USING LEAF EPIDERMIS TO UNLOCK THE ANCIENT FOREST RECONSTRUCTION ENIGMA
This phytolith based LAI proxy, termed rLAI, does not take into account the taxonomy of the phytoliths, as it is impossible to identify which taxa produced them. Instead, it is assumed that cell morphology is environmentally controlled, rather than phylogenetically determined. Here we test whether there is a phylogenetic signal in adaxial pavement cell morphology within angiosperms from ~900 taxa available in the Cuticle Database (http://cuticledb.eesi.psu.edu/). We also compare pavement cell morphology to growth form, phenology, leaf size and Leaf Mass per Area (LMA) data. We find that: 1) there is no phylogenetic signal for cell area and degree of undulation among angiosperms, 2) tree species have smaller cells than other growth forms, 3) there is no difference in cell morphology between deciduous and evergreen species, 4) epidermal cell size is not correlated to leaf size and 5) LMA cannot be predicted from epidermal cell size or density. Additionally, we present preliminary cell morphologic data from dispersed leaf cuticles that suggest LAI can also be predicted from organically preserved leaf cuticles. 1 Dunn, R.E., Strömberg, C., Madden, R., Kohn, M., and Carlini, A., 2015, Linked canopy, climate, and faunal change in the Cenozoic of Patagonia: Science, v. 347, p. 258–261.