In general, the attachment of human hair sheath is considered as one of the most important factors in detecting STR genotypes from single hair samples. Thus, it is difficult to genotype STRs from naturally shed human hair samples using AmpFℓSTR Identifiler PCR Amplification Kit. In this study, we examined the conditions involved in determining STR genotypes from a single hair root of naturally shed human hair focusing on two kinds of commercially available DNA extraction kits; a QIAamp DNA Micro kit (Micro kit) and an ISOHAIR kit (ISOHAIR kit), and the relationship between the concentration of extracted DNA and the numbers of STR loci genotyped.   As a result, based on the protocol of DNA quantification adopted in the police labs in Japan, the DNA amounts were sufficient to quantify in approximately 28% or 36% of 72 naturally shed human hair samples using a Micro or an ISOHAIR kit, respectively. There was no significant difference in genotyping STRs between the two extraction methods. If a quantitative value was estimated based on a Ct (cycle threshold)-value, the extracted DNA was duplicated to amplify by 28 PCR cycles. Alternatively, if a quantitative value showed “ND (not detected)”, the extracted DNA was performed by duplicate PCR amplification for 32 cycles.   Consequently, in the case of a 28 cycle-amplification, when a quantitative value was more than 0.04 ng/μl, the genotyping result was obtained accurately at almost all loci by both extraction methods. On the other hand, in the case of a 32 cycle-amplification, there was a tendency that more accurate genotypes are obtained by a smaller size of an amplicon except a few loci. It indicated that the accurate genotyping rate should not depend only on the size of PCR products.   Therefore, when a DNA sample shows “ND” as an estimated value, more careful interpretation should be necessary to genotype by increasing PCR cycles to 32, or more strict making decision not to proceed to any PCR amplification should be performed.