Diastasis recti abdominis, or midline separation of the abdominal musculature, has not been investigated scientifically. The purposes of this study were to provide data on the incidence and degree of diastasis recti abdominis, to describe the measurement system used, and to determine the interrater reliability of the measurements performed. Forty subjects less than four days postpartum were tested by four raters. All subjects were measured in a supine, flexed-knee position at a standard point of palpation above the umbilicus. During palpation, each subject performed a partial sit-up, and the rater determined the number of finger widths filling the separation. An analysis of variance for repeated measures revealed a highly significant difference between the measurement scores of the four raters. This measurement system, therefore, was found to be unreliable. All subjects had some degree of diastasis recti abdominis; over 60% had separations significant enough to warrant protective exercises. The author proposes that the incidence and degree of diastasis recti abdominis may be underestimated, that selected components of exercise prescriptions may be contraindicated, and that a reliable instrument for measuring the degree of separation is needed.