Phylogeny of the cyclostephanoid diatoms: An investigation of their morphology and stratigraphy.

The phylogeny of the Cyclostephanoid diatoms was investigated. Although several species in this group are important indicators of water quality and biostratigraphy, this is the first formal systematic study of the group. Phylogenetic reconstruction was performed for exemplar taxa within the group. The investigation utilized frustule morphology and stratigraphy with stratocladistic and classic cladistic techniques. Ingroup taxa included recent and fossil specimens from Africa, Asia, North America, and Europe. Forty-one species were detailed using scanning electron microscope techniques. Twelve of these species lack formal taxonomic description and many of the other species have not been previously investigated with the electron microscope. A disproportionate number of fossil species from the Pacific Rim were included in the ingroup, because of the abundance and age of diatomites from this area. The oldest members of Cyclostephanoid diatoms are also present in these diatomites and the area was presumed to represent the paleogeographic origin of the group. Using phylogenetic systematics, the monophylly of major genera was tested. This test was accomplished by including a group of marine Thalassiosira species (the plicated Thalassiosira) in the cladistic analysis and designating these species as the outgroup. Results indicate Cyclotella is not a monophyletic group, but rather a collection of species classified by the presence of relatively pleisiomorphic features. The derived states for many of these features are character losses, which has lead to confusion in previous phenetic based classifications of the group. Proposed generic level classifications are presented, modifying the current classification of Cyclotella to reflect monophyletic groups. This taxonomic revision would require the description of at least one new genus and as many as three new genera. Stratigraphic data contributed to the cladistic analysis, modifying the solution produced by morphological features alone. Exclusion of the stratigraphic information and fossil taxa resulted in a set of solutions consistent with the solution produced using total evidence; however, the solutions produced with the limited data set were less resolved (missing contribution from fossil species) and failed to identify the polyphyletic status of Cyclotella (missing contribution from stratigraphy). These results demonstrate the utility of stratigraphy and fossil species in phylogenetic inference.Ph.D.Biological SciencesBotanyEarth SciencesPaleobotanyPaleontologyUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttp://deepblue.lib.umich.edu/bitstream/2027.42/132604/2/9977188.pd