462 Sir Andreiv NohU [Jan. 18, 



centages in hydrogen. The percentage with the density of "05 

 varies from 8 per cent, to over 20 per cent. The initial percentage 

 in the whole of the explosives increases slightly with increase of 

 pressure, and then (with the exception of the Italian Ballistite) 

 decreases rapidly to about 8 per cent., the curves at the density of 

 • 5 being so close that a difference of little more than 1 per cent, 

 includes all the curves. The rapid decrease of hydrogen is doubtless 

 due to the great ijicrease of CH4. You will, no doubt, have noted 

 that the hydrogen of the Italian Ballistite is nearly constant, the 

 constancy being due to the decrease in the amount of aqueous vapour, 

 and to the small quantity of GH4 formed ; but it is remarkable that 

 with Italian Balhstite there is far less variation, due to density in the 

 products of explosion, than is the case with any other of the 

 explosives. 



I will only trouble you with one more diagram of the products, 

 and that shows the volumes of marsh gas. The order, with one 

 slight exception, is the same as with the carbonic oxide. At the 

 density of • 05 there is, with the whole of the explosives, a trace only ; 

 but, as you see at density • 5, there are both increase and very con- 

 siderable variation, the percentage reached being in one case over 12 

 per cent. 



The next point to which I draw your attention is the units of 

 heat liberated by the explosion. These, as before, I show on the 

 screen, and you will note how all the curves which give the units of 

 heat, the water being gaseous, have approximately the same form, all 

 dropping a little at first, and subsequently rising somewhat rapidly, 

 the higher heat being attributable chiefly to the much larger 

 volume of CO2 formed at the highest pressures, the weight of the 

 gas at the density of * 5 being, with all the explosives named, more 

 than half of the total weight of the gases, and the heat of the Italian 

 Ballistite, as in the case of the volumes of the gases, showing but 

 little variation. 



The difference between the heat of these two explosives, Cordite 

 and Italian Ballistite, as compared with that of the four other 

 explosives, is at the lower densities, including those with which 

 artillerists are concerned, great ; but this difference is much reduced 

 at the high densities, due, as I have said, to the rapid increase in the 

 quantity of CO2 with higher densities. 



I now approach a point concerning which it is necessary to give a 

 few words of explanation. The observations I have laid before you 

 hitherto depend upon direct observation, and these observations have 

 been confirmed in so many ways that they must be accepted as correct, 

 subject to the slight errors from which no human work is exempt ; but 

 the point we have now to determine is, what are the temperatures 

 which the firing of the explosives we are considering produce ? 



I fully admit that many physicists may differ from the view I 

 have taken, but I have a good deal to corroborate my view. 



