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.
Dear authors,
Thank you very much for your disposition to accept all the suggested changes I made in my last letter.
It was an excellent last contribution to have added LSID information for the new clade names and their synapomorphies for to be available in both the ICZN Code and the International Code of Phylogenetic Nomenclature (PhyloCode). Thanks for that, it will be very welcome.
As I have already commented, I am sure that this paper will be a very useful reference for studies on the particular taxa that encompass the Teleosauroidea, but also for researches on related archosaur groups. Indeed, I would hope to find in the future more similar so detailed and clear studies for even unrelated groups.
I have just a little recommendation to do: could you please fix the ornamentation character also in the supplementary description file? You can do it during the process of revising the proofs.
Therefore, I am glad to announce that in my concept, this paper is ready for to be published in PeerJ.
Congratulations!
Best wishes,
Graciela Piñeiro
[# PeerJ Staff Note - this decision was reviewed and approved by Patricia Gandini, a PeerJ Section Editor covering this Section #]
Dear authors,
Thank you very much for your extensive work improving your manuscript on Theleosauroidea systematics and ecomorphologic diversity and distribution, following the constructive recommendations made by all the reviewers. We indeed need to specially thank all of them for their time to make so detailed and useful revision of this interesting and valuable study, which without doubt, will be an ineludible referent in future research on Mesozoic crocodylomorphs. I am glad to note how much you have improved the text and the beautiful and informative figures included. I read carefully the reviewed PDF and the rebuttal letter that you provided and in my opinion, you have completed almost all the reviewer’s concerns which represented substantial modifications for improving specific sections. Therefore, this manuscript would be almost ready to be published in PeerJ. There are just a few modifications that I recommend to be considered to complete the edition at this stage of the review process; you will find them in the annotated pdf attached to this letter. I am sure that it will take just a few minutes to fix them and after that please, submit the revised version of your manuscript for final acceptance.
With my best regards,
Graciela Piñeiro
Dear authors,
We have three review reports for your study on the Teleosaurid phylogenetic relationships, ecology and evolution.
As you will see reviewers have found your manuscript very interesting and relevant to the teleosaurid phylogeny and biology. However, they think that there are many aspects that should be revised and improved. In advance, I think that you should take into account all the reviewers' recommendations and consider them as a valuable input for improving the different sections that comprise this extensive and complete study.
Regarding the last point, I very agree with reviewer 3 in that you can shorten a lot the text making it more fluent to readers, following the guidance that this colleague has kindly provided.
Two reviewers have expressed concerns about the developed methodology to build the phylogenetic analysis and reviewer 1 provided a detailed report which includes valuable recommendations to make your analysis stronger and creditable. So, please, be careful in assessing all these commentaries and recommendations.
The paleoecological inferences need the addition of complementary information and a calibrated cladogram has to be provided. Figures are fine and perhaps some of them will have to be modified. Some useful examples for figure improvement have been provided by reviewer 2 in the annotated pdf.
Hoping that you find useful all of the very constructive and detailed comments provided by the reviewers, I look forward to see your revised manuscript submitted soon.
With my best regards,
Graciela Piñeiro
Good. See below (general comments for the authors) for more.
Good. See below (general comments for the authors) for more.
Good. See below (general comments for the authors) for more.
This looks like a very thorough study, but I should stress that I am not an archosaur specialist. Thus, I have focused on two field for which I have fairly sharp expertise, namely phylogenetics and nomenclature. In general, I like the draft, but I see two problems with the phylogenetic analysis, one that can very easily be fixed, and another one that would be harder to fix, so I don’t think that the authors should be obliged to follow my advice, but I offer both for them to consider. Then, follow more minor, punctual comments.
