Vibrio parahaemolyticus densities in spiked and naturally contaminated seafood samples were enumerated by the MPN method combined with a PCR procedure (MPN-PCR method) targeting the species-specific thermolabile hemolysin gene (tlh), and by the MPN method using subcultivation of alkaline-peptone-water (APW) enrichment culture on thiosulfate-citrate-bile-sucrose (TCBS) agar (MPN-TCBS method). In the samples spiked with both V. parahaemolyticus and V. alginolyticus, the numbers of V. parahaemolyticus enumerated by the MPN-PCR method were similar to, or higher than the numbers of spiked cells, whereas those enumerated by the MPN-TCBS method were below the numbers of spiked cells. In naturally contaminated seafood samples, the numbers of V. parahaemolyticus enumerated by the MPN-PCR method were higher than those by the MPN-TCBS method. In the case of the MPN-TCBS method, isolation of V. parahaemolyticus from some APW cultures was difficult because of the overgrowth of many colonies other than V. parahaemolyticus (e.g., V. alginolyticus) on TCBS agar. In contrast, the PCR technique could detect tlh from APW culture without isolation of V. parahaemolyticus, so the possibility of failing to obtain a positive result in APW culture by the MPN-PCR method was considered to be lower than that by the MPN-TCBS method. Furthermore, utilization of the PCR technique reduces the time and labor required for the biochemical identification tests used in the MPN-TCBS method. For the detection and enumeration of V. parahaemolyticus in seafood, especially for samples that show many colonies other than V. parahaemolyticus on TCBS agar, the MPN-PCR method may be more convenient and reliable than the MPN-TCBS method.