Coding characters from different life stages for phylogenetic reconstruction: a case study on dragonfly adults and larvae,including a description of the larval head anatomy of Epiophlebia superstes (Odonata: Epiophlebiidae)

The exclusive use of characters coding for specific life stages may bias tree reconstruction. If characters from several life stages are coded, the type of coding becomes important. Here, we simulate the influence on tree reconstruction of morphological characters of Odonata larvae incorporated into a data matrix based on the adult body under different coding schemes. For testing purposes, our analysis is focused on a well‐supported hypothesis: the relationships of the suborders Zygoptera, ‘Anisozygoptera’, and Anisoptera. We studied the cephalic morphology of Epiophlebia, a key taxon among Odonata, and compared it with representatives of Zygoptera and Anisoptera in order to complement the data matrix. Odonate larvae are characterized by a peculiar morphology, such as the specific head form, mouthpart configuration, ridge configuration, cephalic musculature, and leg and gill morphology. Four coding strategies were used to incorporate the larval data: artificial coding (AC), treating larvae as independent terminal taxa; non‐multistate coding (NMC), preferring the adult life stage; multistate coding (MC); and coding larval and adult characters separately (SC) within the same taxon. As expected, larvae are ‘monophyletic’ in the AC strategy, but with anisopteran and zygopteran larvae as sister groups. Excluding larvae in the NMC approach leads to strong support for both monophyletic Odonata and Epiprocta, whereas MC erodes phylogenetic signal completely. This is an obvious result of the larval morphology leading to many multistate characters. SC results in the strongest support for Odonata, and Epiprocta receives the same support as with NMC. Our results show the deleterious effects of larval morphology on tree reconstruction when multistate coding is applied. Coding larval characters separately is still the best approach in a phylogenetic framework. © 2015 The Linnean Society of London     

蜻蜓目高级阶元系统发育关系研究进展

综述了对蜻蜓目系统发育的研究。对一些规模较大较全面的研究结果做了描述,并对其进行比较;对最近的基于分子的和基于形态的研究的共同点做了总结,其共同支持的观点包括:束翅亚目处于分支的基部为一并系类群,差翅亚目与间翅亚目关系较近。     

The adipokinetic hormones of Odonata: a phylogenetic approach

Adipokinetic neuropeptides from the corpora cardiaca of the major families of all three suborders of the Odonata were identified by one or more of the following methods: (1) Isolation of the peptides from a methanolic extract of the corpora cardiaca by liquid chromatography, peak monitoring by fluorescence of the Trp residue and comparison of the retention time with those of known synthetic peptides of Odonata. (2) Hyperlipaemic bioassays of the HPLC-generated fractions either in Locusta migratoria or, in a few cases, in Anax imperator or Orthetrum julia. (3) Sequencing of the isolated, bioactive HPLAC fraction by Edman degradation. (4) Mass spectrometric measurement of the isolated, bioactive fraction. Sequence assignment revealed that the investigated Odonata species always contain only one adipokinetic peptide. This is always an octapeptide. The suborder Zygoptera contains the peptide code-named Psein-AKH, the Anisozygoptera and the families Aeshnidae, Cordulegastridae and Macromiidae of the Anisoptera contain Anaim-AKH, whereas Gomphidae, Corduliidae (with the exception of Syncordulia gracilis) and Libellulidae contain Libau-AKH; one species of Libellulidae has Erysi-AKH, a very conservative modification of Libau-AKH (one point mutation). When these structural data are interpreted in conjunction with existing phylogenies of Odonata, they support the following: (1) Zygoptera are monophyletic and not paraphyletic. (2) Anisozygoptera and Anisoptera are sister groups and contain the ancestral Anaim-AKH which is independently and convergently mutated to Libau-AKH in Gomphidae and Libellulidae. (3) The Corduliidae are of special interest. Only Corduliidae sensu stricto appear to contain Libau-AKH, other species placed into this family by most authorities contain the ancestral Anaim-AKH. Possibly, assignments of AKHs can untangle the paraphyly of this family.     

Molecular phylogeny of the higher taxa of Odonata (Insecta) inferred from COI, 16S rRNA, 28S rRNA,and EF1‐α sequences

In this study, we sequenced both two mitochondrial genes (COI and 16S rRNA) and nuclear genes (28S rRNA and elongation factor‐1α) from 71 species of Odonata that represent 7 superfamilies in 3 suborders. Phylogenetic testing for each two concatenated gene sequences based on function (ribosomal vs protein‐coding genes) and origin (mitochondrial vs nuclear genes) proved limited resolution. Thus, four concatenated sequences were utilized to test the previous phylogenetic hypotheses of higher taxa of Odonata via Bayesian inference (BI) and maximum likelihood (ML) algorithms, along with the data partition by the BI method. As a result, three slightly different topologies were obtained, but the BI tree without partition was slightly better supported by the topological test. This topology supported the suborders Anisoptera and Zygoptera each being a monophyly, and the close relationship of Anisozygoptera to Anisoptera. All the families represented by multiple taxa in both Anisoptera and Zygoptera were consistently revealed to each be a monophyly with the highest nodal support. Unlike consistent and robust familial relationships in Zygoptera those of Anisoptera were partially unresolved, presenting the following relationships: ((((Libellulidae + Corduliidae) + Macromiidae) + Gomphidae + Aeshnidae) + Anisozygoptera) + (((Coenagrionidae + Platycnemdidae) + Calopterygidae) + Lestidae). The subfamily Sympetrinae, represented by three genera in the anisopteran family Libellulidae, was not monophyletic, dividing Crocothemis and Deielia in one group together with other subfamilies and Sympetrum in another independent group.     

A molecular phylogeny of the Odonata (Insecta)

Abstract. We estimated the phylogeny of the order Odonata, based on sequences of the nuclear ribosomal genes 5.8 S, 18S, and ITS1 and 2. An 18S‐only analysis resolved deep relationships well: the order Odonata, as well as suborders Zygoptera and Epiprocta (Anisoptera + Epiophlebia), emerged as monophyletic. Some other deep clades resolved well, but support for more recently diverged clades was generally weak. A second, simultaneous, analysis of the 5.8S and 18S genes with the intergenic spacers ITS1 and 2 resolved some recent branches better, but appeared less reliable for deep clades with, for example, suborder Anisoptera emerging as paraphyletic and Epiophlebia superstes recovered as an Anisopteran, embedded within aeshnoid‐like anisopterans and sister to the cordulegastrids. Most existing family levels in the Anisoptera were confirmed as monophyletic clades in both analyses. However, within the corduliids that form a major monophyletic clade with the Libellulidae, several subclades were recovered, of which at least Macromiidae and Oxygastridae are accepted at the family level. In the Zygoptera, the situation is complex. The lestid‐like family groups (here called Lestomorpha) emerged as sister taxon to all other zygopterans, with Hemiphlebia sister to all other lestomorphs. Platystictidae formed a second monophylum, subordinated to lestomorphs. At the next level, some traditional clades were confirmed, but the tropical families Megapodagrionidae and Amphipterygidae were recovered as strongly polyphyletic, and tended to nest within the clade Caloptera, rendering it polyphyletic. Platycnemididae were also non‐monophyletic, with several representatives of uncertain placement. Coenagrionids were diphyletic. True Platycnemididae and non‐American Protoneurids are closely related, but their relationship to the other zygopterans remains obscure and needs more study. New World protoneurids appeared relatively unrelated to old world + Australian protoneurids. Several recent taxonomic changes at the genus level, based on morphology, were confirmed, but other morphology‐based taxonomies have misclassified taxa considered currently as Megapodagrionidae, Platycnemididae and Amphipterygidae and have underestimated the number of family‐level clades.     

