Pollination by fungus gnats (Diptera: Mycetophilidae) and self-recognition sites in Tolmiea menziesii (Saxifragaceae)

Coleeae (Bignoniaceae) are a tribe almost entirely restricted to Madagascar. Coleeae have previously been placed in neotropical Crescentieae due to species with indehiscent fruits, a character otherwise unusual in Bignoniaceae. A phylogeny based on three chloroplast regions (ndhF, trnT-L spacer, trnL-F spacer) identifies a monophyletic Coleeae that is endemic to Madagascar and surrounding islands of the Indian Ocean (Seychelles, Comores and Mascarenes). African Kigelia is not a member of Coleeae, rather it is more closely related to a subset of African and Southeast Asian species of Tecomeae. The molecular phylogeny indicates that indehiscent fruit have arisen repeatedly in Bignoniaceae: in Coleeae, Kigelia and Crescentieae. The characteristic fleshy fruits of species of Coleeae likely arose autochthonously in Madagascar. Within Coleeae Colea and Ophiocolea are sisters, Phyllarthron is sister to Colea + Ophiocolea, and Rhodocolea is sister to the rest of the tribe.     

Floral sources used by the orchid bee Euglossa cordata (Linnaeus, 1758) (Apidae: Euglossini) in an urban area of south-eastern Brazil

Although orchid bees (Apidae: Euglossini) are known as key pollinators in tropical ecosystems, knowledge of their floral sources is still scarce, especially for those species commonly found in urban environments. We aimed to identify the pollen, nectar and resin sources used by the widespread species Euglossa cordata in an urban area in south-eastern Brazil. The residual pollen from 81 brood cells of nine nests reactivated between October 2013 and December 2014 was acetolysed and analysed. A total of 50 pollen types belonging to 20 botanical families were identified in the samples. Pollen sources included species from seven families; five of them were plants with poricidal anthers (Bixaceae, Commelinaceae, Fabaceae, Melastomataceae and Solanaceae). The mass-flowering trees Handroanthus chrysotrichus and H. heptaphyllus (Bignoniaceae) are firstly reported as important pollen sources to an orchid bee species. Nectar was collected primarily from plants with long, tubular corolla as Acanthaceae, Apocynaceae, Bignoniaceae and Convolvulaceae. The vine species Dalechampia stipulacea (Euphorbiaceae) acted as a floral resin source. All pollen sources consisted of native plants whereas some exotic plant species were visited for nectar collection. Although nesting in an area encompassed by a high proportion of invasive plant species, Euglossa cordata females preferred to visit native plants to gather floral resources.     

A molecular phylogeny and classification of Bignoniaceae

Bignoniaceae are woody, trees, shrubs, and lianas found in all tropical floras of the world with lesser representation in temperate regions. Phylogenetic analyses of chloroplast sequences (rbcL, ndhF, trnL-F) were undertaken to infer evolutionary relationships in Bignoniaceae and to revise its classification. Eight clades are recognized as tribes (Bignonieae, Catalpeae, Coleeae, Crescentieae, Jacarandeae, Oroxyleae, Tecomeae, Tourrettieae); additional inclusive clades are named informally. Jacarandeae and Catalpeae are resurrected; the former is sister to the rest of the family, and the latter occupies an unresolved position within the "core" Bignoniaceae. Tribe Eccremocarpeae is included in Tourrettieae. Past classifications recognized a large Tecomeae, but this tribe is paraphyletic with respect to all other tribes. Here Tecomeae are reduced to a clade of approximately 12 genera with a worldwide distribution in both temperate and tropical ecosystems. Two large clades, Bignonieae and Crescentiina, account for over 80% of the species in the family. Coleeae and Crescentieae are each included in larger clades, the Paleotropical alliance and Tabebuia alliance, respectively; each alliance includes a grade of taxa assigned to the traditional Tecomeae. Parsimony inference suggests that the family originated in the neotropics, with at least five dispersal events leading to the Old World representatives.     

