300 



Prof. W. J. Sollas. 



are brought just into contact. Considering the want of exactitude in 

 our knowledge of the specific gravity of silver sulphide, the requisite 

 amount of rotation may be obtained by construction, the errors 

 attendant on this method lying within the errors of observation. Let 



FIG. 9. 



the centres of two octahedral elements be projected on a plane, which 

 cuts the two rectangular axes on which they lie at 45, and is parallel 

 to the third, i.e., parallel to two of the diagonals of the cube of refer- 

 ence. Then let each octahedron be supposed to rotate through a, 

 complete circle, the centres of the paired atoms will then project into 

 an ellipse which may be used as a kind of directrix. The spheres 

 representing the paired atoms will always project as circles. Find the 

 point on the ellipse at which the centres of two circles representing the 

 silver atoms must be situated so as just to touch. The angle which a 

 line joining one of these points to the centre of the octahedron makes 

 with that joining the centres of the two octahedra is that sought. In 

 the case* of silver sulphide it amounts to 21 52'. 



Thus, while preserving the dimensions we had previously obtained 

 for the atoms of silver and sulphur, we arrive at a structure which 

 completely explains the cleavage of silver sulphide, and satisfies the 

 conditions of symmetry imposed by the system to which it belongs. 



Gryrohedry has not yet been described in silver sulphide, but it has 

 been discovered in cuprite by Professor Miers. To this mineral we 

 turn therefore with particular interest. 



The molecular weight of cuprite, Cu 2 0, is 142 -6, its specific gravity 

 has been determined with most discordant results; we select that 

 given in ' Watts' Dictionary of Chemistry,' viz., 5-749 at 4 C. The 

 molecular volume obtained from this is 24-8043, and the edge of the 

 cube of reference measures 5*3. The diameter of an atom of copper 

 in the metal itself has been previously determined as 1-918, that of 

 oxygen on a somewhat doubtful basis, as 1-851. 



Our want of certitude regarding the precise specific gravity of 



