Following out these views, I some time since communicated a paper to the Society on the spectrum of calcium, to which I shall refer more expressly in the sequel.

Differentiation of the Phenomena to be observed on the Two Hypotheses.

When the reductions of the observations made on metallic spectra, on the hypothesis that the elements were really elementary, had landed me in the state of utter confusion to which I have already referred, I at once made up my mind to try the other hypothesis, and therefore at once sought for a critical differentiation of the phenomena on the two hypotheses.

Obviously the first thing to be done was to inquire whether one hypothesis would explain these short-line coincidences which remained after the reduction of all the observations on the other. Calling for simplicity's sake the short lines common to many spectra basic lines, the new hypothesis, to be of any value, should present us with a state of things in which basic molecules representing bases of the so-called elements should give us their lines, varying in intensity from one condition to another, the conditions representing various compoundings. Suppose A to contain B as an impurity and as an element, what will be the difference in the spectroscopic result?

A in both cases will have a spectrum of its own;

B as an impurity will add its lines according to the amount of impurity, as I have shown in previous papers.

B as an element will add its lines according to the amount of dissociation, as I have also shown.

The difference in the phenomena, therefore, will be that, with gradually-increasing temperature, the spectrum of A will fade, if it be a compound body, as it will be increasingly dissociated, and it will not fade if it be a simple one.

Again, on the hypothesis that A is a compound body, that is, one compounded of at least two similar or dissimilar molecular groupings, then the longest lines at one temperature will not be the longest at another; the whole fabric of "impurity elimination," based upon the assumed single molecular grouping, falls to pieces, and the origin of the basic lines is at once evident.

This may be rendered clearer by some general considerations of another order.

Let us further assume that in A there exists a substance a by itself competent to form a compound body ẞ by union with itself or with something else when the temperature is lowered.

Then we may imagine a furnace B in which this compound body exists alone. The spectrum of the compound ẞ would be the only one visible in B, as the spectrum of the assumed elementary body a would be the only one visible in A.

A lower temperature furnace C will provide us with a more compound substance y, and the same considerations will hold good.

Now if into the furnace A we throw some of this doubly-compounded body, we shall get at first an integration of the three spectra to which I have drawn attention; the lines of y will first be thickest, then those of ẞ; finally a will exist alone, and the spectrum will be reduced to one of the utmost simplicity.

This is not the only conclusion to be drawn from these considerations. Although we have by hypothesis ẞ, y, and all higher, that is, more compound forms of a, and although the strong lines in the diagram may represent the true spectra of these substances in the furnaces B, C, and D, respectively, yet, in consequence of incomplete dissociation, the strong lines of ẞ will be seen in furnace C, and the strong lines of y will be seen in furnace D, all as thin lines. Thus, although in C we have no line which is not represented in D, the intensities of the lines in C and D are entirely changed.

In short, the line of a strong in A is basic in B, C, and D, the lines of ẞ strong in B are basic in C and D, and so on.

I have prepared another diagram which represents the facts on the supposition that the furnace A, instead of having a temperature sufficient to dissociate B, y, and 8 into a is far below that stage, although higher than B.

It will be seen from this diagram that then the only difference in the spectra of the bodies existing in the four furnaces would consist in the relative thicknesses of the lines. The spectrum of the sub

*The figures between the hypothetical spectra point to the gradual change as the spectrum is observed near the temperature of each of the furnaces.

hey exist in A would contain as many lines as would 1 of the substances as they exist in D; each line would in

Fig. 2.

c in the whole series of furnaces instead of in one or two

of these General Considerations to Impurity Elimination. as suppose that in the last diagram (Fig. 2) the four resent the spectra of say, iron, broken up into different successive stages of heat. It is first of all abundantly e relative thicknesses of the iron lines observed will vary the temperature resembles that of A, B, C, or D. The the spectra will be the same, but the intensities will vary point. The longest lines, represented in the diagram by ones, will vary as we pass from one temperature to is on this ground that I have before stated that the of impurity elimination must fall to pieces on such an Let us suppose, for instance, that manganese is a comform of iron represented in furnace B, with something. opose again that the photograph of iron which I compare ese represents the spectrum of the vapour at the tempefurnace D. To eliminate the impurity of iron in manganese, ninated it, we begin the search by looking for the longest lines shown in the photograph of iron, in the photograph e taken under the same conditions. I do not find these therefore, that there is no impurity of iron in manganese, the longest iron lines are not there, some of the fainter This I hold to be the explanation of the apparent conich we are landed on the supposition that the elements y..