Diversity and evolution of pollinator rewards and protection by <Emphasis Type="Italic">Macaranga</Emphasis> (Euphorbiaceae) bracteoles

Flowering plants have modified their floral organs in remarkably diverse ways to optimize their interaction with pollinators. Although floral organs represent a major source of floral diversity, many plants also use extrafloral organs, such as bracts and bracteoles, in interacting with pollinators; however, the evolutionary dynamics of non-floral organs involved in pollination are poorly studied. The genus Macaranga is characterized by protective mutualisms with ants that potentially interfere with pollinators on flowers. Macaranga flowers lack perianths and, notably, bracteoles serve the dual function of rewarding pollinators and protecting them from guarding ants; in one group of species, bracteoles provide a nectar reward to generalist pollinators, while in another group, bracteole “chambers” protect thrips or hemipteran pollinators that use these structures as feeding and breeding sites. We examined the diversity and evolutionary dynamics of inflorescence morphology in Macaranga, focusing on bracteoles. We recognized three inflorescence types based on examination of herbarium materials: Discoid-gland, which possess disc-shaped glands on the bracteole surfaces (including all the generalist-pollinated species); Enclosing, in which bracteoles cover flowers (including all the thrips- and hemipteran-pollinated species); and Inconspicuous, in which bracteoles are small, narrow or absent. Ancestral state reconstruction indicated that inflorescence morphologies have changed multiple times in the genus. These findings suggest that morphological changes in non-floral characters (bracteoles) of Macaranga species have occurred as frequently as in the floral structures of many flowering plants. The multiple evolutions of the Enclosing bracteoles, which protect pollinators, might have been facilitated by pollination interference from mutualistic ants.     

Nectar Robbing Positively Influences the Reproductive Success of Tecomella undulata (Bignoniaceae)

The net consequence of nectar robbing on reproductive success of plants is usually negative and the positive effect is rarely produced. We evaluated the influence of nectar robbing on the behaviour of pollinators and the reproductive success of Tecomella undulata (Bignoniaceae) in a natural population. Experimental pollinations showed that the trees were strictly self-incompatible. The three types of floral colour morphs of the tree viz. red, orange and yellow, lacked compatibility barriers. The pollinators (Pycnonotus cafer and Pycnonotus leucotis) and the robber (Nectarinia asiatica) showed equal preference for all the morphs, as they visited each morph with nearly equal frequency and flower-handling time. The sunbirds caused up to 60% nectar robbing, mostly (99%) by piercing through the corolla tube. Although nectar is replenished at regular intervals, insufficient amount of nectar compelled the pollinators to visit additional trees in bloom. Data of manual nectar robbing from the entire tree showed that the pollinators covered lower number of flowers per tree (5 flowers/tree) and more trees per bout (7 trees/bout) than the unrobbed ones (19 flowers/tree and 2 trees bout). The robbed trees set a significantly greater amount of fruits than the unrobbed trees. However, the number of seeds in a fruit did not differ significantly. The study shows that plant-pollinator-robber interaction may benefit the self-incompatible plant species under conditions that increases the visits of pollinators among the compatible conspecifics in a population.     

Visitation rate of pollinators and nectar robbers to the flowers and inflorescences of Tabebuia aurea (Bignoniaceae): effects of floral display size and habitat fragmentation

