Normal limits for the response to caloric irrigation typically are specified as percentage of reduced vestibular response (RVR) and directional preponderance (DP). These measures are calculated using Jongkees' formula without regard to which ear is stimulated first. To assess the influence of the order in which the caloric stimuli are delivered we reviewed the responses to closed loop caloric irrigation of two groups of normal subjects and a group of unselected patients. Results indicated a consistent declining trend in response from the first caloric administration to the last administration in all groups and subgroups except in male patients over 60 years of age. Consequently, a significant bias in RVR was observed. Specifically, with the order of irrigation left cold, right cold, left warm, right warm, the mean percentage RVR for both the patients and control subjects was about a 5% right RVR. The bias resulted in right-sided unilateral hypofunction being diagnosed 2.4 times more often than left-sided hypofunction in patients aged 16-59. Our results differ from the previous literature that describes the effect of order on the response to caloric irrigation; this discrepancy may be related to our use of a closed-loop system for caloric irrigation. We conclude that vestibular laboratories should account for the effect of order in their patients' responses to caloric stimulation. We offer suggestions on how Jongkees' formula can be corrected for the effect of order.