Summary: 1. Pollen limitation may be a consequence of changes in pollinator abundance, diversity and identity. However, no empirical evidence exists concerning the consequences that the spatial variation in pollinator fauna has on pollen limitation intensity and plant reproduction. In this study, we test the effect that changes in flower‐visitor abundance, diversity and identity exert on the occurrence and strength of pollen limitation by experimentally quantifying pollen limitation in eight populations ofErysimum mediohispanicum, a pollination‐generalist plant native to the Iberian Peninsula.2. Pollen limitation was accounted for by using a comprehensive estimator, the net reproductive rate (R₀). Nevertheless, we also determined which components of plant reproduction, from ovule fertilization to seedling survival, were more intensely pollen‐limited. Finally, we explored whether the spatial variation in pollen limitation intensity was related to among‐population changes in flower‐visitor abundance, diversity and identity.3. The whole reproductive cycle ofE. mediohispanicumwas pollen‐limited, although pollen limitation occurred more strongly during the ovule fertilization and seed‐production phases than during fruit ripening or seedling emergence and establishment.4. There was a significant among‐population difference in pollen limitation intensity. Pollen limitation strength was associated with variations in flower‐visitor diversity, and identity. Populations with lower flower‐visitor diversity and with many low‐efficiency pollinators (i.e. beetles) showed stronger pollen limitation.5. Synthesis. Our study shows that the intensity of pollen limitation at the population level may depend on several characteristics of the assemblage of flower‐visiting insects, such as their abundance, diversity and identity. Our results suggest that any impoverishment of pollinator diversity or any alteration in the specific composition of the pollinator assemblage may exacerbate pollen limitation.

How generalist plants diverge in response to pollinator selection without becoming specialized is still unknown. This study explores this question, focusing on the evolution of the pollination system in the pollination generalist Erysimum mediohispanicum (Brassicaceae). Pollinator assemblages were surveyed from 2001 to 2010 in 48 geo-referenced populations covering the entire geographic distribution of E. mediohispanicum. Bipartite modularity, a complex network tool, was used to find the pollination niche of each population. Evolution of the pollination niches and the correlated evolution of floral traits and pollination niches were explored using within-species comparative analyses. Despite being generalists, the E. mediohispanicum populations studied can be classified into five pollination niches. The boundaries between niches were not sharp, the niches differing among them in the relative frequencies of the floral visitor functional groups. The absence of spatial autocorrelation and phylogenetic signal indicates that the niches were distributed in a phylogeographic mosaic. The ancestral E. mediohispanicum populations presumably belonged to the niche defined by a high number of beetle and ant visits. A correlated evolution was found between pollination niches and some floral traits, suggesting the existence of generalist pollination ecotypes. It is conjectured that the geographic variation in pollination niches has contributed to the observed floral divergence in E. mediohispanicum. The process mediating this floral divergence presumably has been adaptive wandering, but the adaptation to the local pollinator faunas has been not universal. The outcome is a landscape where a few populations locally adapted to their pollination environment (generalist pollination ecotypes) coexist with many populations where this local adaptation has failed and where the plant phenotype is not primarily shaped by pollinators.

Premise of the study: Organisms usually show intercorrelations between all or some of their components leading to phenotypic integration, which may have deep consequences on the evolution of phenotypes. One of the main difficulties with phenotypic integration studies is how to correct the integration measures for size. This has been considered a challenging task. In this paper, we introduce an R package (PHENIX: PHENotypic Integration indeX), in which we provide functions to estimate a size‐controlled phenotypic integration index, a bootstrapping method to calculate confidence intervals, and a randomization method to simulate null distributions and test the statistical significance of the integration.Methods and Results: PHENIX is an open source package written in R. As usual for R packages, the manual and sample data are available at: http://cran.r‐project.org/web/packages/PHENIX/index.html. Functions included in this package easily estimate phenotypic integration by controlling a third variable (e.g., the size of the studied organ).Conclusions: PHENIX helps to estimate and test the statistical significance of the magnitude of integration using one of the most‐used methodological approaches, while taking size into account.

Summary: The amount of evolutionary information in phylogeographic studies is usually limited due to the low divergence between closely related organisms. Consequently, obtaining the maximum evolutionary information contained in sequence alignments would be of great importance to infer phylogeographic relationships. SIDIER is a software package that allows inferring evolutionary relationships from gapped alignments using information contained in both substitutions and insertions/deletions (indels). SIDIER estimates the number of indel events that occurred during sequence evolution to obtain a distance matrix. This indel distance matrix may be combined with the substitution distance matrix calculated separately from the same data set. Using this software, the inferred evolutionary events can be represented by means of percolation networks. SIDIER is written in the open source R language and is freely available through the Comprehensive R Archive Network (http://cran.r-project.org/web/packages/sidier/index.html).