Clade-level analysis of body size and life history focused on turritelline gastropods (family Turritellidae) and venericardiine bivalves (family Carditidae), two taxa ubiquitous in the GCP throughout the Paleogene (21, 55). Body size of all available taxa in both groups was quantified through time, and we report both mean and maximum size measures for each (data file S3). For turritellines, body size was given as specimen lengths measured from collections at the Paleontological Research Institution (PRI) and our own field collections, corrected for the frequently missing apex (56). For venericardiines, we used centroid size based on nine homologous landmarks digitized in the lateral (internal) orientation from collections at the PRI, the National Museum of Natural History, the Mississippi Department of Environmental Quality, and our own field collections. In addition, to explore changes in GCP shell size more generally across the PETM, we measured the maximum dimension of specimens of the gastropod genera Athleta, Calyptraphorus, Natica, and Pleurotoma and the bivalve genus Tellina in the PRI collections on either side of the PETM (data file S3) and compared their size distributions using a t test of mean size for Paleocene versus Eocene taxa. Multiple species were measured for each genus (see data file S3).

Size is a function of growth rate and life span, and changes in these two important life history variables give rise to, and can be masked by, body size trends. To better understand the factors determining body size in turritellines and venericardiines, growth rates and life spans were evaluated through stable oxygen and carbon isotope analysis of shell material serially sampled along ontogenetic trajectories. We sampled two turritellines and one large venericardiine each from the BLM and BM, the units straddling the PETM, at high spatial resolution (data file S4). Taxa analyzed are closely related but were probably not direct ancestor-descendant pairs (21, 57). They are, however, the dominant species of their family in each unit. Turritellines were sampled around the spiral growth axis, and venericardiines were sampled from the umbo to commissure. Only one to two individuals from each formation were sampled, a consequence of the need for a large number of samples within an individual ontogenetic trajectory to constrain seasonal cycles. While a greater number of individuals would ideally be preferred, similar data from both turritellines (58) and venericardiines (59) in other settings demonstrate general consistency in size-age relationships among individuals of a species in a given unit.

Annual cycles were identified on the basis of regular deviations from the mean in oxygen and carbon isotope values along the growth trajectory. If growth is sufficiently fast and seasonal variation is large, the number of samples within any year is sufficient to recognize the expected annual sinusoid. However, slow or seasonally truncated growth combined with the muted range of seasonal temperature variation in low-latitude settings can make delineation of years difficult. While turritellines are fast-growing and so posed little problem, age determination in these venericardiine bivalves was especially difficult due to the limited resolution afforded by isotope data for both the aforementioned reasons. To minimize the problem, we sampled on the external surface of shells instead of on cross sections where the need for sufficient material for analysis limits the temporal resolution possible. This strategy, however, eliminated the possibility of identifying growth bands in cross section, which can help ascertain age if they can be shown to be annual. Ultimately, annual cycles in venericardiines were cautiously identified using variation in both oxygen and carbon isotope values, particularly by looking for runs of multiple samples that constrain cycles, but we concede a good deal of uncertainty in these assignments.

Samples were analyzed at the University of Michigan’s Stable Isotope Laboratory or at the University of Kansas Keck Paleoenvironmental and Environmental Stable Isotope Laboratory using a MAT 251 or 253 mass spectrometer coupled to a Kiel automated carbonate preparation system. All values are reported relative to the Vienna Pee Dee Belemnite standard, and precision of values was better than 0.05 ‰.

Note: The content above has been extracted from a research article, so it may not display correctly.

Please log in to submit your questions online.
Your question will be posted on the Bio-101 website. We will send your questions to the authors of this protocol and Bio-protocol community members who are experienced with this method. you will be informed using the email address associated with your Bio-protocol account.

We use cookies on this site to enhance your user experience. By using our website, you are agreeing to allow the storage of cookies on your computer.