This is a discursive account of selected methodological issues encountered in attempts to measure fertility with data collected in a cross-sectional survey of women selected by marital status. The focus of the account is the dependent variable itself, more specifically the ways in which that variable can be refined by collecting progressively more detailed accounts of reproductive behavior. The price paid for this emphasis is neglect of obviously important and difficult questions about the appropriate explanatory framework for that behavior, once it has been properly quantified. Another kind of inadequacy in the account is that it presents far fewer solutions than problems, although that is probably not an unhealthy state for a relatively new and complex discipline.