Otolith Microchemistry and Demographic History Provide New Insight into the Migratory Behavior and Heterogeneous Genetic Divergence of Coilia grayii in the Pearl River
, Qindong Tang 1,3,4,†, Zhi Chen 1,3,4, Dingli Guo 1,3,4, Lei Zhou 1,3,4, Han Lai 1,3,4 and Guifeng Li 1,3,4,*Round 1
Reviewer 1 Report
Wang and co-authors present and interesting and well designed study investigating the population structure and life history of anchovy in the Pearl River river system. I have attached my specific comments, which I feel need to be addressed ahead of publication.
More broadly:
1) I feel many significant pairwise differences in population structure (Table 4) were sufficiently highlighted in the results nor discussed in the discussion section.
2) For readers who do not focus on genomics, the discussion and conclusion could do a better job of integrating the results of the multiple genomic approaches and statistical analyses to provide an overall narrative of the relative importance of the drivers of genetic population structure for anchovy in this system.
3) The conclusion needs to be more an overall "take home message" for readers from all backgrounds on the factors driving population genetic divergence and isolation in this system, rather then a point by point repeat of the results and discussion.
Comments for author File: 
Comments.pdf
Author Response
Response to Reviewer 1 Comments
We appreciate the comments and have spent much effort reorganizing the questions. We provide a point-by-point response to your comments, Any new good revising suggestions will be readily accepted.
General comments
Point 1: I feel many significant pairwise differences in population structure (Table 4) were sufficiently highlighted in the results nor discussed in the discussion section.
Response 1: We only discussed this in discussion 4.3. In the revised manuscript, we have paid particular attention to those issues and details, and we hope that we have identified and corrected them.
Point 2: For readers who do not focus on genomics, the discussion and conclusion could do a better job of integrating the results of the multiple genomic approaches and statistical analyses to provide an overall narrative of the relative importance of the drivers of genetic population structure for anchovy in this system.
Response 2: Thanks for reviewer’s valuable suggestions which we give full consideration to, and we increase some related sentences in the discussion.
Point 3:The conclusion needs to be more an overall "take home message" for readers from all backgrounds on the factors driving population genetic divergence and isolation in this system, rather then a point by point repeat of the results and discussion.
Response 3: We have reconstructed our conclusion under your advisable comments.
Specific comments
1.Is distance really the critical metric in a system with so many dams? I.e. should the a-priori expectation be that there is a tight (linear?) correlation with geographic distance and genetic divergence?
Response 1: This is indeed a tough problem. The challenging from the unclear edges between the effects of geographic distance and the effect of habitat fragmentation. We are incapable to provide more clear evidence and can’t determine statistical correlations between them. Despite all this, there were still some clues left. The population of QY and HM shows greater genetic differences while HZ and HM minor differences are based on F-Statistics, although the relationship is opposite in spatial distance. IBD pattern always was present in species with a wide habitat range, which provide comparative references.
2.Measure of length (total, standard, fork?)
Response 2: It’s the standard length. Modification has been made in a corresponding part of the article.
3.It states above there were 16, not 17
please explain what "low-ratio" indicates (i.e the rational for exclusion)
which were excluded?
Response 3: Low-ratio indicates the poor quality of otolith with low Sr: Ca, even we couldn't detect it. We have modified it.
4.Please summarize the statistical approach
Response 4: Statistical analysis was performed in R. Sr/Ca concentration ratios were traditionally determined and represented as Sr/Ca×1000. The Sr/Ca ratios and X-ray intensity maps of Sr content were used to define the habitat phases in the otolith by comparing them to known patterns of various salinity exposure phases in C. nasus [21,22], where a Sr/Ca ratio of ≤3 is considered to be indicative of a freshwater phase and is colored blue (salinity: < 5); a ratio of 3–7 is considered indicative of a brackish water phase (salinity: 5–25) and is colored green-yellow; and a ratio >7 is considered indicative of a fully sea water phase (salinity: >25) and is colored red.
5.This section could have a more specific heading (something referring to the genetic analysis)
Response 5: DNA extraction, library construction and sequencing
6.Please define how geographic distance was calculated.
Response 6: The latitude and longitude of sites were transformed to corresponding coordinate distances along the surface of the earth by “geoXY” function in R. Then, we obtained geographic distance metric of all sites.
7.Results for 16 otoliths are presented here, including life history transects. Above its stated that 14 of 17 were suitable for analysis.
Response 7: Sorry, it was wrong presentation, and we have modified it.
8.I would suggest referring primarily to the color of the Sr maps as the color is "false". Simply refer to the Sr/Ca values.
Response 8: The color map patterns of otolith Sr were employed to provide more objective and intuitive insights regarding fish migratory patterns and environmental salinity history.
9.are the pairwise watershed level comparisons provided?
Response 9: Yeah, it’s wrong, and we have deleted it in the manuscript.
10.what about HM, which is discussed first below (Genetic Diversity) or the other interesting pairwise differences? Please guide the reader through all the significant and (I believe) insightful significant differences.
Response 10: Thank you for your reminder. New descriptions of genetic differentiation have been provided on paper.
11.This is discussion and be explicit o which barrier you are referring to (distance? Dams?)Be explicit what phenomenon is being referred to here.... which populations were different?
Response 11: The sentence the reviewer referred has been removed. The phenomenon is most genetic differentiation within the farthest population (HX).
12.Be explicit what phenomenon is being referred to here.... which populations were different? The reader should not have to go to the table and extract the relevant result on their own.
Response 12: We rephrased the sentence and added additional explanations.
13.I'm making this comment before reading the discussion, but I'm not sure geographic distance is the proper metric because of the role dams may play.
Also, I'll note that how geographic distance was calculated is not laid out in detail.
Response 13: This is indeed a tough problem. The challenging from the unclear edges between the effects of dams and the effect of geographic distance. However, we find some clues left. The population of QY and HM shows greater genetic differences while HZ and HM minor differences are based on F-Statistics, although the relationship is opposite in spatial distance. IBD pattern always was present in species with a wide habitat range, which provide comparative references. The calculation of geographic distance has been provided in methods and we wrote in Response 6.
14.Is this that no distinct population difference were observed between up and downstream populations in each tributary or overall (all upstream vs. all downstream)? Also, if the microsatellite results were similar to the mtDNA results, i.e. the only sig. different population was HX, is this analysis needed?
Response 14: It’s the wrong presentation. No distinct genetic structure was observed between up and downstream populations except HX.
15.I think this is a vast overstatement. The number of otoliths analyzed simply are not sufficient to determine there are two distinct patterns. While clearly populations above dams will be freshwater residents a larger number of otoliths would be needed to ascertain if some (minority) proportion of the remaining populations unitized brackish habitats over their life histories.
Response 15: Thank you for pointing this out. We agree with this comment. We admitted the draw that this is a small dataset in terms of a number of otoliths. Thus, the limited information we got maybe not be sufficient for two distinct patterns and also be easily suspected. Having otoliths analysed for all individuals will be the best way for research in methods. Sadly, it's not economic and efficient. We didn't find any new types in freshwater populations from the current results and our project funding is not enough to support making electron probes for each individual. From the results of electron probe microanalysis of sixteen samples, it occurs in two types. I believe these interesting questions should have a broader sampling scheme and a rigorous design to be eliminated. We rewrote the sentence on paper.
16.I understand the message of this section but I find it hard to parse out how much of the overall results Hd is driven by HM only and the nucleotide diversity results driven by fish from all sites BUT HM. I think this is a critical discussion point which should be made more transparent to the reader.
Response 16: We agree with this and have incorporated your suggestion throughout discussion 4.2.
17."increasing tendency"? I'm not sure how to interpet that.
Also please define "downstream" in general (ie all populations that are downstream from a dam...its mentioned in the introduction I believe but perhaps redefining it in the methods would help) and in each specific case. For example here are you refering to QY as the "down stream populations", or QY and BL or QY, BL and HM?
Response 17: Agree! We have redefined these in the methods.
"Increasing tendency" --- we want to say “the pairwise F-Statistics were increased, indicating the tendency towards genetic differentiation.”
18.While I realize founder effect is discussed below I would like to see geographic distance,establishment of dams and founder effect discussed together in a holistic manner, as they could likely interact to drive diversity .
I realize the authors do conclude this section by recognizing that the two factors may interact, but I would like to read more of the authors thoughts and the relative importance of the two factors in this study.
Response 18: Thanks for the suggestions, and we re-phrased this part.
19.Remind the reader what you are referring to (anadromous vs freshwater resident?)
Response 19: Yes, we have modified it.
20.suggested, again the otolith evidence is weak due to the limited sample sizes
Response 20: The sentence was modified.
- This sounds like a bit of repeat of results..
For readers who are not population geneticists I'd like to see the conclusion structured a little more "holistically", i.e. rather then a list of individual conclusions and overall perspective on what do the results as whole say about population structure and life history.
Response 21: Thanks for reviewer’s valuable suggestions. This part has been re-written to better clarify its content.
Reviewer 2 Report
Wang et report the migration behaviours and genetic diversity of Coilia grayii inhabiting at the Pearl River of China. Generally, it is a pretty good study and great to see combined methods of genetic analyses and otolith microchemistry to study the fish ecology. Some findings are very valuable for fish biology and fisheries management, such as polymorphic migratory behaviours of C. grayii, a possible spawning area in brackish waters and genetic structure among the geographic areas. Also very appreciate a great effort on genetic analyses, data explanation and further discussion. However, there is an unbalance of sample numbers between genetic and otolith analyses that may arise some issues. I have explained my concern below and pointed out some questions.
General comments:
Introduction- This study focuses on anadromous fishes, but the examples listed in the introduction sometimes mixes with other non-anadromous species. I will suggest introducing general migratory behaviours in one paragraph and then focusing anadromous species in the following paragraph. Or giving examples which are all anadromous species. Otherwise, it is confusing when reading through the paragraphs in introduction.
Materials and Methods-After I read through the whole manuscript, I have questions about the experimental design. The HM population has the highest haplotype diversity among populations, but only three otoliths from this population were analysed. Thus the question is: (1) Would HM population also show high diversity of migration behaviours? (2) Is any possibility to examine the migratory strategy of the same haplotype but in different populations by using otoliths? To answer these questions, the number of otoliths analysed in this study seems not sufficient. I was wondering how the authors’ opinions on these questions and their decision of analysing only a few otoliths among populations.
Results-In some cases, the content and cited figures are not matched. Please carefully examine again. I list several issues in the specific comments.
Discussion-I have two questions which are not mentioned and discussed. (1) Is possible that populations BL and QY migrate within the freshwater areas? If yes, a gene flow may occur at the freshwater areas among populations QY, BL and HM. (2) Across the dam barrier (from up- to downstream), we can find an increase of haplotype diversity based on the Hd from the D-loop analyses. It makes sense and especially, we can observe an obvious difference between QY and LX. But between BL and HZ, only a slight difference is found. Any explanation or possibility?
Specific comments
- Line 57. Strong migratory? Strong migratory connectivity, or?
- Line 112 and Line 116. I expect that 6 populations means the individuals captured from the six sites. Would appreciate if defining it at the beginning.
- Figure 3 caption: how to decide that the Sr/Ca value below 3 is freshwater period? Any experimental validation or literature support? Please explain.
- Line 272-276. “This process produced 40 segregating sites and 37 haplotypes…… Three dominant haplotypes….. while seven haplotypes……. remaining nine unique haplotypes,…”. This paragraph really confused me that among 37 haplotypes and only 19 (3+7+9) were introduced. Am I misunderstanding something?
- Line 292-294. In the content, the result of D-loop and cyt b were allocated with Figure 4a and 4b. However, the figure shows the result of cyt b in 4a and D-loop in 4b.
- Line 339. Should be Figure 5a?
- Line 340. Should be Figure 5c?
- Line 341. Should be Figure 5b?
- Line 533-535. How to infer that the populations is sedentary populations? According to Sr/Ca, we may know they are freshwater residents. However, they may migrate within the freshwater areas that cannot be observed from the otolith Sr/Ca.
Author Response
Response to Reviewer 2 Comments
Thanks for your valuable comments and we have spent much effort reorganizing the questions. We provide a point-by-point response to your comments. Any new good revising suggestions will be readily accepted.
General comments:
Point 1: Introduction- This study focuses on anadromous fishes, but the examples listed in the introduction sometimes mixes with other non-anadromous species. I will suggest introducing general migratory behaviours in one paragraph and then focusing anadromous species in the following paragraph. Or giving examples which are all anadromous species. Otherwise, it is confusing when reading through the paragraphs in introduction.

Response 1: As suggested by the reviewer, we modified the manuscript and gave examples that are all anadromous species in the first paragraph.
