Evolutionary convergence on hummingbird pollination in Neotropical Costus provides insight into the causes of pollinator shifts.

The evolution of hummingbird pollination is common across angiosperms throughout the Americas, presenting an opportunity to examine convergence in both traits and environments to better understand how complex phenotypes arise. Here we examine independent shifts from bee to hummingbird pollination in the Neotropical spiral gingers (Costus) and address common explanations for the prevalence of transitions from bee to hummingbird pollination. We use floral traits of species with observed pollinators to predict pollinators of unobserved species and reconstruct ancestral pollination states on a well-resolved phylogeny. We examine whether independent transitions evolve towards the same phenotypic optimum and whether shifts to hummingbird pollination correlate with elevation or climate. Traits predicting hummingbird pollination include small flower size, brightly colored floral bracts and the absence of nectar guides. We find many shifts to hummingbird pollination and no reversals, a single shared phenotypic optimum across hummingbird flowers, and no association between pollination and elevation or climate. Evolutionary shifts to hummingbird pollination in Costus are highly convergent and directional, involve a surprising set of traits when compared with other plants with analogous transitions and refute the generality of several common explanations for the prevalence of transitions from bee to hummingbird pollination.

Phylogenomic analyses in Phrymaceae reveal extensive gene tree discordance in relationships among major clades.

PREMISE: Phylogenomic datasets using genomes and transcriptomes provide rich opportunities beyond resolving bifurcating phylogenetic relationships. Monkeyflower (Phrymaceae) is a model system for evolutionary ecology. However, it lacks a well-supported phylogeny as a basis for a stable taxonomy and for macroevolutionary comparisons. METHODS: We sampled 24 genomes and transcriptomes in Phrymaceae and closely related families, including eight newly sequenced transcriptomes. We reconstructed the phylogeny using IQ-TREE and ASTRAL, evaluated gene tree discordance using PhyParts, Quartet Sampling, and a cloudogram, and carried out reticulation analyses using PhyloNet and HyDe. We searched for whole genome duplication (WGD) events using chromosome numbers, synonymous distances, and gene duplication events as evidence. RESULTS: Most gene trees support the monophyly of Phrymaceae and each of its tribes. Most gene trees also support tribe Mimuleae being sister to Phrymeae + Diplaceae + Leucocarpeae, with extensive gene tree discordance among the latter three. Despite the discordance, the monophyly of Mimulus s.l. is rejected, and no individual reticulation event among the Phrymaceae tribes is well-supported. Reticulation likely occurred among Erythranthe bicolor and closely related species. No ancient WGD was detected in Phrymaceae. Instead, small-scale duplications are among potential drivers of macroevolutionary diversification of Phrymaceae. CONCLUSIONS: We show that analysis of reticulate evolution is sensitive to taxon sampling and methods used. We also demonstrate that phylogenomic datasets using genomes and transcriptomes present rich opportunities to investigate gene family evolution and genome duplication events involved in lineage diversification and adaptation.

Abiotic Environment Predicts Micro- but Not Macroevolutionary Patterns of Flower Color in Monkeyflowers (Phrymaceae).

Anthocyanin pigments are responsible for many of the vivid pink, purple, red, and blue flower colors across angiosperms and frequently vary within and between closely related species. While anthocyanins are well known to influence pollinator attraction, they are also associated with tolerance to abiotic stressors such as extreme temperatures, reduced precipitation, and ultraviolet radiation. Using a comparative approach, we tested whether abiotic variables predict floral anthocyanin in monkeyflowers (Phrymaceae) across western North America. Within two polymorphic species, we found that abiotic variables predicted flower color across their geographic ranges. In Erythranthe discolor, the frequency of pink flowered (anthocyanin producing) individuals was greater in populations with reduced precipitation. In Diplacus mephiticus, the frequency of pink flowered individuals was greater at higher elevations that had reduced precipitation and lower temperatures but less ultraviolet radiation. At the macroevolutionary scale, across two parallel radiations of North American monkeyflowers, species with floral anthocyanins (pink, purple, or red corollas) occupied areas with reduced precipitation in Erythranthe but not Diplacus. However, after accounting for phylogenetic relatedness, we found no evidence for the joint evolution of flower color and environmental affinity in either clade. We conclude that although abiotic stressors may play a role in the evolution of flower color within polymorphic species, we found no evidence that these processes lead to macroevolutionary patterns across monkeyflowers.

