beck and jennings 2003

Herpetological Monographs, 17, 2003, 111129 2003 by The Herpetologists League, Inc.

HABITAT USE BY GILA MONSTERS:THE IMPORTANCE OF SHELTERS

DANIEL D. BECK1,2 AND RANDY D. JENNINGS3

1Department of Biological Sciences, Central Washington University, Ellensburg, WA 98926, USA3Western New Mexico University, Department of Natural Science, Silver City, NM 88062, USA

ABSTRACT: Many desert organisms cope with extreme and variable conditions by retreating to sub-surfacerefugia, yet little is known of the patterns of refuge use by most desert inhabitants. We investigated shelter useby the Gila Monster, Heloderma suspectum, an ectotherm closely tied to sub-surface refugia in a stronglyseasonal desert environment. We addressed hypotheses that ectotherms may use habitats based on availabilityof shelters, select shelters based on particular cues, respond to seasonal variation in refuge characteristics, andshow fidelity to specific retreats. Using radiotelemetry, we monitored microhabitat use by eight to ten GilaMonsters for six years, recording timing, frequency, and duration of visitation to over 250 specific shelters. Weused transects to assess shelter availability, recorded structural features of all shelters, and used dataloggersto monitor seasonal changes in microenvironments within subsets of shelters for periods up to two years.Shelters, and the habitats where they occurred, were not chosen by Gila Monsters at random. Helodermaspent more time in areas where a higher density of potential shelters was available, and selected shelters basedon rockiness, slope and entrance aspect, depth, humidity, and temperature. Gila Monsters showed longerresidence times and greater fidelity to shelters used during extreme periods (e.g., winter) and these patternswere paralleled by seasonal changes in the characteristics of shelters chosen. Winter shelters tended to besouth-facing, rockier, deeper, and warmer than those used in other seasons, whereas dry-summer shelterswere more soil-like in composition, cooler, and more humid. Gila Monsters showed strong fidelity to specificshelters, some of which were used by two or more lizards, sometimes concurrently. Seasonal variation in useof social shelters coincided with annual cycles of intraspecific behaviors and reproduction. Our resultsunderscore the importance of sub-surface refugia in the ecology of a sedentary desert ectotherm. Becausemany other ectotherms also spend significant periods sequestered in below-ground retreats, it is surprisingthat ecologists have not more extensively investigated this phenomenon. An understanding of the cues usedby desert ectotherms to choose refuge-sites, the spatial and temporal variability in refuge characteristics, andthe fidelity shown to particular retreats, may help better explain patterns of habitat selection, behavior,distribution, and abundance.

Key words: Desert; Gila Monster; Habitat; Habitat selection; Heloderma; Humidity; Microenvironment;Microhabitat; Refugia; Shelter-site; Temperature.

INTRODUCTION

Most small terrestrial animals are closelytied to sub-surface refugia, or shelters. Desertscorpions (Polis et al., 1886), beetles (Rasa,1995), spiders (Henschel, 1998; Humphreys,1978), mammals (Buffenstein, 1984; Schmidt-Nielson, 1964; Randall, 1993; Tracy andWalsberg, 2002), and reptiles (Bulova, 1994;Huey et al., 1989; Norris, 1953; Pough et al.,1978; Zimmerman et al., 1994) all cope withharsh environmental conditions and widefluctuations in availability of resources byexploiting a more benign environment withinshelters. Many ectotherms spend far moretime sequestered in below-ground refugiathan they do above ground (Avery, 1976; Beck,

1990; Huey, 1982; Huey et al., 1989; Nagy andMedica, 1986; Polis et al., 1986). For animalssuch as these, shelters provide escape fromextreme surface temperatures and from pred-ators, and allow access to favorable thermo-regulatory sites, foraging areas, and potentialmates (Blazquez and Rodriguez-Estrella,2001; Bulova, 1994; Christian et al., 1983;1984; Downes and Shine, 1998; Humphreys,1978; Schlesinger and Shine, 1994; Zimmer-man et al., 1994). Shelters are particularlyimportant for ectotherms because their bodytemperatures, and associated resting meta-bolic rates, are determined by the microenvi-ronment within such refugia (Christian et al.,1995; Huey, 1991; Zimmerman et al., 1994).The appropriate choice of a refuge, therefore,has profound implications for the ecology ofdesert animals. Indeed, retreat to refugia is2 CORRESPONDENCE: e-mail, [email protected]

111

arguably the most common and important waythat small desert animals persist in harsh andvariable environments.

Some desert animals, including species ofreptiles (Heatwole, 1977; Pianka, 1966), mam-mals (Nevo et al., 1979; Price, 1978), andarthropods (Krasnov and Shenbrot, 1996; Poliset al., 1986), may select habitats based onstructural microhabitat features, such as avail-ability and quality of subsurface refugia.Population densities of desert rodents areinfluenced by factors affecting the availabilityof burrows (Feldhamer, 1979), and kangaroorat burrows, in turn, influence the abundanceand diversity of ground-dwelling ectotherms(Hawkins and Nicoletto, 1992). Despite theobvious importance of shelters in the ecologyof organisms inhabiting harsh or seasonalenvironments, very little is known about refugeuse by most desert ectotherms, cues used tochoose shelters, fidelity to refuge sites, and thedegree to which shelter availability and qualityinfluences habitat use and local dispersion.

We investigated habitat use by a large,sedentary desert lizard, the Gila Monster(Heloderma suspectum Cope). We used radio-telemetry to determine how this desert ecto-therm may respond to spatial and temporalvariation in the availability and quality ofrefuge sites. In addition, we assessed thefidelity shown by Gila Monsters to particularretreat sites and explored potential roles suchrefugia play in the ecology of this and othersedentary desert ectotherms.

We address the following hypotheses andpredictions: 1) Habitat use is influenced byshelter availability and shelter quality. If so,we predicted Gila Monsters would spend moretime in habitats harboring a higher density ofpotential shelters, and in habitats possessingshelters with certain attributes. 2) Shelters areselected based on certain physical and micro-environmental attributes. If this were true, wepredicted shelters used by Gila Monsterswould differ in specific structural character-istics (such as depth, rockiness, and entranceaspect) and microenvironmental features (e.g.,temperature and vapor pressure) from poten-tial shelters that were not used. 3) Character-istics of shelters chosen by Gila Monsters varyin relation to seasonal changes of microenvi-ronments within shelters, and shelter micro-environments are influenced by physical

characteristics of shelters such as depth,rockiness, and entrance aspect. We predictedthat if seasonal changes in shelter micro-habitats were important, Gila Monsters wouldselect shelters with different characteristicsduring different seasons, and that structuraltraits of shelters would be correlated withmicroenvironments therein. 4) Gila Monstersshow strong fidelity to particular retreat sites.If this were true, we predicted a high pro-portion of shelters would be revisited by GilaMonsters and that re-used shelters might showcharacteristics that set them apart from others.We also predicted that shelter re-use may varyseasonally. We attempt to integrate our resultsinto a fuller understanding of the importanceof shelters in the ecology of this and otherdesert organisms.

Gila Monsters are excellent subjects forinvestigating patterns of refuge use becausemany aspects of their ecology tie them toshelters (Plate 1). They are nest predators,feeding upon juvenile mammals and eggs ofground nesting birds and reptiles (Beck, 1990;Bogert and Martn del Campo, 1956; Loweet al., 1986). Although finding such preyrequires considerable searching, helodermatidlizards spend the vast majority (9598%) oftheir time at rest in shelters (Beck, 1990; Beckand Lowe, 1991; Porzer, 1981). Gila Monstershave very low resting metabolic rates (Beckand Lowe, 1994), can consume large meals(>33% of adult body mass; Beck, 1990; Loweet al., 1986), and store fat within the bodycavity and tail (Bogert and Martn del Campo,1956). These traits make frequent foragingunnecessary. When on the surface, GilaMonsters are particularly vulnerable to pred-ators because, unlike most other lizards, theyare unable to escape by quickly sprinting away(Beck et al., 1995), but rely instead on a painful,venomous bite (Bogert and Martn del Campo,1956). Shelters may, therefore, be important asrefugia from potential predators.

