The phenomenon known as heteroplasmy can be operationally observed in some human mitochondrial DNA (mtDNA) samples. Typically, heteroplasmy manifests itself in an individual presenting two mtDNA species that differ at a single base. Heteroplasmy at two, and even possibly three sites, also may occur, but at very low rates. A recent report (Grzybowski, 2000, see ref. [13]) suggests that much higher levels of mtDNA (point substitution) heteroplasmy can occur in hair. This observation is contrary to the experience of the forensic mtDNA community. There are several explanations for the unusual findings of high levels of heteroplasmy. First, the template quantities of DNA are approximately three orders of magnitude higher than required for mtDNA sequencing, and an excessive number of amplification cycles were used. Thus, the protocol used did not follow routine practices by the forensic community. Second, there are misidentifications and tabular errors that call into question the reliability of the findings. Third, by comparing the natural human mtDNA variation with a reference sample population with that observed in the heteroplasmy in hair study, the data are inconsistent with population genetic expectations. The observation of high levels of heteroplasmy may be due to contamination of the samples and/or possibly the amplification of nuclear pseudogenes. The results observed in the heteroplasmy in hair study do not apply to other methods of mtDNA analysis and cannot be used to question the reliability of the current forensic mtDNA practices.