This project addresses fundamental issues in causal reasoning with the long range goal of developing theoretical foundations that will facilitate building intelligent systems capable of operating autonomously in dynamic and uncertain environments. This project uses causal Bayesian networks to represent and reason about causal relationships and to address several related topics, including axiomatizing causal reasoning, model testing, identifying causal effects, and causal reasoning in structural equation models. In addition, the project will produce a causal reasoning software tool that can answer causal queries. Causal reasoning is intrinsically a multidisciplinary topic and the results of this project, particularly the resulting software tools, would be useful in other fields such as statistics, economics, health care, and social sciences. The educational component of this project includes curriculum development, student mentoring activities, and efforts to involve undergraduate students in research.