In addition, shelters may serve as refugiafrom harsh environmental conditions on thesurface. Gila Monsters show a peak in annualactivity during May, a hot and very dry periodthroughout their range (Beck, 1990; Beck andLowe, 1991; Lowe et al., 1986) yet they alsoprefer relatively low body temperatures (2830 C) while active or basking (Beck, 1990;Porzer, 1981). They become physiologically

112 HERPETOLOGICAL MONOGRAPHS [No. 17

stressed at body temperatures approaching38 C (Bogert and Martn del Campo, 1956).Interestingly, helodermatid lizards also showrates of water loss that appear unusually high

for desert lizards (Lowe et al., 1986; Porter andBeck, unpublished data), a trait that may bebetter understood given their evolutionaryorigins in more mesic environments (Pregill

PLATE 1.Gila Monsters inhabit hot desert regions of North America and show a peak in activity during the driest timeof year. Refuge offered by shelter sites, as shown here, is the primary means by which these lizards, and other ectotherms,cope with harsh seasonal environments of deserts.

PLATE 2.Shelter-site locations (dots) used by six Gila Monsters radiotracked from April to July 1992, and monitored atleast weekly from November 1992 through June 1994. Each color denotes an individual Gila Monster (red 5 male no. 1;yellow 5 female no. 2; green 5 male no. 5; blue 5 male no. 8; orange 5 female no. 9; and pink 5 female no. 10).Minimum convex polygon home ranges are indicated by solid lines (see also Table 5). Not all lizards monitored in thisstudy are shown on this photograph.

2003] HERPETOLOGICAL MONOGRAPHS 113

et al., 1986; Stevens, 1977). Potentially highrates of water loss while Gila Monsters areactive during the dry season may make sheltersimportant not only for escaping the heat, butalso for reducing desiccation.

METHODS

Site Description

Our study area is a Chihuahuan semidesertgrassland (Brown, 1982) in southwestern NewMexico at 328 44.59N latitude and 1088 42.09Wlongitude, 5 km NW of Red Rock, GrantCounty. Topography ranges from 1250 to 1500m in elevation and consists of rocky slopes andsmall dry canyons along the Gila River.Dominant vegetation consists of catclaw (Aca-cia greggi, Mimosa biuncifera), snakeweed

(Guitierezia sp.), ocotillo (Foquieria splen-dens), mesquite (Prosophis juliflora), creosote-bush (Larrea tridentata), juniper (Juniperusmonosperma), cacti (Opuntia spp.), sotol(Dasilyron wheeleri), and grasses. Predomi-nant rocks are strongly weathered Precambriangranites of the Burro Mountain batholith(Hewitt, 1959). Annual precipitation averages363 mm, with 60% falling from July throughOctober (Red Rock weather station, averagingperiod 5 19711993). Temperature variationcan be extreme, with air temperatures com-monly exceeding 40 C in summer anddropping below 0 C in winter (Fig. 1A).Relative humidity also varies greatly, withdaily means below 5% during dry summerperiods and over 50% during some wetsummer and winter periods (Fig. 1B). These

FIG. 1.Variation in ambient temperature (A) and relative humidity (B) over a typical one-year period on the Red Rockstudy site.


patterns of temperature, precipitation, andrelative humidity result in five distinct seasonsat our study site: Winter (December throughFebruary), Spring (March through mid-May),Dry Summer (mid-May through mid-July),Wet Summer (with the onset of the summerrains, mid-July through September) and Fall(OctoberNovember; Fig. 1).

General Study Procedures

We focused on Gila Monsters encounteredwithin a 4.5 km2 core area near a perennialspring. Between April 1992 and December1998, eight adult Gila Monsters (four malesand four females) were outfitted with implant-able, temperature-sensing radiotransmitters(Model IMP-150, Telonics Telemetry, Mesa,AZ). Two more lizards were outfitted withradiotransmitters in 2000 to conduct experi-ments on shelter selection (see below). Trans-mitters were surgically implanted using thetechniques of Beck (1990). Size, sex, andreproductive condition were recorded for eachlizard; movements and shelter use weremonitored with a directional antenna anda TR-2 or TR-4 receiver (Telonics). FromMay 1992 to July 1994, we closely monitoredactivity patterns and shelter use by theselizards (three or more times/wk during theactivity season; once/wk during the winter);more than 200 shelters were marked andcharacterized. During this two year period, weidentified areas showing high use and low use(defined below) by Gila Monsters within thecore area of the study site. From August 1994to December 1998 we monitored shelter useby these lizards less frequently (weekly tomonthly), characterized additional sheltersobserved, and gathered microhabitat datanecessary to address the hypotheses outlinedabove.

Throughout the study we marked shelterswith small rock cairns and tagged them forfuture identification. We marked locationsof shelters on maps drawn from aerialphotographs, and calculated home rangesusing the minimum convex polygon method(Jennrich and Turner, 1969; Rose, 1982). Werecorded the following characteristics: heightand width of shelter entrance, depth (accurateup to about 1 m, after which shelter depth wasestimated using transmitter signal), aspect ofentrance and slope on which shelter occurred,

and rockiness of shelter. To index shelterrockiness, we used an ordinal scale rangingfrom 1 to 5 (1 5 silt-sand, 2 5 sand/gravel,3 5 gravel, 4 5 gravel/rock, 5 5 rock) for boththe roof and the floor of the shelter. Wheneverpossible, we recorded the number of days eachGila Monster spent in each shelter. A fewshelters were not completely characterized;shelters for which we were unable to de-termine specific characteristics (e.g., depth,duration of residency) were not included inanalyses, therefore sample sizes vary slightlyamong analyses.

Hypothesis 1: Is Habitat Use Influenced byShelter Availability and Shelter Quality?

To understand the role of shelter availabilityin habitat use by Gila Monsters, we set up 20100 m32 m transects, 10 in high-use and 10 inlow-use areas, and identified all potentialshelters along the transects. We defined low-use areas as portions of our study site that wereobviously accessible to Gila Monsters (i.e.,within home ranges of lizards we monitored),but that showed , two visit/ha by lizards wefollowed with radiotelemetry during the first2 yr of our study (19921994; see Plate 2).High-use areas received more than twovisits/ha during the same period. We defineda potential shelter as a stable hole or creviceat least 5 cm in diameter and 20 cm deep, thedimensions of the smallest shelter observed toharbor a Gila Monster. We tested for differ-ences in the number of potential sheltersbetween high- and low-use areas using theMann-Whitney test. If habitat use were notinfluenced by shelter availability, we expectedto see no difference in potential shelterabundance between high-use and low-useareas.

To determine whether the quality of poten-tial shelters influenced habitats used by GilaMonsters, we compared features of 28 poten-tial shelters in low-use areas with those of 35potential shelters in high-use areas. Potentialshelters were located by haphazardly search-ing regions of the study area. Because someshelter characteristics were continuous vari-ables (i.e., entrance dimensions and shelterdepth), and others were discrete (roof andfloor rockiness, aspect), and because not allvariables satisfied parametric assumptions ofnormality, we employed a class of nonpara-


metric statistics using ranked data to test thisand other hypotheses (Conover, 1980; Con-over and Iman, 1981). In this case, character-istics of potential shelters (except aspect) wereranked and subjected to a one-way analysisof variance (PROC RANK and PROC GLM,SAS Inst, 1985) to test the hypothesis of nomean ranked difference between shelters inhigh-use and low-use areas. To test hypoth-eses relating to shelter and slope aspect,we grouped observations into one of eightdirectional categories (N, NE, E, SE, S, SW,W, NW) and used G-tests (with Williams cor-rection, Sokal and Rohlf, 1995). If habitat usewas not influenced by the quality of potentialshelters, we expected no differences in poten-tial shelter characteristics between high-useand low-use areas.

Hypothesis 2: Shelter Selection

To address whether Gila Monsters selectedshelters based on certain attributes, we com-pared the physical characteristics of actualshelters in high-use areas with potentialneighboring shelters not known to be used bya Gila Monster. We defined an actual shelteras a shelter observed to have been used bya Gila Monster for at least several hours andwe treated each shelter used by a Gila Monsteras an independent sample. A sample of 48 suchshelters used by seven radio-equipped lizardswas haphazardly chosen from throughout thestudy area. To reduce biases associated withmicrohabitat, aspect, and substrate type, wepaired each actual shelter with a potentialneighboring shelter by walking an ever-increasing spiral with the actual shelter at itsvortex until a potential shelter (fulfillingpotential shelter criteria as described above)was encountered. We assumed that potentialshelters were unused because we neverobserved any to be inhabited by a GilaMonster over the 6 yr they were monitored,although a possibility always existed that theiruse went undetected. Differences in structuralcharacteristics (described above) betweenactual vs. potential shelter pairs were com-pared using the Wilcoxon paired-samplet-tests (PROC MEANS, SAS Inst., 1985). Toeliminate biases associated with treating eachshelter (rather than each lizard) as an in-dependent sample (and to avoid pseudorepli-cation), we used degrees of freedom based on

the number of Gila Monsters (7) selectingactual shelters rather than the number ofshelter pairs (48).

