Modelling the long-term consequences of crop-wild relative hybridization: a case study using four generations of hybrids

1. Hybridization between crops and wild relatives seems possible for many crop taxa. Subsequent introgression of crop-specific traits into wild relatives might lead to the formation of introgressant populations. Until now, few studies have evaluated demographically fitness changes combined with empirically defined heterosis breakdown levels to assess the likelihood of such potential hybrid population formation. 2. For the establishment of case study data, we produced four generations of hybrids between the predominantly autogamous annuals Lactuca sativa (lettuce) and its wild relative L. serriola, along both the autogamous and backcross pathways. Seeds of parental and hybrid lineages were sown individually in field plots and monitored up to and including seed-set. 3. All four hybrid generations were phenotypically similar to L. serriola, but survival rates of hybrids in both pathways were significantly higher in early (1st and 2nd) hybrid generations. Combined with higher germination rates, this resulted in higher λs for all hybrid classes, relative to L. serriola. This fitness surplus decreases through the generations, which is interpreted as heterosis breakdown. 4. These data were entered in a stochastic model. In general, the likelihood of a rapid full displacement of L. serriola at realistic, i.e. low, outcrossing rates seems limited. More likely is the formation of a population containing a wide variety of genotypes with or without L. serriola included. 5. Synthesis and applications. We demonstrate a simple model for providing likelihood estimates for different scenarios of introgression consequences, with the potential to be used for risk assessment of new crops. When including estimates of heterosis breakdown in our predictions, the speed of displacement of the wild taxon is less dramatic than other modelling attempts suggest. Furthermore, although wild relative displacement is the most likely scenario, other outcomes are possible, including no displacement at all.