Now tone-relationship alone, it is clear, would not of itself involve this immediate impulse to end a sequence of notes on one rather than on another. Nor is tonality, in the all- pervasive sense in which we understand it, a characteristic of ancient, or of mediaeval music, while the tendency to end on a certain tone, which we should to-day call the tonic, was always felt. Thus, since complete tonality was developed late in the history of music, while the closing on the tonic was certainly prior to it, the finality of the tonic would seem to be the primary fact, out of which the other has been developed. We speak to-day, for instance, of dissonant chords, which call for a resolution--and are inclined to interpret them as dissonant just because they do so call. But the desire for resolution is historically much later than the distinction between consonance and dissonance.... "What we call resolution is not change from dissonant to consonant IN GENERAL, but the transition of definite tones of a dissonant interval into DEFINITE TONES of a consonant."<1> The dissonance comes from the device of getting variety, in polyphonic music, by letting some parts lag behind, and the discords which arose while they were catching up were resolved in the final coming together; but the STEPS were all PREDETERMINED.<2> Resolution was inevitably implied by the very principle on which the device is founded. That is, the understanding of a chord as something TO BE RESOLVED, is indeed part of the feeling of tonality; but the ending on the tonic was that out of which this resolution- feeling grew. <1> Stumpf, op. Cit., p. 33. <2> Grove, _Dict. Of Music and Musicians_. Art. "Resolution." Must we, then, say that the finality of the tonic is a unique, inexplicable phenomenon? giving up the nature of melody as a problem if not insoluble, at least unsolved? The feeling of finality in the return to 2n is explained by Lipps and his followers, from the fact that the two-division is most natural, and so tones of 2n vibrations would have the character of rest and equilibrium. This explanation might hold if we were ever conscious of the two-division as such, in tones --which we are not; so that it would seem to depend on the restful character of a perception which by hypothesis is never present to the mind at all.