Quantum Physics and the Stars (III): Henry Norris Russell and the Search for a Rational Theory of Stellar Spectra

DeVorkin David H. and Kenat Ralph , “Quantum physics and the stars (I): The establishment of a stellar temperature scale”, Journal for the history of astronomy, xiv (1983), 102–32; “Quantum physics and the stars (II): Henry Norris Russell and the abundances of the elements in the Sun and stars”, Journal for the history of astronomy, xiv (1983), 180–222. Hereinafter referred to as Paper I and Paper II respectively.

The best early example of astronomical observations in need of physical theory is the question of why stars exhibit different spectra. See Meadows A. J. , Science and controversy: A biography of Sir Norman Lockyer (Cambridge, Mass., 1972). Possibly the most important example for the twentieth century has been the source of energy of the Sun and stars. See Hufbauer Karl , “Astronomers take up the stellar-energy problem, 1917–1920”, Historical studies in the physical sciences, xi (1981), 277–303.

See Meadows , op. cit. (ref. 2); and Brock W. H. (ed.), The atomic debates (Leicester, 1967). Bohr's model revealed that the process of ionization was a loss of one or more electrons in a nuclear atom. With it, Saha could explain why different stars displayed different spectra in terms of an atomic theory that had gained wide acceptance. The idea of ionization was also applied by Eddington, with the encouragement of Jeans and others, to attempt to explain the peculiar state of material in the interiors of stars.

The development of quantum theory between 1920 and 1925, resulting in Heisenberg's matrix mechanics, has been treated extensively. All of the comments about the state of quantum theory have been derived from this body of secondary literature. On the state of atomic physics in the early 1920s see Forman Paul , “The doublet riddle and atomic physics circa 1924”, Isis, lix (1968), 156–74; and Cassidy David , “Heisenberg's first core model of the atom: The formation of a professional style”, Historical studies in the physical sciences, x (1979), 187–224. For a general introduction to the “Old quantum theory” see Jammer Max , The conceptual development of quantum mechanics (New York, 1966); Hermann Armin , The genesis of quantum theory (1899–1913) (Cambridge, Mass., 1971).

Russell Norris Henry , “Rubidium in the Sun”, Publications of the Astronomical Society of the Pacific, xxxiii (1921), 202–4.

Russell Norris Henry , “The theory of ionization and the sun-spot spectrum”, Astrophysical journal, lv (1922), 119–44.

Paper II.

Russell to Saha, 3 Aug. 1921 (Russell papers).

Compton K. T. , “Remarks on ionization by cumulative action”, Philosophical magazine, xliii (1922), 531–7.

Russell Norris Henry , “Notes on ionization in the solar atmosphere”, Astrophysical journal, lv (1922), 354–9.

Ibid., 356–7.

Ibid., 357.

This belief was later shown to be erroneous, in part through Russell's own research. See Moore Charlotte E. and Russell H. N. , “On the winged lines in the solar spectrum”, Astrophysical journal, lxiii (1926), 1–12. See also Paper II.

Russell , op. cit. (ref. 8).

Saha M. N. , “On the temperature ionization of elements of the higher groups in the periodic classification”, Philosophical magazine, xliv (1922), 1128–39.

In 1898 Boltzmann had argued that the attractive forces between atoms could not be simple functions of the distance between atoms but that “on the contrary it must be associated with a relatively small region on the surface of the atom. Moreover, it is only with the latter assumption and not the former, that one can obtain a picture of gas dissociation corresponding to reality.” See Boltzmann Ludwig , Lectures on gas theory, trans. by Brush Stephen (Berkeley, 1964), 376.

Saha , op. cit. (ref. 15), 1132.

