LAUSR

It is broadly assumed that polyploidy success reflects an increase in fitness associated with whole‐genome duplication (WGD), due to higher tolerance to stressful conditions. Nevertheless, WGD also arises with several costs in neo‐polyploid lineages, like genomic instability, or cellular mis‐management. In addition to these costs, neo‐polyploid individuals also face frequency dependent selection because of frequent low‐fitness triploids formed by cross‐ploidy pollinations when tetraploids are primarily rare in the population. Interestingly, the idea that polyploidy can be fixed by genetic drift as a neutral or deleterious mutation is currently underexplored in the literature. To test how and when polyploidy can fix in a population by chance, we built a theoretical model in which autopolyploidization occurs through the production of unreduced gametes, a trait modelled as a quantitative trait that is allowed to vary through time. We found that when tetraploid individuals are less or as fit as their diploid progenitors, fixation of polyploidy is only possible when genetic drift is stronger than natural selection. The necessity of drift for tetraploid fixation holds even when polyploidy confers a selective advantage, except for scenarios where tetraploids are much fitter than diploids. Finally, we found that self‐fertilization is less beneficial for tetraploid establishment than previously thought, notably when polyploids harbour an initial decrease in fitness. Our results bring a novel, non‐exclusive explanation for the unequal temporal and spatial distribution of polyploid species.