Can atmospheric composition influence plant fossil preservation potential via changes in leaf mass per area? A new hypothesis based on simulated palaeoatmosphere experiments.

Atmospheric composition, particularly levels of CO2 and O2, impacts all aspects of life but its role in relation to plant preservation in the fossil record is largely unconsidered. Plants, angiosperms in particular, have been widely shown to increase leaf mass per area (LMA) under high CO2 conditions and decrease LMA in low CO2 conditions. Leaf thickness has long been known to be a contributory factor in preservation potential in the plant fossil record, with thicker leaves considered to have a greater recalcitrance than thinner ones. Therefore, any change in leaf density/thickness, through changes to LMA, could lead to an increased or decreased preservation potential of fossil leaves at times of elevated or decreased CO2, respectively. Additionally, the impact of changes to atmospheric O2 and to the atmospheric CO2:O2 ratio on LMA has not been previously considered in detail. This investigation examines the effect of simulated Mesozoic atmospheres, times of high CO2 and low O2, on LMA in a suite of gymnosperms that act as nearest living equivalents for common elements of Mesozoic floras. Exposure to high CO2 (~ 1500 ppm) led to a statistically significant (p < 0.001) increase in LMA in four out of 6 species, and exposure to combined high CO2 and low O2 (~ 13%) induced a statistically significant (p < 0.001) increase in LMA in all six species. The investigation also examined the effects of atmospheric composition on %N, a key plant trait known to co-vary with LMA under modern atmospheric compositions that provides information on plant function and relates to photosynthetic efficiency. Most species showed decreased %N in treatments with increased LMA in agreement with modern ecological studies and supporting the co-varying nature of LMA and %N regardless of CO2:O2 ratio. These findings suggest that atmospheric composition has a pronounced impact on LMA. Based on these results, we propose the hypothesis that atmospheric composition is an important taphonomic filter of the fossil leaf record. Further research is now required to test the significance of atmospheric composition versus other well-known taphonomic filters.