In 1997, a planet of roughly Jupiter's mass was discovered in orbit around the star Andromedae. Two more have now been identified.

In the 1990s, the quest to find planets outside our own Solar System finally succeeded. The first extrasolar planets to be identified are in orbit around a pulsar1, a rapidly rotating neutron star which is quite different from our own familiar Sun. The first planetary companion to a main-sequence star — a hydrogen-burning star like the Sun — was discovered in 1995, using Doppler measurements of the reflex motion (wobble) that the planet's orbit induces in the star 51 Pegasi2. Nearly 20 more main-sequence stars have since been found to possess a single companion of roughly Jupiter mass3. But now we have news of something else again — the discovery, announced last week by Paul Butler and his colleagues4, of a system of three massive planets orbiting the star Andromedae.

All of the planets known to orbit main-sequence stars other than the Sun have been identified using the Doppler technique, which measures changes in the star's velocity component along the line of sight. These observations can be inferred as perturbation of the star by a companion planet. They yield the planet's orbital period and orbital eccentricity, and the product of its mass and the sine of the inclination of its orbital plane to the plane of the sky, M sin(i). Because sin(i) cannot exceed unity, the measured radial-velocity variations of the star provide a lower limit to the planet's mass.

Determining the masses and orbits of three planets travelling about a star using Doppler data is a highly complicated business, because the data have significant intrinsic scatter and the signatures of the planets interfere with one another. Butler et al.4 looked at two sets of data, obtained respectively at the Lick Observatory in California and by the Advanced Fibre-Optic Echelle (AFOE) spectrometer at the Whipple Telescope in Arizona. They analysed these data independently, and when combined, to derive three sets of planetary parameters. The similarity of the values of these parameters (see Fig. 1) lends confidence to the three-planet solution.

Figure 1: The reported orbital parameters for the planets around Andromedae4 are given on.
figure 1

the left side of the plot. The orbit of each planet is represented by three points — the periastron and apoastron (closest and farthest distances from the star, bottom and top points) and the semimajor axis (roughly, the average orbital distance). Triangles represent the outer planet, crosses the middle planet and circles the inner planet. Error bars are shown for the outer two planets; observational uncertainty is far less for the innermost planet. Data from the Lick Observatory are given in green and from the AFOE (Whipple Telescope) in red; combined data are in black. The rest of the plot shows the temporal evolution of the periastra and apoastra of the three planets using the nominal parameters on the left as initial conditions. The systems using the combined and Lick data sets continue to be stable for the entire ten-million-year integration interval. (Calculations and figure by E. J. Rivera of SUNY, Stony Brook, and J.J.L. using the SwIFT symplectic MVS package8.)

The star Andromedae has a mass about 30% greater than that of our Sun, is about halfway through its six-billion-year main-sequence lifetime and is about 44 light years from our Solar System. The innermost planet was first discovered in 1997 (ref. 5). It has a nearly circular orbit with a period of 4.6 days, and a minimum mass about 0.7 times that of Jupiter; these properties are similar to those of the only known planet of 51 Pegasi and to other star-grazing planets.

What about the two newcomers identified by Butler et al. ? The middle planet has a period of about 242 days, a mass about twice that of Jupiter and an orbital eccentricity, e, of approximately 0.22 (comparable to the eccentricities of Mercury and Pluto, and larger than the orbital eccentricities of all other planets known in our Solar System). The orbital period of the outer planet is 1,270 days; e 0.36; and mass 4.1 times that of Jupiter. These latter values are not well constrained, because most of the Doppler data were taken within a 6.5-year interval, less than two orbital periods of the planet. But the masses and eccentricities of the outer two bodies are typical of half-a-dozen known extrasolar giant planets with orbital periods of around a year.

Additional constraints on the Andromedae planets come from astrometric and photometric data, as well as from numerical orbit integrations. The Hipparcos astrometric satellite observed the position of Andromedae on the plane of the sky several times during the early 1990s. No wobble is detectable in these data, which implies that the outermost planet is less than ten and probably no more than five Jupiter masses. Photometric measurements of the brightness of Andromedae around the times that the inner planet would pass in front of (transit) the star if its orbital plane were nearly along the line of sight (sin(i) close to 1) rule out transits deeper than about 0.0002 to 0.0003 magnitude (G. W. Henry, personal communication). This means that the inclination of the planetary orbit must be less than 83° or that we are dealing with an extremely dense, exotic object.

As far as numerical integrations are concerned, the preliminary results of calculations by myself and a colleague, E. J. Rivera, imply that a system having the parameters derived from the AFOE data only, with sin(i) of approximately 1, would tear itself apart within a million years or so (Fig. 1). (The chaotic nature of such a trajectory precludes a precise calculation.) The star, and thus presumably its planets, are about three billion years old, and it would be extremely unlikely that the system just happens to be so near to the end of its expected lifetime. What the calculation implies, then, is that the nominal AFOE parameters are unlikely to be correct. Nonetheless, the characteristics of the system could well be within the error bars quoted for the AFOE data.

By contrast, systems using the Lick or combined Lick/AFOE data sets appear to be stable for at least ten million years and possibly much longer. Additional integrations, not shown in Fig. 1, imply that systems with more massive planets (smaller sin(i) are generally less stable, especially if the orbits are substantially inclined to one another, and that changing the orbital period of the outer planet within the range allowed by the data can have a big effect on the stability of the system.

Neither the star Andromedae nor any of its three planets appear in themselves to be particularly unusual, but the system as a whole stands out in the currently known menagerie of extrasolar planets. Its existence shows that several massive planets can orbit far closer to a star than do the gas giants (Jupiter and Saturn) and smaller ice giants (Uranus and Neptune) of our Solar System. So if giant planets all form far enough out in a protoplanetary disk for ice to condense (the conventional, but not exclusive, view of theorists6), then the migration or mutual scattering (or both) that brings them in towards their star is not always violent enough to destroy themselves or their brethren.

The Andromedae system is just one example of a wide variety of planetary configurations that can be expected to exist in our Galaxy. All of the extrasolar planets thus far discovered orbiting main-sequence stars are more massive than Saturn, and most either orbit very close to their stars or travel on much more eccentric paths than do any of the major planets in our Solar System. The Sun's planets are all either low in mass or travel on distant orbits from their star, and they are therefore more difficult to discover using the Doppler technique. So it could be that most planetary systems will turn out to be like ours.

There is much, much more to come. With the projected advances in detection technology, including schemes to find Earth-like planets7, we are at the beginning of a Golden Age of extrasolar planetary studies.