Distribution and Status of the Southern Bog Lemming, Synaptomys cooperi, in Southeastern Virginia

The Dismal Swamp subspecies of the southern bog lemming, Synaptomys cooperi helaletes, was named based on specimens collected during the 18951898 biological surveys conducted in the Dismal Swamp by the US Department of Agriculture. Unknown in the 20 th Century until re-discovered in 1980, this small boreal rodent was believed to be restricted to the Great Dismal Swamp ofV irginia and North Carolina where the cool damp conditions had permitted it to survive during the Holocene. However, field studies conducted since 1980 have revealed southern bog lemmings to be widespread throughout southeastern Virginia, with populations encompassing an area of more than 3300 km 2, including the cities of Virginia Beach, Chesapeake, and Suffolk, and Isle of Wight County. Lemmings were present on 38 of 165 (23%) pitfall-trapping sites; their frequency was much greater in prime habitats dominated by grasses and sedges on damp organic soils. Thus, southern bog lemmings are distributed widely in southeastern Virginia and, where present, they often are among the most numerous species of small mammal. INTRODUCTION The southern bog lemming, Synaptomys cooperi, distributed from Kansas and Nebraska northward through Minnesota and Manitoba, eastward through Canada, and southward into the Appalachian Mountains of North Carolina and Tennessee (Hall 1981 ), is one of the most enigmatic small mammals in North America. In some Midwestern states, highly trappable and high-density populations coexist with prairie voles in me sic or xeric grassland habitats (Kansas: Gaines et al. 1977; Illinois: Beasley and Getz 1986; Indiana: Krebs et al. 1969). In other permanently wet sites where herbivorous potential competitors often are absent, however, southern bog lemmings are difficult to trap. For example, isolated relic populations associated with permanently flowing springs (now incorporated into state-run fish hatcheries) are known from Meade County in southwestern Kansas and Dundy County in southwestern Nebraska. Other relic populations are believed to be restricted to the Pine Barrens of southern New Jersey and to the Dismal Swamp of southeastern Virginia and adjacent North Carolina. Thus, populations of this small stocky rodent with short tail and. tiny ears are highly patchy in both space and time. For example, in Douglas County in eastern Kansas, where generations of mammalogists have been trained at the University of Kansas since the 1920s, grassland populations existed for about four years starting in the middle 1920s (Lindale 1927, Burt 1928), then disappeared, reappeared in the middle 1940s, disappeared, and then reappeared in the mid-l 960s, since when they have persisted (Rose et al. 1977, Norman A. Slade, University of Kansas, pers. comm., October 2005). Understanding its ·spatial distribution is made difficult because 154 VIRGINIA JOURNAL OF SCIENCE Synaptomys cooperi often is reluctant to enter live traps. For example, Connor (1 959) caught only 38 bog lemmings during four years of study in the swampy habitats of the New Jersey Pine Barrens. By contrast, other populations are readily trapp able . Hundreds of S. cooperi were routinely trapped in two different kinds oflive traps (Rose et al. 1977) in damp and dry oldfields in eastern Kansas, where they reached densities of 42-65 per hectare (Gaines et al. 1977, Gaines et al. 1979). Clearly the name "bog lemming" is misleading because Synaptomys is not restricted to bogs or even to damp places. Synaptomys has been reported from areas of woody vegetation (Hamilton 1941, Coventry 1942, Connor 1959), moist grassy areas (Howell 1927, Stewart 1943, Smyth 1946, Burt 1928, Getz 1961), and from dry, southfacing grassy fields, such as in eastern Kansas (Gaines et al. 1977, Rose et al. 1977, Gaines et al. 1979). First described in 18 5 8 from specimens taken near Jackson, New Hampshire (Hall 1981), the generic name was given because Baird believed it to be a link(= synapse) between the lemmings (Lemmus) and the true mice(= mys). In 1895, investigators from the US Biological Surveys, led by A. K. Fisher, collected southern bog lemmings from cane brakes near Lake Drummond in Virginia's Dismal Swamp which Merriam (1896) described as a new species, Synaptomys helaletes. However, in his revision of the genus, Howell (192 7) demoted the tax on to a subspecies, S. cooperi he la Zetes, a decision accepted by Wetzel (1955) in his taxonomic study of S. cooperi. More recently, Wilson and Ruff (1999) recognize seven subspecies, including the isolated forms in Kansas, Nebraska, and the Dismal Swamp region of Virginia and North Carolina. Fisher collected other southern bog lemmings from the Dismal Swamp as late as 1898, but none was taken thereafter, despite the efforts of several investigators , including Charles 