The detectability of a brief 10‐msec sinusoidal signal was measured in the presence of a sinusoidal masker. The frequency of the masker was varied while the frequency of the signal was fixed at either 250, 1000, or 2800 Hz. In the first study, the masker was either continuous or gated; when gated, it was also 10 msec in duration and occurred simultaneously with the signal. When the masker was continuous, the data agreed essentially with those reported by Wegel and Lane in their classic paper. If the masker was gated and near the signal in frequency, there was little difference in the amount of masking produced by the gated and the continuous conditions, although gating the maskers produced somewhat more masking, about 5 dB. When the frequency of the signal and masker differed, for example, in the case of a masker at 1200 cps with a signal at 1000 cps, the gated condition produced substantially more masking, about 40 dB. A second study measured how the signal level must be changed to compensate for an increase or decrease in the masker level. In the third study, the signal was presented at various delays after the onset of the maskers to determine how long the masker must be present to achieve, effectively, a continuous‐masker condition.