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.
Thank you for dealing with all of the reviewers' comments and mine.
(For Figure 7, please consider not having a y-jitter and rather using an alpha setting of 0.5 for semi-transparent points. Also limit the y-axis to 0 so that the plots do not extend into the negative. )
[# PeerJ Staff Note - this decision was reviewed and approved by Richard Schuster, a PeerJ Section Editor covering this Section #]
Thank you for all of those edits. The manuscript is looking great. I just picked up on one last thing (my apologies for not picking it up earlier), but in Figure 7 - how can you have negative values for Mount Camizard for abundance values? Surely it should be 0 or higher, yes? On line 274 you define abundance as epiphytes per tree, so I assume that is the same for the orchids (it is not actually stated in lines 288-305).
Dear authors,
Thank you for your revision. One of the original reviewers and myself are satisfied with your revisions.
I have suggested a series of minor edits below that should be quick and easy to implement. My comment on Figure 7, however, is important for further investigation.
In figure 4, please include 0 in the y-axis of Brise Fer, and please see if the Mount Camizard pattern remains visible if both are shown on the same y-axis scale. When plotting for comparative purposes, changing the scale makes it very difficult for the reader to readily make a comparison between the sites.
Figures 5 & 6. The lower error bar is not visible with the black background. These look like ggplot figures, so you can plot an broader white error bar underneath, and then the error bar. Code below
ggplot(df, aes(x, y)) +
geom_col(fill = "black") +
geom_errorbar(aes(ymin = y - se, ymax = y + se),
width = 0.2,
colour = "white",
size = 1.8) + # thick white underlay
geom_errorbar(aes(ymin = y - se, ymax = y + se),
width = 0.2,
colour = "black",
size = 0.6)
***NB*** The flat lines for the box and whisker plots in Figure 7 look decidedly odd. You have provided the jittered raw data points which show a good spread, and I don't see any reason for such a single line to appear. I strongly suspect that there is a figure generation error here...
Minor fix. Please put a white fill background behind the subfigure letters (a) and (b) in Figure 8 to make them stand out more. I would also suggest standardising the subfigure lettering for Figs 2 & 3 to white fill background with black text.
Figure 1. You may wish to consider using a coloured digital elevation model. The 100 m contours are exceptionally messy and I battle to interpret them around Brise Fer as they are very bunched up.
Keywords: "ecosystem dynamic" to "ecosystem dynamics"
Ln 270. Please cite R and the version. We cite R because it performs the actual statistical computations, whereas RStudio is only an interface and does not affect scientific results.
Ln 480. "since decades" to "for decades". And can you be more specific. "For over three decades" for example.
Ln 487. "Loping" to "lopping"
Acknowledgements. Please thank the reviewers as they have done a great job in constructively critiquing, and thus improving, the manuscript.
The manuscript has been thoroughly revised and substantially improved in response to the reviewers’ comments. The revisions have addressed the major concerns raised, and the current version meets the necessary standards and is acceptable for publication.
All issues related to the experimental design raised during the initial review have been appropriately addressed and improved by the authors in the revised manuscript. The current version is acceptable for publication.
The same applies to the Results and Findings section. All issues related to the findings, including previously raised overstatements during the initial review, have been appropriately addressed and improved in the revised manuscript. The authors have also included the necessary underlying data, such as tables and figures. The current version is valid and acceptable for publication.
The authors have substantially improved the entire manuscript in response to the reviewers’ comments.
The reviewers have provided a range of very valuable suggestions that I believe will greatly improve your manuscript. One of the major suggestions is to use a GLM to analyse the data. Another important question is the age-old comparing apples with apples issues — are the species readily comparable in terms of their traits and if not, might this bias the results? All reviewers agree that this is an important study, and I look forward to your revised manuscript.
