In 1925 a twenty-five-year-old graduate student worked out what the Sun is made of, arrived at the right answer, and was persuaded to call it wrong.

The reasonable assumption that was wrong

For most of history the composition of a star was beyond asking. They were points of light; what they were made of was anyone's guess. Then, in the nineteenth century, came the astonishing discovery that you could simply look and find out.

Spread starlight through a prism and it fans into a rainbow — but the rainbow is not smooth. Ruled across it are thin dark lines, narrow gaps where colour is missing. Joseph von Fraunhofer mapped hundreds of them in sunlight around 1814 without knowing what they were. The answer came in the 1850s, when Gustav Kirchhoff and Robert Bunsen showed that every chemical element, heated in a flame, emits its own fixed pattern of lines — a barcode unique to that element. The dark gaps in sunlight were those same barcodes in negative: cooler gas in the Sun's outer layers absorbing exactly the colours its elements would emit. Match the pattern and you have read, across ninety million miles, what the Sun contains.

The barcodes in the Sun were crowded with iron, calcium, sodium, magnesium — the very metals that make up the rock beneath your feet. The natural conclusion, and the one nearly every astronomer drew, was that the Sun was built of much the same stuff as the Earth: a hot ball of familiar elements. It seemed almost too obvious to question.

Line strength is a thermometer, not a scale

The flaw was subtle, and it took new physics to expose. A bold, dark line does not simply mean "a great deal of this element." Whether an atom absorbs its line at all depends on its temperature — on whether its electrons are sitting where they need to sit, and whether the atom has been stripped of them altogether. In 1920 the Indian physicist Meghnad Saha wrote down the equation governing this: how, at a given temperature, a gas divides between its neutral and ionised forms.

Saha's equation changed what a spectrum meant. A line's strength blends two things together — how much of the element is present, and how the star's heat has prepared it to absorb. At the temperature of a typical stellar surface, hydrogen mostly sits in a state that barely touches visible light, so it leaves only faint marks even when it is overwhelmingly abundant. Iron, by contrast, is primed to absorb and shouts from the spectrum on a modest ration. To read the true amount of an element, you first have to divide out the temperature. Nobody yet had.

Cecilia Payne does the arithmetic

Cecilia Payne had left England for Harvard, where the observatory held a vast library of stellar spectra painstakingly classified by Annie Jump Cannon. For her 1925 doctoral thesis, Stellar Atmospheres, Payne took Saha's new equation and did the correction properly — right across the sequence of stellar types, from the hottest blue stars to the coolest red ones, untangling temperature from abundance line by line.

When she was done, the numbers were staggering. Corrected for temperature, hydrogen and helium were not minor ingredients at all. They were nearly everything. Hydrogen came out roughly a million times more abundant than the iron and calcium whose lines had so dominated the picture. The Sun was not a ball of Earth-like rock. It was overwhelmingly the lightest, simplest element there is.

"Almost certainly not real"

The result was too much for the field to swallow. Henry Norris Russell, the most influential astronomer in America, reviewed the work and did not believe it — the notion that stars were mostly hydrogen simply contradicted everything the profession took for granted. Under his influence, a sentence went into the thesis stating that the enormous abundance of hydrogen was "almost certainly not real." Payne had found the single most important fact about the composition of the universe and, in the same document, disowned it.

She had discovered that the cosmos is mostly hydrogen — and the discipline's response was to have her doubt it in print.

Four years later, in 1929, Russell reached the same conclusion by a different route and published it. He credited Payne's earlier work, but the discovery travelled for decades under his name. Payne — later Payne-Gaposchkin — stayed at Harvard, and in time became the first woman promoted to full professor there through its own faculty. The astronomer Otto Struve would call her thesis "undoubtedly the most brilliant PhD thesis ever written in astronomy."

What the dark lines were really saying

She was right, completely. The Sun is about three-quarters hydrogen by mass and almost all the rest helium; everything heavier — the iron, the calcium, the oxygen — is a trace, a percent or two. That ratio is no quirk of the Sun's: it is close to the recipe the Big Bang left behind, and the hydrogen is the very fuel the Sun fuses into helium to shine at all. The heavier elements are the universe's afterthought, forged later inside stars and their deaths.

The same dark lines carry more than a shopping list. Their positions shift, ever so slightly, when a star moves toward us or away — and that sliding of the spectral lines became astronomy's finest ruler for cosmic motion. The overall colour of the light already told you how hot the surface is; the fine dark lines threaded through it tell you what the surface is made of, and how fast it is coming or going. One smear of starlight, read closely enough, gives up nearly everything.

It turns the schoolroom picture inside out. The ground under your feet, dense with iron and silicon, is not a sample of the cosmos — it is the rare exception, a scrap of heavy sediment. The universe at large is hydrogen, plain and light and everywhere, and a graduate student read that off a band of coloured light and was told, for a while, to look away.