As you've undoubtedly already read elsewhere by now, a subcommittee of the International Astronomical Union has proposed a definition for planet. I had it delivered to my flap-step while I was camping in Acadia National Park the past week.
The part of "IAU Resolution 5 for GA-XXVI" that describes the planet definition, states "A planet is a celestial body that (a) has sufficient mass for its self-gravity to overcome rigid body forces so that it assumes a hydrostatic equilibrium (nearly round) shape, and (b) is in orbit around a star, and is neither a star nor a satellite of a planet."
The first condition sets a lower limit, it's meant to include objects. It could do with a more quantitative description, but perhaps the words are sufficient if I knew my gravitational fluid dynamics better. If it were all there were though, we would have to include the Moon, Io, Europa, Callisto, Ganymede, Titan, Mimas, Triton, Encealdus, and countless others I can't recall. So we introduce the second condition to limit it: must be orbiting a star. Embedded within that definition is the implication that if two (or more) bodies orbit each other, we must look at the barycenter (center of mass) of the pair - if it is within one body, that is the planet and the other is the moon. If, as in the case of Pluto and Charon, it is between the two bodies, then they are both planets.
There are of course criticisms or quirks of the definition, though I am admittedly beginning to warm to it. There is no third condition for when mulitiple objects are in similar orbits, and this allows mostly round asteroid Ceres to remain a planet. 2003 UB313 (provisionally dubbed "Xena," though it will be officially renamed something else by the IAU later) is bigger than Pluto, and therefore definitely massive enough to be round. Quaoar and Sedna are smaller than Pluto, so we're reserving judgement for now, but we're left with 12 planets in the Solar System if the definition passes. The "planemos" discovered a couple weeks ago would not be planets, as they are not orbiting a star.
There's also an additional clause for concern.
The IAU draft Resolution also defines a new category of planet for official use: "pluton". Plutons are distinguished from classical planets in that they reside in orbits around the Sun that take longer than 200 years to complete (i.e. they orbit beyond Neptune). Plutons typically have orbits that are highly tilted with respect to the classical planets (technically referred to as a large orbital inclination). Plutons also typically have orbits that are far from being perfectly circular (technically referred to as having a large orbital eccentricity). All of these distinguishing characteristics for plutons are scientifically interesting in that they suggest a different origin from the classical planets.
The description of plutons in terms of orbital characteristics rubs me the wrong way. However, I could easily see redefining it to fit the current model of terrestrial and jovian planets. The three sub-categories would then be:
- Terrestrial / Rocky - Mercury, Venus, Earth, Ceres. These planets have densities of 3-5 g/cm^3 and are primarily composed of rocky material (carbon/silicon solids, and may contain liquid metal cores. Asteroids, the Moon, and Titan fit these characteristics though they are not planets.
- Jovian / Gas giants - Jupiter, Saturn, Uranus, Neptune. These planets have densities typically less than 2 g/cm^3 and are primarily composed of atomic hydrogen gas, with trace amounts of other gaseous molecules such as methane. The inner layers transition into liquids due to the intense pressure. Gas giants are NOT undergoing nuclear fusion of either hydrogen or deuterium. Brown dwarfs do not fit into this category, however I believe all extrasolar planets discovered to date do.
- Plutino / Icy - Pluto, Charon, 2003 UB313. Lastly these planets will have a density of roughly 1-3 g/cm^3, and their composition will be predominantly ice - water ice, carbon dioxide ice ("dry ice"), methane ice, and so on. They may have some rocky content or core, but not a significant portion of their composition. Quaoar and Sedna will count in this category if they turn out to be spherical. Europa, Ganymede, and Callisto would be counted if they weren't orbiting Jupiter.
ETA: The BBC says we should keep our eyes on three more asteroids, Vesta, Pallas and Hygiea, that are at the borderline of mass/roundness for whether they count as planets.