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.
The paper is now satisfactory; I have checked it. It is ready for publication.
[# PeerJ Staff Note - this decision was reviewed and approved by Mark Young, a PeerJ Section Editor covering this Section #]
The MS has improved further. It still could use more proofreading, as errors remain. However, the issue of ACCTRAN vs. DELTRAN character optimisation on the tree for character 251 still has not been adequately dealt with in the MS. Changes to Table 1 and the new Figure help, but the critique that one cannot claim synapomorphy unless the phylogenetic analysis unambiguously shows a character state as ancestral at a node in question has not explicitly been addressed by phylogenetic analysis. Data from the character analysis on the tree must be explained to show if it does, or does not, matter whether ACCTRAN or DELTRAN assumptions are required. e.g. see here: https://onlinelibrary.wiley.com/doi/full/10.1111/j.1096-0031.2008.00229.x
Thank you for the revisions.
I've read through the rebuttal and revised MS and have some further comments. The MS is closer to being acceptable. But:
-Further proofreading is needed. There are improvements, but it still needs work. |It is difficult to summarise what is needed, as there are many and diverse kinds of errors needing amendment.
- Rebuttal--- "The main source of confusion that two hypotheses are presented as equally likely" I think this still isn't clear or maybe accurate in the MS. The key question is whether parsimony shows that hypothesis 2 is the only most parsimonious solution needs to clearly be demonstrated via the outcome of the phylogenetic analysis, not just stating that some coelurosaurs have the convex semilunate carpal as in Caudipteridae. It might even help to add a tree with the character mapped onto it, and data from optimising the character onto the phylogeny presented. The prior reviewer noted that ACCTRAN vs. DELTRAN assumptions influence this conclusion, and that needs to be clearly addressed in the MS. It is a bit confusing in the MS as Table 1 implies that only 9 taxa were measured?
-"detected as the most unstable taxa by " by what?
-Table 1 should mention that this is about character 251's states.
-the new figure shown in the Rebuttal is helpful, so why not include it in the paper? I have to agree with the reviewer that Character 251 is about 2 related things (convexity and SLC relative size), not just all about convexity. It doesn't hurt your paper at all to change the character wording to reflect this, so please do so. While prior studies have used "strongly convex" and this paper quantifies it, it is important for future work that the quantification is described in a clear way; i.e., that convexity and SLC relative size are integrated for character 251. I disagree that the states should not be modified.
Please make all of these corrections to avoid further delays. Thank you.
[# PeerJ Staff Note: Please ensure that all review comments are addressed in a rebuttal letter and any edits or clarifications mentioned in the letter are also inserted into the revised manuscript where appropriate. It is a common mistake to address reviewer questions in the rebuttal letter but not in the revised manuscript. If a reviewer raised a question then your readers will probably have the same question so you should ensure that the manuscript can stand alone without the rebuttal letter. Directions on how to prepare a rebuttal letter can be found at: https://peerj.com/benefits/academic-rebuttal-letters/ #]
The reviewer still has some concerns about the analysis. I am reluctant to send the manuscript back to them another time after this, but the authors need to do substantial work to make the manuscript acceptable, and this could continue as numerous more rounds of revision (with me first checking it; hopefully not further peer review) if this is not taken more seriously.
First of all, the entire paper needs thorough proofreading. We do not provide this as a standard service. You may want to hire a service. As it stands, it is not publishable quality of writing. This has been noted repeatedly with examples, but the whole manuscript needs polishing.
There are quite a few points where the phylogenetic analysis remains incorrectly presented or else needs to acknowledge (in the MS itself) uncertainty. In particular, the "statement that this is a significant change at the wrist in Oviraptorosauria is an assertion and not a conclusion" must either be qualified in the MS itself, or changed so that the 2 hypotheses are presented as equally likely, not favouring one or the other.
Please take careful time and attend to all of these issues so that there is no further delay to the manuscript.
**Language Note:** The Academic Editor has identified that the English language must be improved. PeerJ can provide language editing services - please contact us at copyediting@peerj.com for pricing (be sure to provide your manuscript number and title). Alternatively, you should make your own arrangements to improve the language quality and provide details in your response letter. – PeerJ Staff
Issues regarding English language
Thank you for making the changes to the lines I mentioned in the abstract, but these were only examples, and the manuscript would still benefit from a thorough copy-editing, especially prior to final publication.
Issues regarding reporting of methods and results
Thank you for the clarification on the taxa you have pruned to form your reduced strict consensus, and the references to the cladogram in the text as a reduced strict consensus. You should also include that it is a reduced strict consensus cladogram in the Figure 4 caption. I am now able to perfectly replicate your analysis. However, you have reported the tree length as 696 steps (line 205) and it should be 686 steps.
In your rebuttal letter you wrote: “Anomalipes zhaoi and Ganzhousaurus nankangensis were excluded from the reduced strict consensus tree because they were identified as the rogue taxa automatically.” However, you still did not include this information in the manuscript. You should tell the reader in the manuscript that Anomalipes and Ganzhousaurus were the taxa identified as rogue taxa by TNT and that those were the taxa pruned to form the reduced strict consensus.
Previous Issues regarding the description of character 251
Thank you for adding the clarifying diagram. It is now clear which measurements are contributing to your metric in character 251. It seems that one ratio is a measure of convexity and the other ratio is a measurement of semilunate carpal width in comparison to a measurement of manual width. So, the metric is not strictly a measurement of convexity, but convexity and relative size. You can choose whether you want to modify the language of the character states to reflect this.
Issue regarding the labeling in Figure 4
In your rebuttal letter you wrote “In addition, the illustration of the Figure 4 shows that the three characters at the node Caenagnathoidea are “The significant changes of the wrist” rather than the synapomorphies. So, this sentence should not be deleted from the Figure 4.”
The evidence in your manuscript does not support the statement that this is a significant change of the wrist at the node Caenagnathoidea. Because of the limited outgroup sampling in your matrix, the evidence in your manuscript says there are two equally likely scenarios:
1. that the transition from 251:0-->251:1 occurred between Herrerasaurus and Pennaraptora, and was reversed from 1-->0 on the stem of Caudipteridae (ACCTRAN)
2. that the transition from 251:0-->251:1 occurred independently in Avialae (Archaeopteryx) and Caenagnathoidea (DELTRAN)
Thus, the statement that this is a significant change at the wrist in Oviraptorosauria is an assertion and not a conclusion based on your data, because the possible locations of synapomorphy according to your matrix are both inside and outside Oviraptorosauria. Because you have said in the manuscript that it is not a synapomorphy, I suppose it is ok to keep it in the figure, but it is misleading because you cannot unambiguously conclude from the present dataset that this change occurred within Oviraptorosauria.
