I should like to express my thanks to the Ford Foundation, which provided financial support for the research on this topic. A debt of gratitude is owed to Professor Aaron Ihde for his helpful suggestions during the study. Several recommendations by an anonymous referee were also of assistance in the final formulation of this paper.

¹ Bunsen, R. and Kirchhoff, G., ‘Ueber ein neues Alkalimetall’, Journal für praktische Chemie, lxxx (1860), 477–80.Google Scholar

² Bunsen, R., ‘Ueber ein fünftes der Alkaligruppe angehörendes Element’, Journal für praktische Chemie, lxxxiii (1861), 198–200.Google Scholar

³ Crookes, W., ‘On the existence of a new element, probably of the sulphur group’, Chemical news, iii (30 03 1861), 193–5.Google Scholar

⁴ Reich, F. and Richter, H. T., ‘Vorläufige Notiz über ein neues Metall’, Journal für praktische Chemie, lxxxix (1863), 441–2.CrossRefGoogle Scholar

⁵ Boisbaudran, P. Lecoq de, ‘Caractères chimiques et spectroscopiques d'un nouveau métal, le Gallium, découvert dans une blende de la mine de Pierrefitte, vallée d'Argelès (Pyrénées)’, Comptes rendus hebdomadaires des séances de l'Académie des Sciences, lxxxi (1875), 493–5Google Scholar; cited hereafter as Comptes rendus.

⁶ Meadows, A. J., Science and controversy. A biography of Sir Normon Lockyer (London, 1972), pp. 54–60, 194–8.Google Scholar See also Weeks, Mary E., Discovery of the elements, revised by Leicester, Henry (7th ed., Easton, Pa., 1968), pp. 757–8.Google Scholar

⁷ In 1895 Ramsay discovered an inert gas in the rare-earth mineral cleveite; see Ramsay, W., ‘Discovery of helium’, Chemical news. lxxi (29 03 1895), 151.Google Scholar Crookes then confirmed the spectrum of the gas to be that of Lockyer's helium: Crookes, W., ‘The spectrum of the gas from clèveite’, Chemical news, lxxi (29 03 1895), 151.Google Scholar

⁸ See Dennis, L. and Dales, B., ‘Contributions to the chemistry of the rare earths of the yttrium group—part I (historical)’, Chemical news, lxxxv (30 05 1902), 256–8; (6 06 1902), 265–6.Google Scholar

⁹ Lockyer employed the generally accepted notion that white stars (whose chromospheres contain great amounts of hydrogen) are hotter than red stars (whose spectra indicate a lack of hydrogen and a predominance of heavy elements); see McGucken, W., Nineteenth-century spectroscopy: development of the understanding of spectra, 1802–1837 (Baltimore, 1969), pp. 76–7.Google Scholar Lockyer's ideas concerning the complexity of the elements and their reception by chemists are examined in Brock, W. H., ‘Lockyer and the chemists’, Ambix, xvi (1969), 81–99.CrossRefGoogle Scholar

¹⁰ Boisbaudran, P. Lecoq de, ‘Théorie des spectres; observations sur la dernière communication de M. Lockyer’, Comptes rendus, lxxxii (1876), 1264–6.Google Scholar

¹¹ Ciamician, G., ‘Über die spectren der chemischen Elemente und ihrer Verbindungen’, Sitzungsberichte der mathematischnaturwissenschaftlichen Classe der Kaiserlichen Akademie der Wissenschaften, Wien, lxxvi (1877)Google Scholar, Abt. 2, 499–517. ‘Spectroskopische Untersuchungen’, Sitzungsberichte … Wien, lxxix (1879)Google Scholar, Abt. 2, 8–10. See also McGucken, , op. cit. (9), pp. 1091–10.Google Scholar

¹² Liveing, G., Report of the British Association for the Advancement of Science, 1882 (London, 1883), p. 483.Google Scholar Presidential address to the Chemical Section delivered on 24 August 1882.

