Higher‐level phylogeny of diving beetles (Coleoptera: Dytiscidae) based on larval characters

A comprehensive higher‐level phylogeny of diving beetles (Dytiscidae) based on larval characters is presented. Larval morphology and chaetotaxy of a broad range of genera and species was studied, covering all currently recognized subfamilies and tribes except for the small and geographically restricted Hydrodytinae, where the larva is unknown. The results suggest several significant conclusions with respect to the systematics of Dytiscidae including the following: monophyly of all currently recognized subfamilies, although Dytiscinae when considered in a broad context is rendered paraphyletic by Cybistrinae; currently recognized tribes are monophyletic except for Agabini, Hydroporini and Laccornellini; inter‐subfamily and inter‐tribe relationships generally show weak support, except for a few well supported clades; three distinct clades are recognized within Dytiscinae [Dytiscini sensu lato (i.e. including the genera Dytiscus Linnaeus and Hyderodes Hope), Hydaticini sensu lato, and Cybistrini]; and recognition of Pachydrini as a distinct tribe. Other less robust results include: Methlini sister to the rest of Hydroporinae; relative basal position of Laccornini, Hydrovatini and Laccornellini within Hydroporinae; close relationship of Agabinae and Copelatinae; Matinae nested deep within Dytiscidae, as sister to a large clade including Colymbetinae, Coptotominae, Lancetinae and Dytiscinae sensu lato; the sister‐group relationship of Agabetini and Laccophilini is confirmed. The results presented here are discussed and compared with previous phylogenetic hypotheses based on different datasets, and the evolution of some significant morphological features is discussed in light of the proposed phylogeny. All suprageneric taxa are diagnosed, including illustrations of all relevant synapomorphies, and a key to separate subfamilies and tribes is presented, both in traditional (paper) format and as an online Lucid interactive identification key.     

Taxonomic review of the genus Copelatus Erichson (Coleoptera: Dytiscidae) in Korea

A taxonomic review of the genus Copelatus Erichson (Coleoptera: Dytiscidae) in Korea is presented. In the present study, five species of Copelatus are recognized from Korea, including one unrecorded species, C. kammuriensis. We provide a key to the Korean species of Copelatus, diagnosis of one unrecorded species, habitus photographs, scanning electron micrographs of the elytron, and illustrations of the aedeagus.     

Evolution of the Juan Fernández diving beetle,Rhantus selkirki (Coleoptera,Dytiscidae)

Here we provide evidence that confinement in Robinson Crusoe Island (located about 660 km west of continental Chile) over evolutionary time leads to strong morphological modifications in diving beetle (Dytiscidae) larvae. We analysed a large set of morphological larval characters for all currently recognised genera of Colymbetinae as a framework, to infer phylogenetic relationships within the large genus Rhantus Dejean, 1833 and, in particular, of the charismatic Juan Fernández diving beetle, Rhantus selkirki Jäch, Balke & Michat, 2015, comparing our results with a recent phylogeny of the Colymbetinae based on DNA sequence data. We suggest that adaptation to the island's particular habitats resulted in the reversal of certain characters of R. selkirki back to the plesiomorphic states. This may cause the species to be erroneously interpreted as more ‘primitive’ if only morphological characters are analysed. Confinement in the particular, shallow and barely vegetated aquatic habitats of Robinson Crusoe Island for a long time seems to have led to this divergent morphology, particularly in characters related to swimming ability such as several leg and urogomphal setae. In this way, R. selkirki larvae secondarily resemble those of some earlier diverging dytiscid lineages such as Agabinae and Copelatinae, which typically creep on the bottom of water bodies and do not swim well.     

A highly modified stygobiont diving beetle of the genus Copelatus (Coleoptera,Dytiscidae): taxonomy and cladistic analysis based on mitochondrial DNA sequences

Abstract. We report the discovery of a new species of Copelatus that is morphologically highly modified for life in groundwater. Copelatus abditus sp.n. was collected from a 4 m deep bore in central Australia. It was placed in Copelatinae based on morphological evidence. This is the first known stygobiont diving beetle that does not belong to the subfamily Hydroporinae. Sequences from the mitochondrial DNA cytochrome oxidase 1, 16S rRNA, tRNAL and NADH dehydrogenase subunit 1 genes suggest that the species belongs to the subgenus Papuadytes, the morphological apomorphies of which are reduced in the new groundwater species. Copelatus (Papuadytes) abditus very much resembles other world stygobiont dytiscids, which all belong to the subfamily Hydroporinae. We suggest that this striking convergence is due to similar selective pressures imposed on all groundwater beetles. We suggest a scenario in which species from temporary habitats evade drought by entering the hyporheic zone, and may be driven to a subterranean existence by desertification of their ancestral habitats, as can be observed in Australia.     