Large floral displays favour pollinator attraction and the import and export of pollen. However, large floral displays also have negative effects, such as increased geitonogamy, pollen discounting and nectar/pollen robber attraction. The size of the floral display can be measured at different scales (e.g. the flower, inflorescence or entire plant) and variations in one of these scales may affect the behaviour of flower visitors in different ways. Moreover, the fragmentation of natural forests may affect flower visitation rates and flower visitor behaviour. In the present study, video recordings of the inflorescences of a tree species (Tabebuia aurea) from the tropical savannah of central Brazil were used to examine the effect of floral display size at the inflorescence and tree scales on the visitation rate of pollinators and nectar robbers to the inflorescence, the number of flowers approached per visit, the number of visits per flower of potential pollinators and nectar robbers, and the interaction of these variables with the degree of landscape disturbance. Nectar production was quantified with respect to flower age. Although large bees are responsible for most of the pollination, a great diversity of flower insects visit the inflorescences of T. aurea. Other bee and hummingbird species are highly active nectar robbers. Increases in inflorescence size increase the visitation rate of pollinators to inflorescences, whereas increases in the number of inflorescences on the tree decrease visitation rates to inflorescences and flowers. This effect has been strongly correlated with urban environments in which trees with the largest floral displays are observed. Pollinating bees (and nectar robbers) visit few flowers per inflorescence and concentrate visits to a fraction of available flowers, generating an overdispersed distribution of the number of visits per inflorescence and per flower. This behaviour reflects preferential visits to young flowers (including flower buds) with a greater nectar supply.     

Three pollination guilds and variation in floral characteristics of Bornean gingers (Zingiberaceae and Costaceae)

The pollinators of 29 ginger species representing 11 genera in relation to certain floral morphological characteristics in a mixed-dipterocarp forest in Borneo were investigated. Among the 29 species studied, eight were pollinated by spiderhunters (Nectariniidae), 11 by medium-sized Amegilla bees (Anthophoridae), and ten by small halictid bees. These pollination guilds found in gingers in Sarawak are comparable to the pollination guilds of neotropical Zingiberales, i.e., hummingbird-, and euglossine-bee-pollinated guilds. Canonical discriminant analysis revealed that there were significant correlations between floral morphology and pollination guilds and suggests the importance of plant–pollinator interactions in the evolution of floral morphology. Most species in the three guilds were separated on the plot by the first and second canonical variables. Spiderhunter-pollinated flowers had longer floral tubes, while Amegilla-pollinated flowers had wider lips than the others, which function as a platform for the pollinators. Pistils and stamens of halictid-pollinated flowers were smaller than the others. The fact that gingers with diverse morphologies in a forest with high species diversity were grouped into only three pollination guilds and that the pollinators themselves showed low species diversity suggests that many species of rare understory plants have evolved without segregating pollinators in each pollination guild.     

Divergence on floral traits and vertebrate pollinators of two endemic Encholirium bromeliads

Shifts in pollen vectors favour diversification of floral traits, and differences in pollination strategies between congeneric sympatric species can contribute to reproductive isolation. Divergence in flowering phenology and selfing could also reduce interspecific crossing between self‐compatible species. We investigated floral traits and visitation rates of pollinators of two sympatric Encholirium species on rocky outcrops to evaluate whether prior knowledge of floral characters could indicate actual pollinators. Data on flowering phenology, visitation rates and breeding system were used to evaluate reproductive isolation. Flowering phenology overlapped between species, but there were differences in floral characters, nectar volume and concentration. Several hummingbird species visited flowers of both Encholirium spp., but the endemic bat Lonchophylla bokermanni and an unidentified sphingid only visited E. vogelii. Pollination treatments demonstrated that E. heloisae and E. vogelii were partially self‐compatible, with weak pollen limitation to seed set. Herbivores feeding on inflorescences decreased reproductive output of both species, but for E. vogelii the damage was higher. Our results indicate that actual pollinators can be known beforehand through floral traits, in agreement with pollination syndromes stating that a set of floral traits can be associated with the attraction of specific groups of pollinators. Divergence on floral traits and pollinator assemblage indicate that shifts in pollination strategies contribute to reproductive isolation between these Encholirium species, not divergence on flowering phenology or selfing. We suggest that hummingbird pollination might be the ancestral condition in Encholirium and that evolution of bat pollination made a substantial contribution to the diversification of this clade.     