To explore the role of microenvironment(temperature and vapor pressure deficit) inshelter selection, we monitored an additional17 pairs of actual vs. potential neighboringshelters used by five Gila Monsters during theactivity season between 28 April and 10 June2000. For each pair, a datalogger was placed0.4 to 1.0 m deep within an actual shelterinhabited by a Gila Monster, and a seconddatalogger placed (at the same depth) in thenearest potential neighboring shelter thatmatched the inhabited shelter in diameter,rockiness, and aspect. We simultaneouslyrecorded temperature and humidity in eachpair of shelters using Hobo XT and Rhdataloggers (see below) for a minimum of 48 hduring and after occupancy. Vapor pressuredeficit (in kilopascals, kPa) was calculated fromdatalogger measurements of temperature andrelative humidity. Temperature and vaporpressure deficit, during and after occupation,were ranked and compared between the 17actual and potential pairs using a blockedthree-way ANOVA (PROC RANK and PROCGLM, SAS Inst, 1985). Individual lizards wereblocked in the analysis to eliminate potentialbiases and to avoid pseudoreplication (Heffneret al., 1996; Hurlbert, 1984). Whether a shelterwas used by a Gila Monster or not (actual vs.potential shelters) and periods of occupancy(current vs. post-occupancy) were used asmain effects. Interactions among main effectswere also evaluated. We compared periods ofpost-occupancy with current occupancy tocontrol for possible changes in shelter micro-environment, especially vapor pressure deficit(due to evaporative water loss), brought aboutby the presence of a Gila Monster within theshelter. If microenvironment were not a factorin shelter selection by Gila Monsters, wewould expect to see no differences betweenactual and potential shelters in internal tem-perature or vapor pressure deficit.

Hypothesis 3: Seasonal Effects

Based on observations from 19921994,we use the term shelter type to differentiateshelters used in each of three focal seasons:Winter, when shelters presumably serve asrefugia from low temperature, Dry Summer,


when shelters serve as refugia from highsurface temperatures and high vapor pressuredeficit, and Spring when shelters serve moreas temporary refugia between foraging boutsand as potential mating sites.

To assess whether structural attributes ofshelters chosen by Gila Monsters differedamong seasons, we used Kruskal-Wallis tests(PROC RANK and PROC GLM, SAS Inst.,1985) with ranked shelter characteristics(except aspect) as dependent variables andseason of use (i.e., shelter type) as the classvariable. To test hypotheses relating to shelterand slope orientation, we grouped observa-tions into one of eight directional categories(N, NE, E, SE, S, SW, W, NW) and usedG-tests (with Williams correction, Sokal andRohlf, 1995).

To explore seasonal differences in micro-environments among shelter types, we moni-tored temperature and vapor pressure deficitwithin actual Gila Monster shelters betweenJanuary 1995 and August 1996. We pre-setHobo XT and Rh dataloggers (4.534.531.5cm; Onset Instruments, Inc.) to record tem-perature and relative humidity every 72 minand placed them up to 1 m deep withinshelters. Vapor pressure deficit (kPa) wascalculated using relative humidity and tem-perature. Shelters were unoccupied by GilaMonsters while the dataloggers were in place.We attempted to place dataloggers at depthswithin shelters similar to those chosen by GilaMonsters (based on our observations frommonitoring radio-implanted lizards) to assessthe microenvironment experienced by a hypo-thetical lizard inhabiting that particular shelter(Vitt and Sartorius 1999).

We monitored microenvironments withina total of 27 shelters: 9 Winter shelters, 9Dry Summer shelters and, 9 Springshelters. To test for differences in micro-environments among these three shelter types,we subjected ranked temperature and vaporpressure deficit to a blocked, two-way, nestedanalysis of covariance (PROC RANK andPROC GLM, SAS Inst., 1985). Temperatureand relative humidity were sampled for one tofour-week periods between January 1995 andAugust 1996. During a given sampling period,3 shelters of each of the 3 shelter types (a totalof nine shelters at any one time) weremonitored with a temperature and a humidity

datalogger. Temperature and vapor pressuredeficit data were ranked and then averaged foreach sampling period. Sampling periods werenested within season (Winter, Spring, Drysummer, and Wet summer/Fall) and individ-ual shelters were nested within shelter type(Winter, Dry summer, and Spring). Theranked depth at which dataloggers werepositioned within the shelter was used asa co-variate within the overall analysis. Post-hoc (a 5 0.05) differences in microenviron-ments among seasons and shelter types wereassessed using pairwise comparisons of leastsquares means. If Gila Monsters did not selectshelters with different microenvironmentalattributes during different seasons, we wouldexpect to see no differences in shelter micro-environment across seasons.

To assess the potential influence of shelterattributes on shelter microenvironment, weperformed stepwise multiple regressions (withmaxr option) by season (PROC STEPWISE,SAS Inst., 1985). Because temperature andvapor pressure deficit were regressed againstordinal data (roof and floor substrate), allvariables used in the analysis were ranked(PROC RANK, SAS Inst., 1985). Ranks oftemperature and vapor pressure deficit, inturn, were used as dependent variables. Ranksof four shelter attributes: rockiness of the roofand floor, aspect, and depth, were used asindependent variables. Using this approach,we determined which of the four physicalaspects of shelters were best predictors ofmicroenvironment.

Hypothesis 4: Shelter Fidelity

To determine the extent to which shelterswere revisited, seasonal effects on shelterfidelity, and whether revisited shelters differedfrom those not observed to be reused, werecorded the number of times individualshelters were revisited (and by how manylizards), the proportion of total shelters thatwere revisited throughout each year (andseason), and, whenever possible, the GilaMonsters duration of residency within eachshelter. To eliminate the potential for pseu-doreplication, individual lizards, not shelters,were sample units in all statistical tests; samplesizes and degrees of freedom were adjustedaccordingly. Tests for mean rank differencesbetween reused and other shelters were


conducted using the Kruskal-Wallis test(PROC MEANS, PROC RANK and PROCGLM, SAS Inst., 1985) and the Mann-Whitneytest. Seasonal effects upon duration of resi-dency (d) were assessed using the Kruskal-Wallis test. To test for seasonal effects onfrequency of shelter reuse, and for differencesin entrance aspect between reused and othershelters, we used G-tests (with Williamscorrection, Sokal and Rohlf, 1995).

RESULTS

Habitat Use, Shelter Availability,and Shelter Quality

We characterized and monitored 259 shel-ters used by 10 Gila Monsters on the study site.The spatial relationships among most sheltersused by 6 of those lizards are shown in Plate 2.Home range sizes varied greatly, from 6.2 ha to104.8 ha (mean 5 58.1, SE 5 12.4, see plate2). Although Gila Monsters generally used pre-existing cavities as shelters, we occasionallyobserved them modify shelters, especially soilburrows, by digging. High use areas (>2 visits/ha) had more potential shelters (mean 5 3.5potential shelters per transect, SD 5 2.59)than low use areas (mean 5 0.2 potentialshelters per transect, SD 5 0.42; Mann-Whitney U 5 94, n1 5 10, n2 5 10, P ,0.05). High use areas harbored potentialshelters with significantly rockier roof struc-ture (median 5 5 [rocky]) than potentialshelters found in low use areas (median 51.5 [soil/gravel]; F1,62 5 16.0, P 5 0.0001).Potential shelter entrances and slope aspectsin high use areas were significantly morestrongly oriented toward the southeast andsouthwest than were those in low-use areas(GWilliams 5 17.7, P 5 0.013; and GWilliams 513.76, P5 0.033 for entrance aspect and slopeaspect, respectively; Fig. 2).

Shelter Selection

We found actual shelters were significantlydeeper (mean 5 55.0 cm, SD 5 37.04) thannearest potential unused neighboring shelters(mean 5 39.3 cm, SD 5 25.92; t 5 3.89,df5 6, P5 0.009). We detected no differencesbetween actual and potential neighboringshelters in other physical characteristicstested. Actual shelters selected during the ac-tivity season (AprilJune) had a significantly

smaller rank mean vapor pressure deficit(mean 5 0.253 kPa, SE 5 0.008) than poten-tial neighboring shelters (mean 5 0.308 kPa,SE 5 0.008), even after variance explained bydifferences among lizards was removed fromthe model (F1,60 5 9.81, P 5 0.0027; Fig. 3).No differences in rank mean temperaturewere detected for those same shelter pairs(Fig. 3). Shelters did not differ in rank meanvapor pressure deficit nor in rank meantemperature between current occupancy andpost-occupancy periods, nor were there anysignificant interactions between the maineffects.

Seasonal Effects

Many characteristics of shelters used by GilaMonsters varied significantly across seasons.Shelters used during the winter, fall, and drysummer were deeper than those selectedduring the wet summer and spring (F4,169 515.43, P 5 0.001; Table 1). Shelters differedsignificantly across seasons in both roof

FIG. 2.Plots of the aspect of potential shelterentrances, and slopes on which they occurred (percent ofsample falling in each of eight directions). Aspect differedsignificantly between high-use areas (>2 visits/ha/2 yr) andlow-use areas (,2 visits/ha/2 yr). Each concentric circlerepresents 25% of the sample.


(F4,240 5 6.53, P 5 0.0001) and floor (F4,240 54.07, P 5 0.003) substrate, with winter shel-ters being rockier, and dry-summer sheltersbeing composed of more gravel and soil (Table1). Entrance aspect also differed signif-icantly among shelters used in different sea-sons (GWilliams 5 47.3, df 5 28, P , 0.025).Fall and winter shelters occurred on south-facing slopes with entrances also orientedtoward the south; a preponderance of springshelter entrances faced east; and summershelters were more variable in orientation(Fig. 4). Gila Monsters had longer residencetimes in shelters in the winter and fall, andswitched shelters more frequently during

spring and summer (Kruskal-Wallis H 536.68, df 5 4, P , 0.001, Table 2).

As expected, rank mean temperature withinGila Monster shelters differed significantlythroughout the year (F3,20 5 36.35, P 50.0001; Table 3). Rank mean vapor pressuredeficit did not differ across seasons, although itdid differ among sampling periods withinseasons (F20,148 5 2.42, P 5 0.001). Indepen-dent of seasonal or sampling period variation,mean rank temperature and vapor pressuredeficit also differed among shelter types (Fig.5). The dry-summer shelters were significantlycooler throughout the year than winter andspring shelters (F2,1615 25.95, P5 0.0001; Fig5A). Dry-summer shelters also exhibited thesmallest mean rank vapor pressure deficitduring the year, which was significantly smallerthan that of spring shelters (F2,148 5 3.31, P 50.039). Winter shelters exhibited intermediaterank vapor pressure deficits (Fig. 5B). Withinseasons there were also significant differencesin temperature and vapor pressure deficitamong shelter types (based on post-hoc pair-wise comparisons of least squares means, P ,0.05). Dry-summer refugia were generallycooler and more humid than other sheltertypes, while winter refugia were significantlywarmer during winter (Fig. 5A and B).

Two structural features of shelters signifi-cantly predicted shelter microenvironment(Table 4). Roof rockiness exhibited positivepartial regression coefficients with tempera-ture (during fall, wet summer, dry summerand winter) and vapor pressure deficit (drysummer; Table 4). The addition of otherindependent variables provided no significantincrement to R2 or possessed significant pre-dictive power. Based on these analyses, rockyshelters were generally warmer and drierduring the indicated seasons than soil-likeshelters. During the dry summer, shelter

FIG. 3.Comparison of temperature and vapor pressuredeficit from 17 Gila Monster shelters paired with similarpotential neighboring (but unused) shelters monitored for48 h (during and after occupancy) during the activityseason. Values reported are means 6 one standard error(horizontal bars).

TABLE 1.Seasonal variation in structural traits of shelters used by Gila Monsters. Roof and floor substrates were scoredfrom 1 (silt/sand) to 5 (rock), as described in text. Values are mean substrate scores 6 one standard error; those with the

same superscript did not differ significantly.

Spring Dry summer Wet summer Fall Winter

Depth (cm) 49.4 6 6.351 89.5 6 6.292 52.1 6 8.981 110.7 6 11.662,3 128.0 6 9.953

n 54 55 27 16 22Roof structure 4.5 6 0.182,3 3.6 6 0.161 3.9 6 0.231,2 4.3 6 0.331,2,3 5.0 6 0.293

Floor structure 3.0 6 0.16 1 2.7 6 0.141 2.9 6 0.211 2.5 6 0.291 3.7 6 0.262

n 68 90 41 20 26


depth exhibited a negative partial regressioncoefficient with temperature (the deeper theshelter the cooler). Collectively, these resultsindicate that physical aspects of shelterssignificantly predicted shelter microenviron-ment (Table 4).

Shelter Fidelity

Individual Gila Monsters varied consider-ably in the extent to which they reused shelters,ranging from 23% to 54% (mean 5 42.9%,SD 5 11.4, n 5 7, based on per cent of totalshelters used by each lizard; Table 5). Thedegree that shelters were re-used was clearlya function of duration of monitoring; anincreasingly larger proportion of shelters was

reused by Gila Monsters as the study pro-gressed (Fig. 6). Shelters repeatedly used byindividual Gila Monsters had significantlyrockier roofs (mean 5 4.4 5 rock, SE 5 0.29)than shelters that were used once (mean 53.6,5 gravel/rock, SE5 0.27; F1,155 4.54, P50.05), but were otherwise similar in structureand aspect to shelters not observed to bereused. Gila Monsters resided significantlylonger in re-used shelters (mean 5 34.8 d/shelter, SE 5 7.37) than in other shelters(mean 5 4.4 d/shelter, SE 5 0.56; WilcoxinW5 28, n5 7, P5 0.02; Fig. 7). Shelter reusealso was strongly influenced by season (G 520.5, df 5 4, P , 0.005; Fig. 8). A largernumber of shelters was reused during the dry

FIG. 4.Seasonal patterns in the aspect of Gila Monster shelter entrances, expressed as percent of sample size falling ineach of eight directions. Each concentric circle represents 10% of the sample. Numbers in parentheses are sample sizes.


summer (41 of 101 5 40.6%), but a greaterproportion of shelters was reused during thewinter (16 of 22 5 72.7%, Fig. 8).

Of 94 shelters that were reused, 44 (47%)were used by more than one individual (socialshelters, Fig. 9A). Twenty-four of these(54.5%) were used by two or more lizardsconcurrently, most commonly during the latespring and early dry-summer (Fig. 9B) matingseason, when we also observed males andfemales together in shelters. Social sheltersalso differed from individual shelters in theirspatial arrangement. The majority of socialshelters was located near the core of our studysite where lizard home ranges overlapped,whereas the majority of shelters reused byindividual lizards were overwintering refugialocated near the edges of home ranges. Someshelters that were used as winter refugia,however, were also reused during the drysummer.

DISCUSSION

Habitat Use, Shelter Availability,and Shelter Quality

Gila Monsters spent more time in areaswhere a higher density of potential shelterswas available, where potential shelters tendedto be rockier, and where slopes and shelterentrances were oriented more to the south.Together, these results suggest Gila Monstersselected areas of the study site based on theavailability and quality of potential sheltersites. They also suggest that the availability ofsuitable refugia played a role in habitatselection by Heloderma and, in turn, influ-enced their patterns of local dispersion. Otherterrestrial vertebrates and macroinvertebratesin arid regions also show dispersion patternsthat are strongly influenced by the availabilityof shelters and appropriate soils in whichsuitable refugia may be constructed (Kinlaw,1999; Polis et al., 1986).

Shelter Selection

Our results underscore the importance ofmoisture as a factor in microhabitat selectionby Gila Monsters. Comparisons betweenactual vs. potential shelters in our studyrevealed that vapor pressure deficit, nottemperature, significantly influenced shelterselection during the peak activity season.Humidity was also shown to influence activitypatterns in a pit viper inhabiting a seasonaltropical environment (Daltry et al., 1998).When stressed by lack of water, DesertTortoises retreat into burrows (Ruby et al.,1994) and when sleeping or dormant, theirwater loss rates are reduced (Wilson et al.,2001). The consequences of microhabitat useby ectotherms have traditionally been ad-dressed in the context of thermal biology (e.g.,Cowles and Bogert, 1944; Diaz, 1997; Huey,1991; Huey et al., 1989; Humphreys, 1978;Porter and Tracy, 1983; Porter et al., 1973).However, the role of moisture in refugechoice by desert ectotherms has not, to ourknowledge, been specifically investigated (butsee Bulova, 2002). Vapor pressure deficit ispresumably important because burrow micro-environments with lower vapor pressuredeficits provide a reduced gradient forevaporative water loss, thereby facilitatingconservation of body water. A fruitful areaof future research would be to explore theconsequences of spatial and temporal varia-tion in moisture content of desert micro-environments, particularly refugia, for theecology of desert ectotherms (see Tracy andWalsberg, 2002).

