CAPTAIN NOBLE AND ME. F. A. ABEL ON FIRED GUNPOWDER. 
77 
small amounts of this constituent ; in our results (which will presently be compared 
among themselves) the hyposulphite was also found to vary in amount very greatly. 
These fluctuations were found by us, in most cases, to bear definite relation to those of 
the sulphide ; but this is not observed to be the case in the analyses of Linck and Bunsen 
and Schischkoff on comparing them with ours. 
The method pursued by these chemists for obtaining the products of decomposition 
of powder was of a nature calculated to furnish very variable results, which can scarcely 
be accepted as corresponding to those produced when gunpowder is exploded in an 
absolutely closed space or in the bore of a gun. 
By allowing the powder-grains to drop gradually into a heated open bulb, not only is 
their decomposition accomplished under very different conditions to those attending the 
explosion of a confined charge of powder, but the solid products are necessarily subjected 
to further changes during their continued exposure to a high temperature and to the 
action of fresh quantities of powder deflagrated in contact with them. An imperfect 
metamorphosis of the powder-grains themselves and further secondary changes in the 
composition of the residue deposited (which will vary in extent with the duration of the 
experiment), the amount of heat applied externally, and the rate at which the powder- 
grains are successively deflagrated appear to be inevitable results of this mode of opera- 
tion. A comparison of Bunsen and Schischkofu’s results with those shortly afterwards 
obtained by Linck in Bunsen’s laboratory, the same method being pursued for effecting 
the decomposition of the powder, appears to demonstrate this conclusively. 
The differences in the composition of the powders operated upon in the two investi- 
gations would certainly not suffice to account for the important differences exhibited by 
the results of analysis of the residues. The comparatively large proportion of potassium 
sulphide, the much larger proportion of hyposulphite, and the considerably smaller pro- 
portion of sulphate found by Linck appear to indicate that the operation of burning 
the powder was conducted much more rapidly by him, a view which is supported by the 
fact that, while he found a considerable proportion of ammonium carbonate in the 
residue, none existed in the product obtained by Bunsen and Schischkoff, who, however, 
found this constituent in the so-called powder-smoke which they collected in a long tube 
through which the gas escaped. 
Our experiments have demonstrated conclusively that, even when the conditions under 
which the explosion of powder is effected in distinct operations are as closely alike as 
possible, very exceptional results, as regards the composition of the solid residue, may 
be obtained, experiments 7 & 17, 9 & 4, 14 & 70 being illustrations of this. Yet in no 
instance, however apparently abnormal, did any considerable proportion of potassium 
nitrate escape decomposition, the highest amounts discovered in the residues being 0'48 
and 0’56 per cent, (experiments 1 & 8). These percentages existed in the products of 
explosion of powder formed under the lowest pressure ; in only two instances, at 
higher pressures, were similar proportions found. The existence of so large a pro- 
portion as 5 per cent, of potassium nitrate in the residue obtained by Bunsen and 
Schischkoff, the coexistence of 7 ’5 per cent, of hyposulphite and small quantities of 
