Continuous selection for egg production using short-term records

Abstract

A closed flock of White Leghorn has been selected for high egg production, since 1947. Selection has always been based on performance achieved from date of first egg until May 31 (part-period test). Selection has been effective in providing gains of approximately three eggs per generation for the part-period and, at the end of the experiment, there is no evidence that the rate of gain has lessened. Total 72-weeks egg production increased during the earlier years of selection but seems to have plateaued subsequently. This leveling off of total production is due to a decline in performance for the period June 1 to 72 weeks of age. The magnitude of this decline is sufficient to offset the gain in the part-period. Annual selection differentials were calculated for the later years, and generally the realized selection differentials were the same size as that intended, which indicated that natural selection was not opposed to the direction of artificial selection. The estimates of heritability of part-period production index show no decline as a result of selection; the actual response observed during the last years of selection supports the contention of absence of such a decline. Although it cannot be statistically confirmed, there are indications that the genetic correlation between part-period and residual performance has decreased in value with the progression of selection. The size of the correlated decline in the residual production suggests not only a reduction in value of the genetic correlation but also a change of sign. The results obtained imply that continuous selection, based on egg production for the part-period, may be an unrewarding procedure; and that selection based on other characters, favourably correlated genetically with total production, might be more effective in providing genetic gain in total production.