On the insularity of islands

We propose that islands are "less insular" than is generally perceived. Thi s belief results. in part, from the paucity of studies on vagrant species t hat exploit islands but are not permanent residents with continual breeding populations. We show, via anecdotal evidence extracted from the literature and data acquired on Gulf of California islands, that visitors to insular systems are fairly common. We delineate three types of events beyond the bo unds of current biogeographical analysis that make islands less insular: 1) migrants and 'accidental' visitors, 2) individuals of a species whose fora ging areas encompass many islands or the mainland and islands. and 3) speci es who "colonize" islands during opportune periods or years but become exti nct during difficult times (source-sink situations). Such events potentiall y significantly affect the ecology and evolution of island inhabitants by s uch means as increased predation and/or competition, transport of parasites and pathogens, dispersal of seeds and eggs, and genetic introgression and hybridization. Discussion of other "insular" habitats such as freshwater la kes and wildlife refuges illustrate that vagrancy events may be nearly ubiq uitous. Studies addressing the frequency and ecological and evolutionary si gnificance of vagrants are required, especially in light of recent and rapi d extinctions on islands and the increasing fragmentation of habitats.