Comparative floral morphometrics in day-flowering, night-flowering and self-pollinated Caryophylloideae (Agrostemma, Dianthus, Saponaria, Silene, and Vaccaria)

Morphological variation of flowers with different pollination modes was studied in 53 species representing five genera (Agrostemma, Dianthus, Saponaria, Silene s.l., Vaccaria) of the subfamily Caryophylloideae. All species were classified a priori as either diurnal, nocturnal, or selfing. Canonical discriminant analysis (CDA) of 13 floral characters revealed significant correlations between floral morphometry and pollination modes. A CDA of taxonomical groups represented with more than three species (Dianthus, and subgroups of Silene s.l.: Lychnis, Silene s.str., and Viscaria), revealed that the discrimination between these four taxa, based on the same floral characters, is also well supported. Main factors for the discrimination of the species with different pollination modes were characters that (a) define nectar accessibility (calyx length, calyx tooth length), (b) are related to the positioning of anthers and stigmatic areas in relation to pollinators and (c) are important for the visual attractiveness of flowers (plate width). The functional distance between the nectar source (anthophore base) and contact zone with pollinators (style tips), given as the sum of anthophore length, ovary length, and style length (hence called AOS-complex) is better correlated with the calyx length than the single characters. Further, the total AOS-complex length differs significantly between pollination modes suggesting that these characters form a functionally linked complex that is related to the pollen placement on, and stigma contact with, the pollinator's body. However, the contribution of anthophore and style length to the total AOS-complex differed significantly between Silene s.l. and other taxa indicating that the taxonomic groups follow different evolutionary ways for the construction of the functionally linked AOS-complex.     

Contrasting phylogenetic signals and evolutionary rates in floral traits of Neotropical lianas

The diversity of floral forms has long been considered a prime example of radiation through natural selection. However, little is still known about the evolution of floral traits, a critical piece of evidence for the understanding of the processes that may have driven flower evolution. We studied the pattern of evolution of quantitative floral traits in a group of Neotropical lianas (Bignonieae, Bignoniaceae) and used a time‐calibrated phylogeny as basis to: (1) test for phylogenetic signal in 16 continuous floral traits; (2) evaluate the rate of evolution in those traits; and (3) reconstruct the ancestral state of the individual traits. Variation in floral traits among extant species of Bignonieae was highly explained by their phylogenetic history. However, opposite signals were found in floral traits associated with the attraction of pollinators (calyx and corolla) and pollen transfer (androecium and gynoecium), suggesting a differential role of selection in different floral whorls. Phylogenetic independent contrasts indicate that traits evolved at different rates, whereas ancestral character state reconstructions indicate that the ancestral size of most flower traits was larger than the mean observed sizes of the same traits in extant species. The implications of these patterns for the reproductive biology of Bignonieae are discussed. © 2011 The Linnean Society of London, Biological Journal of the Linnean Society, 2011, 102 , 378–390.     

Differentiation of floral color and odor in two fly pollinated species ofMetrodorea (Rutaceae) from Brazil

We investigated if differences in morphological characters in two species ofMetrodorea (Rutaceae) from Brazilian semideciduous forests correspond to some pollination divergence.M. nigra andM. stipularis are sympatric species, display a similar floral morphology, are protandrous, self-incompatible, their flower periods overlap, and both are pollinated by flies.M. nigra main pollinators arePseudoptiloleps nigripoda (Muscidae) andFannia sp. (Fanniidae);M. stipularis major pollinators arePhaenicia eximia (Calliphoridae),Palpada sp. andOrnidia obesa (Syrphidae). The distinct floral odor (disagreeable inM. nigra and sweet inM. stipularis) and color (brownish violet vs. pale yellow) determine the differences on type and number of floral visitors observed. Several species from semideciduous forests initially considered to be pollinated by diverse insects, present flies as main pollinators, stressing the importance of fly pollination in such habitats.     

Genetic rescue of remnant tropical trees by an alien pollinator.