Issue regarding seemingly erroneous scores for character 251 in your matrix
In your rebuttal letter you wrote: “We have added all scored taxa in Table 1, including the Herrerasaurus and Archaeopteryx.” And, later you wrote: “The data of Oviraptor is not here because there are no clear images about the carpal of this genus. The data of Machairasaurus is not here because all the images about the carpal of this genus are not in the standard dorsal or ventral view.”
I understand why you have left them out of the table, but the taxa Oviraptor_philoceratops, Machairasaurus, and “Zamyn_Khondt_oviraptorid are still scored as state 1 for character 251 in your matrix. If you don’t have the measurements you need to change the states for character 251 to “?” for these taxa in your matrix. These were probably left over from the previous qualitative version of this character. Making these changes should not change the length of your most parsimonious trees nor should it change the topology of your reduced strict consensus.
Issue regarding characters marked as ordered in your matrix
I noted in the previous review that you had several two-state characters marked“+[/1” in the “Ccode” block of your matrix. This notation indicates that these characters are ordered, however it is impossible for two state characters to be ordered. It will not affect your results to leave them the way they are, so you can publish it this way if you want to but it is incorrect. You can verify this by loading your matrix in the GUI version of TNT and using the drop-down menus to check which characters TNT interprets as ordered in your matrix.
The MS has improved, although further proofreading is needed. Furthermore, there are still important issues with replicability of this study. We cannot consider the paper further until these are completely fixed and accurately documented in the paper itself. Please pay extreme care to fixing these issues, as per the reviewer's helpful points.
The English used in the manuscript has significantly improved since the first submission, however there are still grammatical errors throughout the text, including a few in the abstract, for example:
Change “to protect the pennaceous feather when they relax” to “to protect the pennaceous feathers when they relax”.
Change “because the specimens with well-preserved wrist are rare” to “because specimens with well-preserved wrists are rare”
The similar placement of your three characters at the node Caenagnathoidea in your Figure 4 suggests their synapomorphy at that node, however as you note in your manuscript the convexity of the semilunate carpal is not a synapomorphy of Caenagnathoidea in your analysis. I would remove “semilunate carpal strongly convex” from the figure in that location.
The appropriate matrix has been provided, however an incomplete dataset for character 251 has been provided in Table 1. The authors should include other scored taxa in the table or else provide some explanation for how the taxa not included in Table 1 were scored for this character.
The methods are not described with sufficient detail to replicate.
a. The taxon Anomalipes is included in the matrix but not in Figure 4, and there is no mention of its exclusion in the text. See below where I discuss the phylogenetic analysis
b. The description of the measurements in character 251 are likely to lead to confusion given the anatomical directional terms used, see below.
It seems that the TNT search strategy employed by the authors did not recover the most parsimonious trees (see below). The authors should more thoroughly describe the search strategy, and perform a more rigorous search. TNT is very fast and there is very little downside to having the search run for a longer period of time. My more rigorous searches found shorter trees and took less than a minute to complete. The authors should set up the search so it runs for more iterations and runs for a longer period of time, and repeat the analysis using different random seeds to ensure that they have found the set of most parsimonious trees.
There are four problems with the phylogenetic analysis which combine to form replicability issues with your study:
1. Char 251: This character has been improved by adding quantitative descriptors to assess convexity, however it is not clear from reading a description of the measurements how the multiplication of these ratios corresponds to a measurement of convexity. You should add a diagram that shows the measurements, especially given that differing research groups use different directional terms when describing the anatomy of the wrist and manus. I suspect that your “anteroposterior length” of the semilunate carpal corresponds to what most researchers would term the proximodistal length, however I can not be sure without a diagram.
You say in Table 1 and its caption that your image reference (based on a “high resolution image”) for Hagryphus is from Zanno and Sampson, 2005, however the only image of the Hagryphus semilunate carpal in that paper is with the manus in ventral view (their figure 2), and the curvature of the semilunate carpal cannot be completely observed in that image nor do I think all of your measurements can be taken from that image. There is a drawing of a dorsal view of the manus carpus included in that paper (their Figure 3), however the semilunate carpal is not in perfect articulation with the metacarpals—it has been slightly displaced proximally and rotated dorsally, which is why the trochleated surface is visible with the manus in dorsal view (see included image). Because Hagryphus is the only member of Caenagnathidae coded for this character in your matrix and it is implied that state 1 may be a synapomorphy of Caenagnathoidea given more complete sampling of early oviraptorosaurs, you should provide some sort of statement about how exactly you achieved these measurements from the image of Hagryphus in Zanno and Sampson, 2005, or otherwise resolve these issues. You say that “…a strongly convex semilunate carpal (character 251: 1) is only found in caenagnathoids among oviraptorosaurs”, however if you can not validate that state 1 is present in Hagryphus, then there is no evidence in your manuscript that state 1 is present in Caenagnathidae and any reference to Caenagnathoidea in this context should be modified to reflect your findings of state 1 in Oviraptoridae.
It is not clear why the measurements of other scored taxa are not included in the table (e.g. Archaeopteryx, Herrerasaurus, Oviraptor, Machairasaurus, etc.). These taxa should not be scored if you did not take the measurements, and if you did take the measurements you should include the data in the table.
2. The taxon Anomalipes zhaoi is included in the data matrix, but is not included in Figure 4. The authors do not describe excluding this taxon from their analysis in the methods section, but I see in the rebuttal letter that the taxon is not in the supplied TNT screenshots either. If you choose to exclude this taxon a priori you should provide some justification, however from my perspective there is no good reason to exclude it a priori unless it is being eliminated because of “safe taxonomic reduction” (Wilkinson and Benton, 1995), because it can be pruned a posteriori in a reduced strict consensus tree (Wilkinson, 1994).
3. The TNT file text provided by the authors marks several characters as ordered despite them being binary characters. These characters are preceded by the text “+[/1” in the “Ccode” block. This is the case for TNT character numbers: 80, 91, 98, 173, and 221—these correspond to your character numbers 81, 92, 99, 174, and 222. Though this did not affect the analysis you should make sure that they are not marked as ordered so as to avoid confusion by anyone who uses your dataset in the future.
