564 DR. H. DEBUS ON THE CHEMICAL THEORY OF GUNPOWDER. 
The by-products contain no oxygen of the saltpetre, and with the exception of the 
potassium and nitrogen in the potassic sulphocyanate, and the nitrogen of the 
ammonia, their elements are derived from the charcoal and sulphur. The chief pro¬ 
ducts should, accordingly, contain all the oxygen of the decomposed saltpetre. This, 
however, is not the case; 0’52 of an atom is wanting, and if the potassic hyposulphite, 
as we must assume, has been formed by the oxidising action of the cupric oxide 
upon the potassic disulphide, then no less than 1*81 atoms or nearly -^g-th of the 
oxygen of the decomposed saltpetre have disappeared. Linck himself finds in the 
products of combustion 1 '75 per cent, or about ^ 2 -nd less oxygen than in the original 
powder, and this loss of oxygen would have appeared still greater if he had not 
assumed that the oxygen of the hyposulphite had been derived from the saltpetre. 
On the other hand, Linck finds in the products of explosion 0'71 per cent, or y^-th 
to 1 - 3 -th more of carbon, and 0*9 or -ygth. to y^th more of sulphur than in the original 
powder. Accordingly, it follows, either that Linck's powder has not the composition 
which he ascribes to it, or that some considerable errors attach to the analytical data 
of the products of explosion. Hence, no near agreement can here be expected between 
theory and experiment. 
If we replace the potassic hyposulphite found, by its equivalent of potassic disulphide, 
express the quantity of the chief products by molecular weights, and calculate the 
composition of that portion of the powder which was transformed into the chief 
products, we obtain 
16KNO 8 +15-06O+7*448-1-06N-1-81O. 
and if in equation (VIII.) we place, 
x—l& y= 15 
2 = 7-5 a— 1*0 
we find the following theoretical values, which are placed by the side of the experi¬ 
mental numbers 
Theory. 
Experiment. 
" k 3 co 3 . . . 
2-14 
2*43 
HKNOg 
K 3 S(\ 
4*21 
3*64 
15C 
>=■ 
ICS, .... 
1*64 
1*90 
7*5S 
co 3 , . . . 
11*85 
11*71 
w 1 
^co . . . . 
1*0 
0*92 
Some mistake appears to have occurred in the determination of the potassic sulphate 
and carbonate. The calculation could likewise be carried out in the following manner : 
The unburnt portion of the powder, and the quantities of the elements contained in 
the by-products, could be subtracted from the composition of the quantity of powder 
taken for experiment, and from the rest, the quantities of the products might be 