The head anatomy of Epiophlebia superstes (Odonata: Epiophlebiidae)

The relic dragonfly family Epiophlebiidae is recovered as sistergroup of Anisoptera (= Epiprocta) by most molecular and morphological analyses. However, in a recent study it was placed within Anisoptera as sister group of Cordulegastridae. In another study, several affinities to Zygoptera in the morphology of the ovipositor and the egg-laying behaviour were pointed out. Here, we present a detailed study of the outer, as well as the inner, head morphology of Epiophlebia superstes. Compared with the last detailed literature account, three additional mandibular muscles were discovered, as well as additional buccal and pharyngeal muscles. The results are compared with the anatomic features of Zygoptera and Anisoptera. A formal character evaluation focused on head characters confirms the sistergroup relationship of Epiophlebiidae and Anisoptera.     

Molecules, morphology and fossils: a comprehensive approach to odonate phylogeny and the evolution of the odonate wing

We undertook a comprehensive morphological and molecular phylogenetic analysis of dragonfly phylogeny, examining both extant and fossil lineages in simultaneous analyses. The legitimacy of higher‐level family groups and the phylogenetic relationship between families were tested. Thirteen families were supported as monophyletic (Aeshnidae, Calopterygidae, Chlorocyphidae, Euphaeidae, Gomphidae, Isostictidae, Lestidae, Libellulidae, Petaluridae, Platystictidae, Polythoridae, Pseudostigmatidae and Synthemistidae) and eight as non‐monophyletic (Amphipterygidae, Coenagrionidae, Corduliidae, Megapodagrionidae, Protoneuridae and Synlestidae), although Perilestidae and Platycnemididae were recovered as monophyletic under Bayesian analyses. Nine families were represented by one species, thus monophyly was not tested (Epiophlebiidae, Austropetaliidae, Chlorogomphidae, Cordulegastridae, Macromiidae, Chorismagrionidae, Diphlebiidae, Lestoideidae and Pseudolestidae). Epiprocta and Zygoptera were recovered as monophyletic. Ditaxinerua is supported as the sister lineage to Odonata, Epiophlebiidae and the lestid‐like damselflies are sister to the Epiprocta and Zygoptera, respectively. Austropetaliidae + Aeshnidae is the sister lineage to the remaining Anisoptera. Tarsophlebia's placement as sister to Epiprocta or as sister to Epiprocta + Zygoptera was not resolved. Refinements are made to the current classification. Fossil taxa did not seem to provide signals crucial to recovering a robust phylogeny, but were critical to understanding the evolution of key morphological features associated with flight. Characters associated with wing structure were optimized revealing two wing character complexes: the pterostigma–nodal brace complex and the costal wing base & costal–ScP junction complex. In turn, these two complexes appear to be associated; the pterostigma–nodal brace complex allowing for further modification of the wing characters comprised within the costal wing base & costal–ScP junction complex leading the modern odonate wing. © The Willi Hennig Society 2008.     

Phylogenetic analysis of the insect order Odonata using 28S and 16S rDNA sequences: a comparison between data sets with different evolutionary rates

Molecular phylogenetic analyses were conducted for the insect order Odonata with a focus on testing the effectiveness of a slowly evolving gene to resolve deep branching and also to examine: (i) the monophyly of damselflies (the suborder Zygoptera); and (ii) the phylogenetic position of the relict dragonfly Epiophlebia superstes. Two independent molecular sources were used to reconstruct phylogeny: the 16S rRNA gene on the mitochondrial genome and the 28S rRNA gene on the nuclear genome. A comparison of the sequences showed that the obtained 28S rDNA sequences have evolved at a much slower rate than the 16S rDNA, and that the former is better than the latter for resolving deep branching in the Odonata. Both molecular sources indicated that the Zygoptera are paraphyletic, and when a reasonable weighting for among‐site rate variation was enforced for the 16S rDNA data set, E. superstes was placed between the two remaining major suborders, namely, Zygoptera and Anisoptera (dragonflies). Character reconstruction analysis suggests that multiple hits at the rapidly evolving sites in the 16S rDNA degenerated the phylogenetic signals of the data set.     

