That earlier scheme was a wildly over optimistic, underfunded rush to a Mars sample return attempt in 2005, which would have probably collapsed through its sheer implausibility even if the 1998 missions had been totally successful.

The revised program is now well defined through 2005 - consisting of two fairly long-range, sophisticated rovers in 2003, and an orbiter in 2005 equipped with an extremely high-powered telescopic camera (among other instruments) and capable of transmitting data back to Earth 12 times faster that Mars Global Surveyor now in Mars orbit can do. Beyond, 2005, however, the program remains flexible while funding positions are more established.

The Bush Administration's proposed budget provides NASA with a hoped-for infusion of money which would allow it to fly a very sophisticated mission in 2007: a big "smart lander" which would test systems allowing soft landers to land within a few kilometers of their target point, detect and dodge dangerously rough or sloped terrain during their actual descent, and survive a very rough landing.

The lander would then deploy a long-range rover that would travel dozens of km over Mars' surface for up to a year, analyzing the surface with as much as 300 kg of scientific instruments.

It would also serve as the test flight for the actual Mars sample-return lander in 2011, which would use a similar rover to collect samples that would then be launched back into Mars orbit by a small rocket where a small container would be ejected to be scooped up by a joint French-American orbiter before blasting out of Mars orbit for the return trip to Earth with its extremly precious Mars samples.

Howqever, given NASA's continuing budgetary woes it's quite possible that Congress may reverse Bush's proposed boost to Mars funding. Indeed, the U.S. Senate is currently trying to cut $50 million out of this year's funding for the program.

This would not be a catastrophe, though - for, even without Bush's boost, the Smart Lander test flight and the first sample return mission would be delayed two years, until 2009 and 2013.

In any event, though, there will definitely be at least one other 2007 U.S. Mars mission - the first Mars Scout mission, a lower cost $300-million mission - essentially a Mars-specialized version of NASA's current Discovery program for general Solar System exploration.

The Mars Scout missions - to be launched every four years - would be selected by an appraisal board from a set of proposals provided by separate competing teams of scientists and engineers.

The selection process for that first Scout mission has already begun with a May workshop in Pasadena where 39 proposals were submitted by various scientific teams.

These were not official submissions - they were, instead, provided as a kind of "practice run" to see what kind of proposals the scientific community is likely to provide when the official Announcement of Opportunity is released next February, and to "identify achievable science objectives and technology development priorities".

The best ten proposals, however, were chosen for 6-month study contracts costing the government $100,000 to $150,000 each - and they will undeniably be in a relatively favorable position when the actual finalists for the Scout mission are chosen in May 2003 (with the final choice being made in January 2004).

This still doesn't make it even probable that one of these ten favored proposals from the Pasadena workshop will be chosen. After all, a similar workshop was held at the start of the Discovery program in 1992 - and among the eight Discovery missions since chosen for flight, only two have come out of the 13 proposals chosen for extended study after that workshop.

But these 10 proposals do provide us with a glimpse of the kinds of missions that will actually be submitted - and eventually chosen - for the 2007 Mars Scout mission.

The actual structure of the Mars Scout program - the range of missions that will be considered for it - has become more flexible since the revised U.S. Mars program was first announced last year.

At that time, the primary role of the Scouts was seen as that of a flock of little, low-cost reconnaissance vehicles to check out various possible landing areas on Mars, both for their landing safety and for their scientific interest, to see if they would be worthwhile targets for later, bigger and far more expensive landers, including the sample return missions.

These vehicles might be durable little hard landers like the 30-kg Beagle 2 to be carried on Europe's 2003 Mars express mission, and Pathfinder in 1997 - which would not only examine an area after landing, but also take photos of it on the way down.

Or they might be aerial vehicles - balloons, gliders, or even powered airplanes - which would use cameras and other instruments to examine Martian terrain from above, in more detail than orbiters can. But while they would be free to carry other instruments, their primary role really would be that of "scouts" for later, more expensive - and less expendable - landers.

However, at its March meeting, NASA's Space Science Advisory Committee (SSAC) reexamined the Mars Scout program and decided to make it far more flexible in structure:

"The Scout program should permit all missions to Mars space that fit within the cost caps and schedule constraints of the Program. In addition to orbital and landed missions to Mars, missions should be permitted that focus primarily on Phobos and Deimos, the upper atmosphere, or 'network science' [long-lived landers for seismic and weather measurements]. Any of these would complement the missions that are part of the main Mars program."

In short, NASA's science committee advised that the Mars Scouts be turned into a Mars-focused version of the Discovery missions, in which virtually any kind of mission is acceptable if the appraisal board thinks that its scientific return will be worth its cost.

Indeed, several Discovery missions have previously been proposed to study Mars - and one, the "Aladdin" mission to return material from Mars' moons to earth at low cost, has been a finalist twice.

The feeling now seems to be that suitable landing sites for the earliest sample return missions can be selected on the basis of the information we get from the current and future Mars orbiters, without the need for detailed on-the-spot inspection of a large number of possible sites.

Sure enough, the 43 proposals offered up at the May workshop - and the 10 selected for detailed study - are a wildly mixed bag.