of energy in excess of the accepted value. This accounts for the additional energy term. 4. When a wave train travels down a tube, there is a region of higher density (where the wave forms) followed by a region of lower density. The density variation seems to be a basic feature of a traveling wave and does not depend on the amplitude. Two additional conclusions, hrst given by Schoch, should be mentioned: 5. One may separate the relation between the energy and the intensity into two parts. One is related to the stored energy, while the other is related to the oscillatory energy.6. As long as the total volume is held 6xed, the net stored energy is zero. This does not mean that the energy may not have a space distribution.One would expect the eRect described in this paper to be unimportant in liquids. Equation (4) is for an ideal gas; however, corresponding forms exist in liquids. ' Combining the expression for water' with Eqs. (10a) and (15a), one may see that Andrejew's terms are not important. One might expect similar results for other liquids. These calculations, however, have not been made.The coeKcients of electrical and thermal conductivity have been computed for completely ionized gases with a wide variety of mean ionic charges. The eftect of mutual electron encounters is considered as a problem of diffusion in velocity space, taking into account a term which previously had been neglected. The appropriate integro-differential equations are then solved numerically. The resultant conductivities are very close to the less extensive results obtained with the higher approximations on the Chapman-Cowling method, provided the Debye shielding distance is used as the cutoft in summing the effects of two-body encounters.