486 REPORTS ON THE STATE OF SCIENCE. 



combination in the rear of the wave except where there are two or more 

 chemical stages in the combustion. 



Section IV. — The Influence of Moisture upon Combustion. 



Thirty years ago H. B. Dixon, in repeating Bunsen's experiments on 

 the division of oxygen between carbon monoxide and hydrogen, both 

 present in excess, discovered that a mixture of carbon monoxide and 

 oxygen, dried by long contact, with phosphoric anhydride, will not explode 

 when sparked in the usual way in a eudiometer, whereas the presence 

 of a trace of moisture or of any gas containing hydrogen (e.g., methane, 

 ammonia, or hydrogen chloride) at once renders the mixture explosive. 

 These experiments, proving as they did the complexity of what at first 

 sight would appear to be one of the simplest cases of combustion, 

 opened up a new field of scientific investigation. 



Tn 1883 H. B. Baker, working in Dixon's laboratory at Balliol 

 College, Oxford, found that purified charcoal, when heated to redness in 

 carefully dried oxygen, burns with extreme slowness and without flame, 

 yielding principally the monoxide, the proportion of the dioxide formed 

 varying inversely with the degree of dryness of the oxvgen. In a further 

 series of experiments he proved that highly purified sulphur or phos- 

 phorus may be repeatedly distilled in an apparatus filled with carefully 

 dried oxygen, without any combustion whatever occurring, although the 

 admission of even a trace of moisture at once causes a vivid burning. 

 Tn subsequent investigations extending over a number of years, Baker 

 has shown that a large number of gaseous interactions are either con- 

 ditioned or greatly accelerated by the presence of moisture. Thus a 

 dried mixture of hydrogen and chlorine does not explode on exposure 

 to sunlight, dried ammonia and hydrogen chloride are mutually inert, 

 and dried electrolytic gas, free from hydrocarbon impurity, can be heated 

 to redness without exploding. 



The amount of moisture required to bring about such chemical 

 chancres as the above is surprisingly small. E. W. Morley has esti- 

 mated that the mere passing of a gas through a long column of 

 phosphoric anhydride leaves only 3 milligrams of water vapour in 

 a million litres (or rather less than 4 molecules of steam per 1,000 

 million molecules of gas), and yet a much more prolonged drying is 

 usually required to inhibit chemical action. Such facts as these, even 

 if they do not raise doubts as to the adequacy of the usually accepted 

 kinetic views of chemical processes, at least suggest the necessity of 

 some less stringent application of them. 



The dependence of the combustion of carbon monoxide upon the 

 presence of water vapour is well illustrated by H. B. Dixon's determina- 

 tion of the rates of explosion for mixtures of carbon monoxide and oxygen 

 in combining proportions containing varying amounts of water vapour. 

 Starting with a ' well-dried ' mixture, the rate of explosion increases 

 with successive additions of moisture, from 1.264 to a maximum of 1,738 

 metres per second for a mixture saturated at 35°, any further addition 

 of steam having a decidedly retarding influence, as follows: — 



