Twenty-one male patients with active duodenal ulcer underwent hourly 24-hr gastric acid collections under controlled, calorically deprived conditions. The 24-hr hourly acid secretory output for the group displayed a statistically significant (p < 0.001) rhythm, with peak rates occurring during the evening hours and low rates during the early morning hours, by population-mean cosinor statistical analysis. Population-mean cosinor analysis also verified the occurrence of a significant (p=0.034) circadian rhythm in unstimulated acid secretion in a group (N=14) of healthy male subjects similarly studied and reported previously. In contrast, population-mean cosinor analysis confirmed the absence of any detectable circadian rhythm in unstimulated acid secretion in a group (N=17) of post-vagotomy and pyloroplasty patients studied 2-11 years after surgery. Population-mean cosinor analysis of 4-hr plasma gastrin determinations, obtained in all groups during the 24-hr gastric acid collection, revealed an absence of any detectable circadian rhythm in plasma gastrin. This latter finding is compatible with the interpretation that the circadian rhythm of unstimulated gastric acid secretion, observed in the clinically healthy and active ulcer groups, is unrelated to changes in plasma gastrin levels. The employment of quantitative chronobiological inferential statistical techniques is important to the analysis of any time-dependent measurement in gastrointestinal function, of which gastric acidity is one example.