ABUNDANCE OF THE CHEMICAL ELEMENTS. 141 
at 56 in iron. Above 56 the elements are comparatively 
rare, and only two of them, barium and strontium, appear 
in my estimates. Below oxygen, hydrogen alone approaches 
one per cent., while between oxygen and iron only scandium 
and vanadium are of neglectable rarity. Furthermore, in 
several elementary groups abundance diminishes with in¬ 
crease of atomic weight. This is plainly seen in the series 
potassium, rubidium, and caesium; in sulphur, selenium, 
and tellurium ; in chlorine, bromine, and iodine; in arsenic, 
antimony, bismuth, etc., etc. The regularity is not certainly 
invariable, but it. occurs often enough to be suggestive. 
Perhaps a part of the difficulty in tracing relations of this 
sort arises from the fact that our field of view is limited to 
the earth and does not include the whole solar system. 
Indeed, several writers, reasoning on the broader basis of 
the nebular hypothesis and noting the low densities of the 
outer planets, have argued that the latter may contain the 
lighter elements mainly, while the heavier substances are 
concentrated at the original nucleus, the sun. Along this 
line, however, close reasoning is impossible, partly because 
evidence is lacking and partly because the conditions of 
temperature and pressure differ so widely between the sun 
and the other heavenly bodies. As thus attacked, the prob¬ 
lem becomes one of enormous complexity, and even the solar 
sprectrum gives us no conclusive evidence. That oxygen 
and silicon are not conspicuous in the sun’s atmosphere we 
may admit, but that non-volatile silica is absent from the 
solar body is by no means certain. We may assume that 
compounds can exist nowhere in the sun, but the assump¬ 
tion is unprovable. As to the composition of the outer 
planets we know practically nothing. How far they may 
differ from each other, from the earth, and from the sun is 
as yet a matter of pure-conjecture. 
If, despite Mendelejeff’s recent demurrer, we assume that 
the elements have been evolved from one primordial form 
of matter, their relative abundance becomes suggesti ve. Start¬ 
ing from the original “ protyle,” as Crookes has called it, the 
process of evolution seems to have gone on slowly until oxy- 
