1907.] on Fifty Years of Explosives. 465 



gives only 818. The other explosives give intermediate differences, 

 and these differences are, as I have said, dependent on the units of 

 heat and the quantity of CO2. 



I now come to the question, can I produce any experiment to 

 corroborate the temperatures I have tentatively assigned ? 



I shall give you two examples — one with the explosive giving the 

 lowest temperature, the other with Mark I. Cordite, which is nearly 

 the highest. 



Eighteen months ago I fired a charge of nearly 3 kilogrammes 

 (6|- lb.) of Nitrocellulose at a density of • 2S. In the centre of the 

 charge a small packet of osmium was placed. 



On opening the vessel after explosion, and after the gases, etc., 

 were removed, the walls of the vessel were scraped, and these scrapings 

 were found to contain osmium. This appears to me to prove that at 

 least a portion of the osmium had volatilised. I do not know the 

 volatilising point of osmium, but its melting point is 2500° C, and 

 I suppose its boiling point is considerably higher. 



In the curve I have given in my paper in the " Royal Society 

 Transactions," the temperature of explosion at the density of '28 is 

 given at 3200° C; and, if allowance be made for the heat necessary 

 for volatilisation, and also for the rapid cooling of the gases, about 

 which I shall have something to say presently, I think the temperature 

 I have given is not very much removed from the truth. 



Again, at the same density, I fired the same charge of Mark I. 

 Cordite, and in the centre of the charge I placed a small packet of 

 electrically deposited carbon. The temperature I have given for this 

 density is" 4300° C. 



The solid and fluid products of combustion were handed over to 

 Sir W. Crookes, for whom I made the experiment. 



In the latest " Physikalisch-Chemische Tabellen " carbon is given 

 as " unschmelzbar," but Mark I. Cordite appears to have melted it, as 

 Sir "W. Crookes, after long and careful examination, obtained crystals 

 which were undoubtedly diamonds. Sir William gives the melting 

 point of carbon at approximately 4400° C. absolute temperature, and 

 I think that this is fairly confirmatory of the temperature that my 

 curve assigns. 



I shall only trouble you on one or two further points. I throw 

 on the screen a diagram which shows the rapidity with which small 

 charges of an explosive (Mark I. Cordite) part with their heat to the 

 walls of the vessel. 



The curves you see are the diagrams on an altered horizontal 

 scale, traced by the explosive itself. You will observe the axis of 

 abscissae gives the time in seconds, and the vertical axis the pressure 

 in tons per square inch. The charges have densities varying from 

 • 1 to • 25. With the higher densities there is considerable oscillation, 

 due to the run of the spring, but it is not difficult to ascertain from 

 the diagram the true pressure. Besides, when accurate time of 

 YOL. XYIII. (No. 101) 2 H 



