SECTIONAL TRANSACTIONS.—A. 361 
different postures of the vocal organs, most of the consonants represent the effect of 
different gestures or movements by which the successive vowel postures are reached. 
The consonants were in general produced by resonators of smaller orifice than the 
vowels, in many cases by more than two resonators, and in many cases by 
resonances which changed in a characteristic way in pitch or amplitude or both. 
In general the human ear identifies the characteristic gestures by means of the 
audibie resonance changes which they produce. 
11. Prof. P. EK. Sasrye.—Research in Architectural Acoustics. 
A brief summary of the Reverberation or Sound Chamber method of 
acoustical measurements is presented. This method makes use of the fact that 
sound from a source within a closed space remains audible for an easily 
measurable length of time after the source has ceased. This time, T, will depend 
upon E, the acoustical power of the source, t, the time during which it speaks, 
V, the volume of air in the room, A, the rate of dissipation of sound energy, and 
, the minimum audible sound energy density. Assuming that the time during 
which the source speaks has been sufficiently great for the density of sound 
energy in the room to reach the steady state I), in which the sound produced 
per second equals the sound dissipated per second, the time required for the 
density to fall to the threshold density i is given by the relation 
E 
1 AT=] Jake 
sa "5 ce) 
A is defined by the equation 
di 
2 —_ =—AI. 
(2) ai I 
A, for sound of any pitch within a given room, is determined initially by 
varying the acoustical power of the source of sound in known ratios and measur- 
ing the duration of audibility of the residuai sound after the source has ceased. 
A is the slope of the straight line obtained by plotting log E as _a function of T. 
; . Hz 
Knowing A, the ratio Fi for any observer follows from Equation (1). 
The arrangement of the Sound Chamber in the Wallace Clement Sabine 
Laboratory at Riverbank, Geneva, Illinois, is described. Illustrating the use of 
such a room, the method and results of a quantitative study of non-musical 
impact sounds, such as the noise from the operation of typewriters and the 
like, are given. 
The results of a research programme on the general problem of the reduction 
of sound transmitted by partition walls are summarised. This research has 
included measurements on so-called sound ‘ deadening quilts’; that is, materials 
which are porous and inelastically compressible and flexible, structural units 
of wood, glass and steel of various types, and standard masonry partitions of 
different materials. 
The experiments show that reduction of sound in transmission by homo- 
geneous felts and the like is a true absorption process, so that I, the density of 
sound energy on the further side of a thickness x of a material of this sort, is 
given by the expression 
i Ty ¢-ar, 
r 
Thus the sound-insulating value of a material of this character may be specified 
by the two experimentally determined coefficients 7 and q. 
In masonry walls the resonance of the structure as a whole and in segments 
is an important factor in sound transmission, so that experiments with a single 
tone are difficult to interpret. Extending the investigation to the whole scale 
of tones, experiments on some fifteen different masonry walls of different 
materials and thicknesses show that the general reduction of sound in transmis- 
sion through such walls is independent of the material and the structural stiff- 
ness of the wall, and is a function only of the weight per unit area. 
