THE ether is usually defined as an imponderable elastic fluid, permeating all bodies and all space. But it must have weight, or mass, if it is matter. Lord Kelvin has computed a minimal mass of 10—grammes per cubic meter. Ether cannot be a mixture of ordinary gases, for these do not penetrate all substances, and they act differently upon those they do penetrate, whereas ether is everywhere the same. Many learned men suggest or express belief in ether as the primordial matter of which atoms are formed, and in which they float just as stars and planets co-exist with un- agglomerated cosmical dust. Some think that atoms are continually being formed and disintegrating, others that they were created once for all and that the ether is a residue or by-product of their formation. With the latter hypothesis, which rests on pure assumption, realists have nothing to do. The former involves the possibility—long accepted by the great mass of mankind—of creating new atoms and annihilating matter. Emmens has claimed that he can make gold from silver, Fittica that he can convert phosphorus into arsenic. Many such transformations were described half a century ago, but every one was based on errors due to inadvertence or prejudice. If we had to do merely with the ether that fills the interplanetary space and conveys energy through it, we could confine our attention to the mass and neglect the chemical nature of the ether. But so negative and bloodless an ether becomes unsatisfactory when we descend from heaven to earth, for the ether must permeate all bodies. This power of penetration may be regarded as the highest development of the power of diffusion, shown by many gases with respect to caoutchouc, and by hydrogen with respect to iron, palladium, and platinum. In the latter case the diffusion is due not only to the lightness and high velocity of hydrogen molecules, but also to a chemical action analogous to solution and the formation of alloys, in which the compounds formed are ether, indefinite or unstable and' readily dissociated by elevation of temperature. But the power of ether to form true compounds must be absolutely nil; in permeating othsr substances the only change it can undergo is a certain condensation. Ten years ago the existence of so inert a substance appeared improbable, but now we know five such, argon, helium, neon, krypton, and xenon, the gases discovered by Ramsay and his associates; which dissolve freely in water, but so far as is known, form no definite compounds with anything. They afford an experimental basis for the conception of ether as a gas incapable of combination. We need not, like Crookes, assume a fourth state of aggregation, and thus we avoid all mysticism. We have assumed nothing inconsistent with the current conception of ether. In 1869, when I pointed out the periodic connection between the properties of the elements and their atomic weights, the existence of absolutely inert elements was not suspected. Therefore the system began with group I and series I, or with hydrogen, the lightest known element, common to both, but I never thought that it must begin with hydrogen. My predictions of the existence and properties of unknown elements were confirmed by the discoveries of gallium, scandium, and germanium. These predictions are examples of what mathematicians call interpolation. The prediction of ether, as a rare gas, is an example of extrapolation which I venture now to attempt, because I have little time to wait, and because the new theory that atoms are composed of much smaller electrons appears to me to have sprung from the want of a definite conception of the ether, an outrush of which will suffice to explain the apparent disintegration of atoms into electrons. The periodic system may be illustrated by an example. If we arrange in two series those elements whose atomic weights lie between 7 and 35.5, thus: Li 7, Be 9, B 11, C 12, N 14, 0 16, F 19, Na 23, Mg 24, Al 27, Si 28, P 31, S 32, CI 35.5, we see that each closely resembles the one below or above it and that the af- flnity for oxygen increases regularly as we go from left to right, the highest oxides of the second series being Na.O, MgjOj^ ALOj, SLO. P.O., S.Oo, Cl,OT. Hence the vertical groups are designated by the Roman numerals I.'to VII. But as the new gases form no compounds, they must be put in a zero group, and the atomic weight of each must fall between those of an element of group VII. of one series and an element of group I. of the following series. This a priori conclusion is fully confirmed by experiment as appears from the following table, which is extended to include group zero and series zero and two hypothetical elements, x and y. The latter, y, must have the fundamental properties of the argon group. Its atomic weight—deduced from ' the variation in successive neighboring groups and series—is probably less than 0.4. This element is probably coronium, whose spectrum, resembling that of helium in simplicity, appears in the solar corona above the hydrogen spectrum millions of miles from the sun's center, a fact which indicates its small density and atomic weight. If it is monatomic, like the helium group, its density is half its atomic weight, or less than 0.2, and its molecular velocity more than 214 times that of hydrogen, so that it may escape from the earth's sphere of influence, though it cannot escape from the sun's, and so cannot be the all-pervading ether. It may aid us, however, to the conception of the lightest and most mobile of elements, which I believe to be the universal ether and for which I suggest the name “Newtonium." I cannot conceive the other elements to be formed of this, and I see no Simplification in a common origin of elements. Unity of a higher order is given by the conception of ether as the final link in the chain of elements. The molecular velocity of a gas may be calculated from the kinetic theory. For hydrogen at 0 deg. C. it, is 1,843 meters per second. For any gas at temperaturo mates of the temperature of space lie bet ween —60 deg. C. and —100 deg. C. Taking the mean, —80 deg. C., we have, 2,191 4,800,000 '0-= , or IV = where v is the molecular velocity in meter.'i per second. PF.RIOIIC '1'ABLK OK KbKMKNTS. I u ® 2 s o t—i o. 3 0 0. o c 0. o 0. s X Cl Cl Cl Cl Cl Cl Cl Cl a 0 X 1 y H 1.M8 2 Hc Li 7.03 Be 9.1 II 11 C 12 N 14 ()4 0 16 F 19 a Ne 19.9 Na 2&0.r> Mit” 24.1 27 Si 28.4 P 31 S ('I 35.41 4 Ar i» K 39.1 Cit 40.1 H.l 'l'1 48.1 V 51.4 <> 62.1 Mil or. Fe('<> Ni (Cu) 59 611 5 Cu 63.6 Zr 65.4 Gil 7(1 fic As 7. 