Length–Weight Relationships and Growth Parameters of Common and Leafy Seadragons (Syngnathidae) from a Public Aquarium

Round 1

Reviewer 1 Report

See the attached file

Comments for author File: Comments.pdf

Author Response

Reviewer 1 commented/criticized only our choice of the von Bertalanffy growth function (VBGF) to interpret the age at length data pairs available for two species of seadragons, and proposes that we instead investigate an array of alternative models. 

This suggestion is not appropriate for the paper we submitted: we wanted to be able to compare the growth of seadragons with that of other fishes, which we can't do if we do not use the VBGF.

The reason for this is that the VBGF is by far the most commonly used growth model used by ichthyologists and fisheries scientist all over the world, e.g., as documented in FishBase (www.fishbase.org). 

Also, the VBGF has parameters that can be easily interpreted in biological terms (see Pauly, D. 2021. The Gill-Oxygen Limitation Theory (GOLT) and its critics. Science Advances, 7(2), https://doi.org/10.1126/sciadv.abc6050), which is not the case for most of the alternative models, which are empirical, and whose 'better' fit is thus purely accidental when it is not due to these models having more parameters.   

However, to accommodate the thought behind the reviewer's request, we have added to the revised manuscript a phrase indicating why we chose the VBGF, and pointed out that our Tables 1 and 2, which include all our length-at-age data pairs, could be fitted by using different growth model by anyone so inclined. 

Reviewer 2 Report

Length-weight relationships and growth parameters of common and leafy seadragons (Syngnathidae) from a public aquarium by Pauly et al. is a rather straight-forward study on the anatomy of sea dragons, their growth parameters in captivity, and some implications relative to the fishes physiology and metabolism. I find the study well focused, adequately presented and with a good data analysis and discussion of the results. Not being an expert in the anatomy and morphophysiology of sea dragons, it is difficult for me to judge the relevance of this study and its general interest. However, being these species poorly known and difficult to study in the wild, in my opinion this study makes a worthy contribution generating a baseline model on the growth of these species compared to other anatomically "conventional" teleosts. I find no obvious errors except a few typos which should be picked at later editorial stages.    

Author Response

We are very pleased with this positive report, which only note a few typos as items to be fixed.  We have endeavored to fix all typos in the revised manuscript.

Reviewer 3 Report

This paper presents morphometric and growth data for 2 species of seadragon kepy at a US aquarium.  While the growth data from known-age individuals is useful, the study is to specific to one aquarium.  These are very basic data, not really testing any hypothesis (I don’t really consider comparing growth of weedy and leafy seadraons particularly useful, or with other species as in Table 6 ) so I’m not sure that the paper contributes much to the literature a d understating on these animals.  Also my understanding is that seadragons in overseas aquaria are all sourced from the Malborne area, so the generality of results obtained here is reduced.

A few specific comments to consider:

Figure 1- unclear why ruby seadragon shown, not part of the paper. 

Tables- in general contain raw data which shod not appear in min ms.

Tale 1 uses scalene length which appears hugely prone to errors and depends how stretched out the animal is. Surely the total length can be easily measured by imaging and software such as Image-J

Line 77- weighing thin appendages that have bene fixed in formalin and stored for tears is problematic, I don’t feel its acceptable

Line 110:  “The resulting mean ratio of ScL to CL is 0.93 for common and 0.82 for leafy 113 seadragon.” I don’t see this calculation as biologically useful given my cone on the arbitrariness of the ”scalene” length /I note above

Fig 5b- surely you can obta a better image here this is very washed out/discoloured

Line 179:  the first paragraph of the Discussion should highlight key outcomes of the research, instead is notes “there has been a tendency to overfeed seadragons and other syngnathid in captivity which causes them to develop fatty liver and excess visceral fat” a minor factor.

Line 208- you use common and weedy throughout, please standardise

Line 202-216- I’m not convinced that observations of few capillaries in seadragon appendages constitutes evidence of low oxygen usage in these tissues,

Line 217 I saw very little evidence for or against the Gill-Oxygen Limitation Theory

Author Response

Reviewer 2 feels that our presentation can be improved in terms of its style, and we have endeavored to do this in the revised version of our manuscript. Notably, we use the plural ‘seadragons’ when dealing with 2 or 3 species, and the singular when dealing with one, even if several individuals are concerned. Please correct our file if this doesn’t correspond to the style of ‘Fishes’.

