General. ---------- PCA: You mention this result in the second sentence of the Abstract and in the last paragraph of the conclusions. However, besides describing the method I didn't think there was much more of a discussion for this seemingly main conclusion. I might have just not read through well enough, but think you might need to just add a bit more of a discussion with regards to the Pricipal Component Analysis either in secton 5 or at least a couple more sentences in section 4.3. --You're not the first to say something along these lines. I have revamped this section somewhat. SED binned by IR/UV: I had really hard time with this section (not because I didn't believe you or find this really interesting...I actually thought this was all really interesting. However, the whole description of the binning procedure is not very complete. For example: 1) What are the bin sizes? I see they are on the Figure, but it would be nice to have a sentence or 2 describing the full range of the TIR/UV ratios along with the widths of the bins in the text. 2) What is the bin numbering scheme and which galaxies are in which bin? I know I can look at each galaxies SED to find the TIR/UV ratio...then figure out what bin the galaxy belongs, but a table listing the galaxies in order of TIR/UV ratio (possibly seperated by bin number using headlines) would make this whole discussion much easier to follow). --Hmmm...interesting idea, but I fear that this paper is already table-saturated. Introduction: ------------- Paragraph 2. "This broadenting in the trend has been ...(versus current) star formation[, and ](e.g. Bell et al. ...)." Remove the ', and'? or was something meant to follow? --Fixed Paragraph 4: I think you could use a little better transition between this and the previous paragraph. Why not start this by: "[The paper is outlined as follows.] Section 2 ..." --Done Section 2: ---------- Last sentence: "...maximum distance of ONLY 30 Mpc" could probably drop "only" --Thought about dropping it, but I guess I like it. Section 3.2: ----------- Paragraph 2: "Most images have photometric ..." (last sent) Why not quantify "Most" here? --Done (all BVRI images for which I've extracted fluxes do, in fact, meet the 5% level. There are some 10% Halpha images that I don't deal with here, plus some RI images that we didn't calibrate and that I thus don't use). Paragraph 3: This bit was a little confusing because you state two (last sent) colors used for ID'ing foreground stars (ie. f3.6/f8 and f8/f24), but then give an example using f24_stellar (ie. f8/f24_stellar > 8)....you never introduce what f24_stellar is. --Fixed Section 4.1 ----------- Paragraph 1 You pick our n1404 and n1482 to exemplify transitions from stellar to dust emisison in the MIR ...one is an E and the other an SA..you might want to state this as it directly shows how early-types differ from spirals. Section 4.2 (for general comments see above) ----------- Paragraph 3: I guess I just didn't really get the point of the discussion of spreads since in some cases you were comparing spreads of a single bin with that of the spread for a number of bins. Actually, I was confused as to what you exactly what you meant by spread...I assume you are quoting the dispersion (std. deviation) of the bins, but I thought you could also be quoting the full range of the bins. --I've changed the wording a bit to be more clear. Section 4.3 (for general comments see above) ----------- Section 5.2 ------------ Paragraph 2: "...,which [starves] the ..." (2nd last sent) I thought "starves" sounds funny, how about "inhibits" --ok Section 5.3 ------------ I guess when I quickly read through your draft in Crete this section didn't jump out at me the way it did now. I think this is a really neat result which has been presented using a very clever method. Nice work. Paragraph 3 You of course reference GXH 04 when mentioning that f24 strongly traces dust from HII regions, but this was something we clearly demonstrated and concluded in our FIR/radio (so you could ref. if you want) --Done Section 5.4 ----------- Paragraph 2: While it may be written correctly, it was a bit (last sent) confusing. I think you need a comma after "Hubble type[,]" or possibly some rewording. SSFR>0.9yr-1, and the later the Hubble type[,] the larger.." --Done Paragraph 2: You state the following results. TIR/UV SSFR Early type galaxies: increases increases => Does this just mean that TIR increases faster than UV for for early type galaxies? Does the Ks lum. need to change? TIR/UV SSFR Late type galaxies: decreases increases => Does this just mean that the UV increaes faster than TIR in late type galaxies? Again, does the Ks lum. need to change I found these results interesting and was hoping for some more discussion to follow. --I've changed the explanation, based on input from several.