The object would have to be at least as massive as Jupiter to create a gravitational disturbance large enough to give rise to the apparent effect on the Oort cloud, but currently favoured theories of how the solar system formed cannot easily explain the presence of a large planet so far from the Sun. It may have been captured into its present orbit after the solar system formed, even though the probability of such an event seems low on the basis of current knowledge. Though a large, distant planet is a fascinating possibility, other possible explanations for the observed clustering of the comet orbits can't be ruled out, though.

Meanwhile the details about what went wrong have become clearer: At fault was the thruster calibration table of the navigation software on the ground that was used to determine the position of the spacecraft. According to AW&ST, it "converted the amount of thruster firing, or 'counts', reported by telemetry into impulse delivered to the spacecraft. The JPL software expected the units of impulse per thruster count to be in Newton-sec., but the Lockheed Martin-provided table numbers were in lbf.-sec., which reduced the computed impulse to 22% of the actual value."

The table problem "was a substantial contributor" to the navigation error, says Mars Surveyor flight operations manager Sam Thurman, but it is unclear whether it produced all the error that became evident only in the last hours of the mission. Many questions remain to be answered, e.g. why the MCO's miss distance at Mars had been set so low (at 224 km, with 150-180 km expected after the last TCM) when a higher altitude (e.g. 400 km) would have provided much more margin for position errors. And one wonders whether NASA can really risk to 'aerocapture' future Mars spacecraft, where even higher precision is mandatory... (AW&ST of Oct. 4, 1999, p. 40-1)