79 Br 79.9. 6 Kr 8l.S Kl> 83 Si- 87.fi V 89 Z> 94 Mo 96 — ltu HliPil (Aif) 101.7 100 llltS.5 7 AK 1W.9 Ccl 112.4 111 114 Pll 119 Sb im TV 187 1 12; 8 9 10 Xe 128 Vs 1.32.8 Ha La 139 Cc 140 - - - - - - (-) - - Yb 173 - Til isa W 184 Os 1r It. (All) WI 193 194.11 11 Au 1W.2 Hg ro Tl 004 I Pb 2Od.9 Hi 208 12 - - Kil 224 - Th 232 - xr 239 This velocity must be great enough to overcome the attraction of the heavenly bodies. A projectile thrown with sufficient velocity will liot return to the earth, the limiting velocity being that acquired in falling to earth from infinity, equal to /2m t it is 1,843 V'- 1 + <r t d where (l is density referred to hydrogen and a is the coefficient of expansion 0.00367. For ether, assumed to be monatomic, this density would be half the atomic weight, CIT 1/:x, Most est!- where r is the earth's radius and m the earth's mass in gravitation or astronomical measure, determined by the m equation g = —. Hence v = 11,190 meters per second. r- The atomic weight corresponding to this molecular velocity is 0.038. Gases heavier than this would remain attached to the earth; lighter ones would escape. To resist the sun's attraction the molecular velocity must be 608,300 meters per second, the atomic weight 0.000013. But ether atoms must be still lighter and swifter to escape the attraction of still larger suns. The masses of some binary stars have been computed from their rotations. The heaviest has 33 times the mass of the sun. There is spectroscopic and other evidence that their densities do not differ greatly from his. Hence, knowing the mass, we can calculate the radius and, therefore, the molecular velocity which would overcome the star's attraction. For a star 50 times heavier than the sun, it would be 608,300 X = 2,240,000 meters per second. This is about 1-130 of the velocity of light, 300,000,000 meters per second, and we may assume the molecular velocity of ether to lie between these limits. The corresponding limits for its atomic weight are 0.0000009G 'and 0.000000000053. In the present state of science it seems impossible to accept the latter value, which would suggest a return to the emission theory of light. I think that many phenomena may be explained by assuming that the x atoms have about a millionth of the mass of hydrogen atoms and a mean velocity of nearly two and a quarter millions of meters per second. While I was making these calculations, I received Prof. Dewar's Belfast address, in which he expresses the opinion that the highest strata of the atmosphere, the region of aurorre, is also the field of hydrogen and the argon gases! It is but a step from this to the assumption of a still lighter gas filling all space and giving a tangible reality to the conception of the ether. Without developing the theory further I turn to certain apparently irrelevant phenomena which have guided my speculations, led me to publish this essay, and induced others to revert to the emission theory or to adopt that of electrons, scarcely conceivable to me, without thereby clarifying our conception of the ether. I refer especially to radio-activity. From the first my impression has been that here we have to do with a condition which is no more peculiar to uranium, thorium, and radium than magnetism is to iron, cobalt, and nickel. These heaviest of all atoms (V 239, Th 232, Rd 224) may be regarded as suns possessing the highest development of that special attractive power which is intermediate between gravitation and chernism, and which is the cause of gas absorption, solution and the like. We must not assume that because ether, like argon, forms no staple compounds, it may not dissolves in accumulate about great centers of attraction such as stars and suns, uranium and thorium atoms. Thoughsuch accumulation might involve a change in velocity, it would be a comet-like rather than a planet-like connection, and it would be most likely to occur with the heavy uranium and thorium (and radium) atoms. Such an ether swarm about the uranium atom would explain various phenomena. I believe that radioactivity indicates a material emanation and that the arrival and departure of ether atoms are accompanied by the disturbances which constitute waves of light. When a flask containing gelatinous zinc sulphide is connected by a tube with a flask of radio-active solution, the sulphide glows as long as the connection is maintained, but the phosphorescence gradually dies away when the connecting tube is closed, and may be renewed by reopening the stop-cock. This experiment, which M. and Mme. Curie performed in my presence, becomes explicable if we assume that a tenuous ethereal gas enters and leaves the radioactive substance as comets enter and leave the solar system. Transverse light waves may be provoked either by molecular motion of other bodies, as in incandescence, or by variation in the motion of ether atoms themselves, that is, by a disturbance of their mobile equi- li brium. A possible cause of such disturbance is the great mass of the uranium atom, as the chief cause of the sun's luminosity is, in my opinion, its great mass and the accumulation of ether due to its attraction. I think that light waves are far more complex than is generally believed, because of the great velocity of the ether atoms. Dewar has observed that the phosphorescence of parafin and other substances is greatly increased by cooling to —193 deg. C. It appears to me that at very low temperatures there is a condensation or an increased absorption of ether in these substances and that the increased phosphorescence iii due to motion of ether atoms. This essay is merely a series of impressions, suggested, however, by actual phenomena. Probably others have had similar ideas but have not developed them. If there is any truth in my theory, it will be elaborated and confirmed; if it is wholly false, its refutation will warn others. I have attempted to give the first approximate answer to the question: What is the chemical nature of universal ether? Or, rather, to bring the question before the parliament of science.