In the following, the reviewer’s  comments (C) are followed by our response (R).

C1: This paper presents morphometric and growth data for 2 species of seadragon kept at a US aquarium.  While the growth data from known-age individuals is useful, the study is too specific to one aquarium. 

R1: We agree that most of our results are specific to one aquarium, but do we not think that this invalidate any of our results. Indeed, most growth and related studies, whether based on field data or, as in our case, on aquarium data, apply to one, or a few sites. Moreover, we do compare our results with those of other authors who worked with other field or aquarium data.

C2: These are very basic data, not really testing any hypothesis (I don’t really consider comparing growth of weedy and leafy seadragons particularly useful, or with other species as in Table 6 ) so I’m not sure that the paper contributes much to the literature and understating on these animals. 

R2: Until known, the growth of seadragons could not be compared with that of other fishes, because length-weight relationships we not available for any species. This was particularly problematic for leafy seadragons, whose appendages seemed to complicate things. The reviewer is entitled to an opinion that this is all banal, and I agree in the sense that this is not rocket science. However, our results will certainly be of interest to researchers who study biological traits across broad fish or animal taxa, to the IUCN Red List group covering syngnathids, and to others groups.

C3: Also, my understanding is that seadragons in overseas aquaria are all sourced from the Malborne area, so the generality of results obtained here is reduced.

R3: See R1

C4: Figure 1- unclear why ruby seadragon shown, not part of the paper.

R4: Ruby seadragon has only be described a few years ago, and growth data on this species are not yet available. 

C5: Tables- in general contain raw data which shod not appear in min ms.

R5: If the editor of ‘Fishes’ asks us to put one or more of our tables in the Supplementary Materials, we will comply. However, we would prefer not to, because these tables, which are all very compact, neatly document the step-by-step approach that we have taken in describing our work.  

 C6: Tale 1 uses scalene length which appears hugely prone to errors and depends how stretched out the animal is. Surely the total length can be easily measured by imaging and software such as Image-J

R6: All but one of the co-authors of this article have handled fresh and/or preserved seadragon, and can attest that attempts to ‘straighten’ seadragons are problematic. Obviously, tracing the main axis of their body along its curvature is possible using imaging software. But considering such measurements as the only one that should be used would imply that we should ignore the available age-length data pairs, which have been measured using scalene lengths. Also, scalene length is usable in the field, with live specimens.

C7L Line 77- weighing thin appendages that have bene fixed in formalin and stored for tears is problematic, I don’t feel it’s acceptable.

R7: Two of the authors of this article collaborated on an earlier paper which we cite (Hay, A. W. Xian, N, Bailly, C. Liang and D. Pauly. 2020. The why and how of determining length-weight relationships of fish from preserved museum specimens. Journal of Applied Ichthyology, doi 10.1111/jai.14014). This paper included a review of much of the literature on the effects, in time, of preservation by alcohol and formalin on the weight of preserved fish specimens.

The result is that weight changes are not predictable with regard to their sign (weight loss or gain), but are not very large.

More importantly, however, we did not use, in our article, the absolute weight that were obtained, but their ratios. Thus, unless the reviewer wants posit the improbable claim that the weight of the appendages would change in a manner opposite that of the rest of the body, preservation was likely not a source of bias.  

C8: Line 110:  “The resulting mean ratio of ScL to CL is 0.93 for common and 0.82 for leafy 113 seadragon.” I don’t see this calculation as biologically useful given my cone on the arbitrariness of the ”scalene” length /I note above

R8: The biological usefulness of the length ratios we propose is that they allow (admittingly approximate) comparisons with the growth parameters estimates of other workers. 

C9: Line 179:  the first paragraph of the Discussion should highlight key outcomes of the research, instead it notes “there has been a tendency to overfeed seadragons and other syngnathid in captivity which causes them to develop fatty liver and excess visceral fat” a minor factor.

R10:  Many (most?) biologists believe that fish growth is a mainly function of the food they get, and thus we presume that they expect, in a paper on seadragon growth, a statement about the food they got. The statement we made was the only useful information we could convey on food and feeding. It conveys that the low growth performance we reported on was not due to the fish being underfed. We believe that this is highly pertinent.

C11: Line 208- you use common and weedy throughout, please standardise

R11: Agreed; weedy is now used only once, to say that it their other name (it also shows up in the references, but this can’t be avoided).

