Field · Emerged 1814 – 1925

Astronomical Spectroscopy

What can the light of a star, spread into its colours, reveal about what the star is made of?

5 chapters4 min read6 turning points 1 open problem

Branched from: Electromagnetism + Old Quantum Theory
Branched into: Exoplanetary Science + Galactic Astronomy + Stellar Astrophysics
Figures: Joseph von Fraunhofer, Gustav Kirchhoff, Robert Bunsen, Pierre Janssen, Norman Lockyer, Edward Pickering, Williamina Fleming, Annie Jump Cannon, Ejnar Hertzsprung, Henry Norris Russell, Meghnad Saha, Cecilia Payne-Gaposchkin

In brief

Spread starlight through a prism and it forms a rainbow crossed by dark lines. Each chemical element absorbs light at its own set of wavelengths, so the lines are a fingerprint of the elements in a star's atmosphere. Their strengths reveal the star's temperature, their shifts its motion towards or away from us, and their widths its pressure and spin.

Spectroscopy turned astronomy, which had measured where stars are, into astrophysics, which asks what they are. Within a lifetime of its invention it had found a new element in the Sun, sorted hundreds of thousands of stars into a single temperature sequence, and uncovered the pattern of stellar evolution. Its biggest surprise, explained only with quantum theory in 1925, was that stars are made almost entirely of hydrogen.

Key ideas

Spectral linesEnters 1859 – 1860: Dark or bright lines at specific wavelengths, produced when atoms absorb or emit light. Each element has its own pattern, fixed by the structure of its atoms.
Spectral classificationEnters 1890 – 1924: Stars sorted by the lines in their spectra, in the sequence O, B, A, F, G, K, M. The sequence turned out to be one of decreasing surface temperature.
Hertzsprung–Russell diagramEnters 1911 – 1913: A plot of stars' brightness against their temperature. Most fall on a band, the main sequence, and the others reveal stages in stellar lives.
Ionisation and temperatureEnters 1925: How strongly a line appears depends on how many atoms are in the right state to absorb, which depends on temperature. Saha's equation, from quantum physics, quantifies it.

Chapter I

Lines in the Rainbow

In 1835 the philosopher Auguste Comte gave, as an example of knowledge forever beyond reach, the chemical composition of the stars. The key to it had already been found. In 1814 Joseph von Fraunhofer, a Bavarian optician who had survived the collapse of the workshop where he was apprenticed as a boy, spread sunlight through a fine prism and found it crossed by hundreds of dark lines. He mapped more than 500 and saw different patterns in the light of bright stars.

In 1859 Gustav Kirchhoff and Robert Bunsen, using Bunsen's clean, nearly colourless gas burner, explained them. Every element heated in a flame glows at its own set of wavelengths, and cooler gas of the same element absorbs those wavelengths from light passing through it. The dark lines in sunlight were the Sun's elements absorbing light from the hotter layers below: sodium, iron, calcium and many more. Chemistry could be done at a distance of 150 million kilometres. Electromagnetism soon explained that light is an electromagnetic wave, and the lines were its wavelengths.

Chapter II

A Sequence of Stars

Spectroscopy found things chemistry had not. In 1868 Pierre Janssen and Norman Lockyer saw a yellow line in the Sun that belonged to no known element, and Lockyer named it helium. William Huggins found in 1864 that some nebulae are glowing gas, not unresolved stars.

Photography made spectra of faint stars possible, and at Harvard Edward Pickering hired women to analyse them at a fraction of a man's wage. Williamina Fleming, first employed as his housekeeper, devised an early scheme. Annie Jump Cannon refined it into the sequence O, B, A, F, G, K, M and classified hundreds of thousands of stars by eye. In 1911–13 Ejnar Hertzsprung and Henry Norris Russell plotted brightness against spectral type and found that most stars lie on a single band, the main sequence. Stars were not a random assortment. They followed patterns that a theory of their lives would have to explain.

Chapter III

Made of Hydrogen

What did the sequence mean? The Sun's spectrum is dominated by lines of iron, calcium and other metals, and it was assumed that stars resemble the Earth in composition. In 1920 Meghnad Saha used the new quantum theory of the atom to show how temperature controls which atoms can absorb which lines. In 1925 Cecilia Payne, a young British astronomer at Harvard, applied Saha's theory to Cannon's spectra. The spectral sequence was a temperature sequence, and once temperature was accounted for, stars had almost the same composition. And that composition was mostly hydrogen and helium, with hydrogen about a million times more abundant than the metals whose lines are so prominent.

