Society for Conservation Biology
The Decline of Amphibians in California's Great Central Valley
Author(s): Robert N. Fisher and H. Bradley Shaffer
Source: Conservation Biology, Vol. 10, No. 5 (Oct., 1996), pp. 1387-1397
Published by: Blackwell Publishing for Society for Conservation Biology
Stable URL: http://www.jstor.org/stable/2386913
Accessed: 31/08/2010 19:39
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://links.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you
have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may
use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://links.jstor.org/action/showPublisher?publisherCode=black.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].
Blackwell Publishing and Society for Conservation Biology are collaborating with JSTOR to digitize, preserve
and extend access to Conservation Biology.
http://links.jstor.org
Great
DeclineofAiphibiansin California's
Th1e
Valley
Central
ROBERT N. FISHER*AND H. BRADLEY SHAFFER
ofCalifornia,
Davis,
SectionofEvolutionand Ecologyand CenterforPopulationBiology,University
Davis,CA 95616, U.S.A.
Abstract: Declines in amphibian populations are rarelyreportedon the communityor ecosystemlevel. We
combined broad-scalefield sampling withhistoricalanialysesof museum recordsto quantifyanmphibiandeclines in California'sGreat Central Valley.Overall,amphibians showed an unambiguouspatternaof decline,
dealthoughtheintensityof decline varied bothgeographicallyand taxonomically.Thegreatestgeograiphicail
cline was detectedin the countiesof theSacramento and SanaJoaquin Valleys.Two species,Rana auroraand
Bufoboreas were identifiedas the mostaffectedby decline,whereas Pseudacrisregillawas the least affected.
The Coast Range countieshad littleor no detectabledecline. Weprovide new evidenceimplicatingintroduced
predatorsas a primary threat.Introducedpredators occur at lower elevations than native species,and our
restriction
to higherelevationsitesfrom
data indicate thatfor some native species therehas been signaificanat
declininganma formerlybroader distribution.Our historicalapproach provides a strategyfor identifying
programs and provides an asphibian communitiesthatcomplementsmore detailed,long-termmnonitoring
sessmentof thepattern of change that is a necessaryprerequisitefor the developmentoffield experiments
thattesthypothesizedmechanismsof change.
La declinaci6nde anfibiosen el GranValle Centralde California
CombiResumen: Las declinacionesde anfibiosraramenteson rep)ortadasa nivel comunidad o ecosiste-ma.
namos muestreosde campo a gran escala con analisis historicosde registrosen museospara cuantificarlas
unl
declinaciones de anfibios en el Gran Valle Central de California.Enageneral, los anfibios maostraroni
patron de declinacion muy claro, aunque la intensidad de declinacion vari6 tanto geograificacomo taixon6micamente.La mayor declinaciongeograficase detect6en los conadadosy los valles de Sacramenitoy San
que
Joaquin. Dos especies,Rana auroray Bufoboreas,fueron las mas afectadaspor la declinaci6n,maientras
Pseudoacrisregillafuela menos afectada. La declinaci6n en los condados defranja costerafuepequefia o no
detectable.Proporcionamos evidencia de que depredadores introducidosson la principal amnanaza. Los
depredadoresintroducidosocurrenen elevaciones menoresque laGsespeciesnativas,y nuestrosdatos indican
a sitiosde mayorelevaci6n al partirde
restrintgidas
que algunas especiesnativas han sido significativamente
una distribuci6noriginal mas amplia. Nuestroenfoquehist6ricoproporicona una estrategiaparatidentificar
la declinaci6n de comunidades de anfibios que complementaa los prograinas de monitoreodetallado y al
largoplazo y proporciona una evaluacion del patr6n de cambio que es un prerequisitopara el desarrollocle
experimentosde campo para probar mnecanismos
hipoteticosde cambio.
'Addresscorrespondenceto R. N Fisherat his currentaddress.DepartmentofBiologv 0116(,Ulniversity
of California-San
Drive,La Jolla,CA 920)93, U.S.A.,(mailfisher@eeves. ucsd.eda).
1, 199f5.
Papersubmitted
July13,199s5;revised
manuscript
accepted
Decemsbe>r
Diego, 9500 Gilmilal
1387
Biology, Pages 138,X713297
(;onseiw;ation
Volomne 1(), No S, October 1996S
1388
DeclineofCaliforniaAmphibians
Introduction
Regionaland global amphibiandeclineshave attracteda
and controversy
greatdeal ofattention
(Blaustein& Wake
1990; Blaustein1994; Blausteinet al. 1994c;McCoy 1994;
Pechmann& Wilbur1994; Travis1994). The controversy
has centeredon the resolutionoftwo questions.First,to
whatextentare "declines"a realphenomenonratherthan
a misinterpretation
of the large fluctuationsin population size thatcharacterizemanynaturalamphibianpopulations(Heyer1979; Corn & Fogelman1984; Pechmann
et al. 1991; Dodd 1992; Hairston& Wiley1993; Kagarise
Sherman& Morton1993)? Second, if declines are real,
are theycaused by globalenvironmental
changeor by localized anthropogenicfactors(Barinaga1990; Blaustein
& Wake 1990; Blausteinet al. 1994c)? Althoughdocumentingthe existence of declines should logicallyprecede the search forcausal mechanisms,the urgencyof
and reversing,the wholesale collapse of
understanding,
worldwideamphibianpopulationshas motivateda search
forcauses even as amphibianbiologistsdebate the realityof globalpatternsof decline(Barinaga1990; Blaustein
& Wake 1990; Blaustein1994; Blausteinet al. 1994a,b;
McCoy 1994; Pechmann& Wilbur1994; Travis1994).
Partof the frustration
in documentingdeclinesin any
is thehistoricalnatureoftheprobspecies or community
lem (Reznicket al. 1994). We mustunderstandthe past
and abundanceof species to make meaningdistribution
fulcomparisonswiththepresent.The keyproblem,then,
is how to quicklyand accuratelycompile both historical
and currentinformation
forentireamphibiancommunities,while maintaining
highstatisticaland fieldmethodologicalstandards.This taskappears daunting,especially
at largegeographicscales,butwe feelthata focusedprogramof fieldsurveysand historicalanalysesof museum
recordscan provideinsightsinto both the patternsand
mechanismsof amphibiandeclinepriorto choosingsites
forlong-term
monitoring
programsor fieldexperiments.