Saha also realized how to incorporate the steric factor into his original theory. Otto Stern (“Zur Theorie der Gasdissoziation”, Annalen der Physik, xliv (1914), 497–529) had recently applied Boltzmann's original idea to the particular case of iodine to provide a complete thermodynamic and mechanical treatment of Boltzmann's model. In so doing, he derived a generalized formula for the entropy of the system which included a factor for the necessary solid angle of approach. Saha was able to simplify and adapt this formula for the entropy and use it in the rederivation of his original equation; thus he created a new equation with an additional term involving the steric factor. Saha could give no a priori values for the steric factor but he could determine values based on the observed facts concerning the rates of ionization. A value of the steric factor of n = 8 could account for the behaviour of barium.

Saha cited Bohr Niels , “Der Bau der Atome und die physikalischen und chemischen Eigenshaften der Elemente”, Zeitschrift für Physik, ix (1922), 1–67. The section on helium begins on p. 33. In 1931 Saha returned to the barium problem in the last chapter of a thermodynamics text written with B. N. Srivastava. He rederived his original formula using the corrections that had been made in the interim by E. A. Milne, R. H. Fowler and Russell, and then added new concepts derived from quantum theory. His new generalized formula accounted for dependence of the entropy of the system upon the configuration of the valence electrons. The magnetic moment of the atom will allow the atom to take up to 2j+1 (where j is the inner quantum number) different states in a magnetic field, thus increasing the entropy. See Saha M. N. and Srivastava B. N. , A text book of heat (Allahabad, 1931), 628. In his revised equation, barium and sodium had different effective ionization potentials because of the different configurations of their electrons. The steric factor thus became superfluous because the apparently anomalous behaviour of barium now could be understood as a natural consequence of his generalized equation.

Plaskett H. H. , “The spectra of three O-type stars”, Publications of the Dominion Astrophysical Observatory, i, #30 (1922), 325–84.

Paper II.

Milne E. A. , “Stars of type O and the theory of high-temperature ionization”, The observatory, xlvi (1923), 113–24.

See Russell H. N. , “The properties of matter as illustrated by the stars”, Publications of the Astronomical Society of the Pacific, xxxiii (1921), 275–90.

See Paper I: Russell to Saha , 3 Aug. 1921; and Saha to Russell , 7 Dec. 1921 (Russell papers), et seq.

See Russell to Fowler Alfred , 18 Jan. 1922 (Russell papers).

On the mounting crisis in physics at the time see Forman , op. cit. (ref. 4), and Forman Paul , “Alfred Landé and the anomalous Zeeman effect, 1919–1921”, Historical studies in the physical sciences, ii (1970), 153–261. Kronig R. , “The turning point”, in Fierz M. and Weisskopf V. F. (eds), Theoretical physics in the twentieth century: A memorial volume to Wolfgang Pauli (New York, 1960), 5–39; Serwer Daniel , “Unmechanischer Zwang: Pauli, Heisenberg, and the rejection of the mechanical atom, 1923–1925”, Historical studies in the physical sciences, viii (1977), 189–256; Cassidy , op. cit. (ref. 4).

Born Max , “Über Quantenmechanik”, Zeitschrift für Physik, xxvi (1924), 379–95, quote on p. 379.

Saunders F. A. , “Series in the spectra of calcium, strontium and barium”, Astrophysical journal, xxxii (1910), 153–78. See also Sitterly Moore Charlotte E. , “Collaboration with Henry Norris Russell over the years”, in Philip A. G. D. and DeVorkin D. H. (eds), In memory of Henry Norris Russell (Albany, New York, 1977), 27–41.

During the First World War, according to Saunders's National Academy biographer, both Russell and Saunders worked at Princeton on sound ranging. See Olson H. F. , “Frederick Albert Saunders”, Biographical memoirs of the National Academy of Sciences, xxxix (1967), 403–16. See also Hickman R. W. Hunt F. V. Oldenberg O. and Kemble E. C. , “Frederick Albert Saunders”, Harvard University gazette, 2 May 1964, 188–9. I am indebted to David Cassidy and to Marjorie Graham for providing this information. Although no surviving correspondence between Russell and Saunders has yet been located that dates prior to the early 1920s, this early contact could well have established each in the mind of the other. Their correspondence began only when Harlow Shapley got the two together. Shapley, newly placed as the Harvard College Observatory director, often orchestrated Russell's official visits — for colloquia or sessions of the visiting committee of the Observatory — when Saunders was likely to be around. See Shapley to Russell , 28 July 1921; and Saunders to Russell , 3 Dec. 1921 (Russell papers).

