Simple formulae are developed which define the effective size (Ne) of populations with overlapping generations, and their use is illustrated using data from a squirrel population. Two mating systems are considered, the random union of gametes and monogamy, in combination with age-independent fecundity. In the simplest case of age-independent (type 2) survivorship in a population of N adults, Ne = N/(2-T-1) where T is the generation time. As T increases, Ne declines asymptomatically to N/2. A generalization of this result (Ne = N/[1 + k-1-T-1], where k influences survivorship) shows that given type 1 survivorship (k greater than 1) this decline in Ne is less severe. A biased sex ratio results in Ne differing between the two mating systems; however, in both systems, a sex ratio bias resulting from survival differences has much less influence on Ne than a sex ratio bias resulting from recruitment differences. Low fecundity can increase Ne, but realistic levels of variation among breeding individuals (Poisson or greater) negate the effect. The effect on Ne of variation resulting from the presence of non-breeders is also considered.