But if such be the nature of polarized light, what notion must we form of the nature of common light? We have seen that common light, in passing through a rhomb of Iceland spar not limited by a screen, is resolved into two polarized streams, which, mixing on emer gence, yield a light having all the properties of common light. Nor does the light reflected from the first surface of the rhomb possess any other properties : it may be polarized like common light, or divided by another rhomb into two polarized streams which mix on emergence just as before, so that there is not the slightest ground for supposing that with the light lost by reflection the original beam loses any ele ment of a different character from that which it retains. It is easy to make out the nature of the vibration which results from the co existence of two series of rectilinear and transverse vibrations propa gated in the same direction and taking place in rectangular planes. The vibrations in such a case would take place in planes perpendicular to the direction of propagation, that is, would lie in the fronts of the waves.
Hence the following important suppositions are adopted in the theory of undulations.
The vibrations of the ether which constitute light, unlike those of air which constitute sound, take place, not to and fro in the direction of pro pagation, bat laterally in the tangent planes of the wares. In polariztd light the vibrations are rectilinear, and either parallel or perpendicular to the plane of polarization, a point which for the present may be left undecided. The polarization of common light consists in the resolution of the vibrations into two rectilinear series, the undulations belonging to one of which are obtained apart from those belonging to the other.
The theory of transversal vibrations was first suggested by Dr. Young, who was led to it by the consideration of certain phenomena of biaxal crystals discovered by Sir David Brewster. It suggested itself independently to the mind of Fresno!, who was led to it by con sidering the laws of interference of polarized light, made out by Arago and himself. It encountered, at first, much opposition, for it was entirely opposed to preconceived notions. It obliges us in fact to suppose that the vibrations which constitute light are carried on, not by forces by which the ether tends to resist condensation or rarefac tion, but by forces by which it tends to resist diatortion,imaccompanied by change of volume. We are driven to conceive that a sliding motion of the ether, layer over layer, is resisted by a tangential force thereby called into play : in fact, to treat the ether, so far as the vibrations of light are concerned, as if it were an elastic solid. Still, it would ill accord with the aplrit of inductive philosophy to reject a conclusion to which we are fairly led by the study of observed phenomena, because it clashes with preconceived notions respecting the nature of a medium of the very existence of which we have no evidence beyond the beautiful and simple explanation it affords of the curious and com plicated phenomena of light.
If once we adopt the theory of transversalvibrations, we are able not only to form a clear picture in the mind of what, to say the least, polarization may be, but to explain in the simplest manner both Malus's law and the laws of interference of polarized light. In fact, Malua's law and laws 2, 4, 5, of interference led to the theory of trans vernal vibrations ; and to explain these laws on the theory we have only to begin at the other end. Laws I and 3, which assert the neces sity of a polarizes in order that any colours should be seen in crystal line plates, follow at once from the consideration that the rectangular components of the vibrations in common light have no fixed and permanent relation to each other, either as to phase or amplitude, so that they are as good as the vibrations belonging to two streams from different sources.
The theory of transversal vibrations, again, enables us to form a perfectly clear notion of the nature of a peculiar modification of light, known as elliptic polarization. When polarized light is transmitted across a crystalline plate, the emergent light is in general elliptically polarized, being only iu particular cases polarized in the way hitherto considered in this article, or plane-polarized, as it is called in contradistinction to elliptic polarization. When tically polarized light is examined by an analyzer which is made to revolve, the light never wholly vanishes, but becomes alternately a maximum and a minimum at every quarter revolution. Thus far it agrees with partially polarized light, from which, however, it wholly differs in otherrespects, both theoretically and experimentally. [Ewen° POLARIZATIO:q Lastly, the theory of transversal vibrations is, to fay the least, very hopeful in its application to the explanation of double refraction, of the polarization of light by reflection, &c. The full explanation of these phenomena is retarded by two causes, first, there are the inherent mathematical difficulties of the problem, and secondly, which is far worse, in our Ignorance of the ultimate constitution of matter, and its relation to the luminiferous ether, we have to grope in the dark in our endeavour to find out what is the real state of things to which our mathematics have to be applied. Still, even in this department many important laws have already been explained, and frequently even pre dicted, on dynamical principles, though not always with perfect rigour; but no dynamical theory has yet accumulated in its favour such a weight of evidence as to exclude others, nor can even the question whether the vibrations of polarized light are parallel or per pendicular to the plane of polarization be yet considered nettled, though the probability seems to be in favour of the latter view, which was that originally adopted by Fresnel.
In this article the subject of double refraction has not been touched execpt so far as was necessary for the explanation of polarization. For further information on this point see a subsequent article, UsranA