C12: Line 202-216- I’m not convinced that observations of few capillaries in seadragon appendages constitutes evidence of low oxygen usage in these tissues.

R12: For tissues to be consuming lots of oxygen (brain, kidney, liver), they must be irrigated by lots of blood flowing through arteries and smaller vessels down to capillaries. Tissues that are less metabolically less active (bones, cartilage, nails) have fewer or no capillaries. What reason is there to believe that the appendages would be metabolically active in the absence of an abundant blood supply?

C13: Line 217 I saw very little evidence for or against the Gill-Oxygen Limitation Theory

R14: We wrote that our results (asymptotic growth, appendages not requiring a high O₂ supply, a low growth performance linked to gills not suitable for a high O₂ throughput, and an L_max/L_m  ratio would value is predicted by the GOLT) corroborate the GOLT. Indeed, if our results were otherwise, they would be counter-evidence to the GOLT. However, to accommodate this comment, we have modified our final statement, which now say that our results are compatible with the GOLT.

In C2, the reviewer stated that our article is “not really testing any hypothesis”. It does.

Round 2

Reviewer 1 Report

I will recommend to the editor the acceptance of the actual version to not be in a conflict with different ideas about the practical use of the VBGF over other equations. I grateful that the authors notice that once their results are published the table 1 and 2 could be used to extend the growth analysis with multi model approach by anyone interested. So, in the future I will do myself and will provided a more ample discussion about the advantage of Multi model approach and extensive explanation about the empiricism of any other growth equation

Author Response

We are very pleased that the reviewer accepted our response to their request to compare different growth curves, which was to add a sentence that the data in our tables 1 and 2 can be used for this purpose by any interested reader. 

Reviewer 3 Report

R3 Comments:  I feel that the authors have not allayed my major concerns (“RESPONSE of R3, below) that the paper is poorly written and structured, uses flawed metrics and is too specific to be of great interest to most readers.

C1: This paper presents morphometric and growth data for 2 species of seadragon kept at a US aquarium. While the growthdata from known-age individuals is useful, the study is too specific to one aquarium.

R1: We agree that most of our results are specific to one aquarium, but do we not think that this invalidate any of our results.Indeed, most growth and related studies, whether based on field data or, as in our case, on aquarium data, apply to one, or a fewsites. Moreover, we do compare our results with those of other authors who worked with other field or aquarium data.

RESPONSE of R3

Sorry, these data are way too specific to be of any significant value to a wider audience- if the authors related growth to another variable such as feed, day length, water temperature etc the data may be of greater value

C2: These are very basic data, not really testing any hypothesis (I don’t really consider comparing growth of weedy and leafyseadragons particularly useful, or with other species as in Table 6 ) so I’m not sure that the paper contributes much to theliterature and understating on these animals.

R2: Until known, the growth of seadragons could not be compared with that of other fishes, because length-weight relationshipswe not available for any species. This was particularly problematic for leafy seadragons, whose appendages seemed tocomplicate things. The reviewer is entitled to an opinion that this is all banal, and I agree in the sense that this is not rocketscience. However, our results will certainly be of interest to researchers who study biological traits across broad fish or animaltaxa, to the IUCN Red List group covering syngnathids, and to others groups.

RESPONSE of R3

I agree with R2 that the data will be of little general interest, including to IUCN Red list, far to specific, only from aquarium specimens etc.

C3: Also, my understanding is that seadragons in overseas aquaria are all sourced from the Malborne area, so the generality ofresults obtained here is reduced.

R3: See R1

RESPONSE of R3

As above I agree with R2 that the data will be of little general interest, including to IUCN Red list

C4: Figure 1- unclear why ruby seadragon shown, not part of the paper.

R4: Ruby seadragon has only be described a few years ago, and growth data on this species are not yet available.

RESPONSE of R3

OK but I don’t think needed, usually images of fish not late dot the paper are not shown.

C5: Tables- in general contain raw data which shod not appear in min ms.

R5: If the editor of ‘Fishes’ asks us to put one or more of our tables in the Supplementary Materials, we will comply. However, we would prefer not to, because these tables, which are all very compact, neatly document the step-by-step approach that we havetaken in describing our work.