Temperature-body size responses in insects: a case study of British Odonata

1. Body size is highly correlated with physiological traits, fitness, and trophic interactions. These traits are subject to change if there are widespread reductions of body size with warming temperatures, which is suggested as one of the ‘universal’ ecological responses to climate change. However, general patterns of body size response to temperature in insects have not yet emerged. 2. To address this knowledge gap, we paired the wing length (as a proxy for body size) of 5331 museum specimens of 14 species of British Odonata with historical temperature data. Three sets of analyses were performed: (i) a regression analysis to test for a relationship between wing length and mean seasonal temperature within species and subsequent comparisons across species and suborders; (ii) an investigation of whether the body size of species has an effect on sensitivity to warming temperature; and (iii) a linear-mixed effects model to investigate factors that potentially affect temperature–size response. 3. The regression analysis indicated that wing length is negatively correlated with mean seasonal temperatures for Zygoptera, whereas Anisoptera showed no significant correlation with temperature. 4. There is a significant decline in wing length of all Zygoptera (but not Anisoptera) with collection date, suggesting that individuals emerging later in the season are smaller. 5. Life-cycle type was not important for predicting wing length–temperature responses, whereas sex, species, and suborder were indicated as important factors affecting the magnitude of temperature–size responses in Odonata. 6. Overall, wing lengths of Zygoptera are more sensitive to temperature and collection date than Anisoptera.     

Insect predation on pike fry

Laboratory tests evaluated the predatory impact of the macroinvertebrates Erythromma najas larvae (Odonata, Zygoptera: Coenagrionidae), Notonecta glauca (Heteroptera: Notonectidae), Ilyocoris cimicoides (Heteroptera: Naucoridae), Libellula depressa larvae (Odonata, Anisoptera: Libellulidae), Dytiscus marginalis larvae (Coleoptera: Dytiscidae) and Anax imperator larvae (Odonata, Anisoptera: Aeshnidae) on 3-, 12-, 21- and 30-day-old pike fry Esox lucius . All these insect predators captured and ate pike fry during the test, although the numbers killed varied among species. Dytiscus marginalis, Anax imperator and Notonecta glauca were the most Voracious predators.     

The phylogeny of Sericini and their position within the Scarabaeidae based on morphological characters (Coleoptera: Scarabaeidae)

Abstract. To reconstruct the phylogeny of the Sericini and their systematic position among the scarabaeid beetles, cladistic analyses were performed using 107 morphological characters from the adults and larvae of forty‐nine extant scarabaeid genera. Taxa represent most ‘traditional’ subfamilies of coprophagous and phytophagous Scarabaeidae, with emphasis on the Sericini and other melolonthine lineages. Several poorly studied exoskeletal features have been examined, including the elytral base, posterior wing venation, mouth parts, endosternites, coxal articulation, and genitalia. The results of the analysis strongly support the monophyly of the ‘orphnine group’ + ‘melolonthine group’ including phytophagous scarabs such as Dynastinae, Hopliinae, Melolonthinae, Rutelinae, and Cetoniinae. This clade was identified as the sister group to the ‘dung beetle line’ represented by Aphodius + Copris. The ‘melolonthine group’ is comprised in the strict consensus tree by two major clades and two minor lineages, with the included taxa of Euchirinae, Rutelinae, and Dynastinae nested together in one of the major clades (‘melolonthine group I’). Melolonthini, Cetoniinae, and Rutelinae are strongly supported, whereas Melolonthinae and Pachydemini appear to be paraphyletic. Sericini + Ablaberini were identified to be sister taxa nested within the second major melolonthine clade (‘melolonthine group II’). As this clade is distributed primarily in the southern continents, one could assume that Sericini + Ablaberini are derived from a southern lineage. Plausibly, ancestors of Sericini + Ablaberini and Athlia were separated by a vicariance event, such as the separation of the African plate from the rest of Gondwana, whereas Sericini and Ablaberini probably diversified during the early Tertiary, with dispersal of some basal Sericini to South America.     

