. Missing data and the design of phylogenetic analyses. J Biomed Inform. 2006; 39(1):34–42. https://doi.org/10.1016/j.jbi.2005.04.001

Winterbottom R. Systematics, osteology and phylogenetic relationships of fishes of the ostariophysan subfamily Anostominae (Characoidei, Anostomidae). Life Sci Contr R Ont Mus. 1980; 123:1–112. https://doi.org/10.5962/bhl.title.52085

Zou Z, Zhang J. Morphological and molecular convergences in mammalian phylogenetics. Nat Commun. 2016; 7:12758. https://doi.org/10.1038/ncomms12758

Authors

^[1] Department of Fisheries, Wildlife and Conservation Sciences, Oregon State University, 104 Nash Hall, 97331 Corvallis, OR, USA. (BLS) brian.sidlauskas@oregonstate.edu (corresponding author), (MDB) burnsmic@oregonstate.edu.

^[2] National Museum of Natural History, Smithsonian Institution, 1000 Madison Drive NW, 20560 Washington, DC, USA

^[3] Department of Ecology and Evolutionary Biology, Tulane University, 6823 St. Charles Avenue, New Orleans, LA, USA.

^[4] Department of Ichthyology, American Museum of Natural History, 200 Central Park West, 10024-5102 New York, NY, USA. (BFM) bmelo@amnh.org.

^[5] Departamento de Biologia Animal e Vegetal, Universidade Estadual de Londrina, Rodovia Celso Garcia Cid, km 380, 86055-900 Londrina, PR, Brazil. (JLOB) josebirindelli@uel.br.

^[6] Scripps Institution of Oceanography, University of California San Diego, 9500 Gilman Drive, 92093-0244 La Jolla, CA, USA. (BWF) bfrable@ucsd.edu.

^[7] University of Washington, Department of Genome Sciences, 3720 15th Ave. NE, 98195 Seattle, WA, USA. (KH) kendralh@uw.edu.

^[8] Department of Ecology and Evolutionary Biology, Museum of Vertebrates, Cornell University, E145 Corson Hall, 14853 Ithaca, NY, USA. (CBD) cbd63@cornell.edu.

^[9] Department of Ichthyology, Academy of Natural Sciences of Drexel University, 1900 Benjamin Franklin Parkway, 19103 Philadelphia, PA, USA. (MHS) mhs58@drexel.edu.

^[10] Instituto de Biociências, Universidade Estadual Paulista, R. Prof. Dr. Antônio Celso Wagner Zanin, 250, 18618-689 Botucatu, SP, Brazil. (CO) claudio.oliveira@unesp.br.

Authors’ Contribution

Brian L. Sidlauskas: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Visualization, Writing-original draft, Writing-review and editing.

Bruno F. Melo: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Validation, Visualization, Writing-original draft, Writing-review and editing.

José L. O. Birindelli: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Visualization, Writing-original draft, Writing-review and editing.

Michael D. Burns: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing-review and editing.

Benjamin W. Frable: Conceptualization, Investigation, Methodology, Writing-review and editing.

Kendra Hoekzema: Data curation, Formal analysis, Investigation, Methodology.

Casey B. Dillman: Conceptualization, Investigation, Methodology, Visualization, Writing-review and editing.

Mark H. Sabaj: Funding acquisition, Investigation, Methodology, Resources, Visualization, Writing-review and editing.

Claudio Oliveira: Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Writing-review and editing.

Ethical Statement

The tissue samples were assembled from the holdings of many natural history collections, where demonstration of proper permission to collect and export from the countries of origin is a standard condition of accession. Some specimens were collected by the authors of this paper under Project N^o 13010120, Protocol N^o 20115, Action N^o 60323, Detail N^o 282180 approved by Drexel University’s Institutional Animal Care and Use Committee (IACUC) or Protocol 4103 approved by Oregon State University’s IACUC. To our knowledge, the tissue samples and voucher specimens were collected in accordance with best practices for the ethical use of fishes in research (Metcalfe, Craig, 2011; Jenkins et al., 2014).

Competing Interests

The author declares no competing interests.

How to cite this article

Sidlauskas BL, Melo BF, Birindelli JLO, Burns MD, Frable BW, Hoekzema K, Dillman CB, Sabaj MH, Oliveira C. Molecular phylogenetics, a new classification, and a new genus of the Neotropical fish family Anostomidae (Teleostei: Characiformes). Neotrop Ichthyol. 2025; 23(1):e240076. https://doi.org/10.1590/1982-0224-2024-0076

Copyright

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

Distributed under

Creative Commons CC-BY 4.0

Diversity and Distributions Published by SBI

Accepted December 18, 2024 by George Mattox

Submitted August 5, 2024

Epub March 14, 2025

Abstract​

Introduction​

Material and methods

Results​

Discussion​

Acknowledgments​

References​