A picky comment: I had trouble identifying the different lines
in the bottom panel of Figure 3 (I am getting old, I know...).
Maybe, using a different color scheme, or even better, since
you are using different line shapes, using all black may
help the `reading' of the figure.

--I switched to all black.

Figure 5: the 70/TIR and 160/TIR are very tightly correlated
and closely follow the models. However, there is a systematic
deviation of the data from the model lines for high 70/160
values. Is this something you think it is worth commenting
on?

--Sure.  There are relatively few 'normal' galaxies that hot, so it was
difficult to constrain the models there with so few data points.  I've added something along these lines to the text.

Section 5.4: one interesting characteristic of the LVL galaxies
is that they not only are located to the right of the starburst
sequence, but also generally to the right (redder UV colors) of
the normal SF galaxies sequence. This may result from the LVL sample
being preferentially later type galaxies (if one looks at Juan-Carlos
plot, his Irregulars have more or less the same position as the
LVL galaxies), but I can see that even the few earlier type
galaxies tend to crowd to the right of the Cortese curve.
The main difference from Juan-Carlos work is that he is looking
at subregions in local galaxies, while you are averaging
across the entire galaxy, which may average over a broader range
of stellar populations.

--That's a good point.  I've noticed this but had not yet addressed it in the text.  I don't quite know what to say.  To see if I made any computation errors, I made a sanity check by seeing if fluxes from the SINGS paper result in the exact same locations in this figure as in the SINGS paper.  They do---whew! :)  Maybe other surveys like SINGS are more biased towards actively star-forming galaxies, whereas LVL is simply observing "everything" out to 11 Mpc.  Hence our sample is more passively evolved than others.  This argument doesn't seem quite right, so I'm unsure what to say in the text.

On Figure 7, when you compare your UV-3.6 color with stellar
population ages, you may want to stress that the comparison
is indicative; if SFRs have been declining with time, you may
be measuring older ages than appropriate (since you have been
accumulating more low-mass stars in the past). I think this is
worth no more than one line-comments, given the uncertainties.

--Done