News & Views | Published:

Extrasolar planets

Giant giants or dwarf dwarfs?

Naturevolume 409pages462463 (2001) | Download Citation


An object at least 17 times the size of Jupiter, discovered orbiting a Sun-like star, has astronomers scratching their heads. Is it a giant planet or a failed star?

Planet hunters keep discovering yet more weird objects around nearby Sun-like stars that resemble nothing found in our own Solar System. The latest discoveries, announced at a meeting earlier this monthFootnote 1, include a three-body system that appears to be composed of a central star orbited by two objects with masses no smaller than 7.5 and 17 times that of Jupiter, the largest planet in the Solar System. The larger of these two behemoths is massive enough to be termed a brown dwarf star. Brown dwarfs are failed stars that are large enough (greater than about 13 Jupiter masses, MJ) to burn some deuterium, but not massive enough to ignite the steady nuclear reactions that make stars shine for billions of years. The discovery of this giant threatens to help overturn the tentative distinctions astronomers have tried to draw between gas giant planets, such as Jupiter, and brown dwarf stars. But reclassifying these beasts may be an almost trivial task compared with figuring out how they formed in the first place.

A worldwide race has been underway for the past five years1 to complete the first census of Jupiter-like planets around Sun-like stars in our local neighbourhood of the Milky Way. The preferred method for detecting these objects is the 'radial-velocity technique', in which telescopes equipped with high-precision spectrographs measure the Doppler shifts of starlight caused by the star orbiting a common centre of mass with its unseen companions. Jupiter, for example, causes the Sun to wobble in space by about 12.5 metres per second during its 12-year orbit.

The latest announcements come from the incredibly successful team led by Geoff Marcy (University of California, Berkeley) and Paul Butler (Carnegie Institution of Washington), who have found roughly two-thirds of the 50-plus extrasolar planets discovered to date. The newest objects2 are in orbit around a Sun-like star called HD168443, located about 123 light years away in the direction of the constellation Serpens.

Previous measurements showed that HD168443 had at least one planet3, but further observations were required to reveal what else might be lurking within the system. The latest data show that the original planet has a mass of at least 7.5 MJ and an orbital period of 58 days, placing it closer to its star than Mercury is to our Sun (about 0.3 astronomical units (au) away from HD168443, where 1 au is the Earth–Sun distance of 150 million kilometres). The second object has a mass of at least 17 MJ and orbits the star with a period of 4.8 years, at a distance of 3 au, corresponding to the distance to the Sun of the asteroid belt in our Solar System. Both objects have eccentric (non-circular) orbits, as do essentially all the extrasolar planets orbiting at these distances.

Because the radial-velocity technique measures only the Doppler shift along the line of sight to the Earth, the actual masses could be much higher if their orbits are seen nearly face-on (rather than edge-on). But this is unlikely, so the masses are probably only about 30% above the minimum values. The absence of a visible wobble in the position of HD168443 seen in astrometric data from the Hipparcos satellite2 limits the mass of the outer companion to less than 42 MJ. Together, HD168443's known companions have at least 25 times the mass of the planets in our Solar System. We're not in Kansas, any more, Toto.

Figure 1 shows how the new system compares to the bestiary of extrasolar planets and brown dwarfs previously discovered around Sun-like stars. There is no obvious break in the mass distribution because nature has managed to populate nearly the entire range of possibilities that have been searched to date, leaving no niche unoccupied (much like the behaviour of terrestrial organisms). The two objects orbiting HD168443 fall uncomfortably close to, or in the middle of, the provisional dividing line between gas giants and brown dwarfs, thought to lie between roughly 10 MJ and 30 MJ. The radial-velocity technique favours the detection of heavyweights, so objects with these masses should be very easy to detect. Yet the fact that very few objects have so far been found in this range suggested that there might be a gap between the masses of the most massive gas giants and the least massive brown dwarfs, however rare the latter might be. But HD168443 throws a monkey wrench into these feeble attempts to classify what we've found so far.

