Audio steganography is a technique for hiding messages in audio signals such that no one except the sender and intended recipient suspects the existence of the messages. This paper proposes a new method for stereo audio steganography that exploits some characteristics of the human auditory system (HAS). Messages are embedded by inserting low-frequency and octave-equivalent pure tones into different channels. Hidden messages are extracted by comparing the frequency domain data of the left channel with those of the right channel. Experimental results reveal that the quality of the message-hiding audio that is generated by the method is only very slightly less than that of the host audio, so attackers cannot perceive the hidden messages.