ii6 



NA TURE 



[June 4, 1908 



SOCIETIES AND ACADEMIES. 

 London. 

 Royal Society, January l6. — " On the Perception of the 

 Direction of Sound." By Prof. C. S. Myers and Prof. 

 H. A. Wilson, F.R.S. 



Lord Rayleigh (Phil. Mag., 1907, xiii., 214) has shown 

 that the determination of the apparent direction of a sound 



sound was only moderately clear 

 right " it was doubtful, while at 

 lateral effect at all. 



If we call the lateral effect 4>: 

 positive and left effects negative, 

 Rayleigh 's 

 stant, .V ii 

 middle point of the scale 



at " middle " or " half 

 " middle " there was no 



considering right effects 



then, according to Lord 



results, ^ =A sin (47rii.-v:/D), where A is con- 



the distance in cm. of the T-piece from the 



the number of vibrations 



per second, and v the 



3SL 



I 6 lb £b 3b -tip sb 6b tIo eta ab idio lio lio lio i4o I jo 1^ ito A I jo gto 



F 



"W 



K 



used in these experiments con- 

 about 2-5 m. long and 2-5 cm. 



is influenced by phase-differences between the vibrations 

 at the two ears. In the present paper this influence is 

 experimentally further investigated, and a theory is 

 advanced which offers a possible explanation of the effects 

 ultimately in terms other than phase-difference. 



The apparatus (Fig. i) 

 sisted of a brass tube AB 



in diameter, at the 



centre of which a short 



T-piece T was soldered. 



This tube was freely 



movable within the 



two slightly larger 



tubes CD "and EF, 



from the ends of which 



tubes were led to two 



softly padded caps, 



supported on retort 



stands, and applied to 



the ears of the 



observer H. Beside 



the tube .\B was 



placed a graduated 



scale DE, which re- 

 corded the position of 



the T-piece. The 



sound entered the T- 



piece from a vibrating 



tuning-fork K held 



near it. By sliding the 



tube .\B, so that the 



T-piece at different 



times pointed to differ- 

 ent divisions of the 



scale, every kind of 



phase-difference b e- 



tween the vibrations 



reaching the two ears 



could be produced. A 



screen SS concealed .'•' 



the position of the T- 



pieee from the 



observer. One of the 



authors varied the 



position of the T-picce, sounding 



it at each position, while the 



observer, stated on which side 



be sounding. The answers of 



graded — "full right,' 



velocity of sound. 



The results of the 

 observations can be 

 conveniently expressed 

 in the form of curves, 

 the abscissa; express- 

 ing the scale-readings 

 and the ordinates the 

 lateral effects. For 

 this purpose a " full 

 right " is reckoned as 

 I, a " half right " as 

 a 3, a " middle or half 

 right " as a J, a 

 " middle " as o, the 

 corresponding left 



effects having equal, 

 but negative, values. 

 ° Figs. 2, 3, 4 show the 



curves obtained with 

 forks of frequencies 

 512, 384. 128. The 

 dotted line in each figure is the curve ifi =sm{^irnxlv). 



It is at once apparent that the theoretically and the 

 experimentally obtained curves agree with one another re- 

 marlvably w'ell. The discrepancies which were sometimes 

 met with were, with the aid of manometric flames, traced 

 to the occurrence of resonance in one or other side of the 

 tube. Interesting results w'ere obtained from experiments 



leocrns. 



Fig. 2. (512) 

 (Observer facing towards fork) 



i6ocms. 



~]-^ 



F10.3 (384) 

 (Facing Cowards) 



3) 



+ 1 



S. 



*c — 



<^ 



*x 



40/ 50 



60 



70 



80 



120 



I60f/n5. 



iba 



right, 

 " full 



the tuning-fork before 



other, acting as the 



the fork appeared to 



the observer could be 



half right," " middle or half 



' "middle," "middle or half left," "half left," 

 left." At "half right" the lateral direction of the 



NO. 2014, VOL. 78] 



Fig. 4- (128) 



(Facing Cowards) 



conducted when the tubes were partially blocked or 

 gradually pinched on one side. Other observations, of 

 psychological rather than physical or physiological interest, 

 are reserved for a future communication. 



It is here suggested that the lateral effects thus pro- 

 duced by differences of phase at the two ears are really 

 and ultimately due to the binaural differences of sound- 



