The family Characidae is a group of freshwater bony fishes that exhibits high species-level diversity and whose members inhabit parts of Texas, Mexico, and Central and South America. Thus far, morphological data have been of limited use in discerning relationships among subfamilies and incertae sedis genera of the family Characidae. In this study, DNA sequence data from GenBank were combined with new sequences for analyses under Bayesian and parsimony schemes. Sequences fell into four gene partitions, with three genes in the mitochondrial subset (12S, 16S, COI genes) and one gene in the nuclear subset (RAG2 gene). Inferred Bayesian and parsimony-based phylogenies reject the monophyly of certain groups (e.g., Astyanax, Hyphessobrycon, and Bryconamericus), do not reject the monophyly of others (e.g., Cheirodontinae and "clade A" of previous authors), and present new sister-group hypotheses (e.g., Brittanichthys sister to Paracheirodon). Sister to clade A is a lineage referred to herein as "clade B" which includes Exodon and exemplars from Cheirodontinae (the most basal lineage within clade B), Aphyocharacinae, Tetragonopterinae, and Characinae (excluding Gnathocharax). "Clade C" is sister to A+B and contains representatives of large incertae sedis genera (e.g., Hyphessobrycon, Hemigrammus), as well as members of Stethaprioninae. Unless certain other subfamilial names are to be disregarded, the use of Tetragonopterinae should continue to be restricted to species of Tetragonopterus because other genera previously referred to this subfamily grouped in clades A or C, quite distant from Tetragonopterus.

BackgroundThe subfamily Stevardiinae is a diverse and widely distributed clade of freshwater fishes from South and Central America, commonly known as “tetras” (Characidae). The group was named “clade A” when first proposed as a monophyletic unit of Characidae and later designated as a subfamily. Stevardiinae includes 48 genera and around 310 valid species with many species presenting inseminating reproductive strategy. No global hypothesis of relationships is available for this group and currently many genera are listed as incertae sedis or are suspected to be non-monophyletic.ResultsWe present a molecular phylogeny with the largest number of stevardiine species analyzed so far, including 355 samples representing 153 putative species distributed in 32 genera, to test the group’s monophyly and internal relationships. The phylogeny was inferred using DNA sequence data from seven gene fragments (mtDNA: 12S, 16S and COI; nuclear: RAG1, RAG2, MYH6 and PTR). The results support the Stevardiinae as a monophyletic group and a detailed hypothesis of the internal relationships for this subfamily.ConclusionsA revised classification based on the molecular phylogeny is proposed that includes seven tribes and also defines monophyletic genera, including a resurrected genus Eretmobrycon, and new definitions for Diapoma, Hemibrycon, Bryconamericus sensu stricto, and Knodus sensu stricto, placing some small genera as junior synonyms. Inseminating species are distributed in several clades suggesting that reproductive strategy is evolutionarily labile in this group of fishes.Electronic supplementary materialThe online version of this article (doi:10.1186/s12862-015-0403-4) contains supplementary material, which is available to authorized users.

Aim Patterns of genetic variation within freshwater fish populations may reflect the historical impact of climate change on either sea-level or environmental conditions. Past sea-level changes enlarged palaeodrainages and so connected currently isolated rivers, whereas changes in environmental conditions reduced forest cover and may have constrained the movement of fish specialized to this habitat. We assayed genetic variation in Hollandichthys multifasciatus, a freshwater fish endemic to the Atlantic Forest of coastal Brazil, to test the relative importance of these factors in shaping current patterns of genetic divergence.Location River drainages along the south-eastern Brazilian coast.Methods A GIS was used to reconstruct palaeodrainages during the Last Glacial Maximum (LGM). Niche modelling was used to infer areas of stability for the southern Atlantic Forest sensu stricto (present and LGM). The contribution of river connections inside or outside areas of stability was evaluated using a calibrated phylogeny, analyses of molecular variance, and Bayesian skyline plots from two mtDNA loci.Results Analyses of 182 individuals from 26 populations and 12 palaeodrainages indicated that structure associated with palaeodrainages explains 75% of the genetic variation among populations, with estimated divergence times occurring within the Pleistocene. The variation explained by palaeodrainages and estimated population sizes was unrelated to the ecological stability of the region.Main conclusions This study demonstrates the importance of Pleistocene palaeodrainages in structuring genetic divergence patterns. The analyses suggest that past connections due to sea-level retreat played a significant role in the diversification of the ichthyofauna along the Brazilian coastal drainages. Moreover, the lack of a signature of habitat stability in structuring genetic variation suggests that refugia may be less important in structuring genetic diversity for freshwater species than for terrestrial species. In addition, our work highlights the utility of a GIS-based approach to recover past connections among coastal basins. Understanding these connections is crucial for studying diversification of riverine organisms and for identifying areas of conservation priority.

