"Jaramillo's intriguing findings provide an evolutionary perspective on a modern crisis," William Laurance of STRI and the Biological Dynamics of Forest Fragments project in Brazil comments, "They suggest that the rapid contemporary loss and fragmentation of rainforests will lead to striking, long-term biodiversity declines."
Jaramillo et al. used cores drilled through 5km of rock in eastern Colombia and western Venezuela to get at the fossil pollen record in a sequence of samples representing 10 to 82 million years before present. Then they correlated pollen diversity with global temperature estimates for the middle part of that sequence (20-65 mybp).
"We found that pollen diversity tracks global temperature through time over millions of years. Diversity increases as the planet warms and decreases as it cools. The mystery is that even when global temperatures vary enormously, average temperatures in the tropics don't change much, so why do we see global temperature patterns reflected in tropical plant diversity?" Jaramillo proposes that changes in area drive speciation and extinction in the tropics.
"There is good correlation between area and number of species: more area implies more species. During global warming, tropical areas expand and diversity goes up, the opposite happens during global cooling. If this is the case, fragmentation of modern tropical forest could be equated to a global cooling period, because forested areas are shrinking dramatically, resulting in plummeting diversity in the forests that remain."