our other experience, though its historical connection with mysticism is intelligible. It affords a valuable means to the study of the associative working of music.
In many respects like the mental conditions spoken of is that accompanying dreams. It is stated that we do not, or do not often, dream in terms of auditory sensation. The author cannot confirm this view. It is hard to decide whether we dream in sound or in music; but the question is not of great importance. Perfect musical hallucinations occur (Macnish). Perhaps unique is a case of the connection of visual images with music. We may compare the connection of colors and tones. An association of touch-sensations with tones is known. (Cf., on the whole subject, Gurney, Myers and Podmore. We need not follow these authors as regards telepathic explanation.)
How, and to what extent, do we memorize music? As a general rule, the manner corresponds to the musical structure itself. The memory for harmony is different from that of melody (cf. the memory of the Hottentot and that of Mozart). The will to remember melodies facilitates their remembrance; in the author's case, such willing means the calling to aid of auxiliary ideas. Bain's explanation of memorizing is not valid: rhythm is the principal thing. Whether memory is always dependent on the intensity of the sense impression is not easy to decide. More probable is Stumpf's view, even in the face of the fact that the imitatory reflex is conditioned by such intensity. Musical memory is better than other memories; yet the impressions are not stronger. For intensity we substitute "mental working over." Cf. reaction-time; the auditory impression is easier associated than others.
Absolute tone-memory is a curious phenomenon. It is not proportional to the relative tone-memory of the same individual. The latter facilitates the appreciation of melody, absolute memory that of harmony. This points to the correlation of a better ear with absolute memory, apart from the question of musical talent. Relative tone-memory is often an indirect tone-memory, which has arisen by association with the time-sense. Relative tone-memory suffices for singing from printed music; the singer may even be disturbed by absolute memory. The latter is not necessary for perfect musical enjoyment. Wagner had no very developed absolute memory (Die Meistersinger); Mozart would seem to have possessed it.
Motor ideas form a great assistance to absolute memory of a tone-pitch. The part played by laryngeal sensations and ideas of move-
