S.L.Olsen 10/3/95
             A First Look at psi'-->omega(780)pi+pi-

   Initial results from an examination of the following psi' decays,
   using the cuts from the psi'-->rho a2 search (sans the omega veto)

           psi'-->omega pi+pi-
 
           psi'-->omega f2(1270)
                           |
                            -->pi+pi-

           psi'-->b1(1235)+/- pi-/+
                     |
                     -->omega pi+/-


  14% rule expectations:  (using acceptance = 0.035 from IHEP psi' group paper)

     -omega pi+pi-   Br(J/psi-->omega pi+pi-) = (7.2 +- 1.0) 10**-3

      expect: 3.44M x 7.2 10**-3 x 0.14 x 0.035  = 121 events
                                                    |
                                               see 46 (+/-9) events
                                              ***omega fit results***
     

     -omega f2(1270)   Br(J/psi-->omega f2) = (4.3 +- 0.6) 10**-3
        (we only look at f2-->pi+pi-;  56% of possible decays)

      expect: 3.44M x 4.3 10**-3 x .14 x 0.035 x 0.56 = 41 events
                                                          |
                                                   see < 15 events (95%CL)
                                                 ***f2(1270) fit results***
                                                  --a suppressed VT mode?--

     -b1(1235)+- pi-+ Br(J/psi-->b1+- pi-+) = (3.0 +- 0.5) 10**-3

      expect: 3.44M x 3 10**-3 x 0.14 X 0.035 = 51 events
                                                  |
                                          see  48 (+-9) events
                                       ***b1(1235) fit results***

  Do all of the omegas in the pi+pi-pi+pi-pi0 data come from 
  psi'--> b1+/- pi-/+ decays?  (In J/psi decay omegas from omega f2
  are at least as many as those from b1 pi.)

  ----data, cuts, etc:

Data:  Use all psi' data. 
        
   Require:  4 charged tracks with |cos theta|<0.74
               Sum charge = 0
               le 1 trks identified as muons
               le 1 showering trks (showering trk = ESHC > 0.8 GeV)
             between 2 and 10 neutral trks
               E > 50 MeV 
               at least 2 ESC layers with significant energy
               angle from nearest chgd trk > 26 deg.  
             
             4-C fit to pi+pi-pi+pi- gamma gamma with chisq<15.  
             (use gamma gamma combination with lowest chisq)  

             No pi+pi-pi0 mass combination within 50 MeV of the J/psi
 
             pi+pi-pi+pi- mass < 3.35  (eliminate chi states)
    
             gamma gamma mass within 35 MeV of the pi0
              
             Unmatched gamma energy<200 MeV

             "omega" = pi+pi-pi0 mass within 80 MeV (2 sigma) of the omega


Results:  (Fig 1a-c = "omega pi pi results"; Fig 2 = "omega fit results"
           Fig 3 = "f2(1270) fit results"; Fig 4a-d = "b1(1235) results"
           Fig 5 = b1(1235) fit results.) 

          -plot pi+pi-pi0 mass vs pi+pi- mass (Fig 1a, 4 entries/event)
         
          -project pi+pi-pi0 mass (Fig 1b)
            Fit region near omega to a gaussian plus polynomial bkgd (Fig 2):
               find 45.6 +- 9.0 events  (this plot has an additional pid cut)  
               mass=0.780 +/- 0.005  <---excellent agreement with 0.782
               FWHM = 61 +/- 11 MeV

          -plot pi+pi- mass recoiling against the omega (Fig 1c)
             Fit to poly bkgd + BW with f2 mass (1.27) & FWHM (0.185) (Fig 3)
               find 4.1 +/- 7.7 events
          
          -plot pi+pi-pi+/-pi0 mass, no omega cuts (Fig4c). Very strong
           b1(1235) signal

          -plot omega pi+/omega pi- mass (2 entries/"omega") (Fig 4d)
             Fit to poly bkgd + BW with b1 mass (1.235) & FWHM (0.142) (Fig 5)
               find 62.3 +/- 10.8 events;