Reprinted from

© 2008 by the Loma Linda University Press. All rights reserved. May not be copied or reused, other than for personal use, without express written permission of the publisher.

Click here for details about the volume and how to order it, or send e-mail to rattlesnakes@llu.edu Pp. 291-302 in W. K. Hayes, K. R. Beaman, M. D. Cardwell, and S. P. Bush (eds.), The Biology of Rattlesnakes. Loma Linda University Press, Loma Linda, California. 291

Thermal Ecology of Hibernation in a Population of Great Basin Rattlesnakes, Crotalus oreganus lutosus

Vincent A. Cobb1,2,3 and Charles R. Peterson1

1Department of Biological Sciences, Idaho State University, Pocatello, Idaho 83209 USA

Abstract.—Although many snake species spend a considerable amount of time hibernating, surprisingly little is

known about body temperature (Tb) selection during this period. To measure Tb variation and associated movements during hibernation, we surgically implanted temperature-sensitive radiotransmitters into 26 adult Great Basin Rattlesnakes (Crotalus oreganus lutosus) over a 3-yr period. All snakes were captured and released at a large den site in the desert on the Idaho National Engineering and Environmental Laboratory in southeastern Idaho. Body temperatures were measured once per week and environmental temperatures were sampled continuously using thermocouples and a datalogger. Snake

entrance into the hibernaculum ranged from 3 October-15 November (median date = 16 October). Mean Tb decreased from approximately 12.5°C in mid-November to a minimum of 7°C in February-April, and then increased to 8°C near

emergence time in May. The lowest Tb experienced by any of the snakes was 4.4°C. During the winter, differences in Tb among individuals measured at the same time were as great as 5°C. Excluding ingress and egress variation, overall mean Tb was 8.9°C and varied only by 1°C between years. Although we could not see snakes in the den, radiotelemetry allowed us to determine the x, y coordinates of individuals. Underground movements by snakes continued until January and by mid-winter most snakes had become relatively stationary approximately 10 m from the den entrance. Movements from January-April were minimal until near emergence time. Emergence times varied from 27 April-26 June (median date = 15 May). Duration of hibernation varied from 189-253 d (median duration = 206 d). Considerable inter-individual

variation in Tb, underground movements, and duration of hibernation suggests that overwintering in snakes is more than just a period of lethargy at low temperatures.

Introduction hibernation sites, including tree roots, rodent and crayfish burrows, ant mounds, sinkholes, water-filled cisterns, and Hibernation in ectothermic vertebrates is a behavioral human dwellings (Gregory, 1982). Although some snakes phenomenon characterized by decreased activity that allows hibernate individually, at more northern latitudes snakes individuals to survive extended periods when metabolic re- typically hibernate communally (Gregory, 1984; Sexton et actions associated with activity cannot be maintained due al., 1992), which has undoubtedly made these few species to shortened photoperiods and low ambient temperatures some of the best studied regarding hibernation ecology. In (Gregory, 1982; Zuffi et al., 1999; McNab, 2002). During regions of severe winter climate, these communal hibernac- this time, organisms retreat to a hibernation site that pro- ula may contain thousands of individuals and multiple spe- tects the individuals from reduced environmental tempera- cies (Gregory, 1984). Such large aggregations are attributed tures (Te), which lower body temperature (Tb) and metabo- to limited numbers of overwintering sites that are appropri- lism, and make physiological processes much less efficient. ate for survival (Viitanen, 1967; Andersson, 2003; Harvey Because the effect of photoperiod and change in ambient and Weatherhead, 2006); however, reproductive tactics for temperature is much greater in temperate latitudes, hiberna- spring breeding in some species (e.g., Thamnophis sirtalis) tion is common in these regions, as opposed to areas with are thought to contribute to the large numbers of the same warmer climates. For organisms in the most northern re- species (Gregory, 1974; Duvall et al., 1992). gions, it is not just inactivity that must be dealt with but Rattlesnakes (Crotalus) are one of the best docu- also escape from freezing temperatures (Storey and Sto- mented taxa of snakes that communally hibernate in large rey, 1996). Although ectothermic vertebrates can success- numbers. Other notable groups include: North American fully hibernate in a variety of microhabitats (Ultsch, 1989), Garter Snakes (Thamnophis; Gregory, 1974; Aleksiuk, a number of species find underground retreats adequate 1976), North American Watersnakes (Nerodia; Kissner and (Gregory, 1982). Weatherhead, 2005), and North American Racers (Coluber; In general, snakes range to higher latitudes in the north- Brown and Parker, 1984; Rosen, 1991). Klauber (1972) ern hemisphere than lizards, turtles, or crocodilians (Pinder documented communal denning of hundreds and thousands et al., 1992). Snakes also exhibit considerable variation in of C. viridis and C. horridus, two of the more northern spe- ______cies. Although numbers observed at hibernacula decades 2 Correspondence e-mail: [email protected] ago have since decreased, there remain some scattered 3 Present address: Department of Biology, Middle Tennessee hibernacula containing populations of several hundred in- State University, Murfreesboro, Tennessee 37132 USA dividuals (Sehman, 1977; Duvall et al., 1985; Macartney, © 2008 Loma Linda University Press 1985; Brown, 1992; Martin, 1992, 2002). 292 V. A. Cobb and C. R. Peterson

