Incompletely reproductively isolated species often segregate into different microhabitats, even when they are able to survive and reproduce in both habitats. Longer term evolutionary factors may contribute to this lack of cross-habitat persistence. When reproductive interference reduces immigrant fitness, assortative mating, including self-fertilization, increases immigrants’ fitness in a single generation, but longer-term, inbreeding depression may reduce the chance of population persistence. Two California monkeyflower species repeatedly segregate into drier and wetter areas in their zone of sympatry. To test whether inbreeding depression may contribute to the maintenance of this segregation pattern, we transplanted outbred and successively inbred Mimulus guttatus and M. nudatus into their native habitats and heterospecific habitats. We measured germination, survival, and seed set and found that recurrent selfing reduced all aspects of fitness in both species, most strongly in foreign habitats. A simulation model, parameterized from the transplant experiment, found that inbreeding reduced fitness to such an extent that sequentially inbred populations of either species would be unable to persist in heterospecific-occupied habitats in the absence of continued gene flow. These results demonstrate that individual immigrants are unlikely to form persistent populations and thus, inbreeding depression contributes to the absence of fine-scale coexistence in this species pair.

Three datasets and two R scripts are included. The germination dataset was collected in a greenhouse and the reciprocal transplant dataset was collected at two field sites at the UC McLaughlin Reserve. Seeds were censused for germination in the greenhouse and a subset of germinants were transplanted into the field. In the field, tranplants were censused for survival and flowering. Fruit were collected at the end of the season, then seeds were counted from every transplant and categorized as viable or inviable based on morphology. The statistical analysis R code tests whether recurrent selfing reduces germination in the greenhouse, and survival and seed production in the field. The simulation R code is parameterized with data from the reciprocal transplant and simulates seed immigration with and without recurrent selfing as described in the manuscript. The simulation results dataset is the output of the simulation code for recurrently selfed and outcrossed seeds. 

MetadataS1.docx is a README file that describes the R code and data files, including column decscriptions for each data file.