Somewhere in the 1925 thesis Stellar Atmospheres sits a sentence its author did not believe. The enormous abundances of hydrogen and helium she had just calculated, Cecilia Payne wrote, were “almost certainly not real.” She was twenty-five, and she had just become the first person to work out what the stars are made of. The line saying her own answer was wrong is the most important thing in the document, because it was put there against the evidence, and because the evidence won anyway.
The number she was made to disown
The answer was hydrogen. Reading the dark absorption lines in starlight, Payne found that the Sun was not a hot version of the Earth at all. For the metals — silicon, carbon, iron, calcium — her figures matched what everyone expected; those were the same materials found in the planet’s crust, in roughly the proportions science assumed. Hydrogen and helium broke the pattern. Her calculations put them not slightly ahead but overwhelmingly ahead: hydrogen roughly a million times more abundant than the metals, helium about a thousand times. The Sun was almost entirely the two lightest, simplest elements in existence, with everything else a trace contaminant.
Nothing in the science of 1925 prepared anyone for that. The Sun and the Earth were supposed to share a composition.
So when Harlow Shapley, the director of the Harvard College Observatory, sent the work to Henry Norris Russell — the Princeton astronomer who was, by common consent, the most influential figure in the field — Russell balked at the hydrogen. It was, he wrote back, “clearly impossible” that hydrogen could be a million times more abundant than the metals.
He was not a fool, and the objection was not pure prejudice. The result genuinely contradicted everything then understood about the Sun, and the young astronomer who’d produced it had every reason to defer to the man who would shape her career. So she added the line. The hydrogen and helium abundances were “almost certainly not real.”
She did not delete the analysis. The number stayed on the page, hedged but visible, where anyone could find it later.
How she got there
The road to that page ran through a lecture hall. Payne was born in Wendover, England, in May 1900, and went up to Newnham College, Cambridge, in 1919 to read botany. That autumn she sat in on a lecture by Arthur Eddington, just back from the island of Príncipe, where his eclipse photographs had caught starlight bending around the Sun exactly as Einstein had predicted. Payne later said the talk reordered her mind; she switched to physics and astronomy and never looked back.
England had nothing to offer a woman who wanted to be an astronomer. Cambridge would not grant her a degree at all — it withheld degrees from women until 1948. So when Shapley visited England, Payne asked him whether there was room for her across the Atlantic. With Eddington’s backing, Shapley arranged a fellowship, and in 1923 she sailed for Massachusetts.
Harvard sat on the largest archive of stellar spectra in the world: hundreds of thousands of glass photographic plates, each one a smear of starlight split into its component colors, crossed by dark lines where atoms in the star’s atmosphere had absorbed particular wavelengths. For decades a team of women — paid by the hour, hired to measure and catalog rather than to theorize — had been classifying these spectra. Annie Jump Cannon had sorted a quarter of a million stars into the temperature sequence astronomers still use.
The dark lines were a fingerprint of what the stars contained. Reading them, though, was treacherous. The strength of a line did not simply track how much of an element was present; it also depended on temperature, which governed how many of those atoms were in the right state to absorb light at all. A faint hydrogen line in a cool star and a strong one in a hot star could mean the same amount of hydrogen.
Payne had a tool most astronomers lacked. The Indian physicist Meghnad Saha had published an equation in 1920 describing exactly how temperature strips electrons from atoms — how it ionizes a gas. Payne applied Saha’s ionization theory to the Harvard plates systematically, working backward from line strengths to the actual abundances of the elements, temperature accounted for. That was how the impossible number came out: not as a guess, but as the thing the equation insisted on.
The thesis was accepted in 1925. Because Harvard’s physics department would not grant a doctorate to a woman, the degree was conferred through Radcliffe, the university’s women’s college — the first PhD in astronomy it ever awarded.
The answer comes back through the other man
Russell kept working on stellar composition, and in 1929 he published his own analysis, reaching the result by a different route. The Sun, he concluded, was overwhelmingly hydrogen. It was the same answer Payne had buried under a disclaimer four years earlier.
He cited her. His paper noted that his figures agreed with hers. But it was Russell’s 1929 paper that the field absorbed as the moment the question was settled, and for decades it was Russell’s name that textbooks attached to the discovery. The conclusion she had reached first, then been argued out of, came back into astronomy through the man who had talked her out of it.
Payne stayed at Harvard. For years she did the work of a faculty member without the title, her courses unlisted in the catalog. Only in 1956 did she become Harvard’s first woman full professor, and then the first woman to chair a department there. Three years before her death in 1979, the American Astronomical Society gave her its highest honor — the Henry Norris Russell Lectureship, named for the man who had pressed her to call her own discovery unreal.
In 1962, the astronomers Otto Struve and Velta Zebergs looked back at the 1925 thesis and called it “undoubtedly the most brilliant PhD thesis ever written in astronomy.”