Communal denning has provided numerous field stud- and feeds primarily upon small mammals (Klauber, 1972; ies with the opportunity of having many individuals readily Diller and Johnson, 1988). Females usually reach reproduc- available, at least during certain times of the year. Several tive maturity in their fourth year (Nussbaum et al., 1983); of these studies have examined a variety of ecological pa- however, in southeastern Idaho, females may not reach rameters associated with hibernation, such as demography sexual maturity until their fifth or sixth year (Cobb and (Sehman, 1977; Martin, 2002), growth and mass change Peterson, unpubl. data). At the more northern latitudes, C. (Hirth, 1966; Gannon and Secoy, 1985; Diller and Wallace, o. lutosus breeds in the late summer, with ovulation and 2002), aggregation behavior (Graves and Duvall, 1987), fertilization occurring in the following spring, and birth in timing of hibernation (Brown, 1992; Martin, 1992, 2002), the late summer (Duvall et al., 1982, 1985; Diller and Wal- body temperature regulation (Jacob and Painter, 1980; lace, 1984, 2002; Gannon and Secoy, 1985; Macartney and Brown, 1982; Macartney et al., 1989), winter activity and Gregory, 1988). Females typically reproduce on biennial or movements (Sexton and Marion, 1981; Brown, 1982; Mac- triennial cycles (Aldridge, 1979; Diller and Wallace, 1984, artney et al., 1989), survivorship (Charland, 1989; Diller 2002; Macartney and Gregory, 1988). However, some pop- and Wallace, 2002), reproductive strategies (Duvall et al., ulations may have longer reproductive cycles, as does C. 1992), and pheromone trailing (Graves et al., 1986). horridus (Brown, 1991; Martin 2002). In the fall of 1989, we initiated a study of hibernation Study area.—The study area was located at Cinder ecology at a large communal hibernaculum of Great Basin Butte Crater on the Idaho National Engineering and En- Rattlesnakes (C. oreganus lutosus) in southeastern Idaho. vironmental Laboratory (INEEL) in Butte County, Idaho In addition to mark-recapture efforts, we felt that this hi- (altitude 1460 m). The INEEL includes an area ca. 2,300 bernaculum (i.e., physical structure of the den and large km2 and is under the jurisdiction of the U.S. Department of number of snakes) offered a unique opportunity to exam- Energy; thus, access is limited to authorized persons. Lim- ine body temperature (Tb) regulation and activity regard- ited access, enforced since 1949, has made the INEEL an ing snake hibernation. First, environmental temperatures area of relatively low disturbance. The landscape is a flat, (Viitanen, 1967) and reversing thermal gradients (Sexton sagebrush-grass community with numerous basalt outcrops and Hunt, 1980) are hypothesized to dictate the movement (Anderson and Holte, 1981). Lava flows underlie most of of snakes into and out of the hibernaculum; however, not the area and the youngest flows are from the late Pleisto- all populations support this idea (Weatherhead, 1989) and cene (Ross and Forrester, 1958). This is a semi-arid region interspecific variation can occur within the same hibernacu- in the northeastern portion of the Snake River Plain that ex- lum (Graves and Duvall, 1990). Second, although activity periences cold winters and hot summers. Snow cover dur- and movements during hibernation may persist in some lo- ing winter usually persists from December-March. Mean calities (Jacob and Painter, 1980; Zuffi et al., 1999), hiber- annual air temperature and precipitation from 1950-1993 nation and associated low temperatures at higher latitudes was 5.56°C and 21.9 cm, respectively (National Oceanic limit underground activity (Viitanen, 1967; Brown et al., and Atmospheric Administration, unpubl. data). During this 1974; Gregory, 1984). Sexton and Hunt (1980) suggested study (1989-1992), mean annual air temperature and pre- that hibernation activity may be associated with thermo- cipitation was 6.49°C and 17.5 cm, respectively. regulation and movement to the warmest available environ- Cinder Butte Crater is a large (ca. 40 ha), irregular- ment. Third, large aggregations of hibernating rattlesnakes shaped volcanic depression ranging from 1-5 m below the could theoretically produce enough metabolic heat to raise desert floor. Edges of the crater are characterized by basalt their Tb considerably above that of ambient temperature boulders, which provide shelter and retreat sites for snakes (White and Lasiewski, 1971), yet this remains unsubstanti- emerging and entering hibernacula. Several sites with south- ated in laboratory trials (Graves and Duvall, 1987) and in ern exposures along the crater’s edge are used by snakes as the field (Brown et al., 1974; Brown, 1982; Weatherhead, hibernacula. These hibernacula consist of layers of horizon- 1989). However, White and Lasiewski (1971) implied larg- tal and vertical cracks and fissures created during the cool- er numbers of snakes than these latter studies tested. Our ing process following lava flow periods. Hibernacula serve approach to the above ideas was to use our mark-recapture as winter retreats for 500-800 C. o. lutosus and smaller data to describe the timing of hibernation, to radiotrack in- numbers (<100) of Desert Striped Whipsnake (Masticophis dividual snakes and monitor underground movements and taeniatus taeniatus), Great Basin Gophersnake (Pituophis body temperatures, and to characterize the thermal condi- catenifer deserticola), and Wandering Garter Snake (Tham- tions of the denning site for defining any trends in either the nophis elegans vagrans; Sehman, 1977; Cobb, 1994). snakes or hibernaculum. Environmental measurements.—Near Cinder Butte, we

measured air, soil, and snake operative temperatures (Te; Materials and Methods Bakken, 1990; Peterson et al., 1993) using a multi-channel datalogger (model CR10, Campbell Scientific, Logan, Utah, Study animals.—Crotalus oreganus lutosus ranges USA) and copper-constantan thermocouples. To determine from southeastern Idaho to northeastern California, south- environmental temperatures available to snakes, we con- ern Nevada, and northern Utah (Wright and Wright, 1957) structed a microclimate weather station that remained in op- Thermal ecology of hibernation in Great Basin Rattlesnakes 293

Table 1. Means (ranges) for the timing of hibernation in a southeastern Idaho population of Great Basin Rattlesnakes (Crotalus oreganus lutosus) for three winters.

Sex N SVL Ingress Ingress Egress Egress Hibernation ______(cm) date mass (g) Date mass (g) duration (d)

1989-90 F 5 78.4 299.2 314.2 132.2 284.3(3) 198 (73–85) (297–303) (205–460) (124–138) (177–405) (192–204)

M 2 94.5 292.5 521 136.5 474 209 (87–102) (284–301) (424–618) (131–142) (394–554) (206–212)

1990-91 F 6 80.5 286.5 392.2(4) 139.8 327.5 218.3 (75–87) (282–294) (237–519) (135–146) (189–480) (206–227)

M 6 79.5 285.5 322.3 152.2(5) 290.3 230.2(5) (67–96) (278–296) (188–502) (137–177) (160–458) (206–253)

1991-92 F 7 83.9 289.3 466.2(6) 125.6 454.0(2) 201.3 (80–91) (276–298) (380–507) (116–132) (443–465) (189–218) M = male; F = female; Julian dates for entrance (ingress) and exit (egress) are provided; subscripts associated with a mean refer to sample size when it differs from N.

eration throughout the duration of the study. We measured ly implant radiotransmitters, we anesthetized snakes with air temperatures at 1 and 200 cm above ground and soil an anesthesia machine (Heidbrink Kinet-O-Meter, Ohio temperatures at depths of 2.5, 10, 15, and 35 cm. Operative Chemical and Surgical Equipment Co., Madison, Wiscon- snake temperatures with hollow copper tubes (30 x 2 cm) sin, USA) using an oxygen/halothane mixture. We implant- sealed with two rubber stoppers and painted to match the ed the radiotransmitter interperitoneally, at ca. 65% of the reflectance of C. o. lutosus skin (Peterson et al., 1993) were snout-vent length posteriorly, following similar procedures placed in the sun on soil, in the shade on soil, in the sun on of Reinert and Cundall (1982). All radiotransmitters were basalt rock, and under typical sagebrush plants that created <4% of snake mass. We held postoperative snakes in the a sun/shade mosaic. Environmental temperatures were sam- laboratory for 2-3 d at warm temperatures, near 30°C, to pled at 1-min intervals and then averaged every 15 min. We promote incision healing. We then released the snakes at determined maximum and minimum environmental temper- their point of capture.

