Combinatorial testing is a specification-based testing criterion, which requires that for each t-way combination of input parameters of a system, every combination of valid values of these t parameters be covered by at least one test case. This approach is motivated by the observation that in many applications a significant number of faults are caused by interactions of a smaller number of parameters. In this paper, we propose new test generation algorithms for combinatorial testing based on two artificial life techniques: a genetic algorithm (GA) and an ant colony algorithm (ACA). The usefulness of these algorithms is demonstrated through experiments. In the case t = 3 in particular, our algorithms exhibited impressive results.