Review History

All reviews of published articles are made public. This includes manuscript files, peer review comments, author rebuttals and revised materials. Note: This was optional for articles submitted before 13 February 2023.

Peer reviewers are encouraged (but not required) to provide their names to the authors when submitting their peer review. If they agree to provide their name, then their personal profile page will reflect a public acknowledgment that they performed a review (even if the article is rejected). If the article is accepted, then reviewers who provided their name will be associated with the article itself.

View examples of open peer review.


  • The initial submission of this article was received on December 1st, 2017 and was peer-reviewed by 2 reviewers and the Academic Editor.
  • The Academic Editor made their initial decision on December 25th, 2017.
  • The first revision was submitted on August 6th, 2018 and was reviewed by 2 reviewers and the Academic Editor.
  • A further revision was submitted on September 17th, 2018 and was reviewed by the Academic Editor.
  • The article was Accepted by the Academic Editor on September 26th, 2018.

Version 0.3 (accepted)

· Sep 26, 2018 · Academic Editor


Thank you for your thorough responses to the expert reviewers. I am pleased to accept your paper into PeerJ at this time. This is an interesting study and I am sure the readers of the journal will find this paper appealing. As your paper moves toward publication, please take the time to check your proofs closely to ensure they are free of typos and that figures and tables are clear and correct. Congratulations on a nice contribution.

# PeerJ Staff Note - this decision was reviewed and approved by Ann Hedrick, a PeerJ Section Editor covering this Section #

Version 0.2

· Aug 28, 2018 · Academic Editor

Minor Revisions

The expert reviewers have now commented on your revision and have only a few, minor suggestions and corrections for consideration at this time. It is my opinion that with these minor changes your manuscript will be ready for publication.

Reviewer 1 ·

Basic reporting

Upon review of the revised manuscript, I accept with minor revisions. See Below.


Line 28 – 29: Delete N. Pustulatus was the most common species in the canopy, and the only species that was more common in the 6m traps than on the ground.

This was repetitive

Line 30 – 31: but our results suggest it is uniquely common in canopy habitats [delete instead that N. pustulatus]
Line 32: [delete between N. pustulatus and other co-occurring burying beetles.]

Line 47: Delete comma after Knollenberg
Line 52: same as above
Line 53: delete comma after Peck
Line 63-64: oviparous snake eggs in general; this idea was supported by an …
Line 65: insert comma after Beresford
Line 66 insert comma after Beresford
Line 74: Further studies have reported N. pustulatus to be more common in the
Line 84: delete that
Line 84: delete their
Line 85: only 1 individual [ not individuals]
Line 100-101: in southeastern Ontario, Canada. [and compared beetles trapped on the ground verses in the canopy]

Line 120: delete comma after Knollenberg
Line 123-124: Silphidae that were caught during our study
Line 133: birds, plants and insects
Line 141: Additional trapping locations at our study site
Line 149: At each trap location, we set two concurrent traps and collected them after…
Line 156-157: ripening and suspended above the saline preservative.
Line 160: Canopy traps were hung 6m high in tree branches; the exact distance between the paired ground and canopy traps …
Line 163: paired ground and canopy traps with the largest distance being within 20m).
Line 226: the Random Forest model provided an overall…
Line 236-238: We omitted traps when no beetles were caught, regardless of trap distance, because traps with zero beetles provided…

Line 274: trap being on the ground increased as abundance of both species increased (Fig. 6).
Line 275: add individuals after Nicrophorus pustulatus
Line 288: delete that

Line 294: Ground traps captured more species and more individuals of burying beetles than canopy traps
Line 295-296: The number of individuals and species increased as the season progressed from May to June, for both ground and canopy traps, indicating there is a greater abundance
Line 297: delete – (July) compared to early in the season (May)
Line 300: classification accuracy), indicating there were repeatable …
Line 302: replace number with abundance
Line 303: delete that
Line 304: replace number with abundance
Line 320: we successfully baited N. pustulatus using chicken, and it has been reported in the laboratory setting, N. pustulatus will behave like
Line 339: Furthermore, the breeding season of vertebrates in the …
Line 348: and their use of canopy resources suggests vertical height is another important…

Literature Cited:
Line 372 replace dash with –
Line 389: Journal should be italicized
Line 397: The Coleopterists Bulletin
Line 430: replace dash with –

Experimental design

The experimental design was relevant and repeatable.

