More discussion of AN dependence on FMS event topology

 I have been looking at run 12 events with a 2 photon  pi0 in a cluster of angular size 0.35 Radians with
a tight pi0 mass cut : Abs(M12-.135)<.08 but no soft energy cut. I then looked at 
three classes of global event topology. 

1) Events (red squares) with additional photons (E>6 GeV) in part of away hemisphere (cos(phiaway-phi)<-.5).
2) Events ( blue circles) with no additional photons. (2 and only 2 photons)
3) Events (green triangles) with additional photons (E>6 GeV) in near hemisphere (cos(phiaway-phi)>0)

For 2 photon 35mR clusters, but additional photons in the event, phi distributions of extra photons is shown below
events (part of event sample).

Phi dist