Net genetic and temporary epistatic or maternal environmental responses to selection for egg production in chickens

The difference between progeny from selected and unselected parents (T) was experimentally partitioned into net genetic change (G(A)), temporary favorable epistatic combinations (G(EP)) and egg-transmitted maternal environment (M) in two strains of Leghorns selected over 14 years for early pure-strain egg production. Differences among progeny from selected sires and dams, selected sires only, selected dams only, unselected sires and dams and the parental generation were equated to expected G(A), G(EP) and M responses for each trait. Total response was 3.3% for early egg number, 3.7% for total egg number, 0.5% for egg weight, 3.8% for early egg mass and 4.2% for total egg mass. Among progeny that survived the test period and were judged to be normal, total response was 2.6% for total number of eggs, 3.0% for early egg mass and 3.1% for total egg mass. The percentage of T attributed to G(A) was 9% for early egg number, 24% for total egg number, 43% for early egg mass and 47% for total egg mass; but 52% for total egg number, 98% for early egg mass and 71% for total egg mass of normal survivors. Temporary maternal selection responses (M) were (1) positive for number of eggs and egg masses, (2) greater for all progeny than for normal survivors, and (3) increased with progeny age. The results suggest that M was caused by reduced egg-transmitted disease. Epistatic selection response was positive for earlier sexual maturity and for number of eggs, but negative for egg weight and thus was small for egg masses. Temporary epistatic and maternal responses can explain overestimation of additive genetic response from offspring-parent regression or from replicated single-generation selection and apparent superiority of mass selection over family or combined selection.