Problem 519

In an attempt to understand pair production, I ran a trivial geometry and
measured pair production cross section versus energy for the standard package
and the low energy package.  By observing the resulting cross section and
comparing it to known values, I found that the standard package takes into
account both nucleus and electric fields (as advertised in the comments in the
code), while the low energy package accounts for the nucleus field only
(despite the claims in the comments in G4LowEnergyGammaConversion.)

The discrepancy was found in versions 4.1 and 5.0, however, the data files used
by G4LowEnergyGammaConversion didn't change in the latest version (5.2).  Thus,
there is no reason to believe that this bug was fixed in the version 5.2.

We need this bug fixed, since we are modeling gamma-ray silicon telescopes.
The difference in cross sections of pair production with and without electric
filed portion, is not negligible in low-Z materials (like silicon) - it can be
as large as few percent.

I have plots demonstrating the bug, and can e-mail it to all interested
parties. (novikova@ssd5.nrl.navy.mil).

Thanks,
-- Elena