Negative impacts of a vertebrate predator on insect pollinator visitation and seed output in<Emphasis Type="Italic"> Chuquiraga oppositifolia</Emphasis>, a high Andean shrub

Studies on plant-pollinator interactions have largely neglected the potential negative effects of the predators of pollinators on seed output, even though anti-predatory behaviour of pollinators may affect visitation patterns, pollen transfer, and therefore potentially, plant reproductive output. We tested the hypothesis that the presence of lizards and insectivorous birds, by reducing pollinator visitation, can have significant negative effects on seed output in the insect-pollinated, genetically self-incompatible lower alpine Andean shrub, Chuquiraga oppositifolia (Asteraceae). The lower alpine belt supports a high density of territorial Liolaemus (Tropiduridae) lizards and low shrubs interspersed among rocks of varying sizes, the latter inhabited by lizards and commonly used by flycatchers Muscisaxicola (Tyrannidae) as perching sites. In a 2×2 factorial predator-exclusion experiment, visitation rates of the most frequent pollinators of C. oppositifolia (the satyrid butterfly Cosmosatyrus chilensis and the syrphid fly Scaeva melanostoma), the duration of pollinator visits, and seed output, were 2–4 times greater when lizards were excluded, while birds had no effect. In a natural experiment, visits by S. melanostoma were 9 times shorter, and pollinator visitation rates of C. chilensis and S. melanostoma, and C. oppositifolia seed output were 2–3 times lower on shrubs growing adjacent to lizard-occupied rocks compared to those growing distant from rocks. Our results, verified for additional Andean sites, suggest that lizard predators can alter the behaviour of pollinators and elicit strong top-down indirect negative effects on seed output. Such effects may be especially important in high alpine plant communities, where pollinator activity can be low and erratic, and pollen limitation has been reported.     

Relative impact of mate versus pollinator availability on pollen limitation and outcrossing rates in a mass‐flowering species

Plant mating systems are driven by several pre‐pollination factors, including pollinator availability, mate availability and reproductive traits. We investigated the relative contributions of these factors to pollination and to realized outcrossing rates in the patchily distributed mass‐flowering shrub Rhododendron ferrugineum. We jointly monitored pollen limitation (comparing seed set from intact and pollen‐supplemented flowers), reproductive traits (herkogamy, flower size and autofertility) and mating patterns (progeny array analysis) in 28 natural patches varying in the level of pollinator availability (flower visitation rates) and of mate availability (patch floral display estimated as the total number of inflorescences per patch). Our results showed that patch floral display was the strongest determinant of pollination and of the realized outcrossing rates in this mass‐flowering species. We found an increase in pollen limitation and in outcrossing rates with increasing patch floral display. Reproductive traits were not significantly related to patch floral display, while autofertility was negatively correlated to outcrossing rates. These findings suggest that mate limitation, arising from high flower visitation rates in small plant patches, resulted in low pollen limitation and high selfing rates, while pollinator limitation, arising from low flower visitation rates in large plant patches, resulted in higher pollen limitation and outcrossing rates. Pollinator‐mediated selfing and geitonogamy likely alleviates pollen limitation in the case of reduced mate availability, while reduced pollinator availability (intraspecific competition for pollinator services) may result in the maintenance of high outcrossing rates despite reduced seed production.     

No evidence for a role of pollinator discrimination in causing selection on flower size through female reproduction

The preference for certain floral phenotypes by flower visiting animals may fuel the evolution of floral traits because variation in flower visitation rates may lead to fitness variation within a population. Here, I examine the importance of flower size for pollinator visitation rate, seed set, and seed mass in two alpine populations of the insect-pollinated herb Ranunculus acris L. during two seasons. There was no pollen limitation of seed set or mass. Pollinators discriminated strongly against flowers experimentally reduced in size. Despite this, there were no signs of any significant impact of flower size on female reproductive success. The results show that although pollinators discriminate strongly among floral phenotypes, this may not always result in female fitness differences within a population because seed set or mass is not limited by pollen availability alone. Probably abiotic environmental constraints prevent plants with high pollinator visitation from capitalizing on the high pollen deposition.     

