HEAT. 65 



tricity, much being dissipated, without change, in the form of 

 heat. 



Heat, however, directly affects and modifies both the mag- 

 net and chemical compounds ; the union of certain chemical 

 substances is induced by heat, as, for instance, the formation 

 of water by the union of oxygen and hydrogen gases : in 

 other cases this union is facilitated by heat, and in many in- 

 stances, as in ammonia and its salts, it is weakened or antag- 

 onised. In many of these cases, however, the force of heat 

 seems more a determining than a producing influence ; yet to 

 be this, it must have an immediate relation with the force 

 whose reaction it determines : thus, although gunpowder, 

 touched with an ignited wire, subsequently carries on its own 

 combustion or chemical combination, independently of the 

 original source of heat, yet the chemical affinities of the first 

 portion touched must be exalted by, and at the cost of, the 

 heat of the wire ; for to disturb even an unstable equilibrium 

 requires a force in direct relation with those which maintain 

 equilibrium. 



Since the first edition of this essay was published, I have 

 communicated to the Royal Society some experiments by 

 which an important exception to the general effect of heat on 

 chemical affinity is removed, and the results of which induce 

 a hope that a generalised relation will ultimately be estab- 

 lished between heat, chemical affinity, and physical attraction. 

 I find that if a substance capable of supporting an intense 

 heat, and incapable of being acted upon by water or either 

 of its elements such, for instance, as platinum, or iridium 

 be raised to a high point of ignition and then immersed in 

 water, bubbles of permanent gas ascend from it, which on 

 examination are found to consist of mixed oxygen and hydro- 

 gen in the proportions in which they form water. The tem- 

 perature at which this is effected is, according to Dr. Robin- 

 son, who has since written a valuable paper on the subject, 

 = 2386. Now, when mixed oxygen and hydrogen are ex- 



