Abstract
Discriminant analysis of patterns of antibiotic resistance in fecal streptococci was used to differentiate between human and animal sources of fecal pollution in natural waters. A total of 1,435 isolates from 17 samples of cattle, poultry, human, and wild-animal wastes were obtained, and their ability to grow in the presence of four concentrations of five antibiotics (chlortetracycline, halofuginone, oxytetracycline, salinomycin, and streptomycin) was measured. When the resulting antibiotic resistance patterns were analyzed, an average of 74% of the known isolates were correctly classified into one of six possible sources (beef, chicken, dairy, human, turkey, or wild). Ninety-two percent of human isolates were correctly classified. When the isolates were pooled into four possible categories (cattle, human, poultry, and wild), the average rate of correct classification (ARCC) increased to 84%. Human versus animal isolates were correctly classified at an average rate of 95%. Human versus wild isolates had an ARCC of 98%, and cattle versus poultry isolates had an ARCC of 92%. When fecal streptococci that were isolated from surface waters receiving fecal pollution from unknown origins were analyzed, 72% of the isolates from one stream and 68% of the isolates from another were classified as cattle isolates. Because the correct classification rates of these fecal streptococci are much higher than would be expected by chance alone, the use of discriminant analysis appears to hold promise as a method to determine the sources of fecal pollution in natural waters.