4. The authors indicate on line 203 that the tree in Figure 4 is a strict consensus cladogram, however I could not replicate this exact strict consensus using either traditional or new technology searches, and using more thorough searches than the authors--each of my search strategies being carried out with and without Anomalipes zhaoi. It appears based on the screenshots of analyses including in the rebuttal letter that the authors found a minimum tree length of 694 steps, however the recovered tree length is not reported by the authors in the manuscript. My searches found a minimum tree length of 686 steps in analyses including Anomalipes and 684 steps in analyses excluding Anomalipes a priori. I have included the details of my analyses with screenshots below. Notably, much of the resolution of Citipatiinae reported by the authors is collapsed into a polytomy near the base of Oviraptoridae in my analyses, and this is relevant to results reported in the manuscript since the authors describe Citipatiinae as the sister taxon of Heyuanninae (lines 207-210).
See attached document for an image of Hagryphus and my own re-analyses of your dataset.
While the manuscript is improved, the reviewers agree that substantial refinements are needed. Please proofread further. Many spelling and grammar errors remain, and some new ones have been added. More importantly, key points as reviewers note must be put into the main text, not solely in the rebuttal to reviewers. That is the point of peer review, not to have an argument but to improve the paper itself constructively. And very importantly, the evidence (phylogenetic matrix and related data) used for the study must be the correct evidence shared with readers. Right now the study appears to be un-reproducible with the files shared and described. It cannot be accepted until this is 100% fixed. The output of the files' analysis must exactly match the results and conclusions reported in the paper (e.g. reviewer's point about purported caenagnathid synapomorphy).
To avoid yet more delay to peer review of your paper, please pay close attention to all of these points and be sure to include them with the resubmitted paper. As at least one reviewer still is not satisfied with important issues in the manuscript, the resubmission may need re-review especially if it is not absolutely clear if and how improvements were made.
**PeerJ Staff Note:** Please ensure that all review, editorial, and staff comments are addressed in a response letter and that any edits or clarifications mentioned in the letter are also inserted into the revised manuscript where appropriate.
**Language Note:** The Academic Editor has identified that the English language must be improved. PeerJ can provide language editing services - please contact us at copyediting@peerj.com for pricing (be sure to provide your manuscript number and title). Alternatively, you should make your own arrangements to improve the language quality and provide details in your response letter. – PeerJ Staff
- The grammar and style still needssubstantial improvement to bring the language up to the standard of “Clear, unambiguous, and professional”.
- Intro, background, and literature are adequate.
- Structure is adequate.
- Figures are high quality.
- Raw data is supplied (although not as a .nex file anymore…)
- Research is original and within scope.
- Research question is well defined, relevant to specialists, and meaningful within the field. Research fills a gap.
- Investigation is basic but appears adequate.
- Methods are now sufficiently described.
- Underlying data have been provided and they are sound
- Conclusions are well stated but in some cases not limited to supporting results. Some speculation that should be removed.
General comments:
- The authors have made most of the changes I suggested, and the expanded description section is now much better (although there are many typos that need correcting).
- I still think more comparison with Hagryphus giganteus is warranted, especially with reference to the shapes of the carpal bones themselves. You do not need to incorporate it into the wrist folding analysis, but it should be compared and contrasted because it is by far the best example of a wrist in caenagnathids. Please add your statements in your rebuttal letter about Hagryphus to the description section, as well as statements comparing the shapes and features of the radiale and semilunate carpals of Heyuannia huangi and Hagryphus giganteus.
- Throughout, the idea that larger radiale angles, flattening of the distal ulna, and a strongly convex semilunate are synapomorphies of Caenagnathoidea is presented. However, the authors do not state whether their analysis actually supports this idea (or at least the wording is ambiguous). If your analysis does indeed show that these are synapomorphies, then I would like a statement like lines 189–191 where you explicitly say: “Our analysis recovers characters XXX, YYY, and ZZZ as synapomorphies of Caenagnathoidea”.
I have not gone through to make every change necessary to style and grammar (this is much easier to do with an editable version of the manuscriot), but I have pointed out some of the major issues below line by line.
Line 24 (abstract) and 33, and elsewhere: “oviraptorosaurid” is not the correct term. It should either be oviraptorosaurian or oviraptorid.
Lines 83–85: This sentence is oddly structured and should be reorganized to improve clarity. It also seems that the sentence from lines 85–86 should be part of this same sentence.
Line 92: I assume you mean ten thousand replicates? Ten replicates is no enough to adequately sample the tree space.
Line 94: add “was” between “manus” and “estimated”.
Lines 99–100: “no significant differences in the articular faces”… if that’s the case, then wouldn’t there be no point estimating range of motion, because it would all be the same? I think this is poorly phrased and should instead say something like “This analysis method has been successfully used in other non-avian pennaraptorans with similar morphology.” And give a few references of cases.
Line 103: “metacarpals not contacted” This phrase is oddly structured and should be reorganized to improve clarity.
Lines 109–111: “The center of a circle […]” This sentence is oddly structured and should be reorganized to improve clarity. Is this what Fig. 4 shows? If so, a reference to this figure would be useful to show this approach.
Lines 125–133: it would be useful to point out any autapomorphies of Heyuannia huangi, if they exist. Also, on line 125, “could” should be “that can be”. On line 132, “could” should be “can”.
Line 128: “Mandibule” should be “mandible”
Lines 135–170: While I appreciate the expanded description, and it is much better than before, this section is replete with grammatical errors and typos. As I do not have an editable copy of the manuscript, I cannot fix all of these errors.
Line 135: It would be useful to label this ventral expansion in Fig. 2.
Line 172: “Serval” should be “several”
Lines 192–193 (and again 250–252): The radiale angle is mentioned here, but I don’t see it plotted on the tree (or reported anywhere for multiple taxa?). If you have measured this for several species, it would be great to plot those values on the tree, and otherwise, a citation or some justification is needed to explain why a radiale angle of ≥ 40 is considered primitive for Caenagnathoidea.
Lines 210–212: It’s unclear why the wrist of oviraptorids is specialized? Is there some reference or some data in the current study that supports this? Also, it seems that the following sentence (“While there is…”) should be part of the first. Overall, these two sentences should be revised and made more clear.
Lines 245–246: This sentence is oddly structured and should be reorganized to improve clarity.
Lines 256–257: as pointed out in my last review, the very notable exception of Hagryphus deserves some comparison here. Even if you can’t include it in your wrist folding analysis, there is lots of information about overall wrist flexibility available from the carpals themselves.
