Roast beef preparation practices were examined in eight foodservice establishments for the likelihood of contamination and the possibilities of survival or growth during each step of the operations. Clostridium perfringens was isolated from raw beef, equipment, and cooked beef. Staphylococcus aureus was isolated from raw beef, equipment, workers' hands, and cooked beef. Salmonellae were isolated from neither meat nor equipment. Numerous opportunities were observed for contamination of cooked beef during operations in most of the establishments. No opportunities for multiplication of foodborne disease bacteria were observed during thawing of frozen beef. From recorded time-temperature data, it was calculated that vegetative foodborne pathogens could survive in 76% of the geometric centers and on 5% of the surfaces of beef during cooking. Survival of these organisms could occur in 36% of the geometric centers and on 11% of the surfaces of the cooked beef during post-oven temperature rise periods. These organisms could have survived in 25% of the geometric centers and on 33% of the surfaces of the cooked roasts during hot holding; they could have multiplied on 25% and 27%, respectively. During cooling, the potential for multiplication of vegetative cells of foodborne pathogens existed in 83% of the geometric centers and 79% of the surfaces of the roasts. During reheating, these organisms would have survived in 71% of the geometric centers and on 13% of the surfaces of the roasts. Recommendations are given for hot holding, cooling, and reheating so as to minimize microbiological problems.