Body weight at different ages or changes in weight between these ages has little effect on egg production. Phenotypic correlations are of the order of — 0.1 and genetic correlations, generally insignificant, appear to be slightly negative. It is concluded that high body weight per se is not necessarily incompatible with high egg number and it should be possible to improve both traits by selection.

Genetic variance in sexual maturity is greatly reduced if not eliminated when expression of the trait is delayed by means of light control. Under these circumstances genetic variance in egg number in the first 6–12 weeks of production appears to be much reduced.

Egg production is adversely affected if commenced below a minimum age which, in the case of this strain, appears to be about 32 to 33 weeks.

The first 6 weeks of egg production is not significantly genetically correlated with total production. The first 12 weeks of production is more satisfactory in this respect.

A form of selection based on part record at 12 weeks followed by retrospective culling at 18 weeks is suggested which combines the advantage of extended, hence accurate, measurement and overcomes most of the difficulties of the 2‐year cycle. The scheme carries the disadvantage of a 57‐week cycle which may occasionally enforce reversion to a 2‐year cycle because of the difiìiculties associated with later summer production.  相似文献