Background and Aims Flower colour plays a major role in the attraction and decision-making of pollinators. Different functional groups of pollinators tend to prefer different flower colours, and therefor may lead to different flower colour compositions among different communities depending on the visual system of the dominant pollinators. However, few studies have investigated the linkage between pollinator fauna and flower colour composition in natural communities, a theme we explored in the present study. • Methods Flower spectral reflectance of 106 Japanese and 96 New Zealand alpine plants in the wavelength range 300-700 nm were measured. The composition of pollinator fauna in the communities and the types of pollinators for each plant species were also investigated. • Key Results Based on bee and fly colour vision models, as well as a principal components analysis, considering phylogenetic non-independence between plant species, flower colours appeared to vary according to pollinator type rather than geographical region. Consequently, flower colour composition differed between the regions, reflecting the bee/fly mixed pollinator fauna of Japan and the fly-dominant pollinator fauna of New Zealand. According to the bee colour vision model, the majority of the colours of hymenopteran-pollinated flowers appeared to be discriminated by bees. In contrast, many of the colours of dipteran-pollinated flowers would not be discriminated by bees and flies. • Conclusion The results suggest that the differences in flower colour composition between Japanese and New Zealand alpine communities are due to differences in the pollinator fauna in these communities rather than differences in abiotic factors between the geographical regions and the phylogenetic origin of the communities.

No abstract

Understanding how landscape composition affects pollinator abundance is crucial for conserving the sustainability of the symbiotic relationship between plants and pollinators. However, field evidence on the effect of landscape composition and spatial scales on the abundance of each pollinator group is biased toward bee pollinators, and non-bee pollinators tend to be excluded. This study aims to evaluate the effect of landscape composition at various spatial scales on diverse taxonomic groups of pollinators, including non-bee pollinators. We selected 22 grasslands with varying landscape compositions in Shiroi City, Chiba Prefecture, Japan. We recorded the abundance of pollinators and floral resource availability at these sites. The effects of landscape composition within a 100-1000 m radius and the floral resource availability were analyzed using Bayesian generalized linear mixed models to determine whether these factors affected the abundance of each pollinator group. As a result, the difference in important factors among the pollinator groups was detected; Scoliidae, Lycaenidae, Stomorhina obsoleta, and Sphaerophoria were influenced by the landscape composition, whereas Bombus diversus and Hesperiidae were not. Scoliidae was affected by the proportion of agri-field and forest areas within a 500-1000 m radius. St. obsoleta and Sphaerophoria showed different response patterns to the proportion of urban land within a 100 m radius. B. diversus and Hesperiidae were influenced by floral resource availability. Our study suggests that considering the spatial scales of landscape composition on the target species is essential to plan effective conservation measures.

Some animal‐pollinated flowers produce no rewards but attract pollinators by imitating rewarding flowers. Termed Batesian floral mimicry, this phenomenon has generally been understood in terms of a common system of the organisms performing one of three roles as model flowers, mimic flowers or signal receivers (a Batesian mimicry trinity). However, organisms other than those of the mimicry trinity may play important roles in the relationships of the trinity members. This study examined the effect of alternative flowers that are dissimilar to either model or mimic flowers on bumblebee discrimination between model and mimic flowers. To this end, we observed bumblebees Bombus ignitus (Smith), foraging in mixed arrays of artificial flowers mirroring either model, mimic or alternative flowers. Bumblebees’ ability to discriminate mimic from model flowers was impacted by the presence of alternative flowers; however, the effect was dependent on the similarity between the model and mimic flowers. This might occur because the presence of alternative flowers imposed an additional cognitive load on the bees’ working memory. Our study suggests that alternative flowers in a community may have beneficial effects on mimic flowers by reducing the discrimination performance of pollinators. A variety of environmental factors that impose complex sensory inputs on pollinators’ working memory might have contributed to the maintenance of Batesian floral mimicry. A free Plain Language Summary can be found within the Supporting Information of this article.