Physicists working at St Andrews University in Scotland have proposed an alteration to Einsteins theory of gravity (general relativity), which potentially solves the problems of 'dark matter' in the universe.
It has been known for a long time that the amount of visible matter in the universe (stars, planets, galaxies etc) accounts for only about 5% of the mass needed to explain the observed expansion of the universe. Also on a more local level, dark matter is needed to explain the fact the galaxies remain in one piece rather than flying apart due to the rate the are spinning. General relativity predicts that there is not enough gravitational force (due the the mass of stars etc) holding the galaxies together to overcome the outward force exerted by the spinning. However, no-one has yet managed to identify this dark matter and so-called dark energy.
The potential for the gravitational laws, developed by Newton and refined by Einstein, to be wrong has been postulated for many years. In that sense this is not a 'new' theory. What the physicists have managed to do now is propose an actual modification to the law which essentially relies on the proposal that gravity is stronger over large distances than current theories predict. If this is the case there would be no need for the dark matter. However, anyone (well OK, anyone with the brain the size of a planet) could tweak a theory to match one set of observations. The key is that this new theory is what scientists call 'falsifiable', that is it make predictions about the universe which can be tested through observation. It those predictions don't match observation, then the theory if false, and it's back to the drawing board.
Lets face it, many of us must retain a healthy amount of scepticism about a theory which requires 95% of the universe to be 'hidden' and then goes on to suggest weird and wonderful types of dark matter and dark energy to make up the remainder.
There's still a long way to go in this leading edge area of cosmology and physics.
See press release from PPARC