Figure 1: Pattern of discoveries for gas giant planets and brown dwarf stars found orbiting nearby stars similar to our Sun.
Figure 1

In most cases, only the minimum mass of the companion is known, although this value is likely to be close to the true mass. The star HD168443 has now been found to have two massive companions (HD168443b and HD168443c) that blur the distinction between giant planets and brown dwarf stars. In particular, the mass of HD168443c falls into the observed gap (from about 15 MJ to 30 MJ) between gas giants and brown dwarfs. The radial-velocity technique preferentially detects massive, close-in planets, as indicated by the detection bias along the dashed line.

Beyond simple classification, there is the looming question of how a system like HD168443 might be created. Could it have the same history as a multiple star system? These are thought to start with the collapse of a dense molecular cloud core to form a flattened protostellar disk, followed by fragmentation into multiple protostars. Our understanding of this process is still hazy, but it seems an unlikely way to give birth to a tight stellar system in which the mass of the primary star is 50 to 100 times more massive than its companions. This is because subsequent growth by gas accretion from the disk and an infalling envelope of gas would add mass preferentially to the smaller protostars, leading to more equal stellar masses4. Furthermore, fragmentation calculations do not favour the formation of low-mass protostars in stable, planet-like orbits around a single much more massive protostar. When multiple protostars form by fragmentation, they begin life in unstable configurations, which are then subject to gravitational scattering during close mutual encounters.

We are therefore left with trying to explain the formation of the HD168443 system by the process of planetary formation which unfolds well after the central protostar has gained the bulk of its mass. There are two competing mechanisms for the formation of gas giants in a protoplanetary disk: core accretion and disk instability. The currently favoured theory, core accretion, was developed to explain the formation of Jupiter and Saturn through the formation of a rock and ice core, followed by the accretion of gas. Core accretion could account for the formation of planets with masses up to about 5 MJ (refs 57), but whether it could produce objects as massive as the companions of HD168443 within the lifetime8 of a typical protoplanetary disk (a few million years) remains to be seen. Disk instability9, on the other hand, is quite capable of rapidly forming massive protoplanets, as large as 17 MJ, with change to spare.

Theorists have their work cut out to explain the formation of HD168443's unexpected companions. The planet hunters, meanwhile, are discovering bizarre solar systems at an alarming rate.


  1. 1.

    *197th Meeting of the American Astronomical Society, San Diego, California, USA, 7–11 January 2001.


  1. 1

    Mayor, M. & Queloz, D. Nature 378, 355–359 (1995).

  2. 2

    Marcy, G. W. et al. Astrophys. J. (submitted). See

  3. 3

    Marcy, G. W., Butler, R. P., Vogt, S. S., Fischer, D. & Liu, M. C. Astrophys. J. 520, 239–247 (1999).

  4. 4

    Bate, M. R. Mon. Not. R. Astron. Soc. 314, 33–53 (2000).

  5. 5

    Bodenheimer, P., Hubickyj, O. & Lissauer, J. J. Icarus 143, 2–14 (2000).

  6. 6

    Kley, W. Mon. Not. R. Astron. Soc. 313, L47–L51 (2000).

  7. 7

    Nelson, R. P., Papaloizou, J. C. B., Masset, F. & Kley, W. Mon. Not. R. Astron. Soc. 318, 18–36 (2000).

  8. 8

    Briceño, C. et al. Science 291, 93–96 (2001).

  9. 9

    Boss, A. P. Astrophys. J. 536, L101–L104 (2000).

Download references

Author information


  1. Institute for Genomic Research, 9712 Medical Center Drive, Rockville, 20850, Maryland, USA

    • Alan P. Boss


  1. Search for Alan P. Boss in:

Corresponding author

Correspondence to Alan P. Boss.

About this article

Publication history

Issue Date



By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing