1. We analyse 7 years of field data on the mating system of a small population of snakes (adders, Vipera berus) to quantify the relative contribution of factors, such as longevity, fecundity and male mating success rates, to overall lifetime variance in reproductive success. 2. Data were available on 32 males and 40 females that attained reproductive age. Litter sizes were known for females, but our analyses on males relied on mating success rather than number of viable progeny. We thus underestimate the variance in male reproductive success because offspring viability is positively correlated with male lifetime mating success and we do not include variance due to differences among males in the numbers of progeny fathered per mating, or the viability of their offspring. 3. In both sexes, about half of the variance in lifetime reproductive success was due to whether or not the individual survived to breed (i.e. juvenile mortality). In females, three factors (longevity, mean litter size and the interaction between these two components) each explained about one-third of the remaining variance. Longevity and mean litter size showed positive covariation (snakes that reproduced more often tended to produce larger litters), because litter sizes increase with body size (and thus, with age and with number of reproductive episodes in the lifespan). 4. Variation among males in mating rates per season was a more important influence on lifetime mating success than was longevity (72% versus 25% of variance), and there was strong negative covariation (-28%) between these two factors. This negative covariation suggests that male adders experience high mortality costs of reproduction, a prediction consistent with other direct evidence on the causes of male mortality. 5. Overall, adders resembled other polygynous vertebrate species in their partitioning of the components of lifetime reproductive success.