Russell to Saunders , 1 Dec. 1921 (Russell papers).

Saunders to Russell , 3 Dec. 1921 (Russell papers).

How willing spectroscopists were to share their hard-won data is a matter of some interest. Evidence exists that Adams and Charles St John at Mt Wilson were less than delighted to share data with Russell, but both accepted his prescient abilities with the data. See Adams to Abbot Greeley Charles , 9 Aug. 1921; Adams to Charles St John, 24 July 1926; and St John to Adams, 4 Aug. 1926 (Adams papers). Russell's long collaboration with William Meggers of the National Bureau of Standards, on the other hand, was mutually supportive, with Meggers supplying laboratory data, and Russell responding not only with calculations and interpretation, but intellectual patronage, especially when he wrote to George K. Burgess, the new director of the Bureau, in the hopes of “getting you [Meggers] more time for pure science”. See Russell to Meggers , 24 Jan. 1924; Russell to Meggers, 21 Dec. 1923 (Russell papers).

Russell to Fowler Alfred , 18 Jan. 1922 (Russell papers).

Russell , op. cit. (ref. 10), 357.

See Russell to King A. S. , 9 Dec. 1921 (Russell papers).

See letters, Saunders to Russell , 1, 3 and 6 Apr. 1923 (Russell papers).

Catalán Miguel A. , “Series and other regularities in the spectrum of manganese”, Philosophical transactions of the Royal Society (London), ser. A, ccxxiii (1923), 127–73.

See Russell to Fowler Alfred , 17 Apr. 1923 (Russell papers).

Carl Eckart recalled that Russell organized a faculty seminar on modern physics during the period of time he was working on “New regularities”. This seminar ran parallel to one in the Princeton physics department that was formed to study current offerings from the Zeitschrift für Physik. The latter was called the “Princeton'sche Physikalische Gesellschaft”. See Eckert Carl , quoted in Sopka Russell Katherine , Quantum physics in America: The years through 1935 (New York, 1988), 94. Sopka (chap. 2) sees the early 1920s as a watershed period in American theoretical physics, where at the beginning of the decade, quantum theory was not generally accepted but where, by the end of the decade, it was established despite its intrinsic difficulties.

See Russell to Saunders , 5 Oct. 1923 (Russell papers). For a short but critical interval of time, Russell could not obtain particular copies of the Zeitschrift at Princeton since they were being bound by the library. Saunders therefore sent him copies from Harvard, and also suggested which ones to read first.

Ibid.

Russell H. N. and Saunders F. A. , “New regularities in the spectra of the alkaline earths”, Astrophysical journal, lxi (1925), 38–69.

Ibid., 61.

Ibid., 62.

Ibid. (emphasis in original).

On the need for the notation, see Russell to Gale Henry , 20 Nov. 1924 (Yerkes Observatory); Russell to William Meggers, 20 Nov. 1924 (Russell papers).

Sitterly , op. cit. (ref. 28).

Russell to Saunders , 16 Nov. 1923 (Russell papers).

In early 1925, the executive committee of the Physics Division of the National Research Council decided to form a Committee on Line Spectra out of its old Committee on Ionization Potentials. Joseph S. Ames, director of the Physical Laboratory at Johns Hopkins and chairman of the executive committee, asked Russell to head the new committee, and select its members. Russell accepted and placed Saunders, Meggers, and other National Bureau of Standards staff on the committee. This allowed Russell to formalize efforts he and Meggers had been making to standardize notation and to coordinate work on both a national and international scale. See Russell to Meggers William F. , 2 Feb. 1925, and other letters between 1923 and 1926 (Russell papers).

