Auditory temporal ordering was investigated in 5 normal-hearing young adults. Ss made diotic 2IFC judgments for pure tones of 0.5, 1, 2, or 4 kc/s and critical-band noises centered at those frequencies, on duration (standard: 40 msec; variables: 10, 20, or 30 msec), level (standard: 70 db HL; variables: 40, 50 or 60 HL), and frequency (standard: 0.5 kc/s; variables: 1,2, or 4 kc/s). Interstimulus interval (ISI) was 5, 10, 20, 50, 100, or 200 msec. All stimuli were at 70 db HL except of course the variable stimuli in the level judgments. Performance improved with increasing ISI. Mean threshold ISIs (performance at 75%-correct) collapsed across pure tones and noise bands were 12.5, 6.25, and 6.25 msec for duration, level, and frequency, respectively. With ISI constant at 100 msec, ANOVAs revealed statistically significant differences among the variables for duration and for level but not for frequency (performance for frequency was nearly perfect at the shortest ISI). There were statistically significant differences between mean performance, and between mean threshold ISIs, for duration, level, and frequency, indicating that auditory temporal acuity is not independent of the kinds of sound, as has been suggested (Hirsh, J. Acoust. Soc. Am., 1959, 31, p. 759).