Reports of elevated nasal nitric oxide (NO) levels in allergic rhinitis suggest that nasal NO levels could be a valuable marker of upper airway inflammation, provided that the reproducibility of nasal NO measurement is acceptable. The aims of this study were to evaluate the precision with which nasal NO levels can be measured at single point in time, and to quantify within-day and between-day variation. Nasal NO was measured using a modified chemiluminescence analyzer. Population data were normally distributed, as judged by testing for skewness and kurtosis. NO levels were not related to age or gender, and there was no evidence of diurnal variation. Sampling rates of 250 mL/minute and 500 mL/minute at a single point in time had acceptable reproducibility (coefficients of variation 10.2% and 6.6%, respectively). However, within-day variation (coefficient of variation 13.4%) and between-day variation (coefficient of variation 11.8%), at a sampling rate of 500 L/minute, were substantially higher. These findings highlight the importance of taking measurement variation into account, in the interpretation of NO levels, in clinical research and, potentially, in routine practice. In individual patients, an alteration of 20-25% in NO levels is required, to ensure that change is genuine and not ascribable to the noise of measurement.