The long‐known, widely documented inverse relationship between body size and environmental temperature (“temperature‐size rule”) has recently led to predictions of body size decline following current climatic warming (“size shrinking effect”). For keystone pollinators such as wild bees, body shrinking in response to warming can have significant effects on pollination processes but there is still little direct evidence of the phenomenon because adequate tests require controlling for confounding factors linked to climate change (e.g., habitat change). This paper assesses the shrinking effect in a community of solitary bees from well‐preserved habitats in the core of a large nature reserve experiencing climatic warming without disturbances or habitat changes. Long‐term variation in mean body mass was evaluated using data from 1704 individual bees (137 species, 27 genera, 6 families) sampled over 1990–2023. Climate warmed at a fast rate during this period, annual mean of daily maximum temperature increasing 0.069°C/year on average during 2000–2020. Changes in bee body mass verified expectations from the size shrinking effect. The mean individual body mass of the community of solitary bees declined significantly, irrespective of whether the analysis referred to the full species sample or only to the subset of species that were sampled in both the old (1990–1997) and recent (2022–2023) periods. On average, body mass declined ~0.7%·year−1, or an estimated average cumulative reduction of ~20 mg per individual bee from 1990 to 2023. Proportional size reduction was greatest among large‐bodied species, ranging from around −0.6%·year−1 for the smallest species to −0.9%·year−1 for the largest ones. Declining rate was steeper for cavity‐nesting than ground‐nesting species. The pollination and mating systems of bee‐pollinated plants in the study region are probably undergoing significant alterations as a consequence of supra‐annual decline in bee body mass.
Within plant populations, space-restricted gene movement, together with environmental heterogeneity, can result in a spatial variation in gene frequencies. In biennial plants, inter-annual flowering migrants can homogenize gene frequencies between consecutive cohorts. However, the actual impact of these migrants on spatial genetic variation remains unexplored. Here, we used 10 nuclear microsatellite and one plastid genetic marker to characterize the spatial genetic structure within two consecutive cohorts in a population of the biennial plant Erysimum mediohispanicum (Brassicaceae). We explored the maintenance of this structure between consecutive flowering cohorts at different levels of complexity, and investigated landscape effects on gene flow. We found that cohorts were not genetically differentiated and showed a spatial genetic structure defined by a negative genetic-spatial correlation at fine scale that varied in intensity with compass directions. This spatial genetic structure was maintained when comparing plants from different cohorts. Additionally, genotypes were consistently associated with environmental factors such as light availability and soil composition, but to a lesser extent compared with the spatial autocorrelation. We conclude that inter-annual migrants, in combination with limited seed dispersal and environmental heterogeneity, play a major role in shaping and maintaining the spatial genetic structure among cohorts in this biennial plant.