1876.] Recent Chemical Researches. 37 
compound, has yet been given. Michaelis proposes to adopt 
the rule that when a doubtful compound yields an atomic 
compound by double decomposition it is itself an atomic 
compound. Inasmuch as ammonium chloride yields am- 
monia (the molecule of which is represented by the formula 
NH 3 ) by the aCtion upon it of lime, we should probably be 
obliged to class this compound among those which are 
atomic. But we know that the aCtion of heat upon ammo- 
nium chloride resolves it into ammonia and hydrochloric 
acid ; is it not, then, very probable that the seeming double 
decomposition which takes place between sal-ammoniac and 
lime is in reality an aCtion between hydrochloric acid and 
lime, and that the first aCtion of the heat is to decompose 
the ammonium chloride with the production of ammonia, 
which escapes, and hydrochloric acid which aCts upon the 
lime ? In the case of such compounds as ammonium 
chloride it is very difficult to determine whether cases of 
seeming double decomposition really belong to this category 
or not. 
But if we agree with Michaelis’s general view of valency 
- — viz., that it is measured by the number of directions in 
which the energy of a chemical element is exercised, but 
that the amount of energy need not be the same in each 
direction — we may imagine that the weaker lines of force 
are those which are concerned in the formation of the so- 
called molecular compounds. Let us take the case of sal- 
ammoniac. Ammonia gas and hydrochloric acid gas, when 
brought together in proper proportions, are no longer to be 
distinguished as such ; their properties become merged in 
those of the new solid body formed. But if this new body 
be heated it is resolved into ammonia and hydrochloric acid ; 
if the mixture of heated gases be allowed to cool the solid 
body is re-formed. Let us suppose that there are five posi- 
tions in which the nitrogen atom can exercise its chemical 
energy during each complete vibration. In ammonia the 
nitrogen atom is brought within the sphere of action of 
three hydrogen atoms, and we may suppose that during each 
complete vibration there are three positions at which the 
attractive force of each hydrogen atom in turn changes to a 
repulsive force ; the nitrogen atom therefore oscillates about 
these three positions, and at the same time about two other 
positions at which it is possible for other atoms to exercise 
an aCtion upon it. And it is at these two latter positions 
that the aCtion of the hydrochloric acid molecule is exhibited. 
But inasmuch as we know that the total energy of chlorine 
is large, although it be exerted in but one direction, while 
