Natural history of a northern population of twin-spotted rattlesnakes, Crotalus pricei

of 17

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
17 pages
0 downs
Natural history of a northern population of twin-spotted rattlesnakes, Crotalus pricei
   Journal of Herpetology, Vol. 36, No. 4, pp. 598–607, 2002Copyright 2002 Society for the Study of Amphibians and Reptiles Natural History of a Northern Population of Twin-SpottedRattlesnakes, Crotalus pricei D AVID B. P RIVAL , 1 M ATTHEW J. G OODE , D ON E. S WANN , C ECIL R. S CHWALBE , 2 AND M ICHAEL J. S CHROFF School of Renewable Natural Resources, 104 Biosciences East, University of Arizona, Tucson, Arizona 85721, USA A BSTRACT .—The twin-spotted rattlesnake ( Crotalus pricei ) is a small-bodied pitviper that has receivedlittle attention in the literature to date. The species reaches the northern limit of its range in southeasternArizona, where it inhabits higher elevations than any of the state’s 10 other rattlesnake species.During1997–2000, we captured, measured, and marked 127 C. pricei in Arizona’s Chiricahua Mountains between 2530and 2900 m elevation. We also used radiotelemetry to track the movements of 16 C. pricei in the study areaduring 1997–1998. Mean (  SE) snout–vent length of C. pricei was 387.8  8.3 mm (range  168–572), andmean mass was 53.5  3.3 g (range  3.6–188.5). Based on fecal analyses, lizards constituted the bulk ofprey (74%), but the diet of C. pricei also included mammals, birds, and a conspecific.Matingwasconcentratedin August and early September and parturition took place during late July and August. Mean number ofembryos was 3.94  0.34 (range  1–6) and female reproduction appeared biennial or less frequent. Basedon shed and growth rates, female C. pricei develop embryos at 4–5 years of age. Gravid females maintainedwarmer body temperatures relative to substrate temperature than nongravid females or males, presumablyby spending more time basking than other snakes. Radiotelemetry revealed that movement patterns variedfrom year to year, as males moved over six times farther per week during the 1998 monsoon season (Julyto September) than during the 1997 monsoon season. Additionally, use of talus slopes by males decreasedduring 1998. During dry years, such as 1998, males may be forced off talus into cooler microclimates whereresources are less concentrated than on talus. Although twin-spotted rattlesnakes ( Crotalus pricei ) were first described over 100 years ago(Van Denburgh, 1895), they have since gonelargely unstudied. Although limited informa-tion about diet (e.g., Klauber, 1972; Gumbartand Sullivan, 1990), reproduction (e.g., Maha-ney, 1997; Goldberg, 2000) and toxicity (Mintonand Weinstein, 1984) can be found in the liter-ature, little effort has been made to obtain eco-logical data beyond that provided by occasionalobservations.In the U.S., C. pricei is restricted to four dis- junct mountain ranges in southeastern Arizona(Chiricahua, Huachuca, Pinalen˜o, and SantaRita Mountains). Its range extends souththrough the Sierra Madre Occidental and SierraMadre Oriental in seven Mexican states (Arm-strong and Murphy, 1979; Campbell and Lamar,1989). The Arizona and Sierra Madre Occidentalpopulations in Mexico comprise one of the twosubspecies, Crotalus pricei pricei . These ambushpredators rarely exceed 600 mm in length andare usually associated with rocky slopes (espe-cially talus), ridges, and occasionally canyon bottoms between 1860 and 3050 m (Van Den- 1 Corresponding Author. E-mail: 2 Present address: U.S. Geological Survey, SonoranDesert Field Station, 125 Biosciences East, Universityof Arizona, Tucson, Arizona 85721, USA  burgh, 1922; Lowe, 1964). The biotic communi-ties associated with C. pricei are Madrean Mon-tane Conifer Forest and Madrean EvergreenWoodland (Brown, 1994; Pase and Brown, 1994). Crotalus pricei is interesting from an ecologicaland evolutionary perspective, because it is oneof the smallest rattlesnakes, inhabits higher el-evations than any of Arizona’s 10 other rattle-snake species, and may be morphologically andecologically similar to early rattlesnakes(Greene, 1997). Conservation of populations of C. pricei in Arizona has been an issue for over30 years (Kauffeld, 1969). Because of the limiteddistribution of the species within the U.S. and athreat posed by collection for the pet trade (Pri-val, 2000), C. pricei is protected by state law inArizona. Although illegal collection for the pettrade may currently be the greatest threat to C. pricei populations in the United States, other po-tential threats include mining, logging, grazing,recreational and other development, and climatechange. Global warming in particular may ul-timately pose a significant threat to Arizona’spopulations because the species is currentlyfound only at the highest elevations of themountain ranges it inhabits. Information aboutthe life history of these snakes will be essentialfor any future conservation efforts (Dodd, 1993).We studied populations of C. pricei in Arizona’sChiricahua Mountains to learn about morphol-ogy, diet, movement patterns, and reproduction.  