Referee 2's Review:

**********
N.B. These are the comments made by the referee when reviewing an earlier version of this paper. Prior to publication the manuscript has been revised in light of these comments and to address other editorial requirements.
**********

The manuscript describes functional characterization of natural Mth alleles using quantitative complementation tests. To my knowledge, this is the first evidence that one of the well-known candidate "aging" genes is functionally relevant to lifespan variation among natural populations. Therefore, the novelty and importance of the data is high.

Technically, the experiments appear to have been carefully conducted and the data analysis is generally appropriate. Quantitative complementation tests are sometimes controversial, but I think the authors have chosen the most appropriate method to measure life history effects of a moderate number of naturally occurring alleles. Precise allele replacement experiments might be the gold standard for such tests, but they are too laborious to extend to the number of different alleles (8) that are compared here.

The manuscript is well written and appropriately summarizes the background information.

I have some suggestions to improve the clarity of the manuscript in places. Overall, however, I am quite enthusiastic about this study.

General Comment:

I think it would be useful to have a table reporting the phenotype means for each genotype, perhaps as Supplementary Information.

Specific Comments:

p. 6. States that flies derived from natural populations were made isogenic for the 3rd chromosome. Were the remainder of the chromosomes (X, 2 and 4) co-isogenic across different lines? If so, this alleviates some of the concern about potential genetic background effects.

p.7 states that the 8 experimental lines were chosen on the basis of haplotype identity and nucleotide diversity at remaining sites. I would like to see more information here. What, exactly were these criteria (the 'nucleotide diversity' criterion is especially obscure)? Perhaps the authors could provide a table that summarizes the differences among genotypes.

p. 7 (bottom) Was CO2 anesthesia used in transfers?

p. 8. In the linear statistical model that is shown, which terms were fixed and which random? I'm guessing that all terms were fixed except for R(LxA), but this should be stated explicitly. Also it would improve clarity if it were stated here that "Line" refers to the lab strains (mutant versus wild type) and "Allele" refers to the wild strains.

p. 8 (bottom) I don't really trust JMP for analyses of mixed linear models (this applies to the fecundity analyses, not the lifespan analyses). I would prefer to see those tests implemented in SAS or some other software (like R) that appropriately handles mixed models. That said, the results appear to be so robust that I doubt that these analyses would change any of the main conclusions of the paper.