In April 1928, at an informal meeting of spectroscopists in Washington, D. C., the three were asked to clarify the system of spectroscopic notation, and spent the next six months querying their colleagues, reviewing criticisms of the system, and finally presenting their new system to the American Physical Society. Shenstone acted as secretary for the effort. See: Russell, Shenstone and Turner to “Dear Sir”, June 1928 and 19 Dec. 1928 (Russell papers). See also: Russell Norris Henry , “Autobiographical statement” (National Academy of Sciences Archives), 23ff, and Russell H. N. Shenstone A. G. and Turner L. A. , “Report on notation for atomic spectra”, Physical review, xxxiii (1929), 900–6. On citation patterns, see Small Henry , Physics citation index 1920–1929 (Philadelphia, 1981).

Saunders to Russell , 21 Nov. 1923 (Russell papers). Russell sent it in twelve days later: Russell H. N. , “Singlet series in the spark spectrum of aluminium”, Nature, cxiii, issue of 2 Feb. 1924, 163. “PDQ” is an abbreviation commonly taken to mean “pretty damn quick”.

See for instance Russell to Saunders , 5 Oct. 1923 (Russell papers).

See letters, Russell to Adams , 20 and 24 Aug. 1923 (Adams papers). Quote from second letter. Although competition loomed large in Russell's mind at the time, he also admitted that the “monotony of the titanium spectrum … has been as much fun as anything I ever did”. Russell continued on with titanium through his vacation time, alternating between sea island chores such as cleaning drift coal off the beach, burying dead seals and mending boat motors, and attending to his beloved spectroscopic puzzles.

See White H. E. , Introduction to atomic spectra (New York, 1934), 204, 267. Russell's preoccupation with titanium reveals that he did not have a clear understanding of what was important about his work; he viewed the coupling scheme as a simple application of the vector model of the atom, while he seemed to see some deeper significance for his exclusion principle. See Russell H. N. and Lang R. J. , “On the spectra of doubly and trebly ionized titanium”, Astrophysical journal, lxvi (1927), 13–42.

Russell to Saunders , 10 Dec. 1923 (Russell papers).

Saunders to Russell , 21 Nov. 1923 (Russell papers).

Russell to Saunders , 26 Nov. 1923 (Russell papers).

Russell to Saunders , 10 Dec. 1923 (Russell papers).

See Serwer , op. cit. (ref. 26); Cassidy , op. cit. (ref. 4); Forman , op. cit. (ref. 4); Kronig , op. cit. (ref. 26); MacKinnon Edward , “Heisenberg, models and the rise of matrix mechanics”, Historical studies in the physical sciences, viii (1977), 137–88.

Wentzel Gregor , “Bemerkungen über Serienspektren an deren Emission mehr als ein Elektron beteiligt ist”, Physikalishe Zeitschrift, xxiv (1923), 104–9.

See Forman , op. cit. (ref. 4); Cassidy , op. cit. (ref. 4).

Serwer , op. cit. (ref. 26); van der Waerden B. L. , “Exclusion principle and spin”, in Fierz and Weisskopf (eds), op. cit. (ref. 26), 199–244; Kronig , op. cit. (ref. 26).

MacKinnon , op. cit. (ref. 59); Cassidy , op. cit. (ref. 4).

Russell to Milne E. A. , 5 Oct. 1925 (Russell papers).

Russell to Laporte O. , 9 Nov. 1925 (Russell papers).

Russell to Babcock H. D. , 30 Oct. 1923 (Russell papers).

Saunders to Russell , 30 Jun. 1924 (Russell papers). They did mention Wentzel's independent speculation on the possibility of two-electron interactions.

Forman , op. cit. (ref. 4), 172.

Saunders to Russell , 29 Aug. 1924 (Russell papers).

Russell and Saunders , op. cit. (ref. 42), 66.

Forman , op. cit. (ref. 4); Serwer , op. cit. (ref. 26); Kronig , op. cit. (ref. 26).

