artist, are most often stamped with a character of vague melancholy, from the influence of which the mind cannot escape. It is especially when the noise of the world is hushed that these powerful harmonies produce the most overpowering and poetical effects."
It characterizes, for example, the sound of the echo, the sound called harmonics, and many others which are included in the range of musical tones, defined further on under the name of chemical and sympathetic music. We have the most remarkable examples of these in the sound of the Æolian harp. Science, as well as philosophy, poetry, and musical art, is interested in the further study of these sounds. In Germany, Goethe and Novalis, in France, Jean Paul, and many others, have eagerly appreciated the bond which unites natural harmonies to the most elevated instincts, and to the most ideal aspirations of the human soul.
Prof. Tyndall has recognized the fact that, in order to render a flame musical, it is necessary that its volume be such that it should explode in unison with the undulations of the fundamental note of the tube, or of one of its harmonics. He also asserts that, when the volume of the flame is too great, no sound is produced; he demonstrates it, by increasing the flow of gas. Prof. Tyndall has also called attention to this fact, that, in order that a flame may sing with its maximum of intensity, it is necessary that it should occupy a certain position in the tube. He shows this by varying the length of the tube over the flame, but he does not specify the proportions which must exist between the flame and the tube for obtaining this maximum intensity of sound. M. Kastner's merit is in having shown that, when two or several flames are introduced in a tube, they vibrate in unison, and produce the musical maximum of sound when they are placed one-third the length of the tube, and, if these two flames are brought in contact, all sound ceases directly, a phenomenon M. Kastner demonstrates to be caused by the interference of sounding flames. Here is a question, lately scarcely thought of, of which M. Frederick Kastner has determined the laws, at the same time making a most remarkable application of them in creating an instrument which reminds one of, and may be mistaken for, the sound of the human voice.
A very simple mechanism causes each key to communicate with the supply-pipes of the flames in the glass tubes. On pressing the keys the flames separate, and the sound is produced (Fig. 1). As soon as the fingers are removed from the keys the flames join, and the sound ceases immediately (Fig. 2). These new experiments made by M. Kastner upon singing-flames should cause all makers of musical instruments to turn their attention to inventions connected with sound. If two flames of suitable size be introduced into a glass tube, and if they be so disposed that they reach one-third of the tube's height, measured from the base, the flames will vibrate in unison. This phenomenon continues as long as the flames remain apart, but the sound ceases as
