PHYSICAL MEASUREMENTS OP AUDITION 169 



With this picture in mind, it is clear why the perception of one 

 tone is interfered with by the presence of a second tone when their 

 frequencies are close together, since the nerves necessary to perceive 

 the first tone are already stimulated by the second tone. Also when 

 their frequencies are widely separated, entirely different sets of nerves 

 carry the impulses to the brain, and consequently there is no inter- 

 ference between the tones except that which occurs in the brain. 

 Although this brain interference may not be entirely negligible, 

 especially for very loud sounds, it is certainly very much smaller 

 than that existing in the ear for tones close together in pitch. 



It is also seen that the reason why the low tones mask the high 

 tones very much more easily that the reverse is due to the harmonics 

 introduced by the transmission mechanism of the ear. Inasmuch 

 as these harmonics are due to the second order modulations, they 

 are proportional to the square of the amplitude and, therefore, be- 

 come much more prominent for the large amplitudes. When two 

 tones are introduced, summation and difference tones as well as the 

 harmonics will necessarily be present (see Appendix B). With the 

 proper apparatus for generating continuously sounding tones, these 

 subjective tones are easily heard. Their frequency can be quite 

 accurately located by introducing from an external source a fre- 

 quency which can be varied until it produces beats with the subjective 

 tone. 



Messrs. Wegel and Lane who are working in this field have ob- 

 served modulation frequencies created in the ear as high as the fourth 

 order. They will soon publish* an account of this work on the 

 vibrational characteristics of the basilar membrane. It is seen 

 that the quality as well as the intensity of the sensation produced 

 by a pure tone should change as the intensity of stimulus is increased 

 due to the increasing prominence of the harmonics. This is in ac- 

 cordance with one's experience while listening to pure tones of varying 

 intensity. The non-linear character of the hearing mechanism is 

 also sufficient to account for the falling off in the ability of one to 

 interpret speech when it becomes louder than about 75 units. The 

 introduction of the summation and difference tones and the har- 

 monics makes the interpretation by the brain more difficult. Its 

 action in this respect is very similar to the carbon transmitter used 

 in commercial telephone work or to an overloaded vacuum tub. 

 This characteristic of the ear also explains why we should expect 

 departures from non-linearity when making loudness balances for 

 complex tones. It also suggests that a similar thing might be ex- 



* Wegel and Lane, see paper already cited. 