Lines 272–275: I don’t think this argument is well supported—there are many other factors involved in limb length and morphology, and there is no test here of the importance of different factors for caudipterygid limb length. This idea is essentially just speculation and I think it should be removed from the manuscript.
Lines 283–285 and again 300–301: I think a bit more subtlety is needed here. I do not think these features of wrist flexion are necessarily strong evidence for the presence of pennaceous feathers (although I also think the general consensus is that oviraptorids did have pennaceous wings). However, it could be argued that they are consistent with the presence of long feathers. I think overall the link between wrist folding and protection of the feathers is overstated. Flexibility of the wrist accomplishes many tasks and enables many features, and there is little evidence to suggest that it is driven by increasing feather length. In fact, the progressive increase in radiale angle from Sullivan et al. 2010 argues against this: we know pennaceous wings evolved by Ornithomimosauria, but consistently increased radiale angles only appear later on in Pennaraptora.
Line 292: “Should be” should be “are”.
Basic Reporting
My comments on the English in the manuscript are similar to my previous review. Clear and unambiguous professional English is not used throughout the manuscript. The English should be improved to ensure that international readers of your manuscript can clearly understand all elements of the text. A non-exhaustive list of examples where the language could be improved including lines is listed under my general comments. I suggest again that you ask a colleague who is proficient in English and who has some expertise in dinosaur paleontology to help copy-edit your manuscript.
Thank you for addressing my comments with your added text about the specimens on display, but I do not understand what you mean by “only the left view could be observed (line 77)”, and based on your text I still do not understand why you can’t see the other two wrists that are visible in Figure 1. Do you mean only the “left” side of the block where left is indicated by the directionality in Figure 1? Do you mean anatomical left even though you’re talking about a right forelimb? If the three wrists that are visible in Figure 1 are no longer visible while on display, you could indicate in the Figure 1 caption that the photograph was taken before the current study and that only the left side of the main block is currently visible.
The authors may have intended to include the raw data for the phylogenetic analysis in accordance with journal policy, however an incorrect dataset was provided again for this submission, just as it was for the first submission (see Experimental Design section).
Please see my comments on figure titles/captions under general comments.
The authors have changed the phylogenetic matrix used in the study to that of Wei et al., 2022. The Wei et al. matrix contains 246 characters and the supplied matrix in this submission contains 255 characters, just as the supplied matrix contained 255 characters in the last submission. Therefore I am unable to evaluate the results of the analysis or findings based on the analyses such as synapomorphic character transformations of the wrist.
Thank you for implementing my suggestions for the character states in your added characters.
Based on your drawings of the angle subtended by the the semilunate carpal, it is not clear to me exactly how the measurement is being taken for this character. How do you define the point from which you draw the angle? Your character 248 refers to the curvature of the proximal margin, but in your figure 4 the angle is about the distal margin of the illustrated semilunate carpals. This is made especially clear by the illustration of Hagryphus in which the proximal trochleated surface is visible on the opposite side of the drawn angle. This issue must be resolved because the lack of clarity for this character may hinder replicability of the character scores by future authors.
I will repeat my comments in the prior review about the supplied data matrix. Please provide a TNT file with the correct data matrix in plain text format for easy re-analysis by reviewers (and so that they do not need to spend time formatting the matrix for re-analysis). If you have analyzed this file you should possess it and there is no reason to withhold it from reviewers.
If possible, it is preferable that you include your matrix in an executable TNT command line format, as the executable file used in your analysis can be perfectly replicated by the program by any reviewer or reader. Then there can be no question about whether the matrix is correct, or whether the analysis is replicable.
I am concerned about your statement on line 98: “This analysis method is adopted because there is no significant difference in the articular faces on the forearms and carpals among all non-avian pennaraptoran.” This is a very broad and strong statement, and it needs a citation or some kind of validation that there are no significant differences in wrist articular surface morphology in non-avian pennaraptorans. In fact, the articular surface morphology of forearm and carpal elements is obscured by matrix in a not insignificant number of non-avian pennaraptoran specimens.
The premise of the range of motion analysis in this paper is that differences in the sizes and shapes of wrist elements confer different ranges of motion within Oviraptorosauria. You have even added characters that delineate differences in the shape of the distal ulna (character 246) and semilunate carpal (character 248), and described differences in the ulna in caudipterids vs other oviraptorosaurs, for example. If I understand correctly what you mean, then your statement that there are no significant differences in the articular faces on the forearms and carpals appears to be negated by your own work in this study.
The authors have made no modifications to the manuscript to discuss the possible influences of soft tissues on range of motion estimates. Even a caveat in the discussion section would be appropriate, just to acknowledge that the ranges of motion presented in this paper are estimated based on bone only, and that soft tissue influence could potentially alter the drawn conclusions about the wing-folding mechanism. For example, similar statements acknowledging the potential influence of soft tissue on range of motion estimates were made in the methods and discussion sections of Senter, 2006.
I will also reiterate from my previous review that cartilages and ligaments were certainly incorporated in the wrist, including cartilages that capped the distal ends of the ulna and radius. It is therefore likely that the center of rotation of the wrist in adduction/abduction motions was set some distance away from the distal ends of the ulna and radius, and the comparative significance of that variation in distance is unknown. Because the variation in these soft tissues is completely unknown, it is therefore unknown what influence they may have on the replicability of range of motion estimates based on bones from study to study and taxon to taxon.
I can not attempt to verify the findings of the phylogenetic analysis because an incorrect matrix was provided for the second submission in a row by the authors.
The discrepancy between the description of character 246 and the newly added illustrations of the semilunate carpal angle in figure 4 that I noted above make me question the validity of the analysis, and notably a portion of the data for the analysis that was added by the current authors.
The conclusion of the authors that a wing-folding mechanism of some oviraptorids is convergent on that of birds is very plausible, and perhaps even likely. However, I still find it difficult to evaluate the findings regarding range of motion in this study, even given the added explanations for the methodology.
I would suggest that the authors do more to convince readers of the replicability of their methodology. However, because the reported ranges of motion are estimates, I do not think my personal questioning of this methodology should prevent publication. What is more troubling is that because the reported ranges of motion are estimates and are acknowledged as such by the authors, the replicability from taxon to taxon of phylogenetic character scores based on these range of motion estimates is questionable.
General Comments
Line 109: “The center of a circle of the curvature of the proximal
margin of the semilunate carpal is defined as the point is equidistant from the highest
point and two terminals. The angle between the radii (from the center of a circle to the
terminal) is a measure of the curvature of the proximal margin of the semilunate carpal.”
