Dr THOMAS THOMSON on a New Combustible Gas. 17 



considerable energy upon the eyes and nose, occasioning a flow 

 of tears, and exciting considerable pain in the eyes. 



3. It is combustible, and burns with a lively bluish-white 

 flame. 



4. Water absorbs it pretty rapidly : one volume of water, in 

 my trials, absorbed five volumes of the gas. The water acquires 

 a pungent taste, and the peculiar smell of the gas. But it does 

 not alter the colour of litmus or cudbear paper. 



5. One volume of oil of turpentine absorbs thirty volumes of 

 the gas ; the oil assumes a light-green colour, and resembles caje- 

 put ; but still retains its peculiar odour. 



6. The gas is neither absorbed by acids nor alkalies. Hence 

 it possesses neither acid nor alkaline properties. 



7. When common air or oxygen gas is mixed with this gas, 

 the usual red fumes of nitrous acid appear, and the volume of 

 the mixture is diminished. It is not, therefore, a homogeneous 

 substance, but contains mixed with it a considerable proportion 

 of nitrous gas. I endeavoured to determine the proportion of 

 nitrous gas in 100 volumes, by mixing it with determinate quan- 

 tities of oxygen gas over mercury. The diminution of vo- 

 lume was noted, and two-thirds of that diminution reckoned 

 as nitrous gas. This method of proceeding is not susceptible of 

 perfect accuracy, because the nitrous acid formed acts upon the 

 mercury. But as the action is not rapid, and the time of each 

 experiment short, I do not think that the error thence arising 

 could amount to so much as 5 per cent. Five experiments made 

 in this way did not absolutely agree with each other. But the 

 discordancy did not exceed 4 per cent. A mean of the whole 

 gave the amount of nitrous gas in 100 volumes of the new gas, 

 63 volumes, or rather more than three-fifths of the whole. 



I tried to determine the proportion of nitrous gas over wa- 

 ter, by causing the water to absorb the new inflammable gas, and 

 then agitating the residual gas in a solution of protosulphate of 

 iron. But this method yields no good results. The new in- 



VOL. XI. PART I. C 