Point 2: Materials and Methods-After I read through the whole manuscript, I have questions about the experimental design. The HM population has the highest haplotype diversity among populations, but only three otoliths from this population were analysed. Thus the question is: (1) Would HM population also show a high diversity of migration behaviours? (2) Is any possibility to examine the migratory strategy of the same haplotype but in different populations by using otoliths? To answer these questions, the number of otoliths analysed in this study seems not sufficient. I was wondering how the authors’ opinions on these questions and their decision of analysing only a few otoliths among populations.
Response 2: We appreciate the comments and have spent much effort on our research limitation. As you advised, having otoliths analysed for all individuals will be the best way for research in methods. Sadly, it's not economic and efficient. We didn't find any new types in freshwater populations from the current results and our project funding is not enough to support making electron probes for each individual. From the results of electron probe microanalysis in HM population, it may occur in two types, and the migratory strategy of the same haplotype but in different populations may be different, whose differences will be possibly observed on phenotype or genomic. I believe these interesting questions should have a broader sampling scheme and a rigorous design to be eliminated.
Point 3: Results-In some cases, the content and cited figures are not matched. Please carefully examine again. I list several issues in the specific comments.
Response 3: done as suggested
Point 4: Discussion-I has two questions which are not mentioned and discussed. (1) Is possible that populations BL and QY migrate within the freshwater areas? If yes, a gene flow may occur at the freshwater areas among populations QY, BL and HM. (2) Across the dam barrier (from up- to downstream), we can find an increase of haplotype diversity based on the Hd from the D-loop analyses. It makes sense and especially, we can observe an obvious difference between QY and LX. But between BL and HZ, only a slight difference is found. Any explanation or possibility?
Response 4: Thanks for the reviewer's valuable questions. (1) Yes, it's possible to occur gene flow, which has been proved in general F-Statistics. Noticeably, we didn't obtain consistency among the three datasets. (2) There are two reasons for a slight difference between BL and HZ: Firstly, their separated time is shorter than that in QY and LX. Secondly, in terms of spatial scale, their gene flow could be easier in the short distance between these two sites. Modification has been made in a corresponding part of the article.
Specific comments
Line 57. Strong migratory? Strong migratory connectivity, or?
Response 1: Thank you very much for your suggestions. done as suggested
Line 112 and Line 116. I expect that 6 populations means the individuals captured from the six sites. Would appreciate if defining it at the beginning.
Response 2: Thank you very much for your suggestions. Modification has been made in a corresponding part of the article.
Figure 3 caption: how to decide that the Sr/Ca value below 3 is freshwater period? Any experimental validation or literature support? Please explain.
Response 3: Thank you very much for your suggestions. The Sr/Ca ratios and X-ray intensity maps of Sr content were used to define the habitat phases in the otolith by comparing them to known patterns of various salinity exposure phases in C. nasus(Yang et al 2006, 2011), where a Sr/Ca ratio of ≤3 is considered to be indicative of a freshwater phase and is colored blue (salinity: < 5); a ratio of 3–7 is considered indicative of a brackish water phase (salinity: 5–25) and is colored green-yellow; and a ratio >7 is considered indicative of a fully sea water phase (salinity: >25) and is colored red. done as suggested.
Line 272-276. “This process produced 40 segregating sites and 37 haplotypes…… Three dominant haplotypes….. while seven haplotypes……. remaining nine unique haplotypes,…”. This paragraph really confused me that among 37 haplotypes and only 19 (3+7+9) were introduced. Am I misunderstanding something?
Response 4: NO, it’s our mistake. The number of unique haplotypes is 27. Modification has been made in a corresponding part of the article.
Line 292-294. In the content, the result of D-loop and cyt b were allocated with Figure 4a and 4b. However, the figure shows the result of cyt b in 4a and D-loop in 4b.
Line 339. Should be Figure 5a?
Line 340. Should be Figure 5c?
Line 341. Should be Figure 5b?
Response 5: done as suggested
Line 533-535. How to infer that the populations are sedentary populations? According to Sr/Ca, we may know they are freshwater residents. However, they may migrate within the freshwater areas that cannot be observed from the otolith Sr/Ca.
Response 6: In the revised manuscript, we have paid particular attention to those issues and details, and we hope that we have identified and corrected them.