atures (Te max and Te min) for each 15-min period by select- Underground snake movements.—During the period ing the maximum and minimum operative temperatures of 1989-1992, we monitored 26 adult rattlesnakes (7-12 snakes the air, soil, and snake model temperatures throughout the each year) from ingress in the fall until egress in the spring. day. We used these data to determine what thermal environ- During ingress and egress, we relocated snakes two to three ments were available to snakes for each day. times per week. Once all snakes retreated underground for During September 1991, we drilled a 15-cm diameter the winter, we monitored them once per week. Because of x 3 m deep hole into the desert floor 11 m from the den en- the parallel arrangement of the historic lava flows, we felt trance. The hole pierced the roof of the hibernaculum at the confident that the cavities between lava flows were parallel approximate location of hibernating snakes during the first too. These cavities appeared to be the sites snakes were using two winters. We placed three thermocouples at 35 cm, 100 as hibernacula. Therefore, to obtain approximate locations cm, and 250 cm to monitor temperatures. The 250-cm ther- of individual snakes, we walked on the desert floor above mocouple was located at the upper edge of the cavity, which the hibernacula and determined the point where the strongest was estimated to serve as the primary hibernaculum, and telemetry signal was detected. We placed a survey flag to this provided measurements of internal den temperatures. indicate the snake’s position. We measured straight-line dis- General telemetry.—We captured snakes during the tances between locations to the nearest cm and took compass summer and fall and transported them to Idaho State Uni- coordinates for each location from the crater’s edge immedi- versity for radiotransmitter implantation. For comparisons ately above the primary hibernacula opening. We then plotted of females of different reproductive condition, we identified snake positions using a polar coordinate system (Zar, 1984). snakes that would give birth during that year by their length/ We calculated distances moved each month by summing all mass relationship and by palpation of follicles. To surgical- straight-line distances between successive relocations. 294 V. A. Cobb and C. R. Peterson

Table 2. Mean monthly body temperatures (ranges) of a south- many individuals retreated underground within a day, oth- eastern Idaho population of hibernating Crotalus oreganus lutosus ers spent several days (e.g., a week) shuttling back and for three winters. forth between boulders during the day and staying only a few meters inside the hibernaculum at night. Ingress for 26 Month 1989-90 1990-91 1991-92 N = 7 N = 10 N = 6 telemetered snakes ranged from 3 October-15 November ______(Table 1), with the mean date being 17 October. Individu- Nov 12.3 12.5 12.8 als that arrived at the hibernaculum early usually remained (10.3 – 18.1) (10.3 – 14.4) (11.3 – 14.6) above ground, while snakes that arrived later entered the hibernaculum quickly. Median ingress date (Julian day = Dec 10.5 10.5 11.0 298) was significantly earlier (14 d) during the 1990-1991 (8.3 – 11.9) (8.5 – 12.0) (9.3 – 12.9) winter (Kruskal-Wallis, H = 7.87, df = 2, P = 0.02), but it was only significantly different from the 1989-90 win- Jan 8.4 8.7 8.9 (6.2 – 10.9) (5.4 – 10.6) (7.3 – 10.9) ter (Dunn post hoc, P < 0.05). However, more variation in retreat time occurred in the 1991-1992 winter (1 SE = 4.2 Feb 6.5 7.7 8.0 d) than in 1989-90 (1 SE = 2.4 d) and 1990-1991 (1 SE = (4.6 – 8.4) (6.5 – 8.9) (6.5 – 10.7) 1.5 d), the other two winters. Sexes did not differ in ingress dates for either of the first two winters in which we tele- Mar 6.4 7.3 7.9 metered both sexes (Mann-Whitney, 1989-90: U = 6.5, P = (4.4 – 8.3) (5.7 – 9.2) (5.9 – 11.1) 0.57, N = 7; 1990-91: U = 41, P = 0.82, N = 12). Snakes emerged from hibernation in the spring as rap- Apr 7.3 7.1 7.9 idly as they entered in the fall, but remained at the crater’s (5.5 – 8.9) (5.1 – 8.2) (5.9 – 11.0) edges longer. Most individuals stayed at the edge several days and again shuttled back and forth between boulders and Mean 8.3 9.0 9.5 the hibernaculum. Egress date (date first observed outside Range of 7.2 – 9.4 8.3 – 9.6 8.9 – 11.5 the den) for 25 telemetered snakes ranged from 26 April-26 individual June (mean date = 16 May; Table 1). Egress dates for the means 1990-91 winter were significantly later than for the 1989-90 and 1991-92 winters (one-way ANOVA: F = 10.18, df = 2, P < 0.001, Holm-Sidak post hoc). Neither egress date for Snake body temperatures.—To measure snake body 1990 (t = -0.78, df = 5, P = 0.47) nor 1991 (Mann-Whitney, temperatures ( s) during the weekly relocations, we used Tb U = 41, P = 0.052, N = 11) showed sex differences. temperature-sensitive radiotransmitters (model CHP-2, Te- The duration of hibernation determined for 25 indi- lonics, Mesa, Arizona, USA) in the rattlesnakes. Prior to viduals was approximately seven months (Table 1). The implantation, we calibrated the radiotransmitters in a water mean 1990-91 hibernation period was significantly longer bath at stable temperatures at 5°C intervals from 0-40°C us- (223 d) than the 1989-90 and 1991-92 periods, which were ing interpulse interval measurements with a signal proces- identical (201 d; one-way ANOVA: F = 10.20, df = 2, P < sor (model TDP-2, Telonics). Temperature curves for each 0.001). This difference likely was associated with the cooler radiotransmitter were determined using 3rd-order polyno- environmental temperatures (Tes) during the 1990-91 win- mial regressions. During the weekly snake relocations, we ter; winter temperatures for the first and third years of the measured interpulse interval measurements for converting study were similar (Fig. 1). Ingress dates were independent into snake Tb. To assure against possible inaccurate inter- pulse interval measurements of single sampling, at least 15 three consecutive readings of similar interpulse intervals 1989-90 were used for Tb determination. 10 o C) 1990-91 Analyses.—Both parametric (Pearson correlation, t- 1991-92 test, analysis of variance [ANOVA]) and non-parametric 5 tests (Kruskal-Wallis, Mann-Whitney) were employed, 0 depending on whether parametric assumptions were met. Statistical tests were conducted using Sigma Stat 3.1 (Sys- -5

tat Software, Inc., Point Richmond, California, USA) with (°C) Temperature Air Air-1 Temperature0 ( alpha = 0.05. -15 Results Oct Nov Dec Jan Feb Mar Apr May Figure 1. Three winters of mean monthly air temperature for the Ingress, egress, and duration.—Upon returning to Cin- Idaho National Engineering and Environmental Laboratory near der Butte Crater from active season migration, rattlesnakes the Cinder Butte Crater hibernaculum. Note the lower temperatures generally stayed close to hibernacula entrances. Although associated with December and January of the 1990-91 winter. Thermal ecology of hibernation in Great Basin Rattlesnakes 295

of egress dates for each of the three winters (1989-90: r = 0.38, P = 0.39, N = 7; 1990-91: r = 0.18, P = 0.59, N = 11; 1991-92: r = 0.20, P = 0.66, N = 7). Also, duration of hibernation was not influenced by sex F( = 3.91, df = 1, P = 0.062) or female reproductive condition (F = 2.92, df = 2, P = 0.059; N = 9 gravid and 7 non-gravid). Overwinter mortality.—Mortality of individuals over the hibernation period was relatively low. Besides two snakes being potentially preyed upon during fall basking at the den (Cobb and Peterson, 1999), only three snakes died during hibernation (Tb and movement data were omit- ted from analyses): two during 1990-91 and one during Figure 2. Hours available to snakes for thermoregulating at >25°C 1991-92. One snake within the hibernaculum in 1990-91 at the time of hibernation entrance during 1991. Maximum op- was only 3-4 m from the outside entrance, and appeared to erative temperatures were derived from five painted copper snake models outside the den entrance at the Cinder Butte Crater hi- have died at the time of a rapid drop in outside air tempera- bernaculum. Last day that snakes were observed above ground ture during December (Fig. 1), after which its Tb dropped was day 292. A cold weather front arriving on day 294 ended all below freezing. During this same period, other telemetered above-ground activity in snakes for the year. snakes made movements further back in the hibernaculum. The other two mortalities could not be explained, but the snakes were at temperatures above freezing.

