Human cortical evoked potentials under conditions of stimuation are registrated in the post-stimulatory phase of a five minutes lasting equally masking white noise (90 dB HL). Changes of the evoked potentials during adaptation, possible analogy with high tone losses after noise representation and the origin of tinnitus are examined. Stimulation was started 3 sec after the off-effect of the noise. For five minutes periodically tone bursts were represented. Each train of stimulation consists of tone bursts of three frequencies: 2 kcs, 4 kcs, 8 kcs. The 0.5 sec lasting tones were separated by pauses of 2 sec. During the experiment stimulation and analysis were controlled by a computer. Changes in latency and amplitudes of the cortical evoked potentials were registered. Changes of the adaptation patterns as a function of the poststimulatory time are discussed.