Shelter-site selection may also have impor-tant consequences for energy conservation.Gila Monsters spend the vast majority of theirtime relatively deep within shelters (>0.75 m),where body temperatures equilibrate toambient temperature (D. Beck, 1990; personalobservation). At our study site, Helodermasuspectum spent >50% of the year at bodytemperatures ,20 C (unpublished data). Such

TABLE 2.Seasonal patterns of shelter occupancy by Gila Monsters on the study site between November 1992 and July1994. Values shown are means 6 one standard error; those with the same superscript did not differ significantly.

Spring2 Dry2 summer Wet2 summer Fall3 Winter1

n 48 43 24 22 9Days in same shelter 5.1 6 1.00 4.5 6 1.08 4.2 6 0.77 13.6 6 2.46 88.7 6 1.21


behavior could be regarded as sub-optimal forvertebrate ectotherms, because foraging op-portunities are lost when time is spent atreduced body temperatures while hidingfrom predators in shelters (Downes, 2001).Moreover, conserving heat while inside shelter-sites may be important for lizards that are ac-tive on a daily basis (Christian et al., 1984; Hueyet al., 1989). Helodermatid lizards, on the otherhand, may actually benefit from lower bodytemperatures, and consequently reduced en-ergy expenditure, during long periods spentinside shelters. This idea is somewhat sup-ported by a previous finding that restinghelodermatids have among the lowest meta-bolic rates of any lizard measured, witha metabolic Q10 of 3 between 25 C and 15 C(metabolic rate decreased three-fold with a 10 Cdrop in temperature) (Beck and Lowe, 1994).Hence, energy savings were greater at reducedtemperatures. For Gila Monsters, the conse-quent reduction in metabolic rate may repre-sent considerable energy savings particularlybecause frequent foraging activity on thesurface is not necessary.

Other reptiles show voluntary hypothermia,or metabolic depression, even when environ-mental conditions provide thermal options foractivity and precise thermoregulation (Chris-tian et al., 1984; 1996a, 1999a; Dorcas andPeterson, 1998; Peterson, 1987; Regal, 1967;Zimmerman et al., 1994). The energeticadvantages of voluntary hypothermia havebeen clearly demonstrated by several lizardspecies inhabiting seasonal tropical environ-ments in Australia (Christian et al., 1996a, b,1999a, b). Reduced body temperatures ofindividuals in shelters also translate intoreduced rates of water loss (Mautz, 1980,1982; Zimmerman et al., 1994). The choice ofan appropriate refuge, therefore, is an impor-tant part of energy and water savings enjoyedby ectotherms that remain hidden belowground (Pough, 1980).

Seasonal Effects

We found that shelters selected by GilaMonsters during the hottest, driest time ofyear (dry-summer shelter type) were signif-icantly less rocky, and remained significantlycooler and more humid throughout the yearthan those chosen in other seasons. Wintershelters, occupied for periods averaging nearlythree months, were significantly rockier,warmer, and deeper than those used duringother seasons. These results indicate GilaMonsters altered their choice of shelters asconditions changed seasonally. During thewinter, when surface temperatures werefrequently below freezing, Gila Monstersselected shelters with south-facing entranceson south-facing slopes that were rockier anddeeper, and consequently warmer than shel-ters used in other seasons. During the drysummer, when surface temperatures were hot,and relative humidity very low, Gila Monstersselected shelters that were more variable inaspect, more soil-like in composition, and,consequently, cooler and moister than sheltersused in other seasons. During the spring, whensurface conditions were more benign and GilaMonsters were foraging, basking, searching formates, and pairing within shelters, shelterswere oriented more strongly toward the east,rockier, shallower, and, more variable intemperature and humidity than other sheltertypes.

Our results also suggest that the physi-cal features of a shelter influenced itsmicroenvironment. Shelter rockiness anddepth significantly predicted shelter micro-environment, suggesting that physical charac-teristics influenced temperature and vaporpressure within Gila Monster shelters. Shal-low, rocky shelters tended to be hotter anddrier than deeper, soil-like shelters. Shelterswith soil floors retained moisture better thanrocky shelters and thus provided better refugefrom the desiccating effects of low humidity

TABLE 3.Seasonal variation in temperature and vapor pressure deficit (VPD in kPa) in 27 Gila Monster sheltersmonitored from January 1995 to August 1996. Values are means for all shelters combined 6 one standard error. Values

with the same superscript did not differ significantly.

Spring Dry summer Wet summmer/fall Winter

Temperature 8C 19.8 6 0.212 26.8 6 0.283 27.2 6 0.263 13.3 6 0.2531

VPD 0.297 6 0.0181 0.260 6 0.0202 0.210 6 0.0172 0.236 6 0.0172


during the hot dry season. South-facing rockyshelters presumably captured more heat fromthe sun, especially during winter; entrances toeast-facing shelters presumably warmed morequickly in the morning.

In other species of lizards and snakes rocksize, especially thickness, affects the thermalproperties of shelters which, in turn, influencerefuge choice (Huey et al., 1989; Kearney,2002). Velvet geckos select rocky-creviceshelters based on rock size, width, and heightabove ground (Schlesinger and Shine, 1994).Thermal properties are important cues inselection of refugia by other ectotherms aswell (Burger and Gochfeld, 1991; Christian etal., 1984; Downes and Shine, 1998; Huey et al.,

1989; Humphreys, 1978). Rockiness may alsobe associated with shelter stability. At ourstudy site, shelters found in soil and gravelsubstrata were susceptible to collapsing, andthus more ephemeral and less predictable asstable refugia.

Based on our results, aspect of shelter-siteentrances may also be important to desert-dwelling organisms. Aspect undoubtedly in-fluences thermoregulatory options at theshelter-site entrance. During the spring(March through May), Gila Monsters showedpeak activity in the morning and often baskedat shelter entrances prior to activity (Beck,1990; this study). Their preference for east-facing shelters may therefore have allowed

FIG. 5.Seasonal variation in mean temperature (A) and vapor pressure deficit (B) among three shelter typesmonitored in this study. Horizontal bars designate 6 one standard error of the mean.


more efficient thermoregulation at shelterentrances at a time of year when baskingbehavior was more frequent.

Gila Monsters chose deeper shelters duringseasons when surface conditions were mostextreme (Winter and Dry Summer). Temper-ature and humidity stabilize beyond a certaindepth within a shelter (Kay and Whitford,1978; van Heerden and Dauth, 1987). Hence,deeper shelters likely provided a more stablemicroenvironment, allowing better refugefrom harsh surface conditions (Kearney,2002; Kinlaw, 1999). Desert Tortoises mayalso reduce water loss rates by choosingdeeper burrows during hot, dry periods(Wilson et al., 2001).

There have been few investigations ofseasonal variation in refuge use by desertectotherms. Burrow use by Desert Tortoises(Gopherus agassizii) varies seasonally accord-ing to their social and thermal environments(Bulova, 1994; Zimmerman et al., 1994). Insouthern Australia, geckos respond to seasonalchanges in thermal microenvironments byswitching from retreats beneath rocks in springto cooler, deeper crevices in summer (Kear-ney, 2002). Similarly, varanid lizards useshallower burrows in summer than in winter

(King, 1980). These results, coupled with ourfindings, suggest that temporal variation insubsurface refugia are an important factor inhabitat use by desert ectotherms and in theirseasonal patterns of activity and dispersion.

