Sir Arthur Eddington smeared the sweat from his brow as he waited on a small island off the coast of Africa for the moon to engulf the sun. Across the Atlantic, two fellow astronomers were casting their eyes skyward near the coast of Brazil, watching for a total solar eclipse making its way along the equator.

This May 1919 eclipse allowed Eddington and his colleagues to measure distant starlight as it passed by the Sun, providing the first piece of hard evidence for a revolutionary theory proposed just four years previous.

A few months before observation day, Eddington recorded the positions of stars sitting directly behind the projected location of the eclipse. On that day in May, if Einstein’s new theory of general relativity was correct, the light from those stars would be diverted by the Sun’s tug of gravity, making it seem as if they had changed their cosmic positions by an ever-so-slight, but precisely predicted, distance.

And that is exactly what they saw. As darkness swept over the equator and stars shone in the middle of the day, the men on two separate continents measured to the same degree a shift in the locations of those stars. Eddington was the first to witness gravitational lensing. Since then, dozens of such phenomena have been catalogued. And now, scientists are using this near-century old idea to probe one of the deepest mysteries in the cosmos.

It began when Einstein introduced a stunning new way to think about gravity and how it works. Space is more than a passive, unchanging box, within which big balls of rock or gas pull back and forth on each other, he said. It’s a pliable fabric that bends and dips in the presence of any matter.

This fabric acts much like a trampoline. Someone standing in the middle makes a depression and the heavier the person, the deeper the dip. The same is true for bodies in space. Jupiter makes a bigger dent than Earth, which makes a bigger dent than the Moon. Even people bend space to a miniscule degree.

Objects moving through space interact with these distortions like a marble on that trampoline. Drop one and it will roll straight down into the toes of the central person. Give the marble a shove along the edge and the tiny ball will start corkscrewing, in wide circles first, then smaller and tighter until it crashes into the middle. This is the Moon orbiting the Earth, the Earth coursing around the Sun. The Moon and Earth have enough forward momentum to keep from spiraling to a crash, but not enough to escape over the edge of the dip and out into the void.

Light, however, is a little different. It whips through space at an unimaginable speed. Nothing can ever catch up to it. It’s fast enough to flit by these massive objects with only a passing nod. It’s as if that marble was flung hard across the surface of the trampoline; it would dip into the sag in space, but it would soon soar out and to the other side, winding up a bit removed from its original path. The amount of the detour depends on how deeply space dips, just as an optical lens curves a light wave’s path according to the contours of its shape (hence the term for this effect, gravitational lensing).

Light from individual stars going by the Sun, as Eddington observed, seems just a little displaced. The light of distant galaxies beaming in unison past the enormous mass of a galaxy cluster, however, gets greatly stretched and distorted, like looking through a fish eye lens. When light skirts around a black hole, it looks to be a giant whirlpool in the sky.

Anything can be a gravitational lenser. It could be our Sun pulling around starlight, or it could be a galaxy bending the light paths of even more galaxies. But the effects are always observable. Modern scientists can do more than just measure the distortion of the light darting through space. Looking at how much the light is stretched and bent can help confirm how much mass the lenser actually has.

This seems straightforward—using gravitational lensing, scientists should record the same amount of matter previously detected through other means. But this isn’t always the case. Some galaxies make dips in space that are simply too large, if all that’s there is the luminous matter scientists observe through a telescope. To them, it looks as if a three-year old child pressed a trampoline all the way to the ground. That doesn’t make sense. There has to be something else there.

Scientists are calling this “something else” dark matter, and they’ve used gravitational lensing, combined with other methods, to determine that about one-quarter of our universe is made up of the stuff. So far, that’s all they really know, and it’s opened up a whole new realm of physics to be explored. And as these 21st century observers search for new ways to see the unseeable, they keep their eyes towards the lensers in attempts to unlock even more secrets of the universe.

This post first appeared on MIT Scope in February 2011