Paleobiogeography of Crown Deer

Round 1

Reviewer 1 Report

I found the article interesting in many ways and certainly up to date but also somewhat verbose in several parts. However, I do not understand the choice of journal to which it was submitted as the content of the paper is probably more appropriate for a biographical journal and far from the aims and scopes of Earth.

The Supplementary files I downloaded from the system were not the correct ones and therefore I cannot comment.

Two important parts of the article (Critical Review of Chinese Forms Cervavitus & Cervoceros, l. 288-389; and Review of the Systematics of the Roussinion Deer, l. 403-465, Fig. 4) are far from the main topic and should be removed. In fact, both of these parts should be the subject of separate analytical articles submitted to the appropriate journals and peer-reviewed. Their inclusion here reduces the coherence of the work.

Figure legends should be more (self-)explanatory

I am skeptical on the scenario of linking early Capreoline radiation with Mediterranean in general, and not fully convinced by the arguments presented by the author. Perhaps a more detailed analysis on a regional (i.e., Mediterranean to Black Sea) and time scale (early – middle – late Miocene) may provide better results to evaluate this hypothesis on a more solid ground.

Multiple minor issues are mentioned directly on the annotated pdf

Author Response

Thank you very much for your detailed and constructive reviewing. I highly appreciated it. Concerning the choice of journal, we agree that biosphere is an important characteristic of Earth that represents the zone of life on Earth. The present paper describes the paleobiogeography of an important systematic group of herbivores (Cervidae) that reveals details of ecosystem evolution at a world scale during the last ca. 10 million years due to the specific for cervids ecological opportunism. The main reason of journal choice is the difficulty to explain some of important events in cervid evolution (almost simultaneous evolutionary radiations of Capreolinae and Cervinae, and the almost complete extinction of the subfamily Capreolinae in the western Eurasia) from the biological point of view. Possibly, an interdisciplinary approach may give more satisfactory answers.  

The responses to the comments:

1) “Two important parts of the article (Critical Review of Chinese Forms Cervavitus & Cervoceros, l. 288-389; and Review of the Systematics of the Roussinion Deer, l. 403-465, Fig. 4) are far from the main topic and should be removed. In fact, both of these parts should be the subject of separate analytical articles submitted to the appropriate journals and peer-reviewed. Their inclusion here reduces the coherence of the work”

 – I AGREE with you that the description of the new genus for “Cervus” ruscinensis should be excluded from the article and published in another paper. However, it is not possible to omit the overview of the Chinese paleontological record of deer and associated taxonomical problems since in this case we are dealing with the diversity of the early evolutionary radiation of the subfamily Cervinae which is biased due to the unresolved taxonomical questions. Therefore, the discussion of taxonomical problems reveals the cervid forms that represent genuine species that are still misunderstood or misinterpreted.

2) “Figure legends should be more (self-)explanatory” – I ADDED some missing details to the figures as you suggested, thank you.

3) “I am skeptical on the scenario of linking early Capreoline radiation with Mediterranean in general, and not fully convinced by the arguments presented by the author. Perhaps a more detailed analysis on a regional (i.e., Mediterranean to Black Sea) and time scale (early – middle – late Miocene) may provide better results to evaluate this hypothesis on a more solid ground” – THIS is the reason why the article is submitted to the journal “Earth”. The causes of almost complete extirpation of the subfamily Capreolinae in Europe during the Late Miocene is not fully clear for me and the answer could lie in the interdisciplinary domain that is outside of my expertise. I proposed the hypothesis that is the most obvious for me as for a biologist-paleontologist.

