Note: The AIP History Weekly Edition is on a summer break and will resume publication of original content on September 4. This article was first published in June in AIP’s semiannual History Newsletter. Part 2 will appear next week.

---

In 1835 the French philosopher Auguste Comte declared that humanity would never know the chemical composition of the stars. It was, he thought, the perfect example of knowledge that would remain permanently beyond a boundary that science could not cross.¹

He was, of course, wrong. Within a generation, scientists began to crack the code of starlight and set astronomy on a collision course with physics that would reshape both disciplines from that moment to the present. The story of how that collision unfolded—the story of how astrophysics, distinct from astronomy, came to exist and flourish—is the story of one of the great transformations in the history of science. And, for anyone in Comte’s time, it is a story that would have been not only impossible to foresee; it would have been absurd to predict.

Spectral lines and a new discipline

A crucial breakthrough came in the late 1850s, when the German physicist Gustav Kirchhoff and chemist Robert Bunsen showed that the dark lines scattered across the solar spectrum corresponded to the bright emission lines of specific chemical elements produced in the laboratory. Iron, sodium, calcium, magnesium could all be identified in the Sun’s atmosphere from millions of miles away. The heavens, it turned out, were composed of the same elements as the Earth.²

This was a radical departure from the astronomy that had prevailed for centuries. Traditional positional astronomy concerned itself with tracking the motions of celestial bodies, which was essential work for navigation, timekeeping, and calendar-making that had been starkly separate from speculation about physical processes. But by the 1890s, the new approach had begun to institutionalize: dedicated journals appeared, including the Astrophysical Journal in 1895, and leading observatories began to take on the characteristics of experimental laboratories.³

In the United States, the door opened to this new kind of inquiry with the total solar eclipse of August 7, 1869. As Jennifer Bartlett and Thomas Hockey have recently shown, when the eclipse swept from Dakota Territory to the Carolinas, it prompted an ambitious expedition organized by the US Nautical Almanac Office, which had no telescopes, no observational mandate, and a staff trained in computation rather than laboratory science. Its superintendent, John Coffin, recruited Dartmouth physicist Charles Young, secured $5,000 in congressional funding (over $100,000 in today’s dollars), and dispatched teams to Iowa armed with borrowed spectroscopes and telescopes.

In Burlington, Iowa, Young made the expedition’s most remarkable observation: through a five-prism spectroscope, he detected a faint green emission line from the solar corona that matched no known terrestrial element. Scientists would puzzle over this line for more than seventy years before the Swedish spectroscopist Bengt Edlén identified it in 1942 as the signature of highly ionized iron, an early clue to the astonishingly high temperatures of the Sun’s outer atmosphere.⁴

Other participants struggled with unfamiliar instruments and unstable mounts; some tried spectroscopy for the very first time. The results were uneven, but the shift in ambition was unmistakable. American astronomy had begun to imagine itself as an experimental science. One of these participants, Edward Pickering, soon took on the directorship of the Harvard College Observatory, where he implemented a program of photography that allowed stellar spectra to be systematically analyzed.

This daytime analytical work was done by a team of women computers Pickering hired that included figures such as Williamina Fleming, Antonia Maury, and Annie Cannon, who would transcend their initially intended roles to creatively develop new classifications of stars. These classifications, in turn, informed new theories about not only the composition of stars, but also, as embodied in the Hertzsprung-Russell diagram, how they evolved over their lifespans.⁵

Relativity transforms the cosmos

If spectroscopy opened a new window to the sky, Albert Einstein’s contributions would prove crucial to understanding what could be seen through it. The special theory of relativity introduced the concept of mass-energy equivalence, which would eventually be needed to determine the source of stars’ energy, but it was the general theory of relativity that would have the most far-reaching consequences. The idea that space and time could be curved by gravity could explain an enduring mystery of positional astronomy, the precessing perihelion of Mercury, and it predicted that light could be perceivably bent as it passed around stars.