¹³ Prout had suggested in 1816 that the ultimate constituent of all substances might be hydrogen; see Prout, W., ‘Correction of a mistake in the essay on the relation between the specific gravities of bodies in their gaseous state and the weights of their atoms’, Annals of philosophy, vii (1816), 111–13.Google Scholar For a good discussion of the use and modification of Prout's hypothesis throughout the nineteenth century, see Farrar, W. V., ‘Nineteenth-century speculations on the complexity of the chemical elements’, The British journal for the history of science, ii (1965), 297–323.CrossRefGoogle Scholar

¹⁴ Dumas, B. A., ‘Mémoire sur les équivalents des corps simples’, Comptes rendus, xlv (1857), 720Google Scholar, and ‘Note sur les équivalents des corps simples’, Comptes rendus, xlvi (1858), 951.Google Scholar

¹⁵ Kopp, H., ‘Ueber die specifische Wärme starrer Körper, und Folgerungen bezüglich der Zusammengesetztheit s.g. chemischer Elemente’, Annalen der Chemie und Pharmacie, cxxvi (1863), 362–72.CrossRefGoogle Scholar

¹⁶ See McGucken, , op. cit. (9), pp. 83–95.Google Scholar

¹⁷ For a detailed biographical study of Crookes one must still consult Fournier D'Albe, , William Crookes (London, 1923).Google Scholar Also helpful is Brock, W. H., ‘William Crookes’, Dictionary of scientific biography (New York: Charles Scribner's Sons (1970–in progress), iii (1971), 474–82.Google Scholar

¹⁸ D'Albe, , op. cit. (17), pp. 26–7.Google Scholar

¹⁹ Crookes, W., ‘On attraction and repulsion resulting from radiation’, Proceedings of the Royal Society, xxiii (1875), 373–8.Google Scholar

²⁰ Crookes, W.. ‘On discontinuous phosphorescent spectra in high vacua’, Proceedings of the Royal Society, xxxii (1881), 206–13.CrossRefGoogle Scholar

²¹ Crookes, W., ‘On radiant matter spectroscopy. A new method of spectrum analysis’, Proceedings of the Royal Society, xxxv (1883), 262–71.CrossRefGoogle Scholar Bakerian Lecture delivered on 31 May 1883.

²² Crookes found that compounds of rare-earth elements—particularly the sulphates—emitted far more pronounced phosphorescent spectra than the pure elements.

²³ Crookes, W., ‘On the fractionation of yttria’, Report of the British Association for the Advancement of Science, 1886 (London, 1887), p. 588.Google Scholar

²⁴ Crookes, W., ‘On radiant matter spectroscopy. Part II. Samarium’, Proceedings of the Royal Society, xxxviii (1885), 414–22.Google Scholar

²⁵ Marignac, J., ‘Sur les terres de la samarskite’, Comptes rendus, xc (1880) 899–903.Google Scholar

²⁶ Crookes, W., ‘On radiant matter spectroscopy: examination of the residual glow’, Proceedings of the Royal Society, xlii (1887), 112.Google Scholar

²⁷ Ibid., 112.

²⁸ Ibid., 115.

²⁹ Ibid., 126.

³⁰ Crookes obtained no meaningful differences between the atomic weights of various yttrium fractions, but he still retained his confidence that different ‘constituents’ of yttrium existed. He perceived quite clearly that the atomic weight obtained from any fraction is an average of the individual weights of the atoms of a particular chemical species in that fraction (that will be seen below). Consequently, small amounts of ‘constituents’ that differ only slightly in atomic weight but emit markedly different spectra could conceivably be present; Crookes was to embrace just this conception to explain away the apparent identity of atomic weights among yttrium fractions.

³¹ Crookes, W., ‘What is yttria?’, Chemical news, liv (23 07 1886), 39.Google Scholar

³² Crookes, W., ‘Genesis of the elements’, Proceedings of the Royal Institution, xii (1889), 45.Google Scholar Lecture delivered before the Royal Institution on 18 February 1887.

³³ Boisbaudran, P. Lecoq de, ‘Sur un nouveau genre de spectres métalliques’, Comptes rendus, c (1885), 1437–40.Google Scholar

³⁴ Crookes, W., ‘On some new elements in gadolinite and samarskite detected spectroscopically’, Proceedings of the Royal Society, xl (1886), 507.Google Scholar

³⁵ Crookes, op. cit. (31), 39. Crookes would repeat this in almost identical words in his address before the Chemical Society on 21 March 1889; see Journal of the Chemical Society, lv (1889), 280.Google Scholar Significantly, in his 1889 statement Crookes added the contention that the lines he had observed by the phosphorescent method back in 1886 and the lines Lecoq had observed with his reversion method ‘do not even in all cases agree … in position’ (my italics). Thus, Crookes continued, ‘Though so accurate an observer, M. de Boisbaudran concluded apparently too hastily that two spectra are identical …’ (p. 280). However, Crookes did not assert that no lines were identical in the two spectra. This 1889 address to the Chemical Society is reprinted in Knight, D. M. (ed.), Classical scientific papers. Chemistry, Series 2 (New York, 1970), pp. 414–27.Google Scholar