599TWIN-SPOTTED RATTLESNAKE ECOLOGYM ATERIALS AND M ETHODS Study Sites .—The Chiricahua Mountains arelocated near the southeastern corner of Arizonain Cochise County and reach an elevation of2986 m. Typically, slightly more than half (52%)of annual precipitation falls during a summermonsoon season characterized by localized,high-intensity thunderstorms from July to Sep-tember (Adams and Comrie, 1997). Most addi-tional precipitation results from widespread,low-intensity winter storms (National WeatherService, unpubl. data). Rustler Park Ranger Sta-tion, located near the study sites at 2560 m, re-ceives an average of 760 mm of precipitationperyear (Bennett et al., 1996).We concentrated our efforts on four exposedtalus slides (Sites A, B, C, and D) within theMadrean Montane Conifer Forest vegetation as-sociation (Pase and Brown, 1994). One of thesites (Site A) is a well-known locality for thespecies. We selected other sites by investigatingunvegetated areas marked on topographicmaps(1:24,000) and visible in aerial photographs.Vegetation at one of the sites (Site B) was burnedseverely during a wildfire in 1994. Study sitesranged in elevation from 2530–2900 m, and talusarea at each site ranged from 0.7–3.3 ha. SitesA and B faced south, whereas Sites C and Dfaced west. Capture Methodology .—Searches involved tra-versing talus, primarily during daylight hours.We spent 340 person-hours searching for snakesduring 1997 and 517 person-hours during 1998,primarily during July to September. We alsospent 170 person-hours searching for snakesover five days in late July to early August 1999at Site A and 195 person-hours searching overfour days in late July 2000 at Site A.We marked 127 C. pricei during the study, in-cluding 83 snakes at Site A, 27 at Site B, 12 atSite C, and 5 at Site D. We recaptured snakes on72 occasions. We captured C. pricei by hand us-ing welding gloves or forceps, and markedthemuniquely with Passive Integrated Transponders(PIT tags; Destron-Fearing Corp., South St. Paul,MN). In addition, we painted up to three rattlesegments of each snake with a unique colorcode to facilitate visual identification of individ-uals and to enable us to determine the numberof sheds between captures. We measured tem-peratures at 1.5 m, 0.5 cm, and substrate levelat every capture and observation site.  Morphology and Growth .—We measuredsnout–vent length (SVL), tail length, mass, and othermorphological characteristics of all snakes cap-tured. After reviewing the data visually, we de-termined that all datasets met the requirementsfor parametric tests except for mass, which wasskewed right. Therefore, we used a natural logtransformation to normalize the mass data forall analyses (Zar, 1996; JMP IN 3.2.1, SAS Insti-tute, Inc., Cary, NC). However, we report non-transformed mean mass in our results. Forsnakes captured more than once, we selectedone capture event at random for most analyses but only used data from the initial capture forradiotelemetered snakes.We determined sex by cloacal probing (Lasz-lo, 1975). We were unable to ascertain the sex ofsome snakes based on their small size. The larg-est snake for which we were unable to confi-dently determine sex measured 301 mm SVL.Therefore, we excluded snakes measuring  301mm SVL from analyses of sex-baseddifferences.We calculated mean growth rates for recapturedsnakes, excluding snakes with fewer than 14days between recapture events. We assessed dif-ferences in growth rate as a function of sex, site,days between recapture events, and initial SVLusing multiple regression. Althoughsexandsiteare categorical variables, we were able to usemultiple regression for this analysis by convert-ing these variables to indicator variables (Ram-sey and Schaefer, 1997). We used a t -test to com-pare growth rates of snakes with and withoutradiotransmitters that had snout–vent lengthsofat least 430 mm, the size of the smallest snakethat carried a radiotransmitter. Using recapturedata from different years, we calculated themean number of sheds per year. Diet and Reproduction .