Mixson’s Bone Bed

University of Florida Vertebrate Fossil Locality LV009


About 1.8 miles (2.9 km) northeast of Williston, Levy County, Florida; 29.41° N, 82.43° W.


Note that in the pre-1980 references cited below, the age of the site is usually given as Pliocene. Hemphillian (and Clarendonian) fossil localities were then correlated with the Pliocene Epoch of the standard geologic time scale. As radioisotopic dating of terrestrial and marine sedimentary rocks became widespread, the correlation between the time scale and North American Land Mammal Ages (NALMAs) were adjusted.

Basis of Age

Vertebrate biochronology. The early mylodontid sloth Thinobadistes is limited to the early Hemphillian interval. The combined presence of the species Amebelodon floridanus, Calippus hondurensis, Teleoceras proterum, and Aphelops malacorhinus favor a Hemphillian 1 age instead of Hemphillian 2, as does the absence of taxa whose first appearance defines the start of the Hemphillian 2 interval: Enhydritherium, Indoarctos, and Machairodus.


Leidy (1884) quoting from a letter written by J. C. Neal: “…the fossils were discovered in a bed of clay, occupying a ridge in the pine forest. They occurred over an irregular area of one hundred feet long by thirty feet wide, and were dug from variable depths of seven feet to the bed-rock.” In the second report on the fauna, Leidy (1885) noted that Neal had later informed him that the deposit now extended across an area of 100 feet by 50 feet and that the bedrock was Eocene limestone. During excavations in 1886 by L. C. Johnson of the United States Geological Survey “…trenches were cut in the [clay] bed to the bottom rock from two and a half to six feet deep, and that the bones were found distributed abundantly without order, through the clay from top to bottom. The bones especially the larger ones, are generally much broken, though the original texture is mostly preserved, and they exhibit no trace of being rolled or water worn” (Leidy, 1887). These clay deposits and similar ones in Alachua County became known as the Alachua Clay (Matson and Clapp, 1909). This unit is no longer formally recognized by the Geological Survey of Florida.

Depositional Environment

Massive clay beds usually indicate some type of quiet water environment, such as a lake. Although the reports on the site mostly focus on the mammals, fossils of turtles and alligator are common. There are no reports of marine vertebrates as is the case at the Love Site and McGehee Farm. Among sites in the region of similar age, the most similar in terms of geology and ancient environment seems to be the Emathela and Moss Acres Racetrack Site.


Excavation History and Methods

According to Leidy (1884), the initial collections were made by Dr. John C. Neal of Archer on the plantation of J. M. Mixson (sometimes spelled Mixon, consult this website for information on variations in spelling of this name) near Williston. Neal sent the fossils to the Smithsonian, which forwarded them to Leidy for study. Recognizing their significance, Leidy requested that the United State Geological Survey investigate the site, so in either 1885 or 1886, W. H. Dall and L. C. Johnson of the survey worked at the site and greatly increased the number of recovered fossils. The last work in this initial phase of excavation was done by famous field paleontologist John B. Hatcher in 1890. All of the specimens from this first excavation phase are housed at the Smithsonian Institution’s National Museum of Natural History in Washington, DC.

The second major and last excavation period at Mixson’s Bone Bed was in the late 1930s and 1940s (excluding the war years). Crews led by Ted Galusha collected the site for Childs Frick (see Galusha, 1975). Following Frick’s death, his entire collection, including those from Mixson’s Bone Bed, were transferred to the American Museum of Natural History in New York and made available for study. The Frick crews used more modern techniques than previous collectors, such as plaster jackets, and were thus able to successfully collect more complete specimens, such as entire skulls. However, these were almost always badly crushed. The total number of specimens in this collection is unknown, but those of the rhino Teleoceras proterum were stated to be in the “thousands” by Prothero (2005:117).

Between these two major periods of excavation, E. H. Sellards of the Florida Geological Survey briefly collected at the site for a few days in October 1914. This resulted in the recovery of about 165 specimens. The Florida Geological Survey collection also has 7 specimens found in 1929 by E. Mixson.


