There are socioeconomic differentials in abortion attitudes and behaviour in the UK: in more deprived areas individuals and communities are less accepting of abortion; and wards have a lower teenage proportion of conceptions ending in abortion. Evolutionary life history theory predicts that with heightened mortality risk, individuals are disinclined to postpone fertility when young due to following a ‘fast’ life history strategy, echoed by Geronimus (1999; 1992, 1996a, 1996b; 1999), who suggested in the ‘weathering hypothesis’ that young women in an environment of heightened morbidity start childbearing at a younger age for similar reasons. Our research investigates whether mortality/morbidity is related to a) likelihood of young women having an abortion and b) young people’s abortion attitudes. We use a population-level analysis using small-area geographical data to test for initial associations between mortality (life expectancy) and morbidity (long-term limiting illness prevalence) and proportion of conceptions ending in abortion. We then use two randomised experiments with mortality and morbidity primes to determine if they increase disapproval of abortion, in an attempt to get at causality and psychological mechanisms. Results of the geographical analysis show that with heightened mortality/morbidity, there is lower under-25 ‘abortion proportion’, in line with predictions; whereas poorer local health means higher ‘abortion proportion’ in older age bands. The experimental results however show no pro-natalist effect of mortality priming on abortion attitudes. Instead it makes women significantly more likely to report they would have an abortion if pregnant now, compared to controls (although this effect is not replicated in the other experiment). Men saw no significant effects on reproductive motivation from treatments. Morbidity Salience had no effect on participants. Despite population-level associations between health and abortion, there either might be other causal pathways from general deprivation to reproductive timing; or the experimental primes had unintended effects.