A central issue in life-history studies is the extent to which organisms are ‘capital’ versus ‘income’ breeders (i.e. using stored resources vs. current food intake). Snakes are primarily capital breeders, but income during vitellogenesis can also contribute to reproductive output. In Manitoba, the garter snake Thamnophis sirtalis has a short active season, with mating occurring upon emergence from hibernation and vitellogenesis shortly thereafter. Given the brevity of this sequence, animals in these populations should be almost exclusively capital breeders. Consistent with this prediction, in this experiment the reproductive output of recently mated snakes was influenced by mass of the mother at the time of mating but not by her food intake shortly after mating. Food eaten during pregnancy contributed significantly to postpartum mass but not to litter mass. However, the initial mass of snakes influenced both postpartum and litter mass. For snakes given a high ration of food, a female's relative mass (i.e. adjusted for body length) significantly influenced the actual amount she consumed (relatively lighter snakes ate more). Perhaps most important, among pregnant snakes, those that were initially relatively heavier gave birth substantially and significantly earlier. Thus, snakes that are relatively heavy when they mate may gain further fitness benefits following pregnancy (e.g. more time to acquire resources and for offspring to grow before winter), which can also be interpreted as indirect effects on reproduction. These life-history variations may be relevant to population dynamics.