The microbiology of domestic kitchens in the homes of subjects who had suffered sporadic Salmonella infection (cases) was compared with control domestic kitchens. Case and control dishcloths and refrigerator swabs were examined for the presence of Salmonella spp., total Enterobacteriaceae counts and total aerobic colony counts. Salmonella spp. were isolated from both case and control dishcloths and refrigerators but there were no significant differences between the two groups. Colony counts were similar in case and control dishcloths and refrigerator swabs. There was no relationship between the total counts and presence of Salmonella. There was no evidence that cases of Salmonella infection were more likely to have kitchens which were contaminated with these bacteria or have higher bacterial counts than controls. Total bacterial counts were poor indicators of Salmonella contamination of the domestic kitchen environment. Further factors which could not be identified by a study of this design may increase risk of Salmonella food poisoning. These factors may include individual susceptibility of the patient. Alternatively, sporadic cases of Salmonella food poisoning may arise from food prepared outside the home.