Genetic parameters for yield in macadamia

Broad-sense heritability, and genetic and phenotypic correlations were estimated for yield (nut-in-shell; kernel yield; and net present value of crop) per tree in macadamia to 10 years after planting from a trial of 40 cultivars replicated over 4locations. In addition, canopy width at 10years of age was measured and used to calculate yield efficiency as yield per square metre of projected canopy area. Stem girth above the graft union was also measured. There was a low broad-sense heritability (0.06 < H <0.22) and correlation of cultivars across locations (r _gloc = 0.14–0.52) for yield per tree. However, the genetic correlation of cumulative yield was high among 3 of the 4locations (g ≥ 0.75), suggesting the higher genetic variance at Rockhampton was the main cause of the genotype by environmental interaction. Heritability was higher for canopy width (H = 0.28) and yield efficiency (H ≈ 0.47) and cultivar performance was highly correlated across locations (r _gloc ≈ 0.70) for these traits. There was a strong genetic correlation (r _g ≥ 0.90) between cumulative yield per tree of cultivars to 7 years and cumulative yield per tree to 10 years. Genetic correlation among all measures of yield per tree were high (r _g ≥ 0.73), but there was no genetic correlation with canopy width. On the other hand, small cultivars tended to have higher yield efficiency (r _g <–0.62). There was also a slight negative correlation between nut-in-shell yield per tree and kernel recovery (r _g = –0.37). These results suggest the use of family information and index selection may improve the efficiency of selection and breeding programs in macadamia. This revised version was published online in July 2006 with corrections to the Cover Date.