Replies to the comments from the manuscript file:

1. i) “I think all previous statements suggest Cervids cannot be opportunists.... are you sure you use the correct word? An ecologically opportunistic species does not depend on a particular habitat and can colonize a wide variety of environments - is that what you mean? How tha need of resources rich in nutrients can correlate with opportunism? Please rephrase or explain better” – SPECIES that are capable to colonize a broad variety of environments are usually described as species with broad ecological valence or species with broad ecological tolerance. This is not the case of cervids. Deer have certain eco-physiological limitation and their ecological opportunism actually is an ecological specialization: deer are able to colonize new areas and new ecosystems that provide them with easily accessible and rich in nutrients forage. This is the ecological opportunism (they are capable to use new opportunities) in the sense proposed by Geist (1998). This understanding of term is used in the Paper. I added a precision to the text.
2. ii) “some more details would be welcome here for the wider audience, i.,e the shortly describing the metacarpals in the two subfamilies” – I ADDED a figure showing telemetacarpal and plesiometacarpal conditions in Cervids, but also initial holometacarpal morphology of deer limbs.

iii) “but how geographically and temporally complete is this 'transitional' record and how does the absence of  spatio-temporal data affect the acceptance or not of the hypothesis?” – ACTUALLY, the completeness of the paleontological record is not a problem here. The available fossil material is very rich and complete enough, but poorly understood. We are dealing with a very typical mistake when paleontologists represent the forms with a different degree of specialization as a chain of successive forms representing a sort of “linear evolution”. We see here the same situation as in early attempts to reconstruct the phylogenetic relationships in other systematic groups. Equidae and Hominidae are the best known examples.

1. iv) “the approximate body size estimation is based on pedicle and burr part measurements of fully grown antlers” - on what basis? Please provide additional information on the relation of these two variables with body mass (in text if based on literature or as supplementary if yours)

– I COMPARED the pedicle measurements of fossil forms with species with already known predicted body mass. The assumed body size remains approximate, but it allows to attribute the poorly known cervid forms to a body-size class.

1. v) “The term "crown systematic group" is applied according to 129 the definition proposed by Cirilli et al “ – but the 'crown group' of Cirilli is restricted to the genus level (Equus) which is much more -let say- easily to be controled. You try to apply it to a subfamily level which is a much more wider and artificial group, especially when fossil taxa are included. How may this affect your data handling and interpretation of results? Who and How defines the 'direct extinct forerunner" of subfamilies used here? – THERE is a universal consensus of cevid specialists that the subfamilies Cervinae and Capreolinae are natural monophyletic groups (I have provided necessary references in the text). The present study considers only cervids that belong to Caprolinae and Cervinae. The term “crown deer” is traditionally applied to Cervinae and Capreolinae. Other cervid forms that are not directly related to the origin of Capreolinae and Cervinae (Eostyloceros, for instance) are not considered in the description of Cervinae and Capreolinae radiation. Cirilli et al. just gave the formal definition of the term “crown systematic group”. This formalization of the already broadly used term is important, therefore I made a reference to Cirilli et al.
2. vi) “from this point down to l. 332 yoy provide a kind of systematic remarks and taxonomic revision of several E Asian taxa. Although it is interesting, I think that they merit a proper and much more detailed presentation as a full review paper. With the way you follow, we are already lost in your thoughts and far from the original topic of the article” – I THINK it is important to keep this part of the article since it explains the list of served forms used for the analysis. I indicate the unresolved taxonomical questions and explain which cervid forms and why are considered in the database. Of course, this paragraph may be developed in a further more detailed paper. But for the moment, I have to explain the list of cervid forms used in the present study.

vii) “The drop in the level of the Mediterranean sea and land connections between 700 Europe and Africa created favourable conditions for multiple dispersals of bovid species” - “but unlike your statements here, the Balkano-Iranian late Miocene Bioprovince was full of bovids ranging from <20 to >150 Kg and with more than 5 taxa per site....” – CERVIDS could not disperse into the Greco-Iranian bioprovince and are represented only by very few forms (and remains) from the contact area between European and Greco-Iranian bioprovinces. There are two main reasons: 1) deer are very week competitors with bovids (Geist 1998); 2) deer could not colonize arid areas. More than 5 taxa of bovids per site means that those species are highly specialized and achieved a good partitioning of ecological resources among them. So, there were no chances for cervids (ecological opportunists and generalists) to survive in arid areas inhabited by bovids. Concerning the puzzling mass extinction of early Capreolinae in Europe, I see only the salinity crisis that possibly extended the Greco-Iranian province and perturbed the late Miocene ecosystems of Europe. But I do agree that possibly the answer is more complex and interdisciplinary. This is the reason of the journal choice.

