This paper reports the results of comparison measurements of “reverberation coefficients” made on identical samples of materials in three reverberation chambers, having volumes ranging from 10,000 to 15,000 cubic feet. The results of these measurements indicate that, in rooms of comparable size, discrepancies in the measured values of the “reverberation coefficients” are not excessive at frequencies of 500 cycles and higher, and that disagreement at frequencies below 500 cycles are not sufficiently great to warrant the abandonment of reverberation methods. Reasons for the disagreements shown are considered. An attempt is made to correlate the results of acoustical impedance measurements made by the closed tube method reported by Beranek with room measurements of reverberation coefficients on eight commercial absorbents. It is shown that at frequencies below 500 cycles, the average values of the measured reverberation coefficients of the eight materials agrees very closely with the coefficient of absorption computed from the equation α = 1 −{|z/ρc| − 1|z/ρc| + 1}2, where z/ρc is the complex value of the normal specific acoustic impedance as given by tube measurements. For frequencies above 500 cycles the relation α = 1 −{|z/ρc| cos θ − 1|z/ρc| cos θ + 1}2 gives close agreement with the average value for the eight materials. Cos θ, the average value of cosine of the angle of incidence is 0.63, corresponding to an average angle of incidence of 51°, instead of the theoretical angle of 60° under the assumption of completely diffuse sound. Explanation is to be found in the arrangement of the diffusing agents used in the reverberation chamber which diffuse normal modes of vibration parallel to the absorbing surface more than the vertical modes.