Genetic Parameters of Stem and Wood Traits in Full-Sib Silver Birch Families

Round 1

Reviewer 1 Report

The main gol of the publication submitted for review is to analyses the phenotypic and genetic variability of silver birch. Scientific research on silver birch, especially on that species breeding, has been conducted for a long time, but their number is small. As the authors wrote, it probably results from the small share of birch in forest management. Birch is a species unjustly underestimated by forest management.  Improving the quality of harvested birch wood, especially in short growth cycles, can change this situation.

The presented research covers the important topic of forestry, improving the quality of birch wood.

The article is well written, the goals are logical and the hypotheses follow. The conclusion responds to the set goals and hypotheses. The research material used is sufficient, and the statistical methodology is up-to-date and correct.

The main mistake of the research is the lack of discussion of the climatic conditions of growth of the tested plants.

I believe that the work requires supplementing the information on the climatic conditions of growth in the chapter meterials. Additionally, detailed fixes:

1) 'Line 39-42- B. pendula and B. pubescens differ in environmental requirements, you should  describe the differences affecting the growth possibilities of species.

 2) Figure 1 should be supplemented with a map with the geographical location of the research area.

 3) Line 145-147 - how many trees were removed during the thinned.

Author Response

I have included the sentences which needed action.  There is also an attached document with both reviewers' comments.

 

The main mistake of the research is the lack of discussion of the climatic conditions of growth of the tested plants.
I believe that the work requires supplementing the information on the climatic conditions of growth in the chapter materials.
Climatic data has been provided.  The key differences in site conditions have been mentioned.  Please indicate if further details are required, and the expected type and format for climate data. 

Additionally, detailed fixes:

1) Line 39-42- B. pendula and B. pubescens differ in environmental requirements, you should describe the differences affecting the growth possibilities of species.
We have mentioned that silver birch often requires fertile and well drained sites, compared to the more tolerant downy birch.  

2) Figure 1 should be supplemented with a map with the geographical location of the research area.
There is now figure 1: (a) Hyssna, (b) Brunsberg, and (c) Mid-south Sweden

3) Line 145-147 - how many trees were removed during the thinned.
Information about thinnings has not been registered for Brunsberg, although the Hyssna stem number was reduced by 30 % during the first thinning.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments to the editor and authors

General

Genetic variation in height and diameter and stem form and wood quality traits was studied in 22 families of silver birch planted with clones at two sites. Genetic parameters were estimated, and relationships between growth traits and stem form and wood properties were described. Discussions were in particular made of the use of non-destructive testing of quality traits in breeding of the species.  

The stem form and wood property traits analysed are of interest in breeding of silver birch. The used of non-destructive testing methods of quality traits is also an important issue. Genetic parameters for these traits are important for developing breeding strategies for the species. However, it is my opinion that the small sample of parents involved in the present material and the limitations of the quantitative genetic analyses performed reduce the conclusions that can be drawn from the estimates of these parameters. I therefore find that too lengthy discussions are made about correlation patterns in 4. Discussion.

My recommendation is that this manuscript should not be published as it is presented here. It may be published in a shorter version, after a new review, when improvements in the quantitative genetic analyses are made, see comments below.          

 

Comments to the quantitative genetic analyses

The 22 full-sib families were generated by crosses between seven mother trees and ten father trees. The crossing design is not presented, by based on the description it could be like that in the following table.

Mother\father	1	2	3	4	5	6	7	8	9	10
1	X	X	X	X	X
2	X	X	X			X
3				X	X	X	X
4					X		X		X
5								X	X
6									X	X
7						X				X

 

This in an uncomplete factorial design with complicated relationships between the full-sib families. This has implications for a proper quantitative genetic analysis.

The model specified for the quantitative genetic analyses presented on page 6 is very general and not complete. What does “genotype” mean, clone or family? In the set of 22 families you have genetic variability between families, which are partly related, and among clones within families.

It is stated that additive genetic variances and heritabilities were estimated. How was that done? The model for these analyses should be stated. It could be based on the uncomplete factorial shown in the table above.

In general, the description of the analyses and the estimation of variance and covariance components is not complete.

A subsample of trees was randomly selected from each family and non-destructive measurements were made on these trees. Was the clonal identity of each selected tree noted? They could be ramets of the same clone of just full-sibs. This again has implications for the analysis and interpretation of the estimated parameters. If family and clonal identity have not been taken into account in the analyses, then the estimates of the genetic parameters may be strongly biased.  

p-values are reported from which tests of significance. What about normal distributions of the traits?

Were any transformations (normal score) made before analyses of the scores of stem straightness?

 

Specific comments (reference to lines in the manuscript file)

The term “plot” is used in two different meanings in Table 1 (single-tree plot, uncomplete blocks).

Both terms “genotypic correlations” and “genetic correlations” are used. What is the difference? (l. 55, l .217, l. 225)

There is an improper use of the terms seedlings, clones and ramets (l. 126-139).

“…between 28 and 35 tested genotypes per family…” What does “tested” mean? (l.135)

Is “age” the years from planting or from vegetative propagation (l. 142-146)?

Subsection 3.1 (l. 284) should be 3.2.

Authors are listed by names on page 11, and not by author numbers.

The very high values of the estimate of the GCV of the grain angle (GA) may be due to mean values close to 0.

 

Author Response

Please see the attachment (reviewer 1's comments are also in the document)

Our comments are red and there are additional figures

 

We have hopefully improved the description of the analysis, but there are some new numbers as we re-did the calculations of grain angle.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

I have checked the revised version of the manuscript. My recommendation is that it could be published in Forests. However, I have two comments that should be considered: • The table of crosses could be placed in an appendix • The sentence in line 153-154: “There were between 2 and 9 clones per genotype with an average of 4.6 clones per genotype” should be rephrased, perhaps as follows “There were between 2 and 9 individuals (ramets) per clone with an average of 4.6 individuals per clone”.

Rationale: A clone is one genotype, ramets or individuals are the members of the same clone and are have the same genotype. A seedling is a plant reproduced from a seed.