The urinary excretion patterns of theophylline metabolites were studied in a subject on a routine, oral dose of 600 mg/day. The highest correlation was observed between the excretions of 1,3-dimethylurate and 1-methylurate (r = 0.73 +/- 0.08). The poorest correlations were observed when the excretion of 1-methylxanthine was compared to those of 1-methylurate and 1,3-dimethylurate (r < or = 0.55, P > or = 0.05). The difference in the quality of the correlations was not due to rate-limiting or class-specific carries. The results suggested that 1-methylurate did not derive solely from 1-methylxanthine, implicating 1,3-dimethylurate as an alternative source. When the theophylline regimen was supplemented by a single dose of caffeine (4.80 mg/kg), the recovery of unaltered caffeine and the yield of metabolites arising from the primary 3-demethylation of caffeine (1-methylxanthine, 7-methylxanthine, 1-methylurate, 1,7-dimethylxanthine and 1,7-dimethylurate) were found to be unchanged. The lack of competition between caffeine and theophylline indicated that caffeine 3-demethylation and the demethylations of theophylline are not catalyzed by the same cytochrome P450 system.