Estimating Abundance of the Endangered Houston Toad on a

Surveys

Estimating Abundance of the Endangered Houston Toadon a Primary Recovery SiteAdam Duarte*, Donald J. Brown, Michael R.J. Forstner

Department of Biology, Texas State University–San Marcos, San Marcos, Texas 78666

Abstract

The Griffith League Ranch is one of the primary recovery sites for the endangered Houston toad Bufo (Anaxyrus)houstonensis. New recovery initiatives have recently been implemented to increase Houston toad abundance;however, no robust estimate of population size has been conducted in the last decade of study, nor from this recoverysite. To assist with inferences regarding efficacy of current and future management actions, we estimated adultHouston toad abundance on the Griffith League Ranch. Houston toads were sampled at breeding ponds during the2010 breeding season using a mark–recapture methodology. We analyzed the data using a modified Jolly–Seber openpopulation model in Program MARK. Models were built whereby the probability of capture remained constant, theapparent survival varied with time or was constant, and the probability of entry varied with time. Model averaging wasused to account for uncertainty and the derived adult male Houston toad abundance ranged from 201 to 307individuals. Using a previously determined Griffith League Ranch Houston toad functional sex ratio of 5:1, weestimated the abundance of the total adult Houston toad population on this primary recovery site to be from 241 to368 individuals. This study is the first to report a robust abundance estimate of a Houston toad population andprovides a foundation for further research to quantify the impact of current and future management actions.

Keywords: abundance; Bufo houstonensis; endangered species; POPAN; Program MARK

Received: July 15, 2011; Accepted: September 9, 2011; Published Online Early: September 2011; Published: December2011

Citation: Duarte A, Brown DJ, Forstner MRJ. 2011. Estimating abundance of the endangered Houston toad on a primaryrecovery site. Journal of Fish and Wildlife Management 2(2):207–215; e1944-687X. doi: 10.3996/072011-JFWM-041

Copyright: All material appearing in the Journal of Fish and Wildlife Management is in the public domain and may bereproduced or copied without permission unless specifically noted with the copyright symbol �. Citation of thesource, as given above, is requested.

The findings and conclusions in this article are those of the author(s) and do not necessarily represent the views of theU.S. Fish and Wildlife Service.

* Corresponding author: [email protected]

Introduction

The Houston toad Bufo (Anaxyrus) houstonensis (Fig-ure 1) was the first amphibian listed under the federalEndangered Species Conservation Act (Gottschalk 1970),the precursor to the Endangered Species Act as amended(ESA 1973). Over the past 50 y this east-central Texasendemic has been extirpated from 3 of the 12 counties inwhich it was known to occupy, with substantial popula-tion declines in the remaining counties (Forstner et al.2007; Supplemental Material, Reference S1; http://dx.doi.org/10.3996/072011-JFWM-041.S1; Gaston et al. 2010).Since the 1970s, the Lost Pines ecoregion in BastropCounty has housed the largest populations of Houstontoads, and has remained the only county without a highshort-term risk of extirpation (Brown 1971; U.S. Fish andWildlife Service [USFWS] 1984).

A number of factors have likely contributed to therange-wide population declines of the Houston toad.Brown (1971) suggested habitat loss and degradationwere leading threats to remaining populations. Giventhat the species is a habitat specialist, showing strongpreference for deep sandy soils and forest cover (USFWS1984; Buzo 2008), development and land-use changeshave undoubtedly been major contributors to localdeclines (Brown 1975; Brown and Mesrobian 2005).Additional factors potentially contributing to populationdeclines across the species’ range include prolongeddroughts (Forstner et al. 2007; Supplemental Material,Reference S1; http://dx.doi.org/10.3996/072011-JFWM-041.S1), interspecific competition and increased oppor-tunity for hybridization caused by habitat alteration(Brown 1971; Hillis et al. 1984), predation by invasivespecies (Freed and Neitman 1988; McHenry et al. 2010),

Journal of Fish and Wildlife Management | www.fwspubs.org December 2011 | Volume 2 | Issue 2 | 207

historically inadequate management toward the conser-vation and recovery of the species (Peterson et al. 2004;Brown and Mesrobian 2005), and anthropogenic changesto habitat likely reducing overall chorus sizes (i.e.,number of individual males calling), which, in turn,decreases reproductive success (Gaston et al. 2010).

