Variability of stomatal density and index in the Upper Permian conifer Quadrocladus Mädler—a taphonomical case study

Abstract

Changes in atmospheric CO 2 concentrations over the geological past have attracted a considerable amount of scientific interest during the last decade. Many authors have demonstrated, that for some extant plants, stomatal densities and indices are negatively correlated with atmospheric CO 2 concentrations. Similarly, changes in stomatal densities and indices in fossil plants are considered to be reliable proxies for changes of atmospheric CO 2 concentrations in the geological past—not only in the Cenozoic but also in the Mesozoic and Palaeozoic. However, systematic investigations on the variability of these parameters in most Mesozoic and Palaeozoic plants are still lacking. In this study we investigate the inherent variability of both parameters in the conifer genus Quadrocladus Mädler from the Upper Permian of Europe. Conifers are generally not considered to be valuable taxa for determining SD and SI as palaeoatmospheric CO 2 proxies, because of their regular arrangement of stomata in rows. Quadrocladus is one of the few Late Palaeozoic conifers which shows a more uniform distribution of stomata on both surfaces, and no such distribution patterns as exhibited by other conifer taxa. Within an individual needle of this taxon, variations of SD and SI increased almost consistently from the base (100%) to the apex (169%) of the needle (both surfaces). In cuticles of Quadrocladus from fluvial deposits from Culmitzsch (Thuringia, E-Germany), the overall range of both parameters was smaller than in cuticles from marginal marine beds of Frankenberg-Geismar (NW-Hesse, W-Germany). This difference is probably due to the wider variety of (micro-)climatic and edaphic conditions experienced by material from the latter locality. Cuticles from Upper Permian deposits of northern Italy, showed no significant changes in SD, whereas SI showed a significant excursion towards higher values within the investigated profile. Mean values from northern Italy were slightly lower than values from the German deposits, which could be interpreted as an elevated atmospheric pCO 2. However, our data suggest that taphonomic distortions (fragmentation of individual needles, mixing of material from different micro-habitats, time averaging) of the data, have to be taken into account when SD and SI of fossil plants, with no closely related living relatives (i.e. same genus or even species), are used as palaeoatmospheric or -climatic proxies, especially when only a small number of cuticles is available for investigation.