Abbe Haiiy’s Observations on Double Refraction, 149 
The following are the general views given by this eminent 
author in his Traite de Miner alogie, vol. i. p. 230, which ap- 
peared in the year 1801. 
“ The quantity of double refraction,” says he, u or, what is 
the same thing, the magnitude of the angle formed by the two 
rays, by means of which the eye sees the two images, varies from 
one substance to another, all other things being similar, accord- 
ing to the nature of the substances themselves. In Zircon , for 
example, the double refraction is very strong, while it is much 
less sensible in Emerald . Besides, this quantity varies in each 
substance from different causes. In general it increases or dimi- 
nishes with the refracting angle, or that which is formed by the 
two faces through which we look at the object. But there is 
another cause of variation, which combines itself with the pre- 
ceding, and which depends on the position of the refracting 
surfaces, relative to the faces of the primitive form ; and such 
is the influence of this cause, that, with two equal refracting 
angles differently situated, we may have distances perceptibly 
unequal between the images of the same object, and there is 
even a limit where the effect of double refraction becomes no- 
thing, that is, when the two images are combined into one. 
u This limit takes place in Quartz and in Emerald , where one 
of the faces containing the refracting angle is perpendicular to 
the axis. It takes place in sulphate of barytes, when one of the 
same faces being parallel to the axis, is at the same time pa- 
rallel to a plane passing through the long diagonals of the base 
of its primitive form. 
“ On this subject I possess only a small number of observations ; 
but it is probable that all the substances which have double re- 
fraction, are included in one or other of the two preceding cases, 
which give the limits of all the positions that the refracting sur- 
faces can have, relative to the primitive form. But, as the po- 
sition parallel to the axis is variable in its turn between several 
limits which correspond to the diagonals, and to the sides of the 
base of the primitive form, it is necessary to know which of 
these last limits is that which belongs to each substance.” 
64 1 shall show, under the article Emerald, how a mistake con- 
ducted me to these results ; and I confess that there remains 
still an uncertainty respecting the refraction of some substances, 
