The Making of Amazonian Diversity The biodiversity of the Amazon Basin is legendary, but the processes by which it has been generated have been debated. In the late 20th century the prevalent view was that the engine of diversity was repeated contraction and expansion of forest refugia during the past 3 million years or so. Hoorn et al. (p. 927 ) analyze findings from a diverse range of disciplines, including molecular phylogeny, ecology, sedimentology, structural geology, and palaeontology, to offer an overview of the entire history of this region during the Cenozoic era (66 million years ago). The uplift of the Andes was a pivotal event in the evolution of Amazonian landscapes because it continually altered river drainage patterns, which in turn put a variety of pressures on organisms to adapt to changing conditions in a multiplicity of ways. Hence, the diversity of the modern biota of the Amazon has more ancient origins than previously thought.

Late Miocene tidal sediments of Acre, Brazilian Amazonia, were deposited in an embayment or interior seaway located in the sub-Andean zone. This late Tertiary embayment system may once have connected the Caribbean with the South Atlantic. The tidal coasts of the embayment-seaway have provided an avenue for the earliest waif (over water) dispersal phases of the great American biotic interchange in the late Miocene. The subsequent change from semimarine to terrestrial environments is of value in assessing the importance of earlier hypotheses on the evolution of the westem Amazonian landscape and gives insight into the formation of several observed biogeographic patterns, especially of aquatic biota.

The fossil-bearing stratigraphic sections of the Solimões Formation (southwestern Brazilian Amazonia) are exposed mainly along the Juruá, Purus, and Acre rivers, and in road cuts. These deposits have provided fossils of the four main lineages of Caviomorpha -Cavioidea, Erethizontoidea, Octodontoidea, and Chinchilloidea, contributing to the understanding on the evolution of tropical Neogene rodents. Herein, our knowledge about fossil rodents from this region is reviewed. New specimens are recorded, including taxa mentioned for this region for the first time, such as basal cavioids, Dolichotinae, Caviodon (Hydrochoeridae), and Drytomomys (Dinomyidae). Unfortunately, the deposits have no absolute ages, and based on palynological data and the biochronology of several taxa (mainly mammals), the encompassed fauna has been constrained to the late Miocene. However, some rodent lineages recorded here seem to be more related to older faunas, from the middle Miocene and Paleogene. Regarding the biogeographic and paleoenvironmental affinities, most of the Neogene rodents from the Acre region show more similarities to those from the Entre Rios, Argentina, and Urumaco, Venezuela, where wet environments were present during Neogene times. An increase in prospecting along southwestern Amazonian rivers looking for rodents (among other vertebrates) associated with methods to better constrain the ages of these faunal assemblages will contribute to a better understanding of the evolution of the tropical rodents as well as the stratigraphy and age of that portion of the basin.

No abstract

The ecomorphological diversity of caviomorph rodents in South America included giant forms, such as the chinchilloid Neoepiblema acreensis from the Upper Miocene of Brazil. The evolution of the brain anatomy and size of these animals can be now studied with non-invasive imaging techniques and exceptional fossils. Caviomorphs show diversity in the traits of the olfactory bulbs, cerebrum, cerebellum, cranial nerves, and blood vessels. Neoepiblema acreensis had a gyrencephalic brain, with an expansion of the frontal lobe, lacking an evident paraflocculus. Compared to the predictions based on extant taxa, even when considering taphonomical effects, N. acreensis , a rodent that weighted almost 80 kg, had a very low encephalization quotient compared to other rodents. The adaptive value of a low energetic cost and other ecological factors could explain the presence of a small brain in this giant rodent––a pattern we also hypothesize for other Neogene giant rodents.