Body mass loss ranged from 2.8-15.5%, with a mean 35 (± 1 SE) of 9.35 ± 0.94%. Mean body mass loss for females 1989-90, 90-91, 91-92 Ma x 30 (N = 12) was 8.8 ± 1.29% and for males (N = 8) was 10.2 Hibernation Periods (N = 23) Me a n ± 1.38 %; however, body mass loss was similar for the two 25 Min sexes (t = 0.69, df = 18, P = 0.50). 20

Snake body temperatures.—We recorded snake Tbs 15 the entire winter for 23 individuals over the 3-yr period. 10 Operative temperatures from snake models indicated that

Body Temperature (°C) Temperature Body Body5 Temperature (°C) snakes must decrease their Tbs because of the lack of time 0 to maintain warm Tbs outside the hibernaculum (Fig. 2). Snakes exhibited gradual cooling for the first three months Oct Nov Dec Jan Feb Mar Apr May of hibernation, from approximately 12°C down to 7°C, v Figure 3. Mean and range of Crotalus oreganus lutosus (N = 24) and then Tb remained relatively steady until shortly before body temperatures over three consecutive hibernation periods at spring emergence (Fig. 3). The last one to two measure- the Cinder Butte Crater hibernaculum in southeastern Idaho.

ments in Tb, <1 wk before emergence, indicated a slight in- crease. After snakes emerged, they began exhibiting diurnal thermoregulatory activities along the edge of Cinder Butte Crater. Although there was considerable variation among individuals (up to 5°C) at any one time throughout a win-

ter, mean Tbs were remarkably similar between months and years (Table 2), despite Te differences. Excluding the time periods overlapping with snake ingress and egress, year-to-

year differences in mean hibernation Tb was only 1°C. The minimum Tb experienced by any snake was 4.4°C, while the mean and maximum lowest Tb was 6.3 and 10.2°C, respectively. Minimum Tb was not influenced by sex F( = 1.43, df = 1, P = 0.24) or SVL (r = 0.01, P > 0.95, N = 23).

Mean minimum Tb for all years combined occurred on 13 March; however, it varied from 18 January-28 May among individuals, and there was no correlation between minimum Figure 4. Environmental temperatures (Tes) of the Cinder Butte Tb and date (r = -0.14, P = 0.52, N = 23). Some snakes Crater den site in southeastern Idaho, with maximum and mini- exhibited the coolest T s in the middle of winter, when b mum snake Tbs for 1992. The 250 cm hole temperature represents air temperatures were the coolest, whereas others experi- internal hibernaculum temperature 10 m back from the den en- enced coolest Tbs just before emergence from hibernation. trance. Note the delay in snake emergence, indicated by rapidly However, there was a significant difference in minimum Tb rising Tbs, compared to increasing Te in the spring and the internal between the 1989-1990 and 1991-1992 winters (Kruskal- hibernaculum temperature lower than snake Tb. 296 V. A. Cobb and C. R. Peterson

Wallis, H = 6.87, df = 2, P = 0.032), and later dates when increases in Tes outside the hibernaculum. Also, snake Tbs minimum Tbs occurred were associated with later egress remained steady during this period until they left the den. dates (r = 0.55, P = 0.008, N = 22). Underground movements.—Positions of snakes

Increase in snake Tb at egress was associated with in- throughout this study strongly indicated that most indi- creases in hibernaculum temperatures and not necessarily viduals were within a large region of the hibernaculum ca. movements toward the den entrance. Deep hibernaculum 8-12 m from the den entrance (Fig. 5). This same area was temperature fluctuated between 4-9°C during most of the the primary site used by most snakes during each year. We winter, remaining slightly cooler than snake Tbs during mid- monitored three individuals for 2 yr, and only one of the winter and within snake Tb ranges in the early spring before three in the second year was located near (x m) its previous egress (Fig. 4). Hibernaculum temperature was only 8°C location. Snakes selected locations close to the den entrance at egress in 1992, and appeared to be buffered from spring (Fig. 5) shortly after ingress. The farthest locations in Fig. 5 were two individuals that spent many days at these sites. During mid winter, one of these two individuals moved to- ward the den entrance to relocate in this region of greatest use. The other remained in its distant location throughout the winter. Underground snake movement was greatest during the first two months of hibernation (Fig. 6). Movements early in hibernation were more concentrated toward the den en- trance, and as the winter progressed, snakes moved further back into the hibernaculum (Fig. 7). Movements among individuals varied considerably. In the first few weeks of hibernation, some individuals moved from near the den en- trance to locations several meters into the den. Some in- dividuals returned to their previous positions by their next relocation. Other individuals gradually moved further into the hibernaculum as the first half of the winter progressed. However, movements from January-April were minimal

until near egress. Figure 5. Spatial distribution of Crotalus oreganus lutosus (N = Effect of movements on snake body temperature.— 7) locations during the 1989-90 winter within the Cinder Butte During hibernation, if snakes remained stationary for mul- Crater hibernaculum in southeastern Idaho. The region 10 m from tiple relocations, Tb changes were small. But, as individual the den entrance is where snakes spent the majority of their time snakes moved from one location to another underground, in hibernation (dark circles). Open circles near the den entrance their T s changed (Fig. 7). Sometimes, movements between represent two ingress locations for five snakes each and open b locations were only 2-3 m, yet relatively large T differ- circles further into the den represent single ingress locations for b two snakes. Open circles with a bar indicate egress locations for ences occurred. However, spatial distributions of snakes three snakes. did not reveal any patterns in snake Tb (Fig. 8). There was a slight tendency for snakes further back in the hibernaculum

to have higher Tbs, yet snakes located within 1-2 m from each other sometimes had Tbs differing by 2°C or more. Thermocouples measuring air temperature inside the hiber- naculum through the drilled hole and 2.5 m into the den entrance from the outside edge indicated that large temper- ature gradients occurred. Air temperatures near the den en- trance commonly fell below freezing throughout the winter, yet internal den temperatures remained higher (4-9°C) and more constant than outside temperature (Fig. 4).

