150 Henry Clifton Sorby, Esq., on the 



gular system, but I shall shew that its primary form is not an octahe- 

 dron but a cube, and that it has very different properties and specific 

 gravity, viz., about 1*89. Anthracite, however, is crystallized in 

 the square prismatic system, and has a specific gravity of about 1*76. 



The manner in which I have been able to ascertain the crystal- 

 line form of these bodies is as follows ; — I bruise a small portion to 

 a fine powder in a mortar along with some ten times the quantity of 

 soft chalk, which is employed to protect the fine fragments from 

 injury after being detached, and afterwards dissolve the chalk in 

 acid, and wash the powder left by subsidence and decantation. Then 

 spreading a portion with the aid of a little water on a piece of thin 

 glass used for microscopic object covers, and drying it, I examine it 

 through the glass with a magnifying power of about 400 linear. 

 By this treatment, the crystalline faces often arrange themselves on 

 the surface of the glass, as can be told by the adjustment of the 

 focus, though many do not ; and, of course, the angles which are then 

 seen are of no value for measurement. Their magnitude is deter- 

 mined by the doubly refracting goniometer, invented by Dr Leeson, 

 by the use of which beautifully contrived instrument, with proper 

 precautions, the angles of fragments of joVo^l^ of ^^ ^^^^ i^ <li^' 

 meter, or even less, can be ascertained within a small fraction of a 

 degree. Of course by this method we can only measure the angles 

 which occur on one plane, and the size of the particles makes it 

 impossible to turn them round. I have, therefore, been compelled 

 to adopt a very different method of procedure to that usually followed 

 in crystallographic researches ; and after measuring a number 

 of angles, sometimes connected, and at others detached, I have, by 

 comparing them by calculation, been able to build up the primary 

 form from which they are derived. 



It is not every specimen of hard coke which shews the crystalline 

 structure well. That which is highly vesicular does not exhibit it 

 at all, nor does that which is smooth and globular, which presents a 

 similar appearance when highly magnified. That which is dense 

 and has a highly metallic lustre is the best, though even with it not 

 many particles with their angles good and well placed for measure- 

 ment are found, as is indeed to be expected from the nature of the 

 substance and the treatment to which it is necessary to subject it. 

 Moreover, I should say that it does not cleave readily, and there are 

 many particles which are obviously broke in a splintery manner, 

 quite independent of structure, but these angles can generally be 

 easily distinguished from those due to crystalline cleavage. Not- 

 withstanding all these necessary difficulties, by very patient examina- 

 tion 1 have found amply sufficient good angles, well placed for measure- 

 ment, to satisfy me that it is crystallized in the regular system. The 

 angles are of 90°, 45°, 60°, 30°, 70^°, 109^", or as near those 

 values as it is possible to ascertain, and I should say that they most 

 certainly do not differ from them by more than |°, or at most ^°, and 