Houston toad breeding activity occurs between lateJanuary and early May (Hillis et al. 1984; Swannack 2007).This species is an explosive breeder (Wells 1977), wherelarge aggregations of calling males occur at irregularintervals throughout the breeding season. Adult toadsare rarely found above ground between breedingintervals (Price 2003; Swannack and Forstner 2007; Brownet al. in press), and breeding activity has been correlatedwith minimum daily temperature, precipitation, andmoon phase (Hillis et al. 1984; Dixon et al. 1990; Price2003; Swannack 2007). Houston toad reproductivebehavior is sex-dependent, such that individual breedingmales return to the same pond within a breeding season(Jacobson 1989), whereas individual breeding femalesrarely visit the same pond more than once within abreeding season (Price 2003).

Auditory call surveys are the primary method used tomonitor Houston toads across their range (Jackson et al.2006), and models have been built to estimate theprobability of extinction for the species under different

scenarios to assist with recovery efforts (Hatfield et al.2004; Swannack et al. 2009). However, abundanceestimates are needed to assess the current status ofpopulations and measure responses to managementactions designed to increase Houston toad abundance,such as prescribed fire and head-starting (i.e., releasingcaptive-reared Houston toads into the wild [Forstneret al. 2007; Supplemental Material, Reference S1; http://dx.doi.org/10.3996/072011-JFWM-041.S1; Brown et al.2011]). Previously documented abundance estimateshave been restricted to ‘‘very good guesses’’ (Forstneret al. 2007; Supplemental Material, Reference S1; http://dx.doi.org/10.3996/072011-JFWM-041.S1), because de-tection problems associated with the secretive behaviorand rarity of the species have made abundanceestimates difficult to derive.

In 2001 we began a long-term Houston toad monitor-ing project on a primary recovery site for the species.Unfortunately, reliable abundance estimates were notfeasible prior to 2010 because of low within-yearrecapture success coupled with low annual adult survi-vorship (Swannack et al. 2009). However, several years ofactive stewardship along with unusually high precipita-tion rates during spring 2010 resulted in extensiveHouston toad calling activity, allowing us to obtainsufficient recaptures to conduct reliable mark–recapture

Figure 1. An adult male Houston toad Bufo (Anaxyrus) houstonensis calling to attract females on the Griffith League Ranch, BastropCounty, Texas in 2010. Image used with permission from photographer Bei DeVolld.

Houston Toad Abundance A. Duarte et al.


analyses. Herein, our objective within the confines of thisstudy is to present the first robust local abundanceestimate for the endangered Houston toad on a primaryrecovery site.

Methods

We conducted this study on the Griffith League Ranch(GLR), a 1,948-ha ranch in Bastrop County, Texas, ownedby the Boy Scouts of America (Figure 2). This large tractof land is located within designated critical habitat forthe Houston toad (USFWS 1984) and is consideredessential for long-term persistence of the species(Hatfield et al. 2004). The GLR is a primarily forestedranch, with an overstory dominated by loblolly pinePinus taeda, post oak Quercus stellata, blackjack oak Q.marilandica, and eastern red cedar Juniperus virginiana,and an understory dominated by yaupon holly Ilexvomitoria, American beautyberry Callicarpa americana,and farkleberry Vaccinium arboreum. The GLR contains 17ponds, with hydroperiods ranging from highly ephem-eral (n = 2) to permanent (n = 3). During 2010 weencountered Houston toads at 12 of these ponds. All 12ponds held water throughout the breeding season and,thus, remained potentially active breeding ponds duringeach sampling event.