Habitat fragmentation is thought to lower the viability of tropical trees by disrupting their mutualisms with native pollinators. However, in this study, Dinizia excelsa (Fabaceae), a canopy-emergent tree, was found to thrive in Amazonian pastures and forest fragments even in the absence of native pollinators. Canopy observations indicated that African honeybees (Apis mellifera scutellata) were the predominant floral visitors in fragmented habitats and replaced native insects in isolated pasture trees. Trees in habitat fragments produced, on average, over three times as many seeds as trees in continuous forest, and microsatellite assays of seed arrays showed that genetic diversity was maintained across habitats. A paternity analysis further revealed gene flow over as much as 3.2 km of pasture, the most distant pollination precisely recorded for any plant species. Usually considered only as dangerous exotics, African honeybees have become important pollinators in degraded tropical forests, and may alter the genetic structure of remnant populations through frequent long-distance gene flow.     

Forests are critically important to global pollinator diversity and enhance pollination in adjacent crops

Although the importance of natural habitats to pollinator diversity is widely recognized, the value of forests to pollinating insects has been largely overlooked in many parts of the world. In this review, we (i) establish the importance of forests to global pollinator diversity, (ii) explore the relationship between forest cover and pollinator diversity in mixed-use landscapes, and (iii) highlight the contributions of forest-associated pollinators to pollination in adjacent crops. The literature shows unambiguously that native forests support a large number of forest-dependent species and are thus critically important to global pollinator diversity. Many pollinator taxa require or benefit greatly from resources that are restricted to forests, such as floral resources provided by forest plants (including wind-pollinated trees), dead wood for nesting, tree resins, and various non-floral sugar sources (e.g. honeydew). Although landscape-scale studies generally support the conclusion that forests enhance pollinator diversity, findings are often complicated by spatial scale, focal taxa, landscape context, temporal context, forest type, disturbance history, and external stressors. While some forest loss can be beneficial to pollinators by enhancing habitat complementarity, too much can result in the near-elimination of forest-associated species. There is strong evidence from studies of multiple crop types that forest cover can substantially increase yields in adjacent habitats, at least within the foraging ranges of the pollinators involved. The literature also suggests that forests may have enhanced importance to pollinators in the future given their role in mitigating the negative effects of pesticides and climate change. Many questions remain about the amount and configuration of forest cover required to promote the diversity of forest-associated pollinators and their services within forests and in neighbouring habitats. However, it is clear from the current body of knowledge that any effort to preserve native woody habitats, including the protection of individual trees, will benefit pollinating insects and help maintain the critical services they provide.     

Asymmetric hybridization and gene flow between Joshua trees (Agavaceae: Yucca) reflect differences in pollinator host specificity

The angiosperms are by far the largest group of terrestrial plants. Their spectacular diversity is often attributed to specialized pollination. Obligate pollination mutualisms where both a plant and its pollinator are dependent upon one another for reproduction are thought to be prone to rapid diversification through co‐evolution and pollinator isolation. However, few studies have evaluated the degree to which pollinators actually mediate reproductive isolation in these systems. Here, we examine evidence for hybridization and gene flow between two subspecies of Joshua tree (Yucca brevifolia brevifolia and Yucca brevifolia jaegeriana) pollinated by two sister species of yucca moth. Previous work indicated that the pollinators differ in host specificity, and DNA sequence data suggested asymmetric introgression between the tree subspecies. Through intensive sampling in a zone of sympatry, a large number of morphologically intermediate trees were identified. These included trees with floral characters typical of Y. b. jaegeriana, but vegetative features typical of Y. b. brevifolia. The opposite combination—Y. b. brevifolia flowers with Y. b. jaegeriana vegetative morphology—never occurred. Microsatellite genotyping revealed a high frequency of genetically admixed, hybrid trees. Coalescent‐based estimates of migration indicated significant gene flow between the subspecies and that the direction of gene flow matches differences in pollinator host fidelity. The data suggest that pollinator behaviour determines the magnitude and direction of gene flow between the two subspecies, but that specialized pollination alone is not sufficient to maintain species boundaries. Natural selection may be required to maintain phenotypic differences in the face of ongoing gene flow.     

