The evolution and diversification of Poaceae
Because of the exceedingly poor fossil record of grasses (Poaceae), little is known about the origin and subsequent taxonomic radiation of the group. Similarly, the
biogeography of early grasses is unclear. We use fossil phytoliths to date major cladogenetic events within Poaceae and study the relative timing of taxonomic radiation and rise to ecological dominance of grasses.

Work with Indian colleagues on presumed dinosaur coprolites from Intertrappean beds in Pisdura, central India, have revealed that several extant sublineages of grasses existed

Grass short cell phytoliths in silificied epidermis from Late Cretaceous coprolites from Pisdura, India. Scale bar = 10 micrometers. See Prasad et al. (2005).

on the Indian subcontinent by the Late Cretaceous (~65 Ma), significantly earlier than dates preferred in molecular dating studies or inferred from other grass fossils (Prasad et al. 2005, 2011). Furthermore, work in North America has shown that the radiation of open-habitat grasses (pooids, PACMAD grasses) occurred by the Late Eocene, 10-20 My earlier than previously thought (Strömberg 2005). 

The research on grass diversification and ecology in the fossil record requires a detailed calibration of diagnostic grass phytoliths (short cells).  For this purpose, a comprehensive study of grass short cell phytoliths is underway in collaboration with Dolores Piperno (Smithsonian Institution) to establish a more explicit, phylogenetic basis for identification of specific grass clades in the fossil record.

Poaceae phylogeny dated using fossil phytoliths from coprolites from Pisdura, India. Red lineages/clades = closed-habitat grasses; yellow clades = open-habitat grasses. See Prasad et al. (2005).