Althoughsuch data will almostnecessarilybe less complete thanone mightdesire,theyoftenprovidethe only
formof historicalinformationavailable for many amphibian communitiesthatare currentlyunder scrutiny.
As Reznicket al. (1994) point out, museumcollections
maybe used to quantifythe local abundancesof individual species (ifcensus data forspecificsitesare available)
or to investigateshiftsin species compositionof entire
drainagesor otherlarge regions.Althoughdirectcomparisons of specificsites are obviouslyimportant,this
approach requiresthatcurrentsamplingbe conducted
at preciselythe same sites,and withthe same methods,
as theywere in the past (Barry& Shaffer1994; Drost &
Fellers1994). Unfortunately,
fewpast studieswere complete and rigorousenough to facilitatesuch directcomof limitedgeneralutility.
parisons,makingthisstrategy
In the absence of well-controlledfieldsamplingprogramsin the past, how can one collect meaningfuldata
ConservationBiology
Volume10, No. 5, October 1996
Fisher& Shaffer
on the historicaldecline of amphibiancommunities?
We
submitthateven withthe relatively
incompletesamples
thattypifymost museum collections,one can stilluse
thisinformation
to glean importantinsightsintohistorical patternsof distribution,
and thuscurrentpatternsof
decline.We providean exampleutilizingthe entirecommunityof pond breedingamphibiansin California'svalleyoak-grassland
and prairiehabitatsthroughout
theGreat
CentralValleyand Coast Range (Fig. 1). This grasslanddominatedlandscape has long been recognizedas a discreteecologicalandbiogeographical
community
(Hickman
1993). We followHickman(1993) in recognizingthree
thatrepresentdifferent
subcommunities
drainagesystems
in Fig. 1: theCoast Range,theSacramentoValley,and the
SanJoaquinValley.Six amphibianspecieswere historically
widespreadin thishabitatand a singleaquatic breeding
sitewouldgenerally
containbetweentwo and fivespecies
(Feaver1971; Barry& Shaffer
1994; Fisher,unpublished).
Methods
Between 1990 and 1992 we surveyedaquatic habitats
forreproductivepopulationsof pond-breedingamphibiNo. ofSpecies
6543210123
-J
*
U
Sacramento
Valley
Sacramento Valley
I
o
_
ucdis
Nativespeciespreae
Nativespeciesabsent
JE|Valley
sii oak-grassland
habitat
B
Recet survey
localty
San Joaquin
No. ofSpecies
6543210123
Coastg
Ul
No. ofSpecies
6543210123
Los Ange
0
50. IO1
00
4
3003
|Kibmetem
Figure 1. Aquatic sitessurveyedfor native amphibians in Californiaduring 1990-1992. VerticalHistograms indicate thenumbersof native and introduced
speciespresent(fromcurrentfieldsurveys)and native
species absent (by subtraction,from historicmuseum
data) by countywithineach subregion.The counties
of each subregionare plottedon thehistogramfrom
top to bottom.Thefirstcountyis theone furthestwest
and north,and thelast one is thefurthesteast and
south,theyprogressin orderfrom westto east.
Fisher& Shaffer
Declinie
ofCaliforniaAniphibians 1389
and California red-leggedlfrog (Rcanaclciturorca).We also
ans. We considereda populationreproductiveifit consurveyed for introdlucedlspecies (frogs andl fish): Gatintainedlarvaeduringthe appropriatemonthsof the winbitsics c{ffinis(mosquito fislh),fish (all other species of
ter or spring.We sampled vernal pools, naturaland
fishfound in any pondts,all of which were usually introartificialponds, and marshlandsfor larval amphibians
dluced), and Rcanaicatesbe/icana(bullfrog). Mosquito fish
duringthelatewinterand springto maximizeour ability
were recorded separately from other fish because they
to detect this breedingactivity(Barry& Shaffer1994;
introdlucedlas biological control agents
are intentionally7
Heyer et al. 1994). A few sites were sampled several
andl are known predlators of amphibian eggs and
timesacross yearsand seasons to determinethe period
hatchlings (L. Kats, personal communication; P. Trenof greatestlarvalabundanceforthe regionalsamplingof
ham, personal communication; (Grubb 1972). We didl
localities. Under natural conditions vernal pools are
species of innot sample intensivelyfor each indtividlual
Therefore,
ephemeraland fillrelativelysynchronously.
troduced fishbecause several species may be present in
mostlarvaetendto use ponds duringthe same periodof
a single ponld, andlthey,appear to have similarecological
time,and sitescould be sampledonce duringa season to
effectson native amphibians. Rather, afterit was estabdeterminethe presence of a species at a site. We samlishedtthat native amphibians were or were not present,
et al.
meshseinenets(Shaffer
pled pondswitlh3-to &mmn
we movedl on to the next pondl. Questionable identifica1994; Fisher1995). From1986 to 1994 we sampledover
1000 aquatic sites,althoughforthisanalysiswe onlyincludedthe315 ponds(from28 counties)forwhichwe had
A
a completesurveyofbioticdata.These sitescovermostof
of the valleyoak-grassland/prairie
the extantdistribution
s
!
,
6- ?
Historic
exceptfora smallsectionoftheMonterey
community
Bay
5-:
coast where the extremelyfragilepopulationsof Santa
,,, - Xh<;'
Cruz long-toedsalamander(Ambystoinamnacrodactylim
3croceum)occur(Russell& Anderson1956).
Xr
Recent
Our samplingprotocolwas designedto sample locali_
2
E
z
ties evenlyacross wide areas to detectregionalpatterns
We generallymaintained
of currentspecies distribution.
0
a distance of 10-30 km between sites unless suitable
habitatwas not available (in which case the dlistances
Sacramento
Valley
CoastRange
between sites were greater).If an abundance of excelSan JoaquinValley
lenthabitatwas availablewe sampled moreintensively.
B
To maximizethe chances of locatingbreedingactivityif
it was present,we concentratedon aquatic sitesthat,in
28our subjectiveview, were "good" amphibianbreeding
24
habitat.In vernal pool complexes with variable pond
20
Xh
~~~Historic
sizes,we concentratedour samplingin the largestpools
0 16because theygenerallyharborthegreatestamphibiandi~12-4 Xr
Xh
O
versity.By concentratingon the (apparently)best am8 0z
phibian habitat,our localitiesrepresentthe minimum
4
number of reproducingpopulations of amphibians.
