This manuscript presents the results of an interesting experiment that artificially selects up and down on death feigning behaviour in Tribolium. While I think that the paper contains some interesting data and results, it needs substantial work including reanalysis before it could be published. Below I provide some suggestions for ways to improve the manuscript (in order of appearance), but please note that this list is not exhaustive and I think the paper needs very careful revision throughout.

Line 9-12: Rather than starting off with how your study differs from others provide background to the question and information on what we do/don't know about this topic.
Line 16: I think this is wrong. Your results show that selection for longer duration of death feigning leads to longer latency to predation
Line 37: Briefly summarise why this is disputed
Line 43 - 52: I agree that good artificial selection experiments require sufficient replication, but I don't think this point needs to take up an entire paragraph of your introduction. I'm more interested to read about what we already know about the evolution of death feigning behaviour, and what might constrain it for instance - talk about the biology/ecology of this topic.
Line 75: at what densities/population sizes?
Line 105: How did you maintain selection lines? Did you continue selection in the same way throughout the experiment? How many generations did you select for? Did you relax selection before you measured traits and ran your predation trials? More detail is needed
Line 108: Need to introduce predation trials with overview - what generation was this done in what was the purpose.
Line 11-113: I find this very confusing are you saying that each predator was used in 6 trials one with a beetle from each line? If so how did you control for potential order effects? how much time between trials? Is it possible that predators became satiated. More detail needed.
Line 115: I thought artifical selection was done with lines of T. freemanni??
Line 127: How old were beetles? what were your sample sizes?
Line 137: Sample sizes???
Line 160: the description of your analysis needs more detail. How was replicate included? Should be nested within selection regime but interations in your results suggest that this is not what you did. You can't have an interaction between replicate and selection regime because your design is nested.
Figure 2 is not really needed you can see this information in figure 1
Figures 3 & 4: It would be better to put the selection regime on the x-axis and group the data according to replicate ie 2 groups of 3 bars
Line 205 - 219: Discuss your results not your methods (unless discussion of your methods offers insight to why your results might differ to other studies)
Line 220 - 229: You compare your results to a previous study, but its not clear whether this is the only previous study that provides similar results or the only previous study that does the same thing
Line 230 - 238: This is more interesting than the stuff that precedes it and could be developed more
Line 241: "is a main controller" be very careful not to infer causation, locomotr activity evolved as a correlated response - that does not mean that it is the cause
Line 239-245: This is interesting and discussion of this relationship could be developed further - potentially links with locomotor activity could constrain the evolution of death feigning behaviour
Line 246 - 255: Again shouldn't really be testing for intereactions with your design
Line 256-257: What is the evidence for this
Line 280: Throughout this paragraph I suspect you mean seconds not minutes, at least your methods and results talk about seconds and this seems like an awfully long time to be feigning death

I have now read the revised version of the manuscript “Death feigning as an adaptive anti-predator behavior: further evidence of its evolution and for field population data”. Although the authors have addressed most of my specific comments they have not addressed my more general comments asking for an introduction that introduces what we already know about death feigning and for careful revision throughout. I apologise if my comment was not made explicit enough previously, but I cannot recommend acceptance of this manuscript in its current form – it needs a thorough revision for readability and clarity, and in some places logic. It must also introduce what has been done previously on the topic in a thorough way. This will require substantial rewriting. The comments I provide below should help, but I expect to see a thorough rewrite of the entire manuscript for it to be acceptable for publication.

