382 Of the Propagation of Heat 



the wire is pointed ; and it is set on fire by being first 

 heated in the flame of a candle, and then plunged sud- 

 denly, while red-hotj into the bottle. The combustion 

 begins the moment the end of the wire enters the oxy- 

 gen gas ; and the metal continues to burn with the ut- 

 most violence, and with a copious emission of intense 

 white light, till the wire, or till all the gas is consumed, 

 affording one of the most brilliant and most interesting 

 sights that can be imagined. 



The product of this combustion is the oxygenation 

 of the iron ; and this metallic oxyd, in a state effusion, 

 and heated to the most intense white Heat, falls to the 

 bottom of the bottle in globules of different sizes. 



To protect the glass against these drops of calx of 

 iron in fusion, it is usual to leave a quantity of cold 

 water in the bottle, enough, for instance, to cover its 

 bottom to the height of about an inch ; but I have fre- 

 quently seen numbers of these globules, much smaller 

 than peas, which have not only descended red-hot through 

 the water, but have remained red-hot at the bottom of 

 the bottle, surrounded by the water, at least two or three 

 seconds, and actually melted the glass on which they re- 

 posed (and as far as I can recollect) without producing 

 the smallest appearance of steam. 



The water could not be decomposed, for the iron was 

 already saturated with oxygen. 



This experiment will, I fancy, be considered as afford- 

 ing an indisputable proof that intense Heat may exist, at 

 least for a short time, in a small particle of matter sur- 

 rounded by a cold Fluid. 



Now, as it has been found by actual experiment, that 

 when a solution of nitro-muriate of eold in water is ex- 



D 



posed to the action of the sun's rays, the gold is revived ; 