We herein analyse the history of the description of the freshwater fish fauna from three drainages in one of the most densely collected areas of Brazil, and possibly of South America, the Rio Grande do Sul State, southern Brazil. An updated inventory of the freshwater fish species from rio Uruguay (partial) in Brazil, Laguna dos Patos (complete) and rio Tramandaí basins (complete) is presented. We found the number of new species described in these drainages increased nearly 56% since 1981, reaching a total of 422 species, but even now 10% of this number still corresponds to undescribed species. This rate of species description suggests that previous estimates of the Neotropical fish fauna are low, and we predict a final number of Neotropical fishes larger than the largest prediction estimate (8,000 species), after other regions of South and Central Americas become densely sampled. We discuss and attempt to demonstrate that species diversity knowledge is historically and strictly related to collecting efforts. We also demonstrate that the ecoregions in eastern South America with the highest density of species per area correspond to the areas more densely sampled in collections, and this may represent a bias in such kinds of analyses. This uneven sampling in Brazilian regions is apparently associated with the uneven distribution of Zoological research centers in different regions of the country. Small-sized species represents an important source of new species, along with little explored regions or little explored habitats, sometimes associated with restricted range species, and species complexes that need revisionary work. In contrast to other Neotropical regions, Atheriniformes are relatively diverse, sharing the fifth place in species richness with Gymnotiformes, and there is a remarkably high number of species of Rivulidae. Eight species are endemic to the rio Tramandaí drainage, 68 to the Laguna dos Patos system, and 78 to the rio Uruguay drainage. Almost 10% of the freshwater fish species are "Critically Endangered", "Endangered" or "Vulnerable" according to the IUCN criteria, with Rivulidae as the family with the largest number of threatened species.

The Neotropical cichlid genus Gymnogeophagus Ribeiro, 1918 is revised. The folio wing species are considered valid and are redescribed: G. rhabdotus (Hensel, 1870), G. gymnogenys (Hensel, 1870), G. labiatus (Hensel, 1870), G. balzanii (Perugia, 1891) and G. australis (Eigenmann, 1907). In addition, two new species are described: G. lacustris, sp. n., from the coastal region of southern Brazil and G. meridionalis, sp. n Cope, 1894) and G. australis. The synonymy proposed by Gosse is significantly re-arranged in the present paper. That author united the four species as a monophyletic assemblage he termed the genus Gymnogeophagus. In his study, however, Gosse primarily used long preserved color-less specimens, resulting in a poor definition of species and a useless key.Kullander, 1981 studying some specimens from Rio Grande do SuI recognized a thick-lipped Gymnogeophagus species, resurrecting G. labiatus from the synonymy of G. gymnogenys.The two characters used by Gosse in redefining Gymnogeophagus are reinterpreted in light of Hennig's, 1966 principles. The spine on top of the first dorsal pterygiophore has proved to be a synapomorphy for the genus, being unique among Neotropical cichlids. The absence of supraneurals, however, although a very good key-character, must be used with caution in a phylogenetic analysis. This reduction seems likely to have occurred independently in various cichlid genera. However, within Geophagines -a probably monophyletic assemblage (Kullander, 1980~), including those genera with an epibranchial lobe: Acarichthys, Biotodoma, Geophagus, Gymnogeophagus, Papiliochromis Retroculus, Apistogramma, Apistogrammoides, Biotoecus and Taeniacara -this reduction evidently occurred only once and may be assumed to be synapomorphic for Gymnogeophagus. The material examined of each Gymnogeophagus species is listed before the description of each species and comparative material is gIven at the end of the paper. Data include museums abbreviations and catalogue number, number of specimens in parenthesis, collecting locality, date and collectors.The synonymy for each species are limited to original species descriptions. The morphometric and meristic data were taken as shown on figure 1. Proportions are expressed as ratios in table 3, with minimum and maximum values given and the mean in parenthesis. The regression data were computed through routine statistical methods and are presented in table 4.Description of color in life are based on observations of living specimens and/or color photos taken just after collection of the specimen.The osteological observations were made on cleared and alizarinstained material (Taylor, 1967 enzyme method) and the drawings sketched with a stereomicroscope and "camera lucida". At least two specimens of each Gymnogeophagus species, except G. australis, were cleared, alizarin-stained and dissected for osteological observations. Radiographs were made of G. australis paratypes. An attempt was made to use specimens of aproximately the same size. Larger specimen...

