Granot and colleagues studied the radiation from a gamma-ray burst – associated with a highly energetic explosion in a distant galaxy – that was spotted by NASA's Fermi Gamma-ray Space Telescope on 10 May this year. They analysed the radiation at different wavelengths to see whether there were any signs that photons with different energies arrived at Fermi's detectors at different times. Such a spreading of arrival times would indicate that Lorentz invariance had indeed been violated; in other words that the speed of light in a vacuum depends on the energy of that light and is not a universal constant. Any energy dependence would be minuscule but could still result in a measurable difference in photon arrival times due to the billions of light years that separate gamma-ray bursts from us.
The Fermi team used two relatively independent data analyses to conclude that Lorentz invariance had not been violated. One was the detection of a high-energy photon less than a second after the start of the burst, and the second was the existence of characteristic sharp peaks within the evolution of the burst rather than the smearing of its output that would be expected if there were a distribution in photon speeds. The researchers arrived at the same null result when studying the radiation from a gamma-ray burst detected in September last year, but could only reach about one-tenth of the Planck energy. Crucially, the shorter duration and much finer time structure of the more recent gamma-ray burst takes this null result to at least 1.2 times the Planck energy.
The paper appeared in Nature advance online publication today.
If this is true, several quantum gravity theories will crash and burn.
Edit: This paper has now officially appeared online. The exact reference is
A.A. Abdo et al., Nature v.462, p.331 (2009).