Nutrient addition extends flowering display, which gets tracked by seed predators, but not by their parasitoids

Although phenological matching between two and three trophic interactions has received some attention, it has largely been disregarded in explaining the lack of strong cascade dynamics in terrestrial systems. We studied the response of the specialist seed predator, Paroxyna plantaginis (Tephritidae) and associated generalist parasitoids (Chalcidoidea) to controlled fertilisation of individuals of naturally growing Tripolium vulgare (Asteraceae) on four island populations (Skeppsvik Archipelago, Sweden). We consistently found evidence of nutrient limitation: fertilised plants increased their biomass, produced more capitula (the oviposition units for tephritid flies), were more at risk of attack by the tephritids, and puparia were heavier in fertilised plants. During some parts of the season tephritids became more heavily parasitized, supporting the presence of cascade dynamics, however net parasitism over season decreased in response to nutrient addition. We found no evidence that capitulum size complicated parasitoid access to the tephritids, however the extended bud production prolonged the flowering season. Thus, tephritids utilized the surplus production of capitula throughout the entire season, while parasitoids did not expand their oviposition time window accordingly. Implications for top down regulation and cascade dynamics in the system are discussed.     

Extremely long-lived stigmas allow extended cross-pollination opportunities in a high Andean plant

High-elevation ecosystems are traditionally viewed as environments in which predominantly autogamous breeding systems should be selected because of the limited pollinator availability. Chaetanthera renifolia (Asteraceae) is an endemic monocarpic triennial herb restricted to a narrow altitudinal range within the high Andes of central Chile (3300-3500 m a.s.l.), just below the vegetation limit. This species displays one of the larger capitulum within the genus. Under the reproductive assurance hypothesis, and considering its short longevity (monocarpic triennial), an autogamous breeding system and low levels of pollen limitation would be predicted for C. renifolia. In contrast, considering its large floral size, a xenogamous breeding system, and significant levels of pollen limitation could be expected. In addition, the increased pollination probability hypothesis predicts prolonged stigma longevity for high alpine plants. We tested these alternative predictions by performing experimental crossings in the field to establish the breeding system and to measure the magnitude of pollen limitation in two populations of C. renifolia. In addition, we measured the stigma longevity in unpollinated and open pollinated capitula, and pollinator visitation rates in the field. We found low levels of self-compatibility and significant levels of pollen limitation in C. renifolia. Pollinator visitation rates were moderate (0.047-0.079 visits per capitulum per 30 min). Although pollinator visitation rate significantly differed between populations, they were not translated into differences in achene output. Finally, C. renifolia stigma longevity of unpollinated plants was extremely long and significantly higher than that of open pollinated plants (26.3±2.8 days vs. 10.1±2.2, respectively), which gives support to the increased pollination probability hypothesis for high-elevation flowering plants. Our results add to a growing number of studies that show that xenogamous breeding systems and mechanisms to increase pollination opportunities can be selected in high-elevation ecosystems.     

Factors affecting pollinator movement and plant fitness in a specialized pollination system

The rate of pollen exchange within and among flowers may depend on pollinator attraction traits such as floral display size and flowering plant density. Variations in these traits may influence pollinator movements, pollen receipt, and seed number. To assess how floral display size and flowering plant density affect parameters of pollinator visitation rate, pollen receipt per flower, seed number per fruit and the between-plant pollinator movements, we studied the self-incompatible plant, Nierembergia linariifolia. Per-flower pollinator visitation rate and bout length increased linearly with increasing floral display size. Pollen receipt per flower increased linearly with increasing flowering plant density. For seed number per fruit, a polynomial model describing an increased seed number per fruit at low density and a decreased seed number per fruit at high density provided a significant fit. Per-flower pollinator visitation rate was not associated with pollen receipt per flower and seed number per fruit. Bees visited plants located near to the center of the population more frequently than plants located at the periphery. Increases in both floral display size and flowering plant density led to an increased chance of a plant being chosen as the center of the pollinator foraging area. These results suggest that even though large floral displays and high flowering plant density are traits that attract more pollinators, they may also reduce potential mate diversity by restricting pollen movement to conspecific mates that are closely located.     

Within-population spatial variation in pollinator visitation rates,pollen limitation on seed set,and flower longevity in an alpine species

Pollen limitation through insufficient pollen deposition on stigmas caused by too infrequent pollinator visitation may influence the reproductive outcome of plants. In this study we investigated how pollinator visitation rate, the degree of pollen limitation, and flower longevity varied spatially among three sites at different altitudes within a population of the dwarf shrub Dryas octopetala L. in alpine southern Norway. Significant pollen limitation on seed set only occurred at the mid-elevation site, while seed set at the other sites appeared to be mainly resource limited, thus indicating a spatial variation in pollen limitation. There was no association between the spatial variation in the extent of pollen limitation and pollinator visitation rate to flowers. However, pollinator visitation rates were related to flower longevity of Dryas; sites with low visitation rates had long-lived flowers and vice versa. Thus, our results suggest within-population spatial co-variation between pollinator visitation rates, pollen limitation, and a developmental response to these factors, flower longevity.     

