TRANSACTIONS OF SECTION B. 451 
Of this ever-growing list of elements not one has been resolved into simpler 
bodies for three-quarters of a century; and we, who are removed by two or three 
generations from the great builders of our science, are tempted to look upon these 
bedies as though they were really simple forms of matter, not only unresolved, 
but unresolvable. The notation we employ favours this view and stamps it upon 
our minds. 
Is it, however, a fact that these reputed elements are really simple bodies? or, 
indeed, are they widely different in the nature of their constitution from those bodies 
which we know to be chemical compounds? Thus, to take a particular instance, 
are fluorine, chlorine, bromine, and iodine essentially distinct in their nature from 
the compound halogens, cyanogen, sulphocyanogen, ferricyanogen, &c.? Are the 
metals lithium, sodium, and potassium essentially distinct from such alkaline bases 
as ammonium, ethylamine, di-ethylamine, &c.? No philosophical chemist would 
robably venture to answer this question categorically with either ‘ yes’ or ‘no.’ 
Tt us endeavour to approach it from three different points of attack—(1) the 
evidence of the spectroscope, (2) certain peculiarities of the atomic weights, and 
(8) specific refraction. 
1. The Spectroscospe.—It was at first hoped that the spectroscope might throw 
much light upon the nature of elements, and might reveal a common constituent in 
two or more of them; thus, for instance, it was conceivable that the spectrum lines 
of bromine or iodine vapour might consist of the rays given by chlorine plus some 
others. All expectations of this have hitherto been disappointed: what we do 
frequently find is a certain similarity of character among the spectra of analogous 
elements, not rays of identically the same refrangibility. Yet, on the other hand, 
it must not be supposed that such a negative result disproves the compound nature 
of elements, for as investigation proceeds it becomes more and more clear that the 
spectrum of a compound is not made up of the spectra of its component parts. 
Again, the multiplicity of rays given out by some elements, when heated, in 
a gaseous condition, such us iron, has been supposed to indicate a more complex 
constitution than in the case of those metals, such as magnesium, which give a 
more simple spectrum. Yet it is perfectly conceivable that this may be due to a 
complexity of arrangement of atoms all of the same kind. 
Again, we have changes of a spectrum at different temperatures; new rays 
appear, others disappear ; or even there occurs the very remarkable change from a 
fluted spectrum to one of sharp lines at irregular intervals, or to certain recurring 
groups of lines. This, in all probability, does arise from some redistribution, but 
it may be a redistribution in a molecular grouping of atoms of the same kind, and 
not a dissociation or rearrangement of dissimilar atoms. 
A stronger argument has been derived from the revelations of the spectroscope 
in regard to the luminous atmospheres of the sun. There we can watch the effect 
of heat enormously transcending that of our hottest furnaces, and of movements 
compared with which our hurricanes and whirlwinds are the gentlest of zephyrs. 
Mr. Lockyer, in studying the prismatic spectra of the luminous prominences or 
spots of the sun, has frequently observed that on certain days certains lines, say of 
the iron spectrum, are non-existent, and on other days certain other lines disappear, 
and that in almost endless variety ; and he has also remarked that occasionally cer- 
tain lines of the iron spectrum will be crooked or displaced, thus showing the 
vapour to be in very rapid motion, while others are straight, and therefore com- 
paratively at rest. Now,asa gas cannot be both at rest and in motion at the same 
time and the same place, it seems very clear that the two sets of lines must originate 
in two distinct layers of atmosphere, one above the other, and Mr. Lockyer’s con- 
clusion is that the iron molecule was dissociated by heat, and that its different 
constituents, on account of their different volatility, or some other cause, had 
floated away from one another. This seems to me the easiest explanation of the 
_ phenomenon ; and, as dissociation by heat is a very common occurrence, there 
is no @ priori improbability about it. But we are not shut up to it, for the 
different layers of atmosphere are certainly at different temperatures, and most 
_ probably of different composition. If they are of different temperatures the 
_ variations of the spectrum may only be an extreme case of what must be acknow- 
i GG@2 
3 