Discussion

One of the most intriguing features of snakes spend- ing such a significant amount of time underground is that half of their lifetime ecology and behavior remains rela- Figure 6. Mean monthly variation in underground movements by tively unknown. Once in hibernation, snakes by and large Crotalus oreganus lutosus (N = 7) from the Cinder Butte Crater are inaccessible for study. One reason this study was suc- hibernaculum during the winter of 1989-90, indicating consider- cessful at describing the movement activities and thermo- able movement during the initial months of hibernation. Thermal ecology of hibernation in Great Basin Rattlesnakes 297 regulatory behaviors of snakes underground was that the (2003) noted for the fall-mating Vipera latastei that adult hibernaculum was situated such that its roof was parallel males emerge from hibernation earlier than adult females. with the desert floor. This structural arrangement at least Andersson (2003) also observed male V. berus emerging allowed a two-dimensional analysis of movement. An in- from hibernation earlier than females. ternal den temperature measurement and others near the Variation in body temperature.—While it is no surprise entrance gave us some indication of temperatures available that snake Tb decreases during hibernation as a result of for snakes underground. cooler Te and depressed metabolic activity (Gregory, 1984; Timing of hibernation.—The C. oreganus lutosus pop- McNab, 2002), few studies have examined snake Tb regu- ulation at Cinder Butte Crater in the eastern Snake River lation during hibernation. Since summarized by Jacob and

Plain of Idaho spends seven months each year underground Painter (1980), several studies have recorded winter Tbs of in communal hibernation. This is not atypical for rattle- snakes; however, few of these have been communally hiber- snakes at this latitude and elevation. The timing of hiberna- nating snakes in natural hibernacula (C. horridus: Brown, tion was almost identical to those reported for C. oreganus 1982; T. sirtalis: Costanzo, 1986; C. oreganus and T. sir- in Utah (Hirth, 1966) and British Columbia (Macartney, talis: Macartney et al., 1989; E. obsoleta: Weatherhead, 1985) and for C. viridis in Saskatchewan (Gannon and 1989). Of these, Brown (1982), Macartney et al. (1989),

Secoy, 1985). Shorter hibernation periods were recorded and Weatherhead (1989) provided temporal Tb variation for for C. oreganus in northern Idaho (Diller and Wallace, individuals. Even in their small samples (N = 2-6), consid- 1984). In Wyoming, C. viridis (Duvall et al., 1985) had hi- bernation periods slightly longer than this study. Patterns of ingress and egress also were similar for C. horridus in the northeastern U.S. (Galligan and Dunson, 1979; Brown, 1992; Martin, 1992, 2002). Because of climate and eleva- tion differences of den sites, regional scale comparisons may not follow general hibernation trends (Sexton et al., 1992). Using northwestern rattlesnakes as an example, the Haystack Mountains site in Wyoming had more severe winters than our study site at similar latitude (Duvall et al., 1990), but a site in north central Idaho (Diller and Wallace, 1984) had milder winters than Cinder Butte Crater. Even comparisons of numerous den sites within single locations for T. sirtalis (Gregory, 1974; Hawley and Aleksiuk, 1975), Elaphe obsoleta (Weatherhead, 1989; Blouin-Demers et al., 2000) and Sistrurus catenatus (Harvey and Weatherhead, 2006) have revealed obvious differences in the timing of hibernation, which implies microclimate and den structure A. as causative factors. The return of snakes to the Cinder Butte Crater hiber- naculum occured in late September and early October, after lengthy active season migrations. Non-gravid female C. o. lutosus moved up to 7 km from the den (Cobb, 1994), and presumably, males were traveling at least as far (King and Duvall, 1990). There was a lack of pattern associated with fall entrance underground except for a slight year differ- ence (1990 was earliest) that could not be linked to tem- peratures in October and November. Emergence dates also were later that same winter, which resulted in snakes be- ing underground longer. Although females emerged in the spring a few days earlier than males on average, a statisti- cal difference was not apparent, which may be due to our limited sample size. Macartney (1985) also suggested that females emerge from hibernation earlier than males, but a large difference was not apparent. Because rattlesnakes B. in north temperate regions do not mate in the spring, early male exodus from hibernation, as in Thamnophis, is not ex- Figure 7. Underground movements (A) and body temperatures pected, and no sex differences were observed in C. viridis (B) of an individual Crotalus oreganus lutosus at the Cinder Butte in Wyoming (Graves and Duvall, 1990). However, Brito Crater hibernaculum in southeastern Idaho. 298 V. A. Cobb and C. R. Peterson

short-term Tb fluctuations of other individuals during peri- ods when air temperatures suddenly became cooler.

Variation in snake Tb was apparent throughout the win- ters, particularly early in hibernation and immediately be-

fore spring emergence. Because Tb variation was occurring, we hypothesized that winter Tb selection could be associated with factors such as sex, reproductive condition, and body size, as is prevalent in surface-active snakes (Peterson et al., 1993). However, we observed no such associations. Per-

haps the lack of Tb effects due to these factors is that snakes (Bennett and Dawson, 1976), especially rattlesnakes (Beck, 1985; Beaupre, 1993; Secor and Nagy, 1994; Beaupre and Duvall, 1998; Beaupre and Zaidan III, 2001; Dorcas et al., 2004), at 5-10°C already have a low metabolic rate and the

effects of small variations in Tb selection during hibernation would be minimal. Also, snake movements associated with microhabitat selection due to many but minor hibernaculum

