others equal to it in energy but of an opposite character situated at right angles to it and to each other, with Dr. Brewster, or as being itself the real axis of polarization. For the resultant axis being the same for all the colours, the partial actions of each of the supposed axes on the former hypothesis, having the same point of compensation for all the colours, must be equal to each other and to the resultant force for them all. The mere fact therefore of a deviation from Newton's scale, however enormous in the tints of any regular crystal with one axis, cannot be regarded as affording of itself any argument for the substitution of two others for it in that particular substance, because each of such axes acting separately would exhibit a scale of tints perfectly identical with that of the axis whose place they supply, and therefore by parity of reasoning should be regarded as the resultant of two others, and so on, ad infinitum. This reasoning appears to me conclusive against any analogy between crystals with one and two axes, founded on a deviation of tints in the rings of the former. But I cannot help regarding the phenomena I have described as affording considerable support to the very ingenious theory of the philosopher just mentioned, as applied to crystals with two axes, inasmuch as they establish the existence of that diversity in the scales of action of the simple or elementary axes, without which their points of compensation (or the poles of the lemniscates they exhibit in polarized light) must of necessity be coincident for all the simple colours, a coincidence which, as has been already remarked at the beginning of this paper, seldom or never takes place. This I conceive to be the view which Dr. Brewster himself has taken of the phenomena of the deviation in crystals with two axes, and to afford ocular demonstration of the existence of what he has called the different dispersive powers of his elementary axes.

SLOUGH, Feb 19, 1820.

F

J. F. W. HERSCHEL.

III. On the Rotation impressed by Plates of Rock Crystal on the Planes of Polarization of the Rays of Light, as connected with certain peculiarities in its Crystallization.

By J. F. W. HERSCH EL, M. A.

FELLOW OF THE ROYAL SOCIETIES OF LONDON AND EDINBURGH,
CAMBRIDGE PHILOSOPHICAL SOCIETY.

AND OF THE

[Read April 17, 1820.]

THE HE phenomena of the developement of colour in plates of Rock Crystal when polarized light is transmitted through them in the direction of the axis of double refraction, and analysed at its egress by a doubly refracting prism, were originally noticed by M. Arago in 1811, and have since been examined in a very masterly manner by M. Biot in two Memoirs communicated to the French Institute in 1812 and 1818. It results from the experiments described in the latest of these, first, that a plate of this substance exposed to polarized light in the manner above described, possesses the singular property of displacing the plane of polarization of an incident ray, and turning it aside in one invariable direction, through an angle always proportional to the thickness of the plate, so that at its egress, the plane of polarization will assume the same position as if it had revolved in one direction during the passage of the ray through the whole thickness with an uniform angular velocity depending on the nature of the ray. Secondly, that this rotation of the plane of polarization, (for so we may be permitted to call it, though we

know not whether any rotation actually does take place within the crystal, but only that the result is as if it were so) though invariable in its velocity for rays of the same colour, differs for those of different colours, being greater for the more refrangible rays, and that in the inverse ratio of the squares of the lengths of their fits in vacuo.

Thirdly, that although the direction of this rotation with respect to the observer (who is supposed to receive the ray in his eye) be invariably the same in the same specimen of rock crystal, whatever be the length of the ray's path within it, or whichever side of the plate be turned towards him, yet it differs in different specimens ; in some being always from right to left, and in others from the left to the right of the observer so placed.

Fourthly, that this singular property is common to rock crystal with a variety of other bodies, and among the rest with various liquids, such as oil of turpentine, solutions of camphor, sugar, &c. in all which precisely the same law of rotation of the differently coloured rays is observed, the absolute velocity only differing. Many of them carry this property with them into their chemical combinations, solutions, and mixtures with other substancesthey preserve it in their solid, fluid, and even gaseous form, with an energy proportional to their actual density, nor can any thing short of the decomposition of their molecules deprive them of it.

From these circumstances, as well as from a number of delicate and accurate experiments on the compensation of opposite rotations by the mixture of different liquids, and others which it is not necessary to enumerate here, M. Biot has concluded that there exists a property inherent in the ultimate molecules themselves of certain bodies, independent of their regular disposition in crystalline forms, their state of aggregation or proximity to each other, in virtue of which each individual molecule turns round the plane of polarization of a ray traversing it, through a certain minute but

determinate angle depending only on the nature of the substance and of the ray. "Cette propriété consiste dans la faculté qu'ont les molécules dont il s'agit de faire tourner d'un certain angle, et certain sens les axes de polarization des rayons

un

"dans lumineux.'

Mysterious as such a property may seem and entirely opposing all our preconceived notions of the nature of the ultimate particles of bodies, unless we deny the precision of M. Biot's experiments, it is hardly possible to refuse our assent to the general tenor of the conclusions he deduces from them. Admitting then, for the present the truth of the inference, we are left to conjecture the circumstances in the intimate constitution of the molecules which can determine an invariable tendency of a ray of light to turn its plane of polarization in one direction rather than another, in whatever way they may be presented to it; and though it seems very difficult to form a definite hypothesis to explain the fact, yet the general impression left on our minds is that of a want of symmetry in the disposition within the molecules themselves, of some of the elementary forces by which they act on light; a thing not incompatible with perfect symmetry in their external forms, or in the distribution of the more powerful forces which determine the laws of their aggregation in regular crystals. Now so far as the action of crystallized media on light has been examined, there appears to exist an intimate connexion between the crystallographical and optical properties of bodies, and as we have every reason to imagine that the forces by which the particles of matter act on light and on each other, do not differ essentially in their nature, it is easy to conceive that any deviation from perfect symmetry in the distribution of even subordinate forces of any kind, will in some degree influence the molecules in their mode of aggregation with one another, and however feeble, yet, being a cause constantly in action, may possibly, under favourable circumstances, produce a sensible