As part of an effort to characterize reproductive modifications in internally inseminating catfishes, ovaries and male reproductive systems were examined histologically in two species of auchenipterid catfishes, Trachelyopterus lucenai and T. galeatus, from southeastern Brazil. Internal insemination was documented in both species by the presence of sperm within ovaries. Although there is some variation in gross morphology of the male reproductive systems between the two species, both have four main regions: spermatogenic lobes, sperm storage regions, and secretory and storage regions of the seminal vesicle. In both species, the anterior portion of the reproductive system is spermatogenic and divided into numerous finger‐like lobes. Posterior to the spermatogenic area is the storage region of the seminal vesicle, a large median structure with a honeycomb‐like appearance. This region is consistently larger in T. lucenai. Attached to the storage region of the seminal vesicle in both species are secretory lobes comprised of tubules lined by secretory cells. These lobes in T. lucenai are small and located on the anterior aspect of the storage region of the seminal vesicle, whereas in T. galeatus the lobes are much larger and located laterally. The sperm storage regions of T. lucenai consist of two large lobes located ventral to the storage region of the seminal vesicle. Highly compact sperm packets (spermatozeugmata) fill the lumina of the ramifying tubules of these regions. Each spermatozeugma consists of elongate nuclei tightly arranged parallel to one another. In T. galeatus two distinct sperm storage regions are present. Just posterior to the spermatogenic lobes a series of small lobes serve as anterior sperm storage regions. Posterior to the secretory lobes of the seminal vesicle is a series of lobes, at the most posterior aspect of the reproductive tract, that serve as posterior sperm storage regions. Both are identical, histologically, to the sperm storage regions of T. lucenai. An absence of compact spermatozeugmata in the T. galeatus specimens may be related to variations in their sexual activity. The descriptions presented here allow for consistent terminology for comparison of regions of the male reproductive system based on presumed function. J. Morphol. 246:131–141, 2000 © 2000 Wiley‐Liss, Inc.

The species of the genus Trichomycterus inhabiting the laguna dos Patos system are reviewed and five species are recognized. Trichomycterus tropeiro Ferrer & Malabarba has a restricted range and is endemic to the uppermost portion of the rio das Antas. Trichomycterus balios, n. sp., is distributed in the upper portion of the rio das Antas and rio Caí basins. Trichomycterus diatropoporos, n. sp., is endemic to the rio da Prata basin, a tributary of the rio das Antas. Trichomycterus poikilos, n. sp., is widely distributed in the upper portion of the rio Jacuí basin and tributaries of the rio Taquari-Antas. Trichomycterus brachykechenos, n. sp., is endemic to the upper portion of the rio dos Sinos. The new species are distinguishable from most congeners, except for T. davisi, T. mboycy, T. naipi, T. payaya, T. papilliferus, T. perkos, T. plumbeus, and T. tropeiro by the lower number of pectoral-fin rays (I+5-6) and by the first pectoral-fin ray not prolonged as a filament. Other characters distinguish the new taxa from these eight species. The distribution of the genus in the laguna dos Patos system is discussed and a taxonomic key is provided.

No abstract