Shelter Fidelity

Gila Monsters re-used nearly half of theirshelters, but we suspect that is an underesti-mate of their actual shelter-site fidelity be-cause the longer we tracked lizards, the morewe observed them reusing previous sheltersites. In terms of time spent within shelters,Gila Monsters showed even greater fidelity tofamiliar sites, spending nearly eight times aslong in reused shelters as in shelters observedto be used only once. Reused shelters tendedto have rockier roofs than other shelters,a result probably attributable to their greaterdurability (shelters with soil roofs were vul-nerable to collapse). Aside from rockiness,reused shelters were no different from sheltersnot observed to be reused. This result suggeststhat familiarity with a particular shelter maybe an important cue in shelter-site selection.Sleepy Lizards (Tiliqua rugosa) use specificvisual cues to recognize shelters and preferfamiliar cues over novel ones; this may playa role in the fidelity they show to home rangeareas (Zuri and Bull, 2000). Cues other thanstructure or microenvironment may also beused; for example, shelter selection by DesertTortoises is influenced by conspecific chemicalcues (Bulova, 1997). Future work on shelter-site cuing in Gila Monsters would add aninteresting dataset to be compared with theinformation on other organisms.

TABLE 4.Physical characteristics of Gila Monster shelters(rockiness, shelter depth) that possessed significant pre-dictive power on shelter microenvironment (temperatureand vapor pressure deficit) by season. Fall and wetsummer shelters were combined in this analysis. Partial

regression coefficients are shown in parentheses.

Season TemperatureVapor pressure

deficit

Winter Roof rockiness(b 5 52.41)

NS

(F1,45 5 7.28,P 5 0.010,R2 5 0.139)

Spring NS NS

Dry summer Roof rockiness(b 5 96.45)

Roof rockiness(b 5 116.26)

Shelter depth(b 5 64.38)

(F1,34 5 7.37,P 5 0.010,

(F2,38 5 5.58,P 5 0.008,R2 5 0.201)

R2 5 0.151)

Fall/wetsummer

Roof rockiness(b 5 75.38)

NS

(F1,44 5 6.14,P 5 0.017,R2 5 0.101)

TABLE 5.Shelter fidelity and home ranges shown by 7Gila Monsters outfitted with radiotransmitters betweenApril and July 1992, and monitored at least weekly fromNovember 1992 through June 1994 (Lizard no. 4s radio

signal died in April of 1993).

Lizardno. Sex

Totalshelters

No.reused

Percentreused

Home rangesize (ha)

1 m 44 14 31.8 88.22 f 42 22 52.4 40.44 m 17 4 23.5 60.05 m 36 16 44.4 41.08 m 67 29 43.3 65.89 f 28 15 53.6 6.2

10 f 37 19 51.4 104.8

Mean 42.9 58.1


The patterns of shelter fidelity shown byGila Monsters also suggest that differentshelters serve different purposes, and thesepurposes varied seasonally. Many of theshelters reused during April through Junewere used by more than one individual, oftentwo lizards at the same time. Males may fightduring spring for access to particular shelters(Beck, 1990), and mating may take place inthese shelters during April and June (Beck,1990; Goldberg and Lowe, 1997; this study).Desert tortoises also show seasonal use ofsocial shelters, where courtship and relatedreproductive behaviors may occur (Bulova,1994). Therefore, these shelter sites apparent-ly played a role in the social structure of thepopulation.

In contrast to social shelters, other shelter-sites used by Gila Monsters during the drysummer appeared more important as refugiafrom harsh conditions on the surface. Theywere used by only one lizard, were deep andsoil-like, and some were inhabited for periodsup to 40 days. With one exception (a male-female pair), winter shelters were inhabited byonly one lizard (interestingly, some individualsused overwintering shelters within a fewmeters of each other). Winter shelters hadsouth-facing entrances, and were predictablyrockier and deeper than shelters used in otherseasons. Gila Monsters showed the greatestfidelity (73%) to these winter retreats, a resultsuggestive of the importance of reliable wintershelters that served as refugia from cold and

predators during >30% of the year (Beck,1990; this study). Some lizards in our studyreturned to the same overwintering shelter forsix consecutive winters.

Fidelity to shelter sites is not unique to GilaMonsters. Many other terrestrial vertebrateectotherms, including snakes (Ciofi and Che-lazzi, 1994; Reinert and Zappalorti, 1988;Webb and Shine 1997), lizards (Norris,1953), tortoises (Bulova, 1994; Burge, 1977;Martin, 1995), and amphibians (Cohen andAlford, 1996) may show strong fidelity tospecific retreats. Desert rodents are known toconstruct elaborate burrow systems that pro-vide protection from predators and harshsurface conditions, and to which they showstrong fidelity (Burns et al., 1989; Goyal andGhosh, 1993; Murie and Michener, 1984;

FIG. 6.Approximately 43% of refugia were used morethan once by Gila Monsters, a proportion that is clearlya function of duration of monitoring and sample size. Thisresult, therefore, may be an underestimate of the actualproportion of shelters that are reused, and suggests thatsuitable refugia may be a limiting resource to GilaMonsters.

FIG. 7.Plot of residence times in shelters. GilaMonsters spent nearly eight times as much time in reusedshelters as in shelters used only once.

FIG. 8.Shelter fidelity was significantly influenced byseason. Gila Monsters showed the greatest fidelity toshelters used during extreme periods, such as winter.


Schmidt-Nielsen, 1964). Desert macroinverte-brates, including burrowing spiders (Hum-phreys, 1978; Marshall, 1997), scorpions (Poliset al., 1986), and beetles (Rasa, 1995) may alsoshow fidelity to retreat-sites. Despite theirdifferences, all these species share their de-pendence on suitable refugia, and theirpatterns of distribution and abundance maybe related to the availability of suitablesubstrate in which burrows and shelter-sitescan occur (Feldhamer, 1979; Kinlaw, 1999).

Conclusions

Our results illustrate how the ecology ofa sedentary desert ectotherm is closely tied toits use of subsurface shelter sites. Shelters, andthe habitats where they occur, were not chosenat random. Gila Monsters selected sheltersbased on specific structural and microenviron-

mental characteristics, altered their choice ofshelters as conditions changed seasonally, andreturned to preferred shelter sites year afteryear. Our results are also important becausethey suggest shelters may be a limiting re-source for desert ectotherms, one that is oftenoverlooked by ecologists interested in otherfactors affecting habitat use and patterns ofactivity, dispersion, and distribution. Selectionof appropriate shelter sites is perhaps the mostimportant means by which most ectothermspersist in extreme and variable desert environ-ments. It is surprising, therefore, to find thatthis phenomenon has not been more exten-sively studied. Although physiological conse-quences of habitat selection for ectothermshave been recognized for some time (e.g.,Huey, 1991), further research on the role ofsubsurface refugia in the ecology of desert

FIG. 9.Seasonal variation in shelter use by Gila Monsters. (A) Individual shelters served as refugia from harsh surfaceconditions, such as winter cold and dry-summer heat. (B) Social shelters were used by more than one individual,sometimes concurrently. Gila Monsters inhabited shelters concurrently primarily during the spring when male-combatand mating occur.


ectotherms might better explain their patternsof activity, behavior, distribution, and abun-dance.

Understanding the ecological importanceand role of shelters in the natural history oforganisms also carries implications for conser-vation. Conservation efforts aimed at identify-ing and protecting suitable habitats forsensitive species, for example, may need tomake specific considerations of shelter sitesas limiting resources. Knowledge of sheltercharacteristics, the abundance of shelters, andseasonal changes in shelter use may providebetter tools for managing desert speciesthreatened with habitat loss and human-re-lated disturbance (Bulova, 1994; Zimmermanet al., 1994).

Acknowledgments.We owe a special debt to CharlesW. Painter of the New Mexico Department of Game andFish (NMDGF) for originally suggesting a study ofHeloderma suspectum at Red Rock and for his unwaveringsupport during all stages of this research. The Share WithWildlife Program of the NMDGF provided funding forthis study. Central Washington University Department ofBiological Sciences and the CWU faculty developmentfund, provided additional support to DB. All procedureswere conducted in compliance with established animalcare guidelines at Central Washington University, anda NMDGF scientific collecting permit (number 1988). Wealso thank NMDGF for allowing us access to the Red RockWildlife Area and for assistance and support received fromNMDGF staff A. Ford, D. Graves, J. McCarter, P. Pirtle,and C. Tyrpak. For field assistance we also thank K.Ernest, S. Fields, J. Worley, J. Roback, S. Jennings,A. Wilds, and especially the heroic field efforts of C. M.Gienger. We thank the anonymous reviewers, and thefollowing colleagues, for providing helpful commentson earlier versions of this manuscript: R. Anderson,K. Christian, D. Darda, K. Ernest, L. Fitzgerald, B.Martin, B. Parmenter, G. Schuett, B. Sullivan, and L. Sun.

