Anachronic Fruit Traits and Natural History Suggest Extinct Megafauna Herbivores as the Dispersers of an Endangered Tree

Round 1

Reviewer 1 Report

This is an interesting and well written paper providing several lines of evidence supporting that fruits from the tree Gomortega keule are anachronic. The observations and experiments conducted by authors are well done. I have some specific concerns that need to be addressed before publication. This is certainly a fascinating study system and I encourage authors to keep working on it.

Specific comments:

Results

L122-125. Please, provide in the text the mean (and error) percentages of germination for each treatment. Currently, we readers don’t know whether most seeds in your experiments germinated or whether most seeds failed to do so. Such information is important for several reasons. For example, if the percentage of seed germination from intact pericarps is very high there would be little possibilities for seed dispersers to increase it. In that case, the adaptive advantages of seed dispersal could only relate to seed movement.

L131-133: What you show in this figure are not the probabilities of germination but the hazard ratios. Please, reword this sentence accordingly.

It is unusual that intact fruits show the highest germination. Please, state the percentages of germination for each treatment in the main text.

Discussion

L136. A brief paragraph (before the first subsection of the Discussion) stating the study’s main results and/or novel perspectives would help.

L182. “…and whether they are effective dispersers or not.”

I agree there is a gap of knowledge and that further studies are needed.  However, the seed dispersal effectiveness is a quantitative variable and not a dichotomic one. Thus, I would better state that little is known about their relative dispersal effectiveness.

L187-188. It is a pity that P. puda was not included in your feeding trials.

L204. What expectation? Please, be as specific as possible.

L206. “their effect is not entirely positive”.  This is not surprising at all. Species interactions (including plant-disperser interactions; Fedriani and Delibes 2011) are typically characterized by benefits and costs.  Quantifying both the costs and the benefits of interactions is needed to get a comprehensive understanding of their nature.

L213. To avoid a dichotomic statement, I suggest to replace “legitimate” with “relatively effective”.

L274. “the loss or strong reduction of…”

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The manuscript entitled „Anachronic fruit traits and natural history suggest extinct megafauna herbivores as the dispersers of an endangered tree” is an interesting piece of communication. Authors described fruit traits of endemic Gomortega keule, biology of the tree and report observations and experimental findings on its potential present-day seed dispersers. Based on these facts (sometimes also anecdotic information) Authors argue that this species is another example of plant with anachronic megafauna dispersal syndrome. Certainly arguments used by Authors and analyses carried out would be not satisfactory in case of the full research article but in case of the Communication paper it may be acceptable if following doubts and problems will be solved:

General comments:

• my main concern is about the conflict between germination results (much lower in case of stones processed by animals) and the main line of discussion about extinct megafauna, which interacted in the process of evolution of the studied tree species and took part in its dispersal in past. If species is adapted to endozoochory, then passing through the intestines should be beneficial or at least neutral to seed. What would be the reason to evolve into direction which causes decrease instead of increase of species fitness?
• are you sure that you use the term “endocarp” in the proper context, when considering the fruit stone of the G. keule fruit in the text starting from line 103? From anatomical point of view an endocarp is an outermost layer of a stone, surrounding a seed, which is inside. Thus, in my opinion, the more proper would be “stone”. E.g.: do endocarps were retrieved from animal faeces or these were whole stones (line117)? do elephants discard only endocarps or whole stones (with seed inside the endocarp) (line 119), etc., etc. – check the whole manuscript from this point of view.
• I advise consultation with a native speaker, turning special attention on tenses: your results should be reported in past (simple) tense and “established knowledge” cited in simple present. This is a standard approach in scientific writing, which allows easy dividing what are the new findings and what is just a citation. There are also several fragments of the text, which need polishing.

Specific comments:

Line 17: endangered tree SPECIES .. and please add name of this “monotypic family”.

L24-25: “Moreover, results from our feeding trials with elephants, may also imply low efficiency of extinct gomphotheres as seed dispersers of this species.” This is the main problem of this paper. The whole story seems to be built around gomphotheres as extinct seed dispersers, which also influenced evolution of dispersal syndrome traits of G. keule, while all experimental evidence is against the hypothesis that endozoochory is the most favourable dispersal mechanism. It needs wider discussion and development of another hypothesis.

L38: what do you mean by over-sized? Over-sized in comparison to what? Guimarães et al. 2008 gives exact range of size, which classifies fruits as megafaunal. Why you did not use these very exact numbers for comparison and discussion of your results?

L44 (and L50, 139): “frugivore” suggests specialisation in fruit consumption, which is overuse in the context of megafauna. It is difficult to imagine megafaunal frugivore. At least contemporary species of megafauna are generally herbivorous, and fruits are just part of their diet.