**PeerJ Staff Note:** Please ensure that all review and editorial 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 review process has identified that the English language must be improved. PeerJ can provide language editing services - please contact us at [email protected] 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
No comment
No comment
No comment
I was happy to read the manuscript entitled “Native pioneer trees can be important phorophytes: Their control for biodiversity conservation on an oceanic island also harms native epiphytes and lianas” (peerj-119098) submitted to PeerJ. This manuscript presents a valuable dataset on epiphyte and liana communities associated with a single pioneer tree species, which serves as a model. Comprehensive datasets on epiphytes and lianas remain rare, especially those that focus on well-defined host trees. Moreover, this was a fluid and easy-to-read manuscript, with clear study proposals, development, and conclusions, and the authors deserve merit for that. However, I have some minor concerns and suggestions for some points in the text, which can be found below:
Major comments:
1. My major concern is regarding the inclusion of a linear regression model in the manuscript. While the study presents important descriptive and qualitative insights into the richness and diversity of epiphytes and lianas associated with native phorophytes, I believe the manuscript would benefit significantly from the inclusion of a more quantitative analytical approach. Specifically, I suggest that the authors explore the use of linear regression models to examine the relationships between richness and diversity metrics (as response variables) and tree characteristics, such as tree size and age (as predictors) This approach would provide a more rigorous test of the influence of host trees (i.e., phorophyte) traits on epiphyte and liana communities and offer a clearer understanding of the strength and direction of these associations. Including such analyses would also enhance the analytical depth of the study and enable more robust interpretations of the patterns observed in the field. In addition, probably there is a hierarchical structure in the data (e.g., trees sampled in Brise Fer and Mount Camizard locations), so the authors might also consider using linear mixed-effects models to account for potential non-independence among samples, using location as a random effect.
Specific comments:
2. Line 124: “parallel host specificity”. I miss the definition of this concept, and I recommend that the authors define that
3. Lines 172 – 173: “In all, 219 woody plants were sampled, including 73 trees of each category.”. Is this the number of sampling units in the analysis?
4. Lines 209 – 212: Good statement here
5. Lines 2019 – 220: “In all, 81% (1,805 of 2,229 individuals) of all epiphytes and lianas observed…” Observed in how many trees? Authors should also provide the number of evaluated trees once that they represent the sampling units of the study
6. Line 231: “The final dataset used for analysis included the respective taxa classified as morphospecies”. Authors could inform here the total number of epiphytes evaluated (it’s 2229 individuals, right?) and the number of studied trees (phorophytes)
7. Lines 233 – 243: I suggest that the authors consider performing linear regression models using richness and diversity metrics of epiphytes and lianas as response variables, and tree size and age as predictors. This type of analysis could provide valuable quantitative insights into how tree characteristics influence epiphytes and lianas communities. Including such models could enhance the robustness of the analytical framework and clarify the strength and direction of these relationships
8. Line 259: “Epiphyte or liana were observed on 116 of 219 (53%) sampled potential phorophytes”. I think that this information should appear earlier in the manuscript, once they represent the number of studied phorophytes (i.e., they are the sampling units of the study)
9. Figure 4, 5 and 6: Please provide panels a) and b) for Brise Fer and Mount Camizard locations to enhance data representation
10. Table 2: All tests were significant, this is right? If yes, authors could provide a statement in the text claiming it
11. Lines 314 – 315: “Our study also aligns with previous findings that tree age and sizes strongly influence phorophytic function”. Good statement here, but this could be examined properly using a linear regression
12. Lines 363 – 368: Excellent statement!
13. Lines 417 – 418: “…where epiphytes and lianas remain a particularly diverse and also largely overlooked component of native plant diversity…”. This is true and may be a general pattern regarding epiphyte communities, once they remain understudied in most ecosystems worldwide, despite their significant ecological roles and high diversity (e.g., see Silva et al., 2025 which examined epiphyllous bryophytes within the Brazilian Atlantic Forest). I encourage the authors to further highlight the novelty and relevance of their focus on epiphytes and lianas within this broader research context. Please see:
Silva et al., 2025. Diversity patterns and knowledge gaps of Atlantic Forest epiphyllous bryophytes: a highly neglected group. Annals of Botany, 135, 1047–1058. https://doi.org/10.1093/aob/mcaf007
I am available if the authors want to clarify any doubts
Best regards,
Jean M. Freitag Kramer
Some sections are ambiguous and would benefit from further clarification. Please avoid overstatements and make sure that relevant results support your claims to enhance the validity and impact of your conclusions. The Figures can be improved by adding more iamges of the epiphytes and lianas. For specific comments and suggestions, please take a look at the comments in the file I've attached. Overall, the rest of the manuscript is acceptable.
