On the Formation of IJ^iier hy Compression, 363 



probabilitv, the violent motion necessary to compress with 

 rapidity prevented us from seeing into the interior of the 

 tube in a manner sufficientlv direct to observe the fugitive 

 light disengaged by the manifest compression, and which 

 in other experiments 1 had myself seen. 



After this trial we introduced into the tube a mixture of 

 hydrogen and oxygen gas ; gave a stroke with tlie piston, 

 and there immediately appeared a bright flash. A strung 

 detonation took place. The glass bottom was driven out ; 

 the copper ring by which it was screwed fast was broken ; 

 and the person who held the barrel had his hand slightly 

 burnt and bruised by the force of the explosion. 



We renewed the experiment, substituting for the olass 

 bottom one of copper, made of one piece, and screwed m. 

 Having then introduced into the barrel a new mixture of 

 the two srases, an explosion similar to the smart crack of 

 a whip was heard on the iirst stroke of the piston; but a 

 second stroke given to the new gas made it detonate, and 

 broke the barrel, or rather tore it, with a violent explosion. 

 After these phjenomena no doubt could remain m regard 

 to the combination of these two gases, since it is known it 

 is that which produces the detonation by the inmiense quan- 

 tity of lieat disengaged when they pass to the liquid state: 

 a heat which is sufficient to reduce -hem immediately into 

 vapour, and to give them in that state an excessive dilata- 

 tion. We did not then think it necessary to repeat any 

 more this experiment, which is not free i'rom danger. 



The theory of these pliEenomena is exceedingly simple. 

 A rapid compression forces the gas to abandon a very large 

 quantity of heat, which, as it cannot be immediately dissi- 

 pated, raises their temperature for a moment, and in that 

 state of compression is sufficient to inflame them. 



We find therefore in the two gases all the elements ne- 

 cessary for combining them, independently of the electric 

 spark or external fire; and it is not improbable that all the 

 gaseous combinations which require an elevation of tempe- 

 rature might be formed iu the same manner without any 

 foreign agent. 



This identity of results sucrgcstcd an idea which I submit 

 to the opinion of philosophers. It is known, and M. Uer- 

 thollet has shown in his Sfutirjne Chunicjue, that electricity 

 in traversing bodies produces in their mojeculae a real com- 

 pression. Thiscff'ect is produced with prodigious velocity, 

 as may be proved by a variety of experiments ; but, as elec- 

 tricity has a similar velocity, it is impossible that it fhould 

 not disengage light from the air, since vvc are able to dis- 

 ci r' aire 



