Page:Herschel - A Preliminary Discourse on the Study of Natural Philosophy (1831).djvu/46

 plane and extraordinarily refracted, while the other followed the ordinary rule. Accordingly this was supposed to be the case; and not only so, but from some trials and measurements purposely made by a philosopher of great eminence, it was considered to be a fact sufficiently established by experiment.

(23.) Perhaps we might have remained long under this impression, for the measurements are delicate, and the subject very difficult. But it has lately been demonstrated by an eminent French philosopher and mathematician, M. Fresnel, that, granting certain principles or postulates, all the phenomena of double refraction, including perhaps the greatest variety of facts that have ever yet been arranged under one general head, may be satisfactorily explained and deduced from them by strict mathematical calculation; and that, when applied to the cases first mentioned, these principles give a satisfactory account of the want of the extraordinary image; that when applied to such cases as those of rock-crystal or Iceland spar, they also give a correct account of both the images, and agree in their conclusions with the rules before ascertained for them: but so far from coinciding with that part of the previous statement, which would make these conclusions extend to all crystallised substances, M. Fresnel's principles lead to a conclusion quite opposite, and point to a fact which had never been observed, viz. that in by far the greater number of crystallized substances which possess the property of double refraction, neither of the images follows the ordinary law, but both undergo a deviation from their original plane. Now this had never been observed to be the case in any