126 Dr Brewster on tlie Refractive Powers, and other 



pidary in Edinburgh, there are no fewer than four strata of 

 cavities nearly parallel to each other, and in the thickness of 

 one-eighth of an inch. 



In the distribution of most of these groupes, accident seems 

 to have had the principal share ; but there are certain modes 

 of distribution that appear to be the result of some general prin- 

 ciple. In a specimen, for example, belonging to Mr Sander- 

 son, an immense number of cavities are arranged in rectilineal 

 groupes, radiating from a centre. Each rectilineal group con- 

 sists of two, or in some places three, rows of cavities, and se- 

 veral of the radiations are bent from their original direction. 

 The spaces between each pair of rows are filled with curiously 

 branching cavities, some of which are half an inch long ; but 

 the remarkable fact is, that these cavities are connected with 

 numerous slender branches, many of which communicate with 

 a single cavity in the nearest rectilineal row of the radiations 

 between which the long cavities are placed. 



In all the cavities of this remarkable specimen capable of 

 being examined, there are found both the new fluids, with the 

 exception of a long branching cavity, from which they had 

 escaped, in consequence of the end being cut by the lapidary. 

 The dense fluid always occupies the filamentous branches. 



The plane in which these cavities lie is perfectly flat, and is 

 nearly perpendicular to the axis of the prism. 



2, On the Form of the Cavities containing the New Fluids. 



In a specimen of topaz belonging to Mr Sanderson, and 

 which is one of the most valuable that I have seen, each cavi- 

 ty consists of a variety of cavities of different lengths and 

 sizes, bounded by parallel lines, and communicating by nar- 

 row channels, which almost escape the cognizance of the mi- 

 croscope. In these cavities, the two new fluids are arranged 

 in the most remarkable manner, the dense fluid occupying all 

 the necks, and angles, and narrow channels, while the expan- 

 sible one is left in the open and less capillary spaces. When 

 the heat of the hand is applied to the specimen, the fluids in 

 the cavities are all set in motion. The dense fluid quits its 

 corners, and assumes new localities ; and the different portions 



