•Succession is ecosystem change caused by discrete disturbances.•Beliefs about successional mechanisms, trajectories and endpoints have changed.•Such beliefs have important implications for restoration and conservation practice.•Models of succession have become more complex, stochastic, and situation-specific.•A single unified theory for successional change is unlikely. Eugene Odum’s 1969 paper, The Strategy of Ecosystem Development, marks a watershed moment in approaches to the study of succession, ecosystem change caused by discrete disturbances. He argued that succession is unique from other kinds of change with regard to mechanisms (modification of the physical environment by the community), trajectory (orderly, directional and predictable), and endpoint (a stable climax ecosystem in which “maximum biomass and symbiotic function between organisms are maintained per unit energy flow”). Odum also argued that understanding successional change was central to the management of a great variety of environmental challenges. Given the important role of disturbance in these ecosystems, this is particularly true for management aimed at restoration and conservation of forests. Although there was considerable debate among ecologists regarding successional mechanisms, trajectories and endpoints in the decades preceding his exegesis, the views outlined by Odum generally prevailed. These significantly influenced answers to three central restoration and conservation questions during that era. (1) What should we restore and conserve? Climax ecosystems. (2) How should boundaries be set for restoration and conservation areas? This was not an important matter. (3) How should restoration and conservation be accomplished? Because succession would inexorably lead to the ultimate climax goal, forest ecosystems should be protected from disturbance. Over the past five decades, virtually every aspect of succession theory as presented by Odum (1969) has come into question. We now understand that there is no single unique or unifying mechanism for successional change, that successional trajectories are highly varied and rarely deterministic, and that succession has no specific endpoint. Answers to the three restoration and conservation questions have changed accordingly. (1) Restoration and conservation goals should include the full range of variation in species diversity and composition associated with disturbance and the succession that proceeds from it. (2) Pattern, scale and context influence patterns of both disturbance and succession, and preserve design really does matter. (3) Restoration and conservation practice must be tailored to the unique mechanisms and post-disturbance ecological legacies that determine the trajectory and tempo of successional change in each particular ecosystem. The search for a grand unified theory of succession apart from other kinds of ecosystem change is futile. Nevertheless, the change caused by discrete disturbances remains an important matter for concern for restoration and conservation practitioners.