Background/Question/Methods A mutation will only alter fitness if it changes an individuals' demography. Making predictions from this simple premise is considerably complicated by the dynamic nature of demography, and feedbacks between demography across individuals within populations. I explore this using the life-history question: when should you flowering if flowering kills you? Results/Conclusions The relatively straightforward empirical system embodied by monocarpic plants permits detailed modeling of the systems' key ecological features including density dependence and environmental stochasticity. From this, expected adaptive dynamics can be inferred. Predictions are in excellent agreement with observed trait distributions. However, questions remain. For example, the persistence of variation in timing of flowering is still mysterious. Is the separation between ecological and evolutionary time-scales as clean as implied by the adaptive dynamics framework? What would the implications of overlap be? I discuss these issues using known patterns in Arabidopsis as a focal system.