Reviewer 2 Report

Material and methods:

not clear how to correlate the predicted body mass can be a proxy for evolutionary differentiation and occupied ecological niches.

Deers with similar body size can occupy different niches or can display totally different antler morphology (fossil species)

Add further explanations.

The data used in the analyses must be free of access, with a correct indication of specimen number and anatomical element used for the body size estimation.

Description.

This section is a little bit confused, being represented partially by a synthesis of data, partially by interpretations of the author and partially being represented by new data.

Is this a review? or is the author giving new hypotheses on evolution of this group of mammals?

The section on the new genus is out of topic, and cannot be placed inside this section. At least the author must organize in a different way this section, including a kind of systematics and then discussing each taxon at species or genus level. Or the author can add a systematic section and then a description (or better a resume) of the evolutionary radiations of the discussed groups.

A figure with the diversity of antlers of Cervinae is needed.

I fig. 3 it is not clear what are the author representing, is the number of species?

A figure with the skulls or remains of selected discussed groups is welcomme.

It is not clear which kind of anatomical element was used to calculate the body size, being this value affected by the type of bone/tooth used in the formula. Are the body masses of all the species calculated using the same anatomical element?

Discussion.

This section needs to be improved. A figure showing the morphological characters described by the author as very important from an evolutionary point of view, suchs as p4 and metapodials, must be added. In this way the reader can see what the author means here. Further, these elements should be described during the description of the different groups, being these elements used to discriminate the different groups. What about the evolution of antlers?

Fig. 6, a time-factor should be added, showing the distribution of body mass by unit of time (possibly not period, but smaller time intervals)

A fgure showing the phyletic relationships within the discussed groups must be added in the introduction or in the description, showing the relationships (or supposed relationships) and the geographic distribution of the groups

Conclusions.

the convergent features must be included within a figure and must be better discussed in the discussion, only in the conclusion the author are giving some explation of the relationships between this feature and external factor.

The short-living species of the Pleistocene are not discussed nor named in the descritpion; so, Who are they?

Author Response

Thank you very much for your valuable comments. They were very helpful. Here are replies to your comments:

1. “Material and methods: not clear how to correlate the predicted body mass can be a proxy for evolutionary differentiation and occupied ecological niches. Deers with similar body size can occupy different niches or can display totally different antler morphology (fossil species)”

– ACTUALLY, I provided the necessary references, explaining how the body mass in cervids ensures the ecological partitioning among deer. This is a well-known phenomenon. Practically all deer species from the same fauna and the same geological epoch have different body sizes. As deer are opportunistic and ecologically flexible feeders, they cannot occupy different ecological niches if the body mass is the same. This is a fundamental ecological difference between cervids and bovids.  Different antler shape ensures the genetic isolation of deer species, as antlers are responsible for intraspecific nuptial communication, but antlers do not protect deer from the competition for the same food resource. See for instance Geist, V. (1998. Deer of the world: their evolution, behaviour, and ecology. Mechanicsburg, Pennsylvania: Stackpole books) who described this phenomenon in details. I added some precisions to the text.

2. “The data used in the analyses must be free of access, with a correct indication of specimen number and anatomical element used for the body size estimation”

– THE most of data used in the analysis come prom scientific articles published in the recognized scientific journals. All bibliographic sources are indicated in the attached bibliographic list. The provided information on specimens comes from the cited sources. it is not possible to add more details on fossils than those indicated by authors (some of them passed away).

