Two methods are described for determination of the microphonic generator transfer characteristic using data from bitonal stimulation experiments. In one method, the filtered high‐frequency microphonic output component is measured at the extreme points of the modulation envelope for a sequence of intensity levels of the lower‐frequency component. This method is applied to data derived from a modulation test procedure making use of a wide range of test intensities for both the high‐ and low‐frequency components and also to data generated in a test using a single sequence of stimuli with increasing intensity of the lower frequency. In a second method, the modulation envelope of the higher‐frequency microphonic response to a bifrequency acoustic stimulus is used to determine a segment of the generator transfer characteristic. The results of a sequence of determinations for different lower‐frequency intensities show good agreement with each other and with the transfer characteristic derived by the first method. However, the second method shows that there is an appreciable level‐dependent hysteresis, the mechanism for which has not been determined. The transfer characteristics determined would give rise to a pronounced dc rectification component in the direction of positive summating potential.