Mixson’s Bone Bed was the first major vertebrate fossil site to be found in Florida and the the first major Neogene land vertebrate site in the eastern United States. Its fossils were initially studied by Joseph Leidy of Philadelphia, the preeminent vertebrate paleontologist of the 19th century in the United States. By the 1880s, Leidy’s research in paleontology had waned due to his antipathy to the “bone wars” of Cope and Marsh. But fossils from Florida rekindled his interest in the subject. As was his typical style, Leidy published a series of brief articles on fossils from Mixson’s in the Proceedings of the Academy of Natural Sciences of Philadelphia, that was to lead up to a more detailed monograph (Leidy, 1884, 1885, 1886a, 1886b, 1886c, 1887). But that major work was unfinished at the time of his death in 1891. Frederic Lucas of the Smithsonian completed that study (Leidy and Lucas, 1896).

Leidy named nine new species based on fossils from Mixson’s Bone Bed, two rhinoceroses, two hipparion horses, three camelids, one gomphothere proboscidian, and one suid. The latter he quickly realized was an error, and was in fact the lower tusk-like incisor of a rhino. Of these, six are still regarded as valid species and now known from other early Hemphillian sites in Florida such as McGehee Farm, Haile 19A, and Tyner Farm. As is typical of faunas of this age, the short-legged rhino Teleoceras is more common than the long-legged rhino Aphelops. The other very common mammals at Mixson’s are the shovel-tusked gomphothere Amebelodon floridanus and giraffe-camel Aepycamelus major. For over a century, the generic affiliation of Leidy’s two hipparion horses from Mixson’s remained controversial. Hulbert (1988) demonstrated that both belonged in the genus Cormohipparion, as Cormohipparion ingenuum and Cormohipparion plicatile. Both are now known from many Miocene Florida localities.

There were several other species present at Mixson’s that Leidy (1884, 1885, 1886a) mentioned but did not formally name or figure because of inadequate specimens. These included a tapir, a calcaneum of a large ruminant artiodactyl, an astragalus of a sloth which he called Megatherium, a fox-sized carnivore, and an alligator or crocodile. To these, Leidy and Lucas (1896) added a pond turtle and a garfish but noted the absence of Carnivora, so the previous report of a fox-sized species may have been an error. Several of these were described by later workers. Oliver P. Hay (1908) of the Smithsonian named the pond turtle Pseudemys caelata. Hay (1919) later named the garfish fossils as a new species of alligator gar (genus Atractosteus), but it is not regarded as distinct from the living species Atractosteus spatula. In the same paper, Hay (1919) described a new genus and species of mylodont sloth from Mixson’s on an astragalus. It would be uncharacteristic of Leidy to mistake an astragalus of a mylodont sloth for that of a megathere, but the fact that both were identified from the same bone supports this conclusion, as does the fact that there have never been any records of megathere sloths from the Hemphillian of North America, and there is only a single sloth astragalus in the Smithsonian collection from Mixson’s Bone Bed.

Other paleontologists, most notably Simpson (e.g., 1929) took issue with Hay’s mylodont ground sloth from Mixson’s, Thinobadistes segnis. They questioned its provenance and age, because at the time there were no other records of mylodont sloths of this age in North America. It was assumed to be an intrusion from the Pleistocene. Fortunately, the Frick Field crews later recovered numerous fossils of Thinobadistes from Mixson’s Bone Bed, included several skulls (Webb, 1989). They showed that it was very distinct from the Pleistocene mylodont Paramylodon, and instead represented one of the earliest-known Neogene dispersal events of a land mammal from South America to North America (Webb, 1989). The American Museum of Natural History donated enough bones of Thinobadistes segnis from Mixson’s Bone Bed to the Florida Museum of Natural History to construct a mounted skeleton, which has been on public display for almost 40 years. It is the only such specimen of its kind in the world.

Although fossils of Carnivora are either absent or very rare in the Smithsonian collection from Mixson’s, the Frick Collection contains specimens of two species of borophagine canids (Wang et al., 1999) and the mustelid Hoplictis (Baskin, 2005). The Florida Geological Survey collection includes the proximal end of a carnivore femur.

As both an alligator and a crocodile are known from the late Miocene of Florida, either or both could occur at Mixson’s Bone Bed. All of the specimens in the Florida Geological Survey collection are alligator, based on ridged osteoderms and relatively short, blunt teeth. No specimens of birds were collected in the first phase of excavations at Mixson’s, but the Frick collection contains a few species representing two grebes and a stork (Becker, 1985, 1987).


This material is based upon work supported by the National Science Foundation under Grant Number CSBR 1203222, Jonathan Bloch, Principal Investigator. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.

Copyright © Florida Museum of Natural History, University of Florida