Evaluation and propagation of selected genetic stock of castanea sativa (MILL.)

Abstract

The performance and variability of 143 open-pollinated families of Castanea sativa (Mill) from six natural contrasting European chestnut populations (Spain, Greece and Italy) growing in contrasting environments were investigated. The experimental design was a single - tree plot (one seedling per family) completely randomized block design with 20 blocks and 3 x 3 m spacing. After four years of growth, height, diameter, volume index and number of leaves were studied for three consecutive years at the Taxiarchis- Chalkidiki provenance - progeny test located in Northern Greece. Survival was 69.9% at the end of the 6th year. The mean yields per provenance for growth traits were 3.93 cm for diameter, 107.8 cm for height and 1493.9 cm3 for volume index six years after planting and the mean number of leaves recorded was 563. Considerable genetic diversity was recorded, both among and within populations and growth traits were under strong genetic influence (control). Judging from the mean yield per provenance, the overall best performing provenances were EU-11 (A. Coruna, Spain)) for diameter (4.05 cm), EU 56 (Pellice, Italy) for height (119.2 cm) and volume index (1725 cm3) and EU-7 (Malaga, Spain) for number of leaves (596.7). The mean yields per family ranged from 2.6 to 5.1 cm for diameter, 67.7 to 170.7 cm for height, 468 to 3084.5 cm3 for volume index and 175 to 1075 for number of leaves. The individual genotype ranged as 1.2 to 7 cm for diameter, 20 to 200 cm for height, 40.7 to 4235.5 cm3 for volume index and 50 to 1550 for the number of leaves. Age x family interaction was not significant indicating that families were quite stable in their ranking at the different ages under study. Coefficients of Variation (CVA), individual, family and within family heritabilities were also estimated for growth traits over a three year period and no particular trend could be associated with age. The genotypic variance components were ? 8.1% for diameter, ? 8.5% for height, ? 7.5% for volume index and ? 5.2% for number of leaves within the three years of studies. Heritabilities for growth traits were found to vary from low-high, indicating that improvement of chestnut could be effective through selection. Four selection methods were used to estimate genetic gain. The estimated genetic gain reached 25% for diameter, 28% for height, 68% for volume index and 49% for number of leaves by employing a 10% selection across all selection methods and ages. Genetic improvement using the combined selection method might be achieved more quickly and selection will favour a larger tree volume index production, however the higher heritabilities recorded for height and diameter than volume did not reflect in the estimated genetic gain.This may be due to the influence of the high error variance recorded for volume. Thus selection based on height and diameter would be much more preferable.