—We palpated snakesfor food boli on 134 occasions during1998–2000.We obtained fecal samples when snakes defe-cated while held individually for processing.Weidentified prey by using a dissectingmicroscopeto compare samples with specimens of animalsknown to occur in the area, based on results oflive trapping, observations, or historicalrecords.We palpated 44 adult female snakes for em- bryos during 1998–2000. We used logistic re-gression to determine whether number of em- bryos was related to SVL. Although Klauber(1972) reported a gravid female measuring 301mm total length, the smallest gravid C. pricei captured during this study measured 364 mmSVL. Therefore, snakes smaller than 364 mmSVL were considered to be juveniles. We useddata obtained from snake palpation,radiotelem-etry (movement patterns), and observations ofmating behavior and paired males and femalesto describe the reproductive cycle of C. pricei inthe Chiricahuas. Two snakes involved in the ra-diotelemetry study were not identified by pal-pation as being gravid, possibly because theywere palpated before embryos were largeenough to be detectable, but we believe theywere gravid based on movement patterns andsubstantial weight loss. These two snakes werenot included in the gravid category for statistical  600 D. B. PRIVAL ET AL. F IG . 1. Size distribution of 127 Crotalus pricei (60male, 43 female, 24 undetermined sex) from the Chir-icahua Mountains, Arizona, 1997–2000. tests other than those related to movement pat-terns and body temperature.  Movement Patterns and Habitat Use .—We sur-gically implanted temperature-sensitive radi-otransmitters with whip antennas (Holohil Sys-tems, Ltd., Carp, ON, Canada) into 16 snakesduring 1997–1998. Most were 1.8-g transmitterswith an expected battery life of four months ( N   13), but some were 3.3-g transmitters de-signed to last six months ( N   3). Surgical im-plantation of radiotransmitters followed Reinertand Cundall (1982) with some modifications(see Prival, 2000).In 1997, we tracked seven males and two fe-males. In 1998, we tracked five males and fivefemales. One male was tracked for 13 months;most other snakes were tracked for less thanfour months each. A TR-4 receiver (Telonics,Inc., Mesa, AZ) and a two-element flexible Yagiantenna were used to locate snakes. We locatedsnakes once per week between July and mid-October and once per month during other timesof the year. We used the radiotransmitter signalto calculate body temperature each time a snakewas located. We also measured temperatures at1.5 m, 0.5 cm, and substrate level with a ther-mometer. Multiple regression was used toassessdifferences in body temperature as a function ofsex, site, month, year, time, and substrate tem-perature (T s ). We converted categorical variablesto indicator variables for this analysis (Ramseyand Schaefer, 1997).We measured distances and angles betweensnake locations with a tape measure and/orrangefinder and a compass and clinometer. Weused Pitter Plotter 1.2 (Concentrics Company,Santa Fe, NM), a cave-mapping program, to es-timate distances between points. Home-rangesize was estimated with Calhome 1.0 (J. Kie, Pa-cific Southwest Research Station, USFS), usingthe 100% minimum convex polygon model.The number of snakes with radiotransmittersvaried over time. The largest number of snakeswas tracked between July and September ofeach year. Analyses of home ranges and move-ment patterns, therefore, only involve observa-tions made during those months unless other-wise stated. All means are reported  1 SE.R ESULTS We marked 51 snakes in 1997, 58 in 1998,eight in 1999, and 10 in 2000. Most snakes (93%)were captured between July and September,during which 92% of the total search effort tookplace.  Morphology and Growth .—SVL for all C. pricei captured averaged 387.8  8.3 mm (range  168–572) (Fig. 1). Mean SVL of males and fe-males did not differ (male  423.2  8.5 mm,female  419.7  10.1, t  0.26, df  101, P  0.79). However, only males were longer than540mm ( N   3). Mass of all snakes captured aver-aged 53.5  3.3 g (range  3.6–188.5). Meanmass of males and females did not differ (male  67.2  4.5 g, female  58.0  5.3, t  0.44,df  101, P  0.66). However, only male snakesexceeded 119 g ( N   8). Overall, 58.3% ( N   60)of snakes  301 mm SVL were males, indicatingthat the population did not differ from a 1:1 sexratio (  2  2.8, df  1, P  0.09). Sixty-threepercent of all snakes captured were adults. Theratio of tail length to total length differed be-tween the sexes (male 95% CI  0.0856–0.0888,female 95% CI  0.0680–0.0708, t  15.9, df  101, P  0.0001). However, there was some over-lap (male range  0.0656–0.0995, female range  0.0590–0.0816). The length-mass relationshipfor C. pricei was highly significant (mass  e (0.2653  0.00882 SVL) ; t  44.2, df  125, P  0.0001, r 2  0.94).Of the variables tested (sex, site, days betweenrecaptures, and initial SVL), growth rate(  SVL/day) was related only to initial SVL (ini-tial SVL: F 3,60  11.6, P  0.0012, r 2 (whole-mod-el)  0.19). Juvenile snakes grew faster thanadult snakes in terms of  SVL/day ( t  2.92, df  67, P  0.0048) and  SVL/shed ( t  2.65, df  39, P  0.012; Table 1). The presence of ra-diotransmitters did not affect C. pricei growthrate ( t  0.62, df  35, P  0.54). On average, C. pricei shed 1.76  0.14 times per year ( N   28). If the estimated growth and shed rates areaccurate, C. pricei become large enough to de-velop embryos (364 mm SVL) at four or fiveyears of age. Diet .—Food boli were detected in 36% of C. pricei palpated, including 44% of males ( N   66palpated) and 25% of females ( N   56;  2  4.8,df  1, P  0.029). We found food boli in 29%of adults ( N   97) and 54% of juveniles ( N   37;  2  7.4, df  1, P  0.0066). During the  601TWIN-SPOTTED RATTLESNAKE ECOLOGY T ABLE 1. Crotalus pricei growth rates (mean  SE). Snakes  364 mm SVL were considered to be juveniles. All snakes N  Juveniles N  Adults N   SVL (mm)/day  SVL (mm)/shed  mass (g)/day  mass (g)/shed0.063  0.0357.26  3.83  0.019  0.0272.06  1.92694168410.253  0.07325.1  7.620.085  0.0585.72  4.091491490.015  0.0372.24  4.04  0.045  0.0301.03  2.1755325432F IG . 2. Prey distribution in Crotalus pricei fecalsamples ( N   31).T ABLE 2. Reproductive condition of recaptured Crotalus pricei in the Chiricahua Mountains, Arizona,1998–2000. G  Gravid. NG  Not gravid. ?  Notcaptured. Snake ID 1998 1999 2000 13750D83683CE42GGGGNG?NGNGNGGG???NG monsoon months, adult males ate most often in July (55% with food boli; N   20 palpated) andSeptember (50%; N   10), and least often in Au-gust (14%; N   14;  2  6.1, df  1, P  0.048).Thirty-three fecal samples were collected andanalyzed, 31 of which contained identifiableprey items (Fig. 2). We found more than onetype of prey in four samples. We found lizardscales in 23 samples (74%). All of the scaleswere Sceloporus , and at least eight samples con-tained scales of Sceloporus jarrovii . Lizardspeciescould not be determined with certainty for theother 15 samples.One fecal sample contained snake scales anda rattle. The rattle size and shape and the col-oration of the scales matched more closely withspecimens of C. pricei than with specimens ofother montane rattlesnakes from southeasternArizona ( C. lepidus, C. molossus , and C. willardi ).The rattle consisted of one segment and a but-ton. A comparison of the ventral scales in thesample with C. pricei specimens suggests thatthe prey snake measured approximately 225mm SVL. According to the length-mass curve,a snake of this size would weigh about 12.4 g.This sample was obtained on 28 September1998from a male snake measuring 466 mm SVL andweighing 70 g.We found hair in nine samples (29%). Bonesin two of these samples were identified as Per-omyscus . Hair from another sample was identi-fied as Neotoma , most likely a Mexican woodrat( Neotoma mexicana ). We could not identify othersamples to genus. We found feathers in twosamples (6.5%). The color gradation of the feath-ers and the rachis color in one of the samplessuggest the prey item was a canyon wren ( Cath-erpes mexicanus ).Seventy-five percent of the 24 samples ob-tained from snakes captured on talus contained Sceloporus , whereas 25% contained mammals.Off talus, Sceloporus accounted for 57% of theprey of the seven snakes producing samples,with mammals comprising the other 43%. Both bird samples and the C. pricei prey were ob-tained from snakes found on talus. Reproduction .— No embryos were detectedbe-fore 9 June or after 1 September. We found mostgravid snakes in July and August. During thosemonths, 50% of 36 captured adult females weregravid. Mean number of embryos detected pergravid snake was 3.94  0.34 ( N   18, range  1–6).Gravid snakes averaged 430  8.6 mm SVL(range  364–500), whereas adult nongravid fe-males captured during July and August aver-aged 442  12.0 mm (range  400–534), indi-cating that adult body size was unrelated to re-productive activity ( t  0.79, df  28, P  0.62).In addition, there was no relationship between body size and number of embryos (  2  12.8,df  1, P  0.17).Between 1998 and 2000, five females werepalpated for embryos in July or August in morethan one year. None of these females was gravidin consecutive years (Table 2). Both snakes cap-tured in alternate years were in the same repro-ductive condition in both years (one gravid, onenongravid).Male and female snakes were observed inclose proximity to each other (  1 m) on 13 oc-casions between 11 August and 21 September1997, and between 9 August and 13 September  602 D. B. PRIVAL ET AL. T ABLE 3. Distance moved and home-range size for radiotelemetered Crotalus pricei, July–September (mean  SE). N   3 for 1998 female home range. 