108 Mr. J, G. and the Rev. T. Knox om Fluorine. 



which we inferred that fluorine and not hydrofluoi'ic acid had 

 been present in the vessel, which was also further confirmed 

 by the absence of fumes when the vessel and its contents had 

 been previously dried. 



Placing inverted over the orifice of the vessel a clear cry- 

 stal of fluor spar, with a small perforation in the centre into 

 which a stopper of fluor spar fitted accurately, on the stopper 

 falling into the vessel the tube was filled with a yellowish 

 green gas, the colour of which deepened with heat, and dis- 

 appeared when cold. On reheating the vessel below, the gas 

 rose again into the crystal above. On removing the crystal 

 while hot to a wet glass plate, it flew to pieces, which pre- 

 vented us from determining whether the coloured gas was bi- 

 chloride of mercury under heat and pressure, hydrofluoric 

 acid, or fluorine. 



Having procured larger vessels with receivers into which 

 ground stoppers were made to fit accurately, we resumed in 

 the present month the experiments we had tried in the be- 

 ginning of April. 



1st Exp. We heated fluoride of lead with oxygen, and 

 afterwards with dry chlorine without action upon the fluoride. 

 When the receiver (its stopper having fallen into the vessel 

 below) was placed over gold-leaf, a chloride of gold was 

 formed. 



2nd Exp. Treating hydrofluate of ammonia similarly with 

 chlorine, there was strong action on glass and formation of 

 chloride of gold as before. 



3rd Exp. Treating fluoride of mercury with chlorine (as 

 we had done in our former experiments), we obtained crystals 

 of bichloride of mercury in the vessel. Leaving the receiver 

 over gold-leaf, there was after a considerable time action on it, 

 producing a yellowish brown appearance. This we placed on a 

 slip of glass, and on adding a few drops of sulphuric acid and 

 evaporating to dryness there was very strong action on the glass 

 where the gold had been, proving that it was a fluoride of 

 gold, and that since gold is not acted on by hydrofluoric acid 

 there must have been fluorine in the receiver. As an addi- 

 tional corroboration there was no hydrogen in the tube, which 

 there would have been had hydrofluoric acid been decom- 

 posed by the gold. From these experiments we conclude that 

 fluorine was present in the receiver, but whether a slight trace 

 of hvdrofluoric acid (to which the action on glass was due) 

 may not have been present with it, we have not yet determined. 

 We hope on a future occasion to be able to give particulars 

 with regard to the properties of the gas ; but we consider that 



