2004 Denver Annual Meeting (November 7–10, 2004)

Paper No. 6
Presentation Time: 1:30 PM-5:30 PM


DUNLAVEY, Tammy, Geology, SUNY at Buffalo, 876 Natural Science Complex, Buffalo, NY 14260, MITCHELL, Charles, Dept. of Geology, SUNY at Buffalo, 876 Natural Sciences Complex, Buffalo, NY 14260 and SHEETS, H. David, Dept. of Physics, Canisius College, 2001 Main St, Buffalo, NY 14208, dunlavey@buffalo.edu

Deformation acts as a taphonomic process accompanied by a loss of biological information. It is this loss of biological information that has burdened accurate taxonomic data recovery for paleontologists for decades. Deformation may enhance subtle features and deformed specimens may differ from undeformed material in shape and the expression of biological features. Hence, confidence levels based on deformed materials are always less than that possible with so called undeformed specimens. In the past paleontologists generally solved these alteration problems by concentrating on rare, well-preserved fossils that allowed them to eliminate the deformation enigma all together. Since the majority of fossils in deformed rocks nonetheless have undergone some degree of alteration in the features used for species discrimination, this is not a satisfactory solution. Fortunately, taxonomic and strain information can be recovered from these altered fossils. In particular, those fossils with bilateral symmetry may be reconstructed by analyzing the strain and restore (retrodeforming) the fossils to their original shape. The primary goal of this poster is to illustrate the pitfalls of the former solution and to introduce a new retrodeformation methodology, which applies a geometric morphometric landmark-based computer program as a plausible resolution to this deformation enigma. A working retrodeformation procedure was tested on material deformed in a mathematical simulation of the strain process produced promising shape retrodeformation results. However, when we applied the process to actual field data from a test set of graptolites with isograptid symmetry and evident shear strain, the procedure unexpectedly did not produce uniformly reduce asymmetry. We are now investigating the degree to which the residual asymmetry is a product of additional deformation associated with flattening, original biological asymmetry, and measurement error. Future work will attempt to isolate these processes by the use of flattened but not sheared and non-flattened material of the same taxa to partition variance in asymmetry among these components.