Interspecific pine hybrids II. Genotype by environment interactions across Australia, Swaziland and Zimbabwe

Abstract

Collaborative research trials of Queensland-bred pine hybrids have been established in many sites outside Australia. These trials enable the estimation of genotype×environment effects, which are important in determining the level of regionalization needed in any tree breeding programme. Correlations across sites which are testing hybrids between Pinus caribaea var. hondurensis and both P. oocarpa and P. tecunumanii established in Australia, Swaziland, and Zimbabwe are reported. Diameter at breast height and height were measured in all trials, stem straightness in Australia and Zimbabwe, and wind firmness and lean in Australia, all at 5 years of age. For each pair of tests, additive genetic correlations and the correlation between full-sib family means were estimated for growth traits across sites. Genetic correlations between test-pairs located within the same country were all strong (0.65 to 0.95); however for pairs of tests in different countries the genetic correlations averaged 0.41 and 0.38 for P. oocarpa and P. tecunumanii hybrids, respectively. Family mean correlations were not as strong as genetic correlations, although they followed a similar pattern. A hybrid-production strategy, involving a two-staged approach of (a) reproducing proven superior families for local deployment and (b), crossing parents with high (hybrid) breeding values to produce additional families for testing across a broad range of sites for future deployment, is proposed.