This manuscript examines the effect of adaptation to larval crowding on heat shock tolerance in Drosophila. Since crowding is not directly related to heat stress, the idea is that crowding triggers a general stress response which could cause resistance to hear stress as a byproduct. The authors found some support for this prediction. Males from the crowding-adapted populations indeed showed greater resistance to heat shock than males from control populations, although this effect was depending on rearing density. There was no difference for females. Larvae from crowding-adapted populations also had higher expression of a heat-shock protein (Hsp70), although there was no difference in the adult stage between crowding-adapted and control populations.

Although I think the results in this manuscript are interesting, especially the evidence for sex-specific evolutionary responses, I found that the structure could be improved and there are a few places where I think more information is needed.

In the introduction, there is much focus on adaptive decoupling hypothesis, but this idea is never returned to in the discussion. Either the results need to be specifically discussed in relation to this hypothesis, or else the introduction needs to be considerably revised.

When it comes to the maintenance of the lines, I find the degree of crowding truly surprising. 800 eggs in 1.5 mL of food is ~530 larvae per mL compared to 10 per mL for the controls. At first I thought this was a typo, but since it's repeated in several places I assume it's correct. But in that case, I don't really understand point 2 on lines 96-98. 40 x 60 = 2400 individuals, while 24 x 800 = 19200 individuals. In what way does this result in similar population sizes? Or is survival in the MCU vials so low that 19200 individuals results in the production of approximately 2400 adults? If so, this should be clarified.

It also wasn't clear to me why the high density treatment was different than the standard conditions in the MCU regime. It seems like it would have been interesting to carry out the experiments at a density that is representative of what the MCU flies usually experience. Maybe that's too harsh for MB flies and causes problems obtaining enough individuals to sample, but if so, this should be explained.

The statistical analysis for the proportion individuals surviving is appropriate, but I wasn't sure if the same was true for the Hsp70 expression data. Was there only a single sample per replicate population? If so, then the analysis is fine, but if not, then replicate population should be included as a nested random effect to avoid pseudoreplication. Please clarify.

The interaction effects for heat shock survival are somewhat complicated and the text in the results is hard to follow. It's good that the results of the pairwise comparisons are included in the supplemental material, but I think these results should be included in the figures somehow, e.g. as letters denoting homogenous subsets, which is probably the simplest option when there are so many possible comparisons. Related to this, I don't find the discussion of the main effects on lines 199-205 particularly helpful since the interaction effects are so extensive. I'm not dogmatic about never looking at main effects in the presence of an interaction, but in this case it doesn't seem particularly illuminating.

I found the structure of the discussion to be kind of confusing. No real mention is made of the adaptive decoupling hypothesis which is a major part of the introduction. And I didn't find the order particularly logical. According to my reading, the order of topics by paragraph is as follows:
1. General summary
2. Sex-specific evolutionary response in heat tolerance in MCU regime
3. Effect of larval crowding on adult heat tolerance in the control regime
4. Increase in larval Hsp70 expression in MCU regime
5. Effect of larval crowding on Hsp70 expression (in MCU regime? MB regime? Both?)
6. Hsp70 expression in adults (both density treatment and regime)
7. Conclusions
I find this confusing because the discussion jumps back and forth between discussion the results of adaptation to larval crowding and within-generation effects of larval crowding. I think it might be more useful to discuss within-generation differences in the MB treatment first, and then compare these to the evolutionary response in MCU.

I was also surprised that there is no mention of potential trade-offs here. From figures 1 and 2 it seems like increased male survival after heat shock in the MCU lines comes at a cost when not raised under crowded conditions.

Minor comments

There are too many acronyms in the manuscript. I counted 6 (ADH, CLC, MCU, MB, HD, LD), not including standard ones such as RNA, Hsp70, etc. It is hard for the reader to keep track of so many, especially since the first two are only used in the introduction.

Line 45: "in itself is stress" should be "in itself is stressful"

Lines 63-64: This sentence seems to be out of place, and it's not clear why it's formatted as the beginning of a list.

Line 86-87: Were the control lines kept separate from each other for a number of generations before creating the MCU lines from them? If so, how long?

Line 101: 18, not 18th

Line 134: vail -> vial

Lines 256-257: I can understand that you don't want to speculate too much here, but it would be useful to briefly describe some sort of potential mechanism which could explain these differences.

Lines 279-280: Again, what is the potential explanation? This statement is very vague. To me this implies that you mean that the differences must increase over development and be greater in the later larval stages. Is this correct?

Figure 2: Both figure 1 and figure 2 are listed as showing results at 37C. I assume this is a typo and figure 2 shows 38C?

Figure 3: Were females used as the reference? This doesn't seem to fit with the description of the methods in the supplementary information.

I think that the revised manuscript is considerably improved, and that the authors have successfully addressed all comments raised in the previous round of review. I have nothing further to add.