Testing an invasion mechanism for Eucalyptus globulus: Is there evidence of allelopathy?

PREMISE: Sparse understory communities, in association with non-native tree species, are often attributed to allelopathy, the chemical inhibition of a plant by another. However, allelopathy is a difficult ecological phenomenon to demonstrate as many studies show conflicting results. Eucalyptus globulus, a tree native to Australia, is one of the most widely planted trees around the world. Sparse understories are common beneath E. globulus plantations and are often attributed to allelopathy, but the ecological impacts of E. globulus on native plant communities outside Austrialia are poorly understood. METHODS: To assess allelopathy as a mechanism of understory inhibition, we tested volatile- and water-soluble leaf extracts from E. globulus, Salvia apiana, and Quercus agrifolia on seed germination of California native plants. We also quantified germination rates and early seedling growth of California native plants grown in soil from E. globulus plantations versus soil from an adjacent native plant community. RESULTS: Volatile compounds from E. globulus did not significantly reduce germination for any species. Inhibition from water-soluble E. globulus compounds was comparable to that of a native tree, Quercus agrifolia (10%). Eucalyptus globulus soil supported germination and early seedling growth of native species equal to or better than coastal scrub soil, although species responses were variable. CONCLUSIONS: In contrast to previous studies, our results fail to support the hypothesis that E. globulus chemically inhibits germination of native species. California native plants germinate and grow well in soils from E. globulus plantations, which may have significant implications for management and restoration of land historically occupied by E. globulus plantations.

Patterns of speciation are similar across mountainous and lowland regions for a Neotropical plant radiation (Costaceae: Costus).

High species richness and endemism in tropical mountains are recognized as major contributors to the latitudinal diversity gradient. The processes underlying mountain speciation, however, are largely untested. The prevalence of steep ecogeographic gradients and the geographic isolation of populations by topographic features are predicted to promote speciation in mountains. We evaluate these processes in a species-rich Neotropical genus of understory herbs that range from the lowlands to montane forests and have higher species richness in topographically complex regions. We ask whether climatic niche divergence, geographic isolation, and pollination shifts differ between mountain-influenced and lowland Amazonian sister pairs inferred from a 756-gene phylogeny. Neotropical Costus ancestors diverged in Central America during a period of mountain formation in the last 3 million years with later colonization of Amazonia. Although climatic divergence, geographic isolation, and pollination shifts are prevalent in general, these factors do not differ between mountain-influenced and Amazonian sister pairs. Despite higher climatic niche and species diversity in the mountains, speciation modes in Costus appear similar across regions. Thus, greater species richness in tropical mountains may reflect differences in colonization history, diversification rates, or the prevalence of rapidly evolving plant life forms, rather than differences in speciation mode.

The effect of range overlap on ecological niche divergence depends on spatial scale in monkeyflowers.

Patterns of niche divergence and geographical range overlap of closely related species provide insights into the evolutionary dynamics of ecological niches. When ranges overlap, shared selective pressures may preserve niche similarity along coarse-scale macrohabitat axes (e.g., bioclimates). Alternatively, competitive interactions may drive greater divergence along local-scale microhabitat axes (e.g., micro-topographical features). We tested these hypotheses in 16 species pairs of western North American monkeyflowers (Erythranthe and Diplacus, formerly Mimulus) with estimations of species' niches, geographic ranges, and a robust phylogeny. We found that macrohabitat niche divergence decreased with increasing range overlap, consistent with convergent selection operating at a coarse scale. No significant relationship was detected for microhabitat niches. Additionally, niche divergence was greater for recently diverged pairs along all macrohabitat niche axes, but greater for distantly diverged pairs along one microhabitat axis related to vegetation cover. For species pairs with partially overlapping ranges, greater microhabitat divergence was detected in sympatry than in allopatry for at least one niche axis for three of four pairs, consistent with character displacement in sympatry. Thus, coarse- and local-scale niche divergence show dissimilar patterns in relation to range overlap and divergence time, perhaps because the relative importance of convergent versus divergent selection depends on spatial scale.

