532 
DR. H. DEBUS ON" THE CHEMICAL THEORY OF GUNPOWDER. 
Saltpetre.74'43 
Sulphur.. 10 '09 
Charcoal— 
Carbon.12*40 
Hydrogen.0’40 
Oxygen. 1*27 
Ash.0-21 
Water. 1*05 
99-85 
The sample used in the first analysis was taken from the top, the one employed in 
the second from the bottom ot the same barrel. Two analyses of powder out of the 
same barrel, executed by the same chemists, gave amounts of carbon which differ from 
each other by no less than 1*54 per cent. ! We will now consider the effect of such a 
difference in the percentage of carbon on the relative quantities of the products of 
combustion. 
If in one experiment 100 grms. of powder containing 75 parts of saltpetre, 10 parts 
of sulphur, and 10’86 parts of carbon, and in a second experiment 100 grms. of powder 
with 74'43 parts of saltpetre, 9 parts of sulphur, and 12*40 parts of carbon had been 
exploded, then, cceteris paribus, in the second experiment more potassic carbonate, 
more sulphide and less sulphate must have formed than in the first, and the quanti¬ 
tative differences of the same products furnished by the two experiments would be 
almost as great as the greatest differences actually observed by Noble and Abel for 
the same description of powder in the whole series of their experiments. We arrive 
in this manner at a very simple explanation of the experimental results upon which 
Noble and Abel have based the conclusions mentioned under No. 1 and 2 on p. 525, 
without the necessity of having recourse to a theory like the one advanced by 
M. Berthelot, or renouncing all explanation like Noble and Abel. 
It follows from Table I. that the differences of composition between P., B. L. G., 
B. F. G., and F. G. powders, one compared with the other, are not greater than the 
differences between two samples of B. L. G. powder taken out of the same barrel. 
From this it appears probable that only one mixture of saltpetre, charcoal, and 
sulphur is prepared at Waltham Abbey, and that from this one mixture the powders 
P., B. L. G., B. F. G., and F. G., differing only in size of grain and, perhaps, density, 
are manufactured. We may take then for the composition of the said powders the 
mean of the numbers of all published analyses. Taking into consideration only the 
saltpetre, sulphur, and carbon, we obtain— 
