ACOUSTICS APPLIED TO PUBLIC BUILDINGS. 227 



continuance of a single sound, and the tendency to confusion in dis- 

 tinct perception will dejiend on several conditions ; first, on the size 

 of the apartment ; secondly, on the strength of the sounder the inten- 

 sity of the impulse ; thirdly, on th» position of the reflecting surfaces ; 

 anil fourthly, on the nature of the material of the reflecting surfaces. 

 In regard to the first of these, the larger the room, the longer time 

 will be required for the impulse along the axis to reach the wall; and 

 if we suppose that at each collision a portion of the original force is 

 absorbed, it will require double tlie time to totally extinguish it in a 

 room of double the size, because, the velocity of sound being the same, 

 the number of collisions in a given time will be inversely as the dis- 

 til nee through which the sound has to travel. 



Again, that it must depend upon the loudness of the sound, or the 

 intensity of the impulse, must be evident, when we consider that the 

 cessation of the reflections is due to the absorption of the walls, or to 

 irregular reflection, and that, consequently, the greater the amount of 

 orig'inal disturbance, the longer will be the time required for its com- 

 plete extinction. This principle was abundantly shown by our obser- 

 vations on different rooms. 



Thirdly, the continuance of the resonance will depend upon the 

 position of the reflecting surfaces. If these are not parallel to each 

 other, but oblique, so as to reflect the sound, not to the opposite,^ but 

 to the adjacent wall, without passing through the longer axis of the 

 room, it will evidently be sooner absorbed. Any obstacle, also, which 

 may tend to break up the wave, and interfere with the reflection 

 through the axis of the room, will serve to lessen the resonance of the 

 apartment. Hence, though the panelling, the ceiling, and the intro- 

 duction of a variety of oblique surfaces, may not prevent an isolated 

 echo, provided the distance be sufficiently great, and the sound suffi- 

 ciently loud, yet that they do have an important effect in stopping the 

 resonance is evident from theory and experiment. In a room fifty feet 

 square, in which the resonance of a single intense sound continued six 

 seconds, when cases and other objects weve placed around the wall, its 

 continuance was reduced to two seconds. 



Fourthly, the duration of the resonance will depend upon the nature 

 of the material of the wall. A reflection always takes place at the 

 surface of a new medium, and the amount of this will depend upon the 

 elastic force or power to resist compression and the density of the new 

 medium. For example, a wall of nitrogen, if such could be found, 

 would transmit nearly the whole of a wave of sound in air, and reflect 

 but a very small porliou ; a partition of tissue-paper would produce 

 nearly the same effect. A polished wall of steel, however, of sufficient 

 thickness to prevent yielding, would reflect, for practical purposes, all 

 the impulses through the air which might fall upon it. The rebound 

 of the wave is caused, not by the oscillation of the wall, but by the 

 elasticity and mobility of the air. The striking of a single ray of 

 sound against a yielding board would probably increase the loudness 

 of the reverberation, but not its continuance. On this point a series 

 of experiments were made by the use of the tuning-fork. In this in- 

 strument, the motion of the foot and of the two prongs gives a sonor- 

 ous vibration to the air, which, if received upon another tuning-fork 



