ARCHITECTURAL ACOUSTICS 105 



acoustics are unobjectionable. This is apparently not borne 

 out by experience, largely because the importance of de- 

 tailed conformity to the model is not realised. As will be seen 

 later, unless the new hall is constructed of similar materials 

 and accurately to the same scale, its acoustics will not follow 

 those of the model. And further, as will be shown, even though 

 the proportions be the same, any great difference in size will 

 alter the acoustics markedly in certain respects. 



The general principle underlying the cure of reverberation 

 is the introduction of material which will absorb the sound. 

 The reverberation in an empty house is notorious, and the im- 

 provement effected by hanging heavy curtains is also a matter 

 of common knowledge. 



The reason is clear when one considers the nature of sound. 

 Sound is a form of energy. It cannot be destroyed, it can only 

 be converted into some other form. Some will escape through 

 open windows, some will be used in imparting mechanical energy 

 to the walls, some will pass directly by friction into heat. A 

 hiss is very soon killed by air friction, and the frictional effects 

 are increased by the presence of porous walls and hangings. 

 It is only by true dissipative forces such as viscosity that sound 

 can be so converted. No mere modifications by the introduction 

 of irregularities are of any avail in this connection. The break- 

 ing up of smooth surfaces by relief work may eliminate echoes, 

 but it does not cure reverberation unless the total absorbing 

 surface is increased by the breaking up. The sound is still 

 present and must be converted into energy of another form 

 before it can cease to be heard. 



The most important experimental work on reverberation 

 was commenced by W. C. Sabine of Harvard in about 1896, 

 and five years later a series of papers appeared embodying the 

 results of these experiments (4). 



The rate of decay of reverberation for a given sound was 

 measured by measuring the duration of audibility of residual 

 sound after the source had ceased. As the rate of disappear- 

 ance of reverberation is proportional to the rate at which 

 sound is absorbed, the experiments could be arranged to give 

 information as to the relative absorbing powers of different 

 substances. These experiments were carried out in the Fogg 

 Art Museum in the University of Harvard. An organ-pipe 

 C 512 was sounded by a constant pressure-blower and the time 

 that elapsed after the wind was shut off before the sound 

 became inaudible was measured by means of a chronograph. 

 Preliminary experiments showed that the duration of audibility 

 was almost independent of the positions of source and observer. 

 The results of introducing cushions from the Sanders Lecture 

 Theatre are given in the following table and curve (Fig. i ), the 



