Here are a few minor comments about the second draft of the paper (dated August 25)
 
Page and line numbers  refor to the double spaced version.
 
                      Cheers daniela

Page 1 line 21

X denote other high-ET final state particle => X denotes other high-ET final state particles or  X denotes another high-ET final state particle

==> Done.

Line 24

New photon timing system and a new model => New photon timing system and a novel model

==> We're sure it's not a novel technique, just new to CDF. Prefer to leave as is.

Page 2 line 10

Both are used to identify. Since the line before ends describing a pair of endcaps, it sound like "both" refers to the end-caps.

==> Done as "The calorimeters are used to identify..."

Line 14

"Our diphoton events with two photons with  " sounds funny

I suggest

Our analysis begins with diphoton events passing the ... The combined ....efficiency is effectively 100% if both photons have |eta <1| and ....

==> Done.

Page 3 ...is there another term we could use instead of "pathologies"

==> We couldn't find any and so we prefer to leave it as is unless somebody
else find something else.

Fig 1

Do we need the arrow ?

==> It shows our optimal cut separate bkg and signal well and we want to show this more explicitly with the arrow. Prefer to leave as is.

Fig 2

You do not use the same notation in the figure and in reference 6. It would be great if you could make it consistent. I personally like better N_messenger and M_messenger than N_m and M_m or N_mess and M_mess but at least they should be the same.

==> Fixed in the reference to be consistent in the notaion, M_m and N_m with other references, i.e. in Ref [17].