³⁶ Boisbaudran, P. Lecoq de, ‘Sur la fluorescence anciennement attribuée à l'yttria’, Comptes rendus, cit (1886), 1536–9.Google Scholar

³⁷ Crookes, , op. cit. (31), 39.Google Scholar

³⁸ Cleve, P., ‘The life work of Marignac’, Journal of the Chemical Society, lxvii (1895), 474–5.Google Scholar

³⁹ Demarçay, E., ‘Les terres rares’, Revue généraled es sciences pares et appliquées, i (1890), 396.Google Scholar

⁴⁰ Though Lecoq's assertion that Crookes's phosphorescent spectroscopy had not detected new elements would eventually be verified by Urbain (see below in this paper), the Frenchman's claims based on reversion spectroscopy were to fare no better. He himself admitted in 1889 that Z β was very possibly identical to terbium, which was later confirmed; see Boisbaudran, P. Lecoq de, ‘Sur le gadolinium de M. de Marignac’, Comptes rendus, cviii (1889), 167.Google Scholar Urbain showed much later that Za was dysprosium; see Urbain, G., ‘Spectres de phosphorescence cathodique du terbium et du dysprosium dilués dans la chaux’, Comptes rtndus, cxliii (1906), 231.Google Scholar Ironically, it had been Lecoq who first isolated dysprosium in 1886 by separating it off from holmium; see Boisbaudran, P. Lecoq de, ‘L'holmine (ou terre X de M. Soret) contient au moins deux radicaux métalliques’, Comptes rendus, cii (1886), 1003–4.Google Scholar

⁴¹ Crookes, , op. cit. (sa), 49–50.Google Scholar

⁴² Gladstone, J. H., Report of the British Association for the Advancement of Science, 1883 (London, 1884), p. 453.Google Scholar Presidential address to the Chemical Section delivered on 20 September 1883.

⁴³ Crookes, W., Report of the British Association for the Advancement of Science, 1886 (London, 1887), p. 558.Google Scholar Presidential address to the Chemical Section delivered on 2 September 1886; reprinted in Knight, , op. cit. (35), Series 1 (1968), pp. 334–52.Google Scholar

⁴⁴ Crookes, , op. cit. (43), 559.Google Scholar

⁴⁵ Ibid., 560.

⁴⁶ Ibid., 561.

⁴⁷ Ibid., 561.

⁴⁸ Ibid., 563. The fact that helium's spectrum contains more than one line was not discovered until 1895; emissions in the red, blue-green, blue, and violet regions of its spectrum were observed. That the yellow line originally discovered in 1868 is really a doublet was also revealed in 1895. See Ramsay, W., Collie, J. Norman, and Travers, M., ‘Helium, a constituent of certain minerals’, Journal of the Chemical Society, lxvii (1895), 687, 698–9.Google Scholar

⁴⁹ Crookes, , op. cit. (43), 564.Google Scholar

⁵⁰ Reynolds, J. Emerson, ‘Note on a method of illustrating the periodic law’, Chemical news, liv (2 07 1886), 1–4Google Scholar; also reprinted in Knight, op. cit. (35), Series 2, pp. 317–20.

⁵¹ Crookes, , op. cit. (43), 566.Google Scholar

⁵² Ibid., 566.

⁵³ Ibid., 568.

⁵⁴ Ibid., 569.

⁵⁵ Ibid., 569.

⁵⁶ Ibid., 571.

⁵⁷ Ibid., 572.

⁵⁸ Ibid., 572.

⁵⁹ Ibid., 576.

⁶⁰ Crookes, , op. cit. (32), 55.Google Scholar

⁶¹ Crookes, W., Journal of Chemical Society, liii (1888), 490.Google Scholar Presidential address to the British Chemical Society delivered on 28 March 1888.

⁶² Ibid.

⁶³ Ibid.

⁶⁴ Ibid.

⁶⁵ Ibid., 502.

⁶⁶ Demarçay, , op. cit. (39), 398.Google Scholar

⁶⁷ Marignac, J., ‘Quelques réflexions sur le groupe des terres rares, à propos de la théorie de M. Crookes sur la genèse des éléments’, Archives des sciences physiques et naturelles, Genève, xvii (1887), 375–6.Google Scholar

⁶⁸ Ibid., 376.

