everything is unknown. Enquiry in such a case is useless, and speculation vain.

The stars, as revealed to us by the spectroscope are all of the same order as the Sun, but they are not all of the same species. Quite a large number of stars, of which Arcturus is one of the best-known examples, show spectra that are essentially the same as that of the Sun, but there are other stars of which the spectra bear little or no semblance to it. Nevertheless, it remains true that, on the whole, stellar spectra bear witness to the presence of just the same elements as we recognize in the Sun, though not always in the same proportions or in the same conditions—hydrogen, calcium, sodium, magnesium, iron, titanium, and many more are recognized in nearly all. It is true that not all the known terrestrial elements have yet been identified in either Sun or stars; but, in general, those missing are either “negative” elements like the halogens, or elements of great atomic weight like mercury and platinum. That elements of one class should, as a rule, reveal their presence in Sun and stars wherever these are placed, and, correspondingly, that other classes should as generally fail to show themselves, indicate that such absence is more likely to be due to the general structure of the stellar photospheres and reversing layers than to any irregularity in the distribution of matter in the universe. It is easy, for example, to conceive that the heavy metals may lie somewhat deeper down within the Sun or star than those of low atomic weight. In the case of the Sun, there seems a clear connection between atomic weight and the distinctness with which the element is recognized in the spectrum of the photosphere, the lower atomic weights showing themselves more conspicuously.

It is clear that not all elements present in a Sun or star show themselves in its spectrum. Oxygen is very feebly represented by its elemental lines, but the flutings of titanium oxide are found in sunspots, and with great distinctness in a certain type of stars. Nitrogen, too, though not directly recognized, proves its presence by the lines of cyanogen. The case of helium is one of particular interest; this element was recognized by a very bright yellow line in the solar prominences before it was known to exist on the Earth; indeed, it received the name helium because it then seemed to be a purely solar constituent. Now it is seen as a strong absorption line in the spectrum of many stars; but for some reason it is not in general seen as an absorption line over the Sun’s disc, and if our Sun were removed to such distance so as to appear to us only as a star, we should have no evidence that it contained any helium at all. So far, then, as the evidence of the spectroscope goes, the elements present in the Earth are present throughout the whole extent of the universe within our view: the same elements and with the same qualities. For the lines of the spectrum of an element are the revelation of its innermost molecular structure, so that we can confidently affirm that hydrogen and oxygen on Sirius, Arcturus, or the Sun, are essentially the same elements as hydrogen and oxygen on the Earth. On a planet attached to any of these stars, the two gases would combine together to form water under just the same conditions as they do here on the Earth; and at suitable temperatures that water would be a neutral liquid, capable of dissolving just the same chemical substances that it does here. It would freeze as it does here; it would evaporate as it does here; it would be water as completely in all its qualities and conditions as earthly water is. And what applies to one element or compound applies to all. Throughout the whole extent of space, the same building materials have been employed, and throughout they retain the same qualities.

Hydrogen is seen in the spectra of nearly all stars, and also in those of nebulæ. The elemental lines of oxygen are not indeed seen in stellar spectra, but that the element is present is shown by the flutings of titanium oxide which distinguish stars like Antares. Nitrogen and carbon again are not recognized by their elemental lines, but the lines of cyanogen are seen in the spectra of comets and of sunspots, and hydrocarbon flutings in the spectra of comets and red stars; while in a few of the hottest stars even sulphur has recently been identified.[10] All the five organo-genetic elements are therefore abundantly diffused through space; the materials for protoplasm, “the albuminous substance with water,” are at hand everywhere. This being so, it is reasonable to infer that if organic life exists elsewhere than on this Earth, its essential feature, there as here, is the metabolism of nitrogenous carbon compounds in association with protoplasm.

But it is objected that “we are not yet able to identify all the lines in solar or stellar spectra; may not some of these lines be due to elements of which we know nothing here, and may not such new elements form complex and unstable compounds with each other, or with some of those familiar to us, that would take the place of the five organo-generators, and so give rise to a physical basis of life, different from that we know on this Earth?”

But the development of Mendeléeff’s Periodic Law has shown that the elements are not to be regarded as disconnected entities. The Law as given in Mendeléeff’s own words, runs: “The properties of the elements as well as the forms and properties of their compounds are in periodic dependence on, or (expressing ourselves algebraically) form a periodic function of the atomic weights of the elements.” In other words, they form a series, not only as it regards their atomic weights, but also as it regards their own properties and the forms and properties of their compounds. We are no longer at liberty, as we might have been many years ago, to call into fancied existence new elements having no relation in their properties and compounds to those with which we are acquainted. New elements, no doubt, will be discovered in the future, as in the past; and indeed we may be able to discover them and learn their atomic weights and properties without ever being able to handle them in a terrestrial laboratory.

In a series of remarkable papers communicated to the Royal Astronomical Society during the past year (1911-1912), Dr. J. W. Nicholson has given the result of his computation of the positions of the spectral lines of two elements of simple structure, and has found that the resulting lines correspond, for one dynamical system, to the chief unidentified lines observed in the spectra of nebulæ, and for the other, to the chief unidentified lines in the spectrum of the corona. The latter element is probably associated with the halogens, but of much lower atomic weight (namely, 1·3), than fluorine; he therefore gives it the name of Protofluorine. The other element, to which he gives the name Nebulium, will have an atomic weight of 2·1. Prof. Max Wolf, of Heidelberg, has recently pointed out[11] the evidence of the presence of two other unknown gases in the Ring nebula in Lyra, and there is no reason to suppose that the process of discovery has come to an end. But we cannot imagine that we shall discover any new elements that are more abundant and more universally diffused than the five which give us protoplasm—“the physical basis of life.” To take an analogy from the solar system: many