On June 13, 2007, the public health authority informed the Austrian Agency for Health and Food Safety about 40 children from two neighboring elementary schools who had fallen ill with abdominal cramps and vomiting on June 8. School milk products consumed on June 8 were suspected as the source of the outbreak. On June 8, the milk products provided by local dairy X to eight elementary schools and two nurseries. The short incubation period - all cases fell ill on the day on which the products were consumed - and the short duration of illness (1-2 days) strongly suggested intoxication. In order to identify the causative pathogen, its reservoir and the mode of transmission, a descriptive-epidemiological and microbiological investigation and a retrospective cohort study were conducted. Six of the 10 institutions served by dairy X completed questionnaires on demographics and food consumption. One school had a 79% response rate (203/258) and was chosen as the basis for our cohort study. A total of 166 of the 1025 children (16.2%) at the 10 institutions fulfilled the case definition. Consumption of milk, cacao milk or vanilla milk originating from dairy X was associated with a 37.8 times higher risk of becoming a case (95% CI: 2.3-116.5). Unopened milk products left over at the affected institutions yielded staphylococcal enterotoxins A and D. Six out of 64 quarter milk samples from three of 16 cows producing milk for dairy X tested positive for S. aureus. The isolates produced enterotoxins A and D, yielded genes encoding enterotoxins and D, and showed spa type t2953. S. aureus isolated from the nasal swab of the dairy owner harbored genes encoding enterotoxins C, G, H and I, and showed spa type t635. Our investigation revealed that the milk products produced in dairy X on June 7 were the source of the outbreak on June 8. The cows - not the dairy owner - the likely reservoir of the enterotoxin-producing S. aureus. From the risk assessment of the production process at the dairy, we hypothesize that staphylococcal toxin production took place during a 3-day period of storage of pasteurized milk prior to repasteurization for the production batch of 7.