Pollinator-Driven Floral Evolution in Subalpine Wildflowers

Knowledge Graph (360 nodes, 5804 connections)

Research Primer

Background

Flowers are advertisements, and pollinators are the audience that ultimately decides which plants leave the most descendants. In the subalpine meadows around Gothic, Colorado, wildflowers such as scarlet gilia (Ipomopsis aggregata), larkspurs (Delphinium), and monkshood (Aconitum) compete for visits from hummingbirds, bumble bees, hawkmoths, and flies. When pollinators visit some flower shapes, colors, or scents more than others — and those preferred plants produce more seeds — the result is pollinator-mediated selection, a form of natural selection that shapes floral traits over generations. Understanding this process matters because Gunnison Basin meadows are warming, snowmelt is advancing, and the partnerships between wildflowers and their visitors are being rearranged in real time.

A few key ideas help make sense of the findings that follow. Pollination syndromes are the classical notion that flower traits (red tubular corollas for hummingbirds, sweet-scented white flowers for moths) reliably predict a flower's main pollinator. Floral volatile organic compounds are the scent chemicals flowers release, which attract pollinators but can also attract seed-eating flies. Nectar robbing happens when bees bite holes in the side of a flower to steal nectar without contacting the reproductive parts. Hybridization, the interbreeding of related species, creates hybrid zones — geographic strips where, for example, scarlet gilia (I. aggregata) and the paler I. tenuituba meet and cross. A cline is a gradient in traits across such a zone. Finally, evolutionary rescue refers to the possibility that natural selection might change a population's traits fast enough to prevent extinction as the climate shifts.

Research in this area links behavior (which pollinator visits which flower, and why), ecology (how visitation translates into seed production), and evolution (how trait variation, heritability, and selection together change populations across generations). Long-term study sites at RMBL, where some experiments now span more than 40 years, make the Gunnison Basin one of the best places in the world to watch these processes unfold.

Foundational work

Early RMBL studies established that pollinator behavior shapes both ecological communities and the evolution of floral form. Inouye (Inouye, 1978) showed experimentally that two bumble bee species in Gothic partition flowers according to how well their tongue lengths match corolla tubes, and that removing one species causes the other to expand its diet — direct evidence that pollinator competition structures plant–pollinator communities. In the same year, Pyke (Pyke, 1978) formalized bumble bee movement as optimal foraging, while Waser (Waser, 1978) demonstrated that Delphinium nelsoni and Ipomopsis aggregata flower in sequence to avoid wasteful pollen transfer between species sharing hummingbird pollinators. Price and Waser (Price & Waser, 1979) added a population-genetic dimension by showing that Delphinium plants achieve highest seed quality when crossed at intermediate distances — outbreeding depression operates at surprisingly fine scales.

A second wave of foundational work clarified how flowers control their own reproductive fates. Harder and Thomson (Harder & Thomson, 1989) showed that bees deposit only a tiny fraction of the pollen they remove, favoring floral designs that ration pollen across many visits. Inouye (Inouye, 1980) provided the vocabulary of floral larceny still used today, and Campbell and Halama (Campbell & Halama, 1993) demonstrated experimentally that seed production in Ipomopsis is jointly limited by pollen receipt and resources. Waser et al. (Waser et al., 1996) then challenged the long-standing idea of tidy pollination syndromes, arguing that most plants are visited by a generalized set of animals — a view later supported by a global test (Ollerton et al., 2009).

Key findings

A central insight from decades of RMBL research is that pollinator identity and behavior translate directly into selection on flowers. Different visitors are not interchangeable: pollen-collecting bees remove far more pollen than nectar-collecting bumble bees but deposit an order of magnitude less on stigmas (Thomson, 1991), and even small experimental removals of a single bumble bee species reduce floral fidelity in remaining pollinators and lower plant seed set (Brosi & Briggs, 2013). Hummingbirds dominate visitation to scarlet gilia, contributing roughly 94% of pollinator visits in some years (Mayfield et al., 2005), and their visitation rate is the single strongest predictor of how much pollen lands on stigmas (Mitchell et al., 2003).

Nectar robbing, once viewed as straightforward cheating, turns out to be more nuanced. Robbers often pollinate while stealing, and their net effects can be positive, neutral, or negative depending on context (Maloof & Inouye, 2000). Robbing rates vary from 0% to 100% across sites and years (Irwin, 2002), and consequences depend on plant mating system: pollen-limited, self-incompatible plants suffer most (Irwin, 2007). In Corydalis caseana, robbing is essentially commensal despite halving nectar rewards (Bronstein et al., 2018), and bees often rob the same flower multiple times because middle-aged flowers refill and remain profitable (Bronstein et al., 2024).

Hybridization between Ipomopsis aggregata and I. tenuituba has emerged as a natural experiment in floral evolution. Hybrids often show higher water-use efficiency than either parent (Campbell et al., 2005), and hybrid traits respond differently to precipitation than parental species do (Campbell & Wendlandt, 2013). Across a hybrid zone, corolla length has shifted measurably over five generations in a direction matching predicted selection (Campbell & Powers, 2018), providing rare direct evidence of contemporary floral evolution. Meanwhile, pollination syndromes have been weakened as a general predictive tool: only about 1% of species cluster cleanly into a syndrome, and predicted pollinators are correct less than a third of the time (Ollerton et al., 2009).