We conducted call surveys at each of the ponds usingthe survey protocol detailed in Jackson et al. (2006). Weprimarily surveyed on nights when Houston toads weremost likely to be active based on perceived weatherpreferences derived from past studies. When Houstontoad calling was detected we searched the ponds andsurrounding areas, captured the toads present, recordedstandard measurements (i.e., snout–urostyle length,head width, and weight), collected DNA from a singletoe clip, and individually marked toads using PassiveIntegrated Transponder (PIT) tags (Camper and Dixon1988; Dodd 2009). Previous research determined thathandling and PIT-tagging Houston toads during callingnights did not adversely impact behavior (i.e., toadscontinued to call after handling), or subsequent returnsto breeding sites (Dixon et al. 1990).

In addition to call surveys, we monitored Houston toadactivity using 26 drift fences (Heyer 1994). Seven of thesedrift fences were located adjacent to (i.e., within 10 m),and parallel with, a pond used by Houston toads duringthe 2010 breeding season. These drift fences consisted ofone 15-m linear arm, a 19-L pitfall trap at each end, andtwo double-throated funnel traps placed near the centeron each side of the flashing. Because of a low number ofHouston toads captured in traps, we did not include themajority of these monitoring data in this study. However,when Houston toads were captured at the seven lineararrays mentioned above on call-survey nights, weconsidered them pond captures for that night, becausewe likely intercepted migration to or from a breedingpond.

We restricted our analysis to the adult male segmentof the population. The return of females to ponds withina breeding season is dependent upon nonsuccessfulreproduction on previous nights, which is rare (Jacobson

1989). Conversely, successful mating does not appear todeter males from returning to breeding ponds (Hilliset al. 1984; Jacobson 1989). It is also important to notethat our sampling design defines the population in thisstudy as the number of adult males attempting toreproduce on the GLR. Thus, for this purpose, ‘‘adult’’ isdefined as a reproductively active male, and if a largenumber of adult males did not participate in chorusing,then our analysis will likely underestimate the number ofadult males. A body condition index was calculated foreach male toad by regressing body lengths (snout-to-vent lengths) on weights and using residuals for the bodycondition index (i.e., positive and negative residuals indi-cate above- and below-average body condition, respec-tively; Schulte-Hostedde et al. 2005). We removed anobvious outlier, but did not transform the data because theuntransformed data fit a linear relationship well (r2 = 0.8).

We transformed adult male Houston toad individualcaptures (Supplemental Material, Table S1; http://dx.doi.org/10.3996/072011-JFWM-041.S2) into encounter histo-ries, resulting in 15 sampling occasions, and imported thedata into Program MARK (White and Burnham 1999).Abundance was then estimated using the POPANsoftware incorporated in Program MARK, which is aCormack–Jolly Seber model design (Schwarz and Arnason1996). This model assumes an open population, allowingfor immigration, emigration, births, and deaths betweensampling occasions. Treating the population as closed wasunreasonable because predation on adult male Houstontoads occurs during the breeding season (Dixon et al.1990; McHenry et al. 2010). Also, toads do not enter thepopulation at the same time; therefore all toads were notavailable for sampling across all occasions (Hillis et al.1984; Jacobson 1989; Swannack 2007).

Program POPAN incorporates four parameter esti-mates, including probability of capture (p), probability ofentry (PENT), apparent survival (Q), and super-populationsize (N). Probability of entry is the probability of anindividual entering the population on a given occasionand Q is the probability of an individual remainingavailable for recapture (survived and did not emigrate).We started with a global model in which p, PENT, and Qwere all time-dependent (p(t) Q(t) PENT(t) N). UsingProgram RELEASE we tested the global model’s good-ness-of-fit. Reduced models were then fit where variationin the parameter estimates was modeled as a function oftime (t) or as a constant (.). Probability of capture shouldbe constant because we based our sampling events tooptimize toad capture, used multiple trained observerseach night, and searched the breeding ponds andsurrounding area until no new individuals were capturedduring each sampling event. Therefore, p was modeledas a constant. Apparent survival was modeled to varywith time or to be constant and PENT was only modeledto vary with time. All possible combinations between ourglobal model and applied constraints yielded a candidateset of three models.

