C H E M I S T R Y samarium is white ; its salts are yellow, and are characterized by definite absorption spectra. Gadolinium.—This element was discovered in 1880 by Marignac among the samarskite earths; it was subsequently studied by de Boisbaudran, Cleve and Bettendorf. Its oxide and salts resemble in appearance those of ytterbium, but it has a much lower atomic weight than that element. Praseodymium and Neodymium.—Didymia—one of the most fully studied of the rare earths—which yields salts either rose-red or violet in colour, was always regarded as a definite substance until Auer von Welsbach, in 1885, by an extended and most laborious series of fractional crystallizations of the mixed nitrates of ammonium, lanthanum and didymium, succeeded in resolving it into two constituents, praseodymia and neodymia. The former gives green salts and solutions, the latter amethyst-coloured salts and rose-coloured solutions. The elements differ in atomic weight by nearly three units, but while the contrast between their physical properties is very marked, they differ chemically to an extraordinarily slight extent. Auer’s observations have been abundantly confirmed, except that, curiously enough, it has been shown by both Brauner and Jones that in assigning atomic weights to the two elements he interchanged the values. But opinion is divided as to the individuality of the products described by Auer. Oleve, while accepting praseodymium as a new element, doubts the individuality of neodymium (cf. Muthman and Stritzel, Ber.deut. chem. Ges. 1899, 32, p. 653). Crookes goes farther, and points out that when the absorption spectra of the two substances are superposed, a spectrum is obtained nearly identical with that of the original didymia, except that two bands are missing. From this circumstance it might be presumed that a third substance is present, but his own work on the fractionation of didymia leads him to conclude that it must not be regarded as compounded of only two, but rather as an aggregation of many closely allied elements. Kriiss and Nilson have arrived at a similar conclusion. Victorium.—In discriminating the elements of the yttrium-cerium group, the spectroscope plays an all important part, several of them being characterized by their absorption spectra, whilst others, although they exert no absorbent action in solution, afford definite spark-spectra. A third method introduced by Crookes consists in submitting the oxides, or preferably the basic sulphates, to the influence of the negative electric discharge in vacuo, and viewing the phosphorescence thus produced with the spectroscope. By a refinement of this method—by photographing the spectrum of the phosphorescent light, using lenses and prisms of quartz—he has been able to study the invisible ultra-violet spectra afforded by the rare earths, and in consequence to detect and separate a new element —victorium—from yttria. Starting from crude yttria, separated in the ordinary way from metals of the cerium subgroup, and subjecting the nitrate prepared from this, time after time, to the well-known process of alternate heating and then boiling the partially decomposed salt with water, thereby separating each time an insoluble basic nitrate from a soluble nitrate, he obtained fractions giving different spectra—a group of lines high up in the ultra-violet attaining a maximum brilliancy in certain intermediate fractions. By fractionally recrystallizing the oxalates prepared from these intermediate fractions a great number of times, further concentration was effected of the substance affording the spectrum referred to. Resort was then again had to the nitrate method; and finally fractions selected with the aid of the spectroscope as likely to be the richest were converted into sulphates, and fractionally precipitated by potassium sulphate. The substance
709
ultimately isolated in this manner contained the new element, victorium. Cleve’s list of “ not yet thoroughly characterized ” elements of the yttrium-cerium group contains holmium, thulium,—these two he himself discovered in ordinary erbia, but has not yet satisfactorily separated,—terbium and dysprosium, and in addition a dozen or more not yet dignified with names the existence of which is suspected by Crookes or de Boisbaudran in various earths (cf. Demargay, G. R. 1900, 130, p. 1185; 131, p. 387). The inchoate state of knowledge of the rare earth elements will be sufficiently inferred from the foregoing statements; it may safely be said that no single element of the group is known in a state of practical purity, or can be regarded as “ thoroughly characterized.” With a view of redetermining the densities of the elementary gases, and thus obtaining data for the further discussion of Prout’s hypothesis that the atomic weights of the elements generally stand in Gases of simple relation to that of hydrogen, Lord Rayleigh was led, in the course of experiments commenced in 1882, to redetermine the density of nitrogen. Operating with the gas prepared by various chemical interactions, as well as with that separated from atmospheric air, he was led, in 1894, to discover that the nitrogen from air was very distinctly denser than that prepared from any nitrogen compound, a litre of the former weighing under standard conditions L2572, and one of the latter only 1-2505 gram. Having considered and tested the various possible explanations of the anomaly, Lord Rayleigh was at last led to ask what was the evidence on which the prevailing view was founded that the inert residue from air is a single substance. It then appeared that this was based entirely on the experiment made as far back as 1785 by the celebrated Cavendish, in which what was then called “phlogisticated air,nitrogen, was combined with oxygen by passing sparks from a frictional electric machine through a mixture of the two gases confined over potash, and so converted into nitre. On repeating the experiment Lord Rayleigh obtained results in accordance with those described by Cavendish, a small residue of uncombined gas being always left; and as he found that this residue was in proportion to the amount of air operated upon, and that its spectrum was not that of nitrogen, the initial discovery of a previously unknown constituent of our atmosphere was made. Meanwhile Ramsay had also succeeded in obtaining a gas differing from Ar nitrogen, by passing air deprived of oxygen, son. etc., over heated magnesium—a well-known absorbent of nitrogen. Pursuing the investigation conjointly, Rayleigh and Ramsay ultimately proved that the superior density of atmospheric nitrogen is due to the admixture of a small proportion (about 1 per cent.) of a gas which they have termed argon (<j.v.)—on account of its complete chemical Indifference—of superior density to nitrogen in the ratio 14-003 :19-9 (cf. Phil. Trans. 1895, A. p. 187). At the time the discovery of argon was made public, Ramsay’s attention was called by Miers to an observation recorded by an American chemist, Hillebrand, that on boiling the mineral uraninite with acid a gas was given off which appeared to be nitrogen. On examining cleveite, an allied uranium mineral, Ramsay obtained a gas almost free from nitrogen, and having observed that its spectrum contained a yellow line close to, but not coincident with, the D lines characteristic of sodium, he submitted it to Crookes for examination. On mapping the spectrum, Crookes found that the line referred to was the line D3 first noticed by Lockyer in 1878 in the spectrum of the solar chromosphere, and ascribed by Frankland and Lockyer to a hypothetical element helium. Helium very
