Henrietta Leavitt’s ruler

At the turn of the twentieth century, Harvard College Observatory was photographing the sky faster than anyone could study it. Its director, Edward Pickering, hired women to examine the glass plates: to measure the brightness and position of stars, classify their spectra and record the results in ledgers. They were called computers, and they were paid a fraction of what a male assistant would earn. Most received about 25 cents an hour. Leavitt’s work was valued enough to earn her 30.

Henrietta Swan Leavitt joined in 1895, first as a volunteer, then as a paid employee. She had studied at Radcliffe, then known as the Society for the Collegiate Instruction of Women, and had taken a course in astronomy. Pickering assigned her to variable stars, which change in brightness over time.

Stars at the same distance

Leavitt studied plates of the Magellanic Clouds, two small galaxies visible from the Southern Hemisphere, taken at Harvard’s station in Peru. Comparing plates taken on different nights, she found variable stars by the hundred. In 1908 she published a catalog of 1,777 of them, and added a remark that reads now like a held breath: “It is worthy of notice that in Table VI the brighter variables have the longer periods.”

The problem she was working around is one of the oldest in astronomy. A faint star might be a dim one nearby or a brilliant one far away, and looking at it cannot tell you which. Leavitt’s insight relied on a shortcut that was available only because of where she was looking. All the stars in the Small Magellanic Cloud are at roughly the same distance from Earth, so differences in their apparent brightness reflect real differences in how luminous they are. The cloud was, in effect, a laboratory in which distance had been held constant.

In 1912 she showed, using 25 of these stars, that their period and brightness follow a tight relationship. Her paper says it without ceremony: a straight line can be readily drawn among each of the two series of points. On her plates, a star whose period is ten times longer is about two magnitudes brighter, which works out at roughly seven times the luminosity.

Measure how long a Cepheid takes to pulse, and you know how bright it truly is.

How the ruler works

Measure how long a Cepheid takes to pulse, and you know how bright it truly is. Compare that with how bright it appears, and the inverse-square law gives its distance: if one Cepheid appears a hundred times fainter than another with the same period, it is ten times further away. No other measurement of distance at that time could reach so far.

It needed one more thing. Leavitt’s law gave relative distances only: this star is ten times further than that one. To make it absolute, somebody had to find the true distance to at least one Cepheid, and none was close enough for easy measurement. She knew this, and wrote that she hoped the parallaxes of some variables might be measured. That task of calibration has never really ended, and it is precisely where today’s fight over the expansion rate is being conducted.

A ruler for the universe

The result was published as a Harvard College Observatory circular under Pickering’s name, with a note that the work had been prepared by Miss Leavitt. Within a year, Ejnar Hertzsprung used it to estimate the distance to the Small Magellanic Cloud. Harlow Shapley used Cepheids to map the size of the Milky Way. In 1923, Edwin Hubble found a Cepheid in the Andromeda nebula and showed that it lay far outside our galaxy, settling a long debate about whether other galaxies existed at all. Hubble’s letter to Shapley reporting the discovery is often said to have prompted Shapley’s remark that this was the letter that destroyed his universe.

The ruler still had surprises in it. In the early 1950s, Walter Baade realized that there are two distinct populations of these pulsating stars, and that astronomers had been mixing them up. Correcting the error roughly doubled the scale of the universe overnight, and with it the size of the Andromeda galaxy and every distance built upon it.

Every one of those results depended on Leavitt’s relationship. It remains the first rung of the cosmic distance ladder, and astronomers increasingly call it the Leavitt law.

Credit, late

Leavitt was often ill and partly deaf, and she was not encouraged to pursue the theoretical questions her discovery raised; Pickering preferred his computers to measure, not to interpret. A century of hindsight makes it tempting to cast this as simple exclusion, and it was partly that. But the hierarchy also did something more specific. It drew a line between the person who finds a pattern and the people entitled to say what it means, and then placed the line so that Leavitt fell on the wrong side of it. Hertzsprung and Shapley drew conclusions from her law that she had been in a better position than anyone to draw.

She died of cancer in December 1921, at the age of 53. Four years later, the Swedish mathematician Gösta Mittag-Leffler wrote to Harvard to propose nominating her for a Nobel Prize, unaware that she had died. The prize is not awarded posthumously.

Harvard’s plate collection, roughly half a million glass negatives of the sky, still exists, and a project to digitize it has made it available to a new generation of astronomers. The plates Leavitt worked from are in it, some still bearing the ink marks the computers made on the glass. The later scanners of bubble chamber film inherited something of her position, skilled, essential and rarely named. The archive is a strange kind of monument: a ruler she built from glass, light and patience, which the rest of astronomy has been measuring with ever since.

Sources

• Leavitt and Pickering, “Periods of 25 Variable Stars in the Small Magellanic Cloud”, Harvard College Observatory Circular 173, 1912
• Leavitt, “1777 variables in the Magellanic Clouds”, Annals of Harvard College Observatory, 1908
• The Marginalian on Dava Sobel’s The Glass Universe