162 BELL SYSTEM TECHNICAL JOURNAL 



The ordinates show the amounts in loudness units that the threshold 

 value of a tone of any frequency called the "masked tone" is shifted 

 due to the presence of another tone called the "masking tone." The 

 frequency of the masking tone is given at the top of each set of curves. 



The experimental procedure was as follows : The threshold values 

 for the two tones were first determined. The intensity of the masking 

 tone (the frequency of which is given above each graph) was then 

 increased beyond its threshold value by the number of units indi- 

 cated just above the curve. The masked tone was then increased 

 in intensity until its presence was just perceived. The amount of 

 this latter increase, measured on the loudness scale is called the 

 threshold shift and is plotted as ordinate in Figs. 8, 9 and 10. The 

 frequencies of the masked tones are given by the abscissae. 



For example, in the fourth chart, the masking effects of the tone 

 having a frequency of 1,200 cycles are shown. It is seen that the 

 greatest masking effect is near 1,200 cycles, which is the frequency of 

 the masking tone. A tone of 1,250 cycles must be raised to 46 units 

 above the threshold to be perceived in the presence of a 1,200-cycle 

 tone which is 60 units above its threshold, or it must be raised to 

 within 14 units of the masking tone before it is perceived. This 

 corresponds to an intensity ratio between the tones of only 25. A 

 tone of 3,000 cycles, however, can be perceived in the presence of a 

 1,200-cycle tone which is 60 units loud when it is only 8 units above 

 its threshold. This means that the intensity ratio between these two 

 tones, under such circumstances, corresponds to 52 units or to a 

 ratio of approximately 160,000 in intensity. However, as the loudness 

 of the masking tone is increased, all of the high tones must be in- 

 creased to fairly large values before they can be heard. For example, 

 the high frequencies must be raised 75 units above the threshold to 

 be heard in the presence of a 1,200-cycle tone having a loudness of 

 100 units. But even for such large intensities for the masking tone, 

 those frequencies below 300 are perceived by raising their loudness 

 only slightly above the threshold value. It should be noticed that 

 in all cases, those tones having frequencies near the masking fre- 

 quency, whether they are higher or lower, are easily masked. 



It is thus seen that Mayer's conclusion, that a low pitch sound 

 completely obliterates higher pitched tones of considerable intensity 

 and that higher pitched frequencies will never obliterate lower pitched 

 tones, is true only under certain circumstances. A low tone will 

 not obliterate to any degree a high tone far removed in frequency, 

 except when the former is raised to very high intensities. Also a 

 tone of higher frequency can easily obliterate a tone of lower fre- 