First, the easy part. The authors have ordered some characters (which is good), but their criteria for doing so are not clear, and obviously, they have omitted to order some characters that should really be ordered. The criterion is simple (but has rarely been simply explained): any character that varies continuously (i.e. forms a cline) should be ordered because in fact, the assumptions that allow to discretize continuous variation into states is exactly the same that stipulates that such states should be ordered. For more details about this and simulations that show what is gained by ordering such characters, see Grand et al. (2013), Rineau et al. (2015, 2018) and Wiens (2001: 693-694). Such characters include all inherently continuous quantitative characters (size, shape, ratios) that have been discretized. These are very common in systematics. Character 5 “Skull width to length ratio” is such a character. The authors should revise and clarify their approach (explain why this was done) starting on lines 317-318, where they list the characters that were ordered. Obviously, more characters need ordering, like new character 294, characters 17, 49, 58, 59, 79, 85, 88, 89, 410, and probably characters 54, 55, 78 (this appears to be a gradient in lachrymal length) and 151, among others (I checke till character 100; after that, I checked just a few, so this list is probably far from exhaustive). Clarifying the rationale for ordering here is important because this is obviously not clear to most systematists (my conversations with several of them shows this clearly). Also note that such ordering also matters in Bayesian analyses, so the latter should also be redone. You did not specify if your Bayesian analyses used the same ordering scheme as your parsimony analyses, but that should be the case. Also, please note that there is a strong caveat to doing Bayesian analyses of morphological data; about this, see Goloboff et al. (2019). I am not saying that you should not do such Bayesian analyses, but this caveat should be discussed briefly.
Second, the harder (optional) part. Some characters are obviously composites of two or more features that have been lumped into a single shape character. Number 2, “Skull geometry, relative position of tooth row, quadrate articular facet and occipital condyle », is such a character. This character is problematic because it mixes the position of three structures (tooth row, quadrate condyle and occipital condyle), which are apparently aligned with respect to a fourth, unidentified structure. As a result, the position of each of these structures, which must vary continuously, cannot be ordered in a meaningful way. I would split this into at least two ordered, multi-state characters. One possibility would be to select an axis aligned on the premaxilla tip and the occipital condyle as a standard reference (though the authors should feel free to choose any pair of homologous points they see fit for this), and determine the position of the other structures (in this case, quadrate and mid-point or caudal end of the tooth row) relative to this axis. This would then allow at least three states to be defined (above, aligned with, or below the axis). Of course, “aligned with”, in all cases, is approximative, so a tolerance threshold should be set (for instance, add or take 1, 2 or 5% of the total cranial length). And such characters could and should be ordered because this represents discretization of continuous variation, a cline. Implementing this for all characters where this might seem warranted on a matrix of this size would be very time-consuming, so I think that this should be optional. But I suggest that the authors try this approach for at least one of their characters (number 2 seems like a good one) and see how it works for them. That way, in their future studies, they can do better.
I have a few more punctual comments, organized by line number of the MS-Word file (I never could download the pdf file):
26, 36 (and possibly elsewhere): « modern » is a very subjective concept ; for some paleontologists, anything after the Proterozoic is modern. For others, it is from the 1990s on… Use more objective terms, like “extant”, “crown”, or name the taxon. Otherwise, only a few specialists may (perhaps) really understand what you mean.
45, 46, 47 (and possibly elsewhere): “genus”, like all Linnaean categories, is an artificial construct. You might want to minimize use of these and use the convenient words “taxon”, “clade”, etc. If you are not familiar with this issue, see, among others, Ereshefsky (2002) and Laurin (2008, 2010).
93-94: “Other taxa were considered insufficient to include in the dataset,…” Unclear.
97-98: “there were no ordered or weighted characters, and multi-state characters were treated as polymorphs (Mueller-Töwe, 2006).” You need to give some background why this is problematic (in your opinion because not weighting is the most common practice). For ordering, Rineau et al. (2015, 2018) are possibilities. For weighting, possibly Goloboff’s papers on implied weights, which you cited elsewhere in the draft.
114-115, 141-142: “The strict consensus (Fig. 1B) produced 67 MPTs and 1462 steps (CI = 0.28; RI = 0.66).” No. the consensus was produced from 67 MPTs. A consensus is a single topology that is typically not fully resolved. And I hope that the reported statistics refer to the MPTs, not to the consensus, which is always longer.
126-127, 129, 186: “the Chinese teleosaurid (IVPP V 10098) was treated as the metriorhynchoid Peipehsuchus”. Do you mean “attributed to”? Please be precise.
