in Fluids. 381 



an oxyd by its union with oxygen, it may be dissolved 

 in and continue suspended in water; and it is clear that 

 the smaller any insulated particle of matter is, at the 

 surface of which Heat is generated in consequence of the 

 absorption of light, the more suddenly must the Heat 

 so generated be dispersed through the whole substance 

 of the particle, and the more equally and more intensely 

 must that particle be heated ; from hence it appears evi- 

 dently, that, if the particles of the oxyd dispersed about 

 in the water are but small enough, the Heat generated in 

 them by the sun's rays will be sufficient to expel the 

 oxygen united to the gold, and revive that metal. 



There is one very obvious objection that will doubt- 

 less bz made to this conclusion, which, however, may 

 easily be removed. The particle of the metallic oxyd 

 which is supposed to be heated is in contact with the 

 water ; how does it happen that a great part of this 

 Heat does not immediately pass off into that cold Fluid ? 

 I might answer, because both water and steam are non- 

 conductors of Heat, and might adduce in support of 

 this reason the well-known fact that a drop of water 

 dropped on a piece of iron, heated to most intense white 

 Heat, will remain some time on the iron without being 

 evaporated, even considerably longer than if the iron 

 were much less hot ; but a circumstance attending the 

 beautiful experiment in which iron is burned in oxygen 

 gas affords a more direct proof of the fact in question. 



As this experiment is commonly made, the iron, which 

 is a piece of small wire, a few inches long, is introduced 

 into a bottle, with a narrow neck, which contains the 

 oxygen gas ; the wire being fixed in its place, by causing 

 its upper end to pass through a cork stopple, which is 

 fitted to the mouth of the bottle. The lower end of 