Reviewer 3 Report
Manuscript titled " Otolith microchemistry and demographic history provide new insight into the migratory behavior and heterogeneous genetic divergence of Coilia grayii in the Pearl River" aims a study on the diversity of the populations of this fish species by mitochondrial and microsatellites markers coupled with otolith analysis related to each constrained habitat.
Study is well designed. Results are clearly presented and sufficiently discussed. Its content in general justifies the length. Language is clear and understandable. Figures are of good quality. Approach, methodologies and analysis used in this work are satisfactory.
My only concerns are related to the software authors have used for genetic differentiation.
In fact, if Genealex and Genpop are good for Amova or general F-Statistics, I think software as Arlequin or Fstat are more appropriate for pairwise FST due to their higher power in permutations.
I took the liberty of doing a quick analysis with your microsatellite data present in the supplementary materials using one of these softwares (Arlequin).
|
FST |
HM |
HX |
LX |
QY |
HZ |
BL |
|
|
HM |
* |
||||||
|
HX |
0.07205 |
* |
|||||
|
LX |
0.0053 NS |
0.03645 |
* |
||||
|
QY |
0.01064 |
0.03903 |
0.01881 |
* |
|||
|
HZ |
0.021 |
0.05835 |
0.03064 |
0.02259 |
* |
||
|
BL |
0.00434 NS |
0.0483 |
0.00592 NS |
0.01661 |
0.00718 NS |
* |
|
|
p values |
|||||||
|
HM |
HX |
LX |
QY |
HZ |
BL |
||
|
HM |
* |
||||||
|
HX |
0.0000 |
* |
|||||
|
LX |
0.20721 |
0.0000 |
* |
||||
|
QY |
0.03604 |
0.0000 |
0.02703 |
* |
|||
|
HZ |
0.00901 |
0.0000 |
0.0000 |
0.0000 |
* |
||
|
BL |
0.27928 |
0.0000 |
0.14414 |
0.00901 |
0.14414 |
* |
|
As you can see from the table, the results show that the differences between some populations are now significant (which they weren't before).
In general, the population of HX shows greater differences (always significant in pairwise) while BL and LX minor differences (values not significant, except with QY which however showed the lowest FST values).
I do not believe that these new findings change the general lines of the discussion and conclusions of the manuscript. However, I believe it is necessary to specify and correctly use the software.
Some shortenings are also suggested below.
L303 not “low” but “lower……. than other populations”
In table 3, please delete or add the line above the fixation indices so that all squares are the same; and put in bold Source of Cyt-b and Source of microsatellites
L339-341 Please, change Figure 5a….5c and ….5b
L353-355 please add more details about a), b) and c) figure 6 images
L377-378 In legend of Figure 7, please add which molecular marker was used
L411 please, add (HM) after Humen population for a faster understanding of the reference
L438 Please, change “Haploid” with “Haplotype” diversity
Author Response
Response to Reviewer 3 Comments
We appreciate the comments and have spent much effort reorganizing the questions. We provide a point-by-point response to your comments, Any new good revising suggestions will be readily accepted.
General comments
Point 1: My only concerns are related to the software authors have used for genetic differentiation. In fact, if Genealex and Genpop are good for Amova or general F-statistics, I think software as Arlequin or Fstat are more appropriate for pairwise FST due to their higher power in permutations.
Response 1: Thank you very much for your suggestions. We have reconstructed our results of pairwise FST under your advisable comments. Modification has been made in a corresponding part of the article and tables.
Specific comments
1.L303 not “low” but “lower……. than other populations”
Response 1: Modification has been made in a corresponding part of the article.
2. In table 3, please delete or add the line above the fixation indices so that all squares are the same; and put in bold Source of Cyt-b and Source of microsatellites
Response 2: Modification has been made in a corresponding part of the article.
3. L339-341 Please, change Figure 5a….5c and ….5b
Response 3: done as suggested
4. L353-355 please add more details about a), b) and c) figure 6 images
Response 4: Modification has been made in a corresponding part of the article.
5. L377-378 In legend of Figure 7, please add which molecular marker was used
Response 5: Figure 7. Allele frequency distribution graph of the six populations based on microsatellite dataset.
