The aim of the present study was to construct and compare two caloric test protocols, one for water irrigation, and one for air insufflation. A set of reference data was constructed and tabulated as well as the intersubject variability. The effect of age, sex, ear, and temperature, as well as a possible priming effect and order effect were investigated. Forty-seven subjects (18-58 years) without otological or vestibular history participated. Four response parameters were investigated: slow component velocity (SCV), frequency, unilateral weakness (UW), and directional preponderance (DP). Statistically higher SCV values were obtained for water compared to air, with statistically higher standard deviations for SCV water values. No influences of age, sex, ear, or temperature could be demonstrated on any of the response parameters. The same applied for the presence of an order effect and a priming effect. Comparing the two protocols to one another led the present authors to favour water as the standard irrigation medium, and air only in situations where water is contra-indicated.