The intelligibility of interaurally alternated speech was measured by two methods: shadowing and subjective estimation. A passage of connected discourse was recorded at two speaking rates, normal (135 wpm) and fast (194 wpm). Shadowing scores and estimations of intelligibility for normal-rate speech were nearly 100% at eight rates of alternation, ranging from 1.0 to 10.0 alternations per second. Estimated intelligibility for fast-rate speech was depressed slightly and shadowing scores approximated 70% for most rates of alternation. In a second experiment a court stenographer transcribed the message, a form of nonvocal shadowing. Although transcription scores varied somewhat with rate of alternation, there was no evidence of a uniquely critical rate at which intelligibility was impaired seriously. The results suggest that (1) interaural alternation does not affect intelligibility appreciably unless the speech is spoken at a very fast rate; (2) no particular alternation rate or region of alternation rates is associated with significantly depressed shadowing scores; (3) estimations of intelligibility made immediately after shadowing are lower than estimations not preceded by shadowing, suggesting that shadowing may interfere with a listener's ability to understand a message.