IT USED TO be that finding even one new planet was enough to make an astronomer’s career. Uranus was discovered in 1781 by William Herschel, who these days has, among other things, space telescopes, asteroids, schools and a street in Paris named in his honour. Urbain le Verrier, who predicted the existence of Neptune in 1846, based on its gravitational influence on Uranus, has likewise given his name to craters, asteroids and bits of the French capital. It is one of 72 engraved into the sides of the Eiffel tower.
These days, though, astronomers can do better. William Borucki has thousands to his name. He is the researcher who conceived of and ran Kepler, a planet-hunting space telescope that was launched in 2009. On October 30th NASA announced that, after nearly a decade in space, Kepler had run out of fuel and would be retired. Kepler has discovered around 2,600 exoplanets—those that orbit stars other than the sun. Another 3,000 candidates await confirmation from ground-based telescopes. The result has been a revolution in astronomy. Its practitioners had long assumed that other stars were likely to have planets of their own. These days, they know that to be true.

Get our daily newsletterUpgrade your inbox and get our Daily Dispatch and Editor’s Picks.

The first exoplanets were detected in 1992, thanks to the gravitational effect they had on the pulsar around which they orbited. (A pulsar is the dead, ultra-dense remnant left over after a supernova.) A trickle of subsequent discoveries followed.
Kepler transformed that trickle into a flood. Yet it almost did not happen. Dr Borucki proposed the mission four times to NASA before it was accepted in 2001. Rather than look for stellar wobbles caused by a planet’s gravity, he suggested monitoring a star’s light itself—looking for tiny dips in brightness as planets, if any, crossed in front of the stellar disc. It is a simple idea. But it required the construction of a light detector 1,000 times more sensitive than anything that had been built before.
The advantage of this “transit” method is that it is well-suited to mass production. Kepler was designed to stare fixedly at a single patch of sky in the constellation of Cygnus, observing around 150,000 stars simultaneously. The consequential torrent of data can be used to draw statistical conclusions about the rest of the galaxy. It seems likely that every one of the Milky Way’s hundreds of billions of stars sports at least one planet.
Many of these are of a type unknown in the solar system. The most common in Kepler’s data are “super-Earths”—rocky worlds intermediate in size between Earth and Neptune. Kepler also helped prove that “hot Jupiters” are common. These are gas giants which orbit implausibly close to their stars. Theorists, who would previously have argued that such planets were impossible, are still debating whether they are able to form in situ or whether they coalesce farther out in their stellar systems and then migrate inward, knocking other planets into deep space as they do so in a game of planetary billiards.
Kepler’s biggest quarry was Earthlike planets at just the right distance around their stars for liquid water to exist on their surfaces. Several have turned up, but the search became harder in 2013, by which time two of the four gyroscopes that kept the telescope stable had broken. Kepler’s engineers came up with an ingenious fix, relying on the radiation pressure exerted by sunlight to re-steady the craft.
With its fuel depleted, no technological rescue is possible this time. But nothing succeeds like success. Exoplanets are now the hottest topic in astronomy. The Transiting Exoplanet Survey Satellite, launched in April, is a Kepler-like mission designed to look for exoplanets in Earth’s immediate neighbourhood. These might be amenable to detailed follow-ups from other, more powerful telescopes.
In theory, it should be possible to sniff the air of such neighbours for signs of life, or even make rough maps of their surfaces. Two similar European missions are scheduled to launch in 2019 and 2026. Exoplanetology is just getting started.