Search

Abstract: Many alpine plants are predominantly outcrossing, thus plant reproductive success is highly dependent on effectiveness of pollinators. How pollinators transfer pollen from one flower to another is of great interest in understanding the genetic structure in plant populations. We studied (1) the role and effectiveness of insect visitors for pollination, and (2) their contribution as pollen vectors for gene dispersal in a Rhododendron ferrugineumpopulation. Various insect visitors were recorded, including Hymenoptera, Diptera, Coleoptera, and Lepidoptera. The most frequent and effective insects were honey bees and bumblebees. Muscid flies were considered as important pollinators, particularly due to their relatively high visitation rate. Syrphid flies, Formicidae, and Coleoptera were ineffective in transporting pollen, while the effectiveness of Lepidoptera and Empididae was negligible. A fluorescence labelling experiment revealed that pollen dispersal was restricted (0 ‐ 2 m) in a dense R. ferrugineumstand and decreased in a leptokurtic fashion. This might lead to geitonogamous self‐pollination that could explain the close relationship between individuals found in genetic studies of R. ferrugineum.However, some pollen grains may travel 40 ‐ 45 m, which implies the occurrence of cross‐pollination through the foraging activities of bumblebees and honey bees.

Many alpine plants are predominantly outcrossing, thus plant reproductive success is highly dependent on effectiveness of pollinators. How pollinators transfer pollen from one flower to another is of great interest in understanding the genetic structure in plant populations. We studied (1) the role and effectiveness of insect visitors for pollination, and (2) their contribution as pollen vectors for gene dispersal in a Rhododendron ferrugineum population. Various insect visitors were recorded, including Hymenoptera, Diptera, Coleoptera, and Lepidoptera. The most frequent and effective insects were honey bees and bumblebees. Muscid flies were considered as important pollinators, particularly due to their relatively high visitation rate. Syrphid flies, Formicidae, and Coleoptera were ineffective in transporting pollen, while the effectiveness of Lepidoptera and Empididae was negligible. A fluorescence labelling experiment revealed that pollen dispersal was restricted (0 - 2 m) in a dense R. ferrugineum stand and decreased in a leptokurtic fashion. This might lead to geitonogamous self-pollination that could explain the close relationship between individuals found in genetic studies of R. ferrugineum. However, some pollen grains may travel 40 - 45 m, which implies the occurrence of cross-pollination through the foraging activities of bumblebees and honey bees.

Genotypic structure of a closed population of the clonal ericaceous shrub Rhododendron ferrugineum is examined in the light of two independent studies previously conducted on this species. In the first study, spatial distribution of genotypes in the closed population was inferred from the amplified fragment length polymorphism (AFLP) procedure. Age of clones was estimated using their spatial extent and the annual growth rate of shoots. In the second study, ramet demography was studied in the three most representative stages of shrub invasion on two different sites, including the site where the population investigated by AFLP grew. The demographic data recorded were the area occupied by ramets and ramet age, and from this information the developmental pattern of Rhododendron populations was determined. Additional data such as genet density and distance between genets were calculated. These two sources of information allow us to propose that all or most of the clones detected in the closed population established at the early successional stage, and that the present genotypic structure was established several hundred years ago, long before the population reached total closure. Hypotheses concerning the future development of this genotypic structure are discussed.

Sexual and vegetative reproduction in the alpine species Rhododendron ferrugineum was studied along a successional sequence (meadow → open heathland → closed heathland) at two sites and in a wet heathland. This study aims to determine (1) the characteristics of sexual reproduction in R. ferrugineum populations (2) when and how these populations develop layering (adventitious rooting) and (3) whether reproductive traits and reproductive strategies develop relative to the degree of population closure and maturity. The variables used to describe sexual reproduction were inflorescence density (per m(2)of Rhododendron cover), number of flowers per inflorescence and per m(2)of Rhododendron cover, and seeds production (per fruit and m(2)of Rhododendron cover). Flowering and fruiting phenologies were also recorded. For describing clonal development, we investigated layering variables such as length and annual growth rate of prostrate stems, rooting occurence and ramet density. The results show that the direction toward which the clones extend is mainly determinated by the topography, and that layering steadily increases with increasing population closure and maturity. Reproductive potential of R. ferrugineum is enormous (0.4–2.4 million seeds m(-2)) but reproductive effort remains low with respect to total biomass of seeds (3–21 g m(-2)). Reproductive effort of R. ferrugineum populations could be reduced as a conterpart of layering development only when the shrub draws more matter and energy in layering stems than to aerial stems. The variations in reproductive traits observed on our sites could be due to primarily to phenotypic response to variable microhabitat features, rather than to genetically deterministic processes.