"TOTAL EVIDENCE" AND THE EVOLUTION OF HIGHLY SOCIAL BEES

Abstract— Recent cladistic analyses of the relationships of the tribes of apine bees have produced conflicting results. The question of single or multiple origins of advanced eusociality in the group is thus unresolved. However, the previous studies have each treated only a limited part of the character data available. A "total evidence" analysis is essayed here, combining the published morphological characters with sequence data from rRNA and mtDNA. The resulting cladogram is: Euglossini + (Bombini + (Meliponini + Apini). This supports a single origin of advanced eusociality, from a stage of primitive eusociality.     

Phylogenetic relationships within the corbiculate Apinae (Hymenoptera) and the evolution of eusociality

We provide a comparison of 520 base pairs of the mitochondrial cytochrome b gene from exemplars of the Meliponini, the Apini, the Bombini and the Euglossini to determine the phylogenetic relationships within the corbiculate Apinae. Our results strongly suggest (91–97% according to bootstrap resampling) that the Meliponini and the Bombini are sister groups. This finding agrees with those of other molecular studies, but is discordant with previous hypotheses based on morphology and the combination of molecular and morphological data. If the Bombini and Meliponini are sister groups and reversal of advanced eusociality is unlikely, then advanced eusociality arose twice within this clade. However, if reversion of eusociality occurred, then it is not possible to discriminate between single or dual origins of advanced eusociality within this group.     

Phylogenetic analysis of the corbiculate Apinae based on morphology of the sting apparatus (Hymenoptera: Apidae)

This study aimed to test the various competing hypotheses regarding the relationships among the four tribes of corbiculate apine bees (Euglossini “orchid bees”, Bombini “bumble bees”, Meliponini “stingless bees”, and Apini “honey bees”) with a completely new set of previously unstudied morphological characters derived from the sting apparatus. The result was one most parsimonious tree of 49 steps, CI = 89, RI = 93 that is perfectly congruent with most studies based on morphological and combined morphological/molecular data, i.e., Euglossini + (Bombini + (Meliponini + Apini)), supporting a well accepted scenario of social evolution for these bees. This data matrix was then combined with other published matrices for this group in order to perform simultaneous analyses. The problem of how to best combine the multiple matrices that did not use the same exemplars was investigated. © The Willi Hennig Society 2007.     

A re-evaluation of the phylogenetic relationships in the Apidae (Hymenoptera)

ABSTRACT .
The relationships between the four tribes in the bee family Apidae are re-examined. Characteristics of the postgena, presternum, antenna cleaner, arolium, female hind tibia and male genitalia, among others, support the placement of these tribes in three subfamilies: Meliponinae (Meliponini), Apinae (Apini) and Bombinae (Euglossini + Bombini). The Apinae is the sister group of the Bombinae. This tribal arrangement was originally proposed, using different characteristics, by Winston & Michener (1977).     

Phylogenetic utility of the major opsin in bees (Hymenoptera: Apoidea): a reassessment

Major opsin (LW Rh) DNA sequence has been reported to provide useful data for resolving phylogenetic relationships among tribes of corbiculate bees based on analyses of 502 bp of coding sequence. However, the corbiculate tribes are believed to be of Cretaceous age, and strong support for insect clades of this age from small data sets of nucleotide sequence data has rarely been demonstrated. To more critically assess opsin's phylogenetic utility we generated an expanded LW Rh data set by sequencing the same gene fragment from 52 additional bee species from 24 tribes and all six extant bee families. Analyses of this data set failed to provide substantial support for monophyly of corbiculate bees, for relationships among corbiculate tribes, or for most other well-established higher-level relationships among long-tongued bees. However, monophyly of nearly all genera and tribes is strongly supported, indicating that LW Rh provides useful phylogenetic signal at lower taxonomic levels. When our expanded LW Rh data set is combined with a morphological and behavioral data set for corbiculate bees, the results unambiguously support the traditional phylogeny of the corbiculate bee tribes: (Euglossini + (Bombini + (Meliponini + Apini))). This implies a single origin of advanced eusocial behavior among bees rather than dual origins, as proposed by several recent studies.     

