Most Comfortable Loudness (MCL) was determined for each of 10 normal-hearing young adults using pulsed pure-tone trains and damped wave trains (DWTs) each at 2.5/sec, at frequencies of .25 and 2 kc/s; 4 MCLs were determined for each of the 4 signal type/frequency combinations. In a bracketing procedure employing steps of 20, 10, and finally 5 db, Ss were permitted to respond only "loud" or "soft" to each short train. Group mean test-retest differences within each stimulus type/frequency combination were only 1-2.5 db for all possible test-to-contiguous-test comparisons, i.e., better than with most other MCL procedures except the descending series in the Method of Limits. However, individual Ss differed by from 15-35 db within the 16 distributions (2 stimulus types X 2 frequencies X 4 replications). No statistical or practical differences were found between pure tones or DWTs, or across frequency. It was suggested that the range of individual differences renders any group mean MCL too imprecise a measure for selecting and prescribing a hearing aid, or for determining the effect that aid may have on an individual's performance.