The first experiment reported here measured thresholds for clicks in forward masking as a function of the masker level and as a function of the temporal relation of a 6-kHz low-pass noise masker to the click (300-msec duration with a 20-msec temporal gap or 10-msec duration with a 5-msec gap). Also varied were the spectral content of the click (low-pass filtered at 1 or 5 kHz) and the interaural phase of the click (0 or pi). The difference in frequency content had no effect on the amount of masking for the 300-msec masker, while with the 10-msec masker greater masking was found for the 1-kHz click. This combination (1 kHz, 10 msec) was also the only one to produce Masking Level Differences (MLDs) when the click was presented dichotically. A second experiment investigated the effects of combining the maskers used in the first experiment. Additional masking (above that predicted by an energy sum) was found, as has been reported elsewhere [Penner and Shiffrin, A. Acoust. Soc. Am. 67, 617-627 (1980)]. However, the magnitude of this additional masking was decreased for certain conditions. These data conflict with the predictions of additivity of masking obtained from a model proposed by Penner and Shiffrin and suggest that modifications to that model are needed. The results of both experiments can be explained by assuming that two processes are acting in forward masking [Duifhuis, J. Acoust. Soc. Am. 54, 1471-1488 (1973)].