6. L411 please, add (HM) after Humen population for a faster understanding of the reference
Response 6: done as suggested
7. L438 Please, change “Haploid” with “Haplotype” diversity
Response 7: done as suggested
Reviewer 4 Report
Wang et al use a multiple techniques to study the demographic history of Coilia grayii in the Pearl River and which factors are currently affecting population connectivity. The combination of genetic data with otolith concentrations of Sr and Ca is an interesting approach providing a more comprehensive view of the population biology of the target system. The methods are well implemented and the result are overall well interpreted. Nevertheless there are some points that need improvement.
My main criticism comes from the fact that the authors only used five microsatellite markers which can be partly the reason why low genetic structure were found. Moreover, these were markers developed for other species and since only the ones that presented a similar amplicon length of the original species were piked it can be that these are the most conserve and therefore less informative. I think these points should be discussed.
Specific comments:
Introduction
Line 59: Gene flow disorder? What is it?
Line 87: Barriers like dams will only contribute to founder effect if some individuals can pass through them. This idea should be made clearer.
Line 93 to 95: I would say that mitochondrial and microsatellite markers are some of the most robust markers in describing dispersal patterns but not necessarily the most effective. For example, genomic approaches allowing the genotyping of thousands of SNPs have shown to work quite well. For that reason, I would recommend changing the statement from “the most robust” to “some of the most robust”.
Line 95 to 98: This is also true for microsatellites. Additionally, it is important to mention that they can serve as a good complement to mitochondrial markers since they have biparental inheritance.
Line 109: The authors use microsatellites fragment length not the sequence.
Methods
Line 167: Since the markers were developed for another species it is possible that the resulting amplicon to have a length quite different from the original species. How different was it?
Line 216: “microsatellite genotypes” instead of sequences.
Results
Line 287: This sentence is wrong. Network methods, as shown here, are based on genetic distances and therefore do not reflect phylogenetic relationships. Phylogenies are based of the distribution of share of derived character states and not genetic similarity.
Line 303: Was a significance test performed? If not please remove the statement that the Ho was significantly lower for these populations. Instead, just say it was lower or much lower.
Line 305: Inbreeding depression refers to the lost of fitness due to inbreeding. Based on the data collected it is impossible to conclude that. All the authors can say is if the populations are inbreed or not.
Discussion
Section 4.2: Why only the mitochondrial genetic diversity are consider? Why not include the information from microsatellite markers?
Line 459: HX is not only just the farthest population. Based on the map, its geographical distance to the closest population seams to be much larger than the largest distance among the remaining populations. Mentioning that would strenghten the IBD hypothesis, especially if the geographical distance values are mentioned.
Line 461: How does the IBD agrees with life history of the species?
Paragraph starting in Line 462: The authors develop the idea that dams contribute to the lower genetic diversity of the upstream populations. However this contradicts the idea of IBD, since if dams were working as a large barrier to geneflow a more abrut genetic structure pattern would have been found. So how these two ideas can be combined?
Figures and Tables:
Figure 1: Additionally, in the legend, upstream corresponds to populations upstream to the dams, correct? If that is the case, please mention that.
Table 1: Please indicate what upstream mean.
All figures are of poor resolution. I recommend remaking them with higher resolution.
Author Response
Response to Reviewer 4 Comments
Thanks for your comments and we have spent much effort reorganizing the questions. We provide a point-by-point response to your comments. Any new good revising suggestions will be readily accepted.
General comments
General comments
Point 1: My main criticism comes from the fact that the authors only used five microsatellite markers which can be partly the reason why low genetic structures were found. Moreover, these were markers developed for other species and since only the ones that presented a similar amplicon length of the original species were picked it can be that these are the most conserved and therefore less informative. I think these points should be discussed.
Response 1: Thank you for the reviewer’s valuable suggestions. This is indeed a tough problem. The challenge was from exploring new microsatellite markers. (1) We admitted the draw that this is a small dataset in terms of microsatellite markers. Thus, the limited information we got maybe not be sufficient for genetic structure and also be easily suspected. However, based on cyt b dataset, it’s also low genetic structure. (2) Regarding "markers developed for other species was most conserve and therefore less informative", we agree partly on the comments, because they are congeneric and have close phylogenetic relationships, though we couldn't promise they are efficient and unconserved.
Specific comments
Introduction
- Line 59: Gene flow disorder? What is it?
Response 1: Apologies for the confused statement, “geographic barrier” may be suitable.
