368 Lf. A. Hill—WNotes on Argon and Helium. 
increased. ‘T’he same is equally true of N,O and H. Why 
not also of argon? ‘The characteristics of argon are preémi- 
nently non-metallic, the difference (2°6°) between its melting 
and boiling points is small, as it should be, and not large as in 
the case of metallic mercury (895°5°); hence the inference 
that like other metalloids its upper limit of combination is far 
removed from its boiling point, much farther so than in the 
case of mercury, that is, far above 800°. But we are asked to 
believe that it is even below the ordinary temperature of the 
laboratory. Everything, however, seems to me to point to the 
contrary conclusion, that argon, like hydrogen, nitrogen, oxy- 
gen, chlorine, bromine and the other non-metallic elements of 
its class, has a very high upper limit of combination, and not a 
low one like the metals; and hence that at ordinary tempera- 
tures it is no more above that limit than are oxygen and nitro- 
gen, but that its chemical inertness is fully explained by 
assuming that the force binding together the atoms within the 
molecule is strong as in the case of nitrogen (and to a less 
degree in hydrogen), and that the contrary view is not justified 
by known facts. 
And this brings us back to our starting point. If the inert 
character of argon is not due to heat, then how can it be mon- 
atomic and also devoid of affinity ? | 
To quote from my previous article—‘ which is the more 
unique, a diatomic gas without rotational energy, or a free 
atom devoid of chemical affinity?’ The presumption is at 
once strongly raised that it is not a monatomic gas but diatomic 
and chemically inert because the two atoms of the molecule 
are very strongly bound together, in fact too strongly for 
any ordinary chemical affinity to overcome the bond, and that 
perhaps afterall there is some hitherto unsuspected weak link in 
the chain of reasoning supposed to prove its monatomicity, or, 
as was stated in the editorial in Nature (Feb. 7, 1895): “ Till 
' further evidence is forthcoming, a heavy strain is thrown on the ~ 
link of the chain of argument which connects the ratio of the 
specific heats with the monatomicity of the gas.” I have 
already pointed out that the conclusion as to monatomicity 
depends on the correctness of various assumptions which have 
already been called in question, viz: 1st, that when the rota- 
tional energy is small the gas must necessarily be monatomic; 
2d, that the ratio of the two specific heats of 1°67 proves 
absence of rotational energy, and hence monatomicity, and 3d, 
that the method used for determining this ratio can give cor- 
rect results. If any one of these propositions fails, monatomicity 
ean hardly be sustained; and any evidence against mon- 
atomicity, like that we have been considering, even though not 
conclusive of itself, yet when construed with these three pro- 