When constructing the models, we assigned Q, p,PENT, and N the link function logit, Mlogit1, and log. Weused the logit link for Q and p because it constrains theestimates from zero to one, which is necessary given that



Figure 2. Locations of the 17 ponds and 26 drift fences on the Griffith League Ranch, Bastrop County, Texas, USA. To estimateadult male Houston toad Bufo (Anaxyrus) houstonensis abundance in spring 2010 we used data collected from the 12 ponds withdetected calling activity during the 2010 breeding season and the 7 drift fences that were located adjacent to, and parallel with,ponds used by Houston toads during the 2010 breeding season.



both parameters are probabilities. We assigned theMlogitl link to PENT because it constrains the sum ofall time steps from zero to one. We used the log link forN, which is required because the abundance estimateshould be .1.

We assessed model fit using Akaike’s InformationCriterion (AICc) corrected for small sample size (Burnhamand Anderson 1998). Model selection was based on AICc

weights (i.e., the best-fit model had the largest AICc

weight). To account for model uncertainty, we used themodel-averaging procedure in Program MARK to aver-age parameter estimates, their corresponding standarderrors, and derived estimates of N.

Results

We conducted 33 call surveys at each of the pondsbetween 13 January and 22 April, and monitored activityusing drift fences from 31 January to 1 May. Out of the 33sampling events, 15 nights resulted in $1 Houston toadcapture. Successful nights occurred between 13 Februaryand 22 April. Intervals between nights with malecaptures ranged from 1 to 13 d and captures variedacross the breeding season for both males and females(Figure 3). One-hundred three individual toads (90 males,13 females) were captured and marked at breedingponds. For the male and female segments of thepopulation we had 32 and 1 recaptures, respectively,that occurred over the entire breeding season. Individualmale body condition index scores ranged from 24.9 to4.1 across all ponds.

The goodness-of-fit test run for the global model inProgram Release indicated underdispersion (c , 1) in ourdata. Underdispersion is not biologically meaningful;therefore, we conducted our model selection assuming

c = 1. Based on the AICc weights, both models in whichp was constant were supported (Table 1). However, themodel in which p was constant and Q and PENT variedacross time (p(.) Q(t) PENT(t) N) was the mostparsimonious (AICc = 355.04). The derived N estimateper survey occasion varied over time and ranged from 4(range using SEs adjusted for minimum number ofindividuals captured = 1–8) to 134 (101–167; Figure 4);however, these estimates are likely biased high becauseof random temporary emigration of individuals. Thegross N estimate was 254 (6SE, 53). The Q ranged from0.43 (60.29) to 0.98 (60.04; Figure 5), and p was 0.24(60.05).

Figure 3. Count of individual male and female Houston toad Bufo (Anaxyrus) houstonensis captures per day on the Griffith LeagueRanch, Bastrop County, Texas, in 2010. We began the x-axis in February because no successful captures occurred during January.

Table 1. Models analyzed and summaries of modelselection for abundance estimates of adult male Houstontoads Bufo (Anaxyrus) houstonensis sampled from the GriffithLeague Ranch, Bastrop County, Texas, in 2010.

Modela AICc DAICc AICc wt.Model

likelihood nPar

p(.) Q(t)PENT(t) N

355.04 0.00 0.74 1.00 30

p(.) Q(.)PENT(t) N

357.16 2.12 0.26 0.35 17

p(t) Q(t)PENT(t) N

385.38 30.35 0.00 0.00 43

a p is the capture probability, Q is the apparent survival probability,PENT is the probability of entry, and N is the super-population size.The characters in parentheses denote the constraint that was put oneach parameter: (.) represents a constant, (t) represents timevariation. AICc is the Akaike Information Criterion corrected forsmall sample size score, DAICc is the difference between the bestmodel score (smallest AICc) and each model, AICc wt. represents therelative likelihood of the model, and nPar is the number ofparameters in each model.