1997 ( N   4 males, 1 female)Distance moved(m/week)Home range(ha)1998 ( N   5 males, 4 females)Distance moved(m/week)Home range(ha) MalesFemales14.0   0.100.8387.8  11.019.7  7.12.29  0.880.20  0.011 1998. Courtship was observed on 21 August1997 between a 492-mm SVL male and a 466-mm SVL female. The behavior was character-ized by constantly intertwined tails and cyclesof about 2.5 min of chin-pressing and tongue-flicking by the male, followed by 4 sec of vig-orous tail movement and 20 sec of motionless-ness from 0851–0949 h. We could not determinewhether copulation occurred.A 441-mm SVL C. pricei gave birth to fourneonates sometime between 2100 h 16 Augustand 1430 h 17 August 1998 while being de-tained temporarily for processing (mean SVL ofneonates  168.8  2.32 mm, range  163–173;mean mass  4.4  0.04 g, range  4.3–4.5). Allfour neonates appeared to be healthy and alert.The female weighed 66.5 g prepartum and 42.5g postpartum.A 390-mm SVL female gave birth to three ne-onates during and following processing on 28–30 July 2000 (mean SVL of neonates  155.0  4.00 mm, range  151–163; mean mass  4.1  0.19 g, range  3.8–4.4). The first neonate wasstillborn; the other two were lethargic and cov-ered by a flaky layer of dead skin. The samefemale also produced an undeveloped ovumand was carrying at least two more embryos orova when released on 30 July. Prepartum massof this female was not obtained; she weighed 57g prior to release.Based on movement patterns and substantialdecreases in mass, we believe three radiotele-metered snakes gave birth during late August1998, although these neonates were never locat-ed. The four gravid snakes measured beforeandafter parturition lost 32.4  5.4 g (range  22.5–44.0 g), representing an average weight loss of39.4  3.4% (range  31.7–47.3%).Other neonate snakes (defined here as snakeswith a rattle consisting of only a prebutton or button) were captured on 15 September 1997(185 mm SVL, 3.6 g), 22 September 1997 (200mm, 7.1 g), 30 September 1997 (176 mm, 6.1 g),and 29 July 2000 (168 mm, 3.8 g).  Movement Patterns and Habitat Use .—Males ( N   9) moved an average of 55.0  14.4 m/weekand females ( N   5) moved 24.2  7.1 m/weekduring the monsoon months (July to September; t  1.5, df  12, P  0.16; Table 3). Variationwas high within each sex (male range  2.6–115.4 m/week, female range  4.7–42.5 m/week).On average, males moved 73.8  13.7 m/week farther in 1998 than 1997 during the mon-soon months ( t  5.4, df  7, P  0.001) (Table3). One male was monitored over most of bothmonsoon seasons. He moved almost four timesfarther per week in 1998 than in 1997.Movement patterns of gravid females tracked before and after parturition in 1998 ( N   3)were characterized by little movement (  3.2m/week) between mid-July and the time we be-lieve they gave birth at the end of August, fol-lowed by greatly increased movement in Sep-tember (mean: 73.1  30.4 m/week, range  38.5–133.6 m/week). The nongravid femalemoved 53.5 m/week between 24 July and 29August 1997, and 30.0 m/week between 29 Au-gust and 30 September.During the monsoon months, mean homerange size for males ( N   9) was 1.37  0.60 ha(range  0.0038–5.34) and for females ( N   4)was 0.36  0.16 ha (range  0.18–0.83; t  0.18,df  11, P  0.86). Home range size of malesdiffered in 1997 ( N   4) and 1998 ( N   5; t  3.5, df  7, P  0.010; Table 3). The nongravidfemale monitored in 1997 had a substantiallylarger home-range size during the monsoonthan the gravid females tracked in 1998.Home range of the male monitored duringmost of the monsoon season in both years wasover 12 times larger in 1998 than in 1997. Hishome range over the entire 13 months he wastracked was 14.7 ha. He did not revisit any ofhis 1997 locations in 1998, and his home rangesduring the two monsoon seasons did not over-lap.Radiotelemetered males differed in use of ta-lus during the monsoon months between years( t  2.4, df  7, P  0.044). In 1997, four maleswere found on talus 87.5  9.5% of the time(range  60–100%), whereas in 1998, five malesspent only 44.5  13.6% of their time on talus(range  0–76.9%). The male tracked during both monsoon seasons was always found on ta-lus slopes in 1997 but never in 1998.Radiotelemetered females varied in their useof talus while gravid in July and August. Three
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!