The purpose of this research was to compile data for use as related information in the education of drivers, and to derive a formula for computing a hazard index for rail-highway intersections. Data for the study were compiled from: (1) all crossings on which field data were collected, (2) reports of 642 accidents, and (3) data collected from individual railroad companies. The hazard-index formula was derived from a 5-year historical record of 140 crossings where one or more accidents occurred between January 1, 1962 and December 31, 1966, and from 140 crossings where no accidents were recorded during the same interval. One hypothesis which was tested, but rejected, stated that vehicle frequency, train frequency, and speed have no effect on the accident rate at grade crossings. Among the conclusions are that: (1) a high rate of speed is not a prerequisite for becoming involved in a railway-highway accident, (2) Tractor-trailer trucks have a more severe accident rate at grade crossings than single-unit trucks and automobiles, and (3) The type of protective device installed at a grade crossing affects the accident potential at the intersection. (Author)