62 On the Variety of Voices. 
duced by the vocal organs, than we do in distin- 
guishing the same in the concords and discords of a 
piece of composite music. The contrary is therefore 
‘true, viz. the intervals that enter into the composi- 
tion of the human voice, and the tones of sonorous 
bodies, are too small to have their terminating sounds 
accurately discriminated by the ear, but sufficiently 
large* to affect it with distinct sensations corres- 
ponding to their relative affections. The certainty 
of the last conclusion can hardly be supected when 
the grounds on which it stands are properly attend- 
edto. The common idea of tone has been consi- 
dered in conjunction with the received definition of 
sound; a definition which easily explains the phe- 
nomenon abstractedly taken, but is incapable of ac- 
counting for its various modifications. But, as 
sound can only be modified t by sound, the three 
* The truth of this assumption rests on the supposition 
that sounds cease to be distinct to sense before they are in 
perfect unison ; an opinion that will be proved in the se- 
quel of the essay. After which it must appear evident, 
that the required effect will take place in intervals, the 
ratios of whose terminating sounds do not exceed a given 
Tatio, 
+ That sound can be modified by nothing but sound must 
be admitted as an axiom in phonics; for, if the contrary be © 
maintained, an absurd consequence will ensue; viz. that 
sonorous bodies can produce in the ear sensible impres- 
sions, arising from their specific or chemical qualities, But 
this is a doctrine repugnant to the common theory, which 
