In the accounts received from travellers who attempt to estimate the scales and character of the native music heard by them, we are accustomed to find a prevalence of the minor mode always affirmed, and the statement is generally accepted as one based upon definite knowledge. It seems to be considered that the mournful and the plaintive in song and in music reflect the temperament of the people, and are its natural expression. I am inclined to question this; for I may doubt the keenness of the observer, doubt his musical capability of ear or mental power of analysis, may perceive a tendency of mind to take a stand on foregone conclusions, and may not be satisfied that the writer is competent upon the subject upon which he writes very positively. Experience has shown me how frequently statements of this kind are not borne out by facts, although the statements have been made in perfect good faith. In this aspect there is a paper by J. F. Fillmore (an American author) which has a peculiar significance. He made a study of the music of the Indian tribes in America, having very special facilities for his work; and he also harmonised many of the melodies, with much satisfaction to the Indians. He says that,— In short, all melodic and harmonic resources to be found in our music, even the most modern and advanced, are also to be found in the primitive music of a people who have no musical notation, no theory of music, no systematized knowledge of it whatever. And then at the end we have this _naïve_ conclusion:— Long before the first week was over, all my preconceived notions of the significance of the incomplete scales, and of the importance of the plain major and minor chords as related to acoustic problems, had wholly disappeared. The truth is that scales are so elusive that they may be read so as to mean anything a system maker desires, and such scrutiny is about as reliable as the reading of character and destiny by the systems of astrology and palmistry. Chinese melodies are never definitely major or minor; are never intended to be so. The intervals are not the same as ours, and our notation does not express them with accuracy such as scientific analysis requires. On the subject of the growth of scales my conclusions have been previously recorded, but I think that here, at the end of the pipe investigation, a brief repetition is desirable to impress the memory with the special view which is of importance to the musician’s survey.