Phylogeny and diversification of diving beetles (Coleoptera: Dytiscidae)

Dytiscidae is the most diverse family of beetles in which both adults and larvae are aquatic, with examples of extreme morphological and ecological adaptations. Despite continuous attention from systematists and ecologists, existing phylogenetic hypotheses remain unsatisfactory because of limited taxon sampling or low node support. Here we provide a phylogenetic tree inferred from four gene fragments (cox1, rrnL, H3 and SSU, ≈ 4000 aligned base pairs), including 222 species in 116 of 174 known genera and 25 of 26 tribes. We aligned ribosomal genes prior to tree building with parsimony and Bayesian methods using three approaches: progressive pair‐wise alignment with refinement, progressive alignment modeling the evolution of indels, and deletion of hypervariable sites. Results were generally congruent across alignment and tree inference methods. Basal relationships were not well defined, although we identified 28 well supported lineages corresponding to recognized tribes or groups of genera, among which the most prominent novel results were the polyphyly of Dytiscinae; the grouping of Pachydrini with Bidessini, Peschetius with Methlini and Coptotomus within Copelatinae; the monophyly of all Australian Hydroporini (Necterosoma group), and their relationship with the Graptodytes and Deronectes groups plus Hygrotini. We found support for a clade formed by Hydroporinae plus Laccophilini, and their sister relationship with Cybistrini and Copelatinae. The tree provided a framework for the analysis of species diversification in Dytiscidae. We found a positive correlation between the number of species in a lineage and the age of the crown group as estimated through a molecular clock approach, but the correlation with the stem age was non‐significant. Imbalances between sister clades were significant for several nodes, but the residuals of the regression of species numbers with the crown age of the group identified only Bidessini and the Coptotomus + Agaporomorphus clade as lineages with, respectively, above and below expected levels of species diversity. © The Willi Hennig Society 2008.     

Phylogenetic Affinities Among the Extant Malagasy Lemurs (Lemuriformes) Based on Morphology and Behavior

The difficulty in achieving a consensus on the phylogenetic relationships of lemuriform primates has been due largely to the lack of a lemur fossil record and to the lack of an appropriate outgroup that would facilitate polarization of character states. Recent findings allow us to polarize some of the bony characters, but to a large extent this problem still remains. In the past, phylogenetic analyses have focused on specialized character sets such as dentition or basicranial traits, or they have employed differential weighting schemes to a more variable set of characters. In the analysis presented here, I combined all relevant characters available in the literature into one data set but restricted my selection to those traits having discontinuous states and for which no contradictory coding schemes were published. I reduced the assumptions in this analysis by removing most external weighting and ordering effects on these data sets. The available data from the literature were supplemented with data from my own observations at the Duke University Primate Center. Data were collected for 25 characters and 20 taxa and were submitted to a cladistic analysis. Some important findings from this study include support for (1) a sister-group relationship between Lepilemur and the Indridae, (2) a sister-group relationship between the Lemuridae (except Varecia) and the Indridae/Lepilemur clade, (3) a monophyletic genus Eulemur, and (4) the exclusion of Varecia from the Lemuridae.     

Limb myology and phylogenetic relationships in the superfamily Dipodoidea (birch mice, jumping mice, and jerboas)

Limb muscles were dissected in seven genera, representing all six superfamilies, of dipodoid rodents and myoloic characters were used to construct a phylogenetic hpothesis of relationships within this cfade. Mologic differences among genera suported tie monophyly of the superfamily Dipodoidea reLtive to the outrou taxon and reveaEd thatSicista is the sister group to all other zapodid and dipodid enera. Tkis picement of Sicista differs markedly from its position in previous classifications where it has been regarded merely as a primitive zapodid genus. The phlograrn based on rnyologic characters also indicated that Cardiocranius is not a rimitive dipodid genus; it is the sister group to the subfamily Dipodinae. Although myologic differences among taxa were not sufficient to warrant the continued separation of zaodids and dipodids into two families, a new classification that places Sicista in its own family, ficistidae, and places the remaining zaodids and dipodids in the family Dipodidae, is proposed. Differences in karyology, genitaP morholoy, and postcranial osteological characters among dipodoid rodents are discussed in light or this pjylogeny.     

Gelechioidea (Insecta: Lepidoptera) systematics: a reexamination using combined morphology and mitochondrial DNA data

Phylogenies of Gelechioidea (Insecta: Lepidoptera) historically have been in disagreement and definitions vary at the family and subfamily levels. Addition of new taxa or new characters drastically changes relationships indicating that current phylogenetic schemes require more investigation. This study is the first phylogenetic analysis of Gelechioidea to include molecular data. Here we present a combined analysis using Maximum Parsimony to investigate sister-group relationships within Gelechioidea. The addition of Cytochrome oxidase I and II to revised published morphological matrices gives 453 parsimony informative characters for the 42 taxa for which we have sequence data. The combined analysis resulted in two trees with mostly novel sister-group relationships. These results challenge current concepts of Gelechioidea, suggesting that traditional morphological characters that have united taxa may not be homologous structures and are in need of further investigation. A combination of morphological data with new molecular data will be the most robust method of study for Gelechioidea phylogenetics.     

Cladistic analysis of the Sepsidae (Cyclorrhapha: Diptera) based on a comparative scanning electron microscopic study of larvae

The result of a phylogenetic analysis of the Sepsidae based on larval characters is presented. It is shown that cyclorrhaphan larvae can be as rich a source of characters as Nematocera immatures when investigated using an SEM. The cladistic analysis comprised fifty-two species in sixteen genera of the Sepsidae and five outgroup species and used fifty-seven morphological characters. It found seven parsimonious trees which only differed with respect to the arrangement of some species within the genus Themira. The basal dichotomies of the phylogenetic trees are particularly well supported, indicating the conservative nature of larval characters. Orygma is confirmed as the sister group of all the remaining sepsids, the Sepsinae. There is good larval evidence that Ortalischema is the sister group of all remaining Sepsinae and that the Toxopodinae constitute an early radiation within the Sepsidae. According to larval data, some genera are paraphyletic ( Themira, Palaeosepsis ), but adult characters appear to contradict these findings. Among the traditionally recognized higher taxa within the Sepsidae, Hennig's Themira species-group and Steysbal's Sepsini have to be rejected as polyphyletic. However, Hennig's Sepsis species-group is confirmed as monophyletic and will probably constitute one major element of a future phylogenetic system of the Sepsidae. States of the strongly modified fore-legs of some adult sepsid males are mapped onto the phylogenetic tree, largely confirming Šulc's ideas about the evolution of these features. The origin and evolution of male sternites with brushes and a gland on the tibiae of the males ('osmeterium') are discussed. Whereas adult characters point to a sister-group relationship between the Sepsidae and the Ropalomeridae, larval characters appear to indicate a sister-group relationship between the Coelopidae and the Sepsidae. The evidence for both hypotheses is critically evaluated.     

Phylogeny and systematics of the subfamily Bryocorinae based on morphology with emphasis on the tribe Dicyphini sensu Schuh,

The classification of the hyperdiverse true bug family Miridae is far from settled, and is particularly contentious for the cosmopolitan subfamily Bryocorinae. The morphological diversity within the subfamily is pronounced, and a lack of explicit character formulation hampers stability in the classification. Molecular partitions are few and only a handful of taxa have been sequenced. In this study the phylogeny of the subfamily Bryocorinae has been analysed based on morphological data alone, with an emphasis on evaluating the tribe Dicyphina sensu Schuh, 1976, within which distinct groups of taxa exist. A broad sample of taxa was examined from each of the bryocorine tribes. A broad range of outgroup taxa from most of the other mirid subfamilies was also examined to test for bryocorine monophyly, ingroup relationships and to determine character polarity. In total a matrix comprising 44 ingroup, 15 outgroup taxa and 111 morphological characters was constructed. The phylogenetic analysis resulted in a monophyletic subfamily Bryocorinae sensu Schuh (1976, 1995), except for the genus Palaucoris, which is nested within Cylapinae. The tribe Dicyphini sensu Schuh (1976, 1995) has been rejected. The subtribe Odoniellina is synonymized with the subtribe Monaloniina and the subtribes Dicyphina, Monaloniina and Eccritotarsina are now elevated to tribal level, with the Dicyphini now restricted in composition and definition. The genus Felisacus is highly autapomorphic and a new tribe – the Felisacini – is erected for the included taxa. This phylogeny of the tribes of the Bryocorinae comprises the following sister‐group relationships: Dicyphini ((Bryocorini + Eccritotarsini)(Felisicini + Monaloniini)).     

