1009] Oil Means of Saviwi Life in Goril Mines. 479 



possible, and that is to have the shot-firing done between the shifts, 

 when the men are ont of the mine. An explosion would thus injure 

 very few people, and if the shot firers took some reserve apparatus 

 with them, they would run a comparatively small risk, even if there 

 were a formidable explosion. For they would be at the face where 

 but little carbonic oxide forms, and if cut off they could be easily 

 rescued, except in those cases w'here falls of roof prevented egress. 



Thus, then, the idea of coal-dust as a principal factor in explo- 

 sions in mines has been firmly estabHshed in Great Britain and in 

 Grermany. 



Curiously enougli, our ingenious and scientific French neighbours, 

 who originally made so many experiments on coal-dust explosions, 

 have of late years been less impressed with the part coal-dust plays 

 in mine disasters. 



When Mr. Atkinson, the Mines Inspector, and I went over after the 

 great explosion at Conrrieres to visit the mine, we found that most 

 of the French miners held the view, which ten years previously our 

 engineers would have held, that gas must have done it. The idea 

 was, of course, incredible : for to attribute an explosion so wide in 

 its effects to gas, one would have to suppose that the mine was full 

 of explosive tire-damp nearly from one end to the other. But how 

 was it to be imagiued that gas in such quantity could be present all 

 over the mine without anyone seeing it on his safety lamp ? 



On the other hand, you had a mine full of the finest and most 

 explosive dust, and as Mr. Atkinson and Mr. Henshaw — who also, at 

 our request, visited the mine — saw at once, the walls were covered 

 with charred coal-dust, exactly similar to that which had so often 

 been seen in our own colliery explosions. The French engineers 

 now entertain no doubt that coal-dust alone is capable both of 

 originating and of propagating an explosion, and are also trying 

 experiments upon a large scale. 



Great as the change of opinion has been with regard to coal-dust, 

 our views have had to undergo a transformation almost equally 

 radical with regard to the causes of death after an explosion. 



When gas is burned thoroughly, you have, roughly speaking, 

 one volume of fire-damp, mixed with two volumes of oxygen, which 

 yields one volume of carbonic acid gas and two volumes of steam : 

 the seven volumes of nitrogen present remaining unchanged. " 



Therefore, after such an explosion the mine ought to l»e full of 

 steam, carbonic acid and nitrogen, all the oxygen having disappeared. 



But in practice this never happens. For in gas explosions there 

 is always an excess of oxygen present. 



But as we have seen, there are no such things as gi^ explosions 

 on a large scale — even if there is any gas present to begin the explo- 

 sion, the main result is always due to dust, and in that case, instead 



