Gauze Tones.—Rijke in 1859 made the discovery that a sound of considerable power may be produced by a heated piece of fine metal gauze stretched across the lower part of a vertical open tube containing air. In the earlier experiments the gauze was heated by a gas flame and the sound was observed immediately after the removal of the flame. Keeping the gauze hot by means of an electric current the sound may be maintained indefinitely. The maintenance depends on the variable transfer of heat due to the motion of air through the gauze, this motion being a uniform upward convection current with a superposed alternating motion due to the resonant vibration of the air in the tube. In the lower half of the tube the alternating flow assists the direct flow a quarter period before the phase of greatest condensation and opposes it a quarter period after this phase. Bosscha and Reiss (1859) demonstrated the complementary phenomenon. If a current of hot air fall on a cold gauze, sound is produced—in this case the phase relationship requires the gauze to be in the upper half of the tube, preferably about a quarter the length of the tube from the top.
The Thermophone. Fine Wires or Strips Heated by Alternating Currents.—When an alternating current i sinnt is passed through a fine wire of resistance R the heat developed is proportional to sin' nt = (1 — cos . Consequently the heat developed will vary between o and at a frequency 2n/271-, that is, at twice the frequency of the current. If a sufficiently large initial direct current be passed through the wire the double frequency term in R(i0-1- i or sinnt — cos2nt may be made negligible—the fluctuations of heating effect then vary with the frequency n/27r- of the current.
Using very fine platinum wires thick), P. de Lange (Roy. Soc. Proc., 91, Ap. 1, 1915) demonstrated that the decrease and increase of heat took place synchronously with the A. C. supply. The air surrounding the wire was thereby rapidly heated and cooled, the corresponding expansions and contractions ap pearing as sound. Ordinary telephonic currents, of speech fre quency, were shown to be sufficient to produce easily audible sounds in a small earpiece containing the fine wire heater and a small volume of air. This device is known as a Thermophone. The reproduction is of good quality, but somewhat feeble relative to the response of electro-magnet telephone receivers. Provided the frequency is not too high, the thermophone has a possible application as a metrical source of sound.
Directional Sources of Sound.—It is evident that a greater range of transmission will result if a given amount of sound energy is confined to a cone with a small angle of divergence instead of spreading uniformly in all directions. Other advantages
of a directional transmitter will be equally apparent. Certain sources of sound exhibit inherent "directional " properties, due to the fact that their dimensions are considerably greater than a wave length of the sound they emit. Other sources may show no inherent directional property but may be made directional by the use of some attachment such as a trumpet or a mirror. A line or an area of equally spaced non-directional sources vibrating in phase may act as a directional compound source.
Sources of Large linear dimensions not small com pared with a wave-length. Rayleigh (vol. p. 138) refers to the directional properties of a speaking trumpet when used to trans mit high pitched sounds, such as a whisper or a hiss. An instruc tive example of this nature is the case of radiation of high fre quency sound waves from one face of a quartz piezoelectric "piston" oscillating with uniform amplitude and phase at all points of its surface (see R. W. Boyle, Proc. Roy. Soc. Canada, III, 1925, p. 167). The sound-distribution round such a source is analogous to that of plane waves of monochromatic light falling on an aperture and forming a diffraction pattern beyond. Con sider the case of a circular piston source of radius a radiating waves of length X (X = c/N where c is the velocity of sound in the medium and N the frequency of vibration). As in the optical case, the sound is a maximum along the axis of the piston where all the elementary disturbances from the various points of the piston arrive in the same phase. In directions inclined to the axis the intensity is less, diminishing steadily to a minimum when the difference in distance to the nearest and the furthest points of the piston is rather more than half a wave-length. In this case, the effect of the further portion of the disc is just neutralised by the effect of the nearer portion. In directions still more inclined, the sound increases again to an intensity equal to 0.017 of that along the axis. This is succeeded by another mini mum value and further maxima of small intensity—correspond ing to the various diffraction rings in the optical analogue. The angle a at which the first silence occurs is Thus the central "beam" of sound will be confined to a cone of small angle when the radius R of the piston is many times the wave-length. The polar distribution of amplitude is somewhat similar to that shown in figure 8 for a line of small nondirectional sources. When the radius R does not exceed X/4 the elementary disturbances from the piston combine without much opposition in phase, and the intensity is nearly the same in all directions—the case of a non directional source of sound.