Calibration of mammoth (Mammuthus) dispersal into

North America using rare earth elements of PlioPleistocene mammals from Florida
Bruce J. MacFadden
a

, a,

and Richard C. Hulbert Jr.a

Florida Museum of Natural History, University of Florida, Gainesville, FL 32611, USA

Received 12 October 2007.

Available online 25 October 2008.

Abstract
The first appearance of mammoth (Mammuthus) is currently used to define the beginning of the Irvingtonian
North American Land Mammal Age at about 1.4 Ma. Thereafter, mammoth fossils are common and
widespread in North America until the end of the Pleistocene. In contrast to this generally accepted
biochronology, recent reports have asserted that mammoth occurs in late Pliocene (ca. 2.5 Ma) alluvium
from the Santa Fe River of northern Florida. The supposedly contemporaneous late Pliocene fossil
assemblage from the Santa Fe River that produced the mammoth specimens actually consists of a mixture
of diagnostic Blancan (late Pliocene) and late Rancholabrean (latest Pleistocene) species. Fossil bones and
teeth of the two mammalian faunas mixed together along the Santa Fe River have significantly different rare
earth element (REE) signatures. The REE signatures of mammoth are indistinguishable from those of
Rancholabrean mammals, yet they are different from those of diagnostic Blancan vertebrates from these
same temporally mixed faunas of the Santa Fe River. Thus, no evidence for late Pliocene mammoth exists
in Florida, and mammoth fossils remain reliable biochronological indicators for Irvingtonian and
Rancholabrean terrestrial sequences throughout mid- and lower-latitude North America.

Mammalian fossils and quaternary biostratigraphy

Abstract
Mammalian fossils are important biostratigraphic indicators in terrestrial Quaternary sequences. Their use in dating and
correlating deposits is based on two main attributes: (1) the presence or absence of particular species in fossil
assemblages; (2) the evolutionary stage of individual species shown by their anatomy. The changes through time in
these features, on which biostratigraphic usage depends, are the result of complex biological phenomena, including
speciation, adaptation, distributional shifts, and extinction. Contrary to common biostratigraphic assumptions, the pattern
of change is rarely simple, and moreover, varies from one species to another. Evolutionary trends show fluctuations of rate
and direction, and may vary in timing across the species' range. Constant remoulding of geographical distributions
produces complex patterns of presence/absence. A proper understanding of these phenomena is essential for the valid
use of mammals in biostratigraphy. The pattern of evolutionary and distributional change in each species must therefore
first be established, based on independently dated samples over a sufficient geographical range, before it can be used in
the dating or correlation of a new sample.

The isotopic ecology of late Pleistocene mammals in North America: Part 1.

Florida
Mammoths and mastodons are common in Pleistocene deposits, yet these proboscideans and many other animals
disappeared suddenly 10,000 years ago. In this study, we reconstruct the diets of proboscideans and associated
mammals through isotopic analysis of carbonate in tooth enamel apatite in order to test nutritional hypotheses for late
Pleistocene extinction. We analyzed specimens from six sites in Florida, ranging from full glacial (>21,000 BP) to late
glacial (14,750 to 10,000 BP) age. The oxygen isotope composition of mammalian apatite covaries with meteoric water
composition, which in turn varies with climate. Consequently, oxygen isotope analysis can be used to assess the potential
for time-averaging or mixing of specimens from different geographic regions within fossil assemblages. The carbon
isotope composition of an herbivore is controlled by the isotopic composition of the plants that it ingests. Carbon isotope
analysis reveals that mastodons ate chiefly C3 plants, presumably trees, shrubs and herbs, whereas mammoths
consumed chiefly C4 grass. Several nutritional hypotheses for late Pleistocene extinction entail the assumption that extinct
taxa had specialized diets. The resource partitioning and focused feeding preferences of Florida's proboscideans
corroborate this assumption, but they do not, in themselves, prove that nutritional stress was the cause of the late
Pleistocene extinction.

Article Outline
1. Introduction
1.1. Paleoecology and paleoclimatology of Florida
1.2. Carbon isotopes and vertebrate paleoecology
1.3. Oxygen isotopes: physiological controls and paleoclimatology
2. Materials and methods
2.1. Sample selection
2.2. Localities
2.2.1. Page-Ladson, Jefferson County

2.2.2. Hornsby Springs, Alachua County

2.2.3. Rock Springs, Orange County
2.2.4. Vero, Indian River County
2.2.5. West Palm Beach, Palm County
2.2.6. Cutler Hammock, Dade County
2.3. Sample preparation and isotopic analysis
2.4. Statistical analysis
3. Results
3.1. Carbon isotope analysis
3.2. Oxygen isotope analysis
4. Controls on isotopic variation in Pleistocene mammals in Florida
4.1. Inferences regarding mixing in fossil assemblages
4.2. Carbon isotopes and the diets of late Pleistocene herbivores
4.3. Seasonal changes in precipitation and diet
5. Conclusion
Acknowledgements
Appendix A: Isotopic data and percent C4 plants in diets for Pleisotocene mammals from Florida
References