Tentative reply to second round of comments from PRD reviewer.
There are a couple things for the paper committee to comment on
if they wish, and one thing still in progress, noted with a []

> This manuscript should be published after the authors have made 
> changes and after they have considered some improvements to the
> manuscript.  The work is a valuable contribution to neutrino 
> physics measuring the Q2 distribution of quasi-elastic events 
> on nuclei.

> I am disappointed that the authors did not take many of the 
> prior suggestions into account to improve the manuscript.  They 
> commented extensively on my previous report, which was helpful 
> for me to understand the paper.  But much of this commentary 
> belongs in the manuscript so that the journal readers will also 
> understand the paper.  For a full length Phys Rev article it is 
> always better to err on the side of completeness.

> The following remarks are to be considered in addition to 
> remarks already made in the made in the first report.


> I suggest the title "Measurement of the quasi-elastic axial 
> vector mass in neutrino interactions on oxygen".  Section IIC 
> and the comparison to previous observations makes it quite clear 
> that the authors consider the oxygen mostly irrelevant.  There 
> is little discussion of the effects of Al, C and H on the result.

Changed.

[For the paper committee:  I guess we should change the title, just
to be expedient.

What I really want to communicate is not that they are "irrelevant"
but that there is no model (that I know of) that indicates that we 
should expect really large differences.  I'm not sure how to make 
this more clear, and I think there is discussion of this issue
in general.  But see also the very last comment below.]


> Could the authors please describe figures with more than one 
> plot on them in a more descriptive manner than "top" and "bottom".  
> In figure 6, for example, the dashed curve is sometimes on top and 
> sometimes not.  The caption uses "top" and "bottom" and is 
> ambiguous.  This caption problem was raised in the first 
> referee report.  Please do not use "top" and "bottom" to describe 
> different traces on the same plot.  Use "solid" and "dashed" etc.

Fixed.

> The first sentence of section IVA should have the words 
> "assumed to be" added to the definition of the muon candidate.  
> This assumption introduces some Q2 bias, particularly in the 
> non quasi-elastic sample.

"muon candidate" and "assumed to be the muon" seem to say the same
thing, though perhaps with different emphasis.  The text now
follows the reviewers suggestion.  The reviewer's comment caused
us to recognize an ambiguity in how we stated the related detail
in Section III-C.  It more clearly indicates that roughly 1% of 
the muon candidates will not actually be a muon.


> Please put the list of Q2 values in the paper.  In the response 
> to the first report it is stated "We can easily supply such a list."
> But it is not in the paper.  Such a table will be a valuable part 
> of the publication record.

Table is added (now presented as Tab. IV).

[Paper committee, please look at the table and the text, and comment
if you have any -- especially about the asterisks in the table.]

> In the discussion of systematic uncertainty, one might expect some, 
> such as the MRD errors to be asymmetrical.  But only symmetric 
> errors are used.

A good caution.  

The muon momentum scale error is effectively symmetric, because it
is a combination of correction and then the error on the corrected
result.  If we had not done the correction we describe, and instead
use the nominal momentum scale, and quoted a (much larger) error 
on the result, we would certainly be discussing an asymmetric error.

The other largest error, collectively coming from the uncertainty
in the Enu spectrum, is also effectively symmetric.

Among the smaller errors, a better model for the nonQE background
shape would likely produce a non-symmetric error, but when it is
folded in with the really large errors, it would not make a
difference, so we have chosen to present the result in the more
simple form.

> The rather odd appearance of figures 8 and 9, in which the fit to 
> all of the data lies at the edge of the fits to smaller subsamples 
> needs an explanation in the text.

We have added text.  [For paper committee -- I am doing one more
check to confirm my explanations here.  This will take a couple days.]

>
>
>

[For paper committee -- 

I propose to add one thing at the very end, based on a conversation
with a colleague in June.  I agree with his philosophy on this point,
and I think this reviewer does also.
Note the middle sentence inserted in the final lines of the paper.

  We do not assume that neutrino interactions on oxygen should be the
  same as for deuterium.  In this sense we have measured the effective
  $M_A$ for oxygen.  However, we find only a small effect on the shape 
  of the $Q^2$ distribution for $Q^2 > 0.2$ (GeV/c)$^2$ from known
  nuclear effects.

If you (paper committee) disagree with this addition, please say so.
Otherwise I will leave it in, and mention the change in the reply
to the reviewer.

]