⁶⁹ Crookes, W., ‘Electricity in transitu: from plenum to vacuum’, Chemical news, lxiii (6 03 1891), 114.Google Scholar Presidential address delivered before the Institution of Electrical Engineers on 15 January 1891.

⁷⁰ Crookes, W., Chemical news, lxxviii (9 09 1898), 134.Google Scholar Presidential address delivered before the British Association for the Advancement of Science on 7 September 1898.

⁷¹ Crookes, W., ‘Photographic researches on phosphorescent spectra: on victorium, a new element associated with yttrium’, Chemical news, lxxx (4 08 1899), 49.Google Scholar

⁷² Crookes, , op. cit. (70), 134.Google Scholar

⁷³ Boisbaudran, P. Lecoq de, ‘Recherches sur le samarium’, Comptes rendus, cxvi (1893), 611–13, 674–7.Google Scholar

⁷⁴ Demarçay, E., ‘Sur la simplicité du samarium’, Comptes rendus, cxvii (1893), 163–4.Google Scholar

⁷⁵ Demarçay, E., ‘Sur les spectres du didyme et du samarium’, Comptes rendus, cii (1886), 1551–2.Google Scholar

⁷⁶ Demarçay, E., ‘Sur un nouvel élément, l'europium’, Comptes rendus, cxxxii (1901), 1484–5.Google Scholar

⁷⁷ Weeks, , op. cit. (6), 818–20.Google Scholar It might be noted here that Urbain's recognition of the inability of chemists to define satisfactorily the chemical element in operational terms had led him to propose a new quantitative technique for such designation. In 1910, stating that atomic weights could not be determined to an accuracy which could unambiguously resolve the controversies surrounding the rare-earths, he offered the possibility that the measurement of magnetic susceptibilities of elements in the rare-earth series could be employed to indicate efficacy of fractional separation; see Urbain, G., ‘Sur l'analyse magnéto-chimique des terres rares’, Comptes rendus, cl (1910), 913–15.Google Scholar

⁷⁸ Urbain, G., ‘Sur les terres yttriques provenant des sables monazités’, Comptes rendus, cxxvii (1898), 107–8.Google Scholar

⁷⁹ Urbain, G. and Lacombe, H., ‘Sur l'emploi du bismuth comme agent de séparation dans la série des terres rares’, Comptes rendus, cxxxviii (1904), 84–5.Google Scholar

⁸⁰ Urbain, G., ‘Europium, gadolinium, terbium, neoytterbium, and lutecium’, Chemical news, c (13 08 1909), 73–5.Google Scholar

⁸¹ Urbain, G., ‘Sur la loi de l'optimum des phosphorescences cathodiques des systèmes binaires’, Comptes rendus, cxlvii (1908), 1472–4.Google Scholar

⁸² Urbain, , op. cit. (80), 75.Google Scholar

⁸³ Few modern attempts to write the history of the rare-earths have been undertaken. Weeks, op. cit. (6), 667–99 is the most exhaustive treatment of this topic. See also Davis, Helen Miles, The chemical elements (Washington, D.C., 1959)Google Scholar; Hopkins, B. S., Chapters in the chemistry of the less familiar elements (Champaigne, Illinois, 1939)Google Scholar; Ihde, Aaron, The development of modern chemistry (New York, 1964), pp. 374–9.Google Scholar

⁸⁴ It is interesting to note, however, that in 1914 Grookes would imply Soddy's isotope concept confirmed ideas on the elements which he had been promulgating for the past thirty years; see Crookes, W., Chemical news, ex (11 12 1914), 290Google Scholar (his presidential address to the Royal Society delivered on 30 November 1914).

⁸⁵ For example, from Crookes's 1888 presidential address to the British Chemical Society, op. cit. (61), 491–2; The atomic weight which we ascribe to yttrium … merely represents a mean value around which the actual weights of the individual atoms of the “element” range within certain limits. But if my conjecture is tenable, could we separate atom from atom, we should find them varying within narrow limits on each side of the mean. ‘The very process of fractionation implies the existence of such differences in certain bodies … We may picture to ourselves some directive force passing the atoms one by one in review, selecting one for precipitation and another for solution, till all have been adjusted. In order that such a selection can be effected there evidently must be some slight differences between which it is possible to select, and this difference almost certainly must be one of basicity … The meaning of ‘slight differences’ among atoms passes easily from differences in atomic weight to differences in basicity in the course of four consecutive paragraphs in this address.