Within-population spatial variation in pollinator visitation rates, pollen limitation on seed set, and flower longevity in an alpine species

Pollen limitation through insufficient pollen deposition on stigmas caused by too infrequent pollinator visitation may influence the reproductive outcome of plants. In this study we investigated how pollinator visitation rate, the degree of pollen limitation, and flower longevity varied spatially among three sites at different altitudes within a population of the dwarf shrub Dryas octopetala L. in alpine southern Norway. Significant pollen limitation on seed set only occurred at the mid-elevation site, while seed set at the other sites appeared to be mainly resource limited, thus indicating a spatial variation in pollen limitation. There was no association between the spatial variation in the extent of pollen limitation and pollinator visitation rate to flowers. However, pollinator visitation rates were related to flower longevity of Dryas; sites with low visitation rates had long-lived flowers and vice versa. Thus, our results suggest within-population spatial co-variation between pollinator visitation rates, pollen limitation, and a developmental response to these factors, flower longevity.     

Pollinator response to flower color polymorphism and floral display in a plant with a single-locus floral color polymorphism: consequences for plant reproduction

Variation in flower color, particularly polymorphism, in which two or more different flower color phenotypes occur in the same population or species, may be affected or maintained by mechanisms that depend on pollinators. Furthermore, variation in floral display may affect pollinator response and plant reproductive success through changes in pollinator visitation and availability of compatible pollen. To asses if flower color polymorphism and floral display influences pollinator preferences and movements within and among plants and fitness-related variables we used the self-incompatible species Cosmos bipinnatus Cav. (Asteraceae), a model system with single-locus flower color polymorphism that comprises three morphs: white (recessive homozygous), pink (heterozygous co-dominate), and purple (dominant homozygous) flowers. We measured the preferences of pollinators for each morph and constancy index for each pollinator species, pollination visitation rate, floral traits, and female fitness measures. Flower color morphs differed in floral trait measures and seed production. Pollinators foraged nonrandomly with respect to flower color. The most frequent morph, the pink morph, was the most visited and pollinators exhibited the highest constancy for this morph. Moreover, this morph exhibited the highest female fitness. Pollinators responded strongly to floral display size, while probed more capitulums from plants with large total display sizes, they left a great proportion of them unvisited. Furthermore, total pollinator visitation showed a positive relation with female fitness. Results suggest that although pollinators preferred the heterozygous morph, they alternate indiscriminately among morphs making this polymorphism stable.     

Effects of flowering plant density on pollinator visitation, pollen receipt, and seed production in Delphinium barbeyi (Ranunculaceae)

Variation in flowering plant density can have conflicting effects on pollination and seed production. Dense flower patches may attract more pollinators, but flowers in those patches may also compete for pollinator visits and abiotic resources. We examined how natural and experimental conspecific flowering plant density affected pollen receipt and seed production in a protandrous, bumble bee-pollinated wildflower, Delphinium barbeyi (Ranunculaceae). We also compared floral sex ratios, pollinator visitation rates, and pollen limitation of seed set from early to late in the season to determine whether these factors mirrored seasonal changes in pollen receipt and seed production. Pollen receipt increased with natural flowering plant density, while seed production increased across lower densities and decreased across higher flower densities. Experimental manipulation of flowering plant density did not affect pollinator visitation rate, pollen receipt, or seed production. Although pollinator visitation rate increased 10-fold from early to late in the season, pollen receipt and seed set decreased over the season. Seed set was never pollen-limited. Thus, despite widespread effects of flowering plant density on plant reproduction in other species, the effects of conspecific flowering plant density on D. barbeyi pollination and seed production are minor.     

