2 Dr. BrewsTER on the Optical Properties 
I. On the Distribution of the Colouring Matter in Topaz. 
In order to obtain results that might be considered as general, 
I ground and polished the flat summits of a great number of 
Brazilian Topazes, and having exposed them to polarized light, 
I observed the following structures. 
1. The internal part of the crystal was almost always of 
a different colour from the external part, as will be afterwards 
more particularly described, but the pink colouring matter was 
confined to two small rhomboidal prisms placed at the acute 
angles of the prism of Topaz as shewn in Fig. 1. When the 
shorter diagonal of the rhomb is in the plane of primitive 
polarization, these small rhomboidal prisms, which are of a pale 
pink tint by common light, assume a deep and brilliant pink 
colour, and they become of a faint blue approaching to a Lilac 
colour when the longer diagonal is m the same plane. In 
some specimens this complementary tint, in place of beng Lilac, 
is yellow of different characters; but in every specimen which 
I have examined, the pink, whenever it exists, is the colour 
which appears when the shorter diagonal is in the plane of 
primitive polarization. 
2. In some crystals the red colouring matter is confined to 
a triangular prism as in Fig. 2. sometimes occupying the apex 
of the acute angles of the outer rhomb, and sometimes the same 
angles of the inner rhomb. In the specimen represented in 
Fig. 2, about one half of the pink portions becomes lilac before 
the other half, and before the longer diagonal comes into the 
plane of primitive polarization; thus indicating a species of 
hemitropism, or a want of parallelism in the principal sections 
or neutral axes of the different portions of the thick prism. 
This effect is shewn in Fig. 9, where no fewer than four 
different colours appear at the same time. 
