Experimentation and deduction are the hallmarks of most scientific studies. Evolution lends itself to these analyses. However, evolutionary mechanisms are difficult to study in traditional educational settings for at least two reasons. First, the typical university semester or secondary school year presents a limited amount of time to study processes that often may take years or decades. Second, the instrumentation and molecular procedures required to study allele frequencies in populations are often well beyond the course budgets of many university departments and secondary schools. This is unfortunate, because the agents of evolutionary change are their own best argument for evolution remaining the central theme of biology and in convincing the general public of its importance. (Troubling statistics have been presented recently in Anonymous, 2004; Moore & Kraemer, 2005; Trani, 2004.) The use of computer models to study evolution removes both the time and budgetary constraints. The majority of available computer models, however, simulate evolution using few genes (frequently, only a single locus) and operate by violating only one or two assumptions of the Castle-Hardy-Weinberg Equilibrium Principle (e.g., selection in an infinite population in Populus[R], Don Alstad, University of Minnesota, http://www. cbs.umn.edu/populus/, freeware) or are expensive commercial products requiring either site licenses or other financial commitments (e.g., EcoBeaker[R], EvoBeaker[R], SimBiotic Software, http://www.simbio.com/). Today's students are accustomed to computer games that are more conceptually sophisticated than the traditional classroom computer simulation. For example, the genre known as Real Time Strategy games requires players to manipulate a large number of interacting components (e.g., The Sims[R], SimLife[R], SimCity[R], Electronic Arts). As a result, many students are effectively exposed to complexities analogous to those of evolutionary mechanisms at work in a rich landscape of fluctuating variables. Critters!, unlike other computer simulations of evolution, is closer to approximating the genetic system and gene/ environment interactions of real populations. [FIGURE 1 OMITTED] The Critters! model contains a variety of traits (Table 1) that can be examined to determine if evolutionary mechanisms are at work during a simulation (i.e., natural selection, non-random mating, genetic drift, gene flow, and mutation are all candidates for student study). The program also enables students to see the effects of the interactions between processes, such as mutation and selection. After running the model or seeing screen shots of the model's organisms, students choose to test one or more traits that might be influenced by an evolutionary mechanism and construct an experimental design to address their hypotheses. Model Details Digital organisms in the model are born, acquire resources, grow, mate, reproduce, and die in a virtual world (Figure 1). Each diploid organism carries 54 chromosomes that contain 102 genes. Except for gender, the phenotypes are polygenic (i.e., determined from multiple genes), and genetic variation is additive (each allele has an effect on the phenotype). For gender, heterozygotes are male and homozygotes are female. When organisms reproduce, they make gametes via meiosis with genetic recombination. Mating conforms to scramble-competition polygynandry (i.e., males scramble and compete to catch fertile females; polygynandry means that males and females usually have a different partner for each mating and multiple matings are possible). Each organism consumes three types of resources that exist in three sizes each (small, medium, and large). Two of the resources are energy-supplying (LNG, long-term resource; and SHT, short-term resource), while the third is structural (STR). Short-term energy resources can be thought of as similar to carbohydrates. …