Human norovirus (NoV) is the leading cause of acute viral gastroenteritis and a major source of foodborne illness. Detection of NoV in food and environmental samples is typically performed using molecular techniques, including real-time reverse transcription polymerase chain reaction (RT-PCR) and less frequently, nested real-time PCR. In this study, we conducted a controlled comparison of two published NoV detection assays: a broadly reactive one-step real-time RT-PCR and a two-step nested real-time PCR assay. A 20% human fecal suspension containing a genogroup II human NoV was serially diluted, genome extracted, and subjected to amplification using the two assays compared via PCR Units. Additional amplicon confirmation was performed by dot blot hybridization using digoxigenin (DIG)-labeled oligonucleotide probes. Both assays displayed similar amplification standard curves/amplification efficiencies; however, the nested assay consistently detected one log.sub.10 lower virus. Dot blot hybridization improved the detection limit of the nested real-time PCR by one log.sub.10 NoV genome copies but impaired the detection limit of the one-step real-time RT-PCR by one log.sub.10 NoV genome copies. These results illustrate the complexities in designing and interpreting molecular techniques having a sufficient detection limit to detect low levels of viruses that might be anticipated in contaminated food and environmental samples.