298 Affinity and Fleat. 
is this study, and how independent of hypothesis, of any 
unknown force, unknown even, from the point of view, of 
the kind of unit to which its exact or approximate mea- 
sure is to be referred. Itis in this case especially that 
affinity, regarded as a force, is an unknown cause, at least 
when it is not merely the expression of a quality of matter ; 
in this case, it should merely serve to designate the fact 
that such and such substances can or cannot combine under 
certain well-defined conditions. 
One point is very urgent—that is, to define combination 
itself. Seeking attentively the state of our knowledge in 
this respect, I find, in the first place, that the definition of 
combination must comprehend solution, and can only ex- 
clude mixture. In fact, the phenomena of solution and of 
combination are joined by a continuous chain, which it is 
impossible to break at any one point. Everyone knows 
that interminable discussions have been raised on this sub- 
ject—-for instance ; in respect to metallic alloys and their 
liquation ; in reference to salts, acids, energetic bases, and 
their action upon water, alcohol, &c. The most certain 
conclusion which can be drawn is, that there are all possible 
intermediate stages between the phenomena of combination 
and the best-characterised phenomena of solution. 
In endeavouring to formulate in the clearest manner 
the ideas now current on combination, I find it cannot be 
better defined than by the fact of change of state. When 
two substances put in presence change condition, they com- 
bine. This change of condition, defined in the ordinary 
manner, shows itself by some new property, whether 
physical or chemical, which discloses new qualities of the 
combined substances—qualities, whether physical or che- 
mical, which distinguish the combination from simple 
mixture. I will give an example. 
The question has been asked, Is air a mixture or a com- 
bination? How has this question been solved? The 
physical and chemical properties of the elements of air 
(oxygen and nitrogen), and of air itself, have been suc- 
cessively studied. They have been found to be identically 
the same, always equal in the case of air to the mean of. 
these properties, determined upon each element simply. 
Hence, it has been concluded that nitrogen and oxygen, 
coming into contact under ordinary atmospheric conditions, 
exert no appreciable action on one another ; that, therefore, 
the condition of the gas has undergone no change; that, in © 
short, air is a mixture, and not a combination, of nitrogen 
and oxygen. ~ 
