which are evolved in Coal-Mines, 129 



quantity of gas could always be brought to a particular gra- 

 duated line by sinking the detonation-tube in the mercury of 

 the pneumatic trough, or by raising it up. In this manner the 

 errors of a geometrical subdivision, owing to the unequal width 

 of the glass tube, were entirely removed. The detonation- 

 tubes were from 12 to 18 inches long. As much of the mix- 

 ture of gases was always employed for detonation as the tube 

 could bear without fear of being broken. The internal level 

 of the mercury generally stood at from 10 to 14 inches above 

 the external, and the gas was therefore about half the density 

 of the atmospheric air. This was advantageous, as the vio- 

 lence of the detonation was thus diminished, and the mea- 

 surement of the gas could thus be made with the same exact- 

 ness as the observation of the height of the barometer, which 

 could not be effected in the same degree if the interior and 

 exterior mercurial level stood at the same height. Of course 



cannot appear surprising, when we consider that, in cooling, platinum and 

 glass do not contract equally. However intimately united, therefore, the 

 two may he during the melting of the glass, yet an opening must be formed 

 after cooling, which, it is true, will for a short time afford no access to the 

 air, but not for a longer period. According-to my experiments, the air- 

 tight inclosure of the wire succeeds more easily in proportion to the slow- 

 ness of the cooling of the glass tube. Hence, immediately after the pla- 

 tinum wire has been included in the tube by fusion, I am in the habit of 

 placing the tube in a large quantity of red-hot sand, and allowing the two 

 to cool together. In order, moreover, to diminish the difficulties, I insert 

 only one platinum wire in the end of the tube, where it succeeds better 

 than at the side. The transmission of the electric spark takes place by an 

 iron-wire being pushed up into the detonation-tube. The mercurial trough 

 has a hole at the bottom, which is closed by a cork, through which the wire 

 is puhed. If the cork be good, the wire quite smooth and a little greased, it 

 can tic easily pushed up and clown, and can yet so close the aperture as to ren- 

 der it impervious to the mercury. It is of course understood that, previous to 

 the introduction of the gas into the tubes, every trace of potash and moisture 

 must be removed. For this purpose I pour some drops of acetic acid into 

 the tube, then rinse it out with water, and afterwards dry it very carefully 

 with blotting-paper. 



If these circumstances arc neglected, we cannot be surprised by the ge- 

 neral complaint among chemists that the analyses of inflammable gas in 

 the detonation-tube afford results which are so little in unison with one 

 another. But by attending to them on the other hand, we obtain results 

 an accurate as can be wished. 



VOL. XXX. NO. LIX. JANUARY 1841. I 