Task-partitioned nectar transfer in stingless bees: work organisation in a phylogenetic context

Abstract 1. The eusocial corbiculate bee tribes comprise the Apini (honey bees), Bombini (bumble bees), and Meliponini (stingless bees). Honey bee foragers ( Apis ) transfer nectar to receiver bees within the nest. This is an example of task partitioning, in which a task is split into sub-tasks connected by material transfer. Nectar transfer does not occur in Bombini. Although it is reported in some species of Meliponini, it has not been subject to detailed study.
2. Nectar transfer was investigated in five genera of Meliponini from Yucatan, Mexico ( Melipona , Trigona , Scaptotrigona , Nannotrigona , and Plebeia ). Nectar transfer occurred in all species and for > 99% of foragers. Multiple transfer, in which a forager unloads nectar to more than one receiver, occurred but at a lower level than in Apis . In M. beecheii , multiple transfer was associated strongly with putative recruitment dances.
3. The data provide some support for the hypothesis that task partitioning is favoured by large colony size, in that the Meliponini never have small colonies because colonies are swarm founded. This ensures that colonies are always large enough to prevent delays in finding a transfer partner imposing high costs. Further tests of this hypothesis are suggested.
4. Viewed in a phylogenetic context, the most parsimonious interpretation is that nectar transfer evolved once in the clade (Apini + Meliponini).     

Comparative morphology of the postmentum of bees (Hymenoptera: Apoidea) with special remarks on the evolution of the lorum

A comparative morphological study of the basal sclerites of the bee labium was undertaken. The term postmentum was applied to the basal sclerite of the bee labium. In contrast to recent interpretations, the undivided postmentum was found to be ancestral for bees and homologous with the single postmental sclerite of other Hymenoptera. This sclerite has previously been incorrectly indentified as two separate sclerites, usually termed mentum and lorum (= submentum), for many short-tongued bees (Colletidae, Halictidae, Andrenidae, Melittidae) and for the long-tongued families of bees (Anthophoridae, Fideliidae Megachilidae) excepting most members of the Apidae. The postmentum, divided into a true mentum and lorum, was found only in certain members of the Apidae. A phylogenetic implication resulting from this study shows that the Andrenidae may be the sister group to the MCL-T group (Melittidae, Ctenoplectridae and long-tongued families of bees). In addition, it is proposed that the Euglossinae should be the sister group to the Apinae (Meliponini, Apini and Bombini). This means that eusociality may be considered a synapomorphy for the Apinae.     

Cladistic analysis of self‐grooming indicates a single origin of eusociality in corbiculate bees (Hymenoptera: Apidae)

Behavioural traits have been used extensively in recent years as an important character source for making phylogenetic inferences. The phylogenetic positions of the members of the Apini subtribe are increasingly being debated, and new characters must be examined. We analysed the presence and absence of certain behavioural patterns, as well as the sequences of some of these patterns, to generate 79 characters. Eleven species comprised the ingroup, and Xylocopini comprised the outgroup. Parsimony analysis showed that the most parsimonious tree was (Euglossina(Bombina(Apina+Meliponina))). This topology is consistent with most studies that use morphological data and the few that use behavioural data, which suggests that advanced eusociality arose only once in a common ancestor of the clade Apina plus Meliponina; however, this hypothesis is inconsistent with our molecular data. Thus we considered behavioural, molecular, and morphological data and recovered the same topology, in which eusociality has a single origin in corbiculate bees.     

Global stingless bee phylogeny supports ancient divergence,vicariance, and long distance dispersal

Stingless bees (Meliponini) are one of only two highly eusocial bees, the other being the well studied honey bee (Apini). Unlike Apini, with only 11 species in the single genus Apis, stingless bees are a large and diverse taxon comprising some 60 genera, many of which are poorly known. This is the first attempt to infer a phylogeny of the group that includes the world fauna and extensive molecular data. Understanding the evolutionary relationships of these bees would provide a basis for behavioural studies within an evolutionary framework, illuminating the origins of complex social behaviour, such as the employment of dance and sound to communicate the location of food or shelter. In addition to a global phylogeny, we also provide estimates of divergence times and ancestral biogeograhic distributions of the major groups. Bayesian and maximum likelihood analyses strongly support a principal division of Meliponini into Old and New World groups, with the Afrotropical+Indo‐Malay/Australian clades comprising the sister group to the large Neotropical clade. The meliponine crown clade is inferred to be of late Gondwanan origin (approximately 80 Mya), undergoing radiations in the Afrotropical and Indo‐Malayan/Australasian regions, approximately 50–60 Mya. In the New World, major diversifications occurred approximately 30–40 Mya. © 2010 The Linnean Society of London, Biological Journal of the Linnean Society, 2010, 99 , 206–232.     