The research questions are well-defined, and the methods are generally appropriate. However, the Methodology section would benefit from clarification regarding how "fitness" was assessed, specifically, whether the parameters used are based on established ecological criteria or were defined uniquely for this study. Clearly outlining the indicators used to measure fitness, such as growth, survival, or reproductive output, will enhance the clarity and ecological relevance of the findings. Additionally, provide more details on the specimen identification process. This should include the taxonomic procedures followed, references to relevant literature or identification keys used, and whether expert consultations were involved. Thorough documentation of the identification process will improve the credibility, transparency, and reproducibility of the study.
All underlying data have been provided and appear robust and statistically sound. However, the results presented in Table 2 include a Chi-square value of less than 1, but the degrees of freedom (df) are not reported. Additionally, the method for determining statistical significance is unclear. Please include the degrees of freedom and specify the criteria used to assess significance, such as the corresponding p-values or threshold levels, to enhance the transparency and interpretability of your statistical analysis. The conclusions are generally well stated; however, some statements require further justification to avoid overstatement.
The manuscript presents relevant and timely findings on the ecological role of Harungana madagascariensis, particularly its function as a phorophyte supporting native epiphytes and lianas. The current title, however, could be improved to better reflect these key findings and their broader conservation implications. I recommend revising the title to highlight both the species’ ecological function and the potential consequences of its removal in restoration contexts, for example: "Harungana madagascariensis as a Key Phorophyte Supporting Native Epiphyte and Liana Recovery in Oceanic Island Forests."
However, some statements in the Abstract and Discussion are overstated and require stronger justification through relevant data. In particular, the assertion that native trees promote “recovery” is not sufficiently supported by the current evidence, which mainly demonstrates the role of Harungana madagascariensis in hosting abundant epiphytes and lianas. Without comparative or temporal assessments, the term “recovery” may be premature or misleading.
Some statements in the Results and Discussion also lack proper referencing. Additionally, while the manuscript raises an important hypothesis that pioneer native trees may become dominant or even invasive during succession, potentially suppressing other ecologically important species, this idea needs clearer justification based on the study’s findings. A more conclusive interpretation of how your results support or challenge this hypothesis, along with the ecological or conservation implications, would significantly strengthen the manuscript.
I find this study interesting and potentially valuable, as its results may serve to guide improved conservation programs on this island that integrate more floristic components of the forest as part of conservation practices. However, it has some conceptual and methodological limitations or flaws that, in my view, question the validity of the conclusions.
In general, the text is clearly written, and professional English is used. However, the language needs improvement in several sections. Some examples where the language could be improved include lines 55-56, 60-61, 139-140, 156-157, 259, 345; the current wording makes comprehension difficult. I suggest you have a colleague who is proficient in English and familiar with the subject matter review your manuscript, or contact a professional editing service.
The Introduction provides sufficient background and context, and the cited literature provides reasonable support for this context.
Figures 4 - 7 could benefit from using bigger lettering, as all labels are too small compared to the size of the graphs.
Generally, the results are relevant to the research questions (the study does not present a hypothesis explicitly), but some improvement is needed, as explained in the next sections.
Although the design of the study is clearly explained and the methods are described with sufficient detail to enable replication, I found some issues that need improvement.
(1) L 43. More species in H. madagascariensis but no more diversity? Does this mean that, in your view, species richness is not part of community diversity? This confusing statement stems from the use of "diversity indices" proposed by different authors over the last century, whose interpretations are obscure, and their comparability is challenging. For example, Fisher's alpha cannot be interpreted in the same way as the Shannon-Wiener (not Weaver) index (see Magurran 2013). For decades, the use of diversity indices instead of diversity measures has hindered our understanding of community diversity. Fortunately, this problem was largely solved with a solution proposed also decades ago, namely Hill numbers (Hill, 1973), which was widely adopted relatively recently (see, for example, Jost 2006, 2010; Chao et al. 2014a, b; Hsieh et al. 2016). The main advantage of using Hill numbers is that you always use the same currency to measure diversity (namely, the effective number of species), calculated with a formula with one parameter (q) whose changes allow for giving increasing emphasis to the relative abundance of the species in the community. Typically (but not necessarily), q = 0, q = 1, and q = 2 are used to roughly measure species richness, abundant species, and dominant species (q is a continuous variable and thus any value can be used). I sincerely invite you to use this popular framework and to abandon the dated indices (none of them being 'measures' but 'indicators' of community diversity, unlike Hill numbers). Since you seem to be skilled R users, this change should represent no problem to you (just use the HillR or the iNExt package). This change is important because this will help you to objectively evaluate whether epiphyte and liana diversity is indeed more or less even in the pioneer or in the non-pioneer species. For example, recording more species on pioneer hosts could be compensated for by a more even distribution in non-pioneer hosts.
