449 
of carbon and hydrogeji, &c. 
Four other experiments gave results all approximating to 
the above.^ The mean result was i hydrogen, ii.57^ carbon. 
Now considering that the substance must, according to the 
manner in which it was prepared, still retain a portion of the 
body boiling at 186°, but remaining fluid at 0°, and which 
substance I find, as will be seen hereafter, to contain less 
carbon than the crystalline compound, (only about 8.25 to 
1 of hydrogen, ) it may be admitted, I think, that the con- 
stant though small deficit of carbon found in the experiments 
is due to the portion so retained ; and that the crystalline 
compound would, if pure, yield 12 of carbon for each 1 of 
hydrogen ; or two proportionals of the former element and 
one of the latter. 
2 proportionals carbon . 1 2 1 ^ bi-carburet of hydrogen. 
1 hydrogen i ) 
This result is confirmed by such data as I have been able 
to obtain by detonating the vapour of the substance with 
oxygen. Thus in one experiment 1092 mercury grain mea- 
sures of oxygen at 62° had such quantity of the substance 
introduced into it as would entirely rise in vapour ; the 
volume increased to 8505, hence the vapour amounted to 413 
parts, or of the mixture nearly. Seven volumes of this 
mixture were detonated in an eudiometer tube by an electric 
spark, and diminished in consequence nearly to 6.1 : these 
acted upon by potash were further diminished to 4, which 
were pure oxygen. Hence 3 volumes of mixture had been 
detonated, of which nearly 0.34 was vapour of the sub- 
stance, and 2.65 oxygen. The carbonic acid amounted to 2.1 
volumes, and must have consumed an equal bulk of oxygen 
gas ; so that 0.55 remain as the quantity of oxygen which 
