81 



the light of an equal volume of coal gas flame. If it be desi- 

 rable that the illuminating power of naphthalized water gas 

 shall be at its maximum, the ratio must be ten and a half im- 

 perial pints to 1000 cubic feet of gas. Any higher ratio only 

 impairs the light. When spirit of turpentine is used with water 

 gas, the ratio should be one imperial gallon to 1000 cubic feet. 



" The more pure and volatile kinds of naphtha are extremely 

 valuable, from the circumstance of their being easily soluble 

 in hydrogen, or in water gas, at ordinary temperatures. We 

 have only to present extensive surfaces of the naphtha to a cur- 

 rent of watergas ; the result will be that the gas will dissolve 

 a quantity of naphtha, and will hold it dissolved even at the 

 temperature of 32°. The mixture of gas and naphtha vapour 

 may be transmitted through tubes in the manner of ordinary 

 coal gas, and burned in the usual way ; its light is white and 

 beautiful. I have described an instrument for producing this 

 gas in the specification of my patent for Scotland. 



" I ascertained that when carbonic oxide is naphthalized, 

 its combustion affords a light of brilliant whiteness, although 

 its natural colour in burning is blue. 



" Since it was thus proved that hydrogen containing a very 

 small quantity of carbon was capable of dissolving naphtha, it 

 became a question whether that variety of the same combina- 

 tion called coal gas would exert a similar agency, and thus 

 be rendered capable of burning with increased brilliancy. I 

 therefore arranged two common gas-burners for comparison, 

 one being supplied with mere coal gas, the other with coal 

 gas which was made to pass through an apparatus properly 

 constructed for naphthalizing it. It required but little obser- 

 vation to decide that the naphthalized flame was much more 

 brilliant and dense. 



" When this trial was made with a gasometer attached to 

 each burner, and an apparatus for determining the quantity 

 consumed, it appeared, after a few hours' trial, that to produce 



L 



