Abstract
Viability may be separated from other components of selection and in turn divided into 2 parts, one dependent on population density and another independent of it. The method of doing so is applied to 2 phenotypes, wild and ebony11, in order to determine the relation between their viabilities and population density. A series of expts. based on different population densities for the cross +/e11 x +/e11 shows that a real though small viability difference exists between the wild and ebony11 progeny when crowding is at a minimum. This difference increases as crowding increases, resulting in a decrease of ebonyll to 15.25% when crowding is at the maximum provided by having 150 individuals of each sex as parents. Most of the decrease in viability is based upon the component dependent upon crowding. Under conditions of the expt., this component ranges from 0 with minimum crowding (1 fertilized female per culture) to 45.9% with maximum crowding (150 individuals of each sex per culture).