RESPONSE of R3

Disagree, the Tables are laborious are not compact but drawn out and should be in a data repository eg Dryas not in the main body of a paper

C6: Tale 1 uses scalene length which appears hugely prone to errors and depends how stretched out the animal is. Surely the total length can be easily measured by imaging and software such as Image-J

R6: All but one of the co-authors of this article have handled fresh and/or preserved seadragon, and can attest that attempts to‘straighten’ seadragons are problematic. Obviously, tracing the main axis of their body along its curvature is possible usingimaging software. But considering such measurements as the only one that should be used would imply that we should ignore theavailable age-length data pairs, which have been measured using scalene lengths. Also, scalene length is usable in the field, withlive specimens.

RESPONSE of R3:

I feel its irrelevant whether or not one can straighten the seadragons, surely they don’t all “bend over” the same amount and surely the true length must be the key variable!  The Scalene metric is flawed.

C7L Line 77- weighing thin appendages that have been fixed in formalin and stored for years is problematic, I don’t feel it’sacceptable.

R7: Two of the authors of this article collaborated on an earlier paper which we cite (Hay, A. W. Xian, N, Bailly, C. Liang and D.Pauly. 2020. The why and how of determining length-weight relationships of fish from preserved museum specimens. Journal ofApplied Ichthyology, doi 10.1111/jai.14014). This paper included a review of much of the literature on the effects, in time, ofpreservation by alcohol and formalin on the weight of preserved fish specimens.

The result is that weight changes are not predictable with regard to their sign (weight loss or gain), but are not very large.

More importantly, however, we did not use, in our article, the absolute weight that were obtained, but their ratios. Thus, unless thereviewer wants posit the improbable claim that the weight of the appendages would change in a manner opposite that of the restof the body, preservation was likely not a source of bias.

RESPONSE of R3

My comment refers to efforts in weight of VERY THIN fins etc when preserved, I’d need to see proof that wet weight s were related to preserved weights for the fins, you have not addressed this

C8: Line 110: “The resulting mean ratio of ScL to CL is 0.93 for common and 0.82 for leafy 113 seadragon.” I don’t see thiscalculation as biologically useful given my cone on the arbitrariness of the ”scalene” length /I note above

R8: The biological usefulness of the length ratios we propose is that they allow (admittingly approximate) comparisons with thegrowth parameters estimates of other workers.

RESPONSE of R3

OK but not so useful I feel given the way length was measured, see above

C9: Line 179: the first paragraph of the Discussion should highlight key outcomes of the research, instead it notes “there has beena tendency to overfeed seadragons and other syngnathid in

captivity which causes them to develop fatty liver and excess visceral fat” a minor factor.

R10: Many (most?) biologists believe that fish growth is a mainly function of the food they get, and thus we presume that theyexpect, in a paper on seadragon growth, a statement about the food they got. The statement we made was the only usefulinformation we could convey on food and feeding. It conveys that the low growth performance we reported on was not due to thefish being underfed. We believe that this is highly pertinent.

RESPONSE of R3

Sorry I persist to say that this comment is veery minor, only anecdotal, and has no place leading off a Discussion.

C11: Line 208- you use common and weedy throughout, please standardise

R11: Agreed; weedy is now used only once, to say that it their other name (it also shows up in the references, but this can’t be avoided).

RESPONSE of R3

Thank you

C12: Line 202-216- I’m not convinced that observations of few capillaries in seadragon appendages constitutes evidence of lowoxygen usage in these tissues.

R12: For tissues to be consuming lots of oxygen (brain, kidney, liver), they must be irrigated by lots of blood flowing througharteries and smaller vessels down to capillaries. Tissues that are less metabolically less active (bones, cartilage, nails) havefewer or no capillaries. What reason is there to believe that the appendages would be metabolically active in the absence of anabundant blood supply?

RESPONSE of R3

Is it possible that capillaries in appendages are slightly higher diameter and have faster flow rates, or that since they are moving more, more blood is pumped?

C13: Line 217 I saw very little evidence for or against the Gill-Oxygen Limitation Theory

R14: We wrote that our results (asymptotic growth, appendages not requiring a high O2  supply, a low growth performance linkedto gills not suitable for a high O2  throughput, and an Lmax/Lm     ratio would value is predicted by the GOLT) corroborate theGOLT. Indeed, if our results were otherwise, they would be counter-evidence to the GOLT. However, to accommodate thiscomment, we have modified our final statement, which now say that our results are compatible with the GOLT.

In C2, the reviewer stated that our article is “not really testing any hypothesis”. It does.

RESPONSE of R3

Thanks for modifying but I don’t feel that this is a strong test of GOL at all.

Author Response

Thank you for your comments.