Here is a serious contender for Astronomy Paper of the Month. This paper uses a very simple and elegant argument based on data from existing astronomical observations and star catalogues to disprove the dark matter theory. It does so by lending significant support to the idea that a few tweaks to the Newtonian equations of gravity remove the need for Dark Matter and Dark Energy.

The Breakdown of Classical Gravity? X. Hernandez, M. A. Jim´enez, C. Allen

I'm uncertain if the paper has been properly refereed, but it is an interesting read.

Dark Matter was invented by cosmologists to explain "missing matter" in the universe. Various large scale cosmological phenomena, such as the speed of galaxy rotations, do not sit well with the laws of Newtonian dynamics unless the galaxies contain a lot of unseen mass. General relativity doesn't help either. As cosmologists can neither see nor directly prove the existence of this mass, it was christened "Dark Matter".

Hopefully we all remember our school physics lessons. The force of gravity decreases with the square of distance. Proponents of MOND (MOdified Newtonian Dynamics) claim that this rule changes at great distances (i.e. at very small accelerations). For our Sun, this means a distance of 7000AU (1 AU = distance from Earth to Sun). Unfortunately It is troublesome to go and measure the force of gravity in this location so a less direct proof is needed.

A lot of the star systems in our galaxy are binaries. Two stars orbiting around each other (remember Luke!) at great distance. The insight of the authors in this paper was to realise that the motions of stars in very wide binary star systems could be used as test cases to see if the law of gravity does need modifying at large distances / small accelerations.

We can't simply observe a wide binary (where the two stars are separated by 7000AU) around its entire orbit (usually several centuries), so we have to make an inference based on the observed proper motion of the stars. Their argument thus becomes a little weakened at this point, but no less convincing. They also make various attempts to "break" their argument by using different data and subsets of the data, but the hypothesis is not disproved.

They conclude that the motion of wide binary star systems points strongly to a need to modify Newtonian gravity. As the distance between the two bodies increases the force moves from the Newtonian inverse square rule of GM/r2 to a form of GaM/r. This means the orbital velocity of wide binaries should cease to decrease with separation and gradually settle on a constant velocity which is only dependent on the entire mass of the star system.

Perhaps we can view gravity as having two separate components, with only one operating on the inverse square rule. Or we can view it in the same way as light. There is a minimium unit of light enery (the "quantum") so why can't we have a minimum unit of gravitational energy?

If it can be proven that the equations of classical gravity need modification at low accelerations, then there is no need to invoke Dark Matter to explain the universe.

A lot of mainstream cosmologists strongly resist the idea of MOND solutions whereas I quite like them. Perhaps I need to do a nice rant against dark matter!