After 96 administrations of metronidazole to 36 subjects, it was found that the clearance could be determined from one plasma sample, the dose, and a volume of distribution estimated from sex, age, body weight, and height, without loss of precision and accuracy compared with conventional clearance determinations (r greater than 0.97). In 230 sample pairs the plasma and saliva concentrations of metronidazole were identical (r = 0.99). In 119 subjects the one-sample clearance of metronidazole was unimodally distributed. Body weight (r = 0.28) and the alcohol consumption (r = 0.23) correlated with the metronidazole clearance. In the same subjects the consumption of tobacco (r = 0.28), alcohol (r = -0.19), coffee/tea (r = 0.27), age (r = -0.24), and sex (r = 0.28) correlated with the antipyrine clearance. The clearances of metronidazole and antipyrine were correlated (r = 0.34). The differential influence of the environmental factors on the elimination rates supports differential metabolism of metronidazole and antipyrine.