In any given population, an increase in the proportion of repeat abortions, and a repeat-abortion rate that is substantially higher than the first-abortion rate, can be anticipated to occur over a number of years after abortion has been legalized. These are the findings of an analysis of repeat abortion for a population of women aged 15--44, using simple mathematical models based on the assumptions that the risk of abortion is constant over time and independent of prior abortion experience. The data generated by the models explain both the rapidly increasing percentage and the high rate of repeat abortion following legalization and refute assertions that making abortion widely available discourages the practice of contraception.