29th Annual Meeting and Symposium of the
Desert Tortoise Council, February 20-23, 2004
Abstracts

Gopher Tortoise Evolution: East vs. West. A Possible Paradigm Shift

Richard Franz and Shelley E. Franz
Florida Museum of Natural History, University of Florida, Gainesville, FL 32611-7800

The fossil record of Gopherus begins with G. laticuneus more than 32 million years ago in the White River sediments of the Late Eocene of Colorado, Nebraska, South Dakota, and Wyoming. Large samples of G. laticuneus from a ranch near Toadstool Park in northwestern Nebraska show that it already had obtained many of the features that link it to more modern gopher tortoise taxa. This gopher tortoise achieved a large body size, and it occurred commonly with another large tortoise in the genus Stylemys. From this meager beginning, gopher tortoises radiated into a plethora of species in the Miocene and Pliocene in the West. Sometime during this period the two clades, one leading to G. polyphemus and G. flavomarginatus, and the other to G. agassizii and G. berlandieri, diverged into separate lineages. Molecular data suggest this split occurred 17-18 million years ago. Unfortunately, the fossil record provides little assistance in sorting the important events of this shift. When and where did this split occur?

Numerous characters have been used to differentiate members of each clade. The most defining characters, unfortunately, concern the head, neck, and front legs, which are rarely preserved as fossils. Bramble (1982) suggested that the peculiar anatomical features found in the members of the polyphemus-flavomarginatus (board-headed) clade are related to the mechanics of digging. These features were either not present or only weakly developed in members of the agassizii-berlandieri (narrow-headed) clade. Bramble (1982, Copeia (4):853-867) placed the latter clade in the genus Scaptochelys (also Xerobates of other authors) based on the absence of these features. Fortunately, one of the defining characters for the polyphemus-flavomarginatus clade is portrayed by a ventral strut that extends from the base of the first dorsal vertebra and attaches to the internal surface of the nuchal plate. Bramble considered the ventral strut as a method of strengthening the neck and body connection during burrowing activities. This strut is marked by a prominent bone scar on the posterior edge of the nuchal plate on closely related fossils, which in our opinion signals them as a specialized burrowing species in the polyphemus-flavomarginatus clade.

Traditional wisdom suggests that gopher tortoises (Gopherus) are recent invaders into the Gulf and Atlantic coastal plains from the West and the East has played little or no significant role in gopher tortoise radiation. However, a succession of more than 60 fossil sites with Gopherus polyphemus from the Late Pliocene (Blancan) and Early, Middle (Irvingtonian), and Late Pleistocene (Rancholabrean) in Florida, Georgia, Mississippi, and South Carolina and recent discoveries of Gopherus-like fossils from the Middle Oligocene (Whitneyan and Arikareean) and Late Miocene (Hemphillian) in South Carolina and Florida tell a different story. The populations of Plio-Pleistocene gopher tortoises from the Southeast best fit with G. polyphemus, although the earliest samples suggest much smaller forms existing in Florida two million years ago, then shifting to larger forms, often exceeding G. flavomarginatus, in the Late Pleistocene. All of these fossil samples from Florida show prominent scars on their nuchal bones, which not only document their taxonomic identity, but also signal their abilities to burrow. Based on this feature and the ecological requirements of modern polyphemus, we conclude that fossil gopher tortoises dug extensive burrows and lived in fire-prone droughty sites, similar to the sandhill, scrubby flatwoods, and scrub habitats that occur in the Southeast today. The presence of gopher frog (Rana capito) and Florida mice (Podomys sp.) fossils with gopher tortoise fossils in Late Pliocene and Pleistocene sites indicate that the unique obligate burrow commensal fauna was already in place in Florida two million years ago.

Several amazing mid-Tertiary fossils from the Southeast suggest that gopher tortoises may have been part of the southeastern coastal plain fauna for 28 million years. The most ancient Gopherus-like tortoise is from the Chandler Bridge Formation, near Charleston, South Carolina. This species, represented by a complete anterior half of the carapace, has many traits of the polyphemus-flavomarginatus clade, including the presence of the ventral strut scar on the nuchal. This fossil species apparently lived in a coastal site, possibly a barrier beach situation, similar to that occupied by gopher tortoise colonies today along Atlantic coast. Two other fossil tortoise taxa, both from Florida, also have gopher tortoise traits, but because of their fragmentary nature, we cannot ascertain with certainty their true identity until more material becomes available. Both fossil samples consist of epiplastral bones and several carapace and plastral fragments. We will attempt to place these ancient fossils in the context of the tortoise fossil record.

In summary, the fossil record, although spurious, suggests that the Southeast may have played a more vital role in the evolution of gopher tortoises than previously believed. We have known that a smaller relative of the modern Gopherus polyphemus suddenly appears in the Southeast two million years ago, bringing with it, not only its ability to burrow, but also its burrow commensal fauna. Did the smaller Late Pliocene form become the larger Late Pleistocene tortoise, the precursor of the modern populations? Did the large gopher tortoises of the Pleistocene disperse westward, then fragment becoming Gopherus flavomarginatus? Or vice-versa? Do the mid-Tertiary tortoises fit with the more widely accepted paradigms? The South Carolina fossil is second in age only to the oldest known species, Gopherus laticuneus. It, however, has the strut, and G. laticuneus does not. Does this indicate that burrowing may have evolved first in the East and spread westward? Much more work is necessary on the fossil tortoise record in North American before we truly understand the evolution of the group.