0
Thus, theyshould not be considereda definitiverefer2
1
4
6
3
5
ence forthe presence or absence of amphibiansin an
species
Rank-ordered
area but rathera reasonableindicationofthe patternsof
Figutre2. The number of nctive amphibican species
activitiesin primehabitatin a given
amphibianbreecding
found in each of 28 couniitiesin thr,eesubregions of the
region.We used counties as our geographicalunit for
Great Centrcl Vat/ley
pistor-icclly a(tndduriing ouir stiranalysisbecause our inabilityto findeven a singlelocalCouinties acrerallnkor-deredwithin each regionab
vneys.
ityfora species in a countyseems a reliableindication
the numinber
of recenitspecies occutrrenices.The miaecani
thatitis eitherrareor absentin a region.
niiiunberof species per county historicailly(XI,) differs
We recordedthepresenceor absence ofthefollowing
fron thactin our survneys(X,.; pa(ired t test,p < 0.000 1)
nativeamphibiansforeach pond: Californian-ewt(Tarof counlitiesin u'nhicheach species
(A) anad the nulmnber
icha spp. includingboth T granulosa and T torosa t[orT toroscasioccurs, historically anad durinigour surveys (B. 1, P.
osa, but notthe streamand riverinhabitant,
Californiatiger
regilla,2, A. californiense; 3, S. hammondii; 4, B. boreas;
errae fromthe SierraNevada footlhills);
Pacifictreefrog
5, Taricha spp.; 6, R. aurora); XI, eandX,. are the miaecian
californienise);
salamander(Amnbystomna
of couintiesper species historicclly and from
numiiiabers
(Pseudacris (Hyla) regilla); western spadefoot toad
our
.
r'espectiveip]
surv2ey7s,
westernToad (Buf0oboreas);
(Scaphiopus bwamJnondif);
-
Bbol)gv
1(), NI) 5. OuCohetr1996t
Q)fllst rv,atbon
Volumelt
1390
DeclineofCaliforniaAmphibians
tions were confirmedby P. Moyle (fishes) and M. Jennings (amphibians). Elevationand several other biotic
and abioticparameterswere measuredat each site fora
separate,ecological studyof amphibianpopulationecology(Fisher1995).
Historicdistributiondata were compiled for the six
native species based primarilyon museum specimens
fromthe CaliforniaAcademyof Sciences and the Museum of VertebrateZoology (Universityof California,
Berkeley)because thesetwo collectionshouse thegreatest numbersof specimensfromthe studyregionand incorporatethemostindividuallocalities.These datawere
supplementedwith unpublisheddata (MarkJennings)
forR. aurora, P. regilla,and S. hammondii.Ambystoma
californiense records were extractedfrom California
Departmentof Fish and Game's NaturalDiversityData
Base, whereasthoseforR. catesbeiana were takenfrom
Buryand Luckenbach(1976). Historicoccurrenceswere
onlyincluded in our determination
of the past distribution of a species if theywere collected fromour samplingregion.We recordedhistoricalelevationsfromthe
CaliforniaDepartmentof Fish and Game's data base for
all availablelocalitiesofA. californienseand determined
elevationsforS. hammondii and B. boreas usingtopographicmaps. We examined these threespecies in detail because the numberof currentand historicrecords
forthemwere verysimilar,suggestingthata directcomparisonof averageelevationsover timewould be meaningful.
Almostall of the historicrecordsin thisdatabase are
fromthe 1890s-1980, and we considerthemto representa reasonablesampleof the recenthistoricrangefor
thesespecies. However,a fewcountiesappear to be systematicallyunder-sampled.We deleted Kings County
O. .X
. 5;A 1.
I
- .^w:F
Fisher& Shaffer
fromtheanalysisbecause onlytwo individualamphibian
specimensare knownfromthe county.We retainedSutter, Yuba, and Placer counties in our analysis,even
thoughseveralspecies thatprobably"should"have been
historically
presentcould not be confirmedin museum
collections.If the historicalrecordsfor these counties
are an inadequaterepresentation
of the truedistribution
of several common species, this will render our estimates of decline conservativebecause these counties
would actuallycontaina greaterloss of species thenwe
detecteddue to inadequatehistoricalsampling.
Results
Distributional
TrendsacrosstheGreatCentralValley
We documenteda statistically
decline in the
significant
number of species currentlyfound in most counties
compared with that found historically,
supportingthe
that a community-wide
interpretation
decline has occurred(Fig. 2A). This decline variesbothgeographically
(Fig. 2A) and taxonomically(Fig. 2B). Geographically,
the Coast Rangehabitatshows relatively
littleoveralldecline (mean difference
in numberof species/countybetween currentand historicalrecords = 0.71), whereas
the Sacramento and San Joaquin Valleys both show
larger,virtuallyidenticaldeclines (San Joaquin Valley,
mean difference= 2.22; SacramentoValley,mean difference = 2.08; Kruskal-Wallis
Test comparingthe three
groups, H corrected for ties = 7.981, DF = 2, p =
0.018). Paired comparisonsof the three areas indicate
thatthe Coast Range has significantly
less decline than
the othertwo areas (t test,p = 0.004 to SanJoaquinVal-
Figure 3. Historic recordsfor Ambystoma californiensefrom the
Great Central Valley (A). The
present distribution of A. californiense in the Great Central Valley
based on our surveys(B). Shaded
area as in Fig. 1.
Conservation Biolog)
Volume 10, No. 5, October 1996
DeclineofCaliforniaAmphibians 1391
& Shaffer
Fisher
".O
A. ,W
,450
,
_
,
1
^
iO,
Figure 4. Historicrecordsfor
Scaphiopushammondiifromthe
Great CentralValley(A). The
~~~~~~~~present
distributionofS. hammondii
jwnlC0,<s;-
3C!t,0; j;SJ
in~~inthe
GreatCentralValleybasedon
our surveys (B). Shaded area as in
Fig. 1.
ley; p = 0.013 to SacramentoValley; correctedusing
Bonferroni
methodformultipletests[Rice 19891).However,the Sacramentoand SanJoaquinValleysdid not sigin amountof decline (p = 0.753).
nificantly
differ
Althoughall species show some decline, theyrange
fromslight(1/28 countieslost forP. regilla) to extreme
(24/28 countieslost forR. aurora) (Fig. 2B). To examine these individualspecies patternsin more detail,we
plotted the recent and historiclocalities in the Great
CentralValleyforfournativespecies and the bullfrog.