Phylogeny of the bees of the family Apidae based on larval characters with focus on the origin of cleptoparasitism (Hymenoptera: Apiformes)

Abstract.  Fifty-four genera of the bee family Apidae comprising almost all tribes were analysed based on 77 traditional and one new character of the mature larvae. Nine, especially cleptoparasitic species, were newly added. Analyses were performed by maximum parsimony and Bayesian inference. Trees inferred from the analysis of the complete dataset were rooted by taxa from the families Melittidae and Megachilidae. Unrooted trees inferred from the analysis of the partial dataset (excluding outgroup taxa) are also presented to preclude possible negative effects of the outgroup on the topology of the ingroup. Only the subfamily Nomadinae was statistically well supported. The monophyly of the subfamilies Xylocopinae and Apinae was not topologically recovered. The monophyly of the tribe Tetrapediini was supported, and this tribe was found to be related to xylocopine taxa. At the very least, larval morphology suggests that Tetrapedia is not a member of the subfamily Apinae. Our analyses support the monophyly of the Eucerine line (Emphorini, Eucerini, Exomalopsini, Tapinotaspidini) and of the Apine line (Anthophorini, Apini, Bombini, Centridini, Euglossini, Meliponini). All analyses support the monophyly of totally cleptoparasitic tribes of the subfamily Apinae. We named this group the Melectine line (Ericrocidini, Isepeolini, Melectini, Osirini, Protepeolini, Rhathymini). In previous studies all these cleptoparasitic tribes were considered independent evolutionary lineages. Our results suggest that their similarities with hosts in morphology and pattern are probably the result of convergence and host–parasite co-evolution than phylogenetic affinity. According to the present analysis, the cleptoparasitism has evolved independently only six times within the family Apidae.     

Phylogeny of the orchid bees (Hymenoptera: Apinae: Euglossini): DNA and morphology yield equivalent patterns

Orchid bees (Euglossini) are spectacular long-tongued Neotropical bees important in the pollination of Neotropical long-corolla flowers, particularly some orchids. Besides remarkably long tongues, males in particular exhibit other flower-related adaptations, including setal brushes on the foretarsi used for rasping the petals of orchids while collecting aromatic compounds. These compounds are stored in large swollen tibiae and are thought to play an important role in courtship behavior. Euglossini are also unusual in lacking sociality; they are the only tribe among the corbiculate bees that are not eusocial, and two of the genera are cleptoparasitic. Each genus exhibits distinct behavioral traits including nest architecture and host-parasite interactions, yet their evolution is unknown. Despite previous phylogenetic studies of on morphological characters, the relationships among the five euglossine genera remain under debate. We investigate euglossine generic relationships using DNA sequence data from four genes and new morphological characters. The morphological and molecular data yield congruent evolutionary patterns, and combining the data gives a fully resolved and well supported phylogeny of Euglossini.     

The Phylogeny of the Gasterosteidae: Combining Behavioral and Morphological Data Sets

Genealogical relationships among the five genera of stickleback (Gasterosteidae: Smegmomorpha) fishes were reconstructed based upon 47 behavioral and 89 morphological characters. Phylogenetic systematic analysis of the combined data set produced a single MPT (consistency index of 82.54%) structured ( Spinachia ( Apeltes-((Pungitius + Culaea )( Gasterosteus aculeatus + G. wheatlandi )))). When analyzed separately, both behavioral and morphological data sets produced the same tree topology, although the consistency index (excluding uninformative characters) was higher for the behavioral (90.70%) than for the morphological (68.13%) characters. This total evidence analysis provides a highly robust estimate of phylogeny for the sticklebacks and reiterates the importance of discarding a priori biases about behavioral evolution when selecting characters for phylogenetic analysis.     

