In the present paper a broad view of animal breeding is given, and some of its problems and prospective in 1998 are described. Issues related with the definition of breeding objectives, genetic evaluation, methods to utilize different types of gene action, optimization of breeding programs and use of new technologies are discussed.  It is concluded that evaluation of animals on the basis of measurements (phenotype) and pedigree for calculating estimated breeding values (EBVs) would remain as the basis for genetic improvement of animals. Improvements in the methods to obtain economic weights for different traits for selection and to maximize the response to breeding programs restricting inbreeding and costs with new optimization and analysis tools for non-linear complex problems such as the genetic algorithms and artificial intelligence are expected. Other improvements are the generalization of methods for genetic evaluation of populations consisting of different breeds and crosses and the development of methods to incorporate genomic information (DNA molecular markers), and the use of new reproductive techniques based on the manipulation of reproductive physiology of animals, gametes and embryos in the design of more efficient animal breeding programs. New advances in reproductive and molecular technology could help in the improvement of the analysis and use of interactions among genes and between genes and the environment.