For this analysiswe plotted up to seven localitiesper
to maintaina similar
countyforthe historicdistributions
10- to 30-kmdistancebetween localitiesas we did for
our fieldsurvey.We also plottedall of our localitiesfor
each species on the recentmaps,forvisualcomparison.
Figures3-7 illustratethe historicand recent distributions of A. californiense,S. hammondii, B. boreas, R.
aurora, and R. catesbeiana, respectively,within the
GreatCentralValley.We did not plot P. regillaor T torosa because the formershowed essentiallyno decline
and the latteris onlyperipherally
presentin partsof the
Sacramentoand SanJoaquinValleys.
we
When compared with theirhistoricdistributions
see a gradientin the degree of decline across the native
species. A. californiense(Fig. 3), which stilloccurs over
much of its previousdistribution,
appears to be in the
initial stages of fragmentation
and decline. It is still
presentin most counties,althoughperipheralpopula-
Figure5. HistoricrecordsforBufo
boreasfromtheGreat CentralValley (A). The present distribution of
B. boreas in the Great Central Val-
ley based on our surveys(B).
Shaded area as in Fig 1.
ConservationBiology
Volume 10, No. 5, October 1996
Fisher& Shaffer
DeclineofCaliforniaAmphibians
1392
.....
..
...
.X
.s
.....
S
E
..
\
LA
tionsin the Sacramentoand easternSan JoaquinValleys
are no longer extant.S. hammondii (Fig. 4) shows a
more severe patternof decline with virtuallycomplete
extirpationfromthe SacramentoValley,and a reduced
densityof populationsin the easternSanJoaquinValley.
B. boreas (Fig. 5) shows a furthercontinuationof this
no recruitment
duringour sampling
trendwithvirtually
period in the Sacramentoor SanJoaquinValleys,but apparentlyviable populationsin the Coast Range. Finally,
R. aurora (Fig. 6) is in thefinalstagesofdecline,and we
founditin onlya fewisolatedponds in the Coast Range.
Thus, at the individualspecies level we see a common
amphibians
sequence of decline in nativepond-breeding
0
.;L4===
Conservation Biology
Volume 10, No. 5, October 1996
Figure 6 Historic records for Rana
aurorafrom the Great Central Valley (A). The present distribution of
R. aurora in the Great Central Valley
. . on
Lt our surveys (B). Shaded
based
area as in Fig. 1.
-A
tFE
l
L}
with losses occurringfirst(and most severely)in the
SacramentoValley,followedby the San JoaquinValley,
and finallyby the Coast Range.
forthe introducedbullfrogare
Changesin distribution
bullfrogshave been found
shown in Fig. 7. Historically,
throughoutthe Sacramentoand easternSanJoaquinValleys in permanentponds, sloughs,and rivers,with apparentlylimitedinvasioninto Coast Range habitats.Althoughour currentsamplingwas oftenin suboptimal
bullfroghabitat,we still found it abundant in ponds
throughoutthe SacramentoValley,but onlyin scattered
localities in the Coast Range and San Joaquin Valley.
However,we did recordit forSan Benitoand Monterey
A
1
-%
Figure 7. Historic recordsfor Rana
catesbeianafrom the Great Central
Valley (A). The present distribution
of R. catesbeiana in the Great Central Valley based on our surveys
(B). Shaded area as in Fig. 1.
Fisher& Shaffer
DeclineofCaliforniaAmphibians 1393
andintroduced
fish
amphibians
ofnative
Table1. Co-occurrence
the1990-1992
during
Central
Valley
inpondsintheGreat
andfrogs
surveys.*
aquatic
Introduced fish and frogs
Native amphibians
Ponds with
Pondswithout
Total
Ponds with
Ponds without
Total
38
195
233
39
77
45
238
82
315
*n = 315 ponds; G value = 28.3, df= 1,p < 0.0001.
counties where it did not occur in 1976 (Bury & Luckenbach 1976) indicating that it is still in the process of
range expansion where appropriate habitat exists.
Species Co-Occurrencesand ElevationalShifts
Because we collected distributional data on the three
major groups of potential exotic predators (mosquito
fish,other fish, and bullfrogs),we can examine the pattern of co-occurrence and habitat shifts of native amphibians and introduced exotics. Table 1 shows that
there is a significant,inverse relationship between introduced exotics and native amphibians (G test comparing
co-occurrence of all native species with all exotics, G
value = 28.3, DF = 1, p < 0.0001), and this pattern
holds individually for virtually all native species for
which we have reasonable sample sizes (Fisher 1995).
When we tested the three individual predator classes,
we also found a significant,inverse relationship with native species, although the frequent co-occurrence of several introduced exotics in the same pond makes it impossible to unambiguously assess the impact of each
predatorclass on native amphibian species. Thus, although
native and introduced species do sometimes co-occur,
the vast majorityof ponds harboring native amphibians
lack introduced species. We tested for an elevational
component to this pattern of negative co-occurrence
both across our entire sampling area (Fig. 8A) and in the
Coast Range (Fig. 9A) because it shows the least level of
decline. In both cases the mean elevation of all native
species was above the mean elevation of our samples,
whereas the mean for introduced fishes, mosquito fish,
and bullfrogs were below the mean elevation of our
samples and below that of all native amphibians. Finally,
for the three species for which we observed a decline
but could stillfind enough samples to reliablydocument
elevationalranges (B. boreas, S. hammondii, A. cali-
forniense), historical localities were significantlylower
in elevation than are current viable populations (Figs.
8B, 9B). This upward shiftin elevation is not due to our
inability to find and sample low-elevation sites in our
currentfield work because the majorityof our sampling
effortwas from0 to 200 m elevation. Rather,most of the
siteswe sampledwere unoccupiedbynative
low-elevation
species,comparedwitlhtheirhistorical
patternof distribution,implying
theyhavebeen lostfromtheseareas.
Discussion
Threeprimaryresultsstemfromour analysisof distributional patternsof native amphibians and introduced
predatorsin the Great CentralValley. First,all native
species examinedappear to show an overalldecline in
theirdistribution.
Second, nativeamphibiansand introtendnot to co-occur,withinduced fishesand bullfrogs
troducedexotics occupyinglow-elevationsites,and native species persistingprimarilyat higher elevations.
