The only direct comment I have on the text is that on page 14 you state nu_0=c/250, and its not immediately clear that c=speed of light (or at least it wasnt to me) why not put nu_0=1.2 THz (250um) instead?

*** Fixed

My major comment however is about the uncertainties for the model fits to the observed SEDs - As it stands the paper is complete, but it may be worth thinking about exploring the relative uncertainties for each parameters in the Draine and Li models, or more importantly the uncertainties in the BB vs DL comparison.  This can be done in several ways, such as monte-carlo sampling of the data within the observational uncertainties, or simply brute force exploring the chi-2 space to determine the probability distributions for the masses.

As I said I dont think this is necessary (as the paper reads as though it is complete), but may improve our understanding better for exactly how significant the difference in masses are between the BB and DL models.

*** A fair comment.  I would definitely be motivated to carry out a Monte Carlo-type analysis of the uncertainties in the fitted parameters if I were including a table of the fitted parameters, but I'll leave that to Aniano et al.