3. “Description: This section is a little bit confused, being represented partially by a synthesis of data, partially by interpretations of the author and partially being represented by new data. Is this a review? or is the author giving new hypotheses on evolution of this group of mammals?”

– THIS is a paleobiogeographic study based on fossil deer species. Some taxonomical questions were not resolved before, therefore it is necessary to clarify them in order to advance toward the research aim.

4. “The section on the new genus is out of topic, and cannot be placed inside this section. At least the author must organize in a different way this section, including a kind of systematics and then discussing each taxon at species or genus level. Or the author can add a systematic section and then a description (or better a resume) of the evolutionary radiations of the discussed groups”

– YOU have a reason. I will keep the name “Cervus” ruscinensis as such and the description of the new genus will follow in another paper.

5. “A figure with the diversity of antlers of Cervinae is needed”

– IT is not possible in the current circumstances. I have provided the diversity of antlers of early representatives of the subfamily Capreolinae. I have studied personally most of the fossil material of early Capreolinae. Thus, an analogous figure of antler diversity of early representatives of Cervinae is needed. Almost all fossil material on early Cervinae is stored in the East Asian institutions are unavailable for me for the moment. I cannot have my firm opinion based only on confused old publications. I indicated the contradictions and outlined the general characteristics of the early evolutionary radiation of the subfamily Cervinae. The aim of this paper is to demonstrate there were two equally diverse evolutionary radiations of Cervidae during the Late Miocene. The general diversity of antler shape in the subfamily Cervinae, including its late representatives does not correspond to the scope of the paper.

6. “fig. 3 it is not clear what are the author representing, is the number of species?”

– YES, this is a number of species, I have added the scale on the right side.

7. “A figure with the skulls or remains of selected discussed groups is welcome” – I HAVE included in the paper illustrations of the skulls originally described as “Cervocerus novorossiae”.
8. “It is not clear which kind of anatomical element was used to calculate the body size, being this value affected by the type of bone/tooth used in the formula. Are the body masses of all the species calculated using the same anatomical element?”

–THE body mass is calculated according to the method proposed by Janis (1990). This is a broadly accepted method and all my calculations from earlier publications are based on this method. I specified in the text which teeth exactly were used for body mass estimations.

9. “A figure showing the morphological characters described by the author as very important from an evolutionary point of view, suchs as p4 and metapodials, must be added. In this way the reader can see what the author means here. Further, these elements should be described during the description of the different groups, being these elements used to discriminate the different groups. What about the evolution of antlers?”

– I ADDED the figures of metapodials since they come from the old publications that may be difficult to find. The evolution of antlers and fourth lower premolar do not represent the aim of this paper, this is a biogeographic study and I included only information that is important from the paleobiogeographic point of view. Actually, a detailed discussion of antler evolution and dentition is already discussed in my recently published paper (Croitor, R., 2021. Early evolutionary radiation and diversity of the Old World telemetacarpal deer (Capreolinae, Cervidae, Mammalia). Neues Jahrbuch fur Geologie und Palaontologie-Abhandlungen, 300(1), pp.33-67); there is no need to repeat the same things again in this paper.

