Scale, succession and complexity in island biogeography: are we asking theright questions?

1 This paper offers a commentary on the development of island ecological th eory since the publication of MacArthur & Wilson's equilibrium theory in th e 1960s. I distinguish the simple model at the core of their Equilibrium Th eory of Island Biogeography (ETIB) and the broader body of their theory, wh ich embraces evolutionary as well as ecological patterns - all, however, wi thin the overarching framework or assumption of equilibrium. 2 The basic problems with the ETIB have long been known, and its status as a ruling paradigm has been the subject of concern for more than two decades . With the development of nonequilibrium ideas in ecology, island biogeogra phers arguably now have viable theoretical frameworks to set alongside or a round the ETIB. Four conditions are highlighted as extremes: i) dynamic equ ilibrium; ii) dynamic nonequilibrium; iii) 'static' equilibrium; and iv) 's tatic' nonequilibrium: together providing a conceptual framework for island ecological analyses. 3 The importance of scale is stressed and attention is drawn to Haila's spa tial-temporal continuum as an organizational device. It is argued that the processes represented within the ETIB (and by extension, other island theor ies) may be prominent within only a limited portion of this continuum, whil e elsewhere they are generally subsumed by other dominant processes. 4 Colonization and ecosystem development of near-shore islands constitute j ust a special case of ecological succession, and thus the development of th eories of island assembly may benefit accordingly from efforts to incorpora te ideas from the ecological succession literature. 5 The desirability of specifying answerable questions is stressed, as is th e need to build a greater degree of complexity into the development of isla nd ecological models. Notwithstanding which, it is also recognized that key advances are often brought about by simple, but bold models, of the form e xemplified elsewhere in this issue.