The decay of the B mesons to muon pairs are quite rare -of the order of one in a billion- but very important, because they proceed via loop quantum diagrams within which existing particles may circulate. And since these loops are virtual, even very massive particles, even ones we do not know yet about, would produce a significant contribution. So measuring the rate of the rare decays allow us to gauge whether there is new physics in store for us, or whether there is just a desert of Standard Model physics awaiting us at the high-energy frontier.
The CDF and DZERO collaborations at the Tevatron collider have sought for these rare decays for two decades, and now the CERN experiments have taken over. The LHCb detector is quite well suited to spot these decays, since it looks at very forward-going proton-proton collisions, ones which produce B quarks in large amounts. Eventually CMS and ATLAS will take over (they have collected a larger integrated luminosity), so now is the time for LHCb to show their stuff. And they do.
The figure below, hot off the press, shows the LHCb data in a reconstructed dimuon mass distribution. A B_s signal seems to be there (the red component; the blue line is the total fit, which includes backgrounds), although the statistical significance is still in the range where fluctuations might be the real source of the bump. LHCb estimates it as a 3.5 standard deviations effect.
For more information on the analysis, see here. Tomorrow a seminar at CERN will provide all the details on this new search. What can be noted already, despite the fact that the decay has not been "observed" as canonically required with five standard deviations of more, is the fact that many new physics theories receive a hard blow by the very good match between Standard Model prediction and observed rate.