An updated phylogeny of Anisoptera including formal convergence analysis of morphological characters

Family interrelationships among Anisoptera (dragonflies) are unresolved. Molecular markers applied thus far have not been particularly useful for resolving relationships at the family level. Previous morphological studies have depended heavily on characters of wing venation and articulation which are believed to display considerable degrees of homoplasy due to adaptations to different flight modes. Here, we present a comprehensive anatomical dataset of the head morphology of Anisoptera focusing on muscle organization and endoskeletal features covering nearly all families. The characters are illustrated in detail and incorporated into an updated morphological character matrix covering all parts of the dragonfly body. Phylogenetic analysis recovers all families as monophyletic clades except Corduliidae, Gomphidae as sister group to all remaining Anisoptera, and Austropetaliidae as sister group to Aeshnidae (=Aeshnoidea). The position of Petaluridae and Aeshnoidea to each other could not be resolved. Libelluloidea is monophyletic with Neopetalia and Cordulegastridae as first splits. Chlorogomphidae is sister to monophyletic [Synthemistidae + (‘Corduliidae’ + Libellulidae)]. In addition, we applied a recently published formal approach to detect concerted convergence in morphological data matrices and uncover possible homoplasies. Analyses show that especially head and thorax characters may harbour homoplasies. After exclusion of possible homoplastic characters, Gomphidae is corroborated as sister group to all remaining Anisoptera.     

Lake acidity,fish predation,and the distribution and abundance of some littoral insects

Densities of Corixidae (Hemiptera), larval Odonata, and large larval Trichoptera were estimated in the littoral zone of small lakes in an acid-stressed area near Sudbury, Ontario. Fish were present in some lakes and absent in others, and fishless lakes occurred across a wide range of pH. Corixidae were significantly more abundant in lakes without fish than in lakes with fish, and their numbers were not related to the pH of fishless lakes. Anisoptera (Odonata) larvae tended to be more numerous in benthic samples from fishless lakes than from lakes with fish, and their exuviae were significantly more abundant around fishless lakes. In most lakes, the assemblage was dominated by three species; Leucorrhinia glacialis, Libellula julia, and Cordulia shurtleffi. In lakes containing white sucker, Catostomus commersoni, Gomphus spp. were most numerous. In the most acid fishless lakes, L. julia was uncommon, and L. glacialis was extremely abundant. In fishless lakes, numbers of Anisoptera larvae and exuviae were negatively correlated with pH, though species richness was positively correlated with pH. Exuviae of Zygoptera (Odonata) were more abundant around fishless lakes, irrespective of pH. Larvae of Limnephilus (Trichoptera) were most abundant in non-acid fishless lakes, and absent at pH<5.2. Abundances of Banksiola (Trichoptera) were negatively correlated with the pH of fishless lakes.     

Evolutionary morphology of whip spiders: towards a phylogenetic system (Chelicerata: Arachnida: Amblypygi)*

The aim of this study is to present a cladogram and phylogenetic system and to use this to discuss the phylogeny and biogeography of the Amblypygi. A total of 29 morphological structures were studied, their plesiomorphic and apomorphic characters or character states were identified, and the resulting data matrix was analysed. As a result, the ‘old’Charontidae or Pulvillata emerge as a paraphyletic group; the genus Paracharon is the sister group of all other amblypygids, which are now termed Euamblypygi. The ‘new’Charontidae (sensu Quintero: the genera Stygophrynus and Charon) are the sister group of the Phrynida or Apulvillata; together they form the Neoamblypygi. The relationships of the genera of the Charinidae cannot be resolved with the available data. They may be a paraphyletic group. The genus Catageus is a possible candidate for being the sister group of the Neoamblypygi. The new system allows a discussion of the phylogeny and biogeography of whip spiders. It also points to unresolved taxa and thus indicates the questions future research should address.     

