No abstract

No abstract

Northern South America harbours a highly diversifi ed forest vegetation. However, it is not clear when this remarkable diversity was attained and how it was produced. Is the high diversity the product of a positive speciation-extinction balance that accumulated species over long time periods, or is it the product of high origination rates over short time periods, or both? Middle Cretaceous fl oras, although very poorly studied, are dominated by non-angiosperm taxa. By the Paleocene, pollen and macrobotanical fossils suggest that the basic phylogenetic composition and fl oral physiognomy of Neotropical rainforests were already present. Hence there was a profound change in Amazonian fl ora during the Late Cretaceous, that still needs to be documented. Levels of Paleocene diversity are much lower than those of modern tropical rainforests. By the Early Eocene, however, pollen diversity was very high, exceeding values of modern rainforests. At the EoceneOligocene a major drop in diversity coincided with an episode of global cooling. The palynological and palaeobotanical records of Amazonia suggest that high levels of diversity existed during the Miocene, a period when the boundary conditions for sustaining a rainforest (e.g. low seasonality, high precipitation, edaphic heterogeneous substrate) were met. The predecessor of the present rainforest was formed during the Paleogene and Neogene when the western Amazon lowlands were affected by Andean tectonism, which radically changed drainage systems and promoted wetland development. An overall global cooling during the Neogene also may have affected the rainforest, decreasing its area and expanding adjacent savanna belts. Recent events like the Quaternary ice ages also played a role in the forest dynamics and composition, although it seems to have been minor. In this chapter we will review the main characteristics of the Neogene palynological and palaeobotanical records in Amazonia, and we will make some comparisons with pre-and post-Neogene records. The data indicate that the Amazonian rainforest is more likely to be a product of a dynamic geological history stretching back over the past 25 million years rather than the last few hundred thousand years.

No abstract