Selection for egg production and correlated responses in the fowl.

HE effectiveness of selection for genetic improvement of egg production in the fowl remains a controversial issue. Mass selection on annual egg records has long been regarded as ineffective as demonstrated by the early work of PEARL (1915), while selection based on family records and progeny testing has seemed to be the key to success as demonstrated by the long-time selection experiments of DEMPSTER and LERNER (1947) and HUTT and COLE (1955). All these studies, however. lacked adequate controls to properly assess time trends. Compared with controls, selection in an “old’ Leghorn strain was ineffective after the first generation, but effective in a “new” population starting from a cross-line foundation ( GOWE, JOHNSON, DOWNS, GIBSON, MOUNTAIN, STRAIN and TINNEY 1959). From an intensive study of records of a commercial Leghorn flock coupled with repeat mating controls, DICKERSON (1955, 1961, 1963) deduced that continuous artificial sclection for overall performance in a domesticated population simulates natural selection for reproductive fitness in a wild-type population. Selection leads to a plateau because of fluctuating year-to-year environments, negative genetic correlations between components o i performance and to random loss of useful genes from inbreeding. Slow progress in breeding for egg production has been attributed, in part, to the time required to obtain a complete annual egg record. DEMPSTER and LERNER (1947) concluded that greater genetic progress is possible by selecting only on the early part of the annual egg record because the generation time interval is halved. MORRIS (1963, 1964) and GOWE and STRAIN (1963) found that selection on early record egg number may not necessarily increase egg production in the total period. OLIVER, BOHREN and ANDERSON (1957) concluded that early record percentage egg production is a more efficient selection criterion than is the actual egg count in the early period. Total egg number was more highly correlated genetically with early rate than with early period egg count. Using early rate as the primary selection criterion in a 10-year selection experiment with White Leghorns, YAMADA, BOHREN and CRITTENDEN (1958) reached a plateau by the fifth selection year. This was attributed to a progressive decline in additive genetic variance. Their study, however, lacked controls to