Seeds’ Early Traits as Predictors of Performance in Direct Seeding Restoration

Round 1

Reviewer 1 Report (Previous Reviewer 1)

The manuscript has changed slightly in relation to the previous text and did not dispel my doubts about the authors' misunderstanding of the concept of seed vigor. Hence, I recommend removing this term (word) from the manuscript and replacing it with the term "field germination (capacity); or “% of field germination”, etc. The authors removed the “direct seeding protocol”.  Lack of reflection on the impact of weed management on germination, emergence and development of plants  (including the repeated use of total herbicides in the experiment)? 

What was the depth of sowing? The paper only states that the seeds are covered with a layer of soil. Were large and small seeds sown at the same depth? Maybe this fact (next to real vibility and effective preparation of seeds for sowing - breaking dormancy) was decisive for effective emergence?

The work is based on a single cycle of research. Repetition of the first stages of the experiment 2-3 times in order to determine the appropriate relationship between seed viability and field germination capacity would effectively help to select species for survival studies and to properly select the proportions of individual species in the set. In the current situation, we do not have an answer whether germability  is related to the species, or whether it results from the characteristics of the seed samples and the conditions of emergence. Perhaps sowing the same seeds with the same viability in the following year or a different place would give a completely different result  of germability, which would lead to different conclusions. I am full of appreciation for the experiment carried out, but this aspect should  be clearly emphasized at manuscript.

Tabela 1 is  difficult  to read in the attached format and therefore difficult to analyze.

There are also a number of minor shortcomings in the work, such as:

incorrect citation order  (page 3, line 13 from top) item 33 follows 27

erroneous of the germability formula  (page 3, line 11 from the bottom)

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report (New Reviewer)

The work titled: “SEED EARLY TRAITS AS PREDICTORS OF PERFORMANCE IN DIRECT SEEDING RESTORATION” represents comprehensive effort to analyze early seed and seedling traits in the context of direct seeding restoration in tropical conditions, implementing multivariate statistical methods in the process.

However, I would suggest corrections listed in the attached file.

Comments for author File: Comments.docx

Author Response

Please see the attachment

Author Response File: Author Response.docx

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Some of the things presented in the publication are incomprehensible to me. The authors take the vigor of seeds as a species trait, while it is shaped by environmental conditions during the formation and maturation of seeds and significantly modified by the conditions of storage, transport, age of seeds and all factors affecting their aging. This is not included in the work. 
The assumption adopted in the work that the quality of the seeds used by the Authors is only a species feature (this is also how it is interpreted in the discussion) is inherently wrong. The sowing value of these seeds can be a property of only this particular seed lot (sample). The use of a different lot of seeds of the same species can give radically different results and conclusions. In the manuscript, seed germination under renewal (field) conditions is interpreted as seed vigor. By definition, this is correct, provided that it concerns the sowing of viable seeds with high laboratory germination. Seeds with low germination under optimal laboratory conditions are assumed to have low vigor. In addition, seeds with 2/3/4% germination and even several species with 0% germination were also used for sowing (Table 1). What was the point of using such seeds? What effect was expected?  
In addition, there is the question of the dormancy of seeds of a number of species, the breaking of which, depending on its nature, may require a specific regime of conditions.  The authors mention the use of appropriate methods of breaking seed dormancy, specific to the species, but without specifying the effect of their use. 
In summary, the use of seeds of such diverse and often low initialy quality for research and assigning its species characteristics is fraught with a significant error and does not allow for the correct interpretation of the results and inference.  The use of the same mixtures of species containing seeds from other lots may give quite different results.  If the authors recommend monitoring vigor to reduce the number of seeds used for renewal, then first this vigor should be reliably assessed and only seeds of the highest quality should be selected to assume renewal. And then evaluate the remaining characteristics given in the work.

In the obtained text of the manuscript, there are also editorial shortcomings in the form of errors in the numbering of subsections in chapters 2 and 3, and the lack of figures 3 and 7.

