Analyzing the processes that determine whether species boundaries are maintained on secondary contact may shed light on the early phase of speciation. In Anacamptis morio and Anacamptis longicornu, two Mediterranean orchid sister-species, we used molecular and morphological analyses, together with estimates of pollination success and experimental crosses, to assess whether floral isolation can shelter the species' genomes from genetic admixture on secondary contact. We found substantial genetic and morphological homogenization in sympatric populations in combination with an apparent lack of postmating isolation. We further detected asymmetric introgression in the sympatric populations and an imbalance in cytotype representation, which may be due either to a difference in flowering phenology or else be a consequence of cytonuclear incompatibilities. Estimates of genetic clines for markers across sympatric zones revealed markers that significantly deviated from neutral expectations. We observed a significant correlation between spur length and reproductive success in sympatric populations, which may suggest that directional selection is the main cause of morphological differentiation in this species pair. Our results suggest that allopatric divergence has not led to the evolution of sufficient reproductive isolation to prevent genomic admixture on secondary contact in this orchid species pair.

Tertiary geological processes and Quaternary climatic oscillations significantly contributed in determining plant species ranges in the Mediterranean basin, making it difficult to distinguish between relative influence of habitat fragmentation and long-distance dispersal in shaping actual species distribution across disjointed regions. We studied the phylogeographic structure of Anacamptis longicornu, a Western Mediterranean orchid species spanning over the main continental islands (Sicily, Sardinia, Corsica, Balearic Islands) and North Africa. We aimed to answer the question whether the highly disjointed distribution of this orchid was determined by: (1) recent over-sea dispersion after the last glacial phase; (2) range fragmentation after the refilling of Mediterranean basin at the end of the Messinian Salinity Crisis (MSC); (3) ancient dispersal events linked to the insurgence of favourable bioclimatic conditions in the basin. Haplotype differentiation, haplotype network, and mismatch analysis support the hypothesis that the current distribution of A. longicornu is more likely explained by allopatric separation and restricted gene flow in connection with fragmentation of a continuous range or a series of ancient dispersal events, after the MSC but before the most recent glaciations. Recent over-sea dispersion, if occurring, must have played a marginal role.

No abstract

No abstract