Current frontier

Research since 2020 has shifted decisively toward climate change as the dominant force reshaping floral evolution. Long-term snowmelt manipulations show that natural selection on floral volatile emissions changes with snowmelt timing and summer precipitation, with the dominant scent compound alpha-pinene responding strongly to advanced snowmelt (Powers et al., 2025). Experimental warming in open-top chambers boosts nectar production in scarlet gilia by 41% and alters daytime floral scent, while also reducing seed-predator fly oviposition by 72% (Wu et al., 2025). Plasticity in leaf traits is also under selection: a 50% precipitation cut reduces stomatal conductance and increases water-use efficiency (Campbell et al., 2022).

The central evolutionary question is now whether selection can keep pace with environmental change. Campbell et al. (Campbell et al., 2025) used long-term data to model evolutionary rescue in Ipomopsis and found that without evolution or plasticity, populations face local extinction under earlier snowmelt; rescue is feasible in only one of two populations studied, and only when strong selection, heritability, and plasticity in specific leaf area act together. A broader synthesis across herbaceous model species reaches similar conclusions, documenting climate-driven maladaptation and population decline at range edges (Kooyers et al., 2025). Meanwhile, drought reshapes interaction networks: drought years bring 4.3-fold higher visitor abundance to scarlet gilia and broader, more connected plant–pollinator networks (Brosi et al., 2021) (Morozumi et al., 2022). New technologies — quantum-dot pollen tracking, microbial community sampling, automated camera traps for hummingbirds — are letting researchers ask sharper questions about pollen flow, floral microbiomes, and phenological mismatch.

Open questions

The most pressing unknown is whether contemporary evolution in floral traits can rescue subalpine wildflower populations from accelerating snowmelt and summer drought, or whether plasticity and dispersal will fall short. Closely related questions include how multi-trait selection — on scent, color, nectar, and leaf physiology simultaneously — resolves trade-offs between attracting pollinators and deterring seed predators; how hybrid zones will migrate or dissolve as climate shifts each parent species' optimal habitat; what roles floral microbes and nectar chemistry play in mediating pollinator choice and reproductive isolation; and whether phenological mismatches between Broad-tailed Hummingbirds and their nectar plants will become severe enough to threaten the pollinators themselves. Answering these will require coupling RMBL's long-term observational records with experimental manipulations, genomic tools, and demographic models that integrate behavior, physiology, and evolution.

References

Bronstein, J.L. et al. (2018). Why are some plant-nectar robber interactions commensalisms? →

Bronstein, J.L. et al. (2024). The hole truth: why do bumble bees rob flowers more than once? Plants. →

Brosi, B.J. et al. (2021). Plant–pollinator interaction niche broadens in response to severe drought perturbations. →

Brosi, B.J. et al. (2022). Simultaneous niche expansion and contraction in plant–pollinator networks under drought. →

Brosi, B.J., Briggs, H.M. (2013). Single pollinator species losses reduce floral fidelity and plant reproductive function. PNAS. →

Campbell, D.R. et al. (2005). Ecophysiology of first and second generation hybrids in a natural plant hybrid zone. →

Campbell, D.R. et al. (2022). Phenotypic plasticity and selection on leaf traits in response to snowmelt timing and summer precipitation. →

Campbell, D.R. et al. (2025). Predicting the contribution of single trait evolution to rescuing a plant population from demographic impacts of climate change. Evolution Letters. →

Campbell, D.R., Halama, K.J. (1993). Resource and pollen limitations to lifetime seed production in a natural plant population. Ecology. →

Campbell, D.R., Powers, J.M. (2018). Clines in traits compared over two decades in a plant hybrid zone. →

Campbell, D.R., Wendlandt, C. (2013). Altered precipitation affects plant hybrids differently than their parental species. →

Harder, L.D., Thomson, J.D. (1989). Evolutionary options for maximizing pollen dispersal of animal-pollinated plants. American Naturalist. →

Inouye, D.W. (1978). Resource partitioning in bumblebees: experimental studies of foraging behavior. Ecology. →

Inouye, D.W. (1980). The terminology of floral larceny. Ecology. →

Irwin, R.E. (2002). Variation in nectar robbing over time, space, and species. →

Irwin, R.E. (2007). Predicting the effects of nectar robbing on plant reproduction. →

Kooyers, N.J. et al. (2025). Responses to climate change – insights and limitations from herbaceous plant model species. New Phytologist. →

Maloof, J.E., Inouye, D.W. (2000). Are nectar robbers cheaters or mutualists? Ecology. →

Mayfield, M.M. et al. (2005). Temporal and spatial variation in pollination of a montane herb: a seven-year study. →

Mitchell, R.J. et al. (2003). Linking pollinator visitation rate and pollen receipt. →

Ollerton, J. et al. (2009). A global test of the pollination syndrome hypothesis. Annals of Botany. →

Powers, J.M. et al. (2025). Natural selection on floral volatiles and other traits can change with snowmelt timing and summer precipitation. New Phytologist. →

Price, M.V., Waser, N.M. (1979). Pollen dispersal and optimal outcrossing in Delphinium nelsoni. Nature. →

Pyke, G.H. (1978). Optimal foraging: movement patterns of bumblebees between inflorescences. Theoretical Population Biology. →

Thomson, J.D. et al. (1991). Heterogeneity among floral visitors leads to discordance between removal and deposition of pollen. →

Waser, N.M. (1978). Competition for hummingbird pollination and sequential flowering in two Colorado wildflowers. Ecology. →

Waser, N.M. et al. (1996). Generalization in pollination systems, and why it matters. Ecology. →

Wu, B.S. et al. (2025). Effects of experimental warming on floral scent, display and rewards in two subalpine herbs. Annals of Botany. →

Species (41) →

Concept (35) →

Protocol (27) →

Author (91) →

Publication (459) →

Dataset (18) →