370-372: “Bremer support values of 10 were also calculated which measure branch support and indicate the number of extra steps required for a clade to collapse (Müller, 2004)” Unclear. Bremer values of up to 10 were computed? And note that Bremer invented that index, not Müller. So, in addition to citing Müller’s paper for the technique, it would be good to cite Bremer (1988).
405: The identity of a type-species should not result from a choice; it must be the oldest species that belonged to that species, according to the rules of the zoological code. Is this what Johnson, Young & Brusatte (in press) did ?
415-416: “we believe that it is necessary to state new proposed teleosauroid genera first” In fact, new genera are not stated, they are erected. Please be precise. Biological nomenclature is a highly technical field and the proper vocabulary must be used.
2153-2154: Some character descriptions are not ideal. Take, for instance, this one: “13. Ornamentation present on lacrimal in dorsal view: yes (0), with shallow to deep pits and/or grooves, or no (1) (Fig. 29).” What if you find a taxon with ornamentation that does not consist of pits or grooves (with ridges or pustules, for instance)? Or if it has pits and-or ridges, but if these are visible only in lateral view (if the lacrimal is vertical, for instance). Thus, you might want to simply have present/absent, but in the description that follows, you could indicate that in the taxa that are scored 0 (present), ornamentation consists of pits and/or grooves. You may even suggest, if another type of ornamentation is encountered (like ridges or pustules), whether or not it should be distinguished as a separate state. But ideally, the set of states that you recognize should cover all possible cases (with some inapplicable scores required, for some situations).
2160-2160: For this character “15. Frontal, extension of ornamentation: extends from the centre of the frontal to lateral- and anterior-most regions (0) or restricted to centre of the frontal (1) (Fig. 29).” What if there is no ornamentation? Will you score this as inapplicable? This might reduce information, unless you have a presence/absence in another character. Alternatively, a third state (2) “absent” should be recognized (and then, this would have to be ordered because there is a gradient from well-developed to absent).
Here a little digression. I notice that you have several ornanamentation characters. Just in case that you did not see this I mention that de Buffrénil has recently published several papers on this, notably one (Buffrénil et al., 2015) that focuses on development, and another one (Clarac et al., 2017) on phylogenetic patterns and possible correlates, in both cases in archosaurs. You might want to look at these (if you have not done so, already) to better understand how ornamentation develops and its significance. But don’t feel obliged to cite these papers in this draft. However, note that to understand what determines variation captures in your character 14 (and probably some other ornamentation characters), Buffrénil et al. (2015) might be relevant.
3127-3129: A better formulation should be found for the states of character 102 “Supratemporal fenestrae, shape is either longitudinal ellipsoid or subrectangular (0), square-shaped (1), transverse (= extended) triangle (2), circular (3), triangle-shaped (4), or parallelogram (5) (Fig. 55).” Indeed, sate 2 seems to belong to state 4 because both are triangles, and states 0 and 1 seem to belong to state 5 because squares are rectangles, and rectangles are parallelograms (from most specific to most general polygon categories). But surely, this hierarchy is not intended.
3782-3783: Italicize Charitomenosuchus, Seldsienean, Deslongchampsina and Machimosaurinae.
4234: What are “semi-marine environments” ? Deltas ? Estuaries ? Mangroves ? Or more generally, any coastal environments?
4295: “on land”, not “onland”.
4301: Note that semi-marine is not the same as amphibious; semi-aquatic is.
This is not an anonymous review; the authors should feel free to contact me if they have questions.
Best wishes,
Michel Laurin
michel.laurin@mnhn.fr
References
Bremer K. 1988. The limits of amino acid sequence data in angiosperm phylogenetic reconstruction. Evolution 42:795-803.
Buffrénil V de, Clarac F, Fau M, Martin S, Martin B, Pellé E, and Laurin M. 2015. Differentiation and growth of bone ornamentation in vertebrates: a comparative histological study among the Crocodylomorpha. J Morph 276:425–445. 10.1002/jmor.20351
Clarac F, De Buffrénil V, Brochu C, and Cubo J. 2017. The evolution of bone ornamentation in Pseudosuchia: morphological constraints versus ecological adaptation. Biol J Linn Soc 121:395–408.
