When light is incident on the generality of crystallised bodies, the ray is refracted In two directions, one of which in uniaxal crystals obeys the ordinary law of refraction, but neither in biaxal crystals. On the theory of emission the forces cannot here be simply normal to the faces of the crystals, but have a connection with the directions of the axes of crystallisation, while on that of undulation the inertia of the ether within such bodies, or else its elasticity, is different in different directions, and the form of the wave-surface ceases to be spherical. The further consideration of that subject will be resumed in the article POLARISATION OP LIGHT.
The formation of foci and images by reflection and refraction follows from the simple laws here discussed, for an account of which the reader may refer to Orrice ; the description of the instruments, con structed to take advantage of the properties of light being given in Less ; MICROSCOPE; MIRROROPTICS, PRACTICAL ; TELESCOPE.
The laws of the reflection and refraction of light admit, as we have seen, of being satisfactorily explained according to either of the two theories respecting which the scientific world was once divided. But while the undulatory theory explains the interference of light, and traces into their most minute details the curious and complicated phenomena of diffraction, no satisfactory account has ever been given of these phenomena on the theory of emission. The remarkable phenomena of polarisation again, so difficult to conceive on the theory of emission, fall simply and naturally into their places on the theory of undulations, when once the doctrine of transversal vibrations is ad mitted. And even to go no farther than the laws of reflection and re fraction, which entail the necessity of making opposite suppositions, on the two theories, respecting the velocity of light within refracting media, the crucial experiment of already alluded to .Annales de Chimie; § 3, tom. 41, p. 129) is decisively in favour of
the theory of undulations.
It must be confessed that there is one phenomenon, perfectly simple according to the theory of emission, which presents a serious diffi culty on the theory of undulations, namely, the astronomical phe nomenon of aberration, according to which a star appears displaced from its true position, in a direction towards the point towards which the earth is moving, by a small angle measured by the ratio of the velocity of the earth to the velocity of light, multiplied by the sine of the angle between two lines, drawn, one in the direction of the motion, the other from the earth to the Ater. The com mon explanation of this phenomenon involves the tacit assumption that the rectilinear propagation of light is not affected by the motion of the earth. We might have expected beforehand that, just as a ship moving through the water would disturb in its neighbour hood the course of little ripples with which the surface might be covered, so the earth in moving round the sun would push the ether out of its way, and thus disturb the course of progress of the little tremors of which light coneiste. To get:rid of this difficulty, many adopt Dr. supposition, that the earth allows the ether to Flail freely through/it as the wind does through a grove of trees. The rectilinear propagation of light, and consequently the phenomenon of aberration, is however explicable without making this rather startling supposition, by supposing the motion of the ether consequent upon the motion of the earth to be of a particular kind. See several articles In the ' Philosophical 31agazine; 3rd series, vole. 27 to 32.
The production of colours by ordinary refraction le considered in the articles and Itatenow ; for that produced by light passing near the edges of bodies, and by interference, the reader may consult Dirreacrfoe and Iereerrersce. For the colours of plates, see