The congeners Rhinanthus angustifolius and Rhinanthus minor, two annual hemiparasites pollinated by bumblebees, are known to hybridise in the wild. Both species are self-compatible, but the capacity for autonomous selfing is higher in R. minor. This suggests a difference in realized outcrossing rates, which have not been determined before in these species. Using microsatellites, both species turned out to have mixed mating systems, but with a much lower multilocus outcrossing rate (0.13) for R. minor compared to R. angustifolius (0.76). We hypothesised that a higher outcrossing rate should lead to a higher chance of heterospecific pollination, and we therefore determined the rate of hybrid formation on each species in an artificial mixed population. Hybrid seeds were produced at low frequency (4.5%), and no significant difference was found between the species. It is therefore likely that post-pollination processes influence hybrid seed formation to counteract the expected difference in heterospecific pollen deposition. We checked fruit set, seed set and the rate of autonomous selfing in controlled crosses in the greenhouse in 2 years, and found that fruit set (2003) or seed set (2010) were lower in R. angustifolius × R. minor crosses relative to the reciprocal cross. Hybrid seeds produced on R. angustifolius also had a much lower germination rate, so most of the established F1 hybrid plants have the R. minor cytoplasm. The formation of advanced hybrids depends on pollinator preference, which is biased towards R. angustifolius if present in sufficient numbers, because it offers more rewards.

No abstract

Hybridization in annual plants is rare, but their short life cycle provides an excellent opportunity to study the dynamics of hybridization. Hybridization occurs between the annual hemiparasites Rhinanthus minor and Rhinanthus angustifolius (Orobanchaceae). Using flower morphology, Kwak (1980) found a prevalence of hybrids close to R. angustifolius in a single population. We aim to find whether this pattern is also found using genetic markers, whether it is generally occurring in mixed populations, and whether these populations are stable over time. We used species-specific genetic markers to determine the number of individuals in a range of hybrid classes in three mixed populations of different ages during four consecutive years. In the young population, F 1 hybrids were found in the first year and mostly hybrids between R. minor and these F 1 s in the second year, but in the years after that, hybrids close to R. angustifolius became more abundant. We also found this in the two older populations, where hybrids close to R. angustifolius always occurred in higher frequencies than hybrids close to R. minor. Over time, R. angustifolius strongly increased in frequency in two populations. Patterns of marker presence and absence suggested that advanced-generation hybrids are mainly formed by backcrossing with one of the parents, predominantly R. angustifolius whenever its frequency in the population is higher than 15%. The dynamics of mixed populations depend on the ecological conditions that regulate the presence of the two parental species, and introgression into R. angustifolius seems prevalent.

Rhinanthus angustifolius and Rhinanthus minor are annual hemiparasitic herbs found in open vegetations. They are closely related, hybridize frequently, and appear to have largely overlapping niches, although some floras report a preference of R. angustifolius for more humid conditions compared to R. minor. We analysed the relative fitness of both species by following their fate from germination to seed production in three different hydric treatments (wet, moist and dry) in a garden experiment. We found that R. angustifolius was either as fit as or fitter than R. minor in all conditions and for all studied parameters, with a slightly higher flower production in R. angustifolius in the wet treatment. Wet conditions had a negative effect on germination and flowering rate, but they increased flower and seed production. These observations show that environmental conditions will influence the composition of mixed Rhinanthus populations.

Fifteen polymorphic microsatellite loci were developed from an enriched genomic library of the annual plant Rhinanthus angustifolius and characterized using 36 individuals. These markers provided high polymorphism ranging from two to 15 alleles per locus. Four loci showed significant departure from Hardy-Weinberg equilibrium, probably because of the occurrence of null alleles. No significant linkage disequilibrium was detected between pairs of loci. Tests of cross-species transferability were performed on four congeners with a success rate of 100% in Rhinanthus minor, 93% in R. mediterraneus and R. glacialis, and 80% in R. alectorolophus. These microsatellite loci will be useful tools to study mating system, gene flow and hybridization in the genus Rhinanthus.