5 Harmonic timbres and their influence on conventional harmony.
To quote Slaymaker:
Dissonance and consonance
Some musical intervals played on instruments that produce periodic tones sound smooth or consonant, while others sound rough or dissonant. Helmholtz pointed out that dissonance results when the upper partials of the tones in a chord produce audible beats. Plomp and Levelt made Helmholtz's observation more specific by showing that the interfering partials had to be within the critical bandwidth of the ear to produce the sensation of dissonance.
The various musical intervals are now classified as dissonant or consonant as a result of long experience of both the musical instruments and the tone combinations possible with them. Once the classification of the intervals as consonant or dissonant had been made, it was possible to judge the musical acceptability of new instruments merely by sounding the various consonant intervals with them. If dissonant sounds were produced, the instrument was inacceptable [sic]. Because of this long process involving the continuous interaction between the tone of the instruments and the form of the rules of harmony, traditionally accepted instruments have evolved a tone structure that is essentially harmonic and a spectral envelope, or formant shape, that prevents serious beats being formed between closely spaced higher partials, which could fall within the critical band of the ear.
Thus traditional harmony results from the characteristics of the traditional instruments of European music, the majority of which fall into the first category discussed in 4.1 above, and also provides positive feedback for the retention of these instruments, the acceptance of new ones which have the same characteristics, and the rejection, or relegation to non-harmonic use, of those instruments which do not conform to or approximate to conformity with those characteristics.
To quote Slaymaker: