A problem that struck the Hubble Space Telescope on Saturday will delay the final space shuttle mission to service it, moving the launching from next month to next year, NASA officials said Monday.

A crew of seven astronauts was scheduled to blast off in the shuttle Atlantis on Oct. 14 for an 11-day visit to the telescope, which for 18 years has been beaming cosmic postcards to Earth from its orbital vantage point above the atmosphere.

During five spacewalks, the astronauts were set to install two new instruments and repair the telescope’s best camera and a spectrograph, both of which had electrical failures. They were also scheduled to replace the telescope’s batteries and gyroscopes, among other things.

But on Saturday, a channel on a control system known as the Hubble Control Unit/Science Data Formatter — which helps relay data to the ground — failed, causing the telescope to go into a “safe mode” and cease observations. Hubble’s managers expect that activating a backup channel will restore the telescope to service later this week.

But that will leave the telescope with no backup if the new channel stops working, so NASA would like to have the astronauts replace the failed control unit with a spare from the Goddard Space Flight Center.

In a telephone news conference with reporters on Monday evening, Preston Burch of Goddard, Hubble’s program manager, said the control unit hangs, attached by 10 bolts, on the inside door of a bay that the astronauts can access easily. With luck, it could be exchanged during a two-hour spacewalk, he said. “We think it’s a relatively straightforward activity.”

It is too soon to tell, the mission managers said, whether replacing the control unit will bump another activity from the servicing schedule. The mission’s five spacewalks are tightly packed with activities, but the lead astronaut, John M. Grunsfeld, has been able in training to complete the camera repair in one spacewalk instead of the scheduled two.

“This may be a doable thing, that we can have our cake and eat it too,” Mr. Burch said.

Understanding what went wrong, testing the spare unit, integrating its installation into the mission schedule and training the crew to install it will take several weeks or more, Mr. Burch said.

The Hubble mission cannot be launched until another shuttle, the Discovery, which is scheduled for a Feb. 12 trip to the space station, is ready to serve as the backup rescue shuttle. As a result, the flight will not take place before February 2009.

DENNIS OVERBYE

Mars Craft Detects Falling Snow

Soil Tests Also Hint at Past Presence of Liquid Groundwater

Icy snow falls from high in Mars's atmosphere and may even reach the planet's surface, scientists working with NASA's Phoenix lander reported yesterday.

Laser instruments aboard the lander detected the snow in clouds about 2 1/2 miles above the surface and followed the precipitation as it fell more than a mile. But because of limitations with the technology, it was unclear whether any of the powdery stuff made it all the way to the surface.

"Nothing like this view has ever been seen on Mars," said Jim Whiteway of York University in Toronto, lead scientist for the Canadian-supplied Meteorological Station on Phoenix. "We'll be looking for signs that the snow may even reach the ground."

In addition to finding snow, the Phoenix team reported discovering material in the Martian soil that had once been dissolved in water -- clays and calcium carbonate (limestone) that could have formed only in the presence of liquid water. Although the lander's instruments earlier found water ice below Mars's polar surface and had photographed surface fog and clouds, it has found nothing like liquid water on the surface.

The presence of nutrients and other material that once dissolved in water, however, plus the continuing presence of water as snow, vapor and ice, is leading researchers to conclude that Mars's polar regions might have supported life in the past -- when the region was much warmer. Because Mars wobbles on its axis far more than Earth does -- in some very long-term cycles, the poles face the sun -- the northern region where Phoenix landed has, in the past, been warm.

"Is this a habitable zone on Mars? I think we are approaching this hypothesis," said principal investigator Peter Smith of the University of Arizona.

In addition, Michael Hecht of NASA's Jet Propulsion Laboratory said that further analysis of the Martian dust by the lander's onboard laboratory has determined that it is as alkaline as ocean water, with a pH of 8.3. He said this finding also suggests that life could have existed on Mars.

Whiteway said the snow, along with frost and fog, began to appear about a month ago, as temperatures cooled on Mars. "This is now occurring every night," he said.

In an interview after the teleconference, Whiteway likened the snow to "diamond dust" that falls in the Arctic and Antarctica.

"What this is telling us is that water does rise from the ground to the atmosphere and then precipitates down," he said. "So there is a hydrological cycle on Mars, and now other experts will study the data and try to determine what it all means."

Although the Phoenix instruments could not determine whether the snow hit the ground, Whiteway said there are some indications that it does. Images of the thin but distinct Martian clouds can be seen on the NASA Web site at http://www.nasa.gov/phoenix.

With daylight quickly diminishing as the Martian winter starts, the Phoenix is not expected to continue operating for many more weeks. The spacecraft has a "Lazarus" feature that could return it to operation when the sun returns, but the brutally cold temperatures during the winter are expected to freeze and crack parts essential to its operation.

Smith said that in the remaining weeks, the scientists plan to turn on a microphone that was designed to record the lander's descent in May. It did not work then, but Smith said, "We are going to try to turn on this microphone and try to listen to Mars for the first time."

Phoenix was scheduled to operate for 90 Martian days, known as sols, but the lander's robotic arm has been digging up soil and ice for more than 120 sols and delivering it to chemistry labs inside the lander. Smith said many of the mission's primary goals have been accomplished, although difficulties transferring the scooped-up material into the lander have interfered with some experiments.

In particular, researchers are eager to know whether the isotopic makeup of the water in the air is the same as the water in the soil, but they have not been able to load the proper material to find out. They have searched in vain for organic (although not necessarily biological) material in the soil.

Scientists have theorized that snow falls on Mars, but they had never before seen it in real time. Future research on the data collected by Phoenix will try to determine where the snow came from -- whether it originated in the ice-covered polar regions or evaporated from the broader Martian surface, or even from the large collections of ice below the surface.

The Phoenix team was surprised in the summer by the presence of the chemical perchlorate in the Martian soil. Used in many industrial capacities and in rocketry, the potentially toxic chemical is also found in the driest deserts on Earth. Smith said its presence on Mars suggests again that there was once liquid groundwater and raises the possibility that life has existed in the planet's seemingly hostile environment.

Because perchlorate changes the freezing point of water dramatically, it could keep water in a liquid state at temperatures of 76 degrees below zero.

"This could form brines for microbes, which could then use the perchlorate as a chemical energy source," he said.

Marc Kaufman