A geographic cline in the ability to self-fertilize is unrelated to the pollination environment.

The reproductive assurance (RA) hypothesis predicts that the ability to autonomously self-fertilize should be favored in environments where a lack of mates or pollinators limits outcross reproduction. Because such limits to outcrossing are predicted to be most severe at range edges, elevated autonomy in peripheral populations is often attributed to RA. We test this hypothesis in 24 populations spanning the range of Campanula americana, including sampling at the range interior and three geographic range edges. We scored autonomous fruit set in a pollinator-free environment and detected clinal variation-autonomy increased linearly from the southern to the northern edge, and from the eastern to the western edge. We then address whether the cline reflects the contemporary pollination environment. We measured population size, plant density, pollinator visitation, outcross pollen limitation and RA in natural populations over two years. Most populations were pollen limited, and those that experienced higher visitation rates by bumblebees had reduced pollen limitation. Reproductive assurance, however, was generally low across populations and was unrelated to pollen limitation or autonomy. Neither pollen limitation nor RA displayed geographic clines. Finally, autonomy was not associated with pollinator visitation rates or mate availability. Thus, the data do not support the RA hypothesis; clinal variation in autonomy is unrelated to the current pollination environment. Therefore, geographic patterns of autonomy are likely the result of historical processes rather than contemporary natural selection for RA.

Self-compatibility is over-represented on islands.

Because establishing a new population often depends critically on finding mates, individuals capable of uniparental reproduction may have a colonization advantage. Accordingly, there should be an over-representation of colonizing species in which individuals can reproduce without a mate, particularly in isolated locales such as oceanic islands. Despite the intuitive appeal of this colonization filter hypothesis (known as Baker's law), more than six decades of analyses have yielded mixed findings. We assembled a dataset of island and mainland plant breeding systems, focusing on the presence or absence of self-incompatibility. Because this trait enforces outcrossing and is unlikely to re-evolve on short timescales if it is lost, breeding system is especially likely to reflect the colonization filter. We found significantly more self-compatible species on islands than mainlands across a sample of > 1500 species from three widely distributed flowering plant families (Asteraceae, Brassicaceae and Solanaceae). Overall, 66% of island species were self-compatible, compared with 41% of mainland species. Our results demonstrate that the presence or absence of self-incompatibility has strong explanatory power for plant geographical patterns. Island floras around the world thus reflect the role of a key reproductive trait in filtering potential colonizing species in these three plant families.

No association between plant mating system and geographic range overlap.

PREMISE OF THE STUDY: Automatic self-fertilization may influence the geography of speciation, promote reproductive isolation between incipient species, and lead to ecological differentiation. As such, selfing taxa are predicted to co-occur more often with their closest relatives than are outcrossing taxa. Despite suggestions that this pattern may be general, the extent to which mating system influences range overlap in close relatives has not been tested formally across a diverse group of plant species pairs. METHODS: We tested for a difference in range overlap between species pairs for which zero, one, or both species are selfers, using data from 98 sister species pairs in 20 genera across 15 flowering plant families. We also used divergence time estimates from time-calibrated phylogenies to ask how range overlap changes with divergence time and whether this effect depends on mating system. KEY RESULTS: We found no evidence that automatic self-fertilization influenced range overlap of closely related plant species. Sister pairs with more recent divergence times had modestly greater range overlap, but this effect did not depend on mating system. CONCLUSIONS: The absence of a strong influence of mating system on range overlap suggests that mating system plays a minor or inconsistent role compared with many other mechanisms potentially influencing the co-occurrence of close relatives.

The scope of Baker's law.

Baker's law refers to the tendency for species that establish on islands by long-distance dispersal to show an increased capacity for self-fertilization because of the advantage of self-compatibility when colonizing new habitat. Despite its intuitive appeal and broad empirical support, it has received substantial criticism over the years since it was proclaimed in the 1950s, not least because it seemed to be contradicted by the high frequency of dioecy on islands. Recent theoretical work has again questioned the generality and scope of Baker's law. Here, we attempt to discern where the idea is useful to apply and where it is not. We conclude that several of the perceived problems with Baker's law fall away when a narrower perspective is adopted on how it should be circumscribed. We emphasize that Baker's law should be read in terms of an enrichment of a capacity for uniparental reproduction in colonizing situations, rather than of high selfing rates. We suggest that Baker's law might be tested in four different contexts, which set the breadth of its scope: the colonization of oceanic islands, metapopulation dynamics with recurrent colonization, range expansions with recurrent colonization, and colonization through species invasions.

