Hybrid zones with multiple independent contact regions between the same species allow to determine the relative importance of intrinsic and extrinsic factors in the evolution of hybrid zones and thus, parallelism in hybridization outcomes. In this study, we take advantage of two hybrid regions between the damselfly species Ischnura elegans and I. graellsii in Spain to measure: i) the extent of parallelism across geographic hybridization replicates, and what factors (intrinsic and extrinsic) drive that variation; and ii) if hybridization has an impact on the ability of species to expand their ranges. RAD sequencing was used to generate 5,702 SNPs to quantify population diversity and population differentiation, and a subset of 381 species-specific SNPs to analyze genotypic composition (individual ancestry and the proportion of individuals in different hybrid classes Our individual ancestry results showed on-going hybridization and introgression with different admixture-class distributions between hybrid regions and between populations explained by i) species proportions, ii) time elapsed since colonization, and iii) asymmetric and reinforced prezygotic barriers and Batson Dobzhansky and Müller (BDM) hybrid incompatibilities, and indicated a role of hybridization as a facilitator of species range expansions. Our study highlights the value of studying complex hybrid zones to gain insights into microevolutionary processes.

The outcome of hybridization is of major interest in evolutionary and conservation biology. Here, we investigate (i) the genomic signal of the hybridization dynamics, (ii) the strength of reproductive barriers preventing copulation in heterospecific and hybrid crosses, and (iii) the population dynamics (stability of species proportions) of the two damselfly species Ischnura elegans and I. graellsii in two differently aged Spanish hybrid regions. RAD sequencing in these hybrid regions and in allopatric populations was used to generate 5,702 SNPs to quantify population diversity and population differentiation, and a subset of 381 species-specific SNPs to analyze individual ancestry and the proportion of individuals in different hybrid classes. Our individual ancestry results showed the presence of F1 and F2 hybrids, in line with on-going hybridization and bidirectional backcrossing in both hybrid regions, with almost complete absence of genetically pure I. elegans and I. graellsii. Different admixture-class distributions were in part explained by 1) different mean strength of reproductive barriers in the hybrid regions, with stronger barrier in the older hybrid region, 2) local dynamics (continuous recolonization events), 3) proximity to introduction site, and 4) time elapsed since colonization. Consistent with theoretical expectations, introgression maintained (in the younger hybrid region) or increased genetic diversity (in the older hybrid region), and reduced genetic differentiation between local populations in both hybrid regions. Whether this will facilitate the ongoing range expansion of I. elegans in Spain is an interesting avenue for future research.