Objectives: These were to determine (1) whether there is a birth rank effect in eating disorders, and (2) whether all-female sibships are overrepresented in the families of those with eating disorders. Method: The study sample consisted of 293 female patients referred from a defined catchment area, the County of Leicester, United Kingdom, to an eating disorders clinic. Diagnoses were made using DSM-III and DSM-III-R. To discern birth rank effect, this core sample was expanded to one of 673 by adding published data sets to our own. Results: No evidence for either effect was found. Discussion: Biases which may have obscured genuine effects are discussed. A comment is made on the implication for family theories of causation of these negative findings. © 1997 by John Wiley & Sons, Inc. Int J Eat Disord 21: 335–340, 1997.