Ereshefsky M. 2002. Linnaean ranks: Vestiges of a bygone era. Phil Sci 69:S305–S315.
Grand A, Corvez A, Duque Velez LM, and Laurin M. 2013. Phylogenetic inference using discrete characters: performance of ordered and unordered parsimony and of three-item statements. Biol J Linn Soc 110:914–930.
Goloboff PA, Pittman M, Pol D, and Xu X. 2019. Morphological data sets fit a common mechanism much more poorly than DNA sequences and call into question the Mkv model. Syst Biol 68:494–504. 10.1093/sysbio/syy077
Laurin M. 2008. The splendid isolation of biological nomenclature. Zoologica Scr 37:223–233. 10.1111/j.1463-6409.2007.00318.x
Laurin M. 2010. The subjective nature of Linnaean categories and its impact in evolutionary biology and biodiversity studies. Contributions to Zoology 79:131–146.
Rineau V, Grand A, Zaragüeta R, and Laurin M. 2015. Experimental systematics: sensitivity of cladistic methods to polarization and character ordering schemes. Contributions to Zoology 84:129-148.
Rineau V, Zaragüeta I Bagils R, and Laurin M. 2018. Impact of errors on cladistic inference: simulation-based comparison between parsimony and three-taxon analysis. Contributions to Zoology 87:25-40.
Wiens JJ. 2001. Character analysis in morphological phylogenetics: problems and solutions. Syst Biol 50:689–699.
no comment
no comment
no comment
All my comments are focused in the Discussion section, especially on paleoecologic and biogeographic implications.
Line 4362- Ecomorphological diversity
According to the text, teleosauroids can be classified in different ecomorphotyps, but I think that criteria are a bit mixed.
Only one criterion is common and applied to all taxa, which is that concerning the general skull shape (longirostrine, mesorostrine, brevirostrine). But then, depending on specie they are classified according its main dietary adaptations (piscivorous/specialist, generalist, macrophagous/durophagous); while other are classified according to adaptations to particular environment (semi-terrestrial or semi-pelagic).
Consequently, that makes the results a bit confusing for a non-expert reader (like me). For example, in both text and Fig. 63 Platysuchus, Teleosaurus and Mycterosuchus are identified as belonging to the longirostrine semi-terrestrial guild, but were they generalist? Durophagous?
The same question arise for Aeolodon + Sericodon + Bathysuchus, which are identified as longirostrine pelagic, but is unclear if they were generalist consumers, or specialized piscivorous or macrophagous/durophagous.
Figure 63 does not help to solve these doubts and neither reflects the ecomorphological diversity of the group. Thus, I suggest changing this figure (please, see my proposal below).
Line 4387 to 4434
Authors describe the occurrence of the different ecomorphologic guilds co-existing in the same time period, and even provide evidence about the appearance/disappearance of certain ecomorphotips along time. But in a more macro-evolutionary viewpoint, it remains unclear if the ecomorphologic diversity of teleosauroids changed or remained stable along 70 Ma. In other words, was the number of guilds constant though time?
Could that ecomorphologic diversity reflect any specific niche partitioning strategy among teleosauroids that live at the same time?
In this regard, it would be useful if authors could provide a time-calibrate phylogeny showing the time distribution of each guild. (please, see my proposal below).
Line 4467 - Biogeographical distribution
I have my concerns about the porpoise of this section. It seems that it is focused on describing the current geographic distribution of the fossil remains rather than paleobiogeographic distribution of teleosauroids during the Jurassic. The alleged global distribution of the clade seems inconsistent when data is included in a global paleogeographic context (see attached figure below ,modified from https://www.paleobiodb.org/navigator/).
Despite of the results might be influenced by sampling, outcrop availability or other biases; a particular patter can be observed. As I said, I am not an expert in teleosaurids, but it looks like that teleosaurid tend to concentrate around the Jurassic tropic belts. This biogeographic distribution is even more noticeable when implementing paleo-climate data (Rees et al., 2000).
