Revision of species of Minerisporites, Azolla and associated plant microfossils from deposits of the Upper Palaeocene and Palaeocene/Eocene transition in the Netherlands, Belgium and the USA

Species of the megaspore genus Minerisporites Potonié, megaspore apparatuses of species of the water fern Azolla Lamarck, and some associated organic-walled microfossils recovered from deposits of the Upper Palaeocene and Palaeocene/Eocene transition in the southern part of the Netherlands and neighbouring Belgium are redescribed on the basis of an examination of specimens under scanning and transmission electron microscopes. Originally studied about 40 years ago by S.J. Dijkstra, the re-examination has enabled emended diagnoses to be produced for six taxa: Minerisporites glossoferus (Dijkstra) Tschudy, M. mirabilis (Miner) Potonié, M. mirabilissimus (Dijkstra) Potonié, Azolla schopfii Dijkstra, A. teschiana Florschütz, and A. velus (Dijkstra) Jain and Hall. In addition, a revised description is provided for massulae of Salvinia Séguier that were originally thought to be megaspores and, hence, named by Dijkstra as Triletes? exiguus. The gross morphology and construction of the exospore of the species of Minerisporites are similar, but nevertheless sufficiently distinct for them to be maintained as separate taxa. Monolete microspores are preserved in hollows in the reticulate surface of some of the specimens of M. mirabilissimus. This is consistent with the presumed isoetalean affinity of Minerisporites. An apparent stratigraphic morphocline from M. glossoferus to M. mirabilis, suggested previously, is confirmed following our reassessment of their characteristics. The species of Azolla are all multi-floated, but they differ from each other in several ways, in particular with respect to the ultrastructure of the megaspore wall. They are also distinct from all other species that have been considered in sufficient detail for satisfactory comparisons to be made. The massulae of A. teschiana are described for the first time. The floats in A. velus are attached to the proximal part of the megaspore only by suprafilosal hairs. There are no maniculae. It is argued that these structures do not exist in either this or any other species of Azolla. Whole fertile fossil remains that demonstrate the nature and habitat of the living plants are known for all three genera. Hence, their reproductive structures can be used with confidence in palaeoenvironmental interpretation. The overall assemblage, which also includes a few specimens of the marsilealean megaspore Molaspora lobata (Dijkstra) Hall, undoubtedly reflects the presence of lowland aquatic and semi-aquatic vegetation.