Tobacco-specific nitrosamines are believed to play a significant role as causes of cancer in people who use tobacco products. Whereas the uptake of one tobacco-specific nitrosamine, 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone, has been shown by analysis of its metabolites in urine, there are no published studies on urinary levels of N′-nitrosonornicotine (NNN), N′-nitrosoanatabine (NAT), and N′-nitrosoanabasine (NAB) or their metabolites in human urine. We developed a method for quantitation of NNN, NAT, NAB, and their pyridine-N-glucuronides NNN-N-Gluc, NAT-N-Gluc, and NAB-N-Gluc in human urine. Total NNN (NNN plus NNN-N-Gluc) was assayed using 5-methyl-N′-nitrosonornicotine as internal standard. Urine was treated with β-glucuronidase. Following solvent partitioning and solid-phase extraction, total NNN was determined using gas chromatography with nitrosamine-selective detection. Total NAT and total NAB were quantified in the same samples. Separate quantitation of NNN and NNN-N-Gluc was accomplished by extraction of the urine with ethyl acetate before β-glucuronidase hydrolysis; NNN was analyzed in the ethyl acetate extract, and after enzyme treatment, NNN released from NNN-N-Gluc was quantified in the extracted urine. Separate analyses of NAT, NAT-N-Gluc, NAB, and NAB-N-Gluc proceeded similarly. Analyte identities were confirmed by gas chromatography-tandem mass spectrometry. Mean levels of total NNN, NAT, and NAB in the urine of 14 smokers were (pmol/mg creatinine) 0.18 ± 0.22, 0.19 ± 0.20, and 0.040 ± 0.039, respectively, whereas the corresponding amounts in the urine of 11 smokeless tobacco users were 0.64 ± 0.44, 1.43 ± 1.10, and 0.23 ± 0.19, respectively. Pyridine-N-glucuronides accounted for 59% to 90% of total NNN, NAT, and NAB. The results of this study show the presence of NNN, NAT, NAB, and their pyridine-N-glucuronides in human urine and provide a quantitative method for application in mechanistic and epidemiologic studies of the role of tobacco-specific nitrosamines in human cancer.