Discussion

Our gross abundance estimate of adult male Houstontoads (i.e., total number of adult male Houston toadsattempting to breed in 2010) was reasonable (201–307individuals). We could not estimate abundance of adultfemale Houston toads in Program MARK because of lowrecapture success (1 female recapture). However, Swan-nack and Forstner (2007) reported a functional male-biased sex ratio of 5:1 on the GLR. This ratio was estimatedfrom successful individual captures utilizing multiplesampling methods over 4 consecutive y. Using thisfunctional sex ratio, we estimated the adult femaleHouston toad population to be between 40 and 61individuals. Thus, the total adult Houston toad populationon the GLR in 2010 was estimated to be between 241 and368 individuals. Though all sampling took place on theGLR, intra- and interyear movement data from 2001 to thepresent, alongside high-resolution population geneticdata including migration inferences (McHenry 2010),support the supposition that our population estimaterepresents the majority of the western habitat patch inBastrop County centered on the GLR.

The breeding-season activity pattern observed duringthis study was similar to previous studies (Hillis et al.1984; Jacobson 1989; Price 2003; Swannack 2007). Maletoads were detected at the breeding ponds prior to thearrival of female toads, and Houston toad calling activitywas greatest in mid-March (Figure 3). Males visited thesame breeding pond throughout the season, althoughonly a relatively small proportion (26.7%) of the malesegment of the population was captured more than

once. The largest number of male captures at breedingponds in one night was 37 individuals. This particularnight had the largest congregation of male toads at asingle pond throughout the entire breeding season(Pond 12; 35 individuals). It is noteworthy that thecapture of 35 individuals was approximately equivalentto the total number of males detected during each of theprevious years, 2006–2009.

Despite the large number of individual male toadscaptured at Pond 12 in one night, subsequent surveynights at this pond (n = 21) detected dramatically fewertoads (n = 9), and only 9% of the individuals captured atPond 12 throughout 2010 were recaptured. This is incontrast to several other ponds, where we obtained farfewer total captures, but a much higher rate of return.For instance, we captured only five individuals at Pond15, but recaptured 60% of the individuals, with oneindividual recaptured three times. We propose threehypotheses concerning the observed high recapturevariability among ponds: 1) probability of detection (i.e.,capture) varied among ponds because of pond charac-teristics; 2) probability of detection (i.e., capture) variedamong ponds because of heterogeneous individual toadbehavior; and 3) within-season survivorship differedamong ponds.

We address evidence supporting each hypothesis insequential order. The estimated probability of capturewas 0.24 (60.05). It is important to note the estimate is afunction of two components: whether we detected anindividual that was present at a pond, and whether anindividual was actually at a pond (and, thus, was able tobe detected). Pond characteristics (e.g., size and vegeta-

Figure 4. Model-averaged-derived abundance estimates of adult male Houston toads Bufo (Anaxyrus) houstonensis per surveyoccasion on the Griffith League Ranch, Bastrop County, Texas, in 2010. Error bars represent 1 standard error corrected for theminimum number of individuals captured.



tion cover) could influence detection. Nevertheless,differences in pond characteristics probably had little orno influence on capture probability during this studybecause we conducted extensive pond searches whentoad calling was detected. Further, even at large ponds,Houston toads on the GLR typically aggregate in thesame localized areas of the ponds night after night (D. J.Brown and M. R. J. Forstner, personal observation). Thus,probability of capture was likely high once toads wereinitially detected during the study.

