More Dark Matter Phenomena Data Points

The following two papers illustrate observations of dark matter phenomena in circumstances when conventional dark matter explanations don't make sense. These two use the simple MOND toy model, which, admittedly has problems. 

But, they key point is that the data points aren't a good fit for dark matter particle theories, and that modified gravity theories are predictive in circumstances other than the ones in which they were formulated.

This paper is notable because wide binary stars would not be expected to have dark matter halos analogous to those alleged to form galaxies and galactic clusters.

Assuming Newton's gravity and GR to be valid at all scales leads to the dark matter hypothesis as a requirement demanded by the observed dynamics and measured baryonic content at galactic and extragalactic scales. Alternatively, modified gravity scenarios where a change of regime appears at acceleration scales a<a0 have been proposed. This modified regime at a<a0 will generically be characterised by equilibrium velocities which become independent of distance. Here we identify a critical test in this debate and we propose its application to samples of wide binary stars. Since for 1M⊙ systems the acceleration drops below a0 at scales of around 7000 AU, a statistical survey of wide binaries with relative velocities and separations reaching 104 AU and beyond should prove useful to the above debate. We apply the proposed test to the best currently available data. Results show a constant upper limit to the relative velocities in wide binaries which is independent of separation for over three orders of magnitude, in analogy with galactic flat rotation curves in the same a<a0acceleration regime. Our results are suggestive of a breakdown of Kepler's third law beyond a≈a0 scales, in accordance with generic predictions of modified gravity theories designed not to require any dark matter at galactic scales and beyond.

Hernandez et al., "Wide binaries as a critical test for Gravity theories" (2012)  https://arxiv.org/abs/1205.5767

A November 2016 paper by Scarpa, et al., updates this result on wide binaries.

Globular clusters at the fringe of a system are another place where dark matter particles are not expected to be present.

Non-baryonic Dark Matter (DM) appears in galaxies and other cosmic structures when and only when the acceleration of gravity, as computed considering only baryons, goes below a well defined value a0=1.2e-8 cm/s/s. This might indicate a breakdown of Newton's law of gravity (or inertia) below a0, an acceleration smaller than the smallest probed in the solar system. It is therefore important to verify whether Newton's law of gravity holds in this regime of accelerations. In order to do this, one has to study the dynamics of objects that do not contain significant amounts of DM and therefore should follow Newton's prediction for whatever small accelerations. Globular clusters are believed, even by strong supporters of DM, to contain negligible amounts of DM and therefore are ideal for testing Newtonian dynamics in the low acceleration limit. Here, we discuss the status of an ongoing program aimed to do this test. Compared to other studies of globular clsuters, the novelty is that we trace the velocity dispersion profile of globular clusters far enough from the center to probe gravitational accelerations well below a0. In all three clusters studied so far the velocity dispersion is found to remain constant at large radii rather than follow the Keplerian falloff. On average, the flattening occurs at the radius where the cluster internal acceleration of gravity is 1.8+-0.4 x 10^{-8} cm/s/s, fully consistent with MOND predictions.

Scarpa et al., "Globular Clusters as a Test for Gravity in the Weak Acceleration Regime" (2006) https://arxiv.org/abs/astro-ph/0601581

There is also a more recent August 2010 paper and a more recent September 2016 globular cluster paper by some of the same authors.