Plant phenology, such as the timing of bud burst in spring for trees or the timing of flowering in many species, shows strong plastic responses to temperature. Interestingly, these phenological traits also often show rapid genetic evolution in the context of a changing climate and there has been discussion about the potential of plasticity and rapid evolution of phenology to mitigate the negative effects of climate change. In this talk I will summarize the findings of several modelling studies conducted in a large collaborative project, which have shed light on this issue. Our work explores in particular the evolutionary consequences of the fact that variation in phenology commonly affects who mates with who in a population. First, individuals can often mate only with other individuals with a similar phenology (e.g. with overlapping flowering time), which is described as assortative mating. Second, individuals with atypical phenology may have less mating opportunities than others. Variation in the number of mates generates sexual selection, which can conflict with natural selection on phenology imposed by climate. We used quantitative genetics models of adaptation to a changing environment to explore these questions, assuming that some phenological trait (e.g. flowering time) is partly determined by genes at several loci, and partly by a plastic response to temperature. Using both individual-based simulation and analytical results, we found that assortative mating and the sexual selection it generates shape the evolution of phenotypic plasticity for phenology along climatic gradients, such that the traits expressed in a changing climate deviate from the optimal values favoured by natural selection. Ours models predict the evolution of either suboptimal plasticity (reaction norms with a slope shallower than optimal) or hyperplasticity (slopes steeper than optimal) in the presence of assortative mating when optimal plasticity would evolve under random mating. Furthermore, a co-gradient pattern of genetic divergence for the intercept of the reaction norm (where plastic and genetic effects are in the same direction) always evolves in simulations with assortative mating, These last predictions do not only provide new explanations for observed patterns of phenotypic and genetic variation of phenology along environment gradients, but also suggest caution when inferring the adaptive value of plasticity from these patterns.