Reviewer 2 Report

This study seeks to determine which species are better for direct seeding with the aim of forest restoration. The topic addressed is timely and very importance, i.e., whether to use direct seeding or previously nursery-cultivated seedlings. There is, in fact, an increasing number of studies and calls to shift (at least partially) to the use of direct seeding for several reasons, including cost, versatility, possibility to access to remote areas, or even the use of drones for forest restoration. However, we still know little about which species are best suited for direct seeding. In this sense, this study is extremely relevant. Nonetheless, the work is a complete mess (I am sorry to say). It is very likely that the data that the authors have are absolutely good and fine. However, in its present form this study does not provide any clear answer to the questions raised, and cannot be accepted for publication. I strongly suggest to address some major concerns and, once this is done, we could judge the validity of the work.

1. To begin, it is confusing which are the variables measured: seed vigor, germinability (that the authors make synonymous), emergence, “probability of success”, etc. They define germinability as “the potential of a seed species to germinate under uncontrolled field conditions”. What this mean is not clear; which uncontrolled field conditions? “Natural” germination? And how can they track germination of a seed that is buried in the ground? I strongly suggest to simplify this, for example as follow:

i) You have seeds (of different species).

ii) You have put on the ground a known number of seeds.

iii) What you measure in the field is EMERGENCE, not germination or germinability.

iv) You monitor and identify most of the seeds that EMERGENCE. This is the great data that you need.

v) You track survival of each individual seedling that has emerged, until day 720. This is also great.

Having this information, you should be straightforward: you have a wonderful value of early seedling recruitment. Why do not work directly with this value and discard so many messy messages? In any case, it would be important to have the germination rate of the seeds used before they are put in the field. This should be done under laboratory conditions to be sure that what we measure is germination under standardized conditions, and this data should be used as a baseline to correct the emergence in the field. Has this been considered in the work?

2. Linked to the previous comments, the experimental design needs several clarifications. Please indicate more clearly what are “groups of three lines”, “five blocks”, “thirty two plots”… and how do they relate each other. It might be helpful a figure showing the experimental design. So far I do not understand the structure of the design, thus I cannot judge the statistical analysis.

3. The statistical analysis does not seem to take into account the spatial structure of the design. In any case, and as indicated above, we cannot judge this point with the information reported in the ms. On the other side, we cannot judge if the data should be analyzed with a survival analysis, including censored data. Please provide details about the pace of seedling emergence (could be that some species emerged during months?).

4. Data analysis section, line 7. You mention here “dispersal syndromes”, but this has not been presented previously.

5. Do not use metrics of diversity to present in the Results the species used! This is simply one of your variables. It does not make sense to talk about this in terms of diversity of legumes, etc. In the same paragraph, you keep presenting data that are irrelevant. For example, that Fabaceae was the family with highest number of species that emerged (12); well it was also the family with more species (15)…

6. Even more, the metrics of biodiversity for seedling recruitment are not the most interesting point of this ms. Above that, I find that the most relevant and novel information is the data obtained for each individual species, and how this might relate to functional traits. Thus, to have more or less diversity is a secondary topic.

7. Do NOT USE in the analysis the mass of the seeds as a categorical variable (small, medium, large). You have to use this variable as a continuous one. This applies to all the analysis and figures. All figures containing the categorization into small, mediun or big are irrelevant.

8. In the same line, please indicate in Table 1 the mass of the seeds per individual seed (mg, for instances). More issues with Table 1: it is useless to indicate the number of plants per hectare; what you should indicate is the percentage of emergence as a measure of establishment success.

9. There is a consideration to the time of the experiment as if it was monitored each single day. For example, in page 10 it is stated that “mortality rapidly increased between 30 and 90 days…”. Does this mean that sampling was done at day 30 and day 90? Or that you sampled every single day? It is stated in Methods that the plots were monitored regularly; please be more precise.

10. It seems that there has been a strong use of herbicide. Please explain/justify clearly that this has no effect on the results, or at minimum that its effect is not different across species.

There are much more issues that could be considered in another version, but that are minor. For example, do not indicate in the text which colors or whatever details are you using in the figures to represent the data. Figure 1 is very nice, but please increase the size of the pictures. The text is too messy in several parts and could be simplified. The discussion is plenty of tautologies and spurious conclusions, a fact that raise from a wrong approach for data analysis, in particular the categorization of the species according to author’s personal views.