Procedures

The study was conducted in 2, 1-hectare alfalfa, Medicago sativa, sites in the University of Illinois Biological Research Area (Philips Tract), 6 km NE of Urbana, Illinois (40o15’N, 88o28’W). We monitored the first site (Alfalfa 2) from 1 March 1982 through 31 July 1984 and the second (Alfalfa 3) from 1 October 1983 through 31 May 1987 (Getz, et al., 1993). A l0-m closely mown strip of grasses separated the two sites. Alfalfa 2 was planted to with M. sativa spring of 1976 and had been trapped as a part of another study (Getz et al. 2001) from March 1977. Alfalfa 3 was planted to M. sativa spring 1982. The study sites are described by Getz et al. (1979, 1987).

Prairie voles at the two sites were monitored by live-trapping directly at nests of social groups. Four to five multiple-capture live-traps (Burt 1940) were placed around each surface nest and at burrows leading to underground nests. The live-traps were constructed of 1.5 cm redwood; inside dimensions of the traps were 6.5 x 6.5 x 20 cm. We baited traps with cracked corn and in summer covered them with vegetation or aluminum shields to shade them from the sun. The wooden traps provided adequate insulation during the winter; we did not provide the traps with bedding. We set traps at nests at 0630 h on Monday morning and checked them at 3-4 h intervals through 2400 h and again at 0630 and 0930 h on Tuesday morning. This schedule was repeated Thursday morning through Friday morning. Because of the short interval between trapping sessions, we did not prebait traps.

Once each month traps were set at stations in a 10-m grid pattern as part of a long-term study utilizing the same sites (Getz et al. 2001). In this study traps were set at 1500 hours Tuesday and checked at 2000 hours, 0800 hours and 1500 hours through Friday afternoon. Midway through the monthly periods of grid trapping, all grid stations > 15 m from known nests were trapped on the above schedule from 1500 hours Wednesday through Friday afternoon.

Voles were weighed and individually marked by toe-clipping at first capture, < 2 toes on per foot. The field protocol was reviewed periodically, and approved by the University of Illinois Laboratory Animal Resource Committee throughout the study, based on University and Federal guidelines, as well as those of by the American Society of Mammalogists, in effect at the time.

Locations of the nests of all social groups were continuously updated as a part of the study of social organization (Getz, et al., 1993). A field coordinator was responsible for ensuring completeness of monitoring social groups. At the beginning of each trapping session (Monday and Thursday morning), a list of all nests to be trapped was provided the field workers, as was a list of animal numbers of all females for which the nest was known. We focused on females in locating social group nests since few male groups or single males occupied permanent nests (Getz et al. 1993). The animal number of each female captured was checked against this list. If the female was on the list, she was released after pertinent data had been recorded. If not on the list, she was dusted with ultraviolet reflective powder (Leman and Freeman 1985) and released; all unmarked adult females also were dusted and released. The color of dust used and where the female was released in relation to the specific nest openings or grid station was recorded on the data sheet for each female. A list was made of all animals dusted during the day: animal number, nest/grid station where released, and color used. That night trackers (teams of 2 individuals, the number employed depended upon the number of dust trails to be traced) went to the field and using U-V lamps, traced the trails to underground or surface nests. When a potential nest was located, a colored wire flag was placed at the burrow/surface nest and the location noted on the sheet. Animals during dusted that night and the next day were added to the list of females not to be dusted the rest of the session. Tracking crews went to the field each night until all the dust trails had been checked.

The coordinator would examine each presumed nest location before the next scheduled session to determine if it appeared to be an inhabited nest. If so, live-traps were placed around the nest and the nest added to the list to be trapped the next session. The female was also added to the list of those known to be residents of a nest. The field coordinator would inspect the captures made during each session and would make a judgment as to which females appeared to be residents at a nest. The others were again dusted when next caught. These animals were added to the up-dated list of known residents provided the field workers the next session. Nests at which none of the residents had been caught for three sessions, 10 days, were removed from the list of trapped nests.

By such a method of monitoring social groups and continuously up-dating the nests trapped and identifying females to be dusted, we were able to account for almost all animals in the population. In excess of 99% of all animals present were captured during any 2 sessions (Ozgul et al. 2004). We thus were able to determine, as well as could be