28G LIGHT SCIENCE FOR LEISURE HOURS. 



the lieat may make for itself an opening which 

 though small would yet suffice to enable the flame 

 w r ithin the lamp to ignite the gas outside. So long, 

 however, as the wire-gauze continues perfect, even 

 though it become red hot, there will be no explosion. 

 No authority is required to establish this point, which 

 has been proved again and again by experiment ; but 

 we quote Professor TyndalPs words on the subject 

 to remove some doubts which have been entertained 

 on the matter : " Although a continuous explosive at- 

 mosphere," he says, " may extend from the air outside 

 through the meshes of the gauze to the flame within, 

 ignition is not propagated across the gauze. The 

 lamp may be filled with an almost lightless flame ; 

 still explosion does not occur. A defect in the gauze, 

 the destruction of the wire at any point by oxidation, 

 hastened by the flame playing against it, would cause 

 explosion; " and so on. It need hardly be said, how- 

 ever, that, imprudent as miners have often been, no 

 miner would remain where his lamp burned with the 

 enlarged flame indicative of the presence of fire- 

 damp. The lamp should also be at once extin- 

 guished. 



But here we touch on a danger which undoubtedly 

 exists, and so far as has yet been seen cannot be 

 guarded against by any amount of caution. Sup- 

 posing the miner sought to extinguish the lamp by 

 blowing it out, an explosion would almost certainly 



