The results of the search were announced by a group led by astronomer Gary Bernstein of the University of Pennsylvania at today's meeting of the Division of Planetary Sciences in Monterey, Calif.

The study's big surprise is that so few Kuiper Belt members were discovered. With Hubble's exquisite resolution, Bernstein and his co- workers expected to find at least 60 Kuiper Belt members as small as 10 miles (15 km) in diameter -- but only three were discovered.

"Discovering many fewer Kuiper Belt objects than was predicted makes it difficult to understand how so many comets appear near Earth, since many comets were thought to originate in the Kuiper Belt," Bernstein says. "This is a sign that perhaps the smaller planetesimals have been shattered into dust by colliding with each other over the past few billion years."

Bernstein and his colleagues used Hubble to look for planetesimals that are much smaller and fainter than can be seen from ground-based telescopes. Hubble's Advanced Camera for Surveys was pointed at a region in the constellation Virgo over a 15-day period in January and February 2003. A bank of 10 computers on the ground worked for six months searching for faint-moving spots in the Hubble images.

The search netted three small objects, named 2003 BF91, 2003 BG91, and 2003 BH91, which range in size from 15-28 miles (25-45 km) across. They are the smallest objects ever found beyond Neptune. At their current locations, these icy bodies are a billion times fainter (29th magnitude) than the dimmest objects visible to the naked eye. But an icy body of this size that escapes the Kuiper Belt to wander near the Sun can become visible from Earth as a comet as the wandering body starts to evaporate and form a surrounding cloud.

Astronomers are probing the Kuiper Belt because the region offers a window on the early history of our solar system. The planets formed over 4 billion years ago from a cloud of gas and dust that surrounded the infant Sun. Microscopic bits of ice and dust stuck together to form lumps that grew from pebbles to boulders to city- or continent-sized planetesimals. The known planets and moons are the result of collisions between planetesimals. In most of the solar system, all of the planetesimals have either been absorbed into planets or ejected into interstellar space, destroying the traces of the early days of the solar system.

Around 1950, Gerard Kuiper and Kenneth Edgeworth proposed that in the region beyond Neptune there are no planets capable of ejecting the leftover planetesimals. There should be a zone, the two astronomers said --- now called the Kuiper Belt -- filled with small, icy bodies. Despite many years of searching, the first such object was not found until 1992. Since then, astronomers have discovered nearly 1,000 from ground-based telescopes. Most astronomers now believe that Pluto, discovered in 1930, is in fact a member of the Kuiper Belt.

Astronomers now use the Kuiper Belt to learn about the history of the solar system, much as paleontologists use fossils to study early life. Each event that affected the outer solar system -- such as possible gravitational disturbances from passing stars or long-vanished planets -- is frozen into the properties of the Kuiper Belt members that astronomers see today.

If the Hubble telescope could search the entire sky, it would find perhaps a half million planetesimals. If collected into a single planet, however, the resulting object would be only a few times larger than Pluto. The new Hubble observations, combined with the latest ground-based Kuiper Belt surveys, reinforce the idea that Pluto itself and its moon Charon are just large Kuiper Belt members. Why the Kuiper Belt planetesimals did not form a larger planet, and why there are fewer small planetesimals than expected, are questions that will be answered with further Kuiper Belt studies. These studies will help astronomers understand how planets may have formed around other stars as well.

The new Hubble results were reported by Bernstein and David Trilling (University of Pennsylvania); Renu Malhotra (University of Arizona); Lynne Allen (University of British Columbia); Michael Brown (California Institute of Technology); and Matthew Holman (Harvard-Smithsonian Center for Astrophysics). The results have been submitted to the Astronomical Journal for publication, and a preliminary report is available on the Web at http://arxiv.org/abs/astro-ph/0308467.

Electronic images and additional information are available at

• http://hubblesite.org/news/2003/25
• http://www.upenn.edu/pennnews/news.php

The Space Telescope Science Institute (STScI) is operated by the Association of Universities for Research in Astronomy, Inc. (AURA), for NASA, under contract with the Goddard Space Flight Center, Greenbelt, MD. The Hubble Space Telescope is a project of international cooperation between NASA and the European Space Agency (ESA).