Predicting the appearance and disappearance of species : from small islands to small genera

In this compilation of studies I address the appearance and disappearance of species. By developing a new model of the turnover of breeding bird species on islands, I test a prediction that this turnover should decrease with increasing area of an island and also decrease with its increasing isolation. The prediction turns out to be correct, but the underlying causes are not those upon which the prediction was based. I then go on to review the status of turnover analysis at various scales of space and time, comparing the approaches taken by ecologists and paleontologists. I find that the models have much in common, but that methods and nomenclature should be standardized. I propose a set of standards, and apply an ecological model of the appearance and disappearance of breeding bird species to a paleontological dataset, where it reveals the existence of 'Lazarus' and 'Elvis' species. Finally, I analyse patterns of species extinction across genera. For mammals, I find that historical and likely future species extinctions are most likely to be drawn from species-poor genera. This also true for historical bird extinctions, but less so for likely future bird extinctions. I suggest that a change from introduced species to habitat destruction as the main cause of bird extinctions is responsible.