Russell told her it was impossible, and she called it "almost certainly not real" in print. Four years later Russell confirmed it himself.

Chapter IV

A Closer Look: Reading Temperature From Colour

A hot, dense body glows with a spectrum whose peak wavelength shifts as its temperature changes, following Wien's law:

\lambda_{\text{peak}} = \frac{2.898 \times 10^{-3}\ \text{m·K}}{T} .

Star	Surface temperature	Peak wavelength	Colour
Betelgeuse (M)	about 3,600 K	about 805 nm (infrared)	red
The Sun (G)	5,772 K	about 502 nm	yellow-white
Sirius (A)	about 9,900 K	about 290 nm (ultraviolet)	blue-white
Rigel (B)	about 12,100 K	about 240 nm (ultraviolet)	blue

The Sun's output peaks in the green-blue, in the middle of the range our eyes evolved to see. The mixture of all its colours looks white from space.

The lines tell a subtler story. Hydrogen's visible lines, the Balmer series, are absorbed only by hydrogen atoms whose electron is already in the second energy level. In a cool star like Betelgeuse, almost all hydrogen atoms sit in the lowest level, so the lines are weak. In very hot stars, most hydrogen is ionised and cannot absorb at all. The lines are strongest in between, around 10,000 K, in A stars like Sirius. In the Sun, only a few hydrogen atoms in every billion are in the right state. That is why hydrogen's lines looked unimpressive, and why astronomers concluded that the Sun contains little hydrogen. Saha's equation lets the tiny fraction be corrected for, and when Payne did so, hydrogen turned out to be by far the most common element. It is now known to make up about three-quarters of the Sun's mass.

Chapter V

Astrophysics

Spectroscopy also measures motion. A star moving away has its lines shifted to longer wavelengths, and one moving towards us to shorter. That shift showed that stars orbit each other, measured the rotation of galaxies, and in the 1920s revealed that the galaxies are receding, the evidence for the expanding universe of physical cosmology. It now reveals planets around other stars by the tiny wobble they cause. The question Payne's result raised, how stars made of hydrogen shine, belonged to stellar astrophysics.