- Line 87: Barriers like dams will only contribute to founder effect if some individuals can pass through them. This idea should be made clearer
Response 2: Agree, we rephrased the sentence and added additional explanations.
- Line 93 to 95: I would say that mitochondrial and microsatellite markers are some of the most robust markers in describing dispersal patterns but not necessarily the most effective. For example, genomic approaches allowing the genotyping of thousands of SNPs have shown to work quite well. For that reason, I would recommend changing the statement from “the most robust” to “some of the most robust”.
Response 3: done as suggested
- Line 95 to 98: This is also true for microsatellites. Additionally, it is important to mention that they can serve as a good complement to mitochondrial markers since they have biparental inheritance.
Response 4: done as suggested
- Line 109: The authors use microsatellites fragment length not the sequence.
Response 5: done as suggested
Methods
- Line 167: Since the markers were developed for another species it is possible that the resulting amplicon to have a length quite different from the original species. How different was it?
Response 6: Yes, for instance, the microsatellites fragment length we obtained may be partial-, mixed, or low-template.
- Line 216: “microsatellite genotypes” instead of sequences.
Response 7: done as suggested
Results
- Line 287: This sentence is wrong. Network methods, as shown here, are based on genetic distances and therefore do not reflect phylogenetic relationships. Phylogenies are based of the distribution of share of derived character states and not genetic similarity.
Response 8: All these concepts are well known, and we have removed the wrong presentation.
- Line 303: Was a significance test performed? If not please remove the statement that the Ho was significantly lower for these populations. Instead, just say it was lower or much lower.
Response 9: No, thanks for suggestions, and we re-phrased this part
- Line 305: Inbreeding depression refers to the lost of fitness due to inbreeding. Based on the data collected it is impossible to conclude that. All the authors can say is if the populations are inbreed or not.
Response 10: Agree! The related statement has been rewritten.
Discussion
- Section 4.2: Why only the mitochondrial genetic diversity are consider? Why not include the information from microsatellite markers?
Response 11:Yes, we should have paid equal attention to the information from microsatellite markers, and we re-phrased this part.
- Line 459: HX is not only just the farthest population. Based on the map, its geographical distance to the closest population seams to be much larger than the largest distance among the remaining populations. Mentioning that would strenghten the IBD hypothesis, especially if the geographical distance values are mentioned.
Response 12: It’s valuable suggestions. The geographical distance between HX and HM is approximate a thousand kilometres, which contributes to IBD hypothesis.
- Line 461: How does the IBD agrees with life history of the species?
Response 13: The bottleneck analysis and allele frequency distribution demonstrated that HX the populations have recently undergone colonization and founder effect. Bayesian skyline plot suggested the species has experienced significant expansion. These phenomena may arise from long-distance dispersal populations, such as HX.
- Paragraph starting in Line 462: The authors develop the idea that dams contribute to the lower genetic diversity of the upstream populations. However this contradicts the idea of IBD, since if dams were working as a large barrier to geneflow a more abrut genetic structure pattern would have been found. So how these two ideas can be combined?
Response 14: We respectfully disagree, and we think the hint to problems related to dams’ contribution to the lower genetic diversity is very important, though a more abrupt genetic structure pattern was not presented in our research. We have introduced an explicit mention of the fact that we cannot distinguish the interaction between the two components and the presence of barriers may have a considerable impact on IBD patterns, which need a broader sampling scheme to test the validity of these hypotheses in the future.
Figures and Tables
- Figure 1: Additionally, in the legend, upstream corresponds to populations upstream to the dams, correct? If that is the case, please mention that.
Table 1: Please indicate what upstream mean.
All figures are of poor resolution. I recommend remaking them with a higher resolution.
Response 15: Yeah, the upstream corresponds to populations upstream to the dams, and we now explicitly stated throughout the paper.
Round 2
Reviewer 4 Report
The authors addressed all my comments and I think the manuscript improved compared to the previous version. My only suggestion left is to mention the limitation of the low number of markers in the discussion and point out that the lack of resolution found may be a consequence of that.
Author Response
Thanks for your suggestions. The sentences (Another important issue must be pointed out: By establishing a set of microsatellite markers to use in each analysis for the given species, the number of the markers to employ should be sufficient and highly discriminating. Considering the limitation of the low number of microsatellite markers in this research, the lack of resolution found may be a consequence of that.) were added to the ending of discussion 4.3.