LITERATURE CITED

AVERY, R. A. 1976. Thermoregulation, metabolism andsocial behaviour in Lacertidae. Linnean Society Sym-posium Series 3:245259.

BECK, D. D. 1990. Ecology and behavior of the GilaMonster in southwestern Utah. Journal of Herpetology24:5468.

BECK, D. D., AND C. H. LOWE. 1991. Ecology of thebeaded lizard, Heloderma horridum, in a tropical dryforest in Jalisco, Mexico. Journal of Herpetology 25:395406.

. 1994. Resting metabolism of helodermatidlizards: allometric and ecological relationships. Journalof Comparative Physiology B 164:124129.

BECK, D. D., M. R. DOHM, T. GARLAND, JR., A. RAMIREZ-BAUTISTA, AND C. H. LOWE. 1995. Locomotor perfor-mance and activity energetics of helodermatid lizards.Copeia 1995:577585.

BLAZQUEZ, M. C., AND R. RODRIGUEZ-ESTRELLA. 2001.Winter refuge characteristics of spiny-tailed iguanas,Ctenosaura hemilopha, in Baja California sur, Mexico.Journal of Arid Environments 49:593599.

BOGERT, C. M., AND R. MARTIN DEL CAMPO. 1956. The GilaMonster and its allies. The relationships, habits, andbehavior of the lizards of the family Helodermatidae.Bulletin of the American Museum of Natural History109:1238.

BROWN, D. E. 1982. Semidesert grassland. In D. E. Brown(Ed.), Biotic Communities of the American South-westUnited States and Mexico. Desert Plants 4:123131. Boyce Thompson Southwestern Arboretum,Superior, Arizona, U.S.A.

BUFFENSTEIN, R. 1984. The importance of microhabitat inthermoregulation and thermal conductance in 2 Namibrodents: a crevice dweller, Aethomys namaquensis, anda burrow dweller, Gerbillurus paeba. Journal ofThermal Biology 9:235241.

BULOVA, S. J. 1994. Patterns of burrow use by deserttortoises: gender differences and seasonal trends.Herpetological Monographs 8:133143.

. 1997. Conspecific chemical cues influence burrowchoice by desert tortoises (Gopherus agassizii). Copeia1997:802810.

. 2002. How temperature, humidity, and burrowselection affect evaporative water loss in desert tortoises.Journal of Thermal Biology 27:175189.

BURGE, B. L. 1977. Daily and seasonal behavior, and areasutilized by the desert tortoise Gopherus agassizii insouthern Nevada. Proceedings of the Desert TortoiseCouncil Symposium 1977:5994.

BURGER, J., AND M. GOCHFELD. 1991. Burrow site selectionby black iguana (Ctenosaura similis) at palo verde, CostaRica. Journal of Herpetology 25:430435.

BURNS, J. A., D. L. FLATH, AND T. W. CLARK. 1989. On thestructure and function of white-tailed prairie dogburrows. Great Basin Naturalist 49:517524.

CHRISTIAN, K. A., C. R. TRACY, AND W. P. PORTER. 1983.Seasonal shifts in body temperature and microhabitatsby Galapagos land iguanas (Conolophus pallidus).Ecology 64:463468.

. 1984. Physiological and ecological consequencesof sleeping-site selection by the Galapagos land iguana(Conolophus pallidus). Ecology 65:752758.

CHRISTIAN, K. A., L. K. CORBETT, B. GREEN, AND B. W.WEAVERS. 1995. Seasonal activity and energetics of twospecies of varanid lizards in tropical Australia. Oecologia103:349357.

CHRISTIAN, K. A., A. D. GRIFFITHS, AND G. S. BEDFORD.1996a. Physiological ecology of frillneck lizards ina seasonal tropical environment. Oecologia 106:4956.

CHRISTIAN, K., B. GREEN, G. BEDFORD, AND K. NEWGRAIN.1996b. Seasonal metabolism of a small, arboreal monitorlizard, Varanus scalaris, in tropical Australia. Journal ofthe Zoological Society of London 240:383396.

CHRISTIAN, K. A., G. S. BEDFORD, AND T. J. SCHULTZ.1999a. Energetic consequences of metabolic depressionin tropical and temperate-zone lizards. AustralianJournal of Zoology 47:133141.

CHRISTIAN, K., G. BEDFORD, B. GREEN, A. GRIFFITHS,K. NEWGRAIN, AND T. SCHULTZ. 1999b. Physiologicalecology of a tropical dragon, Lophognathus temporalis.Australian Journal of Ecology 24:171181.


CIOFI, C., AND G. CHELAZZI. 1994. Analysis of homingpattern in the colubrid snake Coluber viridiflavus.Journal of Herpetology 28:477484.

COHEN, M. P., AND R. A. ALFORD. 1996. Factors affectingdiurnal shelter use by the cane toad, Bufo marinus.Herpetologica 52:172181.

CONOVER, W. J. 1980. Practical Nonparametric Statistics,2nd ed. John Wiley & Sons, Inc., New York, New York,U.S.A.

CONOVER, W. J., AND R. L. IMAN. 1981. Rank trans-formations as a bridge between parametric and non-parametric statistics. American Statistician 35:124129.

COWLES, R. B., AND C. M. BOGERT. 1944. A preliminarystudy of the thermal requirements of desert reptiles.Bulletin of the American Museum of Natural History82:265296.

DALTRY, J. C., T. ROSS, R. S. THORPE, AND W. WUSTER.1998. Evidence that humidity influences snake activitypatterns: a field study of the Malayan pit viperCalloselasma rhodostoma. Ecography 21:2534.

DIAZ, J. A. 1997. Ecological correlates of the thermalquality of an ectotherms habitat: a comparison betweentwo temperate lizard populations. Functional Ecology11:7989.

DORCAS, M. E., AND C. R. PETERSON. 1998. Daily bodytemperature variation in free-ranging rubber boas.Herpetologica 54:88103.

DOWNES, S. 2001. Trading heat and food for safety: costs ofpredator avoidance in a lizard. Ecology 82:28702881.

DOWNES, S., AND R. SHINE. 1998. Heat, safety or solitude?Using habitat selection experiments to identify a lizardspriorities. Animal Behavior 55:13871396.

FELDHAMER, G. A. 1979. Vegetative and edaphic factorsaffecting abundance and distribution of small mammalsin southeast Oregon. Great Basin Naturalist 39(3):207218.

GOLDBERG, S. R., AND C. H. LOWE. 1997. Reproductivecycle of the Gila Monster, Heloderma suspectum insouthern Arizona. Journal of Herpetology 31:161166.

GOYOL, S. P., AND P. K. GHOSH. 1993. Burrow structure of2 gerbil species of Thar desert, India. Acta Theriologica38:453456.

HAWKINS, L. K., AND P. F. NICOLETTO. 1992. Kangaroo ratburrows structure the spatial organization of ground-dwelling animals in a semiarid grassland. Journal of AridEnvironments 23:199208.

HEATWOLE, H. 1977. Habitat selection in reptiles. Pp. 137155. In C. Gans and D. W. Tinkle (Eds.), Biology of theReptilia, Vol. 7. Academic Press, London.

HEFFNER, R. A., M. J. BUTLER IV, AND C. K. REILLY. 1996.Pseudoreplication revisited. Ecology 77:22582562.

HENSCHEL, J. R. 1998. Dune spiders of the Negev Desertwith notes on Cerbalus psammodes (heteropodidae).Israel Journal of Zoology 44:243251.

HEWITT, C. H. 1959. Geology and mineral deposits of thenorthern big burro mountains-Redrock area, Grantcounty, New Mexico. Bulletin 60, New Mexico StateBureau of Mines and Mineral Resources, Institute ofMining and Technology Campus Station, Socorro, NewMexico, U.S.A.

HUEY, R. B. 1982. Temperature, physiology and theecology of reptiles. Pp. 2591. In C. Gans and F. H.Pough (Eds.), Biology of the Reptilia. Vol. 12. AcademicPress, New York, New York, U.S.A.

. 1991. Physiological consequences of habitatselection. American Naturalist 137:S91S115.

HUEY, R. B., C. R. PETERSON, S. J. ARNOLD, AND W. P.PORTER. 1989. Hot rocks and not-so-hot rocks: retreat-site selection by garter snakes and its thermal con-sequences. Ecology 70:931944.

