crystallized surfaces upon light. 149 
tinguished at pleasure in any doubly refracting crystal ; and 
the crystal is thus converted into a singly refracting crystal, 
like certain specimens of agate. 
4. In soft crystals that do not admit of a perfect polish, 
the distinctness of any of the two images may be made a 
maximum, by giving the crystal the best polish of which it is 
susceptible, and then cementing plates of glass upon its sur- 
faces, by a transparent cement of the same refractive power 
as that of the pencil which is to be rendered most distinct* 
If it is required to make the two images equally distinct, the 
refractive power of the cement must be a mean between that 
of the ordinary refraction, and the extraordinary refraction 
which corresponds to the angle which the refracted ray forms 
with the axis of double refraction. 
5. All doubly refracting crystals consist of an ordinary and 
an extraordinary medium, alternating with each other, and 
varying in density according to a law which I have described 
in another paper.* 
I consider the optical structure of agate as demonstrating 
the existence of two media. In quartz, the two media are 
equally perfect and transparent ; but in certain specimens of 
agate, the one medium is seen in a separate state from the other, 
and broken down into small portions like the figures 333. 
The light which passes through these portions, is evidently 
acted upon by a different refractive power from that which 
• The paper here alluded to, was laid before the Royal Society of Edinburgh on 
the 16th of March, 1818; but as it could not have been understood without the 
preceding experiments, its publication was necessarily delayed. The theory which 
it contains embraces also the complex phenomena which arise from the combination 
of two or more axes. Seethe Phil, Trans, Lond, 1818, p. 264. 