0. Handley, Jr., Smithsonian Curator of Mammals, who trapped some of Fisher's sites in 1953, and in other years and places, all without success. Handley (1979) and others (Meanley 1973, Taylor 1974) speculated that since no specimens had been collected since 1898, the Dismal Swamp subspecies might be extinct. However, Rose (1981 ), using pitfall traps placed under power lines in the northwest corner of the Great Dismal Swamp National Wildlife Refuge (GDSNWR), caught 13 specimens from three locations in 1980, laying to rest doubts about its existence. During the 1980s and early 1990s, my students and I conducted survey trapping at over 100 sites throughout southeastern Virginia for the Dismal Swamp southeastern shrew, Sorex longirostris fisheri, then a federally listed mammal; the southern bog lemmings reported here were taken in those same collections. These studies have revealed the Dismal Swamp subspecies, Synaptomys cooperi helaletes, to be widespread in appropriate habitats throughout southeastern Virginia, with populations extending west of the Dismal Swamp at least through Isle of Wight County. METHODS Both live and pitfall traps were used in our studies, with the latter being used more extensively. Systematic live trapping was conducted in the open habitats under a 40-m wide power line in the northwestern corner of the GDSNWR (Stankavich 1984 ). Fitch live traps (Rose 1973), set at 7.6-m intervals in two rectangular grids (0.38 and 0.40 ha), were tended for two days every two weeks from October 1980 to February 1982 . BOG LEMMINGS IN EASTERN VIRGINIA 155 Other live trapping in the following two decades, conducted throughout the region in a range of habitats, has yielded only one other Synaptomys with live traps, except for an (unpublished) study conducted by L. J. Ford in Suffolk during 1987-1988. Most information on distribution and relative abundance comes from pitfall traps set on 0.25-ha grids in a range of habitats in southeastern Virginia (Rose et al. 1990). Placed at 12.5-m intervals on a 5 X 5 grid, each pitfall trap was a #10 tin can placed in the ground flush with the surface and partly filled with water. Earlier studies (e.g., French 1980) had shown that southeastern shrews (and to a lesser extent, southern bog lemmings) are rarely taken in live or snap traps, necessitating the use of pitfall traps to collect distribution and status information on these species. In the initial study, funded by the Office of Endangered Species (Rose 1983, Everton 1985), 37 pitfall grids were set in a range of habitats centering on the GDSNWR. A later study (Padgett 1991 ), funded by the Virginia Department of Game and Inland Fisheries, added 29 grids, mostly placed at greater distances from the GDSNWR in an effort to learn the geographic extent of distribution of the Dismal Swamp southeastern shrew. Another 85 pitfall grids were set at a variety of sites in the region in surveys conducted between 1986 and 1995. Finally, current information on the western limit of distribution comes from a study conducted in 1992 on 14 grids set in the open habitats under powerlines in Isle of Wight County (Rose 2005). Specimens collected in pitfall traps were returned to the lab, measured, weighed and evaluated for reproductive condition, and then saved (mostly as skull and skeleton) . Most of these specimens now are in the collections of the Smithsonian Institution, with a few remaining in the teaching collection at Old Dominion University. Collectively, these surveys provide information on the habitats and extent of distribution of southern bog lemmings in southeastern Virginia. RESULTS Live trapping Biweekly trapping for 17 months on the two live trap grids in the GDSNWR yielded 13 bog lemmings, two on Grid 1 and 11 on Grid 2 (Stankavich 1984). On Grid 2, none was caught until the 10 month, and then all were captured within a period of a few weeks. However, bog lemmings were known to be present from the start because they produce distinctive bright green bullet-shaped fecal pellets, plus they strip and eat the green outer covering from the softrush, Juncus ejfusus, leaving behind the spaghetti-like bits of pith. Ford's year-long mark-and-release study was conducted on a large study grid in a regenerating clearcut near the intersection of Desert and Clay Hill Roads in Suffolk, on a site close to the GDSNWR. She caught several dozen each of bog lemmings and woodland voles (Microtus pinetorum) using modified Fitch live traps (Rose, 1994 ). For unknown reasons, the southern bog lemmings on this site were much more pron~ to entering live traps than the same species had been in Stankavich's (1984) study. The only other Synaptomys taken in live traps was an adult female collected early in 1999 in early successional habitat in a wetland bank now reverting to Dismal Swamp vegetation in southern Chesapeake. I I 156 VIRGINIA JOURNAL OF SCIENCE