496 REPORTS ON THE STATE OF SCIENCE. 



immediately became red hot, and remained so until nearly the whole of 

 the oxygen "had disappeared. In subsequent experiments, Davy proved 

 that hydrogen is far more susceptible to surface combustion than 

 either ethylene or carbon monoxide; also, that the power of inducing 

 surface combustion is by no means confined to the metals of the platinum 

 group, which, however, exhibit it in an eminent degree. 



In 1823 the subject was systematically investigated by Dulong and 

 Thenard, and independently also by Dobereiner, who showed that all 

 solids possess the power of accelerating combustion in varying degrees 

 according to their specific characters and fineness of division. Two years 

 Inter William Henry observed that when a platinum ball is immersed 

 in a mixture of equal volumes of ethylene and electrolytic gas the 

 hydrogen and oxygen alone combine, no combustion of the hydrocarbon 

 occurring unless the original mixture contains a much larger proportion 

 of oxygen. This important result was confirmed by Thomas Graham 

 in 1829. 



Several explanations of surface combustion were put forward by 

 these early investigators. Davy himself suggested an electrochemical 

 one. ' Supposing,' he wrote, ' oxygen and hydrogen to be in the rela- 

 tion of negative and positive, it is necessary to effect their combination 

 that their electricities should be brought into equilibrium or discharged. 

 This is done by the electrical spark or flame, which offers a conducting 

 medium for this purpose, or by raising them to a temperature in which 

 they become themselves conductors. Now platinum, palladium, and 

 iridium are bodies very slightly positive with respect to oxygen. . . . 

 They offer to the gases the conducting medium necessary for carrying 

 off and bringing into equilibrium their electricities without any inter- 

 vening energy, and accumulate the heat produced by this equilibrium.' 

 Dobereiner, who discovered that freshly prepared platinum black absorbs 

 oxygen from the air, and that in this ' oxygenated ' condition it will 

 cause steam to be formed on being plunged into a jar of oxygen, con- 

 tended that the metal merely acts as a carrier of oxygen. On the other 

 hand, Fusinieri (1825) maintained that it is the combustible gas 

 (hydrogen) only which is affected by the surface, being condensed and 

 rendered extraordinarily active by association with the surface. 



The matter formed the subject of a celebrated controversy between 

 Faraday and De la Rive in 1834-35. De la Rive strongly upheld the view 

 that surface combustion essentially consists of a series of rapidly alter- 

 nating oxidations and reductions of the catalysing material. Faraday, 

 whilst not denying that finely divided platinum absorbs oxygen, argued 

 with great force that true surface combustion involves an action quite 

 distinct from that of an oxidised wire or foil upon a combustible gas. 

 The function of the solid is, he contended, to condense both the oxygen 

 and the combustible gas at the surface, thus producing a condition in 

 the surface layer comparable to that of high pressure. After the year 

 1896, however, interest in the subject waned, and was not revived until 

 quite recently. 



It may here be remarked that heated surfaces have undoubtedly 

 a marked influence in accelerating not only combustion but all 

 chemical interchanges in gaseous systems. It is usually considered 



