Actually - I had the 5m radius orbit thing wrong.....although it is a very popularly held misconception. I've been fortuante enough to swop a few emails with a member of the Sojourner team about this very issue.

Sojourner was supposed to try and get back to the lander, but also avoid a 'keep out zone' - and when reaching that Zone - Sojourner would probably have turned right, nudging it around the lander.....so whilst the net result might have been 'circulation' - that wasn't the specific plan. After a while, Sojourners knowledge of its own location would have drifted and it would have started a crazy wandering path who knows where.

Going on that HiRISE image - if that feature is Sojouner ( and the case is fairly compelling given that it appears to be a Sojourner sized rock where no such rock appears from surface imagery ) - then she probably was heading back to the lander from Chimp rock - but gave out after about 7 metres or so. The hows/whys of that autonomous expedition will have to wait for someone to 'do a kilmer'...but kindly...to pull out whatever data resides within Sojourner.

If a second HiRISE observation (for Stereo and probably colour coverage ) shows that dot to have moved.....then all bets are off :)

Doug