Phylogeny of the bee family Megachilidae (Hymenoptera: Apoidea) based on adult morphology

Phylogenetic relationships within the bee family Megachilidae are poorly understood. The monophyly of the subfamily Fideliinae is questionable, the relationships among the tribes and subtribes in the subfamily Megachilinae are unknown, and some extant genera cannot be placed with certainty at the tribal level. Using a cladistic analysis of adult external morphological characters, we explore the relationships of the eight tribes and two subtribes currently recognised in Megachilidae. Our dataset included 80% of the extant generic‐level diversity, representatives of all fossil taxa, and was analysed using parsimony. We employed 200 characters and selected 7 outgroups and 72 ingroup species of 60 genera, plus 7 species of 4 extinct genera from Baltic amber. Our analysis shows that Fideliinae and the tribes Anthidiini and Osmiini of Megachilinae are paraphyletic; it supports the monophyly of Megachilinae, including the extinct taxa, and the sister group relationship of Lithurgini to the remaining megachilines. The Sub‐Saharan genus Aspidosmia, a rare group with a mixture of osmiine and anthidiine features, is herein removed from Anthidiini and placed in its own tribe, Aspidosmiini, new tribe . Protolithurgini is the sister of Lithurgini, both placed herein in the subfamily Lithurginae; the other extinct taxa, Glyptapina and Ctenoplectrellina, are more basally related among Megachilinae than Osmiini, near Aspidosmia, and are herein treated at the tribal level. Noteriades, a genus presently in the Osmiini, is herein transferred to the Megachilini. Thus, we recognise four subfamilies (Fideliinae, Pararhophitinae, Lithurginae and Megachilinae) and nine tribes in Megachilidae. We briefly discuss the evolutionary history and biogeography of the family, present alternative classifications, and provide a revised key to the extant tribes of Megachilinae.     

The phylogenetic position of an undescribed paedomorphic clupeiform taxon: mitogenomic evidence

Molecular characters may offer a useful alternative to confidently estimate the phylogenetic position of paedomorphic taxa otherwise difficult to place based on morphology because of the reduction or absence of characters in their larvae-like adult stage. Here, we sequenced the complete mitogenome of a remarkable undescribed marine paedomorphic clupeiform fish to gain insight into its phylogenetic position. Of a length of 17,507 bp, this mitogenome exhibits a unique gene order within the Teleostei because of the inversion of the contiguous tRNA^Gln and tRNA^Ile within the IQM region and the presence of a putative second control region inserted between these tRNAs. Mitogenomic data from 27 clupeiform species and 22 non-clupeiform species were subjected to partitioned maximum likelihood and Bayesian analyses. All resultant phylogenetic trees strongly supported the placement of this undescribed taxon within the order Clupeiformes, suborder Clupeoidei, and the family Clupeidae, as the sister group of the tribe Spratelloidini (Jenkinsia + Spratelloides) of the subfamily Dussumieriinae. Together, they form a monophyletic group with Chirocentrus and, possibly, Etrumeus. Despite its overall resemblance to Sundasalanx, this undescribed taxa (Clupeidae gen. et sp. indet.) is not closely related to that genus and represents an independent paedomorphic lineage within the Clupeoidei. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

The systematic position of Meruidae (Coleoptera, Adephaga) and the phylogeny of the smaller aquatic adephagan beetle families