Rees PM, Ziegler AM, Valdes PJ (2000) Jurassic phytogeography andclimates: new data and model comparisons. In: Huber BT,Macleod KG, Wing SL (eds) Warm climates in earth history.Cambridge: Cambridge University Press, pp 297–318
Additionally, it seems that they primarily diversified and dispersed around the Tethys Sea.
Consequently, if authors want to keep this section in their study, I suggest changing the direction of the paleobiogeographic section. I also recommend changing current Fig. 64 for a paleogeographic map (midd-late Jurassic) showing the biogeographic distribution of teleosauroids.
Line 4585 - Palaeoenvironment
To me, this section is confusing because I cannot see the main goal of the author in this part. Do they want to prove evidence of “convergent evolution” among certain teleosauroids and extant crocs?
Most of the text seems to be focused in justifying the occurrences of teleosaurids in fresh-water deposits, especially Indosinosuchus and the Chinese teleosauroid.
However, I my opinion the results fall in the speculative spectrum rather than prove strong evidence of adaptations to certain environments.
Consequently, I suggest removing this part from the manuscript.
Line 4651 - Teleosauroid histology
This is a short section that can be described as a review about teleosauroid histology. Given that authors do not provide any new data, and that the results of this discussion seems to be disconnects from the rest of the study (which main goal is the systematic phylogenetic review of the group), I suggest to remove this part from the manuscript.
Finally, I have additional comments:
Line 35- add comma after Buffetaut and Hua
Line 271- review the forma of reference, they should go without parenthesis.
Line 341 and 342- review the forma of reference, they should go without parenthesis.
Line 1033- ilial or iliac process?
Line 1265- Something is missing after Indosinosuchus kalasinensis, the name of the author? sp. nov.?
Line 1328-add Jäger between Macrospondylus and 1832
Line 1337- delete "s" from Figs.
Line 1473- Autopomorphic features of Se. megistorhynchus are observed in lateral view. However, in Fig. 15 the specimen is displayed in dorsal view. It would be appreciate if authors could include a picture in lateral view showing the diagnostic feature.
Line 1565- "nasutus" in italic
Line 1792- delete "s" from Figs.
Line 2004- delete "s" from Figs.
Fig. 19 seems a bit unfocussed, can it be changed?
Note: all line numbers refer to the word document originally submitted, as the review pdf is so large that my browser refuses to open it)
Page 3, line 55: something went awry with the tenses here – should this be ‘only shows superficial differences’?
Page 5, lines 98-100: are these values taken from the original thesis or did you redo the analysis yourself? The text is somewhat ambiguous (and the figure description just states ‘altered from’), it would be good to clarify. Same goes for all of the following phylogenies. If you ran the analyses as well to check the trees, consider adding support values to all of the trees.
Page 7, line 152: should be ‘Hastings’
Page 8, line 183 and following: this is more subjective and ultimately up to the authors, but the detailed comparison of the different taxonomic treatments of species and characters between Wilberg & Martin might be better suited to the supplementary material, as it adds unnecessary detail to an already quite long manuscript, especially as none of the other previously mentioned phylogenies have received this level of detailed comparison. Otherwise it could also be deleted or be described more concisely.
Page 11, lines 256-259: The first sentence isn’t really needed, especially as you just stated this only a few paragraphs before. Can be deleted in my opinion and the section just started with ‘The taxonomic sample…’.
Page 12, line 267 and following: once again, this is to a certain extent up to the authors, but I feel like the full species list (especially those species names already known and used previously) and the specific character changes are something that is better situated in the Supplementary than in the main text, as it only distracts from the main points of the paper.
Page 13, line 304: “which has been substantially updated over time” – you mentioned this just one page ago, can be deleted.
Page 14, line 327 and following: everything from here to line 352 can be placed in the Supplementary, especially since the text contains references to the Supplementary already that appear slightly lost on their own without the context (e.g. the newly added characters, ordered characters).
Page 23, line 525: should be ‘is very wide’ or ‘anterior regions’. Also consider quantifying the width with an actual value, e.g. in respect to total skull table width.