Positive interactions among plant species for pollinator service: assessing the ‘magnet species’ concept with invasive species

Plants with poorly attractive flowers or with little floral rewards may have inadequate pollinator service, which in turn reduces seed output. However, pollinator service of less attractive species could be enhanced when they are associated with species with highly attractive flowers (so called ‘magnet‐species’). Although several studies have reported the magnet species effect, few of them have evaluated whether this positive interaction result in an enhancement of the seed output for the beneficiary species. Here, we compared pollinator visitation rates and seed output of the invasive annual species Carduus pycnocephalus when grow associated with shrubs of the invasive Lupinus arboreus and when grow alone, and hypothesized that L. arboreus acts as a magnet species for C. pycnocephalus. Results showed that C. pycnocephalus individuals associated with L. arboreus had higher pollinator visitation rates and higher seed output than individuals growing alone. The higher visitation rates of C. pycnocephalus associated to L. arboreus were maintained after accounting for flower density, which consistently supports our hypothesis on the magnet species effect of L. arboreus. Given that both species are invasives, the facilitated pollination and reproduction of C. pycnocephalus by L. arboreus could promote its naturalization in the community, suggesting a synergistic invasional process contributing to an ‘invasional meltdown’. The magnet effect of Lupinus on Carduus found in this study seems to be one the first examples of indirect facilitative interactions via increased pollination among invasive species.     

Cost and benefits of attractive floral traits in the annual species Madia sativa (Asteraceae)

The maintenance of flower size variation within populations might be explained by conflicting selection pressures on floral traits that may involve biological agents, such as mutualists and antagonists, and allocation costs associated with floral display. The annual species Madia sativa (Asteraceae) exhibits ample variation in the number of ray florets in natural populations. This field study aimed at evaluating the costs and benefits associated with floral traits in M. sativa. In particular, we addressed two main questions: (1) Is the number of ray florets positively associated with pollinator visitation rate? (2) Is there a fitness cost of ray floret maintenance when pollinators are absent? We detected one benefit of conspicuous ray capitula: a strong preference by insect pollinators. We also confirmed the occurrence of costs: when pollinators were excluded conspicuous ray capitula had a reduced reproductive assurance via autogamous selfing, and there were trade-offs between the number of ray florets and seed mass and seed germination. Results suggest that the maintenance of within-population variation in the number of ray florets in M. sativa is explained, at least in part, by the balance between costs and benefits associated with this floral trait.     

Intrinsic and extrinsic factors acting on the reproductive process in alpine‐snowbed plants: roles of phenology,biological interaction,and breeding system

Pollinator activity and competition for pollinators lead to quantitative and qualitative pollen limitations on seed production and affect the reproductive success of plant species, depending on their breeding system (e.g., self‐compatibility and heterospecific compatibility) and genetic load (e.g., inbreeding depression and hybrid inviability). In alpine ecosystems, snowmelt regimes determine the distribution and phenology of plant communities. Plant species growing widely along a snowmelt gradient often grow with different species among local populations. Their pollinators also vary in their abundance, activity, and behavior during the season. These variations may modify plant–pollinator and plant–plant interactions. We integrated a series of our studies on the alpine dwarf shrub, Phyllodoce aleutica (Ericaceae), to elucidate the full set of intrinsic (species‐specific breeding system) and extrinsic factors (snow condition, pollinator activity, and interspecific competition) acting on their reproductive process. Seasonality of pollinator activity led to quantitative pollen limitation in the early‐blooming populations, whereas in the late‐blooming populations, high pollinator activity ensured pollination service, but interspecific competition for pollinators led to qualitative and quantitative pollen limitation in less competitive species. However, negative effects of illegitimate pollen receipt on seed‐set success might be reduced when cryptic incompatibility systems (i.e., outcross pollen grains took priority over self‐ and heterospecific pollen grains) could effectively prevent ovule and seed discounting. Our studies highlight the importance of species‐specific responses of plant reproduction to changing pollinator availability along environmental gradients to understand the general features of pollination networks in alpine ecosystems.     

A nodding capitulum enhances the reproductive success of Cremanthodium campanulatum (Asteraceae) at high elevations in the Sino–Himalayan Mountains

Background: Flower orientation is considered an evolutionary response to pollinators and abiotic factors. Several members of the genus Cremanthodium (Asteraceae) growing in alpine habitats in the Sino–Himalayas have conspicuously nodding capitula, the function of which is not understood.

Aims: We investigated the influences of nodding capitula on floral thermal conditions, pollinator visitation rate, pollen viability and achene production in Cremanthodium campanulatum.

Results: (1) the nodding capitulum did not modify internal thermal conditions, but the elongated involucral bracts could shelter pollen from UV-B radiation; (2) water and UV-B radiation seriously reduced the viability of pollen grains; (3) capitula artificially held erect set significantly fewer achenes than nodding ones; and (4) no pollinator preference was observed between capitula that were artificially held erect and natural nodding ones.

