Recognition that causal processes change during plant invasion helps explain conflicts in evidence.

Despite intensive research, we still have no general understanding of why plant invasions occur. Many different mechanisms of plant invasions have been proposed, but studies designed to investigate them often produce inconsistent results. It remains unclear whether this unsatisfying state of affairs reflects the complexity of the real world (in which every invasion is unique) or the failure to identify the key processes driving most plant invasions. Here we argue that greater generalization is possible, but only if we recognize that the ecological and evolutionary processes enabling a species to advance into a new area change during the course of an invasion. In our view, an invasion can often usefully be subdivided into a primary phase, in which the abundance of an often preadapted species increases rapidly (typically in resource-rich, disturbed habitats), and a secondary phase, in which further spread is contingent upon plastic responses or genetic adaptation to new ecological circumstances. We present various examples to show how this partitioning of the invasion phase sensu stricto produces new hypotheses about the processes underlying plant invasions. Some of these hypotheses can be conveniently tested by investigating plant invasions along strong environmental gradients such as those that occur in mountainous regions.