Interprovincial patterns: species–area curves in evolutionary time

We can use the definition of a biological island to define the next larger biogeographical unit, the biological province. A biological province is a self-contained region whose species originate entirely by speciation within the region. Of course, no such region actually exists (except for the whole Earth). But the definition sets up a continuum between islandness and province-ness. As the proportion of originations in a region varies from 100% immigration toward 100% speciation, the region is passing from island to province. The continuum helps us see that we need not create two pigeonholes – island and province – only to battle over the place of one or another region. Instead, for practical purposes, we can say that once a region's species are predominately (60%? 80%?) native, we will group that region with the provinces. Often, that means simply recognizing continents as provinces.

When Williams (1964) extended his plant species–area curve to take in whole continents, he saw a much steeper slope than for the within-province curve. Preston (1960) points out the same phenomenon in his bird curve (Figure 2.15). But neither Williams nor Preston paid much attention to the high between province z-values suggested by their graphs.