An extra-solar planet is a planet that circles a star other than Sol (our sun).
There is one planetary system that we know, and reasonably well (compared to those that we haven't the faintest clue about). This is our own system of 8 or 9 planets (depending on designation of Pluto), an asteroid belt, and countless bodies in the Kuniper Belt. Given the trillions of stars in galaxies and billions of galaxies out there, it would be highly improbable at best if our planets were the only ones made.
With 10 to the 11th stars in our galaxy and 10 to the 9th other galaxies, there are at least 10 to the 20th stars in the universe. Most of them may be accompanied by solar systems. If there are 10 to the 20th solar systems in the universe, and the universe is 10 to the 10th years old -- and if, further, solar systems have formed roughly uniformly in time -- then one solar system is formed every 10 to the negative 10 yr = 3 x 10 to the negative 3 seconds. On the average, a million solar systems are formed in the universe each hour.
-- Shklovskii, I.S. and Carl Sagan. Intelligent Life in the Universe . New York: Dell Publishing, 1966. 509p. Pg. 130
The first work on extra-solar planets was done in the 1910's when a rather small and until then insignificant star was found to have a large proper motion of about 10 arcseconds per year (determined by looking at several decades worth of photographic plates). The star was named after its finder, E.E. Barnard and is know today as Barnard's Star. Astronomers investigated this now noteworthy star to find that it was a common red dwarf 5.95 light years away (very close in astronomical terms). Further questions of did this star have any wobble in its movement, and if so, could this wobble be explained by having additional unseen bodies revolving around the star?
Peter van de Kamp poured over 2,000 plates of Barnard's Star from between 1938 and 1962. From those plates, van de Kamp claimed that there was a wobble in the motion caused by a body that was 1.6 Jupiter masses and revolved around the primary every 24 years in an elliptical orbit. Between 1969 and 1982 van de Kamp published a number of papers refining his figures. In 1973, however, a few papers were published that questioned these findings - unable to find any wobble in the motion of Barnard's Star. Another paper indicated that all the stars in the photographic plates from which van de Kamp worked from indicating flawed data - possibly caused by changes in the telescope. To date, no confirmation of planets around Barnard's Star has been established.
The basic methods for detecting an extra-solar planet are:
The method that van de Kamp used is known as astrometric detection which looks at the proper motion of a star using other stars as reference points. Something moving around the star will tug at the star and pull it around. As seen from Earth, this would cause the motion of the star to not be a straight line. Astrometry, however, only works with stars that have an appreciable proper motion - more distant stars would be harder to detect. While the telescopes have gotten better, this is not the most common of means of detection. This method of detection works best with larger planets orbiting in distant orbits because the center is moved away from the center of the star increasing the wobble.
An earth-mass planet in an earth orbit orbiting a solar
mass star 33 light years from us will produce a 0.3 mas wobble in the star's position; while Jupiter at 300 times the mass of the Earth and 5 times the distance from the sun produces a signature that is 1500 times as strong - on the order of 500 mas (mas is a miliarcsecond, or 10-3 arcscond). Jupiter actually pulls the sun 743,000 km away from where it would be otherwise - this is 1.068 solar radius. Saturn pulls the sun another 408,000 km (or 0.586 solar radius). This is known as the barycenter of a system. When all of the Jovian planets are on the same side, the tug is 2.169 solar radii.
From the Moon, the Earth shines as a large blue gem in the sky. And Mars shows as a point of ruby light - and Venus is shining white. Planets reflect the light of the stars. This is the grail of extra-solar planet detection - to be able to see a planet. The problem is that only very large planets would reflect enough light for present day telescopes to detect and the glare of the star outshines the light reflected by the planet. Of the possible bonuses, this would give information on the temperature, albedo, radius, and actual orbit of the planet. Spectroscopy of the light reflected would give the composition of the atmosphere and may show indications of life.
Working once again with the tug of the planet on the star, though this time measuring the difference in the light rather than the proper motion of the star. Here, the Doppler shift of the star light is measured as the star is pulled in our direction and away from us. This shows up as a periodic shift of the spectral lines. With this method, "hot Jupiter's" are easiest to detect - very close to the star orbiting very fast. In the case of Peg 51, the Jupiter class planet was found in an orbit of 4 days around. However, this method is not without faults, it is possible that sunspots and other disturbances could show up as the velocity shifts (the Sun shows such motions in the range of 5 hours). This method works best the Earth is on the orbital plane of the star - and worst when the plane of the orbit is perpendicular to our viewing (then there is no Doppler Shift detectable). The current range of detection with this method only works to 160 light years.
Ground based photometry
By measuring the brightness of a star, dark binaries are detectable - if the observer is on the orbital plane. This is not a large change, but it is one that can be used as a confirmation of the planet with the radial velocity method. Recently, the most distant planet discovered yet (5,000 light years - most other discoveries are within 100 light years) used this method (described as watching a mosquito fly in front of a searchlight 200 miles away). This also works better in crowded star fields (large number of canidates can be observed at once) and uses smaller telescopes (easier to get observing time).
On Oct 6, 1995 the first extra-solar planet was found around 51 Pegasi - a similar star to the Sun, G2. The official announcement was made 19 days later in the International Astronomical Union Circular. The planet was found to have a mass of at least half the mass of Jupiter, and no more than twice the mass of Jupiter at an orbit of 0.05 astronomical units from Peg 51 (Mercury has an orbit of 0.38 AU) spinning about the star every 4.23 days.
Since then, a large number of planets have been discovered
(and even the possibility of detecting a comet around a pulsar). The Extra-solar Planets Catalog (http://www.obspm.fr/encycl/catalog.html)lists 88 planetary systems and 102 planets as of Dec 18, 2002. Of these, a large number are hot and heavy planets or highly eccentric orbits. While no terrestrial planets have yet been detected and thus no place for life as we know it, some large planets have been found, within the habitable zone of the star that could have moons (such as Jupiter and Europa or Saturn and Titan) that could support life.
The next generation of space telescope holds a tantalizing possibility of being able to image the planets themselves, and being able to detect Earth mass planets between 0.5 and 3 AU away from the star.