seem, therefore, that there is a good deal to be said for the adoption of a method which is already in use. This decided the author to employ the ordinary musical system—at any rate for all the preliminary experiments carried out by him—and it was soon found to be so convenient as not to make it worth while to abandon it in the later ones.
The keyboard of the colour-organ has therefore been arranged precisely in the same way as that of an organ or a piano; and, as has already been stated, the spectrum-band has been divided up into similar intervals.
This division of the intervals, or notes, corresponding to those of the musical scale, is not quite as simple as it would first appear to be, because of the fact that the rate of dispersion at one end of the spectrum-band is considerably greater than at the other. At the red end, where the dispersion is greatest, the spaces between the points from which the colour is obtained have to be greater in order to provide for the slower increase in rate of vibration, and at the violet end they are consequently closer together. This, however, is
50