Pauli Wolfgang , “Über den Zussamenhang des Abschusses der Elektronen-gruppen im Atom mit Komplexstruktur der Spektren”, Zeitschrift für Physik, xxxi (1925), 765–83.

The Institute for Scientific Information's Physics citation index 1920–1929 reveals that there were 48 citations between 1925–29 to “New regularities”, 22 of which were in German journals. Those in Germany and the Netherlands responded first; in 1925 there were 8 from Goudsmit, Heisenberg, Hund, Kronig, Wentzel, and Back, etc., mainly in the Zeitschrift für Physik, and one from I. S. Bowen and R. A. Millikan in America. Seventeen of the 22 German citations appeared in the Zeitschrift. During 1926–29, 12 citations appeared in the Physical review and only two in the Astrophysical journal. Twelve of the 48 citations were to the final section on spectroscopic notation in “New regularities” and 29 appeared in specifically spectroscopic works. Of the 48 citations, 22 were attempts to extend the work of Russell and Saunders to the spectra of other substances. Another seven were attempts to solve broader, but nevertheless specifically spectroscopic problems such as finding new regularities in new groups of substances. The remaining 19 papers were attempts to expand upon the theoretical sections of “New regularities”. Finally, in 12 of the papers (including 7 from those designated as theoretical), Russell and Saunders were cited along with the nearly simultaneous contributions of Hund, Heisenberg and Pauli. See Small , op. cit. (ref. 50).

Heisenberg Werner , “Zur Quantentheorie der Multipletstruktur und der anomalen Zeemaneffekte”, Zeitschrift für Physik, xxxii (1925), 841–60. The Zeitschrift was used primarily by the members of the Munich, Copenhagen, and Göttingen schools which centred around Sommerfeld, Bohr and Born respectively. Almost all the papers of Landé, Heisenberg, and Pauli cited here appeared in the Zeitschrift. In contrast, most of Einstein's papers appeared in the Annalen der Physik. Cassidy David C. , in his introduction to “Heisenberg's first core model of the atom: The formation of a professional style”, op. cit. (ref. 4) offers useful insight into the subcommunities within the European theoretical physics community.

Katherine Sopka has studied the American theoretical physics community and its dependence upon the German community in the 1920s. See Sopka , op. cit. (ref. 39), chap. 2, 73. She also examines the reception of “New regularities” as an example of a theoretical contribution to quantum mechanics “by Americans who were not theoretical physicists”. She notes (p. 116) that the 1931 edition of Arnold Sommerfeld's Atombau and Spectrallinian (5th edn, Braunschweig, 1931) contained an extended review of Russell-Saunders coupling. On the state of theoretical physics in the 1920s in America, see also Coben Stanley , “The scientific establishment and the transmission of quantum mechanics to the United States, 1919–1932”, American historical review, lxxvi (1971), 442–66.

Heisenberg , op. cit. (ref. 74).

See Forman , op. cit. (ref. 4); Serwer , op. cit. (ref. 26); Cassidy , op. cit. (ref. 4). Pauli also argued that individual electrons could not have the same four quantum numbers. This exclusion principle was an attempt to avoid the traps into which various atomic models had been led. But the interpretation of the four quantum numbers by those less opposed to classical models would lead to the concept of spin. See Kronig , op. cit. (ref. 26) and van der Waerden , op. cit. (ref. 62).

Heisenberg's attempt to resurrect the Rumpf model has justifiably attracted the most attention from historians. It was, of course, quite a failure. But precisely because it was such a failure, this work turned out to be pivotal for Heisenberg; following this paper in April 1925 Heisenberg abandoned the Rumpf model and the problem of multiple electron atoms and began an attack upon the hydrogen atom using the idea of “virtual oscillators” that he had recently become sensitized to in Copenhagen. This new turn would lead him to develop matrix mechanics by the end of the year. See Serwer , op. cit. (ref. 26), 241–2.

