Official website of the New Zealand Farm Forestry Association

Results of assessments in FR375, Beaumont controlled-pollinated 1999 Douglas fir progeny trial

By M Suontama, C Low, M Miller, K Fleet, R McKinley, December 2015.

Download SWP-T001 (pdf)

Executive summary

The 1999 controlled-pollinated Douglas-fir trial at Beaumont (FR375) was designed to compare selections from the first Douglas-fir breeding population. They had been selected as the best individuals in the best-adapted provenances in the 1957 & 1959 provenance trials, grafted up in 1989 and planted into the Waikuku archive in 1990. Flowers on grafts in the Waikuku archive were bagged, then pollinated with a pollen mix of 10 genotypes from 1994 to 1996, but difficulties in obtaining pollen meant that several different pollen mixes were used for pollinations in different years.

Seed-set was generally low, so it was decided to sow the seed that had been collected in 1997, but to also include some open-pollinated seed that had been collected from some of the same genotypes by Proseed. Germination was also low for most families; but eventually fifty selections were available to be tested in a trial. Scientists in British Columbia had enjoyed success with early assessments of trials on fertile soils (farm-field trials), so this trial was planted at 2500 stems per hectare on a good site, with the expectation of an early assessment and ranking.

Some wet areas developed in the trial, leading to mortality and some toppling, but the majority of the trees grew well. The trees were first assessed when they were three years old. At that time it was noted that some of the trees in the wetter areas appeared to not have grown at all.

A comprehensive assessment was performed in 2015, and the following traits were assessed: breast-height diameter, stem straightness, branching frequency, stem malformation, acceptability, acoustic stiffness, and pith-to-bark increment core wood density.

Heritability was estimated to be 0.31 for diameter and 0.37 for acoustic stiffness, with a lower heritability of around 0.20 for straightness, malformation and acceptability. Heritability of densitywas 0.26 for rings 1 to 5, but lower heritability’s were estimated for rings 6 to 10 (0.17) and rings 11to 15 (0.12). The branching frequency score had the lowest heritability of 0.08. In this trial, provenance had no significant effect on diameter growth.

Genetic correlations estimated are only indicative as most of them have large standard errors. Genetic correlation between diameter and acoustic stiffness was negative, but not significant at - 0.14. Genetic correlations for density rings 1 to 5 with rings 6 to 10 and 11 to 15 were high. Because heartwood density is determined by similar genes in different rings, and rings 1 to 5 have a higher heritability for density than other rings, selection can be undertaken based on rings 1 to 5. Negative, but low genetic correlations were estimated between density rings 1 to 5 and 6 to 10 with diameter, whereas a high (-0.98) unfavourable genetic correlation was estimated between density for rings 11 to 15 and diameter.

Some genotypes had good breeding values for all traits, so will be obvious selections for seed production. In a trial like this, the poly-cross mating design normally limits the number of forwards selections that can be made. It is the intention of this programme to use DNA-based techniques to identify male parentage and limit any inbreeding in forward selections of interest.


No posts yet

Farm Forestry - Headlines