The Neogene transition from C (sub 3) to C (sub 4) grasslands in North America; assemblage analysis of fossil phytoliths

The rapid ecological expansion of grasses with C (sub 4) photosynthesis at the end of the Neogene (8-2 Ma) is well documented in the fossil record of stable carbon isotopes. As one of the most profound vegetation changes to occur in recent geologic time, it paved the way for modern tropical grassland ecosystems. Changes in CO (sub 2) levels, seasonality, aridity, herbivory, and fire regime have all been suggested as potential triggers for this broadly synchronous change, long after the evolutionary origin of the C (sub 4) pathway in grasses. To date, these hypotheses have suffered from a lack of direct evidence for floral composition and structure during this important transition. This study aimed to remedy the problem by providing the first direct, relatively continuous record of vegetation change for the Great Plains of North America for the critical interval (ca. 12-2 Ma) using plant silica (phytolith) assemblages. Phytoliths were extracted from late Miocene-Pliocene paleosols in Nebraska and Kansas. Quantitative phytolith analysis of the 14 best-preserved assemblages indicates that habitats varied substantially in openness during the middle to late Miocene but became more uniformly open, corresponding to relatively open grassland or savanna, during the late Miocene and early Pliocene. Phytolith data also point to a marked increase of grass short cells typical of chloridoid and other potentially C (sub 4) grasses of the PACMAD clade between 8 and 5 Ma; these data suggest that the proportion of these grasses reached up to approximately 50-60% of grasses, resulting in mixed C (sub 3) -C (sub 4) and highly heterogeneous grassland communities by 5.5 Ma. This scenario is consistent with interpretations of isotopic records from paleosol carbonates and ungulate tooth enamel. The rise in abundance of chloridoids, which were present in the central Great Plains since the early Miocene, demonstrates that the "globally" observed lag between C (sub 4) grass evolution/taxonomic diversification and ecological expansion occurred at the regional scale. These patterns of vegetation alteration imply that environmental change during the late Miocene-Pliocene played a major role in the C (sub 3) -C (sub 4) shift in the Great Plains. Specifically, the importance of chloridoids as well as a decline in the relative abundance of forest indicator taxa, including palms, point to climatic drying as a key trigger for C (sub 4) dominance.