12 MEMOIR ON THE 



combustible mass ; only, however, to be followed by a more active collision, result- 

 ing from the subsequent falling back of the conglomerated combustible mass upon 

 the melted nitre. After every such collision, the combustible congeries must have 

 been blown up to a height augmenting with the temperature, the force of the fall, 

 and extent of reciprocal penetration. The force of the fall would, of course, be as 

 the height. Hence the twelve or thirteen successive detonations indicate as many 

 explosive collisions ; while the successive augmentation of the loudness of the 

 reports indicates a proportionable growth of their violence, arising from succes- 

 sively greater elevation and descent. 



16. If I am right in supposing that in fulminating power, the intensely heated 

 nitre and the combustible merchandise were for equal weights equivalent to gun- 

 powders if only a sixth of the 300,000 pounds of nitre held in the store was 

 engaged in the final explosion, it would be equivalent to sixty thousand pounds of 

 gunpowder. 



17. No better way of estimating the force with which the nitre and combustibles 

 were brought into collision for the last time, at which the finishing explosion took 

 place, has occurred, than that of comparing it Avith the blow by which nitre and 

 sugar were exploded as above mentioned, in one of my experiments. 



The weight of the combustible matter contained within the store was 700,000 

 pounds. The store was ninety feet deep by twenty-four wide. Supposing the 

 horizontal area of the sledge as applied, to have been 3X3 = 9 square inches, it 

 seems that for every equivalent horizontal area within the store, there must have 

 been twenty-two pounds, or about three times the weight of the sledge. Hence, in 

 descending from a height of twenty or thirty feet, which there was ample room 

 for it to reach, the combustible congeries may have attained a much greater velo- 

 city than could be imparted to the sledge, and may consequently have produced a 

 much more forcible impact. At the same time, this must have caused an intimate 

 penetration and intensity of compression, which by a dead weight it is almost 

 impossible to create. 



This explanation, so far as it rests upon the assumption that the combustibles 

 were made to dance upon the surface of the melted nitre, is supported by the fact 

 that any combustible mass, when thrown upon the surface of incandescent nitre, 

 will imdergo a dancing motion, so as sometimes to leap out of a deep pot within 

 which the experiment may be made. 



The phenomena are not irreconcilable with the idea that some of the earlier 

 explosions arose from the falling of the liquid nitre upon the combustibles before 

 all the floors gave way ; but it should be recollected that nitre fuses at a low red 

 heat, and at a cherry red gives out oxygen gas. The presence of this gas, as well 

 as the deflagration resulting from contact with the liquid nitre, must have caused 

 the floors to be oxidized with a rapidity far exceeding that which takes place 

 during ordinary conflagrations.* To the causes of quick destruction thus sug- 

 gested, must be added the mechanical force of the explosions directly at war with 



See Note to piiragrapli 9. 



