Such a rover's scientific return -- and the speed with which it can be driven to new spots daily -- obviously greatly depends on how fast the communications link with it is, and how long that link with Earth operates each day.

Every scientific Mars orbiter from now on will do double duty as a communications relay satellite -- but this really isn't enough, especially since mapping orbiters in polar orbits will only fly over a Mars lander briefly twice each day.

This is where Italy comes in. NASA has tentatively dropped its plan to launch a whole set of tiny "Mars Micromission" comsats piggyback on European Ariane 5 missions.

Instead, the new plan is for NASA and the Italian Space Agency to co-build a single medium-sized comsat which will be launched to Mars in 2007 as one of several payloads on an Ariane 5, and which will be put into a high-altitude 4200-km circular orbit inclined 55 degrees to Mars' equator -- and retrograde. (This orbit maximizes the amount of coverage a relay orbiter can provide for landers at various Martian latitudes.) The 2007 rover will be one of its first targets.

A separate 2007 U.S. mission will be the "Mars Scouts", a set of little Martian craft -- perhaps lightweight landers, perhaps balloons or gliders or even powered airplanes -- which will check out the scientific characteristics and landing hazards of a wide variety of possible landing sites for future Mars missions.

They may also carry out individual focused scientific studies of Mars themselves. The design of this mission is completely up in the air at this point -- in fact, the plan is for it to be picked in just the same way as the missions in NASA's "Discovery" program.

As many as 15-20 possible concepts for the Mars Scout mission will be developed by independent scientific teams starting next year; an official Announcement of Opportunity will be issued in 2002, and an appraisal team will select one of the competing proposals in April 2003. The Scouts, too, will depend on the various Mars relay comsats for their contact with Earth.

Finally, France will likely also take advantage of the 2007 opportunity. NASA's new plan is to use scientific Mars missions to also test out the technology needed for future missions whenever possible -- and the 2007 smart lander is the most important example, for the same lander and long-range rover design will be used for the first Mars sample return mission, set for 2011.

It's by no means certain that NASA will still use the previously planned technique for Mars sample return -- in which a small two-stage rocket launches the tiny sample container into low Mars orbit, and a French-built orbiter equipped with U.S. orbital rendezvous equipment chases it down, docks with it, and then uses its own onboard engine to blast out of Mars orbit and return to Earth with it.

The new sample-return technique won't be selected for a year, and it may involve some alternative technique -- such as the lander manufacturing fuel out of Mars' atmospheric gases for a larger Mars ascent rocket which would blast the sample-return spacecraft directly back to Earth.

But there is a very good chance that NASA will decide to stick with the previous U.S./French Mars orbital rendezvous scheme -- in which case it's now thought crucial that a preliminary 2007 test flight of the French sample-retrieval orbiter should also be made in 2007. CNES (the French space agency) would launch and control this orbiter itself, which would test two more new technologies absolutely crucial for the mission.

The first -- in order to minimize the orbiter's weight -- is "aerocapture", which avoids the need for an orbiter to carry a huge supply of fuel simply to brake itself by several thousand km per hour to enter not only an orbit around Mars, but a low circular one.

Instead, the French orbiter, on nearing Mars, will eject a small cruise stage with solar panels -- and the main orbiter, folded up behind a heat shield, will then deliberately make a low skimming dive into the upper layers of Mars' atmosphere down to altitudes as low as 50 km, thus braking itself by thousands of km per hour that way before skimming back out of the atmosphere, ejecting its heat shield, and using small onboard rocket engines to raise its periapsis back out of the atmosphere, leaving itself in a circular orbit only a few hundred km up.

Needless to say, this maneuver requires a precise onboard guidance system to control the craft's atmospheric entry path -- in fact, almost exactly the same kind of system required for a precision landing by a Mars lander -- and such a system must be tested.

The other new technology would be a rehearsal of the orbiter's actual rendezvous and docking with the tiny orbiting Mars sample return container, which requires homing in first using a radar system and then carrying out a precision docking with the capsule (probably using laser sensors).