These sentences are very confusing. If you could add an illustration for this or refer to a figure in another paper it would be helpful.
Results
Line 125: “A oviraptorid dinosaur could be distinguished from other oviraptorids
by the unique combination of characters”
Replace with “An oviraptorid dinosaur that can be distinguished from other oviraptorids by a unique combination of characters.
Line 162: “As in other pennaraptorans, a deep transverse groove that runs
along the entire arc of the proximal semilunate carpal is presented (Xu et al., 2014).”
On several occasions you say “presented” when you mean “present”. Replace with “present”.
Line 164: “The distal surface of semilunate carpal covers the proximal end of
metacarpal I and metacarpal II, as in other caenagnathoids”
The distal surface of the semilunate carpal…
Line 168: “In contrast to the description from Lü (2005), semilunate carpal is not fused with metacarpal I and metacarpal II, an obvious suture between the semilunate carpal and the first two metacarpals is presented.”
In contrast to the description from Lü (2005), the semilunate carpal is not fused with metacarpal I and metacarpal II. An obvious suture between the semilunate carpal and the first two metacarpals is present.”
Line 172: “The phylogenetic position of Heyuannia huangi has been analyzed in serval papers”
Change serval to several
Line 181: “Character 248. semilunate carpal: (0) not strongly convex, curvature of the proximal
margin of the semilunate carpal smaller than 140 degree; (1) strongly convex curvature
of the proximal margin of the semilunate carpal 140 degree or larger.”
Change “degree” to “degrees” in both cases.
Line 192: “The strict consensus tree shows that a strongly dorsoventrally compressed distal
ulna, radiale angle 40 degree or larger and strongly convex semilunate carpal should be
the body plan of Caenagnathoidea.”
You say elsewhere that these features are synapomorphies of Caenagnathoidea. If this is true then you should not say “body plan” here, because its meaning is ambiguous. It is ok to repeat that these are synapomorphies.
Line 199: “…is derived from convergent evolution rather than homologous evolution.”
Replace with “…is convergently evolved in these taxa.” The phrase “homologous evolution” is very unusual, and the fact that they were convergently evolved implies that the transformations were not homologous.
Discussion
Line 220: “The proximal surface of radiale possesses a concave at its center in order to contact the radius.”
Change “possesses a concave” to “possesses a concavity”
Line 250: “According to the phylogenetic result, a strongly dorsoventrally compressed distal ulna, larger radiale angle and strongly convex semilunate should be the synapomorphies of Caenagnathoidea.”
If these are synapomorphies, don’t say that they should be synapomorphies; instead say that they are. I would write:
According to the results of the phylogenetic analysis, a strongly dorsoventrally compressed distal ulna, larger radiale angle, and strongly convex semilunate carpal are synapomorphies of Caenagnathoidea.
Figures
Figure 1 The sentence “The yellow box indicates the forelimb in Figure 2” is repeated.
Figure 2 Give the side of the forelimb and indicate the view in the Figure title
Figure 3 The title says it is a photograph of the forelimb, but this is only the ulna
Two reviewers have provided detailed, constructive critiques. They are supportive of the study but there are some common points. In particular, the estimates of range of motion need to be documented better, and perhaps re-assessed. There is a rich literature on these methods, which can range from simple 2D estimates to very sophisticated 3D digital estimates. Searching the literature for terms like archosaur, dinosaur and range of motion or joint mobility would help inspire and justify how the revised study considers this topic. Some scorings of characters need better explanation, and better descriptions of the osteology are necessary. Further proofreading is needed to improve clarity of the text. Thank you.
**PeerJ Staff Note:** Please ensure that all review, editorial, and staff comments are addressed in a response letter and that any edits or clarifications mentioned in the letter are also inserted into the revised manuscript where appropriate.
**Language Note:** The Academic Editor has identified that the English language must be improved. PeerJ can provide language editing services - please contact us at copyediting@peerj.com for pricing (be sure to provide your manuscript number and title). Alternatively, you should make your own arrangements to improve the language quality and provide details in your response letter. – PeerJ Staff
I have reviewed your manuscript about the wrist of Heyuannia huangi, and find that the work is strong overall, although I nonetheless have a few suggestions to improve the study. I have outlined my general comments below, but throughout the manuscript I have noticed some grammatical or stylistic errors that should be corrected before final publication. As I recommend some substantial changes, I have not flagged all of these errors, but I am happy to do so on a later version.
- My main suggestion for the manuscript is to flesh out the description of the wrist bones, to comprehensively detail the osteology. Zanno and Sampson (2005) would provide a good guide for this, and indeed I was surprised to see that there is not more comparison made between the well-preserved wrist of Hagryphus and Heyuannia. At the moment, the description (only 400 words including references) is fairly superficial, focusing mainly on comparative metrics (sizes, angles), rather than providing details of the shapes and features of the wrist bones themselves. Some of the information currently in the discussion (e.g., lines 161–173) would fit better in this section and help to flesh it out, too.
- A third carpal is mentioned in the description, but no information about it is given. What is its identity/position in the wrist? What is its appearance? The homology of oviraptorid carpals is currently unresolved, because several taxa are described as having only two carpals, although others appear to have three, and the basal condition is four (at least in caudipterids and caenagnathids). These Heyuannia specimens could provide important information on the issue, but this is not given.
- It is also mentioned that a new isolated ulna is described, but no information seems to appear in the description section, and I cannot find reference to the specimen besides the figure. No details are given about why this specimen is referred to Heyuannia huangi rather than another oviraptorid, but this is crucial because multiple oviraptorids are known from the area.
- The phylogeny should be updated to include info from the Funston et al. (2020) paper describing Oksoko—this taxon has an excellently-preserved wrist and so it is important to have it in the phylogeny if you are introducing new carpal characters. The Wei et al. (2022) phylogeny (from their paper describing a new specimen of Yulong) is probably the most appropriate phylogeny to use as a base.
- It would be very useful for future work to provide further description and illustration of the new character states being introduced. For example, it would be very difficult for someone using your new character 256 to discern what is meant by “strongly convex”—all oviraptorosaur semilunates are convex to some degree, where is the boundary between “strong” and “not strong”? A figure showing an example each of the states in these characters would be very useful.
- Lines 126–144: This information should be in the Materials & Methods section.
