Thus, for example, for water at \(\lambda=0.3\) mm the refractive index is approximately the same as in the visible spectrum, while at \(\lambda=30\) mm it is equal to 8.5, which is very close to the square root of the dielectric constant, as it should be for infinitely long electric waves. At the present time this region is being investigated (see the preceding abstract) by a more perfect method than was previously accessible for waves of 27, 18, 11, 8, and 4 mm for the following liquids: water, glycerin, methyl and ethyl alcohol.

The refractive index is calculated by the Cauchy–Quincke formula from the coefficients of reflection and absorption. In addition, an approximate value of the refractive index is obtained directly from the reflection of waves from the two surfaces of a liquid layer poured onto the surface of mercury; if the thickness of the layer is varied, the reflected energy passes through a series of maxima and minima, which make it possible to measure the wavelength in the liquid. The results obtained by the two methods are in good agreement.

S. Rzhevkin.

Minimum sound energy at high frequency perceived by the ear.

C. E. Jane. Minimum Sound Energy for Audition for Tones of high frequency. Phys. Rev., May 1922.

For frequencies from 2,000 to 14,000 \(\dfrac{\text{per}}{\text{sec}}\), it was found that, for the threshold of auditory sensation, an energy of \(7{,}10^{-8}\ \dfrac{\text{erg}}{\text{cm}^2\ \text{sec}}\) is required (energy flux in 1 sec through 1 cm²).

This quantity is practically the same for the entire indicated frequency interval, although small differences exist not only between individual persons, but also between the two ears of one and the same person.

Above 14,000 \(\dfrac{\text{per}}{\text{sec}}\), ever greater quantities of energy are needed in order to produce the sensation of sound, and at frequencies from 18,000 to 20,000 \(\dfrac{\text{per}}{\text{sec}}\) the energy, 1,400,000 times greater than at low frequencies, gives only the threshold of sensation. This upper limit is higher for children and lower for old people.

28/XII 1922.
S. Rzhevkin.

Short electric waves.

E. F. Nichals and J. D. Tear. Short Electric Waves. Journ. of Francl. Inst., V. 194, No. 5, p. 683, Nov. 1922 (Abstract).

The authors of the present work have apparently succeeded, so far as may be judged from the cited preliminary communication, in taking a large step forward in the problem of obtaining short electric waves.

The vibrator is a modified form of Tear’s dipole, with certain additions that make it possible to obtain shorter waves. The spark discharge takes place between the bases of two tungsten cylinders immersed in kerosene. The cylinders were soldered to the ends of thin glass tubes made of fusible glass and were set by means of a micrometer, which made it possible to regulate the spark gap between them; the smallest cylinders used had a diameter of 0.2 mm and a height of 0.2 mm. The high-voltage leads were brought inside the glass tubes to within a distance of several millimeters from the tungsten cylinders; the spark jumped to the cylinders through an air gap, which was constantly blown through with compressed air for the purpose of deionizing and cooling the place where the glass was soldered to the tungsten. With a distance between the cylinders of about 0.01 mm, each discharge produces

electric oscillations in the dipole, strongly damped, whose wavelength, as experiments have shown, is 3–5 times greater than the length of the dipole (the factor 5 applies to small dipoles).

The vibrator was placed on an optical bench at the principal focus of a paraffin lens; then the waves passed as a bundle of parallel rays through an interference apparatus or another optical system and were collected by a second lens onto the receiving apparatus.

As the receiving apparatus Nichols’s radiometer was used, which made it possible to obtain much greater sensitivity than the previous methods (a thermoelement with a galvanometer, a thermogalvanometer); the total weight of the moving system of the radiometer was about 0.5 mgr. The receiving elements consisted either of platinum wire 1 μ in diameter and of suitable length, in order to obtain resonance with the received radiation, or of a metallic film deposited on sheets of mica or quartz thread by evaporation in a vacuum; these receiving elements were placed in place of the usual blackened vanes of the radiometer. Thin mica shields, placed at a distance of about 0.1 mm in front of or behind the receiving element, made the radiometric effect of both vanes one-sided.

The wavelength was measured with a Boltzmann mirror interferometer, and also with a reflecting echelon consisting of precisely calibrated brass plates. The echelon was also used for an approximate analysis of the dipole radiation and for isolating more homogeneous radiation of the desired wavelength.

By the method described, electric waves ranging from several centimeters down to 1.8 mm in length were isolated. Under favorable conditions a wavelength of 0.8 mm was also accessible to observation.

Thus the electric spectrum can apparently be studied more completely than has hitherto been possible,¹ by another two or three octaves toward shorter waves.

S. Rzhevkin.

¹ See, for example, the abstract on the work of Möbius, Advances in the Physical Sciences, Vol. III, issue 1, p. 121.