| A Minnesota population of black spruce seedlings was studied to obtain estimates of the genetic, the environmental, and the genotype x environmental interaction components of variation. These parameters are necessary for estimating responses to selection. The study population was made up of 357 open-pollinated families collected over the species' natural range in Minnesota. Experiments were carried over two growing seasons at four nurseries.;Height growth after eight to eleven months in greenhouse and in the nurseries was found to be under moderate genetic control. Estimated individual-tree heritability and family heritability was .31 and .61, respectively. These are in good agreement with those of other studies.;Although significant ((alpha) =.01) family x nursery interaction was found for both height measurements in both years, the magnitude of variance components for family x nursery was small compared to the phenotypic variance of family means and its adverse effect on estimation of heritability and effectiveness of selection were considered negligible.;Predicted responses to selection estimated for four different selection procedures using the variance components and heritability estimates from combined analysis of second year nursery height were encouraging. The predicted responses to mass, family, simple combined selection, and index selection, assuming a selection intensity of 5%, were 17.2%, 11.9%, 16.0%, and 19.1% of the population mean, respectively.;Among the three approaches of delineating the population into sub-populations tested, the seed zones suggested by Rudolf (1956) appeared to be most effective. A moderate north-south trend in growth rate of black spruce seedlings was observed, indicating a clinal variation pattern. The data suggest that seeds from extreme northeastern Minnesota (seed zones 5c-d and 6) should be avoided when planting other areas.;Significant phenotypic correlations (.36) and genotypic correlations (.36) between 11-month greenhouse height and second year nursery height suggest that a modest reduction in the size of a test population as early as 11 months after germination would have little affect on the gain from selection two years later in the nursery.;A more detailed study of genotype x environment interaction was carried out in a greenhouse using 14 randomly selected families. A factorial experiment with two soil types (Anoka County sand and hemic peat) and three water table depths from soil surface (5cm, 35cm and 65cm) was established using a split-plot design.;Significant treatment effect and significant differences in seedling growth among families were found in all cases. The family x treatment interaction was significant for the second period growth and the total growth, though variance components for the family x treatment were relatively small when compared to the phenotypic variance of family means.;The stability analysis of the greenhouse data based on the two stability parameters, b(,i) and S(,d(,i))('2), indicated that most of the 14 families appeared to respond in the same way to different environments. |
