NASA's joyless Mars team was poised to try one last-ditch effort between midnight and dawn Tuesday to contact its missing lander on the Red Planet.

Hopes had all but faded that they will recover the Mars Polar Lander or the two microprobes it was carrying when the craft entered the Martian atmosphere Friday, handlers admitted early today. Thoughts were already turning to the future and engineers are reassessing the design of upcoming missions.

"Tomorrow night [Tuesday morning EST] will be pretty much the last high-probability chance that we'll have," flight operations manager Richard Cook told reporters at a 2 a.m. briefing today here at the Jet Propulsion Laboratory, which manages the Mars missions for NASA.

If that effort fails, it will almost certainly leave the team in the previously unthinkable position of having lost three landers and one orbiter--the entire $360 million panoply of robots aimed at Mars this year--in what was to have been a vital second wave of research in a sequenced exploration of Mars.

The team, working carefully through a series of branching hypotheses about the missing Lander's condition, has eliminated almost all the most likely single-problem scenarios, Cook said. "Tonight's opportunity is actually a credible single-fault case"--the only such one left, he added.

After that, the assumptions take the engineers deeper into complexity and rapidly lower in probability of success, according to flight operations manager Sam Thurman, though efforts will continue for a week or two before the mission is formally pronounced deceased.

Tonight's exercise would again employ the spacecraft's omni-directional UHF antenna, which does not require accurate pointing. It was to transmit to NASA's orbiting Mars Global Surveyor as it passed 250 miles overhead, carrying an instrument that would act as a relay.

The same gambit failed Sunday, but in the meantime, controllers have sent the lander new commands that would switch it to a different "mode" that could produce better results.

Handlers decided to postpone until Tuesday night a major full-sky search--the largest such scan so far--originally planned for tonight. It would use a totally separate communications path: the spacecraft's main antenna, a medium-speed dish that has to be accurately pointed. Setting up the scan was taking longer than expected, Cook said, adding that there is not much hope for success with this approach in any case.

Managers of the two experimental penetrating microprobes also heard nothing but silence through another long night. The probes, which hitchhiked aboard the Polar Lander on its 416-million-mile journey, were supposed to have separated as the three craft entered the Martian atmosphere about 3 p.m. Friday. They were to slam into the surface at 400 mph, send back a signal and prospect for subsurface water ice.

Not a peep has been heard from any of the craft since Friday.

As NASA braced for the fallout, planetary experts were raising questions about whether the push to make missions cheaper has gone too far and whether the degree of risk accepted on this mission was too high. Some were focusing on the specifics of how to prevent such problems on the next missions in the pipeline to Mars. They have so little information from the spacecraft, their dilemma is figuring out how to cure a "disease" that could remain undiagnosable.

The team plans to position the orbiting surveyor, launched in 1996, to point toward the lander's supposed point of touchdown. Although the 3.5-foot-tall craft verges on being too small to be detected by the camera 250 miles up, the orbiter might be able to see the long shadow the lander casts in the light of the "midnight" sun, low in the skies of the polar summer.

If the lander is at that spot, engineers would then know that the parachute, retro-rockets and other crucial components during the complex descent had worked properly and could be checked off the list, said project scientist Richard Zurek. But flight controllers said that process could take a couple of weeks.

The orbiter orbits at almost two miles per second, Cook noted, and "it's not easy to arrange to point it. If you miss by one second," the craft is two miles off.

There was also considerable speculation about the potential perils of the landing area, whose detailed properties cannot be determined from orbital imagery. Among the dangers: baby-fine dust that could have coated the spacecraft, soil so loose the craft sank into the surface or booby traps like those found on melting glaciers, engineers said.

"We're not trying to make excuses for why we failed," said Mars scientist David Crisp. "We don't want to say 'blame Mars.' We just want to find out what happened."

CAPTION: Randii Wessen, a science systems engineer at the Jet Propulsion Laboratory, stands near model of the Mars Polar Lander while discussing lack of signals.