HUMPHREYS, W. F. 1978. The thermal biology of (Geo-lycosa godeffroyi) and other burrow inhabiting Lycosi-dae (Araneae) in Australia. Oecologia 31:319347.

HURLBERT, S. H. 1984. Pseudoreplication and the design ofecological field experiments. Ecological Monographs54:187211.

JENNRICH, R. J., AND F. B. TURNER. 1969. Measurement ofnon-circular home range. Journal of Theoretical Biology22:227237.

KAY, F. R., AND W. G. WHITFORD. 1978. The burrowenvironment of the banner-tailed kangaroo rat, Dipod-omys spectabilis in south-central New Mexico. Ameri-can Midland Naturalist 99:270279.

KEARNEY, M. 2002. Hot rocks and much-too-hot rocks:seasonal patterns of retreat-stie selection by a nocturnalectotherm. Journal of Thermal Biology 27:205218.

KING, D. 1980. The thermal biology of free-living sandgoannas (Varanus gouldii) in southern Australia. Copeia1980:755767.

KINLAW, A. 1999. A review of burrowing by semi-fossorialvertebrates in arid environments. Journal of AridEnvironments 41:127145.

KRASNOV, B., AND G. SHENBROT. 1996. Spatial structure ofcommunity of darkling beetles (Coleoptera: Tenebrio-nidae) in the Negev Highlands, Israel. Ecography19:139152.

LOWE, C. H., C. R. SCHWALBE, AND T. B. JOHNSON. 1986.The venomous reptiles of Arizona. Arizona Game andFish Dept., Phoenix.

MARSHALL, S. D. 1997. The ecological determinants ofspace use by a burrowing wolf spider in a xericshrubland ecosystem. Journal of Arid Environments37:379393.

MARTIN, B. E. 1995. Ecology of the desert tortoise(Gopherus agassizii) in a desert-grassland communityin southern Arizona. M.S. Thesis, University of Arizona,Tucson.

MAUTZ, W. J. 1980. Factors influencing evaporative waterloss in lizards. Comparative Biochemistry and Physiol-ogy 67:429437.

. 1982. Patterns of evaporative water loss. Pp. 443481. In C. Gans and F. H. Pough (Eds.), Biology of theReptilia, Vol. 12. Academic Press, New York.

MURIE, J. O., AND G. B. MICHENER. 1984. The Biology ofGround-dwelling Squirrels. University of NebraskaPress, Lincoln. 459 pp.

NAGY, K. A., AND P. A. MEDICA. 1986. Physiological ecologyof desert tortoises in southern Nevada. Herpetologica42:7392.

NEVO, E., R. GUTTMAN, M. HABER, AND E. EREZ. 1979.Habitat selection in evolving mole rats. Oecologia43:125138.

NORRIS, K. S. 1953. The ecology of the desert iguanaDipsosaurus dorsalis. Ecology 34:265287.

PETERSON, C. R. 1987. Daily variation in the bodytemperatures of free-ranging garter snakes. Ecology68:160169.


PIANKA, E. R. 1966. Convexity, desert lizards and spacialheterogeneity. Ecology 47:10551059.

POLIS, G. A., C. MYERS, AND M. QUINLAN. 1986. Burrowingbiology and spatial distribution of desert scorpions.Journal of Arid Environments 10:137146.

PORTER, W. P., AND C. R. TRACY. 1983. Biophysicalanalyses of energetics, time-space utilization, anddistributional limits. Pp. 5583. In R. B. Huey, E. R.Pianka, and T. W. Shoener (Eds.), Lizard Ecology:Studies of a Model Organism. Harvard University Press,Cambridge, Massachusetts, U.S.A.

PORTER, W. P., J. W. MITCHELL, W. A. BECKMAN, AND C. B.DEWITT. 1973. Behavioral implications of mechanisticecology: thermal and behavioral modeling of desertectotherms and their microenvironment. Oecologia13:154.

PORZER, L. M. 1981. Movement, behavior, and bodytemperature of the Gila Monster (Heloderma suspec-tum) in Queen Creek, Pinal County, Arizona. M.S.Thesis, Arizona State University, Tempe. 100 p.

POUGH, F. H., D. MORAFKA, AND P. HILLMAN. 1978. Theecology and burrowing behavior of the Chihuahuanfringe-footed lizard Uma exsul. Copeia 1978:8186.

POUGH, F. H. 1980. The advantage of ectothermy fortetrapods. American Naturalist 115:92112.

PREGILL, G. K., J. A. GAUTHIER, AND H. W. GREENE. 1986.The evolution of helodermatid squamates, with de-scription of a new taxon and an overview of Varanoidea.Trans. San Diego Soc. Nat. Hist. 21:167202.

PRICE, M. V. 1978. The role of microhabitat in structuringdesert rodent communities. Ecology 59:910921.

RANDALL, J. A. 1993. Behavioral adaptations of desertrodents (Heteromyidae). Animal Behaviour 45:263287.

RASA, O. A. E. 1995. Ecological factors influencing burrowlocation, group size and mortality in a nocturnal fossorialKalahari detritivore, Parastizopus armaticeps peringuey(Coleoptera, Tenebrionidae). Journal of Arid Environ-ments 29(3):353365.

REGAL, P. J. 1967. Voluntary hypothermia in reptiles.Science 155:15511553.

REINERT, H. K., AND R. T. ZAPPALORTI. 1988. Timberrattlesnakes of the Pine Barrens: their movementpatterns and habitat preferences. Copeia 1998:964978.

ROSE, B. 1982. Lizard home ranges: methodology andfunctions. Journal of Herpetology 16:253269.

RUBY, D. E., L. C. ZIMMERMAN, S. J. BULOVA, C. J. SALICE,M. P. OCONNOR, AND J. R. SPOTILA. 1994. Behavioral

responses and time allocation differences in desert tor-toises exposed to environmental stress in semi-naturalenclosures. Herpetological Monographs 8:2744.

SAS INSTITUTE. 1985. SAS/STAT Users Guide, version 5edition. SAS Institute, Cary, North Carolina, U.S.A.

SCHLESINGER, C., AND R. SHINE. 1994. Selection of diurnalretreat sites by the nocturnal gekkonid lizard Oeduralesueurii. Herpetologica 5:156163.

SCHMIDT-NIELSEN, K. 1964. Desert Animals: PhysiologicalProblems of Heat and Water. Clarendon Press, Oxford.277 p.

SOKAL, R., AND E. J. ROHLF. 1995. Biometry, 3rd ed. W.H.Freeman and Company, New York.

STEVENS, M. S. 1977. Further study of the Castolon localfauna (Arikareean: Early Miocene) Big Bend NationalPark, Texas. Pearce-Sellards Ser. Texas Mem. Mus.(28):169.

TRACY, R. L., AND G. E. WALSBERG. 2002. Kangaroo ratsrevisited: re-evaluating a classic case of desert survival.Oecologia 133:449457.

VAN HEERDEN, J., AND J. DAUTH. 1987. Aspects ofadaptation to an arid environment in free-living groundsquirrel, Xerus inauris. Journal of Arid Envrionments13:8389.

VITT, L. J., AND S. S. SARTORIUS. 1999. HOBOs, tidbits andlizard models: the utility of electronic devices in fieldstudies of ectotherm thermoregulation. FunctionalEcology 13:670674.

WEBB, J., AND R. SHINE. 1997. Out on a limb: conservationimplications of tree hollow use by a threatened snakespecies (Hoplocephalus bungaroides: Serpentes, Elapi-dae). Biological Conservation 81:2133.

WILSON, D. S., K. A. NAGY, C. R. TRACY, D. J. MORAFKA,AND R. A. YATES. 2001. Water balance in neonate andjuvenile desert tortoises, Gopherus agassizii. Herpeto-logical Monographs 15:158170.

ZIMMERMAN, L. C., M. P. OCONNER, S. J. BULOVA, J. R.SPOTILA, S. J. KEMP, AND C. J. SALICE. 1994. Thermalecology of desert tortoises in the eastern mojave desert:seasonal patterns of operative and body temperatures,and microhabitat utilization. Herpetological Mono-graphs 8:4559.

ZURI, I., AND C. M. BULL. 2000. The use of visual cues forspatial orientation in the sleepy lizard (Tiliqua rugosa).Canadian Journal of Zoology 78:515520.


beck and jennings 2003

Documents