Paper No. 8
Presentation Time: 3:15 PM

CONTROLS ON EXCEPTIONAL PRESERVATION IN CONCRETIONS


MCCOY, Victoria, Geology and Geophysics, Yale University, New Haven, CT 06520 and BRIGGS, Derek E.G., Dept. of Geology and Geophysics & Peabody Museum of Natural History, Yale University, 210 Whitney Avenue, P.O. Box 208109, New Haven, CT 06520, victoria.mccoy@yale.edu

Concretions are an important source of exceptional fossils (i.e., those preserving non-biomineralized tissues). In order to better understand the process of exceptional fossilization within concretions and its inherent biases, we scored 88 concretion-bearing sites for 12 variables and analyzed the interaction of these variables using qualitative logistic regression and multiple correspondence analysis (MCA). The regression revealed that only two variables correlated with the presence of soft-tissue mineralization: 1. Fine grained host lithology, which is thought to inhibit decay, and 2. Unchanging δ13C values through the concretion, which may indicate fast growth. Concretion sites with exceptional fossils were distributed randomly throughout the single cluster of concretion sites in the MCA-calculated variable space. This indicates that most of the variation in concretion formation is irrelevant to the presence or absence of soft tissue mineralization. While exceptional fossilization can occur in concretions in almost any environment in which they form, it will not necessarily do so. Exceptional fossilization within concretions is controlled by the same factors that determine its occurrence outside them: factors that inhibit decay and/or promote mineralization. Concretion formation, however, may promote the conditions required for exceptional preservation, in particular by early cementation and a consequent decrease in permeability. The most important variables determining the degree of exceptional preservation within concretions are the timing of nucleation and the rate of concretion growth relative to decay (neither of which can be readily measured as independent variables) which may differ in different settings.