A phylogenetic analysis of Adephaga is presented. It is based on 148 morphological characters of adults and larvae and focussed on a placement of the recently described Meruidae, and the genus‐level phylogeny of the smaller aquatic families Gyrinidae, Haliplidae and Noteridae. We found a sister group relationship between Gyrinidae and the remaining adephagan families, as was found in previous studies using morphology. Haliplidae are either the sister group of Dytiscoidea or the sister group of a clade comprising Geadephaga and the dytiscoid families. Trachypachidae was placed as the sister group of the rhysodid‐carabid clade or of Dytiscoidea. The monophyly of Dytiscoidea including Meru is well supported. Autapomorphies are the extensive metathoracic intercoxal septum, the origin of the metafurca from this structure, the loss of Mm. furcacoxalis anterior and posterior, and possibly the presence of an elongated subcubital setal binding patch. Meruidae was placed as sister group of the Noteridae. Synapomorphies are the absence of the transverse ridge of the metaventrite, the fusion of abdominal segments III and IV, the shape of the strongly asymmetric parameres, and the enlargement of antennomeres 5, 7 and 9. The Meru‐noterid clade is the sister group of the remaining Dytiscoidea. The exact position of Aspidytes within this clade remains ambiguous: it is either the sister group of Amphizoidae or the sister group of a clade comprising this family and Hygrobiidae + Dytiscidae. The sister group relationship between Spanglerogyrinae and Gyrininae was strongly supported. The two included genera of Gyrinini form a clade, and Enhydrini are the sister group of a monophylum comprising the remaining Enhydrini and Orectochilini. A branching pattern (Peltodytes + (Brychius + Haliplus)) within Haliplidae was confirmed. Algophilus, Apteraliplus and the Haliplus‐subgenus Liaphlus form a clade. The generic status of the two former taxa is unjustified. The Phreatodytinae are the sister group of Noterinae, and Notomicrus (+ Speonoterus), Hydrocoptus, and Pronoterus branch off successively within this subfamily. The search for the larvae of Meru and a combined analysis of morphological and molecular data should have high priority. © The Willi Hennig Society 2006.     

A genus-level supertree of Adephaga (Coleoptera)

A supertree for Adephaga was reconstructed based on 43 independent source trees – including cladograms based on Hennigian and numerical cladistic analyses of morphological and molecular data – and on a backbone taxonomy. To overcome problems associated with both the size of the group and the comparative paucity of available information, our analysis was made at the genus level (requiring synonymizing taxa at different levels across the trees) and used Safe Taxonomic Reduction to remove especially poorly known species. The final supertree contained 401 genera, making it the most comprehensive phylogenetic estimate yet published for the group. Interrelationships among the families are well resolved. Gyrinidae constitute the basal sister group, Haliplidae appear as the sister taxon of Geadephaga+Dytiscoidea, Noteridae are the sister group of the remaining Dytiscoidea, Amphizoidae and Aspidytidae are sister groups, and Hygrobiidae forms a clade with Dytiscidae. Resolution within the species-rich Dytiscidae is generally high, but some relations remain unclear. Trachypachidae are the sister group of Carabidae (including Rhysodidae), in contrast to a proposed sister-group relationship between Trachypachidae and Dytiscoidea. Carabidae are only monophyletic with the inclusion of a non-monophyletic Rhysodidae, but resolution within this megadiverse group is generally low. Non-monophyly of Rhysodidae is extremely unlikely from a morphological point of view, and this group remains the greatest enigma in adephagan systematics. Despite the insights gained, our findings highlight that a combined and coordinated effort of morphologists and molecular systematists is still required to expand the phylogenetic database to enable a solid and comprehensive reconstruction of adephagan phylogeny. See also Supplementary material in the online edition at doi:10.1016/j.ode.2006.05.003     

Molecular Support for the Placement of Saiga and Procapra in Antilopinae (Artiodactyla, Bovidae)

The evolutionary history of the bovid subfamily Antilopinae is unclear. Traditionally, this subfamily is subdivided into two tribes: Neotragini (dwarf antelopes) and Antilopini (gazelles and their relatives). Here, we report new sequences for the 12S and 16S rRNA genes in the enigmatic antilopine taxa Procapra gutturosa and Saiga tatarica and analyze the phylogenetic relationships of these taxa relative to other antilopines. Our study demonstrates the close affinity of the saiga antelope to Gazella despite the conventional systematic allocation of Saiga to the Caprinae subfamily. The second member of the Saigini tribe, Pantholops hodgsoni (Tibetan gazelle), falls within Caprinae. In all of our analyses, Procapra gutturosa occupied a basal position in the Antilopinae clade or was a sister-group to the dwarf antelope Madoqua. This suggests early separation of Procapra from other antelopes.     

