Chapter 3

Thedisparityintreespecies

richnessamongtropical,

temperate,andborealbiomes:

TheGeographicAreaandAge

Hypothesis

Paul V.A. Fine, Richard H. Ree, and Robyn J. Burnham

OVERVIEW

According to theoretical models, larger land areas should experience higher speciation rates and lower extinction rates
and thus contain higher species richness than smaller areas, all else being equal. This idea has been applied to explain
the latitudinal gradient in species diversity, and has been named the geographic area hypothesis (GAH). Although
putative differences in the geographic area between tropical and non-tropical biomes within continents have been
linked to the disparity in species richness between biomes, no one has tested the GAH with a global dataset. Using
estimates of tree diversity for 11 biome areas on six continents, we evaluated the importance of geographic area in
explaining patterns in tree diversity at the largest spatial scales. We found that the tree diversity of a biome was not
correlated with its geographic area. However, because area is predicted to influencein situspeciation and extinction
rates within a biome, we considered changes in a biome’s size over tens of millions of years, a time period appropriate
for those processes. We found a significant correlation between current tree species richness and biome size integrated
over time since the Miocene, the Oligocene, and the Eocene. These results suggest that both the wet lowland tropics’
larger area and their longevity have played a significant role in generating and maintaining the extraordinarily high
tree diversity. In addition, minimum biome area during the Pleistocene and current tree diversity were positively
correlated, suggesting that extinction due to contraction of available habitat during glaciation in temperate and boreal
areas may have been important factors reducing diversity at high latitudes. These results support the predictions of a
related hypothesis, the tropical conservatism hypothesis (TCH), and may explain why most tree lineages have arisen in
the tropics, and why tropical forests contain such high tree species richness compared with extra-tropical forests. One
of the implications from our results is that conservation of large areas of tropical forests should be given the highest
priority b ecaus eth es efor ests should b eth emost s ensitiv eto extinction du eto habitat loss.

INTRODUCTION

In th emid-1990s, in his bookSpecies Diversity
in Space and Time, Rosenzweig championed area
as th eprimary factor producing th elatitudinal
gradient in species diversity (Terborgh 1973,

Rosenzweig 1992, 1995). The geographic area

hypothesis (GAH) posits that, all else being equal,

larger areas should promote speciation and reduce

extinction; thus a larg ebiom eadjac ent to a

smaller biome should contain more species. If

th eGAH is tru e, th elatitudinal distribution of