Te changes would likely negate any advantage gained. Figure 8. Body temperature (°C) variation and spatial distribution Movements within hibernacula.—The greater amount of Crotalus oreganus lutosus (N = 7) on 4 November 1989 at the of underground movements in the first few months of hiber- Cinder Butte Crater hibernaculum in southeastern Idaho. nation (c.f. Figueroa et al., this volume) suggests that snakes are not merely entering hibernation and going immediately into a state of lethargy. Possible explanations are numerous, erable inter-individual variation in Tb was observed, as sup- but likely explanations include random exploratory behav- ported by data from this study. ior, sampling microhabitat based on temperature, airflow, The decrease in the availability of Tes outside the den conspecific chemicals, presence of conspecifics, and physi- prior to hibernation appears to be a primary stimulus for the cal structure of the hibernaculum. Early movements were cessation of surface activity. Although snakes may exhibit variable and more localized closer to the den entrance. Lo- activity at Tbs as low as 10°C (Avery, 1982), preferred and cality data on three snakes monitored for 2 yr indicated that field-selectedT bs are usually between 25-35°C (Lillywhite, individuals do not return to the exact same underground 1987; Mori et al., 2002). Studies on thermoregulation in C. locations each year. This could negate the idea of selection oreganus (Gier et al., 1989; Charland and Gregory, 1990; based on chemical cues of self or any other individual. Al- Cobb, 1994) and C. viridis (Graves and Duvall, 1993) sup- though it may be common for snakes to exhibit limited win- port the idea that C. oreganus and related species preferably ter activity in moderate climates (Jacob and Painter, 1980; select Tbs near 30°C. Using these data, we chose to use oper- Sexton and Hunt, 1980; Sexton and Marion, 1981; Zuffi et ative temperatures >25°C to examine the time available for al., 1999; Mori et al., 2002), Tes at our study site were too basking and achieving a high Tb outside the hibernaculum low for snakes to safely emerge once winter began. Even (Fig. 2). Once soil Tes decline, due to the onset of winter, if some days were sufficiently warm, the temperature of they likely prevent den temperatures near the hibernaculum the hibernaculum was relatively constant and snakes likely entrance from warming again until the next year. would not have detected the outside conditions 10 m back The overall Tb pattern for the duration of hibernation from the entrance. was similar between years, showing a gradual decrease the Movements of hibernating snakes have been linked first few months and then becoming more stable until spring to changes in Tbs either directly (Jacob and Painter, 1980; emergence. Body temperatures were similar between years Marion and Sexton, 1984; Macartney et al., 1989) or in- as well, and in mid-winter ranged between 4-10°C. Brown directly using inference from Tes (Sexton and Hunt, 1980; (1982), Macartney et al. (1989), and Weatherhead (1989) re- Sexton and Marion, 1981). Although there was a trend for ported similar Tbs during hibernation at comparable latitudes. snakes further back in the hibernaculum to have warmer The severity of the winter and weather variation between Tbs, this was not always the case. In fact, we only saw lim- years appeared to have little effect on snake Tb at the Cinder ited evidence of snakes moving with the thermal gradient, Butte Crater hibernaculum. Because snakes were ca. 3 m be- as proposed by Sexton and Marion (1981). In these cases, low ground and several meters from the entrance, they were snakes initially were closer to the hibernaculum entrance buffered from drastic temperature fluctuations. For instance, and probably were forced back by freezing temperatures. the 1990-91 winter was much colder in December and Janu- However, we observed many location changes that did not ary than the other years, yet mean snake Tb remained the lead to warmer Tbs (Fig. 7). We have no way of knowing if same. Potentially related observations included the death snakes could move vertically within the hibernaculum, but of one individual that remained closer to the entrance and we do not suspect this was the case. If snakes were follow- Thermal ecology of hibernation in Great Basin Rattlesnakes 299 ing thermal gradients, then all snakes should have moved ing and Environmental Laboratory. Doyle Markham, Tim further into the hibernaculum and would have had similar Reynolds, and Randy Morris of the Radiological and Envi-

Tbs. Yet, snakes often were >5 m apart and had Tb differenc- ronmental Science Laboratory assisted us in gaining access es of 4-5°C. Therefore, underground activity is more than to this snake population. Funding was provided by Idaho merely following thermal gradients. Marion and Sexton State Board of Education, EG&G Idaho/U.S. Department (1984) also discussed the reversal of thermal gradients as of Energy, and Idaho State University Department of Bio- stimulating egress. Like Weatherhead (1989), we found that logical Sciences. Steve Minkin and EG&G Idaho arranged temperatures inside the hibernaculum were so buffered that for us to drill into the hibernaculum. And, we are grateful thermal reversal occurred 3-4 weeks before snake emer- to the following persons for assisting us on various aspects gence, indicating a disassociation between the two. of this project: Jon Beck, Lisa Cobb, Bill Doering, Sarah Because we collected two-dimensional data on move- Doering, Mike Dorcas, and Steve Secor. This research was ments, we know that the hibernaculum has a thermal gradi- approved by the Idaho State University IACUC. ent ranging from approximately 5°C 10 m into the den to at times <0°C near the den entrance. Further back than 10 m, Literature Cited the hibernaculum temperatures could be warmer. In January, Aldridge, R. D. 1979. Female reproductive cycles of the our telemetered snakes had Tbs ranging from 5-8°C above the internal hibernaculum temperature. Several hibernation snakes Arizona elegans and Crotalus viridis. Herpeto- logica 35:256-261. studies have noted that snakes maintain Tbs slightly higher Aleksiuk, M. 1976. Reptilian hibernation: evidence of than Tes. However, in most cases, Tes were not being mea- sured near the snakes because of basic obstacles associated adaptive strategies in Thamnophis sirtalis parietalis. with den structure. Marion and Sexton (1984) examined C. Copeia 1976:170-178. viridis in an artificial den in Missouri and concluded that ———. 1977. Cold-induced aggregative behavior in the temperature differences between snakes and the substrate Red-sided Garter Snake (Thamnophis sirtalis parieta- could only be 2-4°C. lis). Herpetologica 33:98-101. White and Lasiewski (1971) hypothesized that large Anderson, J. A., and K. E. Holte. 1981. Vegetation devel- aggregate numbers of hibernating rattlesnakes could pro- opment over 25 years without grazing on Sagebrush- dominated rangeland in southeastern Idaho. J. Range duce enough metabolic heat to raise Tb 15°C above ambi- ent temperature. While tests from the field (Brown, 1982), Manage. 34:25-29. an artificial hibernaculum (Marion and Sexton, 1984), and Andersson, S. 2003. Hibernation, habitat and seasonal ac- the laboratory (Graves and Duvall, 1987) have not falsified tivity in the Adder, Vipera berus, north of the Arctic White and Lasiewski’s hypothesis, it does remain a poten- Circle in Sweden. Amph. Rept. 24:449-457. tial explanation for the temperature difference we observed. Avery, R. A. 1982. Field studies of body temperatures Our differences were only half of what White and Lasiewski and thermoregulation. Pp. 93-65 in C. Gans and F. H. (1971) predicted. From both field and laboratory observa- Pough (eds.), Biology of the Reptilia, Vol. 12. Academ- tions, Aleksiuk (1977) proposed that cold temperature-in- ic Press, New York, New York. duced aggregations of T. sirtalis may have thermoregulatory Bakken, G. S. 1990. Measurement and application of op- implications. The hibernaculum in our study contains >500 erative and standard operative temperatures in ecology. individuals (Sehman, 1977; Cobb, 1994), whereas White Am. Zool. 32:194-216. and Lasiewski (1971) used 150 hypothetical individuals in Beaupre, S. J. 1993. An ecological study of oxygen con- their calculations. In addition, most snakes hibernated with- sumption in the Mottled Rock Rattlesnake, Crotalus in a 5-m wide band 8-12 m into the den. Theoretically, this lepidus lepidus and the Black-tailed Rattlesnake, Cro- would place many snakes close to each other in a defined talus molossus molossus. Physiol. Zool. 66:437-454. space, with limited air flow at that distance from the den ———, and D. Duvall. 1998. Variation in oxygen con- entrance. We conclude that the possibility exists that there sumption of the Western Diamondback Rattlesnake may be enough metabolic heat generated to create some (Crotalus atrox): implications for sexual size dimor- phism. J. Comp. Physiol. B 168:497-506. of the Tb - Te differences observed, but a conclusive test is needed. Nevertheless, because natural hibernacula typically ———, and F. Zaidan, III. 2001. Scaling of CO2 produc- are complex structures and pose difficulties for experimen- tion in the Timber Rattlesnake (Crotalus horridus), with tal manipulations, future behavioral studies of underground comments on cost of growth in neonates and compara- winter activity may benefit best from artificial hibernacula tive patterns. Physiol. Biochem. Zool. 74:757-768. and simulated hibernation conditions in the laboratory. Beck, D. D. 1985. Ecology and energetics of three sympat- ric rattlesnake species in the Sonoran Desert. J. Herpe- Acknowledgments tol. 29:211-223. Bennett, A. F., and W. R. Dawson. 1976. Metabolism. We thank the U.S. Department of Energy for allowing Pp. 127-224 in C. Gans (ed.), Biology of the Reptilia, us to conduct this project on the Idaho National Engineer- Vol. 5. Academic Press, New York, New York. 300 V. A. Cobb and C. R. Peterson

