128 
John F. Pagels and Charles O. Handley, Jr. 
In the Appalachians S. longirostris seems to be uncommon and to 
be found only at relatively low elevations. Published records reveal a 
general south-to-north gradient of the upper limits of its distribution: 
North Carolina 
Tennessee 
Kentucky 
West Virginia 
Macon Co. 762 m 
Sevier Co. 488 m 
Knox Co. 335 m 
Roane Co. ca. 305 m 
(Gentry et al. 1968) 
(Komarek and Komarek 1938) 
(Caldwell and Bryan 1983) 
(French 1976) 
These localities are all on the west slope of the Appalachians, in the 
Mississippi drainage. Our records from extreme southwestern Virginia, 
with elevations of 427 to 457 m, fit into this gradient. Apparently a 
different south-north gradient, displaced a little to the north, operates 
on the east slope in Virginia: 
Montgomery Co. 
Blacksburg 
610 m 
Rockbridge Co. 
Vesuvius 
457 m 
Greene Co. 
Stanardsville 
411 m 
Rappahannock Co. 
Chester Gap 
411 m 
Page Co. 
Luray 
366 m 
Warren Co. 
Front Royal 
305 m 
We did very little sampling in the Blue Ridge, and most of the 
capture sites there, all in the foothills, were near ones that we had 
reported earlier (Pagels et al. 1982). Nevertheless, it must be assumed 
that the Blue Ridge represents a formidable barrier to distribution of 
Sorex longirostris. Gaps at the Roanoke and James rivers, Chester Gap 
at the head of the Rappahannock River, Manassas Gap at the head of 
Goose Creek, and the Potomac River Gap provide the only access for 
this shrew to the lowlands behind the mountain range (Fig. 2). 
We now have evidence that all of these passages except the Potomac 
River Gap actually have been used by S. longirostris — the Roanoke 
River Gap to Blacksburg, the Roanoke or James gaps (or both) to 
Lexington and Vesuvius, and Chester Gap or Manassas Gap (or both) 
to Front Royal and Luray. Specimens actually were taken in Chester 
Gap and in Manassas Gap (at Linden). 
Although there can be no doubt that Sorex longirostris has used 
the gaps in the Blue Ridge to gain access to the Valley of Virginia, it is 
not clear why its distribution in the Valley is so limited. In this area 45 
of the pitfall localities were below 610 m, the highest elevation at which 
S. longirostris has been found in Virginia (Fig. 2). Within the elevational 
limits represented by the highest point where this shrew was found in 
each basin, there were 23 pitfall localities. Sorex longirostris was caught 
at only 5 (22%) of the 23 localities (2/8 of these were in fields, 3/13 in 
ecotones, and 0/2 in forest). The species has reached the head of the 
Roanoke River and crossed over the divide, barely into the extensive 