The influence of individual differences in callingbehavior on initial detection is more difficult todismiss. Although this was unlikely to be a problemon explosive nights, it could have influenced detectionon nights with minimal calling activity. Calling isenergetically costly, and energetic demands influencea variety of parameters associated with calling activity(reviewed in Wells 2007). Although body size has beenfound to be a poor predictor of nights engaged incalling activity (i.e., chorus tenure) in other toadspecies (Given 2002), body condition has been foundto be positively associated with chorus tenure in frogspecies (Murphy 1994; Judge and Brooks 2001). Thus,differences in body condition of male Houston toadsamong ponds, potentially a function of habitat quality,could account for among-pond variability in chorustenure. However, both extremes of the body conditionindex range calculated were from individuals at Pond12. Also the mean body condition index score for Pond12 was 0.05, indicating that mean body condition ofmale toads at Pond 12 was comparable to toads activeat other ponds on the GLR. Thus, at least from anenergetic demands perspective, it is unlikely that

individuals at Pond 12 were inherently less likely toengage in calling activity on multiple nights than toadscalling at other ponds.

The third potential explanation is within-seasonsurvivorship differed among ponds. We estimated ap-parent survival to range from 0.43 (60.29) to 0.98 (60.04;Figure 5), meaning the likelihood of an individualsurviving and returning to the study area (i.e., breedingponds) from one day to the next was between 43% and98%. The current estimate of annual survivorship foradult male Houston toads is approximately 15% (Swan-nack et al. 2009). Therefore, the lower apparent survi-val estimates were likely primarily a consequence ofemigration. However, it might also be caused bydifferences in predation among chorusing groups. Anincrease in chorus size might elevate exposure topredation. Unfortunately, we were unable to partitionemigration from mortality or derive pond-specificparameter estimates. Nevertheless, we believe among-pond differences in detection probability, migration, orsurvivorship, are all reasonable hypotheses for theobserved recapture heterogeneity, and we encouragefurther investigations concerning this topic.

Our study reports the most extensive, robust estimateof abundance for a Houston toad population to date.From summer 2007 to autumn 2009, Bastrop Countyexperienced significant and exceptional drought condi-tions, which likely negatively impacted Houston toadsurvivorship (Price 2003; Forstner et al. 2007; Supplemen-tal Material, Reference S1; http://dx.doi.org/10.3996/072011-JFWM-041.S1). Thus, the 2010 abundance esti-mate likely represents the adult population size at thelower bound because it follows a series of harsh years

Figure 5. Model-averaged apparent survival estimates of adult male Houston toads Bufo (Anaxyrus) houstonensis per surveyinterval on the Griffith League Ranch, Bastrop County, Texas in 2010. Error bars represent 1 standard error corrected for themaximum probability of apparent survival.



with low reproductive success (M. R. J. Forstner, unpub-lished data).

We assumed the high reproductive success observedin 2010 would have allowed the population to rebound,and the 2011 breeding season would have afforded usthe opportunity to quantify the influence of a goodreproductive year on adult population size. Unfortunate-ly, another severe drought in Bastrop County in 2011resulted in no detected Houston toad calling activity onthe GLR for the first time since we began surveying theproperty. Regardless, this study provides baseline infor-mation regarding Houston toad population size on oneof the primary recovery sites for the species, and will be auseful gauge to infer the efficacy of recovery initiatives inthe future.

Supplemental Material

Please note: The Journal of Fish and Wildlife Manage-ment is not responsible for the content or functionality ofany supplemental material. Queries should be directed tothe corresponding author.

Reference S1. Forstner MRJ, McHenry DJ, Gaston M,Villalobos L, Crump P, McCracken S, Jackson J, SwannackT, Bell J, Gaertner J, Mays S, Hahn D, Dixon JR. 2007. TheHouston toad 2007: annual summary of research andmonitoring. Report to the U.S. Fish and Wildlife Service,Austin, Texas.

Found at DOI: http://dx.doi.org/10.3996/072011-JFWM-041.S1 (497 KB PDF).

Table S1. Capture history of adult male Houstontoads at breeding ponds on the Griffith League Ranch,Bastrop County, Texas, in spring 2010.

Found at DOI: http://dx.doi.org/10.3996/072011-JFWM-041.S2 (18 KB XLSX).

