158 Mary K. Clark, Michael S. Mitchell, and Kent S. Karriker 
short pocosins. Tall pocosins are characterized by a taller under- 
story and are generally dominated by pond pines. 
To establish baseline data on pocosin mammals and to 
describe the changes in these communities associated with intensive 
forest management, M.S.M. and K.S.K. selected 15 stands in 
natural and managed pocosin communities for small mammal 
sampling. Three stands each represented five treatments: three man- 
aged habitats — open canopy, closed canopy, and thinned — and two 
natural stands — short pocosin and tall pocosin. The nine managed 
stands were in pine plantations on the periphery of the Croatan 
National Forest (on Weyerhaeuser Company land), and the six natu- 
ral pocosin areas were within the Croatan National Forest. 
One short pocosin stand was in the interior of the Great Lake 
Pocosin, and one tall pocosin stand was on its periphery. The high- 
est elevations in the National Forest are in the Great Lake Pocosin, 
which has no history of anthropogenic modification and is one of 
the largest contiguous expanses of short pocosin in the state. 
Small mammal trapping — Trapping was conducted for three field 
seasons: summer 1991, winter 1992, and summer 1992. Snap traps 
and pitfall traps were used each field season. Snap traps were ac- 
tive for 5 consecutive nights; pitfall traps were active 7. One-hun- 
dred snap traps (Museum Specials and Victor rat traps), baited with 
a mixture of peanut butter, rolled oats, and raisins, were set in 
each stand. There were five trap-lines per stand. In the first two 
field seasons pitfall traps were located at the end of each trap-line. 
Pitfalls with drift fences were added in the interior of each stand 
in the last field season (see Mitchell 1992 for details). 
Vegetation sampling — Vegetation data were collected to 
provide a context for interpreting faunal community structure. 
Parameters for overstory, shrub layer, herbaceous vegetation, and 
fallen dead material were evaluated for each stand (see Mitchell 
1992 and Karriker 1993 for details). 
Specimen identification — Specimens collected were deposited in 
the North Carolina State Museum of Natural Sciences (NCSM) where 
they were identified and were placed in the research collection as 
vouchers. Skull measurements used by Wetzel (1955) to separate 
subspecies of S. cooperi were compared to measurements taken from 
the skulls of two specimens (NCSM 6778 and NCSM 7190) col- 
lected in the National Forest. The skull of the other specimen (NCSM 
7191) was that of a juvenile and was not compared because Wetzel 
analyzed only adults. 
