So far everything is in order. Now, however, let us make a simple experiment. You know the =stroboscope= or zoetrope that is sold in the toy-shops: a cardboard drum, open at the top, that twirls on a handle; a strip of paper, on which are printed phases of some movement (the flight of a bird, the gallop of a horse), is placed inside, round the bottom of the drum; and you look down at the strip, while the instrument revolves, through vertical slits cut at regular intervals in the upper half of the drum-wall; you then see a continuous movement. Suppose that you make a new strip, on which you draw simply two lines, a vertical and a horizontal; you draw them some distance apart, but in such wise that, if they came together, they would form a right-angle. Turn the drum slowly, and you see the two lines; turn it swiftly, and you see the right-angle, like a letter L; turn it at a middle rate, and you see—according to the direction of turn—the vertical _fall over into_ the horizontal, or the horizontal _rise up into_ the vertical. You see movement, where there is no movement to see! Here, then, is a case of _perception of movement in terms of sheer brain-habit_, of a settled nervous disposition that now has no mental correlate, but whose establishment has depended on the past history of the individual, possibly of the race. Take, as a second instance, the perception of =melody=. Primitive melodies seem to be of two types. In the one, the scale arises by synthesis of small tone-steps or tone-distances, which are approximately ‘whole tones’; the melody consists only of two or three of these steps, and the last and lowest tone is the principal note of the tune. In the other, the scale arises by analysis of the larger consonant intervals, fourth and fifth; these intervals are broken up into smaller steps; the octave appears as a drone-bass; the first and highest tone is the principal note. An intermediate type keeps for the most part to small steps, but shows ascents and descents _portamento_ through octave, fifth and fourth; it, too, makes the first and highest tone the principal note. We can account for a good deal of this development: we know that the voice cannot be evenly sustained in recitative, but naturally drops; we have reason to believe that the memory of absolute pitch is strongly developed in primitive peoples (parrots repeat their tunes at the same pitch, and the same thing is largely true of young children); we know the recurrent tonality of the octave (p. 52); we know that the fourth is the natural drop of the voice at the end of a sentence, and the fifth its natural rise in asking a question; we know that men, women and boys, singing in ‘unison,’ will really sing in octaves, and often in fifths and fourths; we know that the semitone, the final unit of our own scales, is the smallest tone-step that can be accurately sung; we know that musical instruments were invented very early, and that