When the line connecting the two axes of the crystal is inclined 45° to the plane of primitive polarisation, a cross is seen on revolving the potassium nitrate; it gradually assumes the form of two hyperbolic curves, as in Fig. 185. But if the tourmaline be again revolved through half a quadrant, the black cross will be replaced by white spaces, as in the second figure. These systems of rings have, generally speaking, the same colours as those of thin plates, or as those of a system of rings revolving around one axis. The orders of the colours commence at the centres of each system; but at a certain distance, which corresponds to the sixth ring, the rings, instead of returning and encircling each pole, encircle the two poles as an ellipse does its two foci. If the thickness of the plate of _nitre_ be diminished or increased, the rings are diminished or increased according to the thickness of the crystal. Small specimens of various salts may be crystallised and mounted in Canada balsam for viewing under the stage of the microscope; by arresting crystallisation at certain stages, a greater variety of forms and colours will be obtained: we may enumerate salicine, asparagine, acetate of copper, phospho-borate of soda, sugar, carbonate of lime, potassium chlorate, oxalic acid, and all the oxalates found in urine, with the other salts from the same fluid, a few of which are shown in Plate VIII. The late Dr. Herapath described a salt of quinine, remarkable for its polarising properties. The crystals of this salt, when examined by reflected light, have a brilliant emerald-green colour, with almost a metallic lustre; they appear like portions of the elytræ of the cantharides beetle, and are also very similar to murexide in appearance. When examined by transmitted light, they scarcely possess any colour, there is only a slightly olive-green tinge; but if two crystals, crossing at right-angles, be examined, the spot where they intersect appears perfectly black, even if the crystals are not more than one five-hundredth of an inch in thickness. If the light be in the slightest degree polarised--as by reflection from a cloud, or by the blue sky, or from the glass surface of the mirror of the microscope placed at the polarising angle 65° 45′--these little prisms and films assume complementary colours: one appears green, and the other pink, and the part at which they cross is chocolate or deep chestnut-brown, instead of black. Dr. Herapath succeeded in making artificial tourmalines large enough to surmount the eye-piece of the microscope; so that all experiments with those crystals upon polarised light may be made without the tourmaline or Nicol’s prism. The finest rosette crystals are made as follows:--To a moderately strong solution of _Cinchonidine_ add a drop or two of Herapath’s test-fluid.[33] A few drops of this is placed on the centre of a glass slide, and put aside until the first crystals are observed to be formed near the margin.