Acknowledgments

We are grateful to J. Bell, A. Harper, M. Vandewege, M.Gaston, B. DeVolld, and J. Barnett for field assistanceduring this study. The Capitol Area Council of the BoyScouts of America provided access to the GLR, and weappreciate their continuing support of our research. B.Halstead, C. Dodd, J. Hatfield, and the Subject Editorprovided valuable comments that improved an earlierversion of this manuscript.

This research was funded by the U.S. Fish and WildlifeService, Texas Parks and Wildlife Department, and Environ-mental Defense Fund. Trapping and handling permits wereprovided by the Texas Parks and Wildlife Department(permit SPR-0102-191), and the U.S. Fish and WildlifeService (permit TE 039544-0). This research was approvedby the Texas State University - San Marcos InstitutionalAnimal Care and Use Committee (0810_0208_11).

The use of trade, firm, or product names is fordescriptive purposes only and does not imply endorse-ment by the U.S. Government.

References

Brown DJ, Baccus JT, Means DB, Forstner MRJ. 2011.Short-term outcomes for juvenile amphibians after fire

in a southern USA pine forest. Journal of Fish andWildlife Management 2(2):135–145.

Brown DJ, Swannack TM, Forstner MRJ. In press.Herpetofaunal survey of the Griffith League Ranch inthe Lost Pines ecoregion of Texas. Texas Journal ofScience.

Brown LE. 1971. Natural hybridization and trend towardsextinction in some relict Texas toad populations.Southwestern Naturalist 16:185–199.

Brown LE. 1975. The status of the near-extinct Houstontoad (Bufo houstonensis) with recommendations for itsconservation. Herpetological Review 6:37–40.

Brown LE, Mesrobian A. 2005. Houston toads and Texaspolitics. Pages 150–167 in Lannoo M, editor. Amphib-ian declines: the conservation status of United Statesspecies. New Jersey: University of California Press.

Burnham KP, Anderson DR. 1998. Model selection andinference: a practical information-theoretic approach.New York: Springer.

Buzo D. 2008. A GIS model for identifying potentialbreeding habitat for the Houston toad (Bufo housto-nensis). Master’s thesis. San Marcos: Texas StateUniversity - San Marcos.

Camper JD, Dixon JR. 1988. Evaluation of a microchipmarking system for amphibians and reptiles. TexasParks and Wildlife Department, Research Publication7100-159.

Dixon JR, Dronen NO, Godwin JC, Simmons MA. 1990.The amphibians, reptiles, and mammals of Bastropand Buescher state parks: with emphasis on theHouston toad (Bufo houstonensis) and the short-tailedshrew (Blarina sp.). Report to the Texas Parks andWildlife Department, Austin, Texas.

Dodd CK. 2009. Amphibian ecology and conservation: ahandbook of techniques. Oxford: Oxford UniversityPress.

[ESA] Endangered Species Act of 1973, Pub. L. No. 93-205, 87 Stat. 884 (Dec. 28, 1973).

Forstner MRJ, McHenry DJ, Gaston M, Villalobos L, CrumpP, McCracken S, Jackson J, Swannack T, Bell J, GaertnerJ, Mays S, Hahn D, Dixon JR. 2007. The Houston toad2007: annual summary of research and monitoring.Report to the U.S. Fish and Wildlife Service, Austin,Texas (see Supplemental Material, Reference S1; http://dx.doi.org/10.3996/072011-JFWM-041.S1). Texas Jour-nal of Science 40:454–456.

Freed PS, Neitman K. 1988. Notes on predation on theendangered Houston toad, Bufo houstonensis. TexasJournal of Science 40:454–456.

Gaston MA, Fuji A, Weckerly FW, Forstner MRJ. 2010.Potential component allee effects and their impact onwetland management in the conservation of endan-gered anurans. PLoS One 5:e10102.

Given MF. 2002. Interrelationships among calling effort,growth rate, and chorus tenure in Bufo fowleri. Copeia2002:979–987.

