Some Experiments on tJie Comhmion of llie Diamond. 437 



by the long continued action of heated potassium, induced me 

 to suspect in the beginning of these inquiries^ that common 

 charcoal might owe its colour, opacity, and conducting j)OAver, 

 to the circumstance of its containing minute portions ol' the me- 

 tals of the alkalies, or earths, and plumbago to the iron it con- 

 tains; l}ut when I found that charcoal made from oil of turpen- 

 tine, which left no residuum in burning, and charcoal precipi- 

 tated from carburetted hydrogen gas by chlorine, had the same 

 properties, it was necessary to renounce this opinion. 



The only chemical difference perceptible between diamond 

 and the purest charcoal, is, that the last contains a minute j)or- 

 tion of hydrogen ; but can a c|uantity of an element, less in some 

 •cases than -suir^m P^^t of the weight of tlie substance, occasion 

 so great a difference in physical and chemical characters ? This 

 is possible, yet it is contrary to analogy ; and I am more inclined 

 to adopt the opinion of Mr. Tennant, that the difterencS depends 

 upon crystallization. Transparent solid bodies are in general 

 non-conductors of electricity, and it is probable that the same 

 corpuscular arrangements which give to matter the power of 

 transmitting and polarizing light, are likewise connected with its 

 relations to electricity; and water, the hydrates of the alkalies, 

 and a number of other Ijodies which are conductors of electricity 

 when fluid, become non-conductors in their crystallized form. 



The power possessed by certain carbonaceous substances of 

 absorbing gases, and separating colouring matters from fluids, 

 is probably mechanical and dependent upon their porous nature; 

 for it belongs in the highest degree to vegetable and animal 

 charcoal, and it does not exist in plumbago, coke, or anthra- 

 colite. 



The same eonclusions respecting the composition of carbonic 

 ncid may be drawn from these experiments, as from those of 

 Messrs. Allen and Pepys, and Theodore de Saussure. If the 

 calculations be founded upon the difference of the weights of 

 oxygen and carbonic acid gases, which appears the most exact 

 method, carbonic acid gas will contain, according to the esti- 

 mate of the mean sju'cific of the gravities of the two gases given 

 by M. Theodore de Saussure*, thirty parts of oxygen, or two de- 

 finite proportions, to 11 '3 parts of carbon, or one definite pro- 

 portion. 



Supposing the diminution of the oxygen produced in the ex- 

 periments on the common carbonaceous suljstances entirely oc- 

 casioned by the formation of wat'Cr, it is easy to calculate the 

 proportions of hydrogen in them ; but in the case of plumbago 



* Annates fie Cliimie, tome kxi. pa.i;. 261. This rstiiuatian is Uie same 

 as that I have given. Llcinenls of Cl)t'iii. I'liil. pag. 305. 



E e 3 there 



