Was the evolution of hypsodonty in South America a response to the spread of grassland vegetation?: New phytolith records from Gran Barranca, Argentina.

STRÖMBERG, C.; DUNN, R.; KHON, M; MADDEN R.H.; CARLINI A.A.

Pittsburg

Jornada; Society of Vertebrate Paleontology 2010. Annual Meeting; 2010

Faunas from Patagonia, South America show incremental increases in hypsodonty in several meridiungulate lineages from the Middle Eocene (~38 Ma) onward. Hypsodonty is generally assumed to have evolved as an adaptation to feeding on silica-rich grasses or in open, dusty habitats, so this pattern of faunal change has been thought to indicate that savanna grasslandpaleobotanical data from Patagonia either indicate closed forests during the Eocene and Oligocene (macrofossils, palynofloras) or support the idea of open, grassy habitats by the Late Eocene [plant silica (phytolith) assemblages]. We tested the South American early grassland hypothesis through a high-resolution study of phytolith assemblages in the Sarmiento Formation at Gran Barranca, Chubut Province, Argentina. Spanning 42-18.5 Ma, this section is arguably the most complete and important record of South American faunal change, and contains a rich phytolith record. Our preliminary data show that various forest indicators, including palms and woody dicotyledons, dominated phytolith assemblages throughout the section. In particular, palms were very abundant in the Eocene and Oligocene. Grasses were present from the base of the section (42 Ma), but initially consisted primarily of what were likely bambusoid grasses. Phytoliths typical of open-habitat grasses (pooids, PACMADs) do appear by 38 Ma, marking the earliest occurrence of these two clades, but are rare. Late Oligocene/Early Miocene grass phytoliths, now dominantly pooid open-habitat types, increased in abundance to ≤30% of the assemblages. This expansion of grasses coincided with a marked decrease in palm phytoliths, potentially indicating drier or colder climates. Nevertheless, this record rejects the notion that grass-dominated habitats spread in southern South America before at least the late Early Miocene (18.5 Ma). Instead, ingested grit from volcanic ash, rather than silica in grasses, may have been the proximal driver of hypsodonty in South America.