Interactions among coexisting larval Odonata: an in situ experiment using small enclosures

Field experiments using small replicated enclosures focused on interactions between larval populations of Epitheca cynosura and Ladona deplanata (Odonata: Anisoptera) — two species that emerge in early spring. The presence of Epitheca reduced the total biomass of Ladona, but Ladona had no significant effect on Epitheca. These early-emerging species reduced the biomass of small instars of late-emerging Anisoptera which colonized enclosures during the experiments; and the late-emerging Anisoptera seem to have inhibited colonization by Zygoptera larvae. Results are consistent with the importance of predatory (cannibalism or mutual predation) interactions in this community.     

广西大明山蜻蜓调查

2011年8月对广西大明山的蜻蜓物种资源进行了调查。结果表明广西大明山蜻蜓种类隶属2亚目11科46种,差翅亚目Anisoptera4科27种,束翅亚目Zygoptera7科19种。其中广西新纪录种8种(亚种)。     

Wing length allometry in Odonata: differences between families in relation to migratory behaviour

In insects, wing shape and body size are correlated with several aspects of behaviour, and the optimal morphology of wings is a trade-off between a number of functional demands in relation to behaviour (e.g. foraging, migration and sexual display). Dragonflies are spectacularly skilful flyers and present a range of different wing shapes, but to date, no detailed studies have been conducted in this group on wing length allometry in relation to body size. In this paper, we use published data on body length and wing length in all European and North American dragonflies to investigate differences in wing length allometries among Odonata taxa (suborders and families) and to relate these to behavioural patterns. We found different wing allometries between Zygoptera and Anisoptera, which are probably related to the flight mode and wing form of the two suborders. Among the Anisoptera, the Libellulidae showed a distinct wing length allometry from all other anisopteran families and migrants differed from non-migrant species. The first dichotomy is likely to reflect the adaptation of wing morphology of Libellulidae to sit-and-wait behaviour and to brief foraging flights (most species of this family are perchers) with respect to all other families, members of which are typically flyers. The second dichotomy reflects the trend of migrating species to have relatively longer wings than non-migrating members of the same family. Finally, wing length allometry differed among all the zygopteran families analysed, and this pattern suggested that each family evolved a particular wing morphology in response to peculiarities in behaviour, habitat and flight mode.     

Cladistic analysis of the Cirripedia Thoracica

We present a cladistic analysis of the Cirripedia Thoracica using morphological characters and the Acrothoracica and Ascothoracida as outgroups. The list of characters comprised 32 shell and soft body features. The operational taxonomic units (OTUs) comprised 26 well-studied fossil and extant taxa, principally genera, since uncertainty about monophyly exists for most higher ranking taxonomic units. Parsimony analyses using PAUP 3.1.1 and Hennig86 produced 189 trees of assured minimal length. We also examined character evolution in the consensus trees using MacClade and Clados. The monophyly of the Balanomorpha and the Verrucomorpha sensu stricto is confirmed, and all trees featured a sister group relationship between the ‘living fossil Neoverruca and me Brachylepadomorpha. In the consensus trees the sequential progression of ‘pedunculate‘sister groups up to a node containing Neolepas also conforms to current views, but certain well-established taxa based solely on plesiomorphies stand out as paraphyletic, such as Pedunculata (= Lepadomorpha); Eolepadinae, Scalpellomorpha and Chthamaloidea. The 189 trees differed principally in the position of shell-less pedunculates, Neoverruca, the scalpelloid Capitulum, and the interrelationships within the Balanomorpha, although the 50% majority rule consensus tree almost fully resolved the latter. A monophyletic Sessilia comprising both Verrucomorpha and Balanomorpha appeared among the shortest trees, but not in the consensus. A tree with a monophyletic Verrucomorpha including Neoverruca had a tree length two steps longer than the consensus trees. Deletion of all extinct OTUs produced a radically different tree, which highlights the importance of fossils in estimating cirripede phylogeny. Mapping of our character set onto a manually constructed cladogram reflecting die most recent scenario of cirripede evolution resulted in a tree length five steps longer than any of our shortest trees. Our analysis reveals that several key questions in cirripede phylogeny remain unsolved, notably the position of shell-less forms and the transition from ‘pedunculate‘to ‘sessile‘barnacles. The inclusion of more fossil species at this point in our understanding of cirripede phylogeny will only result in even greater levels of uncertainty. When constructing the character list we also identified numerous uncertainties in the homology of traits commonly used in discussing cirripede evolution. Our study highlights larval ultrastructure, detailed studies of early ontogeny, and molecular data as the most promising areas for future research.     