Blouin-Demers, G., K. A. Prior, and P. J. Weather- Duvall, D., S. J. Arnold, and G. W. Schuett. 1992. head. 2000. Patterns of variation in spring emergence Pitviper mating systems: ecological potential, sexual by Black Rat Snakes (Elaphe obsoleta obsoleta). Her- selection, and microevolution. Pp. 321-336 in J. A. petologica 56:175-188. Campbell and E. D. Brodie, Jr. (eds.), Biology of the Brito, J. C. 2003. Seasonal and daily activity patterns of Pitvipers. Selva, Tyler, Texas. Vipera latastei in northern Portugal. Amph.-Rept. ———, L. J. Guillette, Jr., and R. E. Jones. 1982. En- 24:497-508. vironmental control of reptilian reproductive cycles. Brown, W. S. 1982. Overwintering body temperatures of Pp. 201–231 in C. Gans. and F. H. Pough (eds.), Bi- Timber Rattlesnakes (Crotalus horridus) in northeast- ology of the Reptilia, Vol. 13. Academic Press, New ern New York. J. Herpetol. 16:145-150. York, New York. ———. 1991. Female reproductive ecology in a northern ———, M. B. King, and K. J. Gutzwiller. 1985. Behav- population of the Timber Rattlesnake, Crotalus horri- ioral ecology and ethology of the Prairie Rattlesnake. dus. Herpetologica 47:101-115. Natl. Geogr. Res. 1:80-111. ———. 1992. Emergence, ingress, and seasonal captures ———, M. J. Goode, W. K. Hayes, J. K. Leonhardt, at dens of Northern Timber Rattlesnakes, Crotalus hor- and D. G. Brown. 1990. Prairie Rattlesnake vernal ridus. Pp. 251-258 in J. A. Campbell and E. D. Brodie, migration: field experimental analyses and survival Jr. (eds.), Biology of the Pitvipers. Selva, Tyler, Texas. value. Natl. Geogr. Res. 6:457-469. ———, and W. S. Parker. 1984. Growth, reproduction, and Galligan, J. H., and W. A. Dunson. 1979. Biology and demography of the racer, Coluber constrictor mormon, status of Timber Rattlesnake (Crotalus horridus) popu- in northern Utah. Pp. 13-40 in R. A. Seigel., L. E. Hunt, lations in Pennsylvania. Biol. Conserv. 15:13-57. J. L. Knight, L. Malaret, and N. L. Zuschlag (eds.), Ver- Gannon, V. P. J., and D. M. Secoy. 1985. Seasonal and tebrate Ecology and Systematics: A Tribute to Henry S. daily activity patterns in a Canadian population of the Fitch. Univ. Kansas Mus. Nat. Hist., Lawrence, Kansas. Prairie Rattlesnake, Crotalus viridis viridis. Can. J. ———, W. S. Parker, and J. A. Elder. 1974. Thermal Zool. 63:86-91. and spatial relationships of two species of colubrid Gier, P. J., R. L. Wallace, and R. L. Ingermann. 1989. snakes during hibernation. Herpetologica 30:32-38. Influence of pregnancy on behavioral thermoregulation Charland, M. B. 1989. Size and winter survivorship in in the Northern Pacific Rattlesnake Crotalus viridis neonatal Western Rattlesnakes (Crotalus viridis). Can. oreganus. J. Exp. Biol. 145:465-469. J. Zool. 67:1620–1625. Graves, B. M., and D. Duvall. 1987. An experimental ———, and P. T. Gregory. 1990. The influence of female study of aggregation and thermoregulation in Prairie reproductive status on thermoregulation in a viviparous Rattlesnakes (Crotalus viridis viridis). Herpetologica snake, Crotalus viridis. Copeia 1990:1089-1098. 43:259-264. Cobb, V. A. 1994. The ecology of pregnancy in free-ranging ———, and ———. 1990. Spring emergence patterns of Great Basin Rattlesnakes (Crotalus viridis lutosus). Ph.D. Wandering Garter Snakes and Prairie Rattlesnakes in Dissertation, Idaho State University, Pocatello, Idaho. Wyoming. J. Herpetol. 24:351–356. ———, and C. R. Peterson. 1999. Crotalus viridis lu- ———, and ———. 1993. Reproduction, rookery use, and tosus (Great Basin Rattlesnake). Mortality. Herpetol. thermoregulation in free-ranging, pregnant Crotalus v. Rev. 30:45-46. viridis. J. Herpetol. 27:33-41. Costanzo, J. P. 1986. Influences of hibernaculum microen- ———, ———, M. B. King, S. L. Linstedt, and W. A. vironments on the winter life history of the Garter Gern. 1986. Initial den location by neonatal Prairie Snake (Thamnophis sirtalis). Ohio J. Sci. 86:199–204. Rattlesnakes: functions, causes, and natural history in Diller, L. V., and D. R. Johnson. 1988. Food habits, con- chemical ecology. Pp. 285-304 in D. Duvall, D. Mül- sumption rates, and predation rates of Western Rattle- ler-Schwarze, and R. M. Silverstein (eds.), Chemical snakes and Gopher Snakes in southwestern Idaho. Her- Signals in Vertebrates, Vol. 4. Plenum Press, New York, petologica 44:228-233. New York. ———, and R. L. Wallace. 1984. Reproductive biol- Gregory, P. T. 1974. Patterns of emergence of the Red-sided ogy of the Northern Pacific RattlesnakeCrotalus ( Garter Snake (Thamnophis sirtalis parietalis) in the In- viridis oreganus) in northern Idaho. Herpetologica terlake region of Manitoba. Can. J. Zool. 52:1063-1069. 40:182-193. ———. 1982. Reptilian hibernation. Pp. 53-154 in C. Gans ———, and ———. 2002. Growth, reproduction, and sur- and F. H. Pough (eds.), Biology of the Reptilia, Vol. 13. vival in a population of Crotalus viridis oreganus in Academic Press, New York, New York. north central Idaho. Herpetol. Monogr. 16:26–45. ———. 1984. Communal denning in snakes. Pp. 57-75 in Dorcas, M. E., W. A. Hopkins, and J. H. Roe. 2004. Ef- R. A. Seigel, L. E. Hunt, J. L. Knight, L. Malaret, and fects of body mass and temperature on standard meta- N. L. Zuschlag (eds.), Vertebrate Ecology and System- bolic rate in the Eastern Diamondback Rattlesnake atics: A Tribute to Henry S. Fitch. Univ. Kansas Mus. (Crotalus adamanteus). Copeia 2004:145-151. Nat. Hist., Lawrence, Kansas. Thermal ecology of hibernation in Great Basin Rattlesnakes 301