Third,the threespecies of nativeamphibiansforwhich
we have sufficient
samples,have become restrictedto
higherelevationsof occupied sitesover the last several
decades. It thus seems plausible that habitatmodifications,low levels of topographicrelief,or a combination
A
600
500
400
Nativespecies
(n=
(n=19)
i
?
(n=91)
(n=49)
0
~~~(n=57)
>
C) 200-
(n
Introduced
species
I*
=177_1
I
100
0-
I
I
1
(n=17)
(n=43)i
I
2 3 4 5 6 7 8
Rankorderedspecies
9
B
400
(n=49)
-300-
FPcccnt
/(n=57)
i(n=91)
0Z
Bufo
boreas
Scaphiopus Ambystoma
hammondii californiense
Figure8. Currentelevational distributionof native
and introducedspecies rank orderedby elevation: 1,
Tarichlaspp.; 2, B. boreas;3, R. aurora;4, S. hammondii;
5, A. californiense;
6, P. regilla;7,Fish;8, R. catesbeiana;
9, G. affinis(bars representSE and sample sizes [ni are
in parentheses) (A). Historicversusrecentelevations
of threenative amphibians (bars representSE and
sample sizes [n] are in parentheses) (B). The asterisk
( ) is themleanelevationofall 315 ponds in our survey.
(;o1nsewatiofl Biology
Volumen 1(), No 5, Oc:tobecr1996
1394
DeclineofCaliforniaAmphibians
Fisher& Shaffer
A 600
Native species
500
E
400
(n=16)
(n=60) l
(n=5)
ai 200
species
(n=87)
(
300 -
>t
Intoue
(n=23) I(n-35)
J
I(n=2)
(n=6)
i
I
100
0
1
2
3
4
5
6
(n=1
5)
7
8
9
Rank ordered species
B
500 -
Recent
400
E
c
.?
>
W
300-
-n;-60)
(n=36)
CZ~~~~~~~~
200 -
(n=60)
(n=23
(n=35)
-
*
/
(n=16)
/
!
(n=23)
i(=3
Historic
100
Bufo
boreas
Scaphiopus
hammondii
Ambystoma
californiense
Figure9. Currentelevational distributionof native
and introducedspecies in Coast Range onlyrank ordered by elevation: 1, Tarichaspp.;2, S. hammondii;3,
R. aurora;4, B. boreas;5, A. californiense;
6, P. regilla;7,
G. affinis;
8, Fish;9, R. catesbeiana(bars representSE
and sample sizes [ni in parentheses) (A). Historicversus recentelevationsof threenative amphibiansfor
theCoast Range only (bars representSE and sample
sizes [ni are in parentheses) (B). The asterisk(*) is the
mean elevationof all 134 ponds in theCoast Range.
offactorshave allowed exotic species to invadelow-elevationsites,and thishas contributedto the overallpatternof decline seen in thenativeamphibians.
Beforeexaminingthishypothesisin greaterdetail,it is
importantto recognizethe potentiallimitationsto our
analysis.As emphasized by Reznick et al. (1994), the
we wish to makerelyon thequalkindofinterpretations
ityof the museumrecordsforbaselinehistoricaldata as
inwell as our own samplingforcurrentdistributional
formation.
Althoughbothare subjectto manysourcesof
error,we feel theyprovidea reasonablebasis fordocumentinghistoricalpatternsat the coarse scale over
which we are working.Severalpotentialsources of errorhave been identifiedin the literature,
including(1)
and (2) year-tolocal, stochasticpopulationfluctuations
Both of these
yearvariationin patternsof recruitment.
may effectthe reliability
of eitherthe museumrecords
ConservationBiology
Volume 10, No. 5, October 1996
or our own sampling;however, the key question is
whethereitherwill lead to a systematicbias thatwould
be interpreted
as a decline over time.Assumingthatlocal, stochasticpopulationfluctuationsare a normaloccurrence (Shaffer& Fisher unpublished data; Heyer
1979; Corn & Fogelman 1984; Pechmann et al. 1991;
Dodd 1992; Hairston& Wiley1993; KagariseSherman&
Morton1993), we mightnot recordall potentialspecies
at a particularbreedingsite eitherin our currentsamplingor in museumrecords.However,unlessa local extirpation(i.e., across an entirecounty)has occurred,we
shouldrecordthemat a nearbypool and stillscore them
as regionallypresenthistorically
and currently.
Because
we were not quantifying
the numbersof individualsat
each pond, even severe fluctuations
will not affectour
resultsas long as a few individualsreproducemost seasons (Trenerryet al. 1994; Hayek& Buzas in press). In
addition,the size of our samplingunit(the county)ensures thatwe will averageover small-scalefluctuations
and only tallydeclines thatencompass manyhundreds
to thousandsof square kilometers.
due to
The effectsof yearlyvariationin recruitment
factorsare potentiallymore
widespread environmental
severe,especiallybecause mostlocalitieswere onlysampled once duringour 3-yearsurvey.In particular,ifour
surveyswere conducted duringparticularlydry years
and certainspecies did not breed while othersdid, we
a lack ofbreedingdue to low
mightincorrectly
interpret
rainfalllevels as a real,species-specificdecline. Circumstantialevidence suggeststhisis not a seriousconcern,
but may influencea few species. First,our 3 sampling
low-rainfall
winters.Ifwe use
yearswere not uniformly
Davis (Yolo county,a southernSacramentoValleysite)
as representative
of the GreatCentralValley,the 19891990 rainfall(1 July-30June)was 15.85 inches (91% of
the 17.27 inches 100-yearaverage),1990-1991 was 82%
of normal,and 1991-1992 was 100% of normal.Viewed
anotherway, of the 15 yearsfrom1980-1981 to 19941995, our 3 yearswere rankedsixth,ninth,and eleventh
wettest.Thus,althoughnone of our samplingyearshad
experiencedextremelyhighrainfall,
theywere neverso
low thatwe would expect widespread suppressionof
amphibianbreedingactivities.Evidence fromlocalities
thatwe have visitedseveraltimessheds some additional
lighton thisproblem.At two localitiesthatwe visited
multipletimesover severalyears,we noted a consistent
patternofpresenceand absence of species regardlessof
yearlyrainfall:A. californienseand P. regilla larvae
were alwayspresent,whereasthe otherspecies were always absent.Nevertheless,at our long-term
studysiteat
the HastingsReservationin MontereyCounty,H.B.S. has
foundthatA. californiense,P. regilla,and T torosa larvae are reliablypresent regardlessof rainfallbut B.