Non-Euglossine bees also function as pollinators of Sinningia species (Gesneriaceae) in southeastern brazil

Pollination by male and female Euglossini bees, euglossophily, was suggested for a number of neotropical Gesneriaceae species. Information on bee species other than Euglossini as pollinators of neotropical members of this family is limited, and in the tribe Sinningieae data about bee pollination are still lacking. Here, we report on floral biology and bee pollination of four Sinningia species: S. schiffneri, S. eumorpha, S. villosa, and Sinningia "canastrensis". The flower features, such as corolla size, shape, and colour, are very different among the four species, but all conform to the melittophilous syndrome. The average nectar volume and sugar amount is low in S. schiffneri, S. eumorpha, and Sinningia "canastrensis", when compared to that of S. villosa, but low nectar amounts is a general feature of Sinningia species. The main pollinators of the four species are: small Tapinotaspidini (Trigonopedia ferruginea) of S. schiffneri, large Bombini (Bombus morio) and large Centridini (Epicharis morio) of S. eumorpha, large Euglossini (Eulaema cingulata and Eufriesea surinamensis) of S. villosa, and large Euglossini (Eufriesea violascens) and Megachilini (Megachile sp.) of Sinningia "canastrensis". Out of the four species, only S. villosa is exclusively Euglossini-pollinated. The marked differences in flower features and nectar production of these Sinningia species may reflect their pollination by distinct groups of bees. These results strengthen the idea of multiple origins for the pollination systems involving bees within this genus, which is highly supported by molecular phylogenetic analyses.     

Total evidence phylogeny of Gasterosteidae: combining molecular, morphological and behavioral data

Genealogical relationships among Gasterosteidae (Teleostei: Gasterosteiformes) were tested with 84 morphological, 48 behavioral, and 2879 molecular characters. Phylogenetic analysis of the combined data set identified a single (CI = 0.735) best‐supported hypothesis (Spinachia (Apeltes ((Pungitius + Culaea)(Gasterosteus aculeatus + G. wheatlandi)))). This hypothesis is identical to previous phylogenetic propositions proposed on the basis of behavioral, and behavioral plus morphological data. Our hypothesis, however, differed from a molecular‐based phylogeny in the placement of Apeltes. This analysis highlights the importance of combining all available evidence in order to produce the best‐supported proposition of genealogical relationships.     

Mitochondrial genome differences between the stingless bees Melipona rufiventris and Melipona mondury (Apidae: Meliponini)

Within the Meliponini, a widely distributed group of stingless bees, Melipona rufiventris has been considered as a single, cohesive species. Recently, analysis of morphological characters led to the splitting of this species into two species, M. mondury and M. rufiventris. The former occurs in the Atlantic Rain Forest ranging from Santa Catarina to Bahia States, while the latter is found in other parts of Brazil. We used PCR + RFLP to identify genetic marker patterns of the mtDNA between these species. Nine mtDNA regions were amplified and digested with four restriction enzymes (EcoRI, EcoRV, HindIII, and HinfI). Six species-specific restriction sites were identified for M. mondury and M. rufiventris with all enzymes, except for HindIII. The molecular data agree with the morphological classification.     

Subtle visits despite guards: Theft from nest of stingless bee (Meliponini) by orchid bee (Euglossini)

The theft of food items and building materials from nests of stingless bees (Hymenoptera: Apidae: Meliponini) is most commonly carried out by other Meliponini bee species, especially by those related to genus Lestrimelitta Friese, 1903. In this note, we recorded the theft of cerumen and propolis made by a species of the orchid bee, Euglossa annectans Dressler, 1982 (Apidae: Euglossini) from the nest of the stingless bee Tetragonisca angustula (Latreille, 1811). We noticed that the guard workers of T. angustula did not attack the thief bee even though it did not come sufficiently close to the entrance tube of the nest.     

Miocene honey bees from the Randeck Maar of southwestern Germany (Hymenoptera, Apidae)

The Miocene Randeck Maar (southwestern Germany) is one of the only sites with abundant material of fossil honey bees. The fauna has been the focus of much scrutiny by early authors who recognized multiple species or subspecies within the fauna. The history of work on the Randeck Maar is briefly reviewed and these fossils placed into context with other Tertiary and living species of the genus Apis Linnaeus (Apinae: Apini). Previously unrecorded specimens from Randeck Maar were compared with earlier series in an attempt to evaluate the observed variation. A morphometric analysis of forewing venation angles across representative Recent and Tertiary species of Apis as well as various non-Apini controls was undertaken to evaluate the distribution of variation in fossil honey bees. The resulting dendrogram shows considerable variation concerning the wing venation of Miocene Apini, but intergradation of other morphological characters reveals no clear pattern of separate species. This suggests that a single, highly variable species was present in Europe during the Miocene. The pattern also supports the notion that the multiple species and subspecies proposed by earlier authors for the Randeck Maar honey bee fauna are not valid, and all are accordingly recognized as Apis armbrusteri Zeuner.     

