192 M. It. Bunsen on the Lata of Absorption of Gases. 



When this is calculated, negative values for u l + /3 l and for 

 « n + /3 n are found from both absorption experiments, namely, 

 — 0-3325 and —0-34807, instead of the sums of the coefficients 

 found in the tables for methyle and hydrogen, +0*07376 and 

 + 006111. Hence the gas in question cannot consist of equal 

 volumes of hydrogen and methyle. 



If, on the contrary, the same elements are used in the calcu- 

 lation of 7, and <y n , under the supposition that the gas is a simple 

 one, two absorption-coefficients are obtained, which are almost 

 exactly the same as those found in the table for marsh gas, at 

 the temperatures 12°'8 C. and 24°-6 C. The formula 



i^rx = ^ (1) 



gives for the absorption at 12°*8 7, = 0-0439 instead of the 

 actual value 0-0411, and for the absorption at 24°-6 7 B =00333 

 instead of 0-03166. From this agreement, we may conclude 

 that the marsh gas prepared from acetate of potash is neither a 

 mixture of hydrogen and methyle, nor a substance isomeric with 

 natural marsh gas, but that it is actually the same substance 

 which streams from the mud-volcanoes of Bulganak in the 

 Crimea. 



Any general reaction to distinguish between the constituents 

 of a gaseous mixture has hitherto been wanting. The quanti- 

 tative composition of a gas obtained by eudiometrical analysis, 

 depends almost entirely upon certain suppositions concerning its 

 qualitative constitution. If, for instance, eudiometrical analysis 

 points out the presence of marsh gas, it remains quite unde- 

 cided, as I have just shown, whether this gas is a mixture of 

 equal volumes of methyle and hydrogen. If analysis shows the 

 presence of a mixture of marsh gas and hydrogen, it is uncertain 

 whether we are not experimenting upon mixtures of methyle 

 and hydrogen, or of methyle, marsh gas, and hydrogen. All 

 analyses in which the two latter gases occur together may be 

 calculated according to either of these assumptions, without it 

 having been hitherto possible to prove the accuracy of either 

 one. 



It is easy, by means of the law of absorption, to remove these 

 doubts, for the absorption-coefficients serve as the reagents 

 which fail in gas analysis, and they present also the peculiarity, 

 that they not only show the qualitative, but at the same time 

 the quantitative composition of the gas. Let us, for example, 

 suppose that an unknown gas be mixed in an unknown volume 

 x, with an unknown volume y of another unknown gas, we can 

 then, by means of three absorptiometric experiments, determine, 

 — 1st, what gases are present in the mixture, 2nd, in what pro- 

 portions they occur. 