boreas tadpoles and metamorphsfluctuatemuch more
with rainfallpatterns(Shaffer& Koenig, unpublished
data). Althoughthe data are less complete,fieldworkin
Fisher& Shaffer
1993-1994 and 1994-1995 by RNF in southernCalifornia (San Diego, Orange,and Riversidecounties)demonstratesa similarspecies-specificeffect:B. boreas and 5.
variationin recruitment
hammondii show year-to-year
dependingon rainfallpatterns,whereas P. regilla is a
more consistentbreederacross years(Ambystomaand
Rana do not occur at these sites). Thus, it is possible
that the relativelysevere declines we observed in B.
boreas and S. hammondii are in part due to reduced
breedingactivityassociatedwith moderaterainfallduring 1990-1992. Nevertheless,we did observe some
breedingin all species in all years,which confirmsthat
rainfallto induce breedingin the retherewas sufficient
gions we were sampling.Finally,even if the apparent
of rainfall
decline in some species is partiallya ftinction
patterns,itpointsout theinherentdemographicinstabilto conityin thesetaxa and theirpotentialsusceptibility
species thatunIn particular,
tinualhabitatmodification.
in populationsize maybe
dergolargeyearlyfluctuations
extremelysensitiveto habitat changes that interfere
with recolonizationsuccess compared to species that
maintain relativelyconstant population sizes across
years(Travis1994).
What mechanism,or set of mechanisms,mightbe
causing amphibiandeclines in the GreatCentralValley
canIn westernNorthAmericatheprimary
ofCalifornia?
didatesidentifiedto date are increasedultraviolet(particularlyUVB) radiation(Blausteinet al. 1994a), and introduced predatoryspecies (Moyle 1973; Hayes &
Jennings1986; Bradford1989; Bradfordet al. 1993). Althoughwe have no data on LTVBas a mechanismfordefactor
cline in the CentralValley,it maybe an important
in the decline of severalspecies in Oregon(Blausteinet
al. 1994a). However,over the rangeof habitatswe examined, declines are most severe at low elevations,
where the harmfuleffectsof UVB should be lowest.
Thus,in the absence of directevidenceon ambientUVB
levels,it does not appear to be a majorstressorin the
GreatCentralValleysystem.
Consider the causal hypothesisthat introducedspecies are a potentialmechanismof decline,ratherthana
spuriouscorrelation.Most introducedexotics are relafaunaof Caltivelyrecentcomponentsofthe freshwater
ifornia(Moyle 1976; Moyle et al. 1982; Moyle 1986;
Leidy& Fiedler1985; Swiftet al. 1993) and are concentratedat elevationsbelow 150 m in our surveys(Fig.
8A). Ifintroducedspecies have caused declines,then(1)
be missingfromlow
nativeamphibiansshouldcurrently
elevationswhere exotics are common, and (2) there
to highermean
should have been a historicrestriction
elevationsin the native amphibiansas they are eliminatedfromlow-elevationsites.Bothofthesepredictions
are borne out by our analysis(Table 1, Figs.8 & 9), and
it appears most stronglyin the SacramentoValley(Fig.
1). The same process may be in effectin the Coast
Range (Fig. 1), althoughthere is apparentlystillsuffi-
Declineof(CliforniaAmphibians 1395
cient unaffectedhabitatin most countiesthatour very
conservativeestimateofdeclinehas notyetregisteredin
this region. However, we predictthatwidespread declines will become apparentif exotic species continue
to spreadin low-elevationCoast Rangehabitats.A different patternseems to be in effectin the San JoaquinValley(Fig. 1), where we foundfew introducedexoticsyet
catastrophicamphibiandeclines.The San Joaquinis the
farmedand mostmodifiedof the three
mostintensively
of amphibianderegions.One plausible interpretation
clines here is thatthe few remainingvalleyfloorponds
and pools have been so affectedthattheyare no longer
habitableeven forintroducedspecies (Ohlendorfet al.
1988; Parker& Knight1992; Saikiet al. 1992). Thus,the
onlyremaininghabitatin the regionconsistsof higherelevationsitesthatseldomcontainexotics.
Teasingapartthe independentand combinatorialcontributionsof multiplefactors,includingincreasedUVB
(Blaustein1994) and exotics,presentsa majorchallenge
to reversingamphibiandeclines. For UVB thereis reasonable documentationthat ambientlevels can cause
for some species (Blausteinet al.
embryonicmortality
1994a). For exoticsall threeintroducedcategorieshave
been shown to preyon congenersof our nativespecies
or in the field(L. Kats, personal
eitherexperimentally
communication;P. Trenham,personalcommunication;
Werschkul& Christensen1977; Hammerson1982; Jeninnings& Hayes 1985; Kats et al. 1988). Unfortunately,
troducedexotics tendto thrivein highlymodifiedhabitats,confoundinghabitatmodificationand degradation
withthe actual exotic predatoras the real source of decline (Moyle 1973; Moyle 1976; Moyleet al. 1982; Leidy
& Fiedler1985; Baltz & Moyle 1993; Swiftet al. 1993).
breedHowever,our frequentobservationof successfuil
ing activityby nativeamphibiansin extremelymodified
breedingsitesas long as theywere freeof exotics(Barry
& Shaffer1994; Shaffer& Fisherunpubl.) supportsthe
thatthe exotic species themselvesare an
interpretation
elementin thepath to declineand local extirimportant
pation. More complex interactionsmay also be taking
place; forexample,it is possible thatintroducedexotic
predatorsforce egg-layingor young tadpole development to occur in shallow water where the effectsof
UVB are moresevere.
We agree withReznicket al. (1994) thatthe historical
approachwe have takenin thisstudyprovidesa defensible, rapidmethodforassessingpopulationtrendsover a
suitablegeographicscale, and we encourage othersto
conduct similarsurveysforotherlarge regionalbiotas.
To date,onlya handfulof historicaltrendanalyseshave
been conducted on amphibiansin the westernUnited
States(Corn et al. 1989; Fellers& Drost 1993; Drost &
on the
Fellers1994), and we have verylittleinformation
patternsof historicaldecline or introducedspecies in
otherpartsof the world(e.g., Inger& Voris 1993; Richmanyregionslack the necesards et al. 1993). Althlough
Biolo)gy
(,o)lselrviotl)l
Volomlle 1(), No S, October 19)96
1396
DeclineofCaliforniaAmphibians
sary museum records to document historicalshifts,
collections
thereare enoughwithreasonable,long-term
shiftsin communityassemblages
to assess distributional
overa rangeof habitattypes.Such analyses,in combinalocal
tion with ongoingresearchprogramsquantifying
population demographyand experimentalanalyses of
mechanisms,maybe the best strategyto defineand to
beginreversingtrendsin amphibiandeclines.
Acknowledgments
We thankall thosewho providedassistancein the field:
S. Rollmann,S. Stanley,M. Meyers,P. Ustach,J. DiDonato,A. Gluesenkamp,D. Irschick,J.King,K. Lappin,M.
Littlefield,
C. Austin,and S. Sweet. Historicdata were
suppliedbyJ.Vindumof the CaliforniaAcademyof Sciences andJ.Rodriguezof the MuseumofVertebrateZoology. A. Blaustein,T. Case, M. Hellberg,R. Heyer,E.
Main,R. Radtkey,T. Schoener,M. Vicario,D. Wake, H.
read earWilbur,and one anonymousreviewercritically
lierversions.S. Stanleyand S. Rollmannprovidedlab assistance. M. Jenningssupplied unpublished distributionaldata. J. Brode,C. Brown,D. Duncan, R. Hansen,
and A. Roest providedadditionallocalityinformation.
J.
E.
Froke,J. Grant,N. Nediff,P. Patrick,M. Stromberg,
Tanaka,and M. Urittia,in particular,providedaccess to
land. The Departmentof Airand WaterResources(Universityof California,Davis) provided rainfalldata. Aspects of thisworkwere supportedby the CaliforniaDepartmentofFishand Game InlandFisheriesDivision(FG
NSF BSR 90-18686,
9422), the NationalParkService/NBS,
Station.
and theUC DavisAgricultural
Experiment
Literature
Cited
Baltz,D. M., and P. B. Moyle. 1993. Invasionresistanceto introduced
Ecologispecies by a nativeassemblageof Californiastreamfislhes.
cal Applications3:246-255.
Barinaga,M. 1990. Where have all the froggiesgone? Science 247:
1033-1034.
Barry,S. J.,and H. B. Shaffer.1994. The statusof the CaliforniaTiger
at Lagunita:a 5o yearuLpSalamander(Amnbystonci
cacll{forniense)
date.JouLrnal
ofHerpetology28:159-164.
Blaustein,A. R. 1994. Chickenlittleor Nero's fiddle?A perspectiveon
decliningamphibianpopulations.Herpetologica50:85-97.
Blaustein,A. R., and D. B. Wake. 1990. Decliningamphibianpopula5:
tions: a global phenomenon?Trends in Ecologyand EvoluLtion
203-204.
Blaustein,A. R., P. D. Hoffman,D. G. Hokit,J. M. Kiesecker,S. C.
Walls,andJ.B. Hays. 1994a. lN repairand resistanceto solarUV-B
in amphibianeggs: a linkto populationdeclines?Proceedingsof
the NationalAcademyof Science of the UnitedStatesof America
91:1791-1795.
Blaustein,A. R., D. G. Hokit,R. K. O'Hara, and R. A. Holt. 1994b.
Pathogenicfunguscontributesto amphibianlosses in the Pacific
Northwest.BiologicalConservation67:251-254.
neBlaustein,A. R., D. B. Wake, andlW. P. Sousa. 1994c. Amphibianl
of poptLlapersistence,andlsusceptibility
clines: judgingstability,
tionsto local and globalextinctions.ConservationBiology8:60-71.
ConservationBiology
Volume 10, No. 5, October 1996
Fisher& Shaffer
Bradford, 1). F. 1989. Allotypic distribution of native frogs and introduced fislhesin hiiglhSierra Nevada lakes of California-implication
of the negative effect of fish introdLuctions.Copeia 1989:775-778.
Bradford, I). F., F. Tabatabi, and I). M. Graber. 1993. Isolation of rein1tromaining populations of the native frog, Rcmci muIscosu, byT
duced fishes in Sequoia and Kings Canyon National Parks, California. Conservation Biology 7:882-888.
Bury, R. B , andl R. A. Luckenbach. 1976. Introduced amphibians and
reptiles in California. Biological Conservation 10:1- 14.
Corn, P. S., and J. C. Fogelman 1984. Extinction of montane populations of the nortlhernleopalrd frog (Ranai pifiens) in Colorado.
JoUrnalof Herpetology 18:147- 152.
Corn, P. S., W. Stolzenburg, and R. 13.Bury. 1989. Acid precipitation
studies in Colorado and Wyoming: interim report of sulveys of
montane amphibians and water chemistry. Biological report
80(40.26). UI.S.Fislhand Wildlife Setrvice,Washington, D.C.
I)odd, C. K. 1992. Biological diversity of a temporary pond lherpetofauna in north Florida sandhills. Biodiversity and Conservation 1:
125-142.
I)rost, C. A., and (G. M. Fellers. 1994. I)ecline of frog species in tlle
Yosemite section of the Sierra Nevadla. National Biological SuLrvey,
UJnliversity
of California, I)avis.
Feaver, P. E. 1971. Breeding pool selection and larval mortalityof tlhree
California amphibians: Amnb'stomon
tiar/nimn calzJifoniense Gray,
H'la
icyi/ Bairnland (Girard,and .Sa-(p/ioptts cmnmnondii
Girard.
44
M.S. tlhesis.Fresnlo State College, Fresno, California.
Fellers, G. M., and C. A. D)rost 1993. I)isappearance of tlle Cascades
frogRamnac(iscadue at the soutlherlnend of its range, California,
tUSA.Biological Conservation 65:177- 181.
Fisher, R. N 1995. Tracing hunmanimpacts in natural systems: geckos
in the Pacific and amphibians in the Central Valley. Ph.D. dissertation. Univelrsit)of California, I)avis.
(iruLbb,J. C. 1972. I)ifferentialpredation by GaeinibiusiaiaifJilsison the
eggs of seveln species of anuran amplhibians. Americall Midland
Naturalist 88:102-108.
Hairston, N. (G., and R. I1. Wiley. 1993. No decline in salamander (amphibia, caldata) populations-a 20-year stuLdyin the soutlhernAppalachians. Brimlevana 18:59-64.
