DODD & FRANZ: UPLAND SNAKES


components. Ericaceous shrubs often are common understory species. Ferns and
sphagnum moss can form an extensive ground cover on wetter sites.

 Freshwater Marshes.-- Extensive wet prairies occur in four large solution
depressions associated with the Mill Creek basin. Certain of these marshes are
dominated by semi-woody species, such as swamp loosestrife (Decodon
verticillatus), fetterbushes (Lyonia lucida), Virginia willow (Itea virginica), and
buttonwood (Cephalanthus occidentalis), while others are composed of
maidencane (Panicum hemitomon) and various sedges. Both types of marshes
frequently have small to large localized stands of sawgrass (Cladium jamaicense)
associated with depressions in the peat. Fires probably have played important roles
in these marsh systems in the past, which probably helped to control invasive
woody species. Currently, Ordway Preserve managers are not burning these sites.


 METHODS

 Xeric Community Sampling


 In 1989, 100 individually numbered screen wire mesh double-opening funnel
traps (90 cm long by 18-25 cm diameter; see Fitch 1987) were placed at six upland
sites as follows: 31 traps in closed xeric (sand live oak) hammock (11 at the
Fennell homestead [Ordway Preserve site location 12]; 20 south of Enslow Lake
[Ordway Preserve site location 21]); 59 traps in sandhill (high pine) habitat (9 at
the Fennell homestead; 10 in the vicinity of Polecat Flats [Ordway Preserve site
location 19]; 10 in the vicinity of Dry Pond [Ordway Preserve site location 20]; 30
in the vicinity of Single Shot Pond [Ordway Preserve site location 23]); 10 traps in
open xeric (sand live oak) hammock (all in the vicinity of the McCloud homestead
[Ordway Preserve site location 22]). The locations of the sampling areas are shown
in Figure 1.
 In most cases, the traps were set along fallen trees and branches that formed
natural drift fences. At locations 19 and 20, traps were set along drift fences made
of 10 m sections of galvanized metal set in 4-pronged arrays (Campbell and
Christman 1982, fig. 1). The traps were covered with palmetto fronds to prevent
captured animals from overheating in the direct sun and to provide cover. The
traps were checked daily from April 4 through November 17 (23,800 trap nights)
between 0700 and 1200 h.
 Trapped snakes were returned to the laboratory. Prior to measurement, they
were cooled for 1-4 h depending on size. The following data were recorded: snout-
vent and tail length (in mm using a ruler), wet mass body weight (in g using a
Pesola spring scale), sex (using standard reptile sex probes to determine the