A STRAIGHT line is the shortest distance between two points. So light, which never travels farther than it must, moves in straight lines.  On flat surfaces, this poses no difficulty: only one straight line runs From A to B. 0n curved surfaces, there may be many such lines—which would matter little to astronomers, were it not for Einstein. General relativity explains gravity in terms of curves in the multi-dimensional geometry of space and time. If an intervening mass bends space in the required manner, light setting out from A will move along many different “straight-line’ paths to B. Observers at B see a distorted image of A or multiple images of it, because the mass in between acts as a lens.

Most distortions upset astronomers, as the builders of the Hubble space telescope have found. Yet these gravitational distortions fascinate them. Among other things, they allow a new approach to the central problem of modern astronomy: the fact that, although astronomers see only these things that give off light or some other radiation, theory and circumstance tell them that the universe is dominated by matter which makes no such display, and is thus beyond their ken. Since this “dark matter” has gravitational effects, it will sometimes distort the luminous universe, and thus reveal itself.

Theory has it that the place to go for serious amounts of dark matter is a cluster of galaxies. Such clusters boast stars by the uncounted trillions, and those stars are thought to be but a delicate frosting on the underlying bulk of dark matter, whose existence can be inferred from the movements of galaxies and gases. Now the effects of dark matter can be seen directly, even though the stuff itself remains hidden.

In the past few years various teams have produced images that reveal delicate blue arcs around the edges of clusters—light from the backdrop of extremely faint blue galaxies spread across the sky, smeared into arcs by the cluster’s mass. The shape of these arcs, and their apparent positions, prove that there is indeed far more mass in the clusters than meets the eye.

One such cluster-weighing study is helping researchers to answer a different question: how big is the universe? Beyond the blue galaxies lie quasars, the brightest and most distant lights of heaven. How distant, though, is unknown: the relative distance of quasars and far-off galaxies can be worked out, but their absolute distance is uncertain. Because quasars are a long way away but still bright, they are ideal for lensing—the further away an object is the better the chance of something getting in the way and focusing its light. The challenge is to use this lensing effect to deduce the quasars’ distance—and hence to measure the size and age of the universe they see.

An old friend

To do this, many astronomers have turned to the lens they know best—the one they discovered first. It consists of a faint galaxy flanked by two images of a much more distant quasar. Light and radio waves, following the curve in space made by the galaxy’s mass, can pass around it on one side or the other and still fall into the same telescope mirrors and radio dishes on earth. The two possible paths, though, are not quite the same length. If the galaxy’s lensing properties were known precisely, the difference in their lengths could be expressed as a fraction of the distance to the quasar.

This difference in lengths can be measured. The quasar burns with an inconstant flame, and changes in its brightness reach the earth quicker if they take the shorter path. The time between a change in one image and its echo in the other reveals the difference in the paths’ lengths. Given that, and an understanding of the lens, the distance to the quasar can be worked out.

Unfortunately, such understanding is not given; it has to be taken from a universe unwilling to co-operate. The galaxy between the images: is not heavy enough to account for the lensing on its own. It is receiving help from its friends—it sits near the centre of a cluster of galaxies—but nobody knows how much help. This is where cluster-weighing helps. Recent observations show tell-tale blue areas of lensing around the cluster which speak of the mass within. But even that may be insufficient to make sense of the lens. Related studies have revealed another cluster between the first one and the quasar: this too is probably playing a role. It will be a while yet before the distance to the quasar can be worked out accurately.

Many astronomers would like to work with more straightforward lenses. Some exist, but they can be prone to another problem: “microlensing”. As individual stars in the lensing galaxy move across the light’s path, their gravity can act to focus its light and thus make it brighter. So not all the fluctuations in the brightness of the images are due to changes in the brightness of the source; some are due to secondary lensing by individual stars. That makes delays hard to spot.

Microlensing has is uses, though again in the study of dark matter. Although many theorists like to think that most of the dark matter will be made up of exotic new particles, at least some of it is likely to be more mundane: burnt-out stars, still-born stars and planets. Such things might well make up the haloes of dark matter thought to surround galaxies such as the Milky Way—the galaxy that contains the earth—even if the cores of clusters are made of something weirder. If our galaxy is indeed surrounded by such bodies, they should reveal themselves through microlensing.

Three groups—one French, one Polish-American and one American-Australian—are looking at this possibility. The American-Australian programme, just getting under way, is perhaps the most ambitious. Its eye will be set on the stars of the Magallanic clouds, nearby dwarf galaxies. If something in the halo wanders in front of one of them, its microlensing effect will magnify the star, making it look brighter for a few days.

Even if the halo contains a vast number of dark bodies, alignments precise enough to result in microlensing will be rare. So the programme must keep tabs on 3m stars every night in the hope of catching a few events a year. The resulting torrent of data will be processed in ways more familiar to particle physics than astronomy—which is as it should be, since particle physicists are particularly keen students of dark matter, and the programme is co-ordinated by the University of California’s Centre for Particle Astrophysics.

Closer to home

If the halo search proves a success, the same technique could be adapted to a study of less cosmological importance but perhaps broader appeal. Stars in the Milky Way’s central bulge could be microlensed by nearer stars in its arms; and if those intermediate stars have planets, they should show up, too, as transient additional contributions to the lensing. Such a programme is harder to run than one looking outwards for evidence of dark bodies in the halo, but it may soon be feasible.

If planets are discovered around other suns, they will be as hard to study as moths around a searchlight. But gravitational lensing might solve even this problem. The earth is too close to the sun to make use of its potential as a lens; but if you could stand far enough back, the sun’s gravity might reveal continents on earth-sized planets 20 light-years away. “Far enough back” is a long way: 100 billion kilometres or so, far further than any earthly spacecraft has yet travelled. But in time such observations will be possible. Even if, by then, there are easier ways of producing such images, none will be grander.