The telescope sits on the lawn outside Vachon Pavilion at Quebec's Laval University, its gently concave 40-in.-diameter mirror pointing at the sky. Concentrating and reflecting faint starlight into a camera mounted above it, the gleaming face of the mirror seems devoid of the slightest imperfection; it is so smooth, in fact, that it looks solid.
But that is an illusion. The mirror is really a pool of liquid mercury in a shallow wood container. A touch would send ripples racing across its surface, and it must always aim straight up to retain its curvature. As the container is slowly rotated on a turntable; making one revolution every six seconds, the mercury rises gently toward the edges and dips in the middle, the way coffee does when it is stirred in a cup. In perfect deference to the laws of physics, the metal's highly reflective surface takes the form of a parabola, the shape of solid mirrors used in conventional telescopes to focus starlight into a sharp image. Says Ermanno Borra, the Laval astrophysicist who built it: "It's a wonderfully simple arrangement."
That arrangement may help astronomy break through a size barrier that it reached in 1948, when technicians completed work on the famous 200-in. glass mirror for the Hale Telescope at Mount Palomar; beyond that size, glass mirrors tend to sag and distort un-acceptably, affected both by their own weight and by changes in temperature. The only larger mirror in the world, a 236-in. monolith atop Mount Semirodriki in the Soviet Union, is apparently hopelessly flawed and has done little significant work since being completed in 1974. One solution to the size problem is to make several smaller mirrors work together, simulating a single large one. The computer-synchronized Multiple-Mirror Telescope atop Mount Hopkins in Arizona, for example, has six mirrors and the light-gathering power of a 176-in. instrument. But Borra claims that his mercurial approach will make possible virtually flawless single mirrors at least five times as big as the one at Palomar and can do it simply and inexpensively.
Borra emphasizes that the concept of a liquid-mirror telescope is not new; he thinks the idea may have occurred to Isaac Newton, who knew about the behavior of spinning fluids and built one of the first reflecting telescopes. Borra knows that Robert Wood of Johns Hopkins University built a primitive model in 1908. "I am not the inventor," says Borra. "But I am the first to make it work and the first to know what to do with it."
Wood's problem was that the motors of his day could not turn at a constant enough rate of speed. As a result, the curvature of his mirror kept changing. Also he was unable to avoid vibrations, which set up ripples in the metallic pool. Wood was aware of another shortcoming of his telescope: because it always had to face straight up, it could not be swung around to point at interesting stars and galaxies or to take time-exposure photographs by following the celestial objects across the sky as the earth rotated.