Artificial Wasteland · the verification venue

The Blink That Measures the Universe

Some stars breathe. They swell and dim on a clock so steady you could set a watch by it — and in 1912 a deaf, underpaid woman at Harvard proved that the rate of the blink tells you the star's true brightness. Compare true brightness to how bright it looks, and you have its distance. This is the first rung of the ladder that measured the size of the universe. Here is her discovery, rebuilt from her own numbers — and then a stopwatch you can point at another galaxy.

A star that looks faint is either dim, or far. For most of history there was no way to tell the two apart — no way to know whether a smudge of light was a nearby cloud or an island universe a million times more distant. Henrietta Swan Leavitt broke the deadlock with a class of pulsating stars called Cepheids, and a table of twenty-five of them.

1 · Her twenty-five stars

Leavitt catalogued 1,777 variable stars in the two Magellanic Clouds — dim, crowded fields at the edge of the southern sky. Twenty-five of them, all in the Small Magellanic Cloud, had measured periods. Because they sit in the same cloud, they are all at very nearly the same distance from us: any star that looks brighter than another really is brighter. That single fact turns a photograph into a laboratory.

Below are her actual twenty-five, each plotted twice: its brightest point (gold) and its faintest (blue), against its period. On a straight period axis the points sweep a curve. Press “log period.” Watch the curve become a law.

Leavitt's Figure 1 (linear period) and Figure 2 (log period), reproduced from Table I of Harvard College Observatory Circular 173 (1912). The two straight lines are least-squares fits to her own data; they reproduce the residuals she tabulated by hand to about 0.04 magnitudes.

On the logarithmic axis the twenty-five stars fall onto two clean, parallel lines. Leavitt saw it immediately, and stated the slope in one sentence that founded a science:

“A straight line can be readily drawn among each of the two series of points corresponding to maxima and minima… The logarithm of the period increases by about 0.48 for each increase of one magnitude in brightness.” — H. S. Leavitt, Harvard Circular 173, March 1912

Refit her numbers today and the slope comes back at 0.496 (maxima) and 0.488 (minima) in log-period per magnitude — her “about 0.48,” recovered a century later from the same twenty-five dots.

2 · Why the blink knows

A Cepheid is a star ringing. A deep layer of helium acts as a valve: when it ionizes it traps heat and the star swells; the swollen star cools, the helium recombines, the valve opens, the star falls back — and it rings again, brightening fast and fading slow, over and over. The pitch of that ringing depends on the star's mean density, and a bigger, more luminous star is puffier and less dense — so it rings slower. Long period means low pitch means a genuinely bigger, brighter star. Drag the period and watch a Cepheid breathe.

The light curve traces the characteristic Cepheid shape — a fast rise to a brief maximum, a slow decline — exactly as Leavitt described them: “diminishing slowly in brightness, remaining near minimum for the greater part of the time, and increasing very rapidly to a brief maximum.” Longer period ⇒ brighter star ⇒ slower blink. (Reduced-motion browsers get a still light curve.)

3 · The stopwatch becomes a ruler

Here is the whole trick, in three steps you can now run yourself. Time the blink → the law hands you the star's true brightness (its absolute magnitude). Read how bright it looks (its apparent magnitude). The gap between true and apparent is pure distance. That gap is called the distance modulus, and it is just:

distance = 10^{(m − M + 5) / 5} parsecs

Leavitt's 1912 law gave the slope but not the zero-point — she had no distance to any single Cepheid, and said so, hoping “the parallaxes of some variables of this type may be measured.” They were (Hertzsprung 1913; later Shapley, Baade, and the Hubble Space Telescope). The ruler below uses that modern calibration. Point it at Andromeda.

Absolute magnitude from the modern classical-Cepheid relation M_V = −2.43(log P − 1) − 4.05 (Benedict et al. 2007, HST parallaxes). A Cepheid like the one Edwin Hubble found in Andromeda in 1923 — period ≈ 31 days — reads out at ≈ 2.5 million light-years: the measurement that proved the “spiral nebulae” were other galaxies, and that the universe is unimaginably larger than the Milky Way.

4 · The name on the paper

The circular that announced all this opens: “The following statement regarding the periods of 25 variable stars in the Small Magellanic Cloud has been prepared by Miss Leavitt.” Three pages later it is signed Edward C. Pickering — the observatory's director, Leavitt's boss. The discovery is hers; the signature is his.

Leavitt was one of the Harvard “computers,” women hired to measure photographic plates by hand at about thirty cents an hour. She was deaf. She measured brightness for thousands of stars with a precision the men who ran the observatory set her to and then, once, saw straight past the task to the law underneath it. Edwin Hubble, who used her law to prove the universe expands, said she deserved the Nobel Prize. In 1925 the mathematician Gösta Mittag-Leffler wrote to nominate her — and learned she had died of cancer four years earlier, at fifty-three. The prize is not given posthumously. Every galaxy distance in every textbook still rests on the rung she cut.