Someone recently told me there's a new contender for dark matter, a 'new kind of pion', as well as the WIMPs and axions and assorted possibilities that have been suggested before and haven't been seen, so I thought I'd explain why I don't believe there is any form of exotic dark matter in the amounts needed to hold galaxies together.
First, as we all know by now, the rotational velocities of stars around galaxies are too great for them to be held in by the gravity of the mass we can see in them, hence the ad hoc addition of 'dark' matter. Anything ad hoc is bad science in the first place, but the crucial observation here is that this discrepancy always starts at the galactic radius where the stars' rotational acceleration drops below a critical acceleration of 2x10^-10 m/s^2 (Milgrom, 1983). It is difficult to hypothesize any type of matter that would suddenly appear at a specific low acceleration like this, but it is easy to hypothesize a direct link with Hawking-Unruh radiation because at just that crucial acceleration the waves of this radiation become equal to the Hubble scale. This is a fascinating agreement out of which arises MiHsC theory.
Second, globular clusters are smaller, bound, collections of stars usually located in a sphere around the centre of the galaxy (last week I spent some time searching the sky for them when I was on holiday in Cornwall) and wide binaries are bound stars very far apart. Both of these types of system show the same odd behaviour as the much larger galaxies that surround them: at radii where their stars fall below an acceleration of 2x10^-10 m/s^2 they start to rotate around the systems too rapidly for standard physics (Scarpa et al., 2006, Hernandez, 2012). The crucial point here is that in these cases dark matter cannot be used to fix the anomaly since it must be smooth on these scales to allow it to work on galactic scales, so you can't pack a lump of it into the globular cluster or binary system to bind it gravitationally. This implies that whatever effect is happening in globulars (ie: not dark matter!) also applies to the similarly-misbehaving galaxies (MoND theory also fails for globulars which are close to the galactic centre and so the 'external' acceleration is larger than 2x10^-10 m/s^2).
There are also more philosophical reasons for disliking dark matter: science always progresses by presenting old theories with embarrassing new data, so that those theories can be improved. What the dark matter paradigm is doing is defending an old theory by changing the 'observations' (adding unseen mass) which is a complete reversal of the scientific process and more resembles the machinations of the dark ages.
The solution, as ever, is to use crucial observations to force old entrenched theories to make themselves so complex that they collapse in ridicule under the weight of their own convolutions: I hope globular clusters and wide binaries can be useful in this respect.
"You see, the thing that really finishes a Boggart is laughter. What you need to do is force it to assume a shape that you find amusing." - J.K. Rowling (Harry Potter & the Prisoner of Azkaban).
Milgrom, M., 1983. A modification of the Newtonian dynamics as a possible alternative to the hidden mass hypothesis. Astrophys. J., 270, 365.
Scarpa et al, 2006. Globular clusters as a test for gravity in the weak acceleration regime. Proceedings of the 1st crisis in cosmology conference. http://arxiv.org/abs/astro-ph/0601581
Hernandez, X., M.A. Jimenez, C.Allen, 2012. Wide binaries as a critical test of classical grvaity. http://arxiv.org/abs/1105.1873
McCulloch, M.E., 2012. Testing quantised inertia on galactic scales. Astrophys. Space Sci., 342, 575-578. http://arxiv.org/abs/1207.7007