(2) A less important concern is related to another methodological aspect of your study. In Methods, you wrote: "For larger trees (> 8 m height), we restricted our census up to the first section of the canopy (e.g. 1/3 of the branches length), equivalent to "Zone 3" (Johansson, 1974), to avoid observation bias due to the high probability of missing individuals higher up". I am not a specialist in epiphyte ecology, but during my career as a tropical forest ecologist, I have participated in many vegetation surveys in which this guild was considered, and I agree with you that surveying epiphytes above a certain tree height is difficult. However, this methodological decision casts serious doubts on your results and conclusions, as non-pioneer trees usually attain larger sizes than pioneer ones. Thus, you cannot discard the possibility that the upper reaches of non-pioneer trees host a denser and more diverse epiphyte community than you actually recorded. To address this challenge, different procedures can be used, for example, examining recently fallen trees or branches (in addition to climbing trees), but you did not perform any of these procedures. Therefore, your strong conclusions about the inadequacy of the control of the focal pioneer tree must be toned down.
(3) L 39, 48-49, 127, 255, 319. "Fitness" is formally defined in evolutionary ecology as a variable that measures the differential survival and reproduction of the different genotypes present in a population. This is not what you assessed. You compared size (number of leaves), which you interpret as an indicator of differential performance. Therefore, I suggest replacing this term with a more adequate one, such as "size" (what you measured) or "performance" (your interpretation of size). More critical, however, is the fact that you compared this individual plant attribute between pioneer and non-pioneer species for a group of plants belonging to several species of a single genus (Angraecum spp.). To warrant such a comparison, you should first demonstrate that individuals of the different species of this genus growing under the same conditions do not differ in their number of leaves. Without this demonstration, the possibility exists (no matter how small or large) that you are comparing plant size among different species rather than among different host trees. The results described in lines 292-294 ("In addition, there was a significant difference of size (in terms of number of leaves) for Angraecum spp. 2 = 24.78, df = 5, p < 0.05) but not for Angraecum pectinatum 2 = 3.35, df = 5, p > 0.05) among phorophytes across both sites") reflect this concern perfectly. This flaw makes this section the weakest and, in my opinion, it should be removed from the manuscript.
(4) L239-243. Using GLMs (for example, with the lmer function in the lme4 package) would have been a better way to analyze these data and perform the comparisons, as this would have allowed you to introduce the site as a random factor and thus control for the differences between the two sites.
(5) In Lines 238-239 you stated: " Shapiro-Wilk test was used to test the distribution of abundance and species richness data". This choice seems unwarranted, as the Shapiro-Wilk test is a well-known and amply used normality test. Please clarify this procedure or correct it if needed.
References:
Chao, A., Chiu, C. H., & Jost, L. (2014a). Unifying species diversity, phylogenetic diversity, functional diversity, and related similarity and differentiation measures through Hill numbers. Ann Rev Ecol Evol System, 45(1), 297-324.
Chao, A., Gotelli, N. J., Hsieh, T. C., Sander, E. L., Ma, K. H., Colwell, R. K. & Ellison, A. M. (2014b). Rarefaction and extrapolation with Hill numbers: a framework for sampling and estimation in species diversity studies. Ecol Monog, 84(1), 45-67.
Hill, M. O. (1973). Diversity and evenness: a unifying notation and its consequences. Ecology, 54(2), 427-432.
Hsieh, T. C., Ma, K., & Chao, A. (2016). iNEXT: an R package for rarefaction and extrapolation of species diversity (Hill numbers). Methods Ecol Evol, 7(12), 1451-1456.
Jost, L. (2006). Entropy and diversity. Oikos, 113(2), 363-375.
Jost, L. (2010). The relation between evenness and diversity. Diversity, 2(2), 207-232.
Magurran, A. E. (2013). Ecological Diversity and its Measurement. Springer Science & Business Media.
