Conflicting constraints on male mating success shape reward size in pollen-rewarding plants.

PREMISE: Pollen-rewarding plants face two conflicting constraints: They must prevent consumptive emasculation while remaining attractive to pollen-collecting visitors. Small pollen packages (the quantity of pollen available in a single visit) may discourage visitors from grooming (reducing consumptive loss) but may also decrease a plant's attractiveness to pollen-collecting visitors. What package size best balances these two constraints? METHODS: We modeled the joint effects of pollinators' grooming behaviors and package size preferences on the optimal package size (i.e., the size that maximizes pollen donation). We then used this model to examine Darwin's conjecture that selection should favor increased pollen production in pollen-rewarding plants. RESULTS: When package size preferences are weak, minimizing package size reduces grooming losses and should be favored (as in previous theoretical studies). Stronger preferences select for larger packages despite the associated increase to grooming loss because loss associated with nonremoval of smaller packages is even greater. Total pollen donation increases with production (as Darwin suggested). However, if floral visitation declines or packages size preference increases with overall pollen availability, the fraction of pollen donated may decline as per-plant pollen production increases. Hence, increasing production may result in diminishing returns. CONCLUSIONS: Pollen-rewarding plants can balance conflicting constraints on pollen donation by producing intermediate-sized pollen packages. Strictly pollen-rewarding plants may have responded to past selection to produce more pollen in total, but diminishing returns may limit the strength of that selection.

Nectar addition changes pollinator behavior but not plant reproduction in pollen-rewarding Lupinus argenteus.

PREMISE: In addition to its role as the male gamete, pollen is often used as a food reward for pollinators. Roughly 20,000 species of angiosperms are strictly pollen-rewarding, providing no other rewards to their pollinators. However, the influence of this strategy on pollinator behavior and plant reproduction is poorly understood, especially relative to the nectar-reward strategy. We performed a field experiment using the strictly pollen-rewarding Lupinus argenteus to explore how the absence of nectar influences pollinator behavior and plant reproduction. METHODS: We added artificial nectar to Lupinus argenteus individuals to simulate a phenotype that would reward pollinators with both nectar and pollen. We compared bee pollinator behavior, via direct observation, and female reproduction between nectar-added and nectarless control plants. RESULTS: Bees exhibited behavioral responses to the novel reward, collecting nectar as well as pollen and spending 27% longer per flower. Pollen transfer increased with flower visit duration. However, plants in the study population were not pollen-limited; consequently, the observed changes in pollinator behavior did not result in changes in female components of plant reproduction. CONCLUSIONS: The addition of nectar to pollen-rewarding plants resulted in modest increases in per-flower pollinator visit duration and pollen transfer, but had no effect on reproduction because, at the place and time the experiment was conducted, plants were not pollen-limited. These results suggest that a pollen-only reward strategy may allow plants that are visited by pollen foragers to minimize some costs of reproduction by eliminating investment in other rewards, such as nectar, without compromising female plant fitness.

Pollen and vegetative secondary chemistry of three pollen-rewarding lupines.

PREMISE: Optimal defense theory predicts that selection should drive plants to disproportionally allocate resources for herbivore defense to tissues with high fitness values. Because pollen's primary role is the transport of gametes, plants may be expected to defend it from herbivory. However, for many animal-pollinated plants, pollen serves a secondary role as a pollinator reward. These dual roles may present a conflict between selection to defend pollen from herbivores and selection to reward pollinators. Here, we investigate whether pollen secondary chemistry in three pollen-rewarding Lupinus species better reflects the need to defend pollen or reward pollinators. METHODS: Lupinus (Fabaceae) species are nectarless, pollen-rewarding, and produce defensive quinolizidine and/or piperidine alkaloids throughout their tissues. We used gas chromatography to identify and quantitate the alkaloids in four aboveground tissues (pollen, flower, leaf, stem) of three western North American lupines, L. argenteus, L. bakeri, and L. sulphureus, and compared alkaloid concentrations and composition among tissues within individuals. RESULTS: In L. argenteus and L. sulphureus, pollen alkaloid concentrations were 11-35% of those found in other tissues. We detected no alkaloids in L. bakeri pollen, though they were present in other tissues. Alkaloid concentrations were not strongly correlated among tissues within individuals. We detected fewer alkaloids in pollen compared to other tissues, and pollen contained no unique alkaloids. CONCLUSIONS: Our results are consistent with the hypothesis that, in these pollen-rewarding species, pollen secondary chemistry may reflect the need to attract and reward pollinators more than the need to defend pollen from herbivory.

Divergent gametic thermal performance and floral warming across an elevation gradient.

Thermal environments vary widely across species ranges, establishing the potential for local adaptation of thermal performance optima and tolerance. In the absence of local adaptation, selection should favor mechanisms to meet thermal optima. Floral temperature is a major determinant of reproductive success in angiosperms, yet whether gametic thermal performance shows signatures of local adaptation across temperature gradients, and how variation in gametic thermal performance influences floral evolution, is unknown. We characterized flowering season temperatures for the forb, Argentina anserina, at extremes of a 1000 m elevation gradient and generated thermal performance curves for pollen and ovule performance in populations at each extreme. Thermal optima fell between mean and maximum intrafloral temperatures. However, cooler high-elevation populations had ~4 °C greater pollen thermal optima than warmer low-elevation populations, while tolerance breadths did not differ. We then tested whether plants at elevational extremes differentially warmed the floral microenvironment. High-elevation flowers warmed significantly more than low, bringing intrafloral temperatures nearer the pollen optima. A manipulative experiment demonstrated that stronger warming in high elevation was conferred by floral tissues. Elevational divergence in floral warming may be driven, in part, by selection on flowers to meet different thermal demands of the gametophytes.

Noisy communities and signal detection: why do foragers visit rewardless flowers?

Floral communities present complex and shifting resource landscapes for flower-foraging animals. Strong similarities among the floral displays of different plant species, paired with high variability in reward distributions across time and space, can weaken correlations between floral signals and reward status. As a result, it should be difficult for foragers to discriminate between rewarding and rewardless flowers. Building on signal detection theory in behavioural ecology, we use hypothetical probability density functions to examine graphically how plant signals pose challenges to forager decision-making. We argue that foraging costs associated with incorrect acceptance of rewardless flowers and incorrect rejection of rewarding ones interact with community-level reward availability to determine the extent to which rewardless and rewarding species should overlap in flowering time. We discuss the evolutionary consequences of these phenomena from both the forager and the plant perspectives. This article is part of the theme issue 'Signal detection theory in recognition systems: from evolving models to experimental tests'.