Possible contamination by Staphylococcus aureus of the production environment and of the meat of a canned meat production factory was analysed. A total of 108 samples were taken from nine critical control points, 13 of them were positive for S. aureus. None of the isolates produced enterotoxins. To determine how much time can elapse between can seaming and sterilisation in the autoclave without any risk of enterotoxin production by S. aureus, the growth and enterotoxin production of three enterotoxin A producing strains of S. aureus (one ATCC strain and two field strains) in canned meat before sterilisation was investigated at three different temperatures (37, 20 and 10 °C). Two types of meat were used, one with and one without sodium nitrite. In the canned products, the spiked bacteria spread throughout the meat and reached high levels. Enterotoxin production was shown to start 10 hours after incubation at 37 °C and after 48 h after incubation at 20 °C; the production of enterotoxin was always detected in the transition between the exponential and the stationary growth phase. At 10 °C, the enterotoxin was never detected. The statistical analysis of the data showed that the difference between the two different types of meat was not statistically significant (p value > 0.05). Since it is well known that following heat treatment, staphylococcal enterotoxins, although still active (in in vivo assays), can be undetectable (loss of serological recognition) depending on the food matrix and pH, it is quite difficult to foresee the impact of heat treatment on enterotoxin activity. Therefore, although the bacteria are eliminated, the toxins may remain and cause food poisoning. The significance of the results of this study towards implementing good manufacturing practices and hazard analysis critical control points in a canned meat factory are discussed with reference to the management of pre-retorting steps after seaming.