Squamate phylogeny and the relationships of snakes and mosasauroids

Cladistic analysis of extant and fossil squamates (95 characters, 26 taxa) finds the fossil squamate, Coniasaurus Owen, 1850, to be the sister-group of the Mosasauroidea (mosasaurs and aigialosaurs). This clade is supported in all 18 shortest cladograms (464 steps; CI 0.677; HI 0.772) by nine characters of the dermatocranium, maxilla, and mandible. A Strict Consensus Tree of the 18 shortest trees collapses to a basal polytomy for most major squamate clades including the clade (Coniasaurus, Mosasauroidea). A Majority Rule Consensus Tree shows that, in 12 of 18 shortest cladograms, the clade Coniasaurus- Mosasauroidea is the sister-group to snakes (Scolecophidia (Alethinophidia, Dinilysia); this entire clade, referred to as the Pythonomorpha ([[Scolecophidia [Alethinophidia, Dinilysia]], [Coniasaurus, Mosasauroidea]]) is the sister-group to all other scleroglossans. Pythonomorpha is supported in these 12 cladograms by nine characters related to the lower jaw and cranial kinesis. In 6 of 18 shortest cladograms, snakes are the sister-group to the clade (Amphisbaenia (Dibamidae (Gekkonoidea, Eublepharidae))). None of the cladograms support the hypothesis that coniasaurs and mosasauroids are derived varanoid anguimorphs. Two additional analyses were conducted: (1) manipulation and movement of problematic squamate clades while constraining ‘accepted’ relationships; (2) additional cladistic analyses beginning with extant taxa, and sequentially adding fossil taxa. From Test I, at 467 steps, Pythonomorpha can be the sister-group to the Anguimorpha, Scincomorpha, ‘scinco-gekkonomorpha’ [scincomorphs, gekkotans, and amphibaenids-dibamids]. At 471 steps Pythonomorpha can be placed within Varanoidea. Treating only mosasauroids and coniasaurs as a monophyletic group: 469 steps, mosasauroids and coniasaurs as sister-group to Anguimorpha; 479 steps, mosasauroids and coniasaurs nested within Varanoidea. Test II finds snakes to nest within Anguimorpha in a data set of only Mosasauroidea + Extant Squamates; the sistergroup to snakes + anugimorphs is (Amphisbaenia (Dibarnidae (Gekkonoidea, Eublepharidae))). No one particular taxon is identified as a keystone taxon in this analysis, though it appears truc that fossil taxa significantly alter the structure of squamate phylogenetic trees.     

Phylogeny of the Aphnaeinae: myrmecophilous African butterflies with carnivorous and herbivorous life histories

The Aphnaeinae (Lepidoptera: Lycaenidae) are a largely African subfamily of 278 described species that exhibit extraordinary life‐history variation. The larvae of these butterflies typically form mutualistic associations with ants, and feed on a wide variety of plants, including 23 families in 19 orders. However, at least one species in each of 9 of the 17 genera is aphytophagous, parasitically feeding on the eggs, brood or regurgitations of ants. This diversity in diet and type of symbiotic association makes the phylogenetic relations of the Aphnaeinae of particular interest. A phylogenetic hypothesis for the Aphnaeinae was inferred from 4.4 kb covering the mitochondrial marker COI and five nuclear markers (wg, H3, CAD, GAPDH and EF1α) for each of 79 ingroup taxa representing 15 of the 17 currently recognized genera, as well as three outgroup taxa. Maximum Parsimony, Maximum Likelihood and Bayesian Inference analyses all support Heath's systematic revision of the clade based on morphological characters. Ancestral range inference suggests an African origin for the subfamily with a single dispersal into Asia. The common ancestor of the aphnaeines likely associated with myrmicine ants in the genus Crematogaster and plants of the order Fabales.     

Anatomy and phylogenetic analysis of the neotropical callichthyid catfishes (Ostariophysi, Siluriformes)

Based mainly on morphological characters, the phylogenetic relationships among genera and some species groups of the neotropical family Callichthyidae were examined. A study of the osteology of a generalized callichthyid, Callichthys callichthys (Linnaeus), with detailed comparisons among representatives of the remaining genera in the family, is presented and used as a basis for the phylogenetic analysis. A single most parsimonious tree supported the monophyly of the family Callichthyidae based on 28 derived features and the division of the family in the subfamilies Corydoradinae and Callichthyinae. In the subfamily Corydoradinae, the genus Aspidoras is the sister-group of the clade formed by Corydoras plus Brochis. Five derived features support the monophyly of this clade and four support the monophyly of Brochis. No characters, however, were found to support the genus Corydoras. In the subfamily Callichthyinae, Dianema and Hopbstemum are sister-taxa. Megalechis represents the sister-group of Dianema plus Hoplosternum and Lepthoplosternum represents the sister-group to Megalechis plus Dianema plus Hopbstemum. Finally, Callichthys is considered the least derived member of the subfamily, and is hypothesized as the sister-group of the remaining species. A key to all callichthyid genera is provided.