To express the noise reducing effect of a muffler in terms of the difference in the sound power levels radiated from the tailpipe exit with and without the muffler in the system, equations are deduced by means of the four-terminal constants (A B C D) of the equivalent electrical network of the whole muffler system. As a result, the effect L, in dB, is expressed by Eq. (12), and L_s in the equation are given by Eqs. (13) and (14) for constant pressure and constant velocity sources, respectively, where R_2 is the acoustic radiation resistance of the tailpipe exit and Z_1 is the characteristic impedance of the input-pipe, and terms with prime denote the quantities without the muffler. In the case of expansion chamber type mufflers, L_s are simplified as Eq. (20) or (21). by assuming that the cross-sectional area ratio of the chamber to the tailpipe is considerably large. In these equations, as illustrated in Fig. 3, it seems as if the input-pipe and the tailpipe are terminated by zero impedances and the exit of the chamber is closed. And the behavior of the chamber is represented by the term, 20log|Y_fZ_3|, where Y_f is the equivalent open-circuit transmission admittance of the chamber and Z_3 is the characteristic impedance of the tailpipe. For some typical cylindrical chambers, the explicit forms of 20log|Y_fZ_3| are shown is Tables 1 and 2, where it is assumed that the sound wave is a plane wave progressing in the direction of the axis . In the low frequency range, since each element of the muffler system may be regarded as a concentrated constant element, L_s are represented by Eqs. (22) and (23). L, L_s and 20log|Y_fZ_3| are very simple, and their curves as a function of frequency are easily obtained by graphical method as illustrated in Fig. 4. When a loudspeker is used as the source, the measuring systems of 20log|Y_fZ_3| and L are illustrated schematically in Fig. 5 and 6, and the curves are obtained from Eqs. (30) and (31) by the same method as illustrated in Fig. 4. Even if the shape of the chamber is complicated, the curve of 20log|Y_fZ_3| is obtained experimentally by the same process. The results are shown in Fig. 7 and 8. Measured curves are in good agreement for practical use with those by graphical method so far as the sound wave in the chamber can be regarded as a plane wave. In the higher frequency range above the lowest resonance frequency in the direction perpendiculer to the axis of the chamber, the sound wave can not be regarded as a plane wave, and the noise reduction may not be expected.