Is there an inherent instability or limit that prevents natural satellites like the moons of Jupiter from having their own natural satellites?

I looked at other questions such as do moons have moons but they focus on discovery rather than theoretical limits analogous to the Roche limit for satellite formation.

  • $\begingroup$ I would say there isn't. The sun is a natural satellite of the Galactic Center. Jupiter is a natural satellite of the sun. Jupiter's moon is a natural satellite of Jupiter. Why not go another layer deeper? $\endgroup$
    – Mathias711
    Jan 27, 2015 at 18:13
  • 2
    $\begingroup$ possible duplicate of Do moons have moons? $\endgroup$ Jan 27, 2015 at 18:26
  • $\begingroup$ Should be much easier in single lunar systems like Earth or Pluto, than in the well sorted Jupiter gravitational environment. The eccentricity of our Moon prevents any long-term stable orbits. But maybe Charon which is mutually tidally locked to Pluto could allow for it? And occasionally Earth captures an asteroid as a "minimoon". And many asteroids are binary, so... $\endgroup$
    – LocalFluff
    Jan 27, 2015 at 18:31
  • 1
    $\begingroup$ While the proposed duplicate question is focused on discovery, both answers seem to address the question theoretically. You may want to be more specific what those answers are lacking, otherwise the questions are really too similar to be separated. $\endgroup$ Jan 27, 2015 at 23:39

1 Answer 1


Wikipedia says there may be a reason (emphasis mine):

No "moons of moons" (natural satellites that orbit a natural satellite of another body) are currently known as of 2014. In most cases, the tidal effects of the primary would make such a system unstable.

This seems to be because the secondary would have to be very close to the primary - close enough that it doesn't become simply another satellite of the actual primary (i.e. the planet).

The page then goes on to talk about Rhea, a moon of Saturn that may or may not have a ring system. This gives some more information.

  • $\begingroup$ "Not known" does not mean impossible. Tidal effects of the primary can be quite small if it is far enough. $\endgroup$
    – Envite
    Jan 28, 2015 at 12:34

You must log in to answer this question.

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