Gottschalk JS. 1970. United States list of endangerednative fish and wildlife. Federal Register 35:16047–



16048. Available: http://ecos.fws.gov/docs/federal_register/fr27.pdf (August 2011).

Hatfield JS, Price AH, Diamond DD, True CD. 2004.Houston toad (Bufo houstonensis) in Bastrop County,Texas: need for protecting multiple subpopulations.Pages 292–298 in Akcakaya HR, Burgman MA,Kindvall O, Wood CC, Sjogren-Gulve P, HatfieldJS, McCarthy MA, editors. Species conservation andmanagement: case studies. Oxford: Oxford UniversityPress.

Heyer WR. 1994. Measuring and monitoring biologicaldiversity: standard methods for amphibians. Washing-ton, D.C.: Smithsonian Institution Press.

Hillis DM, Hillis AM, Martin RF. 1984. Reproductiveecology and hybridization of the endangered Houstontoad (Bufo houstonensis). Journal of Herpetology 18:56–72.

Jacobson NL. 1989. Breeding dynamics of the Houstontoad. Southwestern Naturalist 34:374–380.

Jackson JT, Weckerly FW, Swannack TM, Forstner MRJ.2006. Inferring absence of Houston toads givenimperfect detection probabilities. Journal of WildlifeManagement 70:1461–1463.

Judge KA, Brooks RJ. 2001. Chorus participation by malebullfrogs, Rana catesbeiana: a test of the energeticconstraint hypothesis. Animal Behaviour 62:849–861.

McHenry DJ. 2010. Genetic variation and populationstructure in the endangered Houston toad in contrastto its common sympatric relative, the coastal plaintoad. Doctoral dissertation. San Marcos: Texas StateUniversity - San Marcos.

McHenry DJ, Gaston MA, Forstner MRJ. 2010. Bufohoustonensis (Houston toad). Predation. Herpetologi-cal Review 41:193.

Murphy CG. 1994. Determinants of chorus tenure inbarking treefrogs (Hyla gratiosa). Behavioral Ecologyand Sociobiology 34:285–294.

Peterson MN, Allison SA, Peterson MJ, Peterson TR, LopezRR. 2004. A tale of two species: habitat conservationplans as bounded conflict. Journal of Wildlife Man-agement 68:743–761.

Price A. 2003. The Houston toad in Bastrop State Park1990–2002: a narrative. Texas Parks and WildlifeDepartment, Report 03-0401.

Schulte-Hostedde AI, Zinner B, Millar JS, Hickling GJ.2005. Restitution of mass-size residuals: validatingbody condition indices. Ecology 86:155–163.

Schwarz CJ, Arnason AN. 1996. A general methodologyfor the analysis of open-model capture recaptureexperiments. Biometrics 52:860–873.

Swannack TM. 2007. Modeling aspects of the ecologicaland evolutionary dynamics of the endangered Hous-ton Toad. Doctoral dissertation. College Station: TexasA&M University.

Swannack TM, Forstner MRJ. 2007. Possible cause for thesex-ratio disparity of the endangered Houston toad(Bufo houstonensis). Southwestern Naturalist 52:386–392.

Swannack TM, Grant WE, Forstner MRJ. 2009. Projectingpopulation trends of endangered amphibian speciesin the face of uncertainty: a pattern-oriented ap-proach. Ecological Modelling 220:148–159.

[USFWS] U.S. Fish and Wildlife Service. 1984. Houstontoad recovery plan. Albuquerque, New Mexico: U.S.Fish and Wildlife Service. Available: http://ecos.fws.gov/speciesProfile/pro file/speciesProfile.action?spcode =D004.

Wells KD. 1977. The social behaviour of anuran amphib-ians. Animal Behaviour 25:666–693.

Wells KD. 2007. The ecology & behavior of amphibians.Chicago, Illinois: University of Chicago Press.

White GC, Burnham KP. 1999. Program MARK: survivalestimation from populations of marked animals. BirdStudy 46(Suppl):120–138.



Documents

Estimating Abundance of the Endangered Houston Toad on a