A graviton is the particle that carries the gravitational field, and it travels at the speed of light. It is a spin-2 particle, the only one. This means that it somehow needs only spin half a revolution before it arrives in the same position. [An analogy of this may be a 'Mobius' strip with not half a twist as required of a Mobius strip, but with two full twists]. Having an integral spin means that it is a boson; and only bosons are force-carrying particles, in this case it carries the the gravitational force. [Photons, with a spin of 1, carry the electromagnetic force]. But unlike as in the the electromagnetic force, which is felt only by fermions (spin ½) and not by other photons, the graviton is able to feel itself and other gravitons. That is, not only does a graviton generate a gravitational field, it is also affected by the very field it creates. This separates it from other force-carrying bosons; the graviton is unique. [However - see Higgs Boson]. But this same self-referral also complicates the equations. Also, the graviton affects both space and time, warping both in it's presence.

Gravitational Waves
A static mass generates a static gravitational field, just like a static electrical charge generates a static electrical field. An accelerating charge generates electromagnetic radiation, carried by photons. However, an accelerating mass generates no gravitational waves, gravitational waves are only generated when the acceleration of the mass is changing. Gravitational waves are carried by gravitons. But with the gravitational force being 1039 times weaker than electromagnetic force, the gravitational waves are feeble by comparison. Thus, a star in the process of gravitational collapse to a black hole would generate strong gravitational waves, but only if the collapse is asymmetric, symmetric collapses cancel out far-field gravity wave formation. The gravitational waves would propagate through space-time at the speed of light, distorting space-time as it passed through it. Because a graviton is a spin-2 particle with a quadrupole moment, the gravitational wave, as it passed through any point in space, would both stretch space in one direction and compress the space in the orthogonal direction. This is a unique signature of the passage of a gravitational wave. A detector that measured both an expansion in a body in one direction, and a compression in the other direction simultaneously would be fairly certain that a gravitational wave had passed through the detector.

If virtual gravitons do carry the gravitational force, (like virtual photons carry the electromagnetic force) then, because of the extreme feebleness of the gravitational force when compared to that of the electromagnetic force, the virtual gravitons must interact with matter extremely rarely. The graviton, however, is 10 21 times less likely to interact with matter than even a neutrino, which themselves only very rarely interact with matter. Theoretically, our Sun could be a source of gravitons, produced by accelerating electrons in a process analogous to bremsstrahlung radiation. But a mass the size of Jupiter orbitting the Sun at Earths orbit would interact with only 1000 gravitons during the lifetime of the Universe. The interaction would produce an electron in a process similar to the photo-electric effect where a photon strikes a semiconductor generating an electron, only it would be called the gravito-electric effect. There would be virtually no chance of detecting such an electron because of the masking effect of utter quintillions of other electrons produced by other means. So not only generating a graviton, but also detecting one is an almost impossible task. Gravitons could be generated by accelerating electrons up to the Planck energy, where gravity becomes comparable in strength to the strong force, but this would require un-imaginably large energies, 1016 times more powerful than that achievable in the Large Hadron Collider which is due to be switched on in 2007.

Despite the extreme feebleness of the gravitational force, it is nevertheless the dominant force in the Universe for the simple fact that all the other, far stronger, forces possess equal and opposite varieties which are often quite close to each other such that their effects cancel out at large distances, like for instance the positive and negative charges within a neutral atom. The nucleus is positively charged, but the electrons are negatively charged. Close to the atom, these separated charges are noticeable, but from a distance the atom appears neutral, neither attractive nor repulsive. But the gravitational force seems to have no opposite repulsive variety, (apart from that due to the so-called 'Cosmological Constant') and thus the gravitational attractive force of masses in the Universe is cumulative throughout the Universe, albeit decreasing with inverse square law.

Detecting Gravitational Waves
The trouble is that even a gravitational wave generated by two colliding black holes would hardly stretch or shrink space by more than the diameter of a proton (a guess by RWD) at the typical distances involved. This calls for the extreme isolation of the detector from other external influences, and extremely sensitive measuring instruments. Two approaches have been tried: (1) - a large suspended cylinder of aluminium designed to resonate (and thus amplify) any sudden changes in its' length caused by the passage of a gravitational wave; and (2) - two kilometre-scale laser beams, at right angles to each other, arranged as akin to a Michelson-Morley interferometer. Neither has irrefutably detected a gravitational wave. This is because, despite the enormous power released as gravitational waves by the coalescence of two black holes, the deformation of space caused by the radiating gravitational wave is extremely small precisely because space is so very stiff, being 1027 times stiffer than diamond, the hardest physical substance known. Another reason is that the gravitational waves generated by two coalescing black holes will have a very long wavelength, much longer than an aluminium cylinder, much longer even than kilometre-scale laser beams. [update - in 2017 Gravitational Waves were detected from the merger of two Neutron Stars by an interferometer consisting of two 1km long tubes with laser beams down them. Many more have since been detected - see 'Nucleosynthesis by Neutron Star Mergers'].