The test came during the total solar eclipse of May 29, 1919. Two British expeditions—one led by Arthur Eddington to the island of Príncipe, off the west coast of Africa, and the other by Andrew Crommelin to Sobral, Brazil—set out to photograph stars near the Sun during totality. If Einstein was right, the Sun’s gravity should bend their light by a precise, predictable amount. The conditions were far from ideal: heavy thunderstorms soaked Eddington’s site for much of the morning, and only a handful of usable plates survived. But the results, presented at a dramatic joint meeting of the Royal Society and Royal Astronomical Society on November 6, 1919, confirmed the prediction. Einstein became an international celebrity overnight.⁶

The 1919 eclipse is rightly remembered as a triumph for Einstein, but it also signaled the beginning of an era in which fundamental physical theories became an important tool for astrophysics, and astrophysics in turn became a proving ground for profound theories about the universe. In the 1920s, Eddington synthesized ideas about radiation, gravitation, ultradense states of matter, general relativity, and quantum mechanics to explain white dwarfs, unexplained outliers on the Hertzsprung-Russell diagram, as stars that had collapsed at the end of their lives.

This advance set up a legendary confrontation in 1935 with a young Subrahmanyan Chandrasekhar, who invoked the effects of relativity on electrons in ultradense matter to posit that sufficiently massive stars could collapse into still denser masses—a conclusion Eddington vehemently rejected. “Eddington realized that the existence of a limiting mass implies that black holes must occur in nature. But he did not accept that conclusion. He said that must be a reduction ad absurdum. … And if he had accepted that he would have been 40 years ahead of anybody else,” Chandrasekhar later lamented to AIP’s Spencer Weart.⁷

Meanwhile, Einstein, Willem de Sitter, Georges Lemaître, and other theorists began to consider the implications of general relativity for the shape of the universe itself. The curvature of spacetime by gravity offered a geometry implying the universe has a finite size while lacking an outer edge. The equations of general relativity also suggested this universe would expand with time—a possibility that became the basis for modern cosmology. Einstein rejected the idea initially and introduced a “cosmological constant” to avoid the expansion, which he famously later looked back on as his “greatest blunder.”⁸

Notes and references

1. Auguste Comte, Cours de Philosophie Positive, vol. 2 (Bachelier, 1835).
2. G. Kirchhoff and R. Bunsen, “Chemische Analyse durch Spectralbeobachtungen,” Annalen der Physik und Chemie 110, no. 6 (1860): 161–189.
3. Donald E. Osterbrock, “Founded in 1895 by George E. Hale and James E. Keeler: The Astrophysical Journal Centennial,” Astrophysical Journal 438, no. 1 (1995): 1–7, doi.org/10.1086/175049.
4. Jennifer Lynn Bartlett and Thomas Hockey, “The Total Solar Eclipse of 1869 as Stimulus for Adoption of Physical-astronomy Techniques in the United States,” Physics in Perspective 27, no. 1 (2025): 3–25, doi.org/10.1007/s00016-025-00322-3; see also Deborah Kent, “The North American Eclipse of 1869,” Physics Today 72, no. 8 (2019): 46–53, doi.org/10.1063/PT.3.4271.
5. Solon I. Bailey, The History and Work of Harvard Observatory, 1839–1927 (Cambridge: Harvard Observatory, 1931); see also Dava Sobel, The Glass Universe: How the Ladies of the Harvard Observatory Took the Measure of the Stars (Viking, 2016). On the significance of classification more broadly, see David H. DeVorkin, “Community and Spectral Classification in Astrophysics: The Acceptance of E. C. Pickering’s System in 1910,” Isis 72, no. 1 (1981): 29–49, doi.org/10.1086/352649.
6. A comprehensive account is offered in Daniel J. Kennefick, No Shadow of a Doubt: The 1919 Eclipse That Confirmed Einstein’s Theory of Relativity (Princeton University Press, 2019).
7. Kameshwar C. Wali, “Chandrasekhar vs. Eddington—An Unanticipated Confrontation,” Physics Today 35, no. 10 (1982): 33–40, doi.org/10.1063/1.2914790; quote from interview of Subrahmanyan Chandrasekhar by Spencer Weart on May 17, 1977, doi.org/10.1063/nbla.kfld.iqgl.
8. The general history of the relationship between relativistic physics, astrophysics, and cosmology is summarized in The Oxford Handbook of the History of Modern Cosmology, edited by Helge Kragh and Malcolm S. Longair (Oxford University Press, 2019), doi.org/10.1093/oxfordhb/9780198817666.001.0001.
   

—

Rebecca Charbonneau
American Institute of Physics
rcharbonneau@aip.org

---

You can sign up to receive the Weekly Edition and other AIP newsletters by email here.

---