Dear Christian,

Thank you for the detailed reaction to all our comments.
As you suggested we like to follow up on a few of those.

Best regards,
  Tjeerd

General comments:
=================
2) As a result of selecting a common CM-rapidity range - with small
overlap - the value of the R_FB changes drastically from 0.94 to 0.094.
An alternative could be to extrapolate both measurements to a larger range
of 1.5 < y' < 4.5 to reduce the statistical dependence. Of course at the
cost of introducing systematic uncertainty due to the extrapolation.

* No, this is not correct. The text says "in the overlap region 2.5<|y���|<4.0". 
An extension of the range does not have any advantage. 
The reason for the small measured R_FB value is just the excess of candidates 
in the backward sample plus a slight excess of candidates at small rapidity 
values in the forward sample.

** If "excess of candidates" is mentioned with reference to the Pythia MC,
in the forward region we have an important lack of candidates (2 with
5.5 expected in 2.5<|y|<4 in MC and in the backward region (4 with 5 expected).
No excess of candidates is presented in Fig.2. Therefore, I cannot follow the
last part of your explanation.

3) A confusion of forward and backward boson production and the forward
geometry is introduced in the text. The footnote which defines forward
and backward contradicts with the text.
One suggestion is offered below, another would be to define forward and
backward early on in the text, referring to the asymmetric acceptance
of LHCb and the direction of the proton beams.

* We changed it at several places including line 28 which is now 
"Due to its acceptance close to the beamline, the LHCb experiment has 
the ability to probe xA down to very small values using data from 
proton-lead collisions where the proton beam points in the forward 
direction and to large values using collisions where the proton beam 
points in the backward direction."
including also your comment about 'the forward geometry'. 
We do not think that the footnote contradicts the text.

** The contradiction between also the new text and the defined 
"forward direction" is that the proton beam can never point in the 
backward direction. By definition "backward" is the lead beam direction.


Detailed comments by line:
==========================
Abstract:
- Replace "The Z production cross section" by
  "The partial Z production cross section" or
  "The fiducial Z production cross section".
  (It is not clear that the cross section is defined in a certain range.)
* Also previous analysis (e.g. Z/W in pp, Upsilon in pPb) just used 
"cross-section". It is probably a point that should be discussed in the EB reading.
** It is never wrong to be precise. Let's wait for the EB.
(Same for line 41 and 43.)


Line 10-14:
- State "from here on R_a^A stands for R_a^A(x_A,Q^2)".
* We think that this is obvious and so the sentence would be superfluous.
** Patrick ??

- Remove "while the prediction of RAa is due to the Fermi motion above 1.0
  for values of close to xA=1.0 [3]".
  (No written information is given about values close to xA=1.0 and to Fermi
  motion in Ref.[3]. There is also no experimental evidence for it.)
  And also remove Ref.[3] for this reason.
* Fig. 2 in arXiv:1401.2345 shows this effect clearly, so I think your 
suggestion is not good. (Be aware Ref. [3] contains two papers)
** Indeed a Fermi motion point is indicated, but again no written statement 
about R>1 at x>0.7 in arXiv:1401.2345. The only experimental points from SLAC
in Fig.1 are consistent (1 sigma) with unity. 
Therefore, no experimental evidence either. In conclusion, it is an artifact 
of the parametrization in Fig.2 in an unexplored x region.
Well there is in fact a measured small enhancement in sigma_Fe/Sigma_d at x=0.85
(Fig.3 in CERN courier April 26, 2013). How that works out in RAa, I don't know.
  
Line 19:
- Replace "deuteron-Au" by "d-Au".
* No, to avoid confusions with the d-quark.
** A confusion is improbable. The d-quark is not mentioned in our paper.
"deuteron-gold" could be an alternative, as "proton-lead" is also used
in line 29.

- Replace "points in the forward direction" by "points toward the detector" and
  "points in the backward direction" by "points away from the detector".
* We changed it to "Due to its acceptance close to the beamline, 
the LHCb experiment has the ability to probe xA down to very small 
values using data from proton-lead collisions in the forward direction 
and to large values using data in the backward direction.  " 
as the forward and backward direction is now defined further above.
** Suggestion to add "with particles detected" before "in the forward direction".

Line 35:
- Replace "This paper presents a first measurement" by
  "In this paper we present a first measurement".
* We have passive form throughout the paper.
** Suggestion to change to "In this paper a first measurement is presented".

Figure 2:
- Add "Z" to the axis label "y'Z" or we may write "Y^CM_Z"
  to contrast against y^lab_Z.
  Same changes in lines 196 and 220 and corresponding equations.
* We changed y_Z,lab to y_lab. Explicitly mentioning Z is not required here.
* y' is changed to y. This should be unambiguous.
** Make sure we do define y as centre-of-mass rapidity per proton-nucleon
pair as we do for sqrt(s_pN). I later realized that y' was only vaguely
defined as "Z rapidity in the centre-of-mass frame of the two beam
configurations" in line 196. 

  
Line 125-126:
- Replace "comes from the GECs, These cuts are based on" by
  "is related to the global event cuts, based on".
  (Explaining epsilon^GEC with GEC does not help. GEC is already defined as plural.)
* We changed it to "The first term, ��GEC, is related to the GECs, 
based on the occupancy in the vertex and tracking detectors."
** Patrick ??


Line 135-136:
- Also here "track multiplicity reweighted Z candidates" is not clear.
* In our eyes it is rather obvious that the reweighting referred here to, 
is the one described before.
** The replacement of "difference" by "ratio" helps to figure it out. Thanks.

- Replace "as a function of the muon pseudorapidity" by
  "as a function of both muon pseudorapidities".
* No, we determine them as a function of the pseudorapidity of one muon 
as the efficiencies of the two muons factorize.
** If the two muons have different pseudorapidities both efficiencies enter.


Line 142:
- Also a remark about the common cut in y^CM-Z would welcome.
* The common cut in y' is only relevant for R_FB.
** In Eq.(3) efficiencies are used that may be different from
those in Eq.(1). Or they can be same but corrected by beta.
It seems natural to discuss this in section 4.

Line 163:
Are all systematic uncertainties uncorrelated?
* Yes, as they have been analyzed with different samples and techniques there is no reason to assume any correlation.
** Patrick??


- What differential measurements? Like in figure 2?
  If so add "as a function of the rapidity of the Z boson".
  Or better, remove the last sentence with "expected in the near future".
  (We don't expect any pA data in the "near" future. Specifically, we do not
  expect any more pA data at 5 TeV at all.)
* Fig. 2 can hardly be called a differential measurement, but possible 
measurements would be in terms of y, p_T, phi*, etc. 
So we'd like to keep it like this as otherwise the sentence gets rather long.
* We removed "expected" and "near".
** Patrick ??


Best regards,
  Tjeerd for the Nikhef group