124 
John F. Pagels and Charles O. Handley, Jr. 
apparently had been lost and could not be re-examined. Thus, the true 
distribution of this shrew in western Virginia remained to be determined. 
Following our earlier work on Sore: c longirostris, Pagels conducted 
a more detailed field study of the distribution and ecology of shrews of 
all species in Virginia. He collected S. longirostris at many localities 
from which it had not been known and located previously collected 
specimens that add significantly to our knowledge of the range of this 
species in Virginia. Although we list and map all of the new localities, as 
an update of our earlier work (Pagels et al. 1982), our present emphasis 
is on the distribution of S. longirostris west of the Blue Ridge. 
MATERIALS AND METHODS 
We placed 1508 pitfall traps (16-oz. aluminum cans) at 140 localities, 
mostly along highways, to form irregular transects in all five 
physiographic regions of Virginia (Fig. 1). The cans, partly filled with 
water early in the study but later with a formalin solution to help 
preserve the specimens, were checked approximately bimonthly for 24 
months between 1983 and 1985. 
Four major cover types were sampled: old field, field-forest edge, 
mixed forest, and hardwood forest. Within a given habitat, traps were 
set irregularly in the best cover rather than at fixed intervals. Because 
this study included all species of shrews, collecting effort sometimes was 
directed toward a particular species. Thus, the intensity of sampling of 
the various cover types was not the same in all regions. 
Although we sampled 140 localities with pitfall traps, only 107 of 
them (52 in the Coastal Plain and Piedmont, and 55 in the mountains) 
were within the known range of Sorex longirostris. The other 33 
localities were above 610 m, the upper limit of S. longirostris at this 
latitude. The boreal habitats of these elevations are not likely to be 
inhabited by this austral shrew. Thus, in Table 1 we listed these 33 
localities as a separate division of the Ridge and Valley Province and 
did not include them in the distributional totals. 
COLLECTIONS 
Altogether we found 115 specimens of Sorex longirostris at 48 
localities in Virginia not listed in the compilation of Pagels et al. (1982). 
The 48 new localities are plotted in Figure 1, and are grouped by 
physiographic province in the following list of specimens (elevations in 
parentheses were estimated by Pagels from topographic maps). 
Because of our collecting technique, the specimens could only be 
prepared as skulls or kept entire in 70% alcohol (skulls of some of those 
