1876.] 
Recent Chemical Researches. 
35 
III. Classification of the Elements. 
We have already referred to Mendelejeff’s system of 
classification according to atomic weights : the more com- 
monly employed systems of classification are based upon 
the chemical value or valency of the individual elements. 
The researches of Michaelis, of Meyer, and of others, 
have important bearings upon the general question of 
valency. 
That the valency of each chemical element is invariable 
throughout the whole of its compounds has been and is 
maintained by many of our most distinguished chemists : 
the dodtrine opposed to this — viz., that the valency is a 
varying quantity — has also been upheld by able supporters. 
Before we are in a position to discuss this question it would 
be well that we should have a clear idea of what chemical 
valency means. Adopting the usually-received definition of 
5C atom ” and “ molecule,” we say that one atom of carbon 
is capable of combining with four atoms of hydrogen, while 
one atom of oxygen is capable of combining with only two 
atoms of hydrogen. The carbon atom is therefore said to 
be “ equivalent to ” four hydrogen atoms. But we must 
remember, as Mills has pointed out, that carbon and hydro- 
gen atoms have never been compared as to the work they 
can do under certain circumstances, and that the relations 
of their potential energies have yet to be determined. We 
are very apt to confuse two things which are, most probably, 
totally distindt — affinity and chemical valency. 
Michaelis imagines a carbon atom performing vibrations 
so that there are four positions during each complete vibra- 
tion in which it is possible for the atom to come within the 
sphere of adtion of another atom : he imagines an oxygen 
atom performing vibrations so that there are but two posi- 
tions during each complete vibration in which it may come 
within the sphere of adtion of another atom. Or we may 
imagine the carbon atom as exercising force in four different 
diredtions, while the oxygen atom exercises force in but two 
diredtions. The number of positions of advantage, as we 
might call them, or of lines of force, may then be indepen- 
dent of the total force exercised. We may imagine an atom 
exercising force in but two diredtions, and nevertheless 
exercising an absolutely greater amount of force than another 
atom the diredtions of the exercise of whose force amount 
to four in number. The total force exercised is therefore 
the affinity, and is dependent upon the nature, position, &c., 
of all the atoms in the compound : the number of diredtions, 
