There is an asymmetry cut in the u-v matching algorithm, regarding whether a u by v = 1xN (Nx1) should be considered a 1x1 or 1x2 (2x1) case.  The value of 0.2 was chosen by Weihong based on studies which used the Iron/air MC.  Jim D. recently conducted a few studies showing that the cut could probably be lowered when the TSIU or TSP clustering algorithms are used.

Attached are files comparing the invarent mass, Z_gammagamma, and pi0 energy for four values of the threshold: 0.05, 0.10, 0.15 and 0.20.  Each page of the pdf is for a different pi0 pT bin, using the 12 runs in Weihong's run list taekn taken on day 156 (i.e. those starting 7156).  Due to the other cuts and the merging procedure, no significant difference is observed between the various values of the cut, and so it is decided to continue to use the standard 0.20 cut.