The Baltimore Eye Survey identified race, intraocular pressure, systemic hypertension, perfusion pressure, family history, and quite possibly the structure and organization of the optic nerve head as potential etiologically significant risk factors in the development of glaucomatous optic nerve damage. These findings have been confirmed by subsequent, sometimes equally rigorous population-based studies in other communities and countries. The Baltimore Eye Survey failed to demonstrate a role for diabetes but revealed that estimates of the significance of diabetes, like that of family history, are subject to considerable selection bias.