Harvey, D. S., and P. J. Weatherhead. 2006. Hiberna- Mori, A., M. Toda, and H. Ota. 2002. Winter activity of the tion site selection by Eastern Massasauga Rattlesnakes Hime-habu (Ovophis okinavensis) in humid subtropics: (Sistrurus catenatus catenatus) near their northern foraging on breeding anurans at low temperatures. Pp. range limit. J. Herpetol. 40:66-73. 329-344 in G. W. Schuett, M. Höggren, M. E. Douglas, Hawley, A. W. L., and M. Aleksiuk. 1975. Thermal and H. W. Greene (eds.), Biology of the Vipers. Eagle regulation of spring mating behavior in the Red-sided Mountain Publishing, Eagle Mountain, Utah. Garter Snake (Thamnophis sirtalis parietalis). Can. J. Nussbaum, R. A., E. D. Brodie, and R. M. Storm. 1983. Zool. 53:768-776. Amphibians and Reptiles of the Pacific Northwest. Hirth, H. F. 1966. Weight changes and mortality of three University of Idaho Press, Moscow, Idaho. species of snakes during hibernation. Herpetologica Peterson, C. R., A. R. Gibson, and M. E. Dorcas. 1993. 22:8-12. Snake thermal ecology: the causes and consequences Jacob, J. S., and C. W. Painter. 1980. Overwinter thermal of body-temperature regulation. Pp. 241-314 in R. A. ecology of Crotalus viridis in the north-central plains Seigel and J. T. Collins (eds.), Snakes: Ecology and of New Mexico. Copeia 1980:799-805. Behavior. McGraw-Hill, Inc., New York, New York. King, M. B., and D. Duvall. 1990. Prairie Rattlesnake Pinder, A. W., K. B. Storey, and G. R. Ultsch. 1992. seasonal migrations: episodes of movement, vernal for- Estivation and hibernation. Pp. 250-274 in M. E. Feder aging and sex differences. Anim. Behav. 39:924-935. and W. W. Burggren (eds.), Environmental Physiology Kissner, K. J. and P. J. Weatherhead. 2005. Phenotypic of the Amphibians. University of Chicago Press, Chi- effects on survival of neonatal Northern Watersnakes, cago, Illinois. Nerodia sipedon. J. Anim. Ecol. 74:259-265. Reinert, H. K., and D. Cundall. 1982. An improved sur- Klauber, L. M. 1972. Rattlesnakes: Their Habits, Life gical implantation method for radio-tracking snakes. Histories, and Influence on Mankind. 2 vols. 2nd ed. Copeia 1982:705-708. University of California Press, Berkeley, California. Rosen, P. C. 1991. Comparative ecology and life history Lillywhite, H. B. 1987. Temperature, energetics, and of the racer Coluber constrictor in Michigan. Copeia physiological ecology. Pp. 422-477 in R. A. Seigel, J. 1991:897-909. T. Collins, and S. S. Novak (eds.), Snakes: Ecology and Ross, C. P., and J. D. Forrester. 1958. Outline of the geol- Evolutionary Biology. McGraw-Hill, Inc., New York, ogy of Idaho. Bull. Idaho Bureau Mines Geol. 15:1-74. New York. Secor, S. M., and K. A. Nagy. 1994. Energetic correlates Macartney, J. M. 1985. The ecology of the Northern Pa- of foraging mode for the snakes Crotalus cerastes and cific Rattlesnake, Crotalus viridis oreganus, in British Masticophis flagellum. Ecology 75:1600-1614. Columbia. M.S. Thesis, University of Victoria, British Sehman, R. W. 1977. Hibernaculum dynamics of the Great Columbia. Basin Rattlesnake (Crotalus viridis lutosus). M.S. The- ———, and P. T. Gregory. 1988. Reproductive biology sis, Idaho State University, Pocatello, Idaho. of female rattlesnakes (Crotalus viridis) in British Co- Sexton, O. J., and S. R. Hunt. 1980. Temperature rela- lumbia. Copeia 1988:47-57. tionships and movements of snakes (Elaphe obsoleta, ———, K. W. Larsen, and P. T. Gregory. 1989. Body Coluber constrictor) in a cave hibernaculum. Herpeto- temperatures and movements of hibernating snakes logica 36:20-26. (Crotalus and Thamnophis) and thermal gradients of ———, and K. R. Marion. 1981. Experimental analysis natural hibernacula. Can. J. Zool. 67:108-114. of movements by Prairie Rattlesnakes, Crotalus viridis, Marion, K. R., and O. J. Sexton. 1984. Body tempera- during hibernation. Oecologia 51:37-41. tures and behavioral activities of hibernating Prairie ———, P. Jacobson, and J. E. Bramble. 1992. Geo- Rattlesnakes, Crotalus viridis, in artificial dens. Prairie graphic variation in some activities associated with Nat. 16:111-116. hibernation in neartic pitvipers. Pp. 337–346 in J. A. Martin, W. H. 1992. Phenology of the Timber Rattlesnake Campbell. and E. D. Brodie, Jr. (eds.), Biology of the (Crotalus horridus) in an unglaciated section of the Pitvipers. Selva, Tyler, Texas. Appalachian Mountains. Pp. 259-277 in J. A. Camp- Storey, K. B., and J. M. Storey. 1996. Natural freezing bell and E. D. Brodie, Jr. (eds.), Biology of the Pitvi- survival in animals. Ann. Rev. Ecol. Syst. 27:365-386. pers. Selva, Tyler, Texas. Ultsch, G. R. 1989. Ecology and physiology of hiberna- ———. 2002. Life history constraints on the Timber Rat- tion and overwintering among freshwater fishes, tur- tlesnake (Crotalus horridus) at its climatic limits. Pp. tles, and snakes. Biol. Rev. 64:435-516. 285-306 in G. W. Schuett, M. Höggren, M. E. Douglas, Viitanen, P. 1967. Hibernation and seasonal movements of and H. W. Greene (eds.), Biology of the Vipers. Eagle the viper, Vipera berus berus (L.), in southern Finland. Mountain Publishing, Eagle Mountain, Utah. Ann. Zool. Fennici 4:472-546. McNab, B. K. 2002. The Physiological Ecology of Verte- Weatherhead, P. J. 1989. Temporal and thermal aspects brates: A View from Energetics. Cornell Univ. Press, of hibernation of Black Rat Snakes (Elaphe obsoleta) Ithaca, New York. in Ontario. Can. J. Zool. 67:2332-2335. 302 V. A. Cobb and C. R. Peterson

White, F. N., and R. C. Lasiewski. 1971. Rattlesnake Zar, J. H. 1984. Biostatistical Analysis. Prentice Hall, En- denning: theoretical considerations on winter tempera- glewood Cliffs, New Jersey. tures. J. Theor. Biol. 30:553-557. Zuffi, M. A. L., M. Macchia, P. Ioalè, and F. Giudici. Wright, A. H., and A. A. Wright. 1957. Handbook of 1999. Winter activity in a coastal population of Vi- Snakes of the United States and Canada. Comstock pera aspis (Reptilia, Viperidae). Rev. Ecol.-Terre Vie Publishing, Ithaca, New York. 54:365-374.