The complete nucleotide sequence and gene organization of the mitochondrial genome of the bumblebee, Bombus ignitus (Hymenoptera: Apidae)

The complete 16,434-bp nucleotide sequence of the mitogenome of the bumble bee, Bombus ignitus (Hymenoptera: Apidae), was determined. The genome contains the base composition and codon usage typical of metazoan mitogenomes. An unusual feature of the B. ignitus mitogenome is the presence of five tRNA-like structures: two each of the tRNALeu(UUR)-like and tRNASer(AGN)-like sequences and one tRNAPhe-like sequence. These tRNA-like sequences have proper folding structures and anticodon sequences, but their functionality in their respective amino acid transfers remained uncertain. Among these sequences, the tRNALeu(UUR)-like sequence and the tRNASer(AGN)-like sequence are seemingly located within the A+T-rich region. This tRNASer(AGN)-like sequence is highly unusual in that its sequence homology is very high compared to the tRNAMet of other insects, including Apis mellifera, but it contains the anticodon ACT, which designates it as tRNASer(AGN). All PCG and rRNAs are conserved in positions observed most frequently in insect mitogenome structures, but the positions of the tRNAs are highly variable, presenting a new arrangement for an insect mitogenome. As a whole, the B. ignitus mitogenome contains the highest A+T content (86.9%) found in any of the complete insects mt sequences determined to date. All protein-coding sequences started with a typical ATN codon. Nine of the 13 PCGs have a complete termination codon (all TAA), but the remaining four genes terminate with the incomplete TA or T. All tRNAs have the typical clover-leaf structures of mt tRNAs, except for tRNASer(AGN), in which the DHU arm forms a simple loop. All anticodons of B. ignitus tRNAs are identical to those of A. mellifera. In the A+T-rich region, a highly conserved sequence block that was previously described in Orthoptera and Diptera was also present. The stem-and-loop structures that may play a role in the initiation of mtDNA replication were also found in this region. Phylogenetic analysis among three corbiculate tribes, represented by Melipona bicolor (Meliponini), A. mellifera (Apini), and B. ignitus (Bombini), showed the closest relationship between M. bicolor and B. ignitus.     

An overview of cytogenetics of the tribe Meliponini (Hymenoptera: Apidae)

The present study provides a comprehensive review of cytogenetic data on Meliponini and their chromosomal evolution. The compiled data show that only 104 species of stingless bees, representing 32 of the 54 living genera have been studied cytogenetically and that among these species, it is possible to recognize three main groups with n?=?9, 15 and 17, respectively. The first group comprises the species of the genus Melipona, whereas karyotypes with n?=?15 and n?=?17 have been detected in species from different genera. Karyotypes with n?=?17 are the most common among the Meliponini studied to date. Cytogenetic information on Meliponini also shows that although chromosome number, in general, is conserved among species of a certain genus, other aspects, such as chromosome morphology, quantity, distribution and composition of heterochromatin, may vary between them. This reinforces the fact that the variations observed in the karyotypes of different Meliponini groups cannot be explained by a single theory or a single type of structural change. In addition, we present a discussion about how these karyotype variations are related to the phylogenetic relationships among the different genera of this tribe.     

THE USEFULNESS OF BEHAVIOR FOR PHYLOGENY ESTIMATION: LEVELS OF HOMOPLASY IN BEHAVIORAL AND MORPHOLOGICAL CHARACTERS

It is widely believed that behavior is more evolutionarily labile and/or more difficult to characterize than morphology, and thus that behavioral characters are not as useful as morphological characters for estimating phylogenetic relationships. To examine the relative utility of behavior and morphology for estimating phylogeny, we compared levels of homoplasy for morphological and behavioral characters that have been used in systematic studies. In an analysis of 22 data sets that contained both morphological and behavioral characters we found no significant difference between mean consistency indices (CIs, which measure homoplasy) within data sets for the two types of characters. In a second analysis we compared overall CIs for 8 data sets comprised entirely of behavioral characters with overall CIs for 32 morphological data sets and found no significant difference between the two types of data sets. For both analyses, 95% confidence limits on the difference between the two types of characters indicate that, even if given the benefit of the doubt, morphological characters could not have substantially higher mean CIs than behavioral characters. These results do not support the idea that behavioral characters are less useful than morphological characters for the estimation of phylogeny.