Validity of the findings

Meaningful results were produced.

Reviewer 2 ·

Basic reporting

No comment

Experimental design

No comment

Validity of the findings

No comment

Additional comments

The authors have adequately addressed my concerns on the previous draft. The adjusted analysis uses mixed effects models to account for non-independence among samples, and the results are similar but more robust. The additional discussion of Random Forests and potential limitations is also helpful. Please see below for a few very minor suggestions.

Minor comments:
87: change to '1 individual' instead of '1 individuals'

92: change to "fewer than 4%"

293: I'm curious why you used log(abundance + 3) instead of log(abundance + 1). I don't recall seeing that before, since the reason is to avoid log(0). The legend for Figure 4 states 'Log(number of beetle per trap + 1)', so one of these might be a typo?

239-240: Perhaps change 'to improve the fit of the models' to 'to meet the assumptions of the models' or similar. This would clarify that rounding was done because of the assumption that errors are poisson distributed, and not as part of some odd fishing expedition for 'more significant' results.

Version 0.1 (original submission)

· Dec 25, 2017 · Academic Editor

Major Revisions

Overall, this an interesting study and a reasonably polished manuscript that will benefit from some further edits. Both reviewers make numerous minor suggestions for text edits and recommendations for moving some sentences and paragraphs to alternate sections. I agree that many of these edits will improve the flow of the manuscript for readers. I would also like to see the hypotheses that are listed in the introduction be touched upon in the discussion. It's clear the results can speak to them if effort is made to state whether they were supported or not and why/why not. These minor changes will make this already well-written manuscript even stronger.

The primary issue at this point, which is what puts this manuscript into the "major revision" category is related to the point raised by Reviewer #2 with regard to how trap stations are treated in analyses. At present, I agree with Reviewer #2 that the two different sampling times should not be treated as independent given that they are repeated measures with spatial autocorrelation highly likely in the data. There are multiple ways to handle this type of analyses and Reviewer #2 offers a valid and commonly used approach of treating the stations as a random effect. Another option is to compute the analyses specifically as repeated measures to account for the 'within-subject' factor. These two approaches are very similar and it is unlikely that alternate analytical approaches will yield substantially different outcomes but this could create issues for random forest analyses which are rarely completed with longitudinal data. However, the package "mvtboost" could support this approach. While random forest approaches are rigorous, variation in underlying error structure could still generate unreliable outcomes. If you disagree with this comment and wish to argue for the current analytical approach, you can choose to do so as long as you offer a clear substantiation in the form of peer-reviewed methods papers or a reputable text. I look forward to your revised manuscript.

Reviewer 1 ·

Basic reporting

The reviewed manuscript takes a novel approach to the examination and discovery of foraging height preference in burying beetles of Ontario. These authors provide data to suggest that Nicrophorus pustulatus prefers foraging in the forest canopy, thus allowing this species to coexist with other closely related species in the same time and space. While this is not the first report of canopy foraging N. pustulatus, the data do provide further evidence in its northern range, as well as presence absence data for sympatric species.

The wording of the “traps being distinguished by predictors” is an odd objective.
In general, the data suggest a preference for a ground or canopy trap. Determining a preference for either trap is important, thus it should be directly stated that these three species clearly have a preference. To me this is the difference between "Predictors verses Observed data".

Experimental design

The first section of the methods should be moved to the Introduction and perhaps the discussion.

Validity of the findings

No Comment

Additional comments

Review of “Ecological divergence of burying beetles into the forest canopy”

Line 25: The wording of the “traps being distinguished by predictors” is an odd statement.
In general, the data suggest a preference for a ground or canopy trap. Determining a preference for either trap is important, thus it should be directly stated that these species have a preference.

Line 27: delete Indeed

Line 101: Study Species: Discussion of the study family should be moved to the Introduction. (Lines 102–111).
State the number of species known to be reported in the study area, and provide reference. You may also provide pertinent information regarding N. pustulatus as this is the species of focus.

Line 114–124. Remove from the methods. These reports could be provided later in the discussion, if seasonality is discussed.

Line 131: State what bait is used here.

Line 140: basswood

Line 142: eastern white cedar

Line 144–145): red oak, etc… be consistent with capitalization

Line 172: The species and sex identification section is unnecessary. Please just state what resources were used for the identification of the study species.

