Chemistry and Physics. 263 
escape combination), it would, in virtue of the velocity of its 
own proper molecular motion, remove itself from our planet, 
and emigrate to a celestial body possessing sufficient gravitational 
attraction to hold it fast. Dr. Stoney suggests this explanation to 
account for the absence of an atmosphere and of water vapour on 
the moon, and for the presence of an atmosphere of hydrogen 
on the sun. It would also account for the absence of helium in 
our atmosphere, and for the presence of the chromospheric line 
D,. Of course if an element can form compounds, or if it is 
absorbed by solids, as helium appears to be, it will, like hydro- 
gen and helium, be found on the earth. 
The inertness of these gases would favor their existence in 
the free state. And argon exists in the atmosphere, precisely 
because it forms no compounds. Similarly nitrogen is a con- 
stituent of air, because in the first place those elements with 
which it combines directly are comparatively rare, and also 
because such compounds are mostly decomposed by water; and 
the excess of nitrogen therefore occurs in the free state. Simi- 
larly, the occurrence of free oxygen is due to the fact that some 
remains over, after all or almost all the readily oxidized sub- 
stances have already united with oxygen. If there exist gases 
similar to argon in inertness, they two may be looked for in air. 
Now if argon possesses the atomic weight 40, there is no place 
for it in the periodic table of the elements. And up to now 
there is no exception to this orderly arrangement, if the doubtful 
ease of tellurium be excluded. Rayleigh and Ramsay have 
shown that the high density of argon can hardly be accounted 
for by supposing that molecules of A, are mixed with molecules 
of A,; and excluding as untenable the supposition that argon isa 
compound, the only remaining suggestion is that it is a mixture. 
No attempts have as yet been made to test the correctness of this 
idea; but experiments have already been started which, it is 
hoped, will throw light on this question. 
The density of argon is too high; to fill its place in the 
periodic table, between chlorine and potassium, its density 
should be about 19 and its atomic weight 38. We might 
expect the presence of another element with a density of 41 
‘and an atomic weight of 82, to follow bromine, as argon follows 
chlorine; and this element would probably also be a gas, since 
its density would be only a little higber than that of chlorine. 
But here we meet with a difficulty. There are certain lines 
in the spectrum of helium coincident with lines in the argon 
spectrum. There can be only one explanation, excluding the 
extremely improbable hypothesis, which is not verified in any 
instance, that two elements may give spectra containing identical 
lines. That explanation is, of course, that each contains some 
common ingredient; and there appears to be a place for one with 
density 10 and atomic weight 20, to follow fluorine in the periodic 
table. The density of helium is, however, so low, that there 
does not appear room for any large quantity of a heavier gas; 
and to fit the periodic table, the density of argon should be 