Low local but high beta diversity of tropical forest dung beetles in Madagascar

Aims We have compared local (alpha) and regional (beta) species diversities of dung beetles in wet forests in the main tropical regions including Madagascar. Madagascar is exceptional in lacking native large herbivorous mammals which produce the key resource for dung beetles elsewhere. Location Central and South America, mainland Africa, Madagascar and Southeast Asia. Methods Trapping data on dung beetles and data on mammalian faunas were obtained from published and unpublished studies. We used our original data for Madagascar. Results Species richness of dung beetles and that of large‐bodied (> 15 mm length) species in particular were highly significantly explained by the regional number of large‐bodied (> 10 kg) mammals (R² from 50 to 80%). For a given pairwise spatial distance between two communities, beta diversity was significantly higher in Madagascar than elsewhere, explaining the very high total species richness in Madagascar in spite of low local diversity. Main conclusion The presence and numbers of large herbivorous mammals greatly influence the species richness of dung beetles in tropical wet forests. The lack of native large herbivores rather than a limited species pool explains the low local diversity in Madagascar. Exceptionally high beta diversity in Madagascar suggests a pattern of old radiation involving extensive allopatric speciation.     

High gene flow in two thrips‐pollinated South‐East Asian pioneer trees: genetic diversity and population structure of Macaranga hypoleuca and M. beccariana (Euphorbiaceae)

As a result of intensive exploitation, disturbed forests now dominate large areas of lowland tropical rainforest in South‐East Asia. The genus Macaranga comprises some of the most important pioneer tree species of the region, among them M. beccariana and M. hypoleuca, two closely related obligate ant‐plants pollinated by thrips. We used nuclear and plastid DNA markers to address questions of genetic diversity and population structure. Twelve plastid haplotypes were detected among 281 samples, three of which were shared between the two study species. Hybrids between the two species appear to be rare. Overall, genetic diversity in both species was moderate to high, with low levels of population differentiation, consistent with other tropical pioneer trees. Genetic structure was generally more pronounced in plastid than in nuclear data, indicating that gene flow via pollen may be more efficient than via seeds. Thrips apparently also serve as efficient pollinators over long distances, perhaps through a combination of passive dispersal by wind and active search for inflorescences in the target area. Our results indicate that M. beccariana and M. hypoleuca populations from recently disturbed habitats do not yet suffer from reduced genetic diversity or increased inbreeding. © 2013 The Linnean Society of London, Botanical Journal of the Linnean Society, 2013, 173 , 606–621.     

Pollination biology ofOpuntia polyacantha andOpuntia phaeacantha (Cactaceae) in southern Colorado

The phenology, compatibility system, and pollinator fauna ofOpuntia polyacantha Haw. andO. phaeacantha Engelm. in southern Colorado were studied and compared. The total blooming periods overlap, but the peak of blooming differs between the species withO. polyacantha blooming first and for longer. Neither species is apomictic,O. polyacantha is largely self-incompatible andO. phaeacantha is self-compatible but may not automatically fully self-pollinate. Examination of flowering structures showed that the former has larger flowers with more perianth parts and heavier (but fewer) anthers. Further the flowers of the allogamous species are visited by greater diversity (but similar abundance) of pollinators. Medium to large bees of the generaDiadasia, Lithurge, Melissodes, Bombus, Agapostemon andMegachile were found to be effective pollinators of the Opuntias studied. The two species were compared as to their floral attributes, breeding systems, and pollinators, and the possible role of competition for pollinators is discussed with respect to its role in their evolutionary paths.     

