202 SIE ANDREW NOBLE: RESEARCHES ON EXPLOSIVES. 



M. BERTHELOT, in the memoir to which I have referred, considers that the view 

 which we took was contrary to all that was known in chemistry. 



It is no light thing to differ from so great an authority as M. BERTHELOT ; but the 

 innumerable experiments I have since made with various modern explosives, in which 

 the decomposition is of a simpler nature than that of fired gunpowder, have only 

 confirmed me in the opinion that Sir F. ABEL and I then expressed. 



Thus in a paper published in the ' Proceedings of the Royal Society,'* I pointed 

 out that when gun-cotton was fired under a great variation of pressure, the variations 

 in the proportions of the resulting gases were both great and regular. In passing, 

 for instance, from explosions under a pressure varying from 1'5 ton per sq. inch 

 (2287 atmospheres) to 50 tons per sq. inch (76217 atmospheres) the volume of 

 carbonic anhydride increased from 26'49 per cent, to 36'18 per cent., while the 

 carbon monoxide decreased from 36'G6 per cent, to 27'57 per cent. 



There were also other differences, though not quite so marked, such as the steady 

 decrease of free hydrogen and the large and steady increase of marsh gas. 



In the researches on gun-cotton to which I have alluded, certain data, such as the 

 units of heat and the quantity of water formed by the explosion, although deter- 

 mined, were not determined under the varying conditions with regard to pressure 

 and the quantity and nature of the gases generated, under which the explosion 

 took place. 



In the researches I am about to refer to, all the data connected with the explosion 

 have been carefully determined, and I preface an account of the experiments 

 themselves by a description of the varied apparatus adopted, or specially designed, 

 for determining the tension of the gases generated by the explosion, the volume of 

 the permanent gases and their nature, the quantity of water formed, the units ot 

 heat generated, the time taken to complete the explosion under different pressures 

 and different dimensions of the cords, tubes, or ribbons, these being the forms under 

 which the explosives are generally made up. 



I have made experiments also to determine the time in which the exploded gases 

 part with their heat to the walls of the vessel in which they are confined. 



These investigations have opened out many suggestive points, but in the present 

 paper I propose to confine myself to a description of the apparatus used and the 

 results obtained, giving also a resume of the calculations made to test the accuracy 

 of the observations. 



Commencing with the apparatus for firing the explosives experimented with at 

 different densities, obtaining the gases for analysis, and measuring their total volume, 

 the vessel A., in Plate 1, is one of the explosion .cylinders used for these experiments ; 

 B is the plug closing the vessel, on which also is shown the arrangement by which, 

 when desired, the gas is allowed to pass at a small pressure through the tubes, either 



* ' Roy. Soc. Proc.,' vol. 56, p. 209. 



