256 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1953 



It is believed that low-frequency transients help also in this respect. 



A factor of similar origin which will make localization of the music 

 of the wind in an orchestra more certain than that of the strings is that 

 the former are mostly playing parts which are not doubled, whereas 

 the latter show a chorus effect. The fact that two or more strings 

 playing the same part can never exactly reproduce each other's effect 

 as to loudness, pitch and timbre, and time of initiation and duration 

 of transients (even if they were to be played by a machine) masks 

 any directivity, introduces slight but rapid vibrato, and smudges indi- 

 vidual characteristics. The same is true of an organ solo stop played 

 over chorus foundation stops on another manual. This chorus effect 

 is, in fact, of sufficient importance to be imitated — than which nothing 

 is easier ! — by the makers of electrophonic instruments. 



Recording for scientific purposes is properly done in a heavily 

 lagged room in order that the sounds of the instruments shall not be 

 overlaid with extraneous echoes, but since one usually listens to music, 

 whether directly or over the radio, under circumstances in which some 

 reverberation is superposed on the original sounds, it is proper to 

 consider the modifications introduced by the acoustics of the buildings 

 in which the music is produced or reproduced. The general effect of 

 excessive reverberation is a smudged rendering. The desirable 

 amount of reverberation depends on the individual's taste and expe- 

 rience. Some reverberation is desirable if only for the sake of the 

 players who would find an anechoic chamber very unsympathetic to 

 their efforts. What one is willing to tolerate in this respect is largely 

 a matter of what one is accustomed to. Since we most often hear a 

 large organ under a high vault and a piano in our own drawing room 

 we incline to favor long and short reverberation times respectively 

 for these two instruments. 



Another concomitant of reverberation is loss of directivity. In a 

 hall of hard wall surfaces sounds will bombard the listener's ears from 

 several directions at once, thus masking the direct sound by which he 

 estimates the true direction of the source. Furthermore, there will 

 be marked foci and deaf spots at certain frequencies in such a cham- 

 ber, a condition which is not only unfair to the audience but makes it 

 impossible for the radio engineers to locate their microphones to 

 give faithful reproduction. 



From the point of view of the latter persons, indeed, the ideal 

 would be to diffuse the sound equally in all parts of the auditorium, 

 a condition obtained in broadcasting studios by lining the walls with 

 half cylinders or other protuberances of varying size. This would, 

 however, further spoil the directivity by which, as I have already 

 pointed out, the audience is able to judge which instruments are play- 

 ing at a given instant and to pick out the soloist in a concerto. 