Page 52, line 1177: there is one ‘total’ too many in here somewhere
Page 92, line 2076 onwards: perhaps somewhat controversially, I would suggest the entire Character Description section or at least a good part of it (at least in this amount of detail) could well be moved to the Supplementary, and perhaps a shorter description of the new characters be retained. Especially how each taxon is scored is already visible from the provided table in the supplementary, and any peculiarities in scoring for particular taxa can be explained in the supplementary material. The manuscript is 250 pages; I think losing these 60 pages will ultimately be beneficial to the manuscript, especially given that the information and all the hard and impressive work that went into it wouldn’t be lost, but still available to read for those who need it. Smaller paragraphs, such as the differences in ornamentation between species pointing towards more pelagic life histories can be retained and moved to other parts of the manuscript.
In addition, I think the character numbers can also well be integrated into the previous section of systematic palaeontology; instead of just naming autapomorphies and emended diagnoses, the described features can well be supplemented with character numbers.
Alternatively, if these suggestions are utterly unwelcome, the section can still be shortened by grouping the characters together into broader morphological areas (e.g. premaxilla/anterior snout, dentition, etc), where several characters can be talked about at the same time in a description of how certain morphological areas change between different members of Teleosauroidea. This way, the life history/behviour/ecology implications of these changes can be retained for the main manuscript, whilst the detailed scores for single taxa can go into the supplementary.
Page 106, line 2384: ‘forms’ should be singular
Page 153/Results section in general: since you have such a rich and impressive dataset, it would be great if the full tree you obtained would be visible somewhere, for example in the Supplementary.
Page 167, line 3826: ‘support’ instead of ‘supports’
Page 170, line 3902 and following: I feel like the entire ‘excluded taxa’ section would be better situated in the Methods section, when describing the dataset.
Page 178, line 4083/Figure 63: Figure 63 would benefit from a legend for what the different colours mean directly in the figure, not just in the image description. This could also be combined with a silhouette of the snout shapes in question in dorsal view to really illustrate the point. Same goes for Figure 65.
Page 187, line 4295: space missing in ‘onland’.
Page 190, lines 4369 to 4370: one ‘were’ too many – delete the second one.
Page 16, line 374: I would caution against the use of implied weighting only. It has been pointed out in the past that implied weighting can lead to artificially inflated character weights due to missing values, which is of particular concern when using it on a fossil dataset (Goloboff, 2014). I would suggest using extended implied weighting instead, which was specifically developed by Goloboff to counteract that issue. (I also see below on Page 155 that you are citing the Goloboff 2014 paper – did you perchance use extended implied weighting already? If so, correct throughout the paper)
Page 20, line 452: as has been done for most features described here, a concrete number should be given for the length of the external vs the internal antorbital fenestra, rather than just stating ‘significantly’, as this is quite subjective. This goes for several other places in the manuscript, too, including:
- Page 23, line 528, numbers instead of ‘significant’
- Page 36, line 815, numbers instead of ‘relatively reduced’
- Page 56, line 1264, can you quantify ‘slightly elongated’? (as opposed to ‘strongly elongated’)
- Page 67, line 1502, can you quantify ‘premaxillae dorsoventrally high’ with actual ratios somehow?
- Page 79, line 1780, numbers instead of ‘small’, if possible
- Page 80, line 1782, numbers instead of ‘reduced’, if possible
Page 157, line 3607: from which set of trees did you produce the agreement subtree? Only the weighted analysis or all analyses? Add the answer.
no comment
First of all, I would like to congratulate the authors on a monumental piece of work that speaks of an enormous effort from all the parties involved and provides a major step forward in teleosauroid taxonomy. I'm sure this work will provide the foundation for even more exciting things to come and be cited wildly.
The vast majority of my comments are minor and mostly cosmetic in nature, with only two major things (elaborated upon in the specific sections above):
1) Implied Weighting is not the appropriate method for data analysis in this case. You should use Extended Implied Weighting instead, which is the improved version of Implied Weighting. Using implied weights only introduces bias against characters with too many missing scores, which is undesirable in a palaeontology dataset. Hopefully the results shouldn't be too different.
2) Some of the sections (in particular the detailed character discussion) can be shortened/integrated with previous sections and/or moved entirely to the Supplementary in my opinion, to created a 'punchier' paper.
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.