Crater II is a dwarf galaxy that should not look the way it does. It is enormous for its luminosity — roughly the size of Fornax but far dimmer — and its stellar density is so low that it barely registers against the background. Cold dark matter simulations struggle to produce anything like it. The galaxy is too big, too diffuse, too cold in its internal kinematics.
Atzberger et al. present the clearest evidence yet for why: the Milky Way is tearing it apart. Narrowband calcium photometry across 128 square degrees reveals tidal tails extending at least 7 degrees — about 95 kiloparsecs — from Crater II's center, with 37 candidate member stars identified in the streams. The galaxy has lost at least 25% of its initial stellar mass, and the tails extend beyond the survey footprint. The true extent of the disruption is unknown.
The metallicity gradient is the finding that constrains the history. Stars in Crater II's center are more metal-rich than those in its outskirts and tails, with a measured gradient of -0.34 dex per degree. This means the galaxy was once more concentrated — it had a radial structure, with processed material sinking to the center and primordial material in the outer envelope. The Milky Way's tidal field preferentially strips the outer, metal-poor stars, inflating the galaxy's apparent size while lowering its density. What we observe now is the skeleton of a galaxy that was once denser and more compact.
The dark matter question remains unresolved. Whether Crater II's halo is cored (flat-density center, as some alternative dark matter models predict) or cuspy (density rising toward the center, as CDM predicts) cannot yet be determined from the data. Both halo profiles can reproduce some of the observed features, but neither reproduces all of them. The tidal disruption complicates the measurement: the galaxy we observe is not the galaxy that formed, and the dark matter profile we infer is the profile after billions of years of stripping, not the original.
This is the fundamental difficulty of inferring initial conditions from disrupted systems. The process that makes Crater II interesting — its ongoing destruction — is the same process that obscures the information we most want to extract.