1874.] on Fired Gunpowder. 413 



distinct experiments made at varied pressures, differed, in several in- 

 stances, quite as greatly as those found in some of the residues of powders 

 which presented decided differences in composition. 



Although, for the reasons already given, the authors cannot attempt 

 to offer any thing approaching a precise expression of the chemical 

 changes which gunpowder of average composition undergoes when ex- 

 ploded in a confined space, they feel warranted, by the results of their 

 experiments, in stating, with confidence, that the chemical theory of the 

 decomposition of gunpowder, as based upon the results of Bunsen and 

 Schischkoff and accepted in recent text-books, is certainly as far from 

 correctly representing the general- metamorphosis of gunpowder as was 

 the old and long-accepted theory, according to which the primary products 

 were simply potassium sulphide, carbonic anhydride, and nitrogen. 

 Moreover, the following broad facts regarding the products furnished by 

 the explosion of gunpowder appear to them to have been established by 

 the analytical results arrived at, 



1 . The proportion of carbonic oxide produced in the explosion of a gun- 

 powder in which the saltpetre and charcoal exist in proportions calculated, 

 according to the old theory, to produce carbonic anhydride only is much 

 more considerable than hitherto accepted. 



2. The amount of potassium carbonate formed, under all conditions 

 (as regards nature of the gunpowder and pressure under which it is ex- 

 ploded), is very much larger than has hitherto been considered to be pro- 

 duced, according to the results of Bunsen and Schischkoff and more recent 

 experimenters. 



3. The potassium sulphate is very much smaller in amount than found 

 by Bunsen and Schischkoff, Linck, and Karolyi, even in the highest results 

 obtained in the authors' experiments. 



4. Potassium sulphide is never present in very considerable amount, 

 though, generally, in much larger proportion than found by Bunsen and 

 Schischkoff ; and -there appears to be strong reason for believing that, in 

 most instances, it exists in large amount as a, primary result of the explo- 

 sion of gunpowder. 



5. Potassium hyposulphite is an important product of the decomposi- 

 tion of gunpowder in closed spaces, though very variable in amount. It 

 appears probable (the reasons being fully discussed in the paper) that its 

 production is in some measure subservient to that of the sulphide; and it 

 may perhaps be regarded as representing, at any rate to a considerable 

 extent, that substance in powder-residue -4. e. as having resulted, partially 

 and to a variable extent, from the oxidation, by liberated oxygen, of sul- 

 phi.le which has been formed in the first instance. 



6. The proportion of sulphur which does not enter into the primary 

 reaction 011 the explosion of powder is very variable, being in some 

 instances high, while, in apparently exceptional results^ the whole amount 

 of sulphur contained in the powder becomes involved in the rnetamor- 



