Chemical and Physical Papers. 93 



cessful manufacture and use in England, but it is believed that no 

 potassium nitrate black powders are used in this country. Appar- 

 ently it is a fact that the use of sodium nitrate makes no important 

 difference in the gaseous decomposition products. 



The most careful examination of the decomposition products of 

 gunpowder are those made by Abel and Noble of England and 

 published in 1875 and 1880. Their experiments seemed to indi- 

 cate, though not very conclusively, that the percentage of carbon 

 monoxid decreases as the percentage of the volume of the contain- 

 ing vessel occupied by the powder increases; or, in other words, as 

 the density of the products of composition increases. If this is 

 the case, it would seem that the products of a blown-out shot in 

 which the j^owder burns without doing much mechanical work 

 should contain a comparatively large amount of carbon monoxid. 



In order to test this point and others, a large number of experi- 

 ments have been carried on. Powder has been burned in three dif- 

 ferent ways: First, by dropping the grains into a heated iron tube 

 so arranged that the gases could be collected. This gave the gases 

 at nearly atmospheric pressure. Second, by placing a considerable 

 quantity in an iron tube and igniting it by means of a fuse, the 

 gases being collected in a larger tube for the purpose of excluding 

 air. This gives conditions approaching those of a blow-out shot. 

 Third, by exploding the powder in a gas-tight bomb from which 

 the gases could be drawn off as desired. The experiments have not 

 confirmed the results indicated by the experiments of Abel and 

 Noble, in that apparently the percentage of carbon monoxid does 

 not decrease as the density of the composition products increases, 

 but the change, if any, is in the opposite direction. The series of 

 experiments is hardly sufficiently extended to make it advisable to 

 give this as a positive statement, but the experiments so far con- 

 ducted indicate this. It is sometimes stated that mining powder 

 gives as much as thirty-three per cent, of carbon monoxid. If this 

 statement is true, it evidently refers to the mining powder which 

 has sometimes been used in England and which differs from gun- 

 powder in the reduction of the percentage of saltpeter, and not to 

 mining powder as used in the United States, which is fully ni- 

 trated, and gives about seven to ten per cent, of carbon monoxid. 



To determine as far as possible the effect of coal upon the 

 gaseous products of the powder, coal in the form of fine dust and 

 in the form of larger grains has been mixed with the powder. It 

 was found that carbon monoxid, methane and hydrogen appeared 

 in greatly increased quantities, and that, in some cases, the gaseous 