- Line 197–198: While this is true, there is one very relevant exception: Hagryphus giganteus, which has an exquisitely-preserved carpus that has been described in detail. The information from that study is sufficient to make some comparisons at least, even if the ulna and radius are not known. For example, is the arrangement of the carpals more similar in Hagryphus to caudipterids or to oviraptorids?
- Figure 4: It looks like the illustrations of Khaan and Heyuannia are identical—was this intended? I do not see any mention of Khaan’s wrist mobility in the main text, and it is unclear why this taxon is shown instead of, say, Citipati, Machairasaurus, Nemegtomaia, or Oviraptor, each of which have articulated hands preserved.
- Figures 4, 5: It would be useful to annotate the illustrations with quantitative measures of the angles of flexion and extension.
- I would like to see more details about how the range of motion of Heyuannia huangi was actually determined. Was it through physical manipulation of the bones? Or simply by rearranging two-dimensional images? Was it based solely on the range of articulation of the holotype specimen? If so, why are these values considered maxima and minima of flexion? No methodology is given for determining the values and positions. Some of the findings are questionable based on the articulation of the semilunate and radiale that I have observed in other oviraptorosaurs like Oksoko, where the ball-and-saddle nature of the radiale-semilunate contact confer extensive flexibility in the wrist. For example, in one of the articulated holotype specimens, the wrist is preserved in undisturbed articulation at an angle of just 24° (see Funston et al. 2020 Fig. 3c)! To me, this suggests that the three-dimensional features of the bones, and in particular the nature of the intracarpal joints, have a major effect on flexibility. It may be that you’ve taken all this into account, but your methodology for determining range of motion hasn’t been outlined in the materials and methods section.
- Looking at the supplementary information, it seems that the newly introduced characters 255 and 256 have not been coded for any taxon, including Heyuannia huangi? There may have been a frameshift somewhere, as the character scores provided only include 255 characters, not 256. In future submissions I would recommend submitting the actual .TNT or .NEX file as a supplementary file, because I cannot easily reproduce the phylogenetic analysis based on the data provided. However, I have done so, and I was not able to replicate the result that the authors describe. Whereas the authors state they recovered 2 most parsimonious trees (but do not report what length), I recovered 6, which have the same topology, except that the authors’ figure is missing Incisivosaurus gauthieri, which is present in their data matrix. This omission is not explained in the text or figure caption. I suspect these errors arise from different versions of the matrix being used for the text vs. the actual analysis, but they should be sorted out so that the analysis is replicable.
- Without more details about how the range of motion was determined, it is difficult to evaluate whether the conclusions about wrist evolution in oviraptorosaurs are valid. In particular, it is unclear whether changes in the detailed morphologies of the wrist bones might enhance flexibility by allowing movement between them—this was described by Sullivan et al. (2010) as the main articulation in the wrist of extant birds, but the focus in this manuscript seems to be on the ulna’s effect on flexibility.
- Overall I have little reason to doubt that oviraptorosaurs show a progressive increase in wrist mobility through their evolution, so I think ultimately the conclusions of the manuscript are reasonable. However, the methods and data as outlined are insufficient to evaluate the details of this progressive increase, or for future workers to build on these results and add new data to the framework. These issues could be addressed by outlining the methods and data in more detail throughout the manuscript.
Though I am critical of the manuscript in my review below, please understand that my comments are meant to be constructive and I have attempted to convey the collegial tone of my comments. Your research subject is interesting and is deserving of detailed treatment in the literature, and I hope that my comments will help you improve your manuscript for re-submission.
Clear and unambiguous professional English is not used throughout the manuscript. The English should be improved to ensure that international readers of your manuscript can clearly understand all elements of the text. A non-exhaustive list of examples where the language could be improved include lines is listed under my general comments. I would suggest that you ask a colleague who is proficient in English and who has some expertise in dinosaur paleontology to review your manuscript before resubmission.
The authors appropriately discuss and cite references covering the background and context of the subject, however, the authors may wish to frame their study by raising conclusions made by recent treatments of forelimb morphology in Heyuannine oviraptorids such as Funston et al., 2020; and of wrist morphology in theropod evolution more generally such as Botelho et al., 2014 and Xu et al., 2014.
The structure and organization of the article is professional and the raw data are shared. Figures are of appropriate resolution, though a higher resolution photograph of the holotype block shown in Figure 1 would be a valuable addition to the manuscript. It would be helpful if you label in Figure 1 the forelimb shown in Figure 2.
If possible, the authors should provide additional images of the skeletal morphology of the region to better support their description. The authors provide figure 2 showing the right forelimb and manus of HYMV1-2, but otherwise do not provide any alternate angles of the morphology that is the topic of the paper. Based on Figure 13.3 in Lu et al., 2005, it is possible for a firsthand observer to view the same wrist in Figure 2 in three dimensions and from the opposite side. Furthermore, there are three oviraptorosaur wrists preserved on the holotype block, so it is not clear why these other wrists are not figured or mentioned in a manuscript dedicated in part to the wrist morphology of this taxon. If it is no longer possible to view the figured specimens in multiple dimensions because of further preparation, display conditions of the specimens, or some other reason, then the authors need to describe this and acknowledge it as a limitation of the study.
The authors intended to include the appropriate raw data for the phylogenetic analysis in accordance with journal policy, however an incorrect dataset was provided (see Experimental Design section).
The research question is well-defined, relevant, and meaningful. The authors summarize their aims: “we provide a detailed description of the osteology of the wrist of Heyuannia huangi and discuss the function of the wrist based on the careful comparison between the morphology of the wrist of Heyuannia huangi and other pennaraptorans with completed wrist preserved”. These are valid aims that are deserving of study, however I do not find that the authors have described the morphology in sufficient detail or that they have made enough appropriate morphological comparisons to relevant taxa.
The methods employed in the phylogenetic analysis are described with sufficient detail and information. However, the authors did not provide a data matrix that matches the one described in the text. The authors say that they have added three characters to Qiu et al., 2019 for a total of 256, however the supplied data matrix only contains 255 character states per taxon. I analyzed the provided matrix with 255 characters under the protocol described by the authors, and it resulted in 6 most parsimonious trees and a strict consensus similar to the one presented in Figure 4. Though I cannot determine whether the results are perfectly replicable based on the available information, I suspect that they will be replicable upon provision of the correct matrix in a future submission. Please provide a TNT file with the correct data matrix in plain text format when you resubmit for easy re-analysis by reviewers (and so that they do not need to spend time formatting the matrix for re-analysis), and preferably include this file as part of the supplementary information. If possible, it is preferable that you include your matrix in an executable TNT command line format, as the executable file used in your analysis can be perfectly replicated by the program by any reviewer or reader. Not all authors use the command line version of TNT though, so it is understandable if you are unable to do this.