Hammerson, (J. A. 1982. Bulllfrogeliminating leoparld frogs in Colorado? Herpetological Review 13:115-116.
Hayek, L.-A. C., and M. A. Buzas In press. Surveying natural popuLlaPress. New York.
tions. Columbia UJniversity
Hayes, M. P., and M. R. Jennilngs.1986. Decline of ranid frog species in
westerln NortlhAmerica: are buLllfrogs(Ranci cattesbeianca) responsible? Journalof Herpetology 20:490-509.
Heyer, W. R. 1979. AnllUal variation in larval amphibian populations
witlhina temperate pond. Journal of the Washington Academy of
Science 69:65-74.
Heyer, W. R., M. A. Donllelly, R. W. MeDiarmid, L.-A. C. Hayek, and
M. S. Foster, editors. 1994. Measuring and monitoring biological div7ersity:standard methods for amphibians. Smithsonian Institution
Press, Washington, I).C.
Hickman, J. C., editor. 1993. ThleJepson manual: Higlherplants of C alifornia. Uniiversityof California Press, Berkeley.
Inger, R. F., and H. K. Voris. 1993. A comparison of amphibian communities tlhrouLghl
time and from place to place in Bornean forests.
Journal of Tropical Ecology 9:409-433.
Jennings,M. R., and M. P. Hayes. 1985. Pre-1900 overharvestof California
forbullretl-leggetlfrogs (R(mna aur-ora(cl(rcto,nf): the inducement
41:94-103.
frog(Raina c(ltesbei(Il(I) introtduction.
Herpetologica
KagariseSherman,C., and M. L. Morton.1993. Populationdeclinesof
Yosemitetoads in the easternSierraNevada of California.Journal
of Herpetology27:186-198.
defenseand
Kats,L. B.,J.W. Petranka,and A. Sib. 1988. Antipredaltor
the persistenceof amphibianlarvaewithfishes.Ecology69:1865187(0.
Leidy,R. A., anldP. L. Fiedler.1985. Humandisturbanceand pratterns
DeclineofCaliforniaArnpbibians 1397
Fisher& Shaffer
in the San FranciscoBaydrainage.Biologiof fishspecies diversity
cal Conservation33:247-268.
McCoy,E. D. 1994. "Amphibiandecline":a scientificdilemmain more
ways thanone. Herpetologica50:98-103.
Rana catesbeiana,
Moyle,P. B. 1973. Effectsof introducedbullfrogs,
on the native frogsof the San JoaquinValley,California.Copeia
1973:18-22.
of California
Moyle,P. B. 1976. Inland fishesof California.University
Press,Berkeley.
Moyle,P. B. 1986. The invasionsof fishintoCalifornia.Pages 151-179
in H. A. MooneyandJ.A. Drake,editors.Ecologyof biologicalinvaNew York.
sions of NorthAmericaand Hawaii. Springer-Verlag,
Moyle,P. B., J.J. Smith,R. A. Daniels, T. L. Taylor,D. G. Price,and
D. M. Baltz. 1982. Distributionand ecology of streamfishesof the
of
Sacramento-SanJoaquindrainagesystem,California.University
CaliforniaPublications,Zoology 115.
H. M., R. L. Hothem,and T. W. Aldrich.1988. BioaccumulaO)hlendorf,
tionof seleniumbysnakesand frogsin the SanJoaquinValley,California.Copeia 1988:704-710.
inhabiting
Parker,M. S., and A. W. Knight.1992. Aquaticinvertebrates
saline evaporationponds in the southernSan JoaquinValley,California.Bulletinof the SouthernCaliforniaAcademyof Science 91:
39-43.
J.P. Caldwell,L.J.Vitt,
Pechmann,J.H. K., D. E. Scott,R. D. Semlitsch,
and J. W. (Gibbons.1991. Declining amphibianpopulations: the
problem of separatinghuman impacts fromnaturalfluctuations.
Science 253:892-895.
Pechmann,J.H. K., and H. M. Wilbur.1994. Puttingdecliningamphiband humanimian populationsin perspective:naturalfluctuations
pacts. Herpetologica50:65-84.
studiesoftropReznick,D., R.J.Baxter,andJ.Endler.1994. Long-term
ical streamfishcommunities:the use of fieldnotes and museum
collectionsto reconstructcommunitiesof the past.AmericanZoologist34:452-462.
Rice, W. R. 1989. Analyzingtables of statisticaltests. Evolution43:
223-225.
Richards,S. J., K. R. McDonald, and R. A. Alford.1993. Declines in
frogs.Pacific
populationsof Australia'sendemictropicalrainforest
ConservationBiology1:66-77.
Russell,R. W., and J. I). Anderson.1956. A diisjunct populationof the
long-toedsalamanderfromthe coast of California.Herpetologica
12:137-140.
Saiki,M. K., M. R. Jennings,and T. W. May. 1992. Seleniumand other
fishesfromthe irrigatedSanJoaquinValley,
elementsin freshwater
126:109-137.
California.Science and theTotal Environment
H. B., R. A. Alford,B. D. Woodward,S. J.Richards,R. G. Altig,
Shaffer,
and C. Giascon.1994. Quantitativesamplingof amphibianlarvae.
Pages 130-141 in W. R. Heyer,M. A. Donnelly,R. W. McI)iarmid,
L.-A.C. Hayek,and M. S. Foster,editors.Measuringand monitoring
standardmethodsforamphibians.Smithsonian
biologicaldiversity:
Institution
Press,Washington,D.C.
Swift,C. C., T. R. Haglund,M. Ruiz,and R. N. Fisher.1993. The status
fishesofsouthernCalifornia.Biuland distribution
ofthe freshwater
letinof the SouthernCaliforniaAcademyof Science 92:101-167.
Travis,J. 1994. Calibratingour expectationsin studyingamphibian
populations.Herpetologica50:104-108.
M. P., W. F. Laurance,and K. R. McDonald. 1994. FurtherevTrenerry,
idence forthe precipitousdecline of endemic rainforestfrogsin
tropicalAustralia.PacificConservationBiology1:150-153.
predation
Werschkul,D. F., and M. T. Christensen.1977. Differential
by Lepomis inacrochiruson the eggs and tadpoles of Ranam.Herpetologica33:237-241.
R
,
IL~~~~~~~~
)~~~~~~~~~~~~~~~~~~~~
-I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
(Conlsenationl Biolog)
VoluLme10, No. 5, Octobcr 1996