10. “Fig. 6, a time-factor should be added, showing the distribution of body mass by unit of time (possibly not period, but smaller time intervals)” – THE diagrams on this figure show the body mass diversity of deer that resulted from the thee important evolutionary radiations of cervids: the Late Miocene radiation of Capreolinae, the Late Miocene radiation of Cervinae, and the Pleistocene radiation of Capreolinae in South America. The time factor has no sense here, this is not the case. The time intervals smaller than periods, unfortunately, are not available. The paleontological record of Europe is well-dated and allows a fine geo-chronological periodization due to MN zones. The geochronological dating pf the cervid paleontological record of Eastern and Southeastern Asia and South America still needs a refinement.
11. “A fgure showing the phyletic relationships within the discussed groups must be added in the introduction or in the description, showing the relationships (or supposed relationships) and the geographic distribution of the groups” – THE phylogenetic relationships of all the deer of the world is outside the scope of this paper. I cannot do it alone since this study requites an extensive revision and study of cervid remains from all the world. The hypothetically supposed relationships will remain speculative. This is what I want to avoid in my article.
12. “the convergent features must be included within a figure and must be better discussed in the discussion, only in the conclusion the author are giving some explation of the relationships between this feature and external factor” – AS you suggested, I added a figure illustrating a striking convergent evolution of Morenelaphus brachyceros from the Pleistocene of South America and Cervus elaphus angulatus from the Pleistocene of Europe.
13. “The short-living species of the Pleistocene are not discussed nor named in the descritpion; so, Who are they?” – I MENTIONED in the text some short-living deer species from the Pleistocene of Eurasia as you suggested.

Reviewer 3 Report

The manuscript by Croitor "Paleobiogeography of crown deer" provides a review of the fossil record of Cervinae and Capreolinae arranged by biogeographic realms and chronological time-spans, accompanied by selected discussions on certain morphological and taxonomical aspects. The author also introduces a new genus, "Eocervinus", for "Cervus" ruscinensis.

The paper offers an interesting review with a reasoned synthesis of much research. However, it has, in my opinion, at least one important flaw and other minor but still significant points that must be considered.

My first and most relevant concern is on the newly introduced taxon. The content of this paper is that of a review and a species-level analysis of diversity. Some taxonomic opinions are necessarily expressed on previously published species and specimens, if these are deemed of doubtful or divergent applications of what previously published, but the choice of naming a new taxon here sounds peculiar. What I mean is that the analysis would remain unchanged if the species is treated as "Cervus" ruscinensis, and a proposal of a new genus should be corroborated by a dedicated and broader taxonomical study, rather than been introduced “between the lines” of a work with a different focus. It is not entirely clear what comparative material has been considered for supporting this attribution, what specimens are included in E. ruscinensis that provide evidence for the traits considered diagnostic, what implications, if any, on the relationships within the genus and between other species can be drawn... I'm confident that the author justification for this proposal is stronger than appears, but here does not seem the appropriate place. Indeed, including an even longer digression on "Cervus" ruscinensis aimed at better supporting its reassignment would be beyond the scope of a paper that otherwise works quite well.

In sum, I strongly recommend to refer to the taxon as "Cervus" ruscinensis in this paper.

I have two other relatively minor concerns I ask the author to consider. Both are related, let's say, to acknowledge some degree of uncertainty.

1) There is much debate on the attributions of several taxa treated in the text, with radically different opinions expressed on the validity of certain species and genera. This has, evidently, implications for the reconstruction of diversity patterns.

I suggest adding considerations at the genus level, which is frequently adopted in macroecological studies, to partly diminish the problem. The identified genera can be added in figure 3 (or preparing a separate figure of similar structure) and salient differences between patterns observed at the species and genus levels discussed.

2) The author used body mass as "a proxy for evolutionary differentiation and the range of occupied ecological niches". This is common practice. However, body mass values are necessarily estimated from equations fitted on linear measurements. This is also common practice, but such estimates have error ranges that should be kept into account.

Aside from discussing it in the text, I suggest to redrawn figure 7 (which is misnamed as figure 6, by the way) including a "buffer zone" derived using minimum and maximum values of the estimates (not just the mean). The error ranges are provided in the same work followed for the estimates.

Partly related to the latter point, but also for general clarity. I recommend including points of the single specimens in the boxplots of figure 5. The methods for constructing the graph should also be specified

One last minor thing:

- figure 3 needs a scale with the number of species to be interpreted. It can be added manually on the right side.

Author Response

Thank you very much for your valuable and constructive comments and suggestions. They were very helpful. My replies to you questions and suggestions:

1. i) “My first and most relevant concern is on the newly introduced taxon … In sum, I strongly recommend to refer to the taxon as "Cervus" ruscinensis in this paper”

– I TOTALLY agree with you, the description of new genus is excluded from this study and will be published later in another paper.

