80 On the supposed Chemical Affinity 



1 . The constant proportion of azot and oxygen in the 

 atmosphere is considered as an argument for their being 

 held by affinity. So indeed it may ; but it is equally in 

 favour of my hypothesis, and, therefore, nothing tending 

 to decide the question can be obtained from it. For, let 

 part of the oxygen be abstracted any where from the atmo- 

 sphere, then the azot may be supposed to attract the ox- 

 ygen from the vicinity, and thus the equilibrium be resto- 

 red; but it is certainly equally satisfactory to suppose that, 

 the oxygen in the vicinity meeting with less repulsive power 

 from the deficient quarter, nothing prevents its diffusion 

 into that quarter but the azot previously there, which* by 

 hvpothesis, can only retard, but by no means prevent, the 

 effect. Thus, then, whether the azot attract the oxygen, 

 or the oxygen repel itself, the effect is precisely the same. 

 From this fact simply it is impossible, therefore, to decide, 

 the merits of either theory; but if it be found that any 

 one gas diffuses itself in any other with nearly the same 

 celerity, it will be a presumption in favour of my hypothe- 

 sis' ; if otherwise, it may be urged that the quicker diffusion 

 is owing to the stronger affinity. I have made a great num- 

 ber of experiments on this head, but could not find any 

 remarkable difference in the time and circumstances of dif- 

 fusion of the same gas. 



2. The experiments of Morozzo and Humboldt show that 

 air possesses different properties, from a mere mixture of its 

 two component parts. I do not credit the experiments. Hum- 

 boldt finds a variable quantity of oxygen, from 25 to 30, or 

 more, per cent, in the air ; whereas others, who are more 

 accurate, find but 21, or at most 22, and that constant. It 

 is no wonder then, if he mix 28 oxygen and 72 azot, that 

 the mixture diminishes nitrous gas more than air, and sup- 

 ports combustion and animal life for a longer time. 



3. iC Different combustibles are capable of absorbing dif- 

 ferent portions of oxygen from a given quantity of air : 

 phosphorus 22 per cent, sulphur 8, &c." The only infe- 

 rences I draw from these facts are, that phosphorus will 

 burn in oxygen of any density; that sulphur will not burn 

 in oxygen, unless it be of \ of atmospheric density, or more. 

 The difference in the phenomena of combustion in air and 

 in oxygen is not to be ascribed to the combination of azot 

 and oxygen, but to the less density of the latter, ~ of what 

 a pure atmosphere of the same gas would be. From an in- 

 cidental, but imperfect, trial I made, in conjunction with 

 Mr. Davy last winter, I have no doubt but iron wire would 

 burn in common air of five times the density with brilli- 

 ancy, 