Conclusions: This study indicated that the nodding capitulum of C. campanulatum may protect pollen from being damaged by UV-B radiation, being washed away from the anther and stigma, and enable it to avoid rain damage during the wet monsoonal period. Our results suggest that nodding capitula in C. campanulatum possibly confer a selective advantage and may be driven by non-biological agents rather than pollinator attractors.     

Effect of elevation on sexual reproduction in alpine populations of Saxifraga oppositifolia (Saxifragaceae)

Self-compatibility in high arctic and alpine areas is regarded as an adaptation to low pollinator abundance. However, high genetic variability as a consequence of outcrossing is, with regard to population persistence, favorable in highly stochastic environments such as tundra habitats. To evaluate these contradictory scenarios, I performed in situ pollination experiments to examine the breeding system of the predominant outcrosser Saxifraga oppositifolia in ten populations at two different elevations in the Swiss Alps. Pollinator limitation was detected at both elevations, but fruit set in naturally pollinated flowers was only slightly less at the higher elevation. Increased pollinator limitation at high compared with low elevation thus could not be demonstrated in this experiment. Hand-crossings yielded equal mean proportion seed set at both elevations, and so did hand-selfings. This constant pattern of the breeding system in S. oppositifolia indicates selective factors that lead to the maintenance of a high level of outcrossing even in high-elevation populations. Based on sex allocation models, it was expected that a high ovule number should be selectively advantageous in a plant-pollinator system where chance visitation or selfing play important roles. However, female reproductive offer in terms of ovule number per flower did not change from low to high elevation. Since neither increased pollinator limitation nor increased seed set in selfed flowers was found at high compared with low elevation, the prerequisites for testing the hypothesis were not given. This study contradicts the hypothesis that inimical environmental factors in alpine or arctic habitats necessarily select for increased selfing rates in a preferentially outcrossing species like S. oppositifolia. Received: 28 April 1997 / Accepted: 20 October 1997     

Relationships between species’ floral traits and pollinator visitation in a temperate grassland

Knowledge about plant–plant interactions for pollinator service at the plant community level is still scarce, although such interactions may be important to seed production and hence the population dynamics of individual plant species and the species compositions of communities. An important step towards a better understanding of pollination interactions at the community level is to assess if the variation in floral traits among plant species explain the variation in flower visitation frequency among those species. We investigated the relative importance of various floral traits for the visitation frequency of all insects, and bumblebees and flies separately, to plant species by measuring the visitation frequency to all insect-pollinated species in a community during an entire flowering season. Visitation frequency was identified to be strongly positive related to the visual display area and the date of peak flowering of plant species. Categorical variables, such as flower form and symmetry, were important to the visitation frequency of flies only. We constructed floral similarity measures based on the species’ floral traits and found that the floral similarity for all species’ traits combined and the continuous traits separately were positively related to individual visitation frequency. On the other hand, plant species with similar categorical floral traits did not have similar visitation frequencies. In conclusion, our results show that continuous traits, such as flower size and/or density, are more important for the variation in visitation frequency among plant species than thought earlier. Furthermore, differences in visitation frequency among pollinator groups give a poor support to the expectations derived from the classical pollination syndromes.     

Drivers of plant individual-based pollinator visitation network topology in an arid ecosystem

Interactions with pollinators underlie the structure and function of plant communities. Network analysis is a valuable tool for studying plant-pollinator interactions, but these networks are most frequently built by aggregating interactions at the species level. Interactions are between individuals and an advantage of individual-based networks is the ability to integrate inter-individual variation in traits and environmental context within complex ecological networks. We studied the influence of inter-individual variation on pollinator sharing among foundation shrubs and cactus in a desert ecosystem using plant individual-based pollinator visitation networks. We hypothesized that the traits that alter attractiveness of plants to pollinators will also influence an individual plant's role within the visitation network. Foundation plants growing with higher densities of nearby blooming shrubs had higher pollinator visitation rates and had greater access to the conspecific mating pool, suggesting widespread and diffuse pollination facilitation within this community. Further, shrub density influenced the role of betweenness centrality and the effective number of partners (e^H). Floral display size also influenced the effective number of interaction partners but did not directly influence the centrality measures for individual plants or other measures of network structure despite increasing visitation rates. The individual-based visitation networks were significantly modular and module membership was predicted by species identity and pollinator visitation rates. Ecological and individual context mediate the outcome of pollinator-mediated interactions and are fundamental drivers of whole community structure. This study shows that the density of immediate neighbours can influence the overall structure of plant-pollinator interaction networks. Exploring the contribution of intraspecific variation to community interaction networks will improve our understanding of drivers of community-level ecological dynamics.     

