Investigating the unaccounted ones: insights on age-dependent reproductive loss in a viviparous fly

Most empirical and theoretical studies on reproductive senescence focus on observable attributes of offspring produced, such as size or postnatal survival. While harder to study, an important outcome of reproduction for a breeding individual is whether a viable offspring is produced at all. While prenatal mortality can sometimes be directly observed, this can also be indicated through an increase in the interval between offspring production. Both direct reproductive loss and presumed losses have been found to increase in older females across several species. Here, we study such reproductive loss (or "abortion") in tsetse, a viviparous and relatively long-lived fly with high maternal allocation. We consider how age-dependent patterns of abortion depend on the developmental stage of offspring and find that, as per previous laboratory studies, older females have higher rates of abortion at the late-larval stage, while egg-stage abortions are high both for very young and older females. We track the reproductive output of individual females and find little evidence that experiencing an abortion is an adaptive strategy to improve future reproductive outcomes. After an abortion, females do not generally take less time to produce their next offspring, these offspring are not larger, and there is no sex bias towards females, the sex with presumed higher fitness returns (being slightly larger and longer-lived than males, and with high insemination rates). Abortion rates are higher for breeding females experiencing stress, measured as nutritional deprivation, which echoes previous work in tsetse and other viviparous species, i.e., humans and baboons. We discuss our results in the context of studies on reproductive loss across taxa and argue that this is an important yet often overlooked reproductive trait which can vary with maternal age and can also depend on environmental stressors.