To what extent is evolution repeatable? Little is known about whether the evolution of hybrids is more (or less) repeatable than non-hybrids. We used field experimental evolution in annual sunflowers (Helianthus) in Texas to ask the extent to which hybrid evolution is repeatable across environments compared to non-hybrid controls. We created hybrids between Helianthus annuus (L.) and H. debilis (Nutt.) and grew plots of both hybrids and non-hybrid controls through eight generations at three sites in Texas. We collected seeds from each generation and grew each generation × treatment × home site combination at two final common gardens. We estimated the strength and direction of evolution in terms of fitness and 24 traits, tested for repeated versus non-repeated evolution, and assessed overall phenotypic evolution across lineages and in relation to a locally adapted phenotype. Hybrids consistently evolved higher fitness over time while controls did not, though trait evolution varied in strength across home sites. Repeated evolution was more evident in hybrids versus non-hybrid controls, and hybrid evolution was often in the direction of the locally adapted phenotype. Our findings have implications for both the nature of repeatability in evolution and the contribution of hybridization to evolution across environmental contexts.

Data were collected from common gardens in 2017 and 2019. Data were collected on paper, transcribed to Microsoft Excel, then processed using R.