620 SIR DAVID BREWSTER ON CIRCULAR CRYSTALS. 
bells already described, in which oblique refraction and the thinness of the film 
combine to produce beautiful coloured rings, with a black cross. 
In other crystals, such as Muriate and Citrate of Quinine, Codeine, and Nitrate 
of Codeine, I have observed the luminous sectors, and the black cross round the 
air-bubbles, which are formed after fusion, a phenomenon exactly the same as 
that which takes places round cavities in diamonds, amber, and other substances. 
Having thus described the principal phenomena of circular crystals, I shall 
now proceed to make a few observations on their formation and decomposition. 
Circular crystals are abnormal aggregations, which owe their existence to some 
disturbing cause. The natural tendency of the elementary molecules of the most 
perfect of them, is to combine with their homologous axes parallel to one another, 
and to form regular crystals; and it is only when this tendency is counteracted 
by the quick application of heat or cold, by pressure, or by the nature of the sol- 
vent or of the combined ingredients, as in the case of borax and ‘phosphoric acid, 
that the molecules are constrained to arrange themselves round a centre, not 
merely in radiating prisms, as in Wavellite and some other minerals, but accord- 
ing to laws which could not have been anticipated from any known principles of 
crystallisation. If, owing to any disturbing cause, two molecules should be de- 
posited with their axes at right angles to each other, or four with their similar 
poles directed to the same point, this will lead to the formation of a circular dise, 
which will be of limited thickness, if the crystallisation takes place between two 
plates of glass pressed together, or to the formation of a spherical crystal, as in 
the Lithoxanthate of Ammonia, when there is room for its growth in all directions. 
The disc, or the sphere, might thus increase to a considerable size, if there was 
only one centre of crystallisation, but as the same causes have been operating all 
around, the size of the circular crystal is limited by the number of molecules 
within its sphere, or by its junction with the other discs around it. In this last 
case, they form a sort of mosaic, in which their shape is not circular, but hexa- 
gonal, as in manna, oil of mace, and many other substances. 
In the greater number of circular crystallisations, the tints are a minimum at 
the centre of the disc, and increase outwards,—that is, the molecules form a 
thinner film at the centre, which increases in thickness towards the circumference ; 
but in other cases the reverse of this takes place, and in the disc represented in 
Fig. 3, where the tints are those of Newron’s rings, some cause, which we cannot 
even conjecture, must have determined the atoms to unite according to the com- 
plex law which connects these tints with the thicknesses at which they are pro- 
duced. A cause of an opposite kind must have given birth to the disc shewn in 
Fig. 4, where the molecules form a thick film at the centre, which diminishes in 
thickness from nine to three as the tint passes from the central blue to the white 
at the circumference. 
