ICMSF methods studies. VII. Indicator tests as substitutes for direct testing of dried foods and feeds for Salmonella.

The correlation between the occurrence of indicator organisms and salmonellae in the same sample homogenate of various classes of dried foods was studied. In the first phase of the work involving 5837 samples, 1943 contained fecal coliforms and 100 contained Salmonella. Fecal coliforms were not detected in 53 of the Salmonella-positive samples. In the second phase, an additional 637 samples were tested for the presence of Enterobacteriaceae, coliforms, fecal coliforms, and Salmonella. Enterobacteriaceae were detected in 538 samples, coliforms in 506, fecal coliforms in 343, and Salmonella in 49. All Salmonella-politive samples were also positive for Enterobacteriaceae and coliforms. However, 6 of 49 Salmonella-positive samples were negative for fecal coliforms. Therefore, the use of the fecal coliform test as an index of Salmonella contamination would result in the rejection of Salmonella-negative foods and the acceptance of Salmonella-positive foods. It is concluded that none of these indicator groups are reliable as an index of Salmonella contamination in dried foods.