Abstract: The occurrence of maternal senescence, the decrease of offspring performance with increasing mother age, varies a lot across the tree of life. There is some evidence of maternal senescence among mammals, but its strength remains weak and varies a lot among offspring traits. However, our understanding of maternal senescence is currently limited, and very few studies to date have investigated the variation in several offspring traits within a given species using populations facing different environmental contexts. We fill this knowledge gap by taking advantage of two detailed long-term monitoring of female roe deer (Capreolus capreolus) in populations subjected to markedly different environmental contexts. At Chizé in southwestern France, roe deer face with limited food resources due to low soil quality and frequent summer droughts, whereas at Trois Fontaines in northeastern France, roe deer benefit from more favorable conditions thanks to high quality soils providing a high forest productivity. From over 100 females monitored throughout their entire life in each population for their reproductive status and the fate and phenotypic quality of their offspring, we tested for a decrease in offspring performance with increasing maternal age. Offspring performance was assessed by early growth, weaning mass, first-year survival, adult body mass, and lifespan. We explicitly tested for offspring sex effect (we expected no between-sex differences in maternal senescence of roe deer due to its low intensity of sexual selection generated by a weak sexual size dimorphism and a low level of polygyny) and for an influence of environmental conditions (we expected maternal senescence to be more acute in females facing strong resource limitation (i.e. born in poor years within a population or at Chizé compared to Trois Fontaines)). Our findings support the existence of a weak but detectable maternal senescence. We found support for the first prediction that maternal senescence was similar in male and female offspring. On the other hand, our findings did not support the expectation of a more acute maternal senescence in the population facing more limited food resources. Increasing experience with maternal age and age-specific changes in reproductive allocation are likely to explain the limited magnitude of maternal senescence and its low susceptibility to varying environmental conditions.