Geographic range size is predicted by plant mating system.

Species' geographic ranges vary enormously, and even closest relatives may differ in range size by several orders of magnitude. With data from hundreds of species spanning 20 genera in 15 families, we show that plant species that autonomously reproduce via self-pollination consistently have larger geographic ranges than their close relatives that generally require two parents for reproduction. Further analyses strongly implicate autonomous self-fertilisation in causing this relationship, as it is not driven by traits such as polyploidy or annual life history whose evolution is sometimes correlated with selfing. Furthermore, we find that selfers occur at higher maximum latitudes and that disparity in range size between selfers and outcrossers increases with time since their evolutionary divergence. Together, these results show that autonomous reproduction--a critical biological trait that eliminates mate limitation and thus potentially increases the probability of establishment--increases range size.

Pollinator-mediated competition influences selection for flower-color displacement in sympatric monkeyflowers.

PREMISE OF STUDY: When coflowering plant species share pollinators, pollinator-mediated competition may favor divergent floral characters associated with pollinator attraction. One potential outcome of this process is that sympatric populations will display increased divergence in floral traits compared with allopatric populations. We developed a new system to study the pattern and process of character displacement. In the central Sierra Nevada of California, USA, Mimulus bicolor is a spring wildflower with two flower-color morphs, one of which resembles coflowering M. guttatus. METHODS: We documented a fine-scale geographic pattern of character displacement in sympatric and allopatric patches and, using experimental arrays, measured seed set in M. bicolor color morphs in the presence versus absence of M. guttatus. KEY RESULTS: In sympatric arrays yellow, guttatus-like M. bicolor morphs had lower relative fitness (0.35 ± 0.05) and reduced conspecific pollen deposition compared with the distinct alternative morph, whereas in allopatric arrays yellow, guttatus-like morphs were occasionally strongly favored. CONCLUSIONS: Pollinator-mediated competition with M. guttatus is consistent with ecological character displacement in M. bicolor and likely contributes to a geographic pattern of character displacement.

Niche and range size patterns suggest that speciation begins in small, ecologically diverged populations in North American monkeyflowers (Mimulus spp.).

Closely related species (e.g., sister taxa) often occupy very different ecological niches and can exhibit large differences in geographic distributions despite their shared evolutionary history. Budding speciation is one process that may partially explain how differences in niche and distribution characteristics may rapidly evolve. Budding speciation is the process through which new species form as initially small colonizing populations that acquire reproductive isolation. This mode of species formation predicts that, at the time of speciation, sister species should have highly asymmetrical distributions. We tested this hypothesis in North American monkeyflowers, a diverse clade with a robust phylogeny, using data on geographical ranges, climate, and plant community attributes. We found that recently diverged sister pairs have highly asymmetrical ranges and niche breadths, relative to older sister pairs. Additionally, we found that sister species occupy distinct environmental niche positions, and that 80% of sister species have completely or partially overlapping distributions (i.e., are broadly sympatric). Together, these results suggest that budding speciation has occurred frequently in Mimulus, that it has likely taken place both inside the range and on the range periphery, and that observed divergences in habitat and resource use could be associated with speciation in small populations.

Increased floral divergence in sympatric monkeyflowers.

Sympatric sister species are predicted to have greater divergence in reproductive traits than allopatric sister species, especially if mating system shifts, such as the evolution of self-fertilization, are more likely to originate within the geographic range of the outcrossing ancestor. We present evidence that supports this expectation-sympatric sister species in the monkeyflower genus, Mimulus, exhibit greater divergence in flower size than allopatric sister species. Additionally, we find that sympatric sister species are more likely to have one species with anthers that overtop their stigmas than allopatric sister species, suggesting that the evolution of automatic self-pollination may contribute to this pattern. Potential mechanisms underlying this pattern include reinforcement and a stepping stone model of parapatric speciation.