Selection is not, in itself, generally believed to be a creative process. Ideally, it is a means by which the good, bad and indifferent elements comprising a given population are identified and classified. As such, selection is necessarily fundamental to any breeding system, whether the aim is consistent improvement or merely the maintenance of advances already achieved. Without some differentiation of the material, no firm basis can be established upon which to carry out further work. The fact that in dairy cattle breeding in particular, it is seldom possible, where characteristics of economic importance are concerned, to conduct an intimate inquiry into the different genetic elements comprising the given subject with any degree of certainty, should not be permitted to detract from the value of preliminary selection as a foundation for more comprehensive determination of worth. With such "aids to selection" as progeny tests, production records, pedigree estimates and type valuations, a fairly accurate estimate of hereditaty constitution can frequently be arrired at in so far as it affects the functions of economic value. The mode of operation of inheritance is now known in considerable detail and the breeders pursuing a broad programme of improvement may "act as if he knew the genes themselves" and make selections accordingly.