1. ii) 1) “There is much debate on the attributions of several taxa treated in the text, with radically different opinions expressed on the validity of certain species and genera. This has, evidently, implications for the reconstruction of diversity patterns. I suggest adding considerations at the genus level, which is frequently adopted in macroecological studies, to partly diminish the problem. The identified genera can be added in figure 3 (or preparing a separate figure of similar structure) and salient differences between patterns observed at the species and genus levels discussed” – I UNDERSTAND the problem and the solution you proposed. Unfortunately, practically it is impossible to do because many of taxonomical questions are still unresolved on the genus level. The genus taxonomic level is not safe at all and does not guarantee unbiased character of data (the case of “Cervus” ruscinensis is a good example). I added to the paper a figure with skulls of two clearly different cervid forms from the Tertiary of China that should be placed in two different genera, but both originally are described as “Cervocerus novorossiae”. Therefore I prefer to take in consideration “cervid forms” that obviously represent a taxonomically genuine species, but still requires a taxonomical revision.

 iii) 2) “The author used body mass as "a proxy for evolutionary differentiation and the range of occupied ecological niches". This is common practice. However, body mass values are necessarily estimated from equations fitted on linear measurements. This is also common practice, but such estimates have error ranges that should be kept into account. Aside from discussing it in the text, I suggest to redrawn figure 7 (which is misnamed as figure 6, by the way) including a "buffer zone" derived using minimum and maximum values of the estimates (not just the mean). The error ranges are provided in the same work followed for the estimates”

– THE figure in question is a kernel density estimate plot (kde plot) that presents the distribution of body masses within the subfamilies and the body mass estimation error range is not possible here. The diagram presents a sort of “density” of body mass distribution.

5. iv)  “Partly related to the latter point, but also for general clarity. I recommend including points of the single specimens in the boxplots of figure 5. The methods for constructing the graph should also be specified”

– I INCLUDED the version of the figure 5 with outliers as you suggested.

1. v) “figure 3 needs a scale with the number of species to be interpreted. It can be added manually on the right side”

– I ADDED the scale for the number of species, thank you.

Round 2

Reviewer 1 Report

The new version of the article is significantly improved-some minor linguistic and editing improvements are still needed (see annot. pdf).  While I do not necessarily agree with all of the author's conclusions, I believe that his reasoning is sufficient and supports his opinion. I still cannot understand S2 in its present form and how it can be used by experts.

Comments for author File: Comments.pdf

Author Response

I am very thankful to you for your careful work with the article text. I accepted practically all your suggestions. My special thanks for English editing.

Reviewer 2 Report

Dear Author,

thank you for your revised version and for your reply.

I saw that almost all my advices have been accepted and the text has been modified accordingly or some corrections have been made.

Only a remark here:

I think that a figure with the evolution of p4 would be helpful for the reader, similarly to that on the metapodials (added by the author in the new version); this would simplified the lecture of the text by no-specialists. But this is my opinion anyway.

All the best

Author Response

Thank you very much for your revision that improved the quality of the paper. I was ready to accept your suggestion to include one more figure on lower P4 morphology (a compilation of figures from my older publications), but finally I arrived to the conclusion that the literature providing a detailed description of lower P4 morphology and evolution is easily accessible on the web. Nonetheless, the figures on lateral metacarpals and skulls are necessary in the paper, since they are published in old and difficult-to-access papers.

Best regards,

Reviewer 3 Report

In the revised version of the manuscript, the author either adhered to my suggestions or adequately reply to them. I have no further concerns precluding the publication of this work in the present form.

But in the caption of figure 7 correct "asapted from ****" when revising the proof.

Author Response

Thank you very much for your revision that improved the quality of my paper. I have corrected the caption to the figure 7, thank you for indicating this mistake.