Negative Effects of Conspecific Floral Density on Fruit Set of Two Neotropical Understory Plants

Plant reproductive success is usually positively related to conspecific floral density, but neutral or negative effects of floral density on reproduction have also been reported. Differences in the relationship between reproduction and floral density largely originate from a trade‐off between increasing attractiveness versus increasing competition for pollinators at high floral densities. Although floral densities strongly vary in the understory of tropical forests, for instance, due to variation in light availability, little is known about the density dependence of reproduction in tropical understory plants. We used path analyses to disentangle direct and indirect effects of canopy openness and floral density on fruit set and analyzed the relationship between pollen load and floral density for two Neotropical understory plants, Heliconia metallica and Besleria melancholica. In both species, fruit set was not directly related to canopy openness, but decreased with increasing floral density. In H. metallica, canopy openness had an indirect negative effect on reproduction mediated by its effects on floral density. Effects of floral density on pollen loads were species‐specific. In B. melancholica, pollen loads linearly decreased with increasing floral density, indicating competition for pollinators at high densities. In H. metallica, pollen loads were reduced at both low and high densities, indicating an interplay of facilitative and competitive effects of floral density on pollen deposition. In contrast to other studies, we found negative density dependence of reproduction in both understory species. Negative effects of floral density on reproduction appear to be related to pollinator‐mediated effects on reproduction rather than to variation in abiotic conditions.     

First report on generalized pollination systems in Melastomataceae for the Andean páramos

Melastomataceae is a megadiverse family with records of transitions from specialized to generalized pollination systems for several species. These transitions are associated with the colonization of new, unpredictable and/or impoverished pollinator habitats or habitats where specialized pollinators are scarce (e.g., in highland environments). The bee species diversity is low in highlands. Therefore, autonomous breeding systems such as apomixis and self-pollination emerge in these environments. In this paper, we studied the floral traits associated with the generalization of pollination systems and registered the floral visitors of two species in the Colombian Andes: Miconia cataractae and M. elaeoides. We investigated the breeding system of M. elaeoides. Both species presented small flowers, short anthers of medium pore size, and nectar-producing stomata on the base of the anthers. Miconia cataractae produced an average of 1.62 μl nectar/flower, a sugar concentration of 6.78%, whereas M. elaeoides produced 0.09 μl nectar/flower, a sugar concentration of 6.13%. We recorded a wide diversity of pollinators for both species, mainly insects from the orders Hymenoptera and Diptera. Miconia elaeoides presented a mixed breeding system and was also capable of setting fruits by apomixis. We conclude that flower and anther morphology, combined with nectar production, thus represent convergent traits resulting in a generalist pollination system shared by M. cataractae and M. elaeoides. Here, we presented the first generalist pollination system recorded for Miconia (and the Melastomataceae) in the Andes, the first report for a species from the small-pored section Amblyarrhena, and the first report for a species from the large-pored section Cremanium in Colombia.     

Repeated evolution of vertebrate pollination syndromes in a recently diverged Andean plant clade

Although specialized interactions, including those involving plants and their pollinators, are often invoked to explain high species diversity, they are rarely explored at macroevolutionary scales. We investigate the dynamic evolution of hummingbird and bat pollination syndromes in the centropogonid clade (Lobelioideae: Campanulaceae), an Andean‐centered group of ∼550 angiosperm species. We demonstrate that flowers hypothesized to be adapted to different pollinators based on flower color fall into distinct regions of morphospace, and this is validated by morphology of species with known pollinators. This supports the existence of pollination syndromes in the centropogonids, an idea corroborated by ecological studies. We further demonstrate that hummingbird pollination is ancestral, and that bat pollination has evolved ∼13 times independently, with ∼11 reversals. This convergence is associated with correlated evolution of floral traits within selective regimes corresponding to pollination syndrome. Collectively, our results suggest that floral morphological diversity is extremely labile, likely resulting from selection imposed by pollinators. Finally, even though this clade's rapid diversification is partially attributed to their association with vertebrate pollinators, we detect no difference in diversification rates between hummingbird‐ and bat‐pollinated lineages. Our study demonstrates the utility of pollination syndromes as a proxy for ecological relationships in macroevolutionary studies of certain species‐rich clades.