Delayed life history effects, multilevel selection, and evolutionary trade-offs in the California tiger salamander.

Delayed life history effects (DLHEs) occur when fitness in one life stage affects fitness in subsequent life stages. Given their biphasic life cycle, pond-breeding amphibians provide a natural system for studying DLHEs, although these effects are not restricted to species with biphasic life histories. In this study, we used multiple mark-recapture techniques enabled by a large trapping array to monitor components of fitness and resulting DLHEs in a population of the endangered California tiger salamander (Ambystoma californiense). We found that DLHEs are prominent across all life stage transitions and that there is variation in whether selection acts primarily at the individual or cohort level. We also demonstrated that there is more than an order of magnitude variation in mean cohort fitness, providing tremendous variation for DLHEs to act upon. We documented an evolutionary trade-off between mass at emergence and date of emergence, which may play a role in maintaining the variation in mass (fitness) at emergence. A literature review revealed that such high levels of intercohort variation occur in many other pond-breeding amphibians, and that appropriately documenting the magnitude of intercohort variation requires long-term studies (roughly two population turnovers). Given the profound effect that DLHEs can have on population dynamics, quantifying intercohort variation in mean fitness and the level(s) at which selection acts will be very important for developing accurate models of population dynamics. In general, when developing models of population dynamics, more attention should be paid to variation in mean fitness and not just variation in total numbers.