XV. An Instrument of precision for producing monochromatic light of any desired wave-length, and its use in the investigation of the optical properties of crystals
In the optical investigation of crystals it is of great advantage to command a ready means of illuminating the field of the observing instrument with light of any desired wave-length. The red, yellow, and green monochromatic light emitted by incandescent salts of lithium, sodium, and thallium has hitherto been considered sufficient for most crystallographical investigations. The disadvantages of employing such a source of monochromatic illumination are threefold. In the first place it is difficult to remove the last traces of the relatively more powerfully illuminating sodium from the lithium salt employed. The admixture of yellow with the red light is a very great inconvenience when determining refractive indices by the method of total reflection and when measuring the optic axial angle of biaxial crystals by observations of the separation of the hyperbolic brushes of the interference figures. In the latter case, owing to more or less dispersion of the axes for light of different wave lengths, the effect of the admixture of even a little of the highly illuminating yellow sodium light with the red lithium light is to diminish the definition of the brushes, the interference figures for the two colours being superposed, and thus to destroy the possibility of accurate measurement of the separation of the axes for lithium light. In the second place, the poisonous nature of the fumes of the volatile thallium salts renders it imperative that the green flame should be produced in a draught cupboard, and all observations conducted in front of it, a condition which it is frequently inconvenient to fulfill. The third and most weighty objection to this mode of producing monochromatic light is that it confines the observations to three wave-lengths, at considerable intervals apart, ceasing, however, with the yellowish-green, and leaving the blue end of the spectrum out of consideration altogether. For substances whose crystals exhibit very slight dispersion of the optic axes this may, perhaps, be conceded to be sufficient, although, even in these cases, the observations cannot be considered as complete. For the numerous substances, however, whose crystals are endowed with sufficient dispersion to exhibit considerable differences of optic axial angle, and (in crystals belonging to the two systems of least symmetry) differences in the directions of stauroscopic extinction, observations with light of only these three wavelengths are insufficient. Moreover, in the cases occasionally met with—such as the rhombic form of titanium dioxide known as brookite, the rhombic triple tartrate of sodium potassium and ammonium, and the monoclinic ethyl-triphenylpyrrholone described three years ago by the author,*—in which the dispersion is so large that the axes for red light lie in a plane perpendicular to that which contains them when illuminated by blue light, observations with lithium, sodium, and thallium light are totally inadequate to enable us to follow the change which must occur as the wave-length of the light is altered, and, except by mere fortuity, afford no means whatever of observing the interesting point when the wave-length is such that the axes coincide in the centre of the field and the biaxial crystal simulates a uniaxial one. It is evident, therefore, that for the complete investigation of the optical properties of crystals, an arrangement for procuring monochromatic light must be adopted which will enable us to illuminate the field of the observing instrument with the whole of the spectrum colours in succession. A step towards supplying such a requirements has been made by Fuess, the well-known crystallographical optician of Berlin, in his larger axial angle goniometer. In front of the objective of the polariscope are placed a small prism and a collimating tube, arranged at such an angle to the polariscope | that the light from a lamp passing through the slit of the collimator is dispersed by the prism into a spectrum, the whole of which is seen in the field on observing through the polariscope. The prism is capable of rotation, the amount of which is registered by a micrometer. It is intended that the readings of the micrometer shall be recorded for the coincidences of the vertical cross-wire of the polariscope with the principal lines of the solar spectrum, so that light of any particular wave-length may be brought into the centre of the field when using any artificial source of white light In practice, however, the author finds this arrangement unsatisfactory. The smallest amount of “backlash” in the working of the endless screw and wheel by which the rotation of the prism is effected introduces a considerable error in the reproduction of the setting for any solar line. But, even assuming the construction perfect at first and to remain so after use, the arrangement labours under the great inconvenience I that the whole, or, when the second power is employed, almost the whole, of the spectrum is visible at once. Although it may be true that a fair approximation to the value of the optic axial angle for any wave-length may be obtained in cases where the dispersion of the axes for different colours is not considerable, by bringing light of that wave-length to the vertical cross-wire (or between the pair of cross-wires) to which the hyperbolic brushes are also successively adjusted, still the rings and lemniscates surrounding the axes are distorted more or less according to the amount of dispersion by the other portions of the spectrum in the field of view. In cases where the dispersion of the axes is great the method fails altogether, for the interference figures become perfectly unintelligible.