I will add here that the provided data matrix does not contain any character scores that are scored for polymorphism or ambiguity. This does not conform to the expectations of scoring this many fossil taxa and thus does not represent a matrix that is on par with the current highest standards of the field. Though I am reasonably confident based on my own modification of the tree search strategy that the authors have found the most parsimonious trees, the authors’ chosen heuristic search methods of 1000 traditional search replications is not a very rigorous search. A more rigorous search strategy using “New Technology” in TNT with parsimony ratchet iterations, etc. would add only negligible time to the analysis and would provide a stronger test of the authors’ phylogenetic hypothesis.
The authors can improve their added morphological characters such that an independent observer could reach the same character scores. The following suggestions have the potential to significantly improve the new characters and make them more valuable to future researchers who wish to incorporate the phylogenetic information into their own studies:
“254. Distal end of ulna: (0) enlarged; (1) strongly compressed.”
The state “enlarged” suggests a comparison to an unenlarged distal ulna, while even the compressed distal ulnae of some oviraptorids appears “large” when viewed from a dorsal or ventral perspective. It would make your character more clear if state 0 is “not strongly compressed dorsoventrally” and state 1 is “strongly dorsoventrally compressed”.
“255. Radiale: (0) smaller than 40 degrees; (1) larger than 40 degrees”
This should read “255. Radiale angle: (0)…”. I would also make one of the states include exactly 40 degrees so there is no ambiguity as to how to score such a taxon in the future. You can make state 1 “40 degrees or more”.
“256. Semilunate” carpal: (0) not strongly convex; (1) strongly convex”
Is there a way you can add some quantitative description to each of these states or else provide photographs or illustrations of each state? If you do not, then each researcher who uses the character may come to their own subjective conclusions about what constitutes relative strength of convexity.
There is no description of the methods used to determine adduction and abduction angles, range of motion, anatomical position at rest, etc. as discussed in the text and drawn in Figures 4 and 5. Because the forelimb and manus as figured perfectly match the orientation of the photograph of the holotype block, I assume that the described forelimb is still encased in matrix. Therefore, the authors cannot have engaged in manual manipulation of the elements to test ranges of motion, and the authors do not describe any CT scanning that would enable them to generate physical or digital casts to accomplish this. The authors therefore seem to be relying on manipulation of drawings to determine ranges of motion as best I can tell from the manuscript. Regardless of the methods employed, it is imperative that the authors describe their methodology in detail, and defend their methods of determining range of motion against anticipated counter-arguments. For example, I would anticipate arguments that range of motion is difficult to determine in the absence of soft tissue. It is clear from the specimen’s preservation in articulation and with another exquisitely preserved articulated specimen that the distal carpal for the third digit is not missing and was most likely cartilaginous at death. The same morphology of a distally displaced third metacarpal is found in other articulated oviraptorosaur hands including the heyuannine Machairasaurus, among others. Thus, some might argue that there is ontogenetic variability in the structure of the carpus and that the material is from an ontogenetically immature individual that could be interpreted as having incompletely ossified superficial margins of the carpals that would bias range of motion estimates.
The findings of the phylogenetic analysis are valid based on my re-analysis of the provided dataset. However, again it is important that the authors provide the correct dataset upon re-submission, and I again note that an executable command line file will be perfectly replicated by any reviewer.
I find it difficult to evaluate the findings regarding range of motion, principally because the authors did not describe their methods for determining the limits of motion and because it is difficult to assess the extent of possible error in a range of motion study such as this one.
I can not conclude from the data and arguments in the manuscript that the following conclusion is supported by sufficient evidence: “The morphology of its wrist indicates though the center of rotation of the wrist is located at the joint of the ulna and radius”. I consider it likely that some cartilaginous carpals were present in life including the intermedium and the distal carpal of the third digit, and possibly the ulnare. Furthermore, other cartilages and ligaments certainly would have supported the wrist, including cartilages that capped the distal ends of the ulna and radius. It is therefore likely that the center of rotation of the wrist in adduction/abduction motions was set some distance away from the distal ends of the ulna and radius.
I cannot conclude based on the information presented in the paper that the referral of the ulna HYMV 2-8 to Heyuannia huangi is grounded in sufficient evidence. Notably, there is no morphological information regarding the ulna in your revised diagnosis of this taxon, so it is not clear from this manuscript what morphologies of the ulna are being used to refer HYMV 2-8 to Heyuannia huangi. I suspect that the referral is based in part on taphonomic evidence since Lu, 2005 indicates that the ulna was collected from the same quarry. If the referral is based on similarity of HYMV 2-8 to other Heyuannia huangi specimens and on taphonomy, then the authors should state that though there are no autapomorphies of the ulna, the ulna matches the morphology of HYMV 1-2 and was collected from the same quarry, or make some similar argument depending on what information you have. This is important because there are several other oviraptorid taxa known from contemporaneous deposits of southern china, and given that you have not identified autapomorphies of the ulna or differential diagnostic characters of the ulna in this taxon, HYMV 2-8 must bear similarity to the ulnae of these other taxa.
The authors describe their motivation of adding new characters to their phylogenetic analysis as an intent to “discuss the evolution of the wrist during the evolution of oviraptorosaurs”. However, there is little information in the results or discussion sections regarding whether any of these added characters result in homologous transformations of the wrist that can be optimized as synapomorphies and/or where on the cladogram these synapomorphies occur. These topics need to be addressed in these sections, and the conclusions of the paper should address whether the principal experiment of the paper (the phylogenetic analysis) revealed any new information about the sequence of character transformations that produced the observed evolutionary distribution of wrist morphologies in Oviraptorosauria, and especially in heyuannine Oviraptoridae.
This manuscript could provide significant enduring value to the literature by adding more detail to the anatomical description, description of other wrists on the same block, and detailed photographs of the relevant morphology from the different wrists in multiple angles. I would suggest that the authors focus on improving this aspect of the paper. Below are some more general (non-exhaustive) thoughts on specific areas of the paper.