A major concern I have about your valuable work is related to the main message you convey in your manuscript. After reading your paper, I was left with the impression that there is a big conservation conflict in Mauritius that practitioners and scientists are facing: either conserve a high tree diversity or conserve a high epiphyte and liana diversity. To me, both are relevant goals, of course, but I cannot imagine a healthy forest with a very low tree diversity and no high epiphytic diversity is possible without a healthy tree community. Therefore, I wonder if this is a real conflict or rather a situation that must be tackled in a more holistic way, without giving higher priority to one side of the coin or the other. I understand that you are truly concerned about the future of epiphyte and liana communities in Mauritius' forests, but I believe that you are equally concerned about your island's tropical forests. Therefore, I would like to invite you to consider rewriting the Discussion, particularly in lines 363-368 and in the Conclusions, to convey a more balanced message. This revised Discussion should specifically address the fact that H. madagascariensis is a short-lived species (you mention a decline of its individuals after 10 years), and the best conservation decisions must be made for the mid- and long-term. I agree with you that this pioneer species plays an important role in the initial years of forest recovery in the large gaps created by the elimination of invasive species, but whether or not it can continue playing this role when the slow-growing mature forest species take over the forest canopy is uncertain.
You could consider using a shorter, more concise, and appropriate title. Particularly worrisome is the use of the word "also" in the second sentence, as this implies that the control of Harungana madagascariensis is, according to your results, detrimental not only to the epiphyte and liana communities but also to something else, but you did not study anything else than these communities. I understand that you are concerned about the future of H. madagascariensis populations, but you did not investigate this issue, and therefore, you should not mention it in the title.
Other minor comments:
L 30. This opening sentence of the Abstract is vague: "threats" to what? That is, what is being threatened that drives the rapid biodiversity loss?
L 33 and 37. Change "within" to "in".
L 34. Which attributes of these communities did you compare? This should be stated here, because this statement is vague.
L 44. 'phorophytes'
L 50. 'benefits'
L 58. 'epiphyte and liana guilds'
L58-59. The sentence that reads "Our study underscores how native pioneer trees can help accelerate ecosystem recovery" is an overstatement. You didn't assess rates of ecosystem recovery in your study, did you?
L 70. 'suffer' (present tense, as they still suffer).
L 87. Delete 'the inclusion of'.
L 102. 'few studies have investigated'.
L 122. 'compare as a phorophyte'.
L 166. Please, report the exact sampling method that guaranteed randomness in the sampling.
L 168 and 198-212. What was the rationale to use tree age as a classifying variable of trees being compared? Determining age required a lot of time and effort, according to your description of methods, but your decision to include this attribute in your analysis is unclear.
L 173. 'phorophytes'.
L 176 and elsewhere. I suggest that you use the standard terms 'vegetative' and 'reproductive' instead of 'vegetating' and 'reproducing' throughout the text.
L 234. 'was analysed'
L 234. The correct name of this index is Shannon Index or Shannon-Weaner Index (not Shannon Weaver). However, as mentioned in my main comments, diversity assessment should be done based on true diversity measures (Hill numbers). By the way, Hill numbers of order q = 1 are equivalent to the exponential Shannon Weaner Index (i.e., the transformation of the index whose units are bits, logits, or decits, depending on the logarithm base used for its calculation, to effective numbers of species.)
L 259. 'Epiphytes and lianas were recorded...'
L263-264. These figures reflect an extremely uneven distribution of species abundances, which could be adequately captured by using Hill numbers (i.e., much lower order q=2 than order q = 0 Hill numbers).
L 272-273. This statement differs totally from what you said in lines 43-44 of the abstract! Please check this.
L 274-275. I am not sure exactly what groups you are comparing here with the Kruskal-Wallis test and thus I can't figure out why you have five degrees of freedom. Please, explain this in the methods better.
L 307. 'therefore'? 'furthermore'? 'relatively'? You may delete these words for increased clarity.
L 321. As argued above, your results about individuals of Angraecum spp. having more leaves on Harungana trees are not convincing simply because various species (each one potentially having different numbers of leaves) were combined for this analysis.
L 333. 'it is'
L 336. 'are found' (richness and abundance are two different things).
L 352. 'including some of the'
L 359. 'other species that grow'
Sincerely,
Jorge A. Meave
Professor Emeritus
Facultad de Ciencias, Universidad Nacional Autónoma de México
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.