TIMMERMAN: CROTALUSADAMANTEUS ON ORDWAY PRESERVE Table 7. Arboreal, terrestrial, and fossorial habitat use by eight eastern diamondback rattlesnakes at the Ordway Preserve, Putnam County, Florida. Numbers in parentheses represent percentages of the observations in that season. Arboreal' Terrestrial2 Fossorial3 Unknown4 "Winter" December-February 0 112 95 11 (0.0) (55.4) (44.6) "Summer" March-November 1 403 35 85 (0.2) (91.8) (8.0) SIn tree or bush 2 On surface ofground ' In burrow or hole below the ground * Snake was not visible due to thick vegetation-it could not be determined whether snake was on surface or below ground. 2 to 4 weeks. Klauber (1972) reported several instances when temperate-zone rattlesnakes were killed by early cold snaps as the snakes tried to make their way to dens. Winter Habitats and Den Sites.- Snakes used Mesic Hammocks, Xeric Hammocks, High Pine forests, and Lake Meadows as overwintering habitats. Old Fields, Swamp Forests, and Wet Prairies were not used. Of the 76 observations of burrow use, 94% of which were made in the winter (December through February), only 12 were gopher tortoise burrows. Four records were of unidentified holes, and 57 were identified as dens of the nine-banded armadillo (Dasypus novemcinctus) although it should be noted that this animal often commandeers and reshapes old Gopherus holes. Means observed that his study animals used gopher tortoise burrows, stump holes, or some other reliable shelter during winter. Ordway rattlesnakes often used dilapidated holes as a base of operations and were often visible within a meter of the burrow mouth. The cooler weather at Tall Timbers probably dictates that rattlesnakes choose good den sites that will be reliable during a frost. Ordway diamondbacks should be able to use less-protected den sites because of the slightly milder winters here. Snakes always used more than one winter site during this study. In a few cases, movements between these winter locations included changes in habitat. For example, M3 began the winter of 1986-87 in a Mesic Hammock. But, on 7 January he moved to a Xeric Hammock, and used four different sites within this vegetation zone. Then, on 21 January, he moved back to a Mesic Hammock and disappeared into a palmetto thicket, probably taking refuge in channels among the root systems. He did not reappear above-ground until 16 March. This same snake