which are evolved in Coal-Mines. 33l 



deposited ; and since, according to Henry,* carbonic oxide gas 

 undergoes no change from 1100 small sparks ; it appeared to 

 me, that to electrify the pit-gas, would be a very appropriate 

 method of ascertaining the presence or absence of the latter. 

 Pit-gavS, which had been purified by washing with a solution 

 of potash, was introduced into a detonation tube over mercury, 

 and dried by means of chloride of calcium. 6200 sparks from a 

 a charged jar were passed through the gas, for which a period of 

 thirty hours was required. Long before this large number of 

 sparks had been transmitted, I could remark no farther increase 

 of volume. Every thing, therefore, had been accomplished 

 which could be attained by this method. One might other- 

 wise easily be deceived respecting the actual increase of vo- 

 lume, when, as is the case during so long a continuance of tlie 

 electricity, interruptions occur ; for the gas expands somewhat 

 in consequence of the heat excited by the sparks, and again 

 contracts during the interruption. Besides, the increase of vo- 

 lume during continued electricity always goes on diminishing, 

 so that towards the end, hundreds of sparks may pass through 

 without a change of volume being perceptible. Unfortunately 

 I could not measure the true increase of volume, as the de- 

 tonation tube was not quite air-tight, for, during the long 

 continuance of the -experiment (four to five days), some atmo- 

 spheric air gained admittance between the platinum wires and 

 the glass, t 



• New System of Chemistry, vol. ii. p. 177. 



t Detonation tubes may bo perfectly tight for mercury, and yet not air-tight. 

 I have very frequently found that they were filled with mercury, or that the 

 hermetically sealed end was turned downwards, and that not a drop was allow- 

 ed to get through for some days ; but that, in the opposite iiosition, when 

 standing in the pneumatic trough, after some days some air entered. This 

 air frequently amounted to so little, that it was only after some days that the 

 mercury in the tube fell a few lines. There is no doubt that the pressure 

 of the atmosphere, which in the last case acts on the tube, not only causes 

 this, but that the greater subdivision of the air, in comparison with the mer- 

 cury, has a share in it. I had previously tried the detonation tube employed 

 in the above experiments, and found that no air entered in twenty-foui* 

 hours. I can, therefore, make no other supposition, but that, during the 

 shock of so lai^e a number of electric sparks, a space, however small it may 

 have been, was opened between the platinum wires and the glass. I may 

 further remark, that, at the commencement of the experiment, the mercury 

 Btood at 12, and at its termination at 8 inches above the external leveL 