Application of these Considerations to Known Compounds.

Now to apply this reasoning to the dissociation of a known compound body into its elements

A compound body, such as a salt of calcium, has as definite a spectrum as a simple one; but while the spectrum of the metal itself consists of lines, the number and thickness of some of which increase with increased quantity, the spectrum of the compound consists in the main of channelled spaces and bands, which increase in like manner.

In short, the molecules of a simple body and a compound one are affected in the same manner by quantity in so far as their spectra are concerned; in other words, both spectra have their long and short lines, the lines in the spectrum of the element being represented by bands or fluted lines in the spectrum of the compound; and in each case the greatest simplicity of the spectrum depends upon the smallest quantity, and the greatest complexity (a continuous spectrum) upon the greatest.

The heat required to act upon such a compound as a salt of calcium so as to render its spectrum visible, dissociates the compound according to its volatility; the number of true metallic lines which thus appear is a measure of the quantity of the metal resulting from the dissociation, and as the metal lines increase in number, the compound bands thin out.

I have shown in previous papers how we have been led to the conclusion that binary compounds have spectra of their own, and how this idea has been established by considerations having for a basis the observations of the long and short lines.

It is absolutely similar observations and similar reasoning which I have to bring forward in discussing the compound nature of the chemical elements themselves.

In a paper communicated to the Royal Society in 1874, referring, among other matters, to the reversal of some lines in the solar spectrum, I remarked:

"It is obvious that greater attention will have to be given to the precise character as well as to the position of each of the Fraunhofer lines, in the thickness of which I have already observed several anomalies. I may refer more particularly at present to the two H lines 3933 and 3968 belonging to calcium, which are much thicker in all photographs of the solar spectrum [I might have added that they were by far the thickest lines in the solar spectrum] than the largest calcium line of this region (4226-3), this latter being invariably thicker than the H lines in all photographs of the calcium spectrum, and remaining, moreover, visible in the spectrum of substances con

* "Phil. Trans.," vol. clxiv, part 2, p. 807.

taining calcium in such small quantities as not to show any traces of the H lines.

"How far this and similar variations between photographic records and the solar spectrum are due to causes incident to the photographic record itself, or to variations in the intensities of the various molecular vibrations under solar and terrestrial conditions, are questions which up to the present time I have been unable to discuss.

An Objection Discussed.

I was careful at the very commencement of this paper to point out that the conclusions I have advanced are based upon the analogies furnished by those bodies which, by common consent and beyond cavil and discussion, are compound bodies. Indeed, had I not been careful to urge this point the remark might have been made that the various changes in the spectra to which I shall draw attention are not the results of successive dissociations, but are effects due to putting the same mass into different kinds of vibration or of producing the vibration in different ways. Thus the many high notes, both true and false, which can be produced out of a bell with or without its fundamental one, might have been put forward as analogous with those spectral lines which are produced at different degrees of temperature with or without the line, due to each substance when vibrating visibly with the lowest temperature. To this argument, however, if it were brought forward, the reply would be that it proves too much. If it demonstrates that the h hydrogen line in the sun is produced by the same molecular grouping of hydrogen as that which gives us two green lines only when the weakest possible spark is taken in hydrogen inclosed in a large glass globe, it also proves that calcium is identical with its salts. For we can get the spectrum of any of the salts alone without its common base, calcium, as we can get the green lines of hydrogen without the red one.

I submit, therefore, that the argument founded on the overnotes of a sounding body, such as a bell, cannot be urged by any one who believes in the existence of any compound bodies at all, because there is no spectroscopic break between acknowledged compounds and the supposed elementary bodies. The spectroscopic differences between calcium itself at different temperatures is, as I shall show, as great as when we pass from known compounds of calcium to calcium itself. There is a perfect continuity of phenomena from one end of the scale of temperature to the other.

Inquiry into the Probable Arrangement of the Basic Molecules.

As the results obtained from the above considerations seemed to be so far satisfactory, inasmuch as they at once furnished an explanation

VOL. XXVIII.

N