Peripheral blood samples were collected from 128 unrelated, healthy individuals of Serbian origin (including minorities). Informed consent was given by all participants prior to their inclusion in this study. DNA was extracted according to the protocol of Sambrook et al. [5]. Amplification of APOB, D1S80, HUMTH01 and HUMVWA was performed using original primers and protocols [6-9]. The PCR products were separated by electrophoresis on 6% native polyacrylamide gel (APOB), 8% native polyacrylamide gel (D1S80), 6% denaturing polyacrylamide gel (HUMTH01 and HUMVWA), and visualised by silver staining [10]. The designation of allels was made in comparison with allelic ladders, and allelic frequencies were estimated by gene counting. The nomenclature of alleles corresponds to the number of repeats they contain [7,8,11,12]. The global c² test was performed to check the confirmity of the observed genotype frequencies with their respective Hardy-Weinberg expectations [13]. The p value of 0.05 was taken as a statistical significance limit. In the case of the D1S80 locus, all expected values less than 0.01 were pooled to form a residual group to minimize the large c² values resulting from sampling events. The forensic usefulness of the analyzed systems was measured using two different statistics: the likelihood of coincidental match and the power of exclusion [14].