Turning points6

1814DISCOVERY
Fraunhofer's dark lines
Joseph von Fraunhofer, a Bavarian glassmaker testing prisms for telescope lenses, maps more than 500 dark lines crossing the spectrum of sunlight. He finds the same lines in light reflected from the Moon and planets, and different patterns in bright stars. He did not know what caused them.
Joseph von Fraunhofer
› Sources (1)
- Fraunhofer, J. (1817). Bestimmung des Brechungs- und Farbenzerstreuungs-Vermögens verschiedener Glasarten. Denkschriften der Königlichen Akademie der Wissenschaften zu München 5: 193–226.
1859 – 1860DISCOVERY
Each element has its own lines
Gustav Kirchhoff and Robert Bunsen in Heidelberg show that every element, heated in a flame, emits light at its own characteristic wavelengths, and absorbs at the same wavelengths when cooler light shines through it. Fraunhofer's lines are the signatures of sodium, iron and other elements in the Sun. The chemistry of the stars could be read from their light.
Gustav Kirchhoff, Robert Bunsen
› Sources (1)
- Kirchhoff, G. & Bunsen, R. (1860). Chemische Analyse durch Spectralbeobachtungen. Annalen der Physik und Chemie 110: 161–189.
1868DISCOVERY
A new element found in the Sun
Observing a solar eclipse, Pierre Janssen sees a bright yellow line in the Sun's prominences that matches no known element. Norman Lockyer attributes it to a new element and names it helium, after the Greek for the Sun. Helium was found on Earth only in 1895, by William Ramsay.
Pierre Janssen, Norman Lockyer
› Sources (1)
- Nath, B. B. (2013). The Story of Helium and the Birth of Astrophysics. Springer.
1890 – 1924DISCOVERY
The Harvard classification
At Harvard College Observatory, Edward Pickering employs a team of women, the "computers", to classify stellar spectra from photographic plates. Williamina Fleming, Antonia Maury and above all Annie Jump Cannon build the scheme still used today, O, B, A, F, G, K, M. Cannon alone classified around 350,000 stars, and the Henry Draper Catalogue became the basis of stellar astronomy.
Edward Pickering, Williamina Fleming, Annie Jump Cannon
› Sources (2)
- Cannon, A. J. & Pickering, E. C. (1918–1924). The Henry Draper Catalogue. Annals of Harvard College Observatory 91–99.
- Sobel, D. (2016). The Glass Universe: How the Ladies of the Harvard Observatory Took the Measure of the Stars. Viking.
1911 – 1913DISCOVERY
The Hertzsprung–Russell diagram
Ejnar Hertzsprung in Denmark and Henry Norris Russell at Princeton independently plot stars' true brightness against their spectral type. Most stars lie on one band, the main sequence, and a few are bright giants or faint dwarfs. The diagram became the central tool for understanding how stars live and die.
Ejnar Hertzsprung, Henry Norris Russell
› Sources (1)
- Russell, H. N. (1914). Relations between the spectra and other characteristics of the stars. Popular Astronomy 22: 275–294.
1925PARADIGM-SHIFTContested
Stars are made of hydrogen
Using Meghnad Saha's quantum theory of ionisation, Cecilia Payne shows in her Harvard doctoral thesis that the differences between stellar spectra reflect temperature, not composition, and that stars are overwhelmingly hydrogen and helium. It was thought impossible, because the Sun's spectrum is full of lines of iron and other metals. Otto Struve later called it the most brilliant thesis ever written in astronomy.
What is disputed
Henry Norris Russell, reviewing the thesis, told Payne that the hydrogen abundance was "clearly impossible", and she described her result as "almost certainly not real". In 1929 Russell reached the same conclusion by other methods and is sometimes credited with it. Historians now credit Payne with the discovery, and her case is often cited as an example of a woman's result credited to a senior man.
Henry Norris Russell, Meghnad Saha, Cecilia Payne-Gaposchkin
› Sources (2)
- Payne, C. H. (1925). Stellar Atmospheres. Harvard Observatory Monographs 1. Harvard College Observatory.
- Saha, M. N. (1920). Ionization in the solar chromosphere. Philosophical Magazine 40: 472–488.

Applications

Where it is used

Chemistry
New elements from their spectra
Within two years of inventing spectral analysis, Bunsen and Kirchhoff discovered caesium and rubidium from unfamiliar lines in mineral water and minerals. Thallium, indium, helium and several rare gases followed the same way.
› Sources (1)
- Kirchhoff, G. & Bunsen, R. (1861). Chemische Analyse durch Spectralbeobachtungen. Zweite Abhandlung. Annalen der Physik und Chemie 113: 337–381.
Environmental analysis
Measuring trace metals
Atomic absorption spectroscopy, developed by Alan Walsh in 1955 on Kirchhoff's principle, measures traces of lead, mercury and other metals in water, blood and food, and is a standard tool of environmental and clinical laboratories.
› Sources (1)
- Walsh, A. (1955). The application of atomic absorption spectra to chemical analysis. Spectrochimica Acta 7: 108–117.

Open problems

Where the map runs out

Open

The solar abundance problem

Open as of 2026. Revised abundance analyses have narrowed but not closed the gap.

In the 2000s, improved models of the Sun's atmosphere lowered the measured abundance of carbon, nitrogen, oxygen and other heavy elements. But helioseismology, which probes the Sun's interior through its vibrations, agrees much better with the older, higher values. Either the spectroscopy, the models of the Sun's interior, or the physics of how radiation passes through hot matter is wrong.

Why it is hard: Each input, the three-dimensional modelling of the Sun's turbulent surface, atomic data for thousands of lines, and the opacity of plasma at millions of degrees, is difficult to test independently. Laboratory measurements of iron opacity at solar interior conditions have themselves disagreed with theory.
What resolving it unlocks: The Sun is the reference for measuring every other star's composition, so its abundances set the scale for stellar ages, galactic chemistry and the properties of exoplanet hosts.

› Sources (1)

Asplund, M., Grevesse, N., Sauval, A. J. & Scott, P. (2009). The chemical composition of the Sun. Annual Review of Astronomy and Astrophysics 47: 481–522.