In the case of short duration sounds it is well known that temporal loudness summation occurs within a certain time period (i.e., critical duration), and a trade-off relationship is established between the sound intensity and the duration of the sound. Therefore, for the estimation of the loudness of brief sounds such as impulsive noise the temporal loudness summation must be considered. This paper deals with artificial impulsive sounds with a relatively short rise time and a long decay, as usually observed for actual impulsive sounds, and whether the loudness of an impulsive sound is determined by the peak level of the sound or by the total energy of the sound including the decaying part . LAX in ISO 1996, which is the single event exposure level, is chosen as an energy index of the sound in this paper. It is probable, however, that the time varying pattern of the sound or auditory after effect has an effect on the loudness of impulsive sounds. Such effects, however, are not reflected either in LAX or in the peak level measurement. Accordingly additional experiments have been carried out to examine the temporal characteristics of the hearing mechanism in relation to the loudness of impulsive sounds. As a result, it seems that LAX is a good measure of the loudness of impulsive sounds which are not too short, but when the duration of the sound is less than 60 ms the contribution of the auditory after-effect on the loudness needs to be taken into consideration. No relation could be found between the peak level and the loudness of impulsive sounds.