Using data from England and Wales, we test the hypothesis that legalizing abortion reduces crime. The timing of changes in crime rates in aggregate data is generally inconsistent with this hypothesis. Using panel data on recorded crime from 1983 to 2001, we are able to replicate the negative association between abortion rates and reported crime that J. J. Donohue and S. D. Levitt found for the United States. However, this association breaks down under the scrutiny of robustness checks and is not present when we examine data on convictions broken down by age. Overall, we find no clear, consistent relationship between abortion and crime in England and Wales.