This Virtual Experiment investigates the response of a population to artificial selection. A population of corn (Zea mays ) is simulated, with each genotype encoding an oil content. Corn is commonly selected for high oil content, so fitness is calculated according to how much oil each individual contains.

Each generation follows this cycle: the fitness of each individual is calculated and the whole population is ranked in order of fitness. The next generation is formed by breeding from the fittest individuals; the rest are discarded. The response of any population to artificial selection depends crucially upon how many individuals are discarded in each generation. You can enter different input values for this variable before running the experiment. This value is taken as a percentage of the population - between 1 and 99.

There is an optimum number of individuals to discard each generation - a value at which mean fitness increases at its fastest. Try to find this optimum value by running experiments with different values; you should repeat each value several times to make sure that the results are typical. Why does evolution sometimes come to a halt after so many generations, so that mean fitness doesn't change? Why do changes in mean fitness seem random at some values?