EESEAKCHBS UPON THE CHEMICAL PROPERTIES OF GASES. 157 



its crystallized form and containing therefore 7 mols. water. This salt and the ferri- 

 cyanide of potassium were ground separately and mixed just previously to the ex- 

 periment (anhydrous sulphate of iron was found not to give satisfactory results). 

 A little of the mixture placed in the end of the glass tube was found to be a delicate 

 indicator for moisture, assuming quickly a deep blue color. 

 Uxpt 7.— 



Air 96.9 



Ethylene 3. 1 



This gas mixture was passed through soda solution and then dried by oil of vitriol. 

 It then traversed the palladium asbestos, which was gradually heated in the oven. 

 At the far end of the same glass tube the gas passed over about 0.1 gra. of the green 

 vitriol and ferricyanide of potassium mixture (cold), and finally into lime water. It 

 was found that the lime water became milky a few moments before the powder turned 

 blue. This change to blue took place, however, immediately afterwards and before 

 any further increase in the temperature of the oven had occurred. As the lime water 

 is a much more sensitive reagent towards CO., than is the green vitriol mixture 

 towards moisture, it is natural that the CO2 should produce its effect a little in 

 advance of the water vapor. Similar experiments were tried with methane. The 

 results were the same as with ethylene. 

 Expt. 8.— 



Air. ; 96.9 



Methane 3.1 



This mixture (moist) was found, after passing the palladium asbestos, to cause no 

 reduction in palladium chloride solution (see reactions in solution), and hence no oxi- 

 dation of CHi to C2H4 had occurred. There is therefore no reason to suppose that, 

 under the circumstances which prevailed in the apparatus above described, either 

 constituent of the hydrocarbon is oxidized before the other. The hydrocarbon yields 

 directly CO. and H.O. 



PARAFFINS. 



2. Methane. — This hydrocarbon was prepared by the method of Gladstone 

 and Tribe {Jour. Cliem. Soc, 1881, p. 1541). Methyl iodide of normal boiling point 

 was caused to flow in admixture with alcohol upon '"'copper-zinc couple." The resulting 

 gas was freed from alcohol vapor by oil of vitriol. It was then washed with bromine 

 water, and the bromine vapors subsequently removed by ferrous sulphate solution. 

 Finally it was passed over dry palladium chloride at 50°, to remove any possible 



A. P. S. — VOL. XVJL U. 