Refuse attracts? Effect of refuse dumps of leaf‐cutting ants on floral traits

The nutrient‐rich organic waste generated by ants may affect plant reproductive success directly by enhancing fruit production but also indirectly, by affecting floral traits related with pollinator attraction. Understanding how these soil‐nutrient hot spots influence floral phenotype is relevant to plant–pollination interactions. We experimentally evaluated whether the addition of organic waste from refuse dumps of the leaf‐cutting ant Acromyrmex lobicornis (Hymenoptera: Formicidae: Attini) alters floral traits associated with pollinator attraction in Eschscholzia californica (Ranunculales: Papaveraceae), an entomophilous herb. We analysed flower shape and size using geometric morphometric techniques in plants with and without the addition of refuse‐dumps soil, under greenhouse conditions. We also measured the duration of flowering season, days with new flowers, flower production and floral display size. Plants growing in refuse‐dumps soil showed higher flower shape diversity than those in control soil. Moreover, plants in refuse‐dumps soil showed bigger flower and floral display size, longer flowering season, higher number of flowering days and flower production. As all these variables may potentially increase pollinator visits, plants in refuse‐dumps soil might increase their fitness through enhanced attraction. Our work describes how organic waste from ant nests may enhance floral traits involved in floral attraction, illustrating a novel way of how ants may indirectly benefit plants.     

Shrub canopy matrix decreases reproductive output of a sheltered plant via pollinator exclusion

Shrubs establish microenvironments under their canopies that can favor the growth of other plants. However, the shrub canopy could impede pollination by reducing the number of pollinator visits to sheltered plants, resulting in pollen limitation and decreased reproductive output. We assessed whether the presence of a nurse shrub species (Flourensia thurifera) alters the reproductive output of a sheltered cactus (Eriosyce coimasensis) via the restriction of access by the giant hummingbird (Patagona gigas) to E. coimasensis flowers. During two consecutive years (2018 – 2019), we excluded hummingbirds from individual cacti (using cages) and studied fruit set and seed production in two microhabitats: underneath shrubs and in open sites. In addition, we compared the reproductive mode of E. coimasensis in the two microhabitats. We observed that shrubs limit the reproduction of E. coimasensis, which strongly depends on P. gigas for seed production. Plants in open sites produced 80% more fruit and 76% more seeds than those growing underneath shrubs. The reproduction of caged individuals was low and similar to those growing beneath shrubs. In addition, plants underneath shrubs, but not in open sites, may suffer pollen limitation. Our results offer novel insights into plant-plant interactions and suggest potential trade-offs for sheltered cacti between the mild microclimatic conditions under the canopy, that could lead to larger plants and pollinator preclusion, which decreases the reproductive performance of sheltered plants.     

Experimental reduction of pollinator visitation modifies plant–plant interactions for pollination

The strength of interactions between plants for pollination depends on the abundance of plants and pollinators in the community. The abundance of pollinators may influence plant associations and densities at which individual fitness is maximized. Reduced pollinator visitation may therefore affect the way plant species interact for pollination. We experimentally reduced pollinator visitation to six pollinator‐dependent species (three from an alpine and three from a lowland community in Norway) to study how interactions for pollination were modified by reduced pollinator availability. We related flower visitation, pollen limitation and seed set to density of conspecifics and pollinator‐sharing heterospecifics inside 30 dome‐shaped cages partially covered with fishnet (experimental plots) and in 30 control plots. We expected to find stronger interactions between plants in experimental compared to controls plots. The experiment modified plant–plant interactions for pollination in all the six species; although for two of them neighbourhood interactions did not affect seed set. The pollen limitation and seed set data showed that reduction of pollinator visits most frequently resulted in novel and/or stronger interactions between plants in the experimental plots that did not occur in the controls. Although the responses were species‐specific, there was a tendency for increasing facilitative interactions with conspecific neighbours in experimental plots where pollinator availability was reduced. Heterospecifics only influenced pollination and fecundity in species from the alpine community and in the experimental plots, where they competed with the focal species for pollination. The patterns observed for visitation rates differed from those for fecundity, with more significant interactions between plants in the controls in both communities. This study warns against the exclusive use of visitation data to interpret plant–plant interactions for pollination, and helps to understand how plant aggregations may buffer or intensify the effects of a pollinator loss on plant fitness.