L45-46: “compared” is rather not a proper term here and the whole sentence is strange (if smaller vertebrates do not ingest the whole seed, then they can’t disperse it because it would be damaged and digested by them) ... rewrite it.

Figure 1: I would delete “megafaunal” from the figure caption. This is part of the results - you did not discussed yet if G. keule fruits fit or don’t fit description (traits) of the megafaunal fruits.

Table 2 is difficult to follow because limits between the lines are not clear – which examples belong to which trait? I would also suggest to replace “Big sized fruit” by exact range of dimensions given by Guimarães et al. 2008 or other sourced. Another question: do you mean here large seed or large stone?

L104: “whole fruits” or at least big chunks of them together with stone.

L107: really only a few (less than 10)?

L120: what do you mean here by “the second year”? Second year of feeding fruits to elephants?

L123: endocarp can’t germinate .. seed can. Change it.

L125: how seed can be damaged if endocarp is intact? Unless all stones retrieved from animal dung were damaged? Why you did not show how many endocarps of the retrieved stones were (optically) intact and how many were cracked in your germination experiment? Don’t you think that seed may be still undamaged just they changed their physiology and became dormant? Explain/discuss.

Figure 3: how large percentage of intact and retrieved seeds germinated? This information is very important for understanding dispersal and germination patterns of the species.

L142: thus, why you did not use them? In table 2 you gave names of plant species considered to be megafaunal but you did not use descriptors of these traits used in literature.

L146-148: but it seems not protecting seed of G. keule … can explain why?

L165-166: whole fruits or at least big chunks of them.

L174-176: lack of seedlings far from the mother plants does not contradict efficiency of long distance dispersal of seeds by livestock – there are many secondary factors, which may reduce number of seedlings even if seeds are transported abundantly: seed predation by rodents, lack of proper symbiotic fungi needed for germination, not proper (micro)habitats where seeds are stored, damage made to seedlings by livestock or other animals.

L189-190: adaptation to browsing (selective feeding) does not contradict ability to disperse seeds.

L196: rather needed than warranted?

L198: the title is misleading because there is no “present-day megafauna” native to this region; I suggest changing it.

L202: fruit or fruit stone were processed .. not endocarp.

L205: damaging the endocarp could be profitable for germination (I would expect that hard woody endocarp prevents germination), what you mean here is probably damage done to seed?

L207: is Cuvierionius the most recent scientific name of this genus? I was also sure that only one species from this genus was present in the studied region – were there more of them?

L209-213: this contradiction is the biggest problem of this manuscript: if seed dispersal by gomphoterus was not beneficial to G. keule due to high percentage of damaged seeds, then the tree dispersal syndrome should not evolve into endozoochoric direction …

L224-225: dispersal may take place also without animal vectors … it is much more safe to write that dispersal mechanism is unclear or unknown

L236-237: no, you did not carry such a comparison. At least it was not obvious in your results.

L254: how it is possible that 100 fruits were translated into only 4-15 pots five seeds each? Add this information in the Methods.

L255: how many control pots were established?

L263: which R packages were used? Cite them with their corresponding references.

L280: delete “very nicely” – such expressions are not acceptable in the scientific texts.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

The MS has been significantly improved by the Authors. I have jest a few remarks, which in my opinion need to be addressed in the final version:

The story is still very one-sided. During the second peer-review I’ve found the paper by P. Jordano (1995) in which the coevolutionary relationship between plants with fleshy fruits and their seed disperser animals has been questioned (Jordano P (1995) Angiosperm fleshy fruits and seed dispersers: a comparative analysis of adaptation and constraints in plant-animal interactions. Am Nat 145(2):163-191.) Please add this doubt somewhere in the Introduction (e.g. in L36-42), Discussion (e.g. L168) or Conclusions, to acknowledge that there also opinions suggesting that plant dispersal syndromes might have been evolved also independently from seed disperser presence/interaction.

L116: rather ‘diverse’ than ‘different’?

L233-234: Please consider that high germination rates (60-70%) in the germination experiment confronted with very low number of seedlings/saplings observed in situ suggests that there must be another factor(s) limiting establishment of G. keule. This should be acknowledged in the Discussion or Conclusions. This factor could be an inhibition of seed germination by mother tree (or soil fungi under the mother tree), damage of seedlings by pathogenic fungi or by invertebrate or vertebrate herbivores. These obstacles should be revealed by future studies to allow conservation of the species.

L243: in L73 you stated that this is a false drupe … check and unify.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf