The cosmologists’ widely accepted “concordance” model asserts that only about 15% of the mass of matter in the cosmos is baryonic—made of protons and neutrons. The “dark matter” that predominates is thought to consist of particles yet unknown. Particle theory provides an attractive candidate: weakly interacting massive particles (WIMPs) predicted by supersymmetric extensions of the theory’s standard model. Presumably created in the Big Bang, those stable neutral particles, about 100 times heavier than the proton, would be abundant enough to account for gravitational effects observed in the rotation and clustering of galaxies. The Cryogenic Dark Matter Search (CDMS) collaboration has now released its analysis of all the data taken by its CDMSII detector in three years of looking for WIMPs deep inside an old Minnesota iron mine. CDMSII is a 5-kg cryogenic array of germanium and silicon crystals micro-instrumented to detect the recoil of a nucleus in a rare collision with a WIMP as Earth sweeps through the halo of dark matter presumed to envelop the Milky Way. An instrument of CDMSII’s limited mass was predicted to find, at most, a statistically marginal handful of WIMP collisions in a three-year run. The community is working to decide which of several competing detector technologies can best be scaled up to provide a robust result. In its final year, the detector recorded two collision events that showed no evidence of coming from the enormous background of electron recoils or from a neutron collision. But the group calculates a 23% chance that both events were background imposters that squeezed past the analysis cuts that reduced backgrounds a millionfold. So the paper makes no claim of significant evidence for WIMP interactions. But it does present the most stringent upper limits to date on the WIMP-nucleon scattering cross section. The figure shows those limits as a function of the putative WIMP mass, together with a range of predictions from supersymmetric theories. (Z. Ahmed et al., CDMS collaboration, http://arxiv.org/abs/0912.3592. —Bertram Schwarzschild