Research Training Program

Stephen Schellenberg
University of South Florida
Tampa, Florida

Douglas Erwin, Ph.D.
Supervising Scientist
Department of Paleobiology

"The RTP broadened and deepened my appetite for science by providing both an overview and a personal, in-depth, expedition into the wide breadth of systematic research."

Preliminary Phylogenetic Analysis of the Late Paleozoic Gastropod Family Pseudozygopleuridae with a Comparison to Current Superspecific Taxonomy

This research focused on elucidating the phylogeny of the Late Paleozoic gastropod family Pseudozygopleuridae and consisted of two studies. In the first study, the text, diagrams, and dichotomous keys of J. Brooks Knight's 1930 Pseudozygopleurinae monograph were assumed to be a proxy for the taxonomic paradigm under which Knight created the genera and subgenera of the family. These data were transformed into a subgeneric-level character matrix and examined cladistically to determine the degree to which Knight's key taxonomic characters provide phylogenetic information. Consensus trees produced a large polytomy at the generic level. The only exception was an 86% supported monophyly of the Cyclozyga and the Hermizyga subgenera with the synapomorphic characters of obliquely reversed and whorl restricted collabral ornamentation. Within this clade the Hemizyga subgenera clade/polytomy was differentiated from Cyclozyga by the synamorphy of both collabral and spiral ornamentation on Hemizyga subgenera. The large number of equally parsimonious primary trees and their secondary consensus trees indicate that characters utilized by Knight to circumscribe species provide little to no phylogenetic information. This is dependent on the worker's willingness to infer relationships based on non-strict consensus of primary trees.

In the second study a search for informative synapomorphies in the Pseudozygopleuridae was performed using twenty representative species of genera and subgenera from the NMNH Paleobiology Department collections. Forty-three characters and their states were constructed and examined cladistically. A survey of all trees and character paths showed only the P. (Pyrogzyga) species to form a monophyletic group in total agreement with Knight's taxonomic nomenclature. Characters supporting this clade were a linear outline in lower shell whorls and adpressed suturing. The remaining genera are widely dispersed throughout both trees with fewer synapomorphies occuring towards and increasing homoplasies away from both tree's roots.

The plethora of homoplasy and general paucity of good characters in both analyses currently prevent any confident resolution of evolutionary relationships within the Pseudozygopleuridae. In their own way, however, both characters lacking phylogenetic information were utilized to construct a majority of the current taxonomy. The second study of individual species reveals that many characters exhibit a high degree of homoplasy and few monogeneric clades. Thus it is logical to develop and examine new characters. It is unclear as to where such characters will be derived as those employed were believed to cover the spectrum of morphological variety.

It is possible that relationships within the Pseudozygopleuridae cannot be resolved, cladistically or otherwise, now or in the future. Permian gastropod assemblages generally consist of fewer than five species of a particular genus; Knight circumscribed over thirty taxa to P. (Pseudozygopleura) from two Pennsylvanian assemblages alone. A high degree of phenotypic plasticity in the employed characters would explain both an oversplitting of taxa by Knight and the current unresolvability of the group's clade.

This research was supported by a grant from the National Science Foundation Research Experiences for Undergraduates Program ( Award: DIR-9200203).