Arguments are adduced to show that the doublet centred at 4152 Å occurring in the spectrum of all fluorescing diamonds arises from ‘forbidden’ transitions analogous to the forbidden³P-¹S transitions in the spectrum of C I. On the basis of this analogy, fluorescence lines are also expected to occur at about 8816 Å and 7849 Å, analogous to the³P-¹D and¹D-¹S transitions in C I. Of these, the former should also occur in absorption, while the latter should not occur in absorption and should be emitted only if the exciting radiation has a wavelength shorter than 4152 Å. A line has actually been found at 7930 Å, satisfying the latter conditions. The former line could not be recorded being outside the limit of sensitivity of the photographic plate.

Phosphorescence patterns in a dozen cleavage plates of diamond revealing the variations in intensity of phosphorescence in different portions of the same diamond have been successfully recorded by the method of contact photography. Comparison of these with the fluorescence patterns in the same diamond shows that the yellow phosphorescence emission corresponds to the blue fluorescence patterns, and that the yellow fluorescence patterns, if present, are not recorded in phosphorescence. Geometric patterns such as parallel bands, triangles, hexagons and spirals have been obtained in many cases.

When a plane-polarised beam of the intense λ 2537 radiation from a water-cooled, magnet-controlled mercury arc is sent along the optic axis of a perfectly clear and transparent sphere of quartz crystal free from inclusions, the track when photographed from a direction transverse to the beam exhibits striking fluctuations in intensity along its length. The effect, which is reproduced in the paper, is obviously connected with the rotation of the plane of polarisation of the polarised λ 2537 radiation as it traverses the crystal. The distance between one dark band to the next corresponds closely to a rotation of 180° of the plane of polarisation of the λ 2537 radiation. A similar effect was observed and photographed in smoky quartz by the present Lord Rayleigh in 1919 using the Tyndall scattering of visible light by the inclusions in the crystal. In the present case the effect is due to the diffusion of light arising from the atomic vibrations in the crystal lattice, and the clearness of the bands indicates that such scattering is strongly polarised.

The electronic line at λ 4156 present in the fluorescence spectrum of all diamonds sharpens and increases in peak intensity on cooling the diamond;per contra, it broadens and decreases in peak intensity on increasing the temperature of the diamond and finally at 350°C. is lost in the background. The integrated intensity of the line has been investigated and found to be unaltered to any marked degree between − 180° C. and +150° C.

The phosphorescence of diamond at various temperatures has also been dealt with. At low temperatures the after-glow is weak and greenish-yellow in colour, while at 400° C., the glow is extremely intense and blue. The effect of quenching by red or green light on this bright and blue glow has been reported in this paper.

As in the case of quartz (Chandrasekharan, 1948), using the property of optical activity, the selection rules for Raman lines of sodium chlorate have been experimentally verified using λ 2537 excitation for two different orientations of the crystal. Its 15 Raman lines have been classified as follows into the various symmetry classes possible for a cubic crystal:— 70 (E), 83 (F), 103 (F), 123 (F), 131 (A), 179 (E), 482 (E), 487 (F), 627 (A+F), 933 (F), 936 (A), 959 (E), 966 (F), 984 (F), 1026 (−). The results agree with earlier ones of Couture and Mathieu (1948) except for the Raman line 131 cm.⁻¹ which they classify as belonging to Class E.

The Brillouin components are well polarised and weaker than the symmetric Raman line 936 cm.⁻¹

For the first time, the theory of Doppler shifts in thermal scattering of light in birefringent crystals is worked out and the magnitude of the shift Δν of the components is given by$$\Delta v/v = \pm v_e /c)\sqrt {n_i ^2 + n_s ^2 - 2n_s n_s cos \theta } $$, where Δν is the frequency of the incident light,c the velocity of light in vacuum,v_e, the velocity of the elastic wave effective in scattering andn_i andn_s are either of the refractive indices of the crystal for the incident and observation directions. Sincen_i andn_s can each take two values, there are four pairs of values (n_i,n_s) and furtherv_e takes three values. Therefore, there mustin general betwelve pairs of Doppler components in the light scattered along a particular direction. The twelve pairs can be divided into four species each with a specific pair of values (n_i,n_s) and consequently specific polarisation character. They can be studied individually by the use of proper polarising devices in the incident and scattered paths. Each species consists of three pairs of components arising from the elastic waves of wavelength$$\lambda _e = \lambda /\sqrt {n_s ^2 + n_s ^2 - 2n_s n_s cos \theta } $$, whereλ is the wavelength of the incident light in vacuum and propagated along a specific direction. For any particular species, the scattering must be appropriately regarded as “coherent reflection” or “coherent refraction” of light waves by the effective elastic waves according as cosϑ<n_i/n_s andn_s/n_i or cosϑ>n_s/n_i orn_s/n_i. There can in general be 3 pairs of Doppler components withfinite shifts in the exactlyforward scattering.

In singly refracting crystals (n_s/n_i=n_s/n_s=n) the expression for shift reduces to the familiar expression Δν/ν=±(n_s/v_e/c)2n sinϑ/2 and in this case there could only be three pairs of Doppler components arising from “specular reflection” of light by elastic waves.

Sodium chlorate is the first crystal belonging to the tetrahedrite class (T) of the cubic system for which photoelastic constants have been measured. Since the crystal exhibits optical activity and no birefringence in the absence of stress, special techniques have to be adopted for measuring the birefringence introduced by stress. This has been done by the use of a petrological microscope in conjunction with the stressing apparatus and measuring the ellipticity and other characteristics of the light transmitted by the crystal. A particularly simple method is to use the “elliptic analyser”. Since the crystal does not possess four-fold axes, the relative orientation of the X- Y- and Z-axes was determined by means of X-rays. From observations on crystals compressed along [100], [110] and [111] directions,q₁₁–q₁₂,q₁₁–q₁₃ andq₄₄ were evaluated. All the four constants were independently obtained by combining these with polarisation measurements of light diffracted by ultrasonic waves in the crystal. The values are:q₁₁=1·48,q₁₂=3·88,q₁₃=2·89,q₄₄=−1·58×10⁻¹³ cm.² dyne⁻¹;p₁₁=0·173,p₁₂=0·258,p₁₃=0·223,p₄₄=−0·0187.q₁₂ andq₁₃ are different as is to be expected from Bhagavantam’s theory for crystal classes T and T_h. For stress along X-axis, the values of birefringence for observation along the Y- and Z- axes differ by as much as 70%, which is the largest observed so far for a cubic crystal.