Endogenous formation of N-nitrosodi-n-butylamine (NDBA) was studied in rats after administration of sodium nitrite or sodium nitrate and N,N-dibutylamine (DBA) by monitoring the urinary excretion of NDBA and its metabolites, N-butyl-N-(4-hydroxybutyl)-nitrosamine (BBN) and N-butyl-N-(3-carboxypropyl)nitrosamine (BCPN). Animals were given sodium nitrite (0.2%) or sodium nitrate (0.5%), dissolved in the drinking-water. This treatment was started 24 h before DBA administration and was continued throughout the experiment. Animals were fasted overnight before receiving DBA, which was administered by gavage as three doses of 50 mg/kg, 8 h apart; 24-h urine samples were collected on ammonium sulfamate. NDBA, BBN and BCPN were extracted and analysed by GC-TEA, according to a method previously described. Under the experimental conditions reported, NDBA and BBN (free or glucuronic acid-conjugated) were not detected in the urine of animals given nitrite or nitrate and DBA, but the presence of BCPN indicated that N-nitrosation had occurred in both groups of animals. These results suggest that, when studying nitrosamines that are extensively metabolized, quantitative analysis of urinary metabolites is a better indicator of nitrosamine exposure than measurement of nitrosamine itself.