Another experiment, MiniGRAIL, has cooled a 1.4 ton sphere of copper-aluminium allow to just 0.068 Kelvin, to detect the passage of gravitational waves, this being by far the largest mass ever cooled to such a low temperature. It is tuned to 3kHz and will thus respond best to gravitational waves with a 0.33 millisecond time-scale, and being very cold, should be very sensitive. However, because of its very narrow bandwidth, it may not be the first detector to detect a gravitational wave. But if 100 such balls with differing masses were made into gravitational detectors, they would together cover a much broader frequency range.

Inverse-Square Law
A gravitational field, just like an electromagnetic field, decreases in inverse proportion to the square of the distance (from point sources). If it were any different (inverse cube law, for instance) then there would be no stable orbit in which two bodies could gyrate. Only when gravity obeys inverse square law can bodies like the Earth and Sun orbit each other.

Geodetic Effect
One of the predictions of Einsteins' General Theory of Relativity is called the 'Geodetic Effect': the way that the gravity of the Earth slightly distorts the surrounding space-time, like the way a heavy ball bearing resting on a rubber sheet will deform the rubber sheet into a hyperbola. Measuring this effect for the Earth will require extreme precision, as the effect is miniscule in the case of the Earth.

Frame Dragging / Gravito-Magnetic Effect / Lense-Thirring Effect
Another prediction of General Relativity is that any spinning mass, like the Earth for instance, will generate a force that will slightly twist space-time, a process known as 'frame-dragging'. This effect is known as the gravito-magnetic effect and also as the Lense Thirring effect. For Earth, this force will be even more feeble than that due to the Geodetic effect above, being hundreds of times weaker. For a gyroscope orbiting around the Earth, the Lense-Thirring effect would cause the axis of spin of the gyroscope to rotate by about 42 milli-arcseconds per year, a miniscule and hardly measurable amount. [This amount of rotation could easily be masked by rotation due to either the Earths magnetic field, thermal radiation from the gyroscope after being heated by the sun, or variations in Earths gravity over the surface of the Earth]. But for a rapidly rotating massive compact object like a neutron star, the effect could generate considerable shear forces. But measuring the precession in spin of a neutron star produced by the gravito-magnetic effect will be fraught with difficulties, as great assumptions will have to be made about the precise way in which the neutron star spins: it's moment of inertia. A satellite called the 'Gravity Probe B' will be launched to orbit around the Earth to try to measure this effect. But they have been beaten to it by another group of researches who are utilising an already orbiting group of satellites, the LAGEOS satellites, which have mapped the Gravity of the Earth. They are re-analysing 11 years of data and re-using it to calculated the magnitude of any gravito-magnetic effect, whilst correcting for any other effects that could influence the result.

The gravito-magnetic effect is the gravitational analogue of the magnetic field produced by an electrical current and is produced by a 'mass current'.

And just as a time-varying magnetic field produces a time-varying electric field, so too should a time-varying gravito-magnetic field generate the equivalent time-varying gravito-electric field. This secondary gravito-electric field is dipole, and always forms closed loops just like magnetic field lines. There thus exists the gravito-magnetic and gravito-electric field equivalent of Maxwells' magnetic and electric field equations. H.W. Wallace believed that any secondary gravito-electric field generated by a moving gravito-magnetic field would result in the exclusion of the primary (generating) gravito-magnetic field, just like the presence of an electric current in a superconductor excludes the magnetic field from existing within that superconductor. In other words, that there is a corresponding Meissner Effect for gravito-magnetic / gravito-electric fields.

In the 1970s H.W. Wallace proposed that any spinning mass with un-paired nucleons (that is, isotopes with odd nuclear spin) should produce a gravito-magnetic effect. The spinning of the disc aligns the un-paired nucleons to spin in the same direction. He proposed that this induced gravito-magnetic effect and the secondary gravito-electric field might, by way of the exclusion of the gravito-electric field described above, produce a kind of gravitational shielding, or anti-gravity. Any mention of anti-gravity in the literature immediately gives rise to suspicions of crankiness in the wider scientific community, and nothing more has been heard of it since...

In 1991 Ning Li and Douglas Torr proposed that the gravito-magnetic field (that they propose is generated within the superconducting vortices of a current-carrying superconductor) is 1011 times greater than the magnetic field which that current generates. The gravito-magnetic field may be produced by the spinning lattice ions around each current-carrying vortex. A gravitational shielding effect is claimed, but reported measurements of up to 2% weight loss in spinning superconducting discs have not been confirmed by others...