The sense of smell in Odonata: An electrophysiological screening

Volatile chemicals mediate a great range of intra- and interspecific signalling and information in insects. Olfaction has been widely investigated mostly in Neoptera while the knowledge of this sense in most basal insects such as Paleoptera (Odonata and Ephemeroptera) is still poor. In the present study we show the results of an electrophysiological screening on two model species, Libellula depressa (Libellulidae) and Ischnura elegans (Coenagrionidae), representatives of the two Odonata suborders Anisoptera and Zygoptera, with the aim to deep the knowledge on the sense of smell of this insect order. The antennal olfactory sensory neurons (OSNs) of these two species responded to the same 22 compounds (out of 48 chemicals belonging to different functional groups) encompassing mostly amines, carboxylic acids or aldehydes and belonging to green leaf volatiles, vertebrate related volatiles and volatiles emitted by standing waters bacteria. The properties of Odonata OSNs are very similar to those of ionotropic receptors (IRs) expressing OSNs in other insects.     

Phylogenetic analysis of Staphyliniformia (Coleoptera) based on characters of larvae and adults

Abstract. One hundred and twenty-one morphological characters of larvae and adults of the series Staphyliniformia were scored (multistate coding) and analysed to determine the family group relationships of the polyphagan groups Scarabaeoidea, Histeroidea, Hydrophiloidea and Staphylinoidea. Cladograms were rooted with exemplars of Adephaga, Archostemata, Myxophaga and the polyphagan families Dascillidae, Derodontidae, Eucinetidae and Scirtidae. Analyses of the same dataset with multistate characters re-coded as presence/absence (144 characters) produced cladograms that were similar to those produced from analyses of the original characters. Cladograms produced from partitioned larval and adult characters differed strongly, with adult-only trees more similar to those produced by combined data. The results confirm the monophyly of Hydrophiloidea + Histeroidea and of Staphylinoidea (including Hydraenidae). The Epimetopidae + Georissidae are the only strongly supported clade within Hydrophiloidea. A clade comprising Hydrochidae, Spercheidae and Hydrophilidae, and a sister-group relationship between the latter two families were confirmed in analyses of the data with presence/absence coding. Helophoridae, Epimetopidae and Georissidae are probably not a monophyletic unit, and additional evidence is needed for a reliable placement of Helophoridae. Scarabaeoidea are placed as a sister taxon of Hydrophiloidea + Histeroidea, but support for this relationship is weak. The branching pattern ((Hydraenidae + Ptiliidae) + (Leiodidae + Agyrtidae)), and a clade comprising Scydmaenidae, Silphidae and Staphylinidae (= ‘staphylinid group’) are well founded. The branching pattern (Orchymontiinae + (Prosthetopinae + (Ochthebiinae + Hydraeninae))) within Hydraenidae is confirmed. Poor resolution at the base of the trees and the placement of some nonstaphyliniform taxa (Dascillidae, Derodontidae, Scirtidae and Eucinetidae) as a sister group to a clade comprising Scarabaeoidea, Hydrophiloidea and Histeroidea suggests that Staphyliniformia may be paraphyletic. It is recommended that series names are eliminated from the classification of Polyphaga, at least for the more ‘primitive’ groups.