Although Hund later stated that Heisenberg thought of the coupling scheme independently, Heisenberg gave full credit to Russell and Saunders in this study. See Hund Friedrich , The history of quantum theory (New York, 1974).

Hund Freidrich , “Zur Deutung verwickelter Spektren insbesondere der Elemente Scandium bis Nickel”, Zeitscrift für Physik, xxxiii (1925), 345–71; “Atom theoretische Deutung des Magentismus der seltenen Erden”, Zeitscrift für Physik, xxxiii (1925), 855–9.

Heisenberg Werner and Jordan P. , “Anwendung der Quantenmechanik auf das Problem der Anomalen Zeemaneffekte”, Zeitschrift für Physik, xxxvii (1926), 263–75.

In 1927 Hund brought together the accomplishments of the previous few years into a comprehensive treatment of the spectra of most of the elements. See Hund Friedrich , Linienspektren und periodisches System der Elemente (Berlin, 1927).

Russell to Webster David , 11 Oct. 1926 (Russell papers).

Paper II, 197.

Russell , op. cit. (ref. 83). On the state of quantum mechanics and theoretical physics in general in the United States in the 1920s, see Sopka , op. cit. (ref. 39) and Coben , op. cit. (ref. 75).

Russell H. N. Dugan R. S. and Stewart J. Q. , Astronomy I ; II (Boston, 1926; 1927).

Russell to Laporte Otto , 9 Nov. 1925 (Russell papers).

Candler Albert C. , Atomic spectra and the vector model (Cambridge, 1937), 4; see also the 1964 edition, p. 3.

Johnson R. C. , Atomic spectra (London, 1946), p. v.

“Henry Norris Russell”, in Finkelstein Louis (ed.), Thirteen Americans: Their spiritual autobiographies (New York, 1953; reissued, 1969), 31–45, p. 42.

Stewart J. Q. , “Henry Norris Russell”, Monthly notices of the Royal Astronomical Society, cxviii (1958), 311–12, p. 311. Russell did not maintain the best of relations with Stewart in later years. Although much of Stewart's commentary in this obituary notice required further elaboration, the statement cited here is confirmed by the present analysis.

Russell , op. cit. (ref. 40).

Meggers William F. , “Multiplets and terms in technetium spectra”, Journal of research of the National Bureau of Standards, xlvii, # 1 (July 1951), 7–14, p. 7.

Catalán , op. cit. (ref. 37).

See the monograph by Van Vleck John H. , Quantum principles and line spectra (Bulletin of the National Research Council, x, pt 4 (Mar. 1926)). See also Cassidy , op. cit. (ref. 4) and Forman , op. cit. (ref. 26).

Candler , op. cit. (ref. 88, 1964 edition), 64–67.

See Landé Alfred , “Über den anomalen Zeemaneffekt (Teil I)”, Zeitschrift für Physik, v (1921), 231–41; “Zur Theorie der anomalen Zeeman- und magneto-mechanischen Effekte”, Zeitschrift für Physik, xi (1922), 353–63; “Termstruktur und Zeemaneffekt der Multipletts”, Zeitschrift für Physik, xix (1923), 112–23.

See Forman , op. cit. (ref. 26) and Van Vleck op. cit. (ref. 95), 230. Jammer , op. cit. (ref. 4), chap. 3.3, offers a very good review of the Zeeman effect.

See Heisenberg Werner , “Über eine Abänderung der formalen Regeln der Quantentheorie beim Problem der anomalen Zeemaneffekte”, Zeitschrift für Physik, xxvi (1924), 291–307; “Zur Quantentheorie der Linienstruktur und der anomalen Zeemaneffekte”, Zeitschrift für Physik, viii (1922), 273–97.

See MacKinnon , op. cit. (ref. 59); Cassidy , op. cit. (ref. 4).

See Landé A. , “Termstruktur und Zeemaneffekt der Multipletts”, Zeitschrift für Physik, xv (1923), 189–205.

The best reviews of the vector model are in Van Vleck , op. cit. (ref. 95); and Candler , op. cit. (ref. 88).