The aim of this study was to investigate the reproducibility of genetic typing of two tetrameric short tandem repeat (STR) loci and the extent of genetic variation in the Swedish population. An automated, fluorescent-based Applied Biosystems 373A sequencer was used for typing of the HUMTH01 and HUMACTBP2 loci (also named SE33). The former locus has seven alleles in the size range of 154-174 bp, while the latter is a complex locus with more than 32 alleles in the range of 227-316 bp. Using different fluorescent dyes, polymerase chain reaction (PCR) products from the two STR loci were sized in one lane using an internal size standard. In order to compare within- and between-gel reproducibility of fragment size estimates, a control sample was typed three times on each of 20 gels. Within the gel, the standard deviation (SD) of fragment size variability was less than 0.1 bp for four fragment sizes between 158-291 bp. Standard deviations between gels were slightly higher for the two shorter fragment sizes (HUMTH01), while the larger fragments varied between 0.3 and 0.4 bp (HUMACTBP2). The amount of genetic variation was investigated in samples from three Swedish cities (n = 301). Seven alleles were found at HUMTH01 and the observed heterozygosity was 0.77. At the HUMACTBP2 locus more than thirty alleles were found and the observed heterozygosity was 0.96. The observed genotype frequencies at HUMTH01 and HUMACTBP2 did not deviate significantly from Hardy-Weinberg expectations. No indication of a significant excess of homozygotes was found at any of the loci. We conclude that both HUMTH01 and HUMACTBP2 can be reliably typed using the method described. However, the latter locus requires an allelic ladder to be run on each gel.