Abstract
It was relatively easy to produce alternative community states as a function of variability in a sequence of species invasions employed to assemble a community. Numerous mechanisms and processes are capable of producing both temporary and nonrecoverable differences in community structure. They include priority effects, intransitivities, emergent properties (e.g., vulnerability to invasion, specific topologies), and effects specific to differences in assembly sequences themselves. In ecological systems, the existence of alternative states presents a difficult comparative problem in the search for unifying principles that might underlie community-level organization. This problem can be solved only if we understand the cause of alternative states and determine whether such states are persistent or transient. Otherwise, ecological communities appear so variable that general principles and mechanisms elude detection, even if they exert a powerful influence over the structure of the system. Mechanisms that are documented as controlling community organization and the population dynamics of component species are not necessarily the mechanisms responsible for observed patterns. Factors uncovered in any extant community may be responsible only for maintaining the status quo. The reason these factors (e.g., competition, predation, parasitism) are currently operating is often a direct function of community-assembly mechanics. Community-assembly processes and dynamics must be understood if we are to discern properly between maintenance and causal mechanisms.