Selection experiments with salmon: III. Genetic and environmental sources of variation in length and weight of Atlantic salmon in the freshwater phase

Atlantic salmon (Salmo salar) brood stock were taken from 32 strains in Norway for 3 consecutive years. Each year, after fertilization, the offspring in each family from each strain were reared in separate egg trays and fingerling tanks. Individual weights and lengths were recorded on a sample of fish from each tank. Environmental and genetic factors influencing the growth rate were studied. The length and weight at a standardized age were significantly affected by hatching date and number of fish held in a tank. There was also a significant ‘tank effect’ which influenced the heritabilities estimated. Significant differences between strains were found for all three year-classes (P<0.01). The estimates of heritabilities found for different year-classes are very consistent. When ‘tank effect’ was removed the heritabilities estimated from the total material were 0.08 for weight and 0.12 for length calculated from the sire components of variance. Calculated from the dam components, the heritabilities were 0.15 for weight and 0.17 for length. The genetic correlation between length and weight estimated from the total material was 1.00. For the ‘trait’ condition factor there were no significant sire and dam components of variance when tank effect was removed.