Lines 2-3: Perhaps change title to “further evidence for its evolution from artificial selection and natural populations”
Lines 10- 13: Rephrase to “Previous studies on Tribolium castaneum have shown that prey which death feign have a fitness advantage over those that do not. Whether these effects are repeatable across species or whether they can be seen in nature, is however, unknown.”
Line 16-17: reword: “collected from field sites with and without predatory bugs.”
Line 20 – 23: I think these two sentences could be reduced to simply say – “demonstrating that previous results are repeatable across both prey and predator species” you can go into more detail about the predators in the discussion.
Line 23: I think it would be best to state the species name so that people don’t forget and think it is the same species as experiment 1
Line 23 - 25: reword: “In the second experiment, wild beetles from populations with predators death feigned for longer than wild beetles from predator free populations.”
Line 25 – 27: reword: “When combined our results from the two experiments, and those from previous studies, provide strong evidence that predators drive the evolution of longer death feigning durations.”
Line 41: here, do you mean there is little evidence that it is adaptive or little evidence of what drives its evolution? Revise to make this clear
Line 42: Not sure this sentence should start with “on the other hand” perhaps “further” would be more appropriate? Also, do you mean heritable variation in death feigning behaviour or the survival benefits of death feigning behaviour – the former makes more sense to me, but what you have written sounds like the latter.
Line 49: change to “from field populations with and without the predatory bugs.”
Line 53: change to “using different prey and predator species to that used previously” After this sentence perhaps state why this is important – simply because it provides across species replication, or because the species you use is more suitable???
Line 60: This sentence needs revising for correct gramma
Line 60: This and the previous paragraph doesn’t really feel like an introduction, but rather a synopsis of your methods. I would like to see a more thorough synthesis of what work has been done previously on this topic (for instance what we know from studies like Nakayama et al 2009 and others). And how your study extends on these (not simply that you provide stronger evidence or more replicates but how your study will improve our understanding of the evolution of death feigning behaviour.
Line 64 -68: I don’t think this journal usually ends the introduction with a summary of the results, perhaps it would be better to give predictions?
Line 98: rephrase – “…time between touching the beetle and observation of its first visible movement, such as movement of a leg or antenna.”
Line 100: reword – “All observations in this experiment were conducted by one person (KK).”
Line 101: would be good to give an indication of how often this occurred.
Line 104-107: rewrite this sentence to make it clearer that 14 days after eclosion is when the behavioural trials were conducted.
Line 107-109: So you selected 20 individuals to go into the next generation – is that 10 males and 10 females? and how many did you actually measure the behaviour of? This information is needed to enable calculations of the selection pressure (which you should provide). Ok I see this information in line 112. However, I think it could be reorganised to make the flow more logical .
Line 119-131: This paragraph needs to be weaved into the text better, at the moment it doesn’t flow well. I think you could focus more on the reasoning for using 3 replicates, rather than focusing on what previous studies did.
Line 138: Not sure you need to say “the experiments was designed as below.” You describe the design in this paragraph, by saying “below” it makes me think it is going to be in a later paragraph.
Line 138-140: this paragraph still needs to be revised to make the design clearer.
Line 154: It would be good to show the sample size for each strain type. Also move “(7-21 days old, virgin) into text for better readability. For instance this info could go into the next sentence eg “For each trial a virgin beetle was placed in a petri dish without food when they were between 7 and 21 days old.”
Line 156: reword sentence for readability “Locomotor activity was measured by an infrared light beam which passed through the clear petri dish and projected onto a photomicrosensor…”
Line 166-172: So you didn’t measure body size of the individuals used in predation or locomotor trials?
Line 188: delete “those for the method”
Line 197 and throughout: you refer to “selection strain” throughout, I think a more usual terminology is “selection regime” to me “strain” would be the same as “line”
Tables 2,3, & 4: These tables don’t show the results for the nested effect of line. I think it is important to report all results from the model.
Results section: I think it would also be nice to add some information about what happened in each replicate line. That is it would be nice to know whether the direction of the effect was the same in all the replicates or is it a really strong effect in one replicate driving the pattern. This would give extra confidence in the conclusions.
Throughout results and elsewhere: For clarity, I would spell out lines selected for longer/shorter death feigning durations rather than using the shorthand L strains/S strains.
Line 238: change to “Females were heavier than males”
Line 242: Here the important information about the population is not the location but whether predators were present or not – so I would revise to “… beetles from the population with predators (ie the population from Okinawa) showed… than the predator free populations (ie. base lab population).”
Line 248: This sentence doesn’t make sense. Also I think that although your study might show stronger evidence than the Miyatake study the more important point is that it provides additional evidence, and that the evidence comes from another species – suggesting that it is a general phenomena. Revise.
Line 252-262: This gramma in this paragraph needs to be revised for readability and clarity.
Line 259-260: This seems like an important addition in your experiment, would have bben nice to talk about this more in your introduction
Line 260-262: This sentence does not belong here. Or you need to weave it in better
Line 263-274: This paragraph needs revising, the logic makes a bit of a jump. I don’t think you can say that all this evidence in flour beetles with invertebrate predatos can explain the evolution of death feigning across the entire tree of life. It suggests the pattern is more general, but there are limits to that generality.
Line 277: I still disagree with this, what is the previous evidence that “activity” is the controller of death feigning. I think you need to explain this logic more fully. Genetic links do not mean that one “controls” the other
Line 275-290: Gramma in this paragraph needs revising for readability and clarity. Also some of the information presented here should go in your introduction.
Lines 291-294: As stated in my previous review you cannot have interactions with nested factors. I suspect this has not been revised since you reanalysed your data.
Lines 291-301: The point of this paragraph is not clear to me, it also needs revising for gramma.
Line 302-310: Did you also find sex differences in death feigning behaviour? If not what does this tell us?
Lines 311-323: this paragraph needs revising for clarity.
Lines 324-350: This could be made more succinct and reduced to one paragraph.
Line 351-352: My understanding is that one population is from the field and the other is from the lab. Is that correct? So in actual fact there could eb many things other than just the absence of predators that differ between these populations. I would make this clearer

