Grimsdalea magnaclavata Germeraad, Hopping & Muller: an enigmatic pollen type from the Neogene of northern South America

Abstract

Grimsdalea magnaclavata was first described by Germeraad, Hopping and Muller from Miocene and Pliocene deposits of northern South America, specifically Colombia and Venezuela. The specific epithet, ‘magnaclavata', was selected by the authors to draw attention to the distinctive clavate processes of the pollen. A revised taxonomic description is supported by illustrations of fossil specimens from outcrop and well sections that show the range in morphology not previously recorded in the original description; the key difference is the wide variation in process morphology from clavae to clavae with interspersed echinae. Based on the new observations of the pollen morphology we conclude that the records from Cretaceous and Paleogene sediments in North and West Africa are invalid. G. magnaclavata is clearly restricted to northern South America from the Miocene to Pleistocene. The parent plant of G. magnaclavata probably grew in upper coastal plain or around swamps, commonly associated with shrubs and herbaceous savannah plants. The botanical affinity of G. magnaclavata has always been in question. Based on a review of pollen from several palm genera presently extant in northern South America we consider that the nearest living relatives, if indeed G. magnaclavata is an ancient palm, are the genera Mauritia and Mauritiella. The basis for this conclusion is the distinctive foot layer morphology of the processes in the fossil and the modern examples.

Keywords:

• Grimsdalea magnaclavata
• fossil
• extant
• palm
• pollen
• Paleogene
• Neogene
• Pleistocene

Acknowledgements

James Solomon and the staff of the Missouri Botanical Garden graciously allowed us to collect pollen samples from the herbarium (MO) and provided overnight accommodations. We thank Susan Jarzen for her assistance in making the collections of palm pollen from the Missouri Botanical Garden and for her careful reading of several versions of the manuscript. We also extend our appreciation to two reviewers and to Madeline Harley and Alan Graham for useful discussions on botanical affinities.