METHODS 



Surveys at Azure Cave and other potential roost sites in the Little Rocky Mountains were 

 conducted during two visits in 1997 (19-20 April, 7-9 October), and three visits in 1998 (17-18 

 April, 16-18 September, 12-13 November). Counts of hibernating bats were made in April of 

 each year and during the November visit. Netting sessions and inspection of other potential 

 roosts occurred during September and October visits. 



During bat counts in Azure Cave, care was taken to minimize disturbance. Therefore, 

 bats encountered were not handled nor marked while crews were in the cave. Because of 

 concern about disturbance, and inaccessibility of some hibernating bats, species identification of 

 most bats in the cave was not possible. Counts of all bats were made independently by two 

 observers to reduce bias of the total estimate. Route of movement through the cave during 

 counts was similar for each observer. Cave rooms were counted simultaneously by both 

 observers, one room at a time. Location of occurrence within the cave was recorded to document 

 rooms and passages used most often by the bats. Air temperature and relative humidity were 

 recorded using a thermometer and Bacharach sling psychrometer. Remains of bat skulls were 

 collected, tentatively identified, and verified by Mike Bogan (USGS, Museum of Southwestern 

 Biology, University of New Mexico). 



Bats were captured at the mouth of Azure Cave using two mist nets (50 denier: one 6 m, 

 one 9 m) set in the same configuration each night. An ANABAT II ultrasonic bat detector was 

 operated during netting to indicate when bats were nearby. Trapping occurred two consecutive 

 nights during each session. Nets were set 30 min before dusk and operated for at least four 

 hours. Captured bats were identified, sexed, measured (weight with a Pesola 30 g scale, forearm 

 length with a dial caliper), examined for reproductive staUis, and marked with colored celluloid 

 bands. A unique combination of colored bands was used for each night of trapping: 7 October 

 1997 (single white on right wing), 8 October 1997 (white + purple), 16 September 1998 (white + 

 blue or white + white), 17 September 1998 (single white on left wing). 



Rock-shelter caves were located using topographic maps or directions from area 

 residents. Caves were explored to their full extent, while recording climate data and taking 

 preliminary measurements of cave dimensions; the presence or absence of bat spoor was noted. 

 Locations of abandoned mines were obtained in 1996 from maps or the aid of Zortman Mining, 

 Inc. personnel. Visits were limited to external inspection of workings. 



Sources for additional data on bats from the Little Rocky Mountains used in the analyses 

 presented here include Chester et al. (1979), Butts (1993), Hendricks and Genter (1997), and Don 

 Sasse (Ashland District, Custer National Forest: personal communication). Where results are 

 statistically analyzed, I follow standard procedures as described in Sokal and Rohlf (1981), with 

 statistical significance assumed when P < 0.05. Analysis of variance (ANOVA), two-sample t- 

 test procedures, simple regression, and correlation analyses were run using STATISTIX software. 