This is the second time I have reviewed this paper, and I must say it is much improved and the authors have done a good job of addressing my previous comments. The English is much improved, although small errors still remain (I think this is a problem associated with having English corrected by professional translators who are not necessarily familiar with scientific English and expressions) they generally do not detract from the meaning.
I must say that I do still feel that the introduction doesn’t do a fantastic job of introducing the background to the topic of the evolution of death feigning behaviour and instead focuses too much on what you did, but again this generally doesn’t detract from the scientific merit of the study.
Below is a list of sentences that I still found confusing, and suggestions for improvement.

Line 21-22: This sentence repeats what was said on line 15-16. I prefer the way it is written on 15-16, but think 21-22 is a more appropriate place to say it.
Line 42: wording here still a bit strange do you mean “studies showing survival benefits of death feigning behaviour along with estimates of heritability”
Line 42-43: I think this “ie” should be in brackets
Line 43: I would put a full stop after “limited” and start the next sentence with “An exception is previous work by Miyatake et al (2004, 2009) which does XXX”
Line 47-49: this is difficult to understand. Do you mean “…to investigate how selection on death feigning behaviour influences survival in the presence of the predatory insect Amphibolus venatar.”
Line 58-59: I’m not sure this sentence logically follows on from the previous. The locomotor activity stuff comes a little out of the blue. Perhaps a bit more of an introduction to this idea would help.
Line 66-73: This needs to be integrated better. And actually is probably better put in the methods
Line 79: This hypothesis is a bit weird. You hypothesized that the results from your experiment would provide evidence that predators drive the evolution of death-feigning behaviour in nature? But you don’t and can’t really test this.
Line 84: I’m not sure this line is needed
Line 275- 279: the writing here is a bit clunky. Perhaps restate as something like “The artificial selection experiment and population comparison presented here, provide evidence for heritable variation in death feigning behaviour as well as evidence for it’s role in deterring predation and increasing survival. Combined with previous studies showing similar results in other species our results suggest that these patterns can be generalised across species.
Line 290-293: I don’t understand why the fact that both predators are generalists means that prey immobility is effective across predator species and why that suggests immobility can be said to evolve under natural conditions. There is something wrong with the logic in this sentence.
Line 324: I don’t understand what you mean by “more test interactions”
Line 327 & 329: I’m not sure assumed is the right word here. Perhaps replacing with “could be” or “could have” would be better
Line 342: lifespan not longevity