Figure 4.
a. “The major changes of oviraptorosaur’s wrist”
Change to something like “Major changes of the wrist during oviraptorosaur evolution”
b. “The line drawings are not to scale”
This is written twice in the caption.
c. “All the left wrists are in the minimum abduction and all the right wrists are in the maximum abduction”
Minimum abduction is probably better written as maximal adduction. This caption is also confusing as written because all of the figured illustrations are from the anatomical left.
d. The illustrations of Khaan and Heyuannia as drawn look very unnatural so as to be dislocated in their maximally adducted state.
e. Nemegtomaia barsboldi is spelled incorrectly in the figure
Figure 5.
Again, minimum abduction is probably better written “maximum adduction”. I don’t understand why the maximally adducted wrists of Deinonychus and Meleagris show a medial metacarpal in line with the radius, but the same position in Heyuannia as interpreted results in a medial metacarpal aligned with the ulna. There is no legend or indication in the figure as to what the different colors represent—this is clear to me but it may not be to many readers
INTRODUCTION
Lines 33-34 “found from the Upper Cretaceous in the Heyuan Basin”
You should probably include some information about the Dongyuan Formation here.
Lines 35-36 “where thousands of dinosaur eggs were found”
This information is not very relevant to the article, and sticks out as such when left alone. If the authors wish to expand on this information and provide more relevant context about the fossil record of the Dongyuan Formation that would constitute part of a proper introduction.
Line 35 “The deposits contained this dinosaur”
Revise English
Line 38 “unique clade”
I would not use ‘unique’ to describe a clade, as by definition the content of all clades are unique with respect to all other clades.
Lines 45-46
Revise English
Lines 53-54
Revise English
Lines 58-59 “…in the derived oviraptorosaurs”
The use of basal vs derived is a somewhat outdated construct, as early-diverging oviraptorosaurian clades such as Avimimidae also survived into the latest Cretaceous and were therefore equally “derived” in terms of branch length. I would be more specific and say it is a good example of a wrist in a Heyuannine oviraptorid.
MATERIALS AND METHODS
There is no description of the methods used to determine adduction and abduction angles, range of motion, anatomical position at rest, etc. as drawn in Figures 4 and 5.
There is also no description of the phylogenetic analysis under materials and methods, which would be appropriate in this section and is less appropriately placed in the results section.
It would be helpful to include in this section your methodology for measurement of the “radiale angle”. Or, at least redescribe the information given in Sullivan et al., 2010.
Line 68-74 “…in Heyuan Museum, a research and educational non-profit museum formed in 1982…”
This information about the status of the museum and the timing of discovery of the holotype of Heyuannia is not relevant to the materials and methods. This information could go in the introduction.
Line 76-77 “…described briefly by Lü (2002).” You have already given this information in the introduction.
Line 77 “preserve elements”
Revise English
RESULTS
Systematic Paleontology
A section should be included here that describes the hypodigm: the holotype and any specimens that you are referring to the taxon. Most importantly you need to include the ulna (HYMV 2-8) that you describe. Notably, there is no morphological information regarding the ulna in your revised diagnosis of this taxon, so it is not clear what morphologies of the ulna are being used to refer HYMV 2-8 to Heyuannia huangi.
Line 93-94
“more groove-like” as compared to what? I would just say “groove on the quadrate demarcating the articulation with the quadrate”, or something similar.
Line 95 of revised diagnosis
Change “mandibular” to “mandible”
DESCRIPTION OF THE WRIST
Line 110
You have identified the proximal carpal as the radiale here, but it is known that the radiale is fused to the intermedium in many coelurosaurian theropods including birds. It may be true that this is only a radiale and that the intermedium is not present and fused to this element, but you should provide some information about your hypothesis of homology. You should also provide an argument rejecting the hypothesis that the preserved proximal carpal is an intermedium without a radiale. You could state that you are using “radiale” for simplicity of comparison with other oviraptorosaur research, but as this paper is dedicated to oviraptorosaur wrists you should be precise here. Your paper will be one of the first results for anyone who is looking for information on the wrists of oviraptorosaurs. Botelho et al., 2014 termed the radiale+intermedium the scapholunare, though this terminology has not been used broadly by theropod researchers.
Line 115
You have “semilunate” in quotes here and elsewhere in the paper, but it is not in quotes on line 113. The semilunate carpal of birds is known (e.g. Botelho et al., 2014) to be derived from the distal carpals of the two medial digits (I+II of some authors and II+III of others) and is interpreted to be derived from the same two distal carpals in coelurosaurian dinosaurs by nearly all authors that I am aware of. I would refrain from using it in quotes, but if you want to indicate some doubt as to its homology then you need to discuss this as well.
The distal carpal associated with the third metacarpal is not discussed. This carpal should be presumed present in oviraptorosauria, as the carpometacarpi of
OTHER COMMENTS
The authors recover Heyuannia huangi within the clade Heyuanniinae following the revised definition of Yun, 2019 rather than the original definition of Osmolska, 2004 of Ingeniinae (=Heyuanniinae), but they neither discuss Yun’s revised definition of this clade nor do they label the clade in Figure 4. Yun, 2019 defines the clade as the least inclusive clade containing Heyuannia huangi and Conchoraptor gracilis, but recent research by Funston et al., 2020 employs a different definition in their Figure 6. The authors need to address this conflict.
Taxa within Heyuanniinae should be of elevated importance for anatomical comparison and discussion in this manuscript. Though Conchoraptor, Jiangxisaurus, Heyuannia yanshini, Nemegtomaia, and Machairasaurus are all recovered as members of Heyuanniinae and at least a partial if not complete forelimb and manus is known from each of them, some of these taxa are never mentioned in the manuscript (aside from their labels in Figure 4) and there is only sparing discussion of their wrist morphology, with some discussion for a single character of the radius in Machairasaurus and Nemegtomaia. While the forelimb and manus of the oviraptorid taxon Oksoko avarasan is also completely known based on several specimens preserving this region, this taxon is only mentioned in the context of carpal anatomy and is not included in the phylogenetic analysis. The authors should discuss synapomorphies of the forelimb and manus within Heyuanniinae and contrast these morphologies with non-heyuannine Oviraptoridae, and provide more extensive discussion regarding the contrasting morphologies of these taxa and other oviraptorosaurs. The authors should consult the publicly available dissertation of Gregory Funston (https://era.library.ualberta.ca/items/32edf6a6-9791-478f-9b95-7deef3c6fd09) for several unpublished images of oviraptorosaur forelimbs that will be valuable to the authors.
The authors state that the semilunate carpal is not fused with the metacarpals, which is in direct conflict with the interpretation of Lu et al., 2005, however there is no mention of this differing interpretation. The authors should address this conflict and provide additional details that support their interpretation.
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.