Here are the first results for A_LT for kinematics A and D from my 
student, Dasuni Kalhari.  These are for all three foils and the  entire 
Q^2  and omega range.  This is the result of a fit of A + B cos(phi) to 
each missing momentum bin.  The plots show A, B and B/A.  B/A should be 
just A_LT.  A should be related to the cross section and B to the LT 
part of the cross section.  We did not propagate  errors (which might 
account for the bumpiness of the fit).

The A and A_LT results are not consistent from kinematics A to D, but 
that could be due to the Q^2 and omega bin size.

The A_LT results are somewhat consistent with previous results, because 
both start slightly negative and become increasingly negative with 
increasing p_miss.

We plan to
 a) plot the individual p_miss plots with phi fits to show the fit 
quality (so we can examine the bins with peculiar fits)
 b) bin the data in Q^2  and p_miss
 c) implement error propagation
in that order as time permits.