Lycopsids from the Glossopteris flora: contributions to poorly understood Permian Gondwanan plants

Abstract

The Permian witnessed a transition from an icehouse to a greenhouse world. Elucidating the vegetation response to that transition is a remarkable goal, but only achievable if we have a sound understanding of the plants that thrived during those times. Here, I focus on the problematics of the lycopsids from the Permian Glossopteris flora of Gondwana. The fossil record of these lycopsids comprises mainly fragmented remains of sub- to arborescent forms, and little is known about them. This thesis includes four papers on these plants. Publication 1 deals with late Palaeozoic lycopsid fossils from the Paraná Basin in South America. It delimits their updated stratigraphical distribution in the basin and provides a baseline for future research on these lycopsids. Publication 2 deals with Azaniadendron Rayner from the Guadalupian of South Africa. Its reproductive structures are re-interpreted as arranged in fertile zones, and evidence for a ligule is reported, solving, at least for this genus, a dilemma about the presence of this feature in lycopsids from Glossopteris flora. Also, its leaf cushion variations are described, using an approach that may help to overcome the challenges in reliably identifying and circumscribing lycopsid fossil-taxa represented by fragmented axes. Publication 3 describes lycopsid axes from the Cisuralian of Brazil, using methods that include a vinyl polysiloxane casting technique, which revealed their epidermal features. Based on these fossils, we proposed the new fossil-species Nothostigma sepeensis Spiekermann et al. This paper reveals how the lack of studies of long-ignored remains, has been obscuring the lycopsid diversity of the Glossopteris flora. Publication 4 re-examines an anatomically preserved axis from the Kungurian of Brazil first interpreted as an Euramerican Sigillariaceae. The characters of this specimen indicate that it is not a lycopsid but a member of Cycadales. Based on its set of characters, we proposed the new-fossil genus and species Iratinia australis Spiekermann et al. It is the oldest known anatomically preserved cycad axis. There remains no evidence of sigillarians or any other typical Euramerican arborescent lycopsid in the Glossopteris flora. Together, these papers bring important novelties on the lycopsids from the Glossopteris flora and push the research on these plants forward. However, we are still far from understanding these plants. Unknowns, such as their growth architecture and systematic position, can be solved only with the discovery of more complete specimens.