> Typo: section 2.4, magnetic and not magnetc

Fixed.

> Section 3.1: It may be better to add a line as to what Radler 1969
> and Radler and Stepanov 2006 found, and then compare with their result.
> From the present text it is not clear in what way the simulation results
> are "consistent" with the above papers.

In order to address this point and to compare with their results we
have now added the new Appendix B where the comparison is discussed
in detail.

> Section 3.2; second paragraph, The last line "If confirmed, the critical
> value.." is ununderstandable to me as it stands.

We fully agree, and don't understand it ourselves any more. We replaced
this sentence by: "However, within the error bars of our simulation
there is no support for the shear--current effect."

> Figure 11 caption: The last line of the caption should read I think
> "The slope D_etaS/Dalpha S = 1/2 (corresponding to k_1/k_f = 1/5) is
> indicated by a dashed line"

Thanks, well spotted! We corrected this.

> The authors say the Yousef et al results can not be due to the shear-current
> effect. Would they like to speculate if this is also due to the
> incoherent alpha-Omega effect?

We have added now the sentence: "However, the incoherent alpha--shear
effect seems to be a plausible candidate." at the end of the corresponding
paragraph.

> A minor complaint: Rotation is discussed in the paper in section 3.1 and then
> is not at all crucial in the rest of the paper. A line in the conclusion
> as to why the authors thought it important to include rotation would
> be useful for the reader.

In the section "Effect of rotation" we have now added the sentence:
"This also lends support to the validity of the test field procedure in
determining the sign of the off-diagonal comments of the eta_ij tensor
correctly." In the conclusions we added the sentence: "The validity of
the correct determination of the sign of eta_21 is supported by the fact
that the sign of $\delta$ agrees with the analytics results of Radler
(1969) and Radler & Stepanov (2006)."

> Appendix B: some more comments on how these stochastic equations are solved
> would aid the reader.

After (B5) We added the sentences: "We solve Eq.(B1) using a
third-order Runge-Kutta time stepping scheme. At each time step of
length delta t, the fluctuations of alpha_ij and eta_ij are taken from
a normalized Gaussian distribution, scaled by alpha_ij^rms/sqrt(delta
t) and eta_ij^rms/sqrt(delta t)$, respectively." and, after (B7),
"This equation is solved simultaneously with Eq.(B1) using the same
third-order Runge-Kutta time stepping scheme."