Evolution of Fitness. Ii. Correlated Effects of Natural Selection on the Productivity and Size of Experimental Populations Ofdrosophila Serrata

The average size of a population at equilibrium in a certain environment may be used as a measure of the performance, or fitness, of the population in that environment (Carson, 1961; Dobzhansky and Pavlovsky, 1961; Cannon, 1963; Ayala, 1965a, 1966a, 1967). The average number of individuals emerged in the population per unit time and their biomass may also be used as measures of the fitness of populations. The two measures, average size and productivity, are mutually dependent but need not be strictly correlated. Natural or experimental populations of sexually reproducing organisms are not likely to remain genetically stationary for many generations. Under the action of natural selection, a population evolves genetically to improve its Darwinian fitness in the environment in which it lives. Some attempts have been made to observe and measure improvements in the ability of a population to exploit the resources of a particular environment (Dobzhansky and Spassky, 1947; Buzzati-Traverso, 1955; Carson, 1958; Strickberger, 1963; Ayala, 1965b, 1966b, etc.). Ayala (1965b) studied the performance of four experimental populations of D. serrata over one and a half years, using population size and productivity as fitness estimates. A gradual increase in the average size of the populations was observed throughout the experimental period. The number of flies emerging per unit time remained, however, approximately constant from the 16th week, when an approximate equilibrium was reached, until the end of the experiment. A number of questions are raised by such results. First, is the gradual increase in population size due to genetic changes in the population, or is this increase simply due to the improving ability of the experimenter in handling the populations? Second, has the productivity of the population remained constant, or did it improve genetically during the first few generations of selection when the population was rapidly growing and the experimental design did not allow estimation of fitness improvements? Finally, the relationship between the productivity and the size of a population is brought into question: to what extent may natural selection result in increases of the average size of the population when its productivity is not increasing? The two experiments reported here were designed to answer these questions.