used to convincingly demonstrate that organism interactions with the earth system have contributed to the evolution of life on Earth over the past several billion years. One dramatic example is the evolution of the earliest photosynthesizing unicellular organisms, which radically altered the early earth system by adding free oxygen to the atmosphere and thereby eventually providing the conditions for animals to survive and diversify.

The fossil record also demonstrates that the causative linkages and feedbacks do not always occur in simple or immediate ways—careful and creative investigations are usually required to demonstrate cause-and-effect relationships. A chemist can replicate an experiment many times to demonstrate a cause-and-effect relationship and can thereby reject a hypothesis when it is not supported by the replicated results. However, for historical sciences, our “experiment” has been run and it cannot be precisely replicated. In addition, there often are multiple causative factors as well as complicated feedbacks that controlled events recorded in the fossil and archaeological records. Accordingly, the task of historical scientists studying evolution is to test hypotheses through other means (e.g., Frodeman, 1995):

• By looking for robust correspondences of events in time and in the predicted cause-before-effect order. This requires an accurate and precise understanding of the ages of events.

• By testing whether the predicted cause-and-effect outcome took place multiple times, either under similar situations at different geologic times or, in the case of evolution and ecology, across multiple taxa (different organisms) for a given event. For example, multiple groups of animals with similar characteristics can be analyzed to determine whether their fossil records responded in similar ways to a proposed causative event (e.g., Vrba, 1988, 1992, 1995; Potts, 1996a, 1998).

• By “rerunning” this historical experiment multiple times with computer models, to test and understand the underlying dynamics of the possible cause-and-effect relationship as informed by a combination of hypothesized causal factors (climate forcing functions), initial environmental conditions, and findings from the fossil record.

An important consideration in any discussion of causality is the possibility that hominin evolution was largely unaffected by climate change—the evolution–environment “null hypothesis.”

Ecological factors such as predation, competition, and disease among organisms operate in all environments, and these interactions have an important influence on their evolutionary history. Such interactions can be—but are not necessarily—strongly shaped by climatic conditions with their resultant habitat characteristics, and thus detailed climate studies can provide a critical context for understanding evolution. For example, animals preying upon other animals