I often had arguments with Mr. Ellis upon these points, and after the study of Arabic and Persian scales for his comparative examination of “The Musical Scales of Various Nations” he came at last to the same conclusion. The fourth always seemed to me the most naturally selected interval for the origin of the primitive scales. It prevails in Arabia, Persia, China, and the East generally.

The instrument which is here illustrated is Japanese, and is called a clarinet on account of the similarity in the relation of its sounds, its second series being 12ths, not octaves. The most noticeable peculiarity of the little instrument is its reed, which is as broad as the tip of our bassoon reed; but unlike that, is broader at the bass end, which is inserted in the pipe (as you will understand by the drawing, which shows the reed cut through at mid-section).

The vibrating portion is at the tip, to the extent downward of three eighths of an inch, which evidently has been pinched together and then dried in some particular way. The two lips from the centre expand outwardly under moisture, and leave a fine ovate opening, which, under the suction of the passing stream of air closes, and then reopens by its own elasticity. The reed does not consist of two separate parts bound together, but is itself tubular, its diameter at the bottom being three eighths of an inch.

Then a little clip of cane with bound ends forms a ligature to keep the lips of the reed in proper relation during blowing; and as it is pressed down tightly or loosely, affects in some degree the pitch. Also the lower end of the reed is bound with a strip of soft paper, where it fits into the pipe; and so, whether it is allowed to be set far into the pipe or not, will likewise affect the pitch considerably. This will account for some discrepancies in the statements as to the normal pitch of the Hichi-riki. Again, in China, the same kind of instrument is found differing in length, and having the name Kwan-tze, The Japanese instrument is no doubt a refined copy of the Chinese model, which itself is so ancient that it may have been brought from some region of the Caucasus. My own instrument measures in pipe length 8in., and with the reed fitted in, 9-1/8in. In the Brussels Museum, one is noted which is 8-5/8in. in pipe length, and the lowest note is F; but this instrument has another thumb hole between the third and fourth holes in addition to the hole which appears in my pipe between the sixth and seventh hole.

The pipe also, it should be remarked, is not cylindrical, but in a musical sense is more so; since, by its being a cone inverted, the flattening influence of form on the pitch is increased. As it was in the old German flute, which, like this, was an inverted cone, and so conduced to the better production of the lowest notes.

The scale of the Hichi-riki, on the authority of the Musical Institute of Tokio, is given with the following tablature:—

The open pipe length for the lowest note would therefore be twice the length of this pipe, so we say that the Hichi-riki speaks double depth tone. And when blown with higher pressure, the first series of harmonics is not one of octaves, but of twelfths. An interesting circumstance is that when a smaller reed such as we use for the oboe is inserted, then the tone leaps a fourth (not an octave) higher, and its harmonic series is one of octave relation; in fact, it is the original twelfth acting, slightly modified by being elicited by a smaller reed, and hence emphasizing the compound nature of results from pipe and reed associated. With one reed, I remember that the pipe rose a fifth, its twelfth being then really transfigured only, yet becoming its octave, being, as elicited, the same note.

Another curious fact connected with the Hichi-riki is that—if the upper end of the pipe is placed full within the mouth, and is blown through without any reed whatever, and without any action of the lips—clear and powerful notes are elicited, varied as the openings of the holes are varied; provided one of the upper holes is left open. Then the pitch of the issuing notes corresponds to such as are calculated according to the length between the distant holes as an open pipe length. It is, further, indifferent whether the end of wide diameter or that of narrow diameter is taken into the mouth; either way sounds are readily produced. The upper finger hole thus corresponds to the twelfth hole in the clarionet—according to the argument upon this question in a previous chapter—and the length of pipe above it is to be disregarded.