Line 210 – indent paragragh

Line 243: Random Forest (be consistent either way stated)

Line 270–271: First sentence is unnecessary as same results are stated in next sentence


The entire first paragraph of the discussion is repetitive from the results section, as if restating. Be brief and combine with second paragraph.

Line 304: delete different

Line 306: delete (Fig. 1,2) – don’t need to refer to figures in the discussion.

Line 319: provide four previously stated references here from the Introduction.

Line 321: awkward sentence

Line 323: Our study supports the findings of Dyer and Price (2013), suggesting that N. pustulatus can be common, but only within the forest canopy.

Line 328–344: add Hocking et al. reference [Hocking, M. D., R. A. Ring, and T. E. Reimchen. 2006. Burying beetle Nicrophorus investigator reproduction on Pacific salmon carcasses. Ecological Entomology 31(1): 5–12.]. These authors found N. pustulatus on fish, and Dyer and Price (2013) suggest this as a possible resource, especially in areas of abundant raptor nests.
Literature Cited
Bold the authors, and Volume number. Also use a – between pages for range, not a dash.
Figure 3: State why Log transformation was used. Readers need to be reminded of differing N values.
Table 1: Carrion beetles captured in ground and canopy traps at …

Reviewer 2 ·

Basic reporting

This manuscript is very well written with clear language and structure. I had no issues with the literature or background, and the results and discussion were directly relevant to the clearly stated hypotheses and predictions.

Experimental design

1. The methods included some extraneous information (i.e., the description of how each species was identified -- this is in the key cited but should be moved to an appendix if the authors would like to include it) and a couple of places where more information is needed, detailed below.
2. Station should be included as a random effect in your analyses since each Station was sampled twice. It likely will not affect the overall conclusions, but currently n is being artificially inflated in the statistical analyses since collections from the same station are not independent.

Validity of the findings

The data seem adequate for the questions and analyses, and the conclusions the authors draw are justified (pending the station random effect noted above).

Additional comments

This study was really interesting and is a nice, concise blend of basic natural history and basic community ecology. The introduction is really well written. It flows smoothly, covers the relevant background well, and very quickly grabbed my interest. With a few (probably) minor changes to the statistical analyses, this article would be a good contribution to the body of literature on carrion beetles. Please find comments for particular line numbers below.

79-80: " only four studies" seems to imply that there have been many studies with baited elevated traps, but most were not successful in capturing N. pustulatus. Is this the intended meaning? Or rather that there have only been a few studies that used baited elevated traps?

91-99: Having the questions *and* predictions so explicitly listed here is very helpful.

160: "...hung 6m high in tree branches" -- could you clarify? Were there any guidelines for proximity to branches, distance from the branch holding the trap, etc?

177-203: This section seems unnecessary since this information is included in Anderson & Peck 1985 which you cited as your key.

208: Were there traps that did catch beetles that also showed evidence of disturbance?

210-222: I think you should include Station as a random effect in your model since each trap was sampled twice. It doesn't seem justified to assume that they're independent just because they are a month apart.

265: The 'n' in MuMIn should be italicized

304-317: Good, concise summary.

324: 21, 12, and 1 in how many traps? Is this a per-trap average? If not, that would probably be the most informative.

328: Doesn't the observation of larvae feeding on ringneck snake eggs mean that gray ratsnake eggs are *not* the only known breeding resource? One counterexample is all that is needed to show that 'only' is not reality. I recognize this is a bit nitpicky, but I think it's worth rewording to say either that this is a common (or historical?) assumption (despite the evidence laid out in the rest of the paragraph), or that they are thought to nearly exclusively reproduce on gray ratsnake eggs or similar.

Figure 3: Since random forests are still relatively uncommon in ecology, could you briefly explain the scale for relative accuracy in classification?

Figure 6: I think changing the y-axis to direct probability of presence rather than logit(prP)/2 would make this figure more effective. Constraining ylim(0,1) would then make the panels easier to compare. Also, what is the rug on the x-axis?

Table 1: For the 'Number of traps with each species present' column, could you add the corresponding % in parentheses?

Typo in 'round' description in Variable_descriptions.txt

All of your code ran with no issues and the results matched those reported in the manuscript. One minor suggestion for the future to help make your code easier to read is to follow the recommended style guide for R code:

All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.