D'Andrea, F., F. Fridez, and R. Coqouz. 1998. Preliminary experiements on the transfer of animal hair during simulated criminal behavior. Journal of Forensic Sciences. 43:1257-1258.

Dog hair is the most common animal-derived sample in criminal investigations. Often the hairs have been naturally shed and lack substantial hair roots. Although STR typing is always attempted if the root end of the hair is present, it may be unsuccessful, particularly if only a few hairs are presented as evidence. In that event, mitochondrial DNA (mtDNA) is extracted from the hair shaft. MtDNA is also more easily recovered from blood and tissue samples that have undergone environmental degradation. This is possible because mtDNA is present in hundreds of thousands of copies in each cell (as opposed to nuclear DNA in which there is only one copy per cell). MtDNA is inherited from the mother; animals related through the maternal line will have the same haplotype and can provide a reference sample if needed.

The region of mtDNA analyzed for forensic identification is called the control region, the hypervariable region, or the D-loop. The analysis consists of PCR amplification of the control region in either one large or two smaller sections followed by DNA sequencing of the entire region. The different sequence variants of the control region are termed "haplotypes". QuestGen Forensics has a database of 312 unrelated dogs with 60 unique haplotypes recognized thus far. The chart above shows the distribution of the haplotypes in the database to date. There are a few haplotypes that are found more commonly (with the most common seen in 18% of the dogs in the database) but many haplotypes are rare. As in human mtDNA typing, in the event of a match the convention for reporting the significance is to simply state the number of times the haplotype has been seen in the database. Canine mitochondrial typing has been admitted as evidence in a court case in California (see Closed Casework - State of California v. David Westerfeld).