PCR and direct DNA sequencing methods were used to analyse the prevalence of mutations in exon 2 of the p21waf1 gene in 14 oral squamous cell carcinomas (OSCCs) and 8 non-malignant oral mucosal lesions from Sudanese toombak dippers. For comparison, OSCCs (14 from the Sudan, 16 from Norway, 11 from Sweden, 21 from the USA and 14 from the UK) and non-malignant oral mucosal lesions (3 from the Sudan) from non-snuff-dippers were included. The prevalence of mutations in exons 2 & 3 of the S100A4 gene were analysed in the 14 OSCCs from toombak-dippers and in 25 cases of OSCCs from the control non-snuff-dippers. Of the 14 OSCCs investigated from toombak-dippers, mutations in the p21waf1 exon 2 were found in 43% (6 out of 14), compared to 14% (2 out of 14), 22% (6 out of 27) and 14% (5 out of 35) found in those from non-snuff-dippers from the Sudan, Scandinavia and the USA/UK, respectively. OSCCs from toombak-dippers showed 13 different mutations distributed as 10 (77%) transitions and 3 (23%) transversions. OSCCs from non-snuff-dippers from the Sudan, Scandinavia, the USA and the UK showed 33 different mutations distributed as 14 (42%) transitions and 19 (58%) transversions. In the OSCCs examined, cases with mutations in the p21waf1 also had p53 gene mutations. Only exon 2 of the S100A4 gene was found mutated in 3 cases of OSCCs (one from a toombak-dipper and two from the non-snuff-dippers). The toombak-dipper OSCC had 4 mutations (one transition, 3 transversions), compared to the OSCCs from non-snuff-dippers which showed 3 mutations each (one transition, 2 transversions). All these 3 cases were negative for mutations in the p21waf1 and p53 genes. No mutations of p21waf1 or S100A4 were found in the non-malignant oral mucosal lesions from the snuff-dippers/non-dippers. These findings suggest that; (i) p21waf1, together with p53, is a target gene of oral carcinogenesis in OSCCs from toombak-dippers, with the tobacco specific nitrosamines present in toombak possibly acting as principal carcinogens in these OSCCs; (ii) findings of p21waf1 exon 2 mutations in the OSCCs unrelated to snuff use further demonstrate that this gene may play an important role during the pathogenesis of OSCCs caused by smoked tobacco use; (iii) mutations in the S100A4 gene are rare in OSCCs, but appears to be complementary to p21waf1 and p53 mutations. Since molecular analysis of OSCCs can provide clues to endogenous or environmental factors contributing to the high risk of OSCCs, further analysis of the role of the p21waf1 gene mutations as a biomarker of malignant transformation, which is linked to the p53 gene, is necessary, especially in habitual users of toombak from the Sudan.