Effects of ad libitum feeding on performance of different strains of broiler breeders

Abstract

Tolerance to ad libitum feeding was compared in three genotypes of broiler breeder hens: a standard broiler breeder fed ad libitum (SA) or restricted (SR), a slow growing ‘label’ broiler breeder (L) and an experimental dwarf heavy broiler breeder (E). Two similar experiments were conducted in two distinct research centres.

Feed intake and body weight were measured every 3 weeks from hatch to 40 to 49 weeks of age. Egg production and egg abnormalities were recorded. The number of yellow follicles in ovaries was counted at the age of 32 weeks.

Body weight was stabilised at 2·2, 3·7 and 5·4 kg after 24 weeks of age in L, E and SA hens, respectively. Growth of the SR hens was similar to that of L up to 20 weeks and stabilised at a similar level to that of E hens after 30 weeks of age.

Sexual maturity was delayed by 6 weeks in restricted breeders compared to ad libitum fed hens that started to lay at 20 weeks. SA hens had low egg production and a high proportion of defective eggs, which was largely compensated for by feed restriction. However, productivity of SR hens remained lower than that of L breeders.

Compared to the low viability and reproductive fitness observed with SA hens, the E dwarf broiler breeder tolerated ad libitum feeding and had better egg production, fewer egg abnormalities and yellow follicles per ovary and a higher egg production. However, laying rate was still lower than that of the SR and L groups. Energy conversion (kJ/g egg) from 32 to 40 weeks of age was much higher in the SA group than in the other three groups.

The feasibility of feeding a dwarf broiler breeder ad libitum calls for further research on implications of specific IGF and GH-receptor expression at the level of the ovary in dw chickens.

Acknowledgements

We are grateful to Florence Favreau, Claude Bouchot, Maxime Quentin, André Respen, Rony van Aerschot and Carine Borgers for their essential contribution to bird maintenance, controls and statistical analyses. This work was supported by a European Community grant (QLRT-2000-1732). Ariane Heck gratefully acknowledges the financial support of the Région Centre, France. Veerle Bruggeman is a postdoctoral fellow of the Fund for Scientific Research—Flanders (F.W.O.-Vlaanderen, Belgium).