*****************************************
 *  Minutes of the meeting held on nov.5 *
 *****************************************

1. Marie Noelle reports that she is trying to extract
   the number of hadrons per bunch for the 95 data.
   This has been requested by Brigitte in order to
   cross-check the luminosity per bunch.
   Working with MINIs Marie Noelle gets incoherent results
   depending on whether she relies on the bunch train number
   or on the t0 from the Ecal wires.
   To clarify this, she is now starting the selection from DSTs.

2. Arnaud reports that he is advancing with the note and that he
   has generated 7 million qqbar at Kingal level to finalise
   the numbers for efficiency and gg background for the data
   from 90 and 91.

3. We went back to the old discussion on what is the 
   minimum invariant mass square used in the analyses.
   This bound is set in the Monte Carlo event generation
   when evaluating detector acceptances, and the same value
   should appear in the fitting functions.
   In the latest Koralz version this bound is hard wired
   to 4*(m_tau)**2 . So this is the s' min to be set in MIZA
   for taus and muons (andrea's analysis).
   (What about Imma's analysis???) 
   For the Bhabhas the situation is more complicated. There
   are 3 steps involved:
   1.efficiency correction using UNIBAB 
   2.t-channel subtraction with ALIBABA
   3.Acceptance correction to evaluate impact of cuts on costheta*
     and acollinearity with KORALZ.

   Step 3 clarly uses the same s'_min of 4*m_tau**2, so it is coherent
   with the other leptons.
   It was not clear at the meeting whether we should also worry about
   steps 1 and 2.

4. The discussion went on on how to deal with events with fermion
   pair production (llv).
   
 - Pair production from the initial legs.
   This situation is treated in the fitting program together with 
   the beam spread corrections, since it is equivalent to a smearing
   of the beam energy.
   But the Monte Carlo does not contain a correction for these pairs, 
   therefore the acceptance is not exact. 
 - Pair production from  final legs. 
   These type of events should be kept since they contribute to the 
   physical width, but again KORALZ doesnt have this type of evts
   and therefore a 4-fermion generator should be used. 

   Clearly it should be kept in mind that pair production is suppressed
   by a term proportional to (alpha/pi)**2), so that we are talking of 
   small effects.

   Andrea is trying to clarify these points with the help of Manel.

   Next meetings.
   -------------

   We will meet again next tuesday 12 nov. at 14:30 
   in room 13-3-005 to review what
   will be discussed at the EW meeting of friday 15.