The purpose of the paper is to offer methods for use in measuring abortion rates and in undertaking abortion diagnostic investigations. In addition, some of the associations found between demographic and reproductive features of the dam and subsequent abortion will be described. The cohort life table method calculates the risk of fetal loss (abortion), for pre-defined time intervals (eg. days or weeks) during the gestation period, and the overall proportion of cows that abort. In contrast, the abortion density method calculates abortion risk for a defined calendar time (eg. per month) as the number of abortions per cow-days-at risk. Both methods are standardized to account for the changing number of cows at risk of abortion, as a result of culling, death, abortion, and calving, thus tend to produce higher, but more realistic, estimates for abortion than rates typically calculated. These measures and statistical modeling have been used to estimate the expected magnitude of abortion for dairies (ranging from 8% to 19%) and to identify associations between demographic or reproductive features of the dam and subsequent risk of abortion. Some of the factors to be discussed include effects of a previous abortion, dam gravidity and age, days open at the time of conception, and season. Analytic methods also will be presented for herd-based diagnosis of abortion related to infectious agents or other exposures. These methods permit estimates of the risk or proportion of abortions attributable to exposure to an infectious agent or to other putative abortifacient exposures.