This is the mass distribution of Higgs boson candidates, detected by collecting events featuring four charged leptons (electrons or muons). The leptons are combined in pairs, and one pair (of same flavour) will usually yield a mass in the ballpark of the Z boson mass; the other pair will have lower energy if it is produced by an off-shell Z boson. Of course all four-lepton pairs compatible with quality criteria are kept, and the mass distribution thus contains a majority of events due to standard model production of ZZ pairs. This is the peaking background at masses above 180 GeV.
A different component in the histogram is due to decays of a single Z boson to four leptons. This rare process, seen for the first time by CMS two years ago, arises when a Z->ll decay is followed by an "internal" radiation of a energetic photon, which then converts in another lepton pair. One will then reconstruct with the four leptons a mass equal to 91 GeV, as for any other Z boson decay.
And then there is the Higgs... Which decays to two Z bosons, one on-shell (at 91 GeV) and the other off-shell (i.e. with mass of the two leptons far below the 91 GeV pole). In the gif one can see how the signal nicely grows in the 125 GeV region, to the very beautiful peak we have now with the total data accumulated in the 2011+2012 statistics.
The gif is available at https://twiki.cern.ch/twiki/pub/CMSPublic/Hig13002TWiki/HZZ4l_animated.gif .
This begs the question: will CMS also produce an animated gif of the diphoton mass distribution ? The question is not an idle one, since CMS has not updated their result on diphoton Higgs searches in the recent past. However, watch out for new results this week at Moriond!