Since the publication of Galambos and Davis″ work, electrophysiological data have been accumulated which show that the analysis of sound is performed in a special way in the cochlea. The examination of discharge patterns and the response areas of single primary auditory neurons in the cochlear nerve bundle reveals that they are sending enough informatoion about intensity of sound, but not about the frequency of analysed sound. This fact recalls to us Rutherford's telephone theory which had once been supplanted by Helmholtz's place theory of the cochlear function. The telephone theory tells us that the frequency analysis of sound is not performed in the cochlea at all, but it is done in the auditory cortex. This hypothesis seems in a sense to be valid for us, because the primary auditory neurons send only very poor information about the frequency of sound in spite of the fact that cats and monkeys and, especially, human beings are remarkably capable of discriminating the freqency of sound. In order to explore this problem, the present author tried to record the responses of neurons with capillary microelectrodes filled with 3M-KCI solution from several relay nuclei along the auditory tracts in the brains of cats and monkeys.