Jupiter has four and Saturn seven spherical satellites. Jupiter's moon Ganymede and Saturn's Titan are larger than Mercury and they're the most massive moons at 0.0248 M♁ (Ganymede) and 0.0225 M♁ (Titan). Do these moons 'shepherd' the other moons' and asteroidal satellites' orbits? Ganymede is three times the mass of Europa and quite close to it, so does it dominate Europa's orbit? Callisto on the other hand is twice as far as Ganymede and the 3rd most large and massive known moon while quite lonely in its orbit. Would it qualify for a planet too? If Jupiter was a star, I think Ganymede and Callisto would qualify as planets, Europa would be a 'dwarf planet' (because of being dominated by Ganymede's and perhaps Io's orbit) and Io I don't know, it probably would be more controversial.

Saturn's moon Titan is more massive than the other six spherical moons altogether, so I think it would clearly be a planet. The other moons (except Iapetus) are quite close to each other so I think they'd be considered 'dwarf planets' except Iapetus which is very far from Saturn and also quite lonely, so it might qualify as a planet (there might be an issue concerning its shape but my question is about the 'clearing the neighbourhood'/planetary discriminant point). Mimas would near-certainly be a 'dwarf planet' because it's about within the rings which alone are about 2/5 the mass of Mimas. So I think Titan and Iapetus would be considered 'planets', Mimas a 'dwarf planet' and the other spherical moons of Saturn probably too.

What would you say? Which Jovian and Saturnian moons would be considered planets and which ones 'dwarf planets' as per the proposed planetary discriminants if Jupiter and Saturn were stars?


1 Answer 1


I've noted before that the IAU naming critera are guidelines rather than laws of nature. If applied to Jupiter's moons the four Galilean satellites would probably be "planets" (they don't share their orbits with anything else of comparable size and are large enought to be in hydrostatic equilibrium) There is a mean-motion resonance 4:2:1 between them but this is a more or less symmetric relationship without one body dominating another.

The other moons are tiny by comparison. There are 4 major bodies orbiting Jupiter.

For Saturn, Titan certainly dominates its orbit (Hyperion is much smaller.) Dione and Rhea also seem to have a value of mu that are over 100 (though determining the value is tricky because not all moons and moonlets have been discovered) Enceladus and Tethys and Iapetus also seem to qualify, As you note, Mimas is too small and borderline spherical. The other moons are just so much shrapnel by comparison.

So there would probably be the six pre-1800 moons, minus Mimas.

It is possible that the rules might be bent, the Rings are no permanent features, so they might be ignored for the sake of planetary determination. Remember the rules exist to serve us. Humans make the rules. Some can be bent, some broken.

  • $\begingroup$ Somebody watched the Matrix too often. Thank you for stating how vague the 2006 definition is. My question was more precisely concerning the planetary discriminant proposals but you're right in that 'orbital neighbourhood' is too vague and must be defined arbitrarily perhaps. $\endgroup$
    – Greenhorn
    Dec 27, 2020 at 15:24
  • $\begingroup$ Precisely. It's not meant to be some kind of definition that is mathematical and unambiguous. It just says "Some of the rocks orbiting the sun are big and special, some are less big and less special in their context. The big special ones are planets (and there are eight of them)" If you start to get all legalistic about it, you are not engaging with the spirit of the definition. $\endgroup$
    – James K
    Dec 27, 2020 at 17:45
  • $\begingroup$ But there must be some kind of precision in order to neither include Pluto, Eris or Ceres nor exclude Mercury or Mars for instance (all of which is possible with the 2006 definition). The planetary discriminants offer an equation that determines what a planet is from a certain value on. These equations could also be applied to the moons in this what-if-scenario. $\endgroup$
    – Greenhorn
    Dec 27, 2020 at 18:15
  • $\begingroup$ @Greenhorn: The planetary discriminant of the smallest planet is so vastly different from that of the largest non-planet that there would be no need to figure out what the cutoff value should be unless a body is found for whose value would be between those two values, but not even close to either of them. On the other hand, I think it's more useful to classify Pluto as the first discovered moon/planet hybrid which is in a resonant orbit with a planet, than as anything else. $\endgroup$
    – supercat
    Feb 27, 2023 at 20:42

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .