effects of ultraviolet radiation on amphibians: field experiments

JAN 2 a 1999AMER. ZOOL., 38:799-812 (1998)

Effects of Ultraviolet Radiation on Amphibians: Field Experiments'

ANDREW R. BLAUSTEIN,2* JOSEPH M. KIESECKER,! DOUGLAS P. CHIVERS,^

D. GRANT HOKIT,§ ADOLFO MARCO,^ LISA K. BELDEN,* AND AUDREY HATCH*

*Department of Zoology, 3029 Cordley Hall, Oregon State University, Corvallis, Oregon 97331-2914^School of Forestry and Environmental Studies, Yale University, 370 Prospect St.,

New Haven, Connecticut 06511^Department of Biological Sciences, 5751 Murray Hall, University of Maine, Orono, Maine 04469

^Department of Biology, Carroll College, Helena, Montana 59625^Departamento de Biologia Animal, Universidad de Salamanca, 37071, Salamanca, Spain

SYNOPSIS. Numerous reports suggest that populations of amphibians from a widevariety of locations are experiencing population declines and/or range reductions.In some cases, unusually high egg mortality has been reported. Field experimentshave been used with increasing frequency to investigate ultraviolet radiation asone of the potential factors contributing to these declines. Results from field ex-periments illustrate that hatching success of eggs is hampered by ultraviolet ra-diation in a number of species, while other species appear to be unaffected. Con-tinued mortality in early life-history stages may ultimately contribute to a popu-lation decline. Although UV-B radiation may not contribute to the population de-clines of all species, it may play a role in the population decline of some species,especially those that lay eggs in open shallow water subjected to solar radiationand in those that have a poor ability to repair UV-induced DNA damage.

INTRODUCTION

The fundamental problem in ecology isto determine the causes that limit the dis-tribution and abundance of organisms(Krebs, 1994). Examining this problem isespecially relevant today because popula-tions of many organisms are undergoing de-clines and range reductions as part of anoverall loss in biodiversity (Wilson, 1988;Ehrlich, 1997). As part of this "biodiversitycrisis," many amphibian populations havebeen declining and undergoing range re-ductions {e.g., Semb-Johansson, 1989;Wake, 1991; Crump et al, 1992; Richardset al, 1993; Pounds and Crump, 1994;Blaustein and Wake, 1995; Stebbins andCohen, 1995; Drost and Fellers, 1996; Fish-er and Shaffer, 1996; Laurance et al, 1996;Reaser, 1996; Pounds et al, 1998; Lips1998). Unfortunately, the causes for am-phibian population declines have been dif-ficult to assess. Much of the information onamphibian declines comes from observa-tional or anecdotal accounts (Blaustein etal, 1994a; Stebbins and Cohen, 1995). Hy-

1 An invited mini-review.2 E-mail: [email protected]

pothesized causes for the declines includehabitat destruction (the most obviouscause), pathogens, introduced exotic spe-cies, pollution, and increased ultraviolet ra-diation (Blaustein and Wake, 1995; Steb-bins and Cohen, 1995). These agents mayact alone or in combination to contribute tothe decline of amphibian populations. How-ever, few experimental tests of key hypoth-eses have been conducted, especially in thefield under natural conditions. This is sur-prising because field experiments have beenextremely useful tools in examining basicecological questions such as how amphibi-ans interact with one another, with other or-ganisms, and with their abiotic environment(Hairston, 1989, 1994).

The diversity of locations where amphib-ian populations have declined promptedconsideration of atmospheric factors suchas increased ultraviolet irradiance associat-ed with depletion of stratospheric ozone.Although still relatively rare, several inves-tigators have used field experiments to ex-amine the potential role of ultraviolet-B ra-diation (UV-B; 280-315 nm) in amphibianpopulation declines by measuring mortalityof embryos {e.g., Blaustein et al., \997a;Broomhall et al, 1999; Anzalone et al,

799

by guest on June 25, 2015http://icb.oxfordjournals.org/

Dow

nloaded from

http://icb.oxfordjournals.org/

800 A. R. BLAUSTEIN ETAL.

1998; Corn, 1998; Lizana and Pedraza,1998). Continuous high mortality in earlylife stages may ultimately contribute to adecline at the population level.

In this review, we briefly describe themethods, evidence and implications of stud-ies that have incorporated field experimentsto test the effects of UV-B radiation on am-phibian eggs and embryos. We also discusssome concerns and misinterpretations re-garding the use of field experiments in stud-ies of UV and amphibians. We have at-tempted to include all studies that have in-corporated field experiments or a field com-ponent to investigate the effects of UV-Bon amphibian embryos. Details of the meth-ods are included so that a more thoroughinterpretation can be made of the results.Laboratory studies are not reviewed.

UV SENSITIVITY HYPOTHESIS

Since 1979 we have been studying theecology of amphibians at several locationsin the Pacific Northwest (USA). As part ofthis research, we have monitored egg de-velopment and hatching success of Cas-cades frog (Rana cascadae), Pacific tree-frog (Hyla regilla) and western toad (Bufoboreas) embryos in the field (Blaustein andOlson, 1991; Blaustein et al., 19946; Kie-secker and Blaustein, 1997). From the1950s through the mid 1980s, egg mortalityin Oregon was no more than 10% for thesespecies (Kiesecker and Blaustein, 1997).Through the 1990s this remains the case forH. regilla but not for the other two species.At several sites, R. cascadae and B. boreassuffered high embryo mortality (greaterthan 50%). In some years, mortality wasover 90% at some sites (Kiesecker andBlaustein, 1997). Moreover, when eggswere taken from these high mortality sitesand reared in the laboratory, more than 99%survived through metamorphosis (unpub-lished data, A.R.B.). Heavy metals and pes-ticides were not detected in lake water andpH values were near neutral. Consequently,we formulated the hypothesis that ambientlevels of UV-B radiation contribute to am-phibian egg mortality in nature. We pro-posed this for several reasons: 1) egg mor-tality was occurring at relatively high ele-vation sites (> 1,500 m), which under cer-

tain circumstances may receive moreambient UV-B than sites at lower eleva-tions, 2) eggs were usually laid in open,shallow water exposed to solar radiation 3)previous laboratory studies showed thatamphibian embryos were susceptible toUV-B radiation and 4) there were no otherobvious causes of egg mortality at the sitesof oviposition.

As we planned our experiments, ourworking hypothesis was modified and stat-ed as the "UV Sensitivity Hypothesis."One basic prediction of this hypothesis isthat the effects of ambient UV-B radiationon eggs and embryos will vary accordingto their natural exposure to sunlight andtheir ability to repair UV-induced DNAdamage (Blaustein et al., 19946). For ex-ample, the eggs of species that are not laidin sunlight (e.g., under forest canopies, indeep water, under debris) would not bedamaged by UV-B radiation. Furthermore,we predicted that the eggs of those specieswith a relatively high capacity to repair UV-induced DNA damage would be more re-sistant to UV-B than those with a lower ca-pacity.

Field experiments in Oregon andWashington: General methods

Most of our field experiments were con-ducted at natural oviposition sites. Methodsand materials were similar in all studies. Weplaced newly laid eggs, with their surround-ing jelly matrix in enclosures (38 X 38 X7 cm). Eggs were exposed to ambient levelsof UV-B radiation or shielded from UV-Bradiation with filters (50 X 50 X 7 cm). Allenclosures had clear Plexiglas frames withfloors of 1 mm2 fiberglass mesh screen. Inall experiment but one (Blaustein et al.,1997a; see below), we used the followingtreatments: 1) a UV-B blocking filter madeof mylar was placed over one third of theenclosures; 2) an acetate filter that trans-mitted UV-B was placed over another thirdof the enclosures (a control for placing amylar filter on some enclosures); 3) the re-maining enclosures had no filters. Trans-mitting properties of the mylar and acetateused on enclosures were assessed beforeand after experiments.

Enclosures were always placed in a lin-


Dow

nloaded from


EFFECTS OF UV ON AMPHIBIANS 801

ear array in a randomized block design,with the three treatments randomly assignedto enclosures within each block (e.g., Blau-stein et al, 1994ft, 1995). Each treatmentwas replicated four times. Experimentswere terminated when the original embryoseither hatched or died, as assessed by dailycount of embryos. All embryos were ac-counted for at the end of each experiment;there was no predation. Survival was mea-sured as the proportion of hatchlings pro-duced per enclosure. Temperatures weretaken within enclosures in each treatment.

Pacific treefrog, Cascades frog, andwestern toads in Oregon

Our initial field experiments were de-signed to assess the effects of ambient lev-els of UV-B radiation on embryos of thePacific treefrog (H. regilla), western toad(B. boreas), and Cascades frog (/?. casca-dae) as they developed at natural oviposi-tion sites in the Cascade Range of Oregon(Blaustein et al, 1994ft). We chose thesespecies because their eggs are laid in open,shallow water and are highly exposed tosunlight (Blaustein et al., 1994ft; Nussbaumet al., 1983). Moreover, populations of B.boreas and R. cascadae are in decline,whereas those of H. regilla appear to berobust (Blaustein et al., 1994ft). Further-more, unusual embryo mortality has beendocumented in both B. boreas and R. cas-cadae but not in H. regilla (Blaustein andOlson, 1991; Kiesecker and Blaustein,1997). Therefore, differences in sensitivityto UV-B may help us explain differences inthe population status of these species.

Experiments were conducted at severalsites from 1,220-2,000 m elevation inspring 1993 (Blaustein et al., 1994ft). Toensure that effects were not due to site dif-ferences, we tested each species at two dif-ferent sites, including one site where allthree species were found together. Weplaced 150 newly deposited eggs (<24 hrold) in each enclosure. Enclosures wereplaced parallel to the water's edge at depthsof 5-10 cm, where eggs are laid naturally(Blaustein et al., 1994ft).

There were no differences in H. regillahatching success among the three treatmentregimes. Thus, H. regilla embryos were

highly resistant to ambient levels of UV-Bradiation (Table 1). The hatching success ofB. boreas and R. cascadae, however, wassignificantly greater under sunlight lackingUV-B than under either unfiltered sunlightor filtered controls (Table 1). Therefore, weconcluded that embryos of B. boreas andR. cascadae in Oregon were susceptible todeleterious effects of UV-B radiation.

Red-legged frogs, and Northwestern andlong-toed salamanders in Oregon

We conducted field experiments similarto those described above to investigate theeffects of UV-B radiation on Northwesternsalamander (Ambystoma gracile) and red-legged frog (R. aurora) embryos (Blausteinet al, 1995, 1996). Experiments on red-leg-ged frog embryos were conducted in theWillamette Valley, Oregon (76 m elev.)where they were formerly abundant but arenow rare (Nussbaum et al, 1983). Studiesof northwestern salamanders were conduct-ed at 183 m elev. in the Coast Range ofOregon. Long-toed salamanders (A. macro-dactylum) were studied at 2,000 m elev. ata site in the Oregon Cascade Range.

The red-legged frog has disappearedfrom large portions of its historical range(Nussbaum et al, 1983; Blaustein et al,1996; Kiesecker and Blaustein, 1998). Thepopulation status of the Northwestern sala-mander and the long-toed salamander is un-known. All three species commonly layeggs in shallow water, exposed to solar ra-diation (Blaustein et al, 1995, 1996,1997a). However, there are no reports ofunusual egg mortality in any of these spe-cies.

We used the same methods and materialsdescribed above for R. cascadae and B. bo-reas except that enclosures were placed insmall plastic pools (110 cm diameter, 18 cmdeep). Within the pools, eggs were im-mersed in 5-10 cm of natural pond water,a depth at which eggs are naturally laid.Because there were fewer eggs available,we used only the mylar (UV-B blocking)and acetate (mylar control) regimes whentesting A. macrodactylum. We have neverseen significant differences in hatching suc-cess among the acetate and open regimes.

Hatching success of A. gracile and A.


Dow

nloaded from



TABLE 1. Effects of ultraviolet radiation on amphibians in field experiments.

Species

Rana cascadaeCascades frog

Rana auroraRed-legged frog

Rana auroraRana pretiosa

Oregon Spotted FrogRana luteiventris

Columbia Spotted FrogBufo boreas

Western toadBufo boreas'Bufo bufo

Common toad

Bufo calamitaNatterjack toad

Hyla regillaPacific treefrog

Hyla regillaHyla regillaHyla cadaverina

California treefrogLitoria verreauxii

Alpine treefrogLitoria aurea2

Bell frogLitoria dentata

Keferstein's treefrogLitoria peronii

Peron's treefrogCrinia signifera

Eastern frogletAmbystoma macrodactylum

Long-toed salamanderAmbystoma gracile

Northwestern salamanderTaricha torosa

California newt

uvEffect

Yes

No

NoNo

No

Yes

NoYes

No

No

NoNoYes

Yes

9

No

No

Yes

Yes

Yes

Yes

Location

Oregon

Oregon

Western CanadaWashington

Washington

Oregon

ColoradoSpain

Spain

Oregon

CaliforniaWestern CanadaCalifornia

Australia

Australia

Australia

Australia

Australia

Oregon

Oregon

California

Population status

Declining

Declining

DecliningDeclining

Declining

Declining

DecliningDeclining in

some areasof Europe

Declining insome areasof Europe

Persistent

PersistentPersistentUnknown

Declining

Declining

Persistent

Persistent

Persistent

Unknown

Unknown

Unknown

Investigators

Blaustein et al., \994b

Blaustein et al., 1996

Ovaska et al, 1997Blaustein et al..

unpublishedBlaustein et al.,

unpublishedBlaustein et al., 1994b

Corn, 1998Lizana and Pedraza, 1998

Lizana and Pedraza, 1998;see also Gasc, 1997


Anzalone et al., 1998Ovaska et al., 1997Anzalone et al., 1998

Broomhall et al., 1999

van de Mortel andButtemer, 1996



Broomhall et al., 1999

Blaustein et al., 1997a


Anzalone et al., 1998

1 Goebel (1996) suggests that B. boreas in Oregon and Colorado may belong to different species.2 Possible effects; see text for discussion.

macrodactylum was significantly greaterunder treatments blocking UV-B radiationthan under those in which eggs were ex-posed to UV-B. Furthermore, more than90% of A. macrodactylum embryos weredeformed under the UV-transmitting regime(Blaustein et al, 1991 a). There were nosignificant differences in hatching successamong the three sunlight regimes in tests ofR. aurora.

Spotted frogs in WashingtonEffects of UV-B radiation on embryos of

spotted frogs, Rana pretiosa and R. luteiv-

entris, were studied at three elevations inWashington (unpublished data, A.R.B.).Populations of both species are in decline(McAllister et al., 1993; Leonard et al.,1993). Unusual egg mortality has not beenreported in these species, but detailed in-formation on their reproductive dynamics islacking.

Rana pretiosa was examined near sealevel and at a site about 600 m elev. Ranaluteiventris was examined at about 1,700 melev. Hatching success was not affected byUV-B radiation in any of the spotted frogexperiments.


Dow

nloaded from


EFFECTS OF LTV ON AMPHIBIANS 803

Treefrogs and newts in CaliforniaEffects of UV-B radiation on California

and Pacific treefrog (//. cadaverina and H.regilla) embryos and on embryos of theCalifornia newt (Taricha torosd) were stud-ied at a low elevation site (290 m) in theSanta Monica Mountains, near Los Ange-les, California (Anzalone et al, 1998). Em-bryos were placed in small plastic chambers(25 cm long, 15 cm wide). Both open endswere covered with fiberglass screen to al-low water flow, and chambers were placedin shallow areas of a stream so that the openends were parallel with the current. Embry-os were immersed in 4—8 cm of water underthe surface of the stream. Chambers wereexposed to three treatments: 1) uncoveredchambers allowing unfiltered sunlight 2)chambers with UV-B blocking mylar filtersand 3) chambers with Acrylite UV-B trans-mitting filters.

Solar UV-B radiation significantly re-duced the hatching success of H. cadaver-ina and T. torosa embryos. Hyla regillaembryos were not affected by ambient UV-B radiation.

Toads in ColoradoPopulations of toads (Bufo boreas bo-

reas) have experienced widespread declinesin Colorado (Carey, 1993; Corn, 1998).Corn (1998) studied the effects of UV-B ontoads at two Colorado lakes (2,810 and3,266 m elev.). Between 18 and 37 recentlylaid eggs at the 2-32 cell stage were placedin 8.9 cm diameter polystyrene petri dishesand assigned to one of three treatments: 1)petri dishes with Saran Wrap® over themthat transmitted about 80% of ambient UV-B, 2) petri dishes with covers that trans-mitted about 50% ambient UV-B, and 3)petri dishes with a cover, a mylar disk anda mylar collar around them that blocked allUV-B transmission. The depths at whicheggs were placed varied from 3.4 to 9.8 cm.These depths also varied between sites andwithin sites. There were no differences inhatching success among regimes at eithersite.

Treefrogs and red-legged frogs in WesternCanada

Experiments were conducted at 90 melev. in a forest clearing in British Colum-

bia, Canada, to examine the effects of UV-B radiation on H. regilla and R. aurora(Ovaska et al, 1997). Thirty-six 12-literplastic buckets were buried up to their rimsin a 6 X 6 grid within a 4 X 4 meter area.Plastic tubs (24 X 24 cm, 10 cm height)were fitted into the upper portion of eachbucket and filled with pond water. Approx-imately equal numbers of newly laid eggswere placed in each bucket. In one treat-ment, an acrylic filter blocked UV wave-lengths <450 nm. In another treatment, anacrylic filter allowed transmission of all UVwavelengths. A third treatment had no fil-ters and was open to solar radiation. By us-ing fluorescent lamps, two UV enhance-ment regimes were also employed: lampwith a filter, which provided a 15% increaseover noon ambient UV; and lamp without afilter which provided a 30% increase.

Hatching success of H. regilla did notdiffer among regimes. Hatching success ofR. aurora was similar in the treatmentswhere UV was blocked and those whereeggs were exposed to ambient levels. How-ever, it was lower in the UV-enhanced treat-ments when data from both enhanced treat-ments were combined.

These results are similar to those of Blau-stein et al. (1994&) and Anzalone et al.(1998) who showed that hatching successof H. regilla was unaffected by UV-B inOregon and California, respectively. More-over, as in Blaustein et al. (1996), hatchingsuccess of R. aurora was unaffected by am-bient UV-B radiation.

Ovaska et al. (1997) also observed larvaeof H. regilla and R. aurora for two monthspost-hatching under the UV regimes de-scribed above and found that survival wassimilar in no UV and ambient UV regimes.However, larval survival was lower in en-hanced regimes.

Frogs in AustraliaPopulations of several species of frogs

have disappeared in Australia (Richards etal., 1993). At least two studies have usedfield experiments to investigate the possiblerole of UV-B radiation in these declines.

van de Mortel and Buttemer (1996) stud-ied the effects of UV-B on Litoria aurea,L. dentata and L. peronii at sites in the II-


Dow

nloaded from



lawarra and Southern Highlands of NewSouth Wales. Populations of L. aurea seemto be in decline while populations of theother two species appear to be stable (vande Mortel and Buttemer, 1996).

An aluminum raft was constructed ateach site. Each raft had nine Perspex boxes(40 cm wide X 40 cm long X 7 cm deep)with 1 mm2 fiberglass mesh bases. The box-es were placed on the rafts in 2 cm of water.A muslin insert was placed over the meshto prevent egg and tadpole losses. Freshlylaid eggs (150 per box) were subjected tothree UV treatments: 1) unfiltered sunlight,2) a mylar sunlight filter to remove UV-B,and 3) a polyethylene control filter that al-lowed 80% UV-B transmission.

In one experiment, the hatching successof L. aurea was significantly higher in re-gimes where UV-B was blocked. However,in other experiments with this species andthe other species there were no differencesin hatching success among treatments.

Using similar methods, Broomhall et al.(1999) studied the effects of UV-B on thehatching success of Crineria signifera andLitoria verreauxii alpina at three differentsites (1,365 m, 1,600 m and 1,930 m elev.)in the Snowy Mountains of southeast Aus-tralia. Populations of L. v. alpina are in de-cline while those of C. signifera do not ap-pear to be declining (Broomhall et al.,1999). However, high numbers of dead C.signifera eggs have been observed in shal-low alpine ponds (Broomhall et al., 1999).

Tanks containing six Perspex trays (41cm long X 41 cm wide X 9 cm deep) witha 1 mm2 mesh base were supported on analuminum floating raft. A layer of curtainmesh lining was attached to all four rims ofeach tray to retain larvae. Trays were im-mersed in 6—9 cm of water. Three treat-ments were used: 1) unfiltered sunlight, 2)sunlight with mylar filters that removedUV-B, and 3) polyethylene filters that trans-mitted UV-B. The numbers of embryoshatching and larvae reaching metamorpho-sis were recorded for four weeks.

The exclusion of UV-B radiation signif-icantly enhanced the survival of both spe-cies at all sites. The probability of dyingwas considerably higher in L. v. alpina, a

species whose populations are in decline,than in C. signifera, a nondeclining species.

Toads in SpainField experiments investigating the ef-

fects of UV-B radiation on the hatchingsuccess of two toad species, Bufo bufo andBufo calamita, were conducted at 1,920 melev. in the Sierra de Gredos, Central Sys-tem of Spain (Lizana and Pedraza, 1998).Populations of both species have declinedin various parts of Europe (Blaustein et al.,1994a; Gasc, 1997).

One hundred fifty newly laid eggs (<24hr old) were placed in each of 12 plasticenclosures (27 X 27 X 14 cm) that wereperforated to allow water and air circula-tion. Enclosures were placed under 1) Liu-mar Window Film stuck to a PVC plasticsheet that blocked UV-B radiation, 2) PVCplastic that blocked 80-85% UV-B, or 3) 2-cm plastic mesh netting that permittedtransmission of UV-B radiation. Eggs wereplaced in 15 cm of water.

There were significant differences inhatching success of Bufo bufo eggs amongthe three regimes. Survival was about 16%in open enclosures, 49% in enclosures thatpartially blocked UV-B, and 78% in enclo-sures that totally blocked UV-B radiation.There were no differences among treat-ments in Bufo calamita.

SYNERGISTIC EFFECTSIn nature, more than one environmental

agent may affect amphibians as they devel-op. Field experiments have been used to ex-amine at least three factors that may interactsynergistically with UV-B: a pathogenicalga (Saprolegnia ferax), low pH, and fluor-anthene, a polycyclic aromatic hydrocarbonthat may pollute aquatic environments im-pacted by petroleum contamination.

UV/AlgaThe alga Saprolegnia ferax has contrib-

uted to mortality in several amphibian spe-cies inhabiting the Pacific Northwest (Blau-stein et al., 1994c; Kiesecker and Blaustein,1997). In a series of field experiments sim-ilar to those described above, Kiesecker andBlaustein (1995) tested the hypothesis thatthere is a synergism between UV-B radia-


Dow

nloaded from



tion and Saprolegnia. This hypothesiswould be supported if the effects of bothUV-B radiation and Saprolegnia togetherwere greater than those of either factoralone. Tests were conducted on eggs of R.cascadae, B. boreas, and H. regilla.

Experiments were conducted at three nat-ural oviposition sites in the central OregonCascade Range. Each species was tested attwo sites including one site where all threespecies were tested (from 1,190-2,000 melev.). Enclosures identical to those de-scribed in Blaustein et al. (19946) wereplaced in small plastic pools so that thepresence or absence of Saprolegnia couldbe controlled. Within the pools eggs wereimmersed in 5—10 cm of natural lake water.Pools with enclosures were placed in a lin-ear array parallel to the water's edge in a 2X 3 randomized block design, with threesunlight treatments crossed with two algaltreatments (four replicates per treatment;150 eggs per enclosure) at each site.

Saprolegnia was cultured in the labora-tory and randomly added to half of the en-closures in each sunlight treatment. An an-tialgal agent was added to the remainder ofthe enclosures to remove any naturally oc-curring Saprolegnia. A second experimentwas conducted simultaneously with the firstone using identical techniques except thatenclosures were placed directly into thelake. Thus, in the second experiment, em-bryos were exposed to all three sunlight re-gimes and natural levels of Saprolegnia.

In the first experiment, all three specieshad reduced hatching success in the pres-ence of Saprolegnia. In addition, when ex-posed to both Saprolegnia and UV-B, em-bryos of Bufo and Rana experienced sig-nificantly higher mortality than when theywere exposed to either factor alone. Hatch-ing success of Hyla regilla was not affectedby UV-B radiation but it was affected in thepresence of Saprolegnia.

In the second experiment, hatching suc-cess of B. boreas and R. cascadae was alsogreater in treatments shielded from UV-Bradiation. Hatching success of H. regilladid not differ among the regimes.

UV/pHSeveral studies have shown that habitat

acidification can have adverse effects on

amphibians (e.g., Pierce, 1985; Dunson etal., 1992; Kiesecker, 1996). In regionswhere acid pollution is a concern, theremay be synergistic effects between low pHand other factors such as UV-B radiation.Long et al. (1995) investigated the potentialeffects of such a synergism on amphibianembryos. They studied eggs of the leopardfrog (Rana pipiens) exposed to combina-tions of three levels of UV-B and three lev-els of pH.

Experiments were conducted outdoors(250 m elev.) but natural UV-B was sup-plemented with artificial sources. One re-gime had no UV-B levels. Another regimehad UV-B levels simulated for high eleva-tions (intermediate UV-B level) and a thirdregime had UV-B levels simulated to ap-proach levels predicted with a 30% loss inozone (high UV-B level). These UV levelswere crossed with pH levels of 4.5, 5, and 6.

With low pH and simulated intermediateand high UV-B levels, hatching success wassignificantly lower compared to the otherregimes. There was no increase in mortalitywith either UV-B or pH alone. This studydemonstrates the potential effects of a syn-ergism between UV-B radiation and lowpH.

UVIFluorantheneHatch and Burton 1998 investigated the

potential synergistic effects of UV andfluoranthene on two species of frogs, Xen-opus laevis and Rana pipiens, and the spot-ted salamander, Ambystoma maculatum.Experiments were conducted outdoors inexperimental chambers on the roof of abuilding in Ohio. In one experiment, theymonitored the hatching success of R. pi-piens embryos under full spectrum sunlightand under approximately 6% sunlight. In asecond experiment using the same materialsthey monitored the survival of X. laevis andA. maculatum larvae under three light re-gimes crossed with three concentrations offluoranthene. The light regimes were fullsunlight, a regime that transmitted about40% solar radiation (by covering the bea-kers with cheesecloth), and a regime thattransmitted about 4-12% solar radiation.The lowest light regime in all experimentswas obtained by coating the beakers with


Dow

nloaded from



Coppertone® Sport SPF 30 sun block andcovering them with Saran Wrap® and blackplastic. In all experiments, eggs or larvaewere placed in 250-ml beakers (10 individ-uals per beaker; two replicates per treat-ment). Water depth was 5 cm. A random-ized block design was used in all experi-ments and beakers were placed in a tub fullof water to control for evaporation andoverheating.

There were no effects on hatching suc-cess in any treatment regime. However, un-der the highest fluoranthene doses and fullsolar radiation, mortality rates of larvaewere significantly increased in X. laevis andA. maculatum in a manner that was corre-lated with increasing UV intensity and in-creasing fluoranthene concentration.

SUMMARY OF FIELD EXPERIMENT RESULTS

Results of field experiments by severaldifferent investigations strongly indicatethat the hatching success of at least ninespecies of amphibians, from widely sepa-rated locales, is reduced under ambient UV-B radiation (Table 1). This includes twofrog species, one toad species, two sala-mander species and a new from NorthAmerica, two frog species from Australia,and a species of toad from Europe. Thesespecies comprise a diverse group taxonom-ically that includes two orders, six familiesand seven genera of amphibians. Some ofthese species are found in montane areas,others at sea level. A key characteristicshared by these species is that they oftenlay their eggs in shallow water, where theyare exposed to solar radiation.

Hatching success of several other speciesof frogs in North America and Australia,and toads in Europe and Colorado, were notaffected by UV-B radiation (Table 1). Thisis not surprising because many studies havedemonstrated differential sensitivity of am-phibians to various abiotic factors {e.g.,Blaustein, 1994).

Results of several experiments examin-ing synergistic interactions suggest thatthere is a potential for UV radiation to actsynergistically with a wide variety ofagents.

DISCUSSION

The power of field experiments in ecol-ogy has been discussed extensively {e.g.,see discussions in Hairston, 1989; Jaegerand Walls, 1989; Wilbur, 1989; Real andBrown, 1991; Walls 1991; Ramsey andShafer, 1997). It is apparent, however, thatthere is some misunderstanding regardingthe design, power and interpretation of theresults of field experiments.

Where should field experiments beconducted?

We believe that field experiments inves-tigating particular causes for amphibianpopulation declines, egg mortality and re-lated phenomena should be done in the pre-cise area, and if possible, microhabitatwhere these phenomena are occurring. Insome cases, field experiments may be con-ducted in areas where a species was for-merly abundant but where it is now rare orperhaps extinct. Most of the studies de-scribed above were performed in areaswhere the animals live or where they werehistorically found. For example, Blausteinet al. (1994b) erected enclosures among eggmasses that were laid communally to ex-amine egg mortality in frogs and toads inthe Pacific Northwest. Anzalone et al.(1998) conducted their experiments in areasof streams where the amphibians they ex-amined naturally lay their eggs. Interpretingthe results of an experiment where a speciesis brought into an area far from its nativesite is difficult and should be done withcaution.

What varies in a field experiment?Field experiments are designed so that all

factors vary naturally and simultaneouslybetween experimental and control treat-ments, except for the variable(s) of interest(see Hairston, 1989 and papers cited there-in). Thus, in a randomized block design,{e.g., Blaustein et al., 1994Z?; 1997a, b)treatment combinations are assigned ran-domly to subjects within a block and sep-arate randomizations are conducted for eachblock (Ramsey and Shafer, 1997). This con-cept seems to be insufficiently understoodby some investigators. For example, Licht(1996) erroneously claimed that in earlier


Dow

nloaded from



UV/amphibian experiments investigators"controlfed] all variables but UV-B trans-mission." Actually, UV-B transmission isthe only variable that is strictly controlled.All other factors vary naturally.

Variability of abiotic and biotic factors iswhat makes a field experiment ecologicallyrelevant. For example, it may be cloudy oneday and sunny the next. But if the experi-ment is designed appropriately, all enclo-sures are subjected to the same amount ofsunny and cloudy days. The inherent vari-ability of field experiments may seem un-intuitive or unacceptable to laboratory sci-entists who strictly control all parameters inan artificial setting. A good experimentaldesign contains adequate replication (tocapture a range of the variability present),controls treatments (to account for the ef-fects of any manipulations), and often usesblocking (to find environmental gradientswithin the study area).

Controls

Obviously, adequate controls are neces-sary in a field experiment. Controls shouldbe treated the same as experimental units.If enclosures are used, the control and ex-perimental enclosures should be the samedimensions and made of the same material.If shields are placed over experimental en-closures such as those used in many of theUV-B experiments described above, shieldsshould be placed over control enclosuresbecause there may be a "shield effect."

Replicates

"Replication means applying exactly thesame treatment to more than one experi-mental unit" (Ramsey and Shafer, 1997).True replicates have to be conducted withthe same methods on the same species andin the same experiment. Employing an ad-equate number of replicates for each treat-ment will help to ensure that results are notunique to a particular series of treatmentsor blocks. In addition, we suggest, thatthere be replicate sites whenever possible.It is possible that the results obtained fromone site are unique to that site. Some of theUV-B studies described above do not usemore than one site. Site replicates enhance

the significance and generality of the re-sults.

Comparisons between studiesEven if the exact same methods and ma-

terials are used it is difficult to compare theresults of studies that are conducted in dif-ferent systems. It is even more difficult tocompare studies that use different methods,conduct tests at different times, and exam-ine different species. This is especially truefor studies examining the effects of UV-Bradiation on amphibians. The amount ofUV-B radiation exposure varies with manyfactors, including cloud cover, precipitation,altitude, latitude, and local pollution events.Aquatic habitats differ in turbidity, amountof dissolved organic carbons, temperature,animals, plants, microorganisms and manyother factors. These differences may resultin differences in how UV radiation affectsamphibians between habitats. Thus, it is notsurprising that Corn's (1998) results on theeffects of UV-B on B. boreas differed fromours. Our group {e.g., Blaustein et al.,1994&) used relatively large open enclo-sures with mylar and acetate niters andCorn (1998) used small petri dishes, somecovered with Saran Wrap®. The Coloradosystem differs dramatically from Oregon inseveral abiotic and biotic parameters, aswell as atmospheric conditions. There alsomay be geographic variation in the effectsof UV-B in such a wide-ranging species.Indeed, based on molecular-systematic data,toads examined in Colorado may constitutea different species from B. boreas inOregon (Goebel, 1996). Comparisons be-tween studies are difficult to make and theycertainly should not be considered repli-cates of one another. We suggest greatercaution when comparing results from dif-ferent studies.

Will measuring levels of UV-B radiationenhance the significance of experimentalresults?

Measurements of UV-B radiation are notnecessary to document a detrimental effectof radiation on amphibian development. Ina well-designed experiment, if hatchingsuccess is lower in enclosures where eggsare exposed to UV-B than in enclosures


Dow

nloaded from



where they are shielded from UV-B it isobvious that UV-B radiation has affectedhatching success.

Measuring UV-B levels under shields ob-viously should show that levels underblocking regimes are less than levels undertransmitting regimes. If done preciselyenough, UV-B measurements may revealthe minimum dose that will cause egg mor-tality. Nevertheless, UV-B levels changeconstantly. They change with weather pat-terns, cloud cover, water flow, and turbidity.Due to temporal changes, UV-B levels mustbe monitored constantly for a meaningfulinterpretation of their effects on biologicalsystems. Because there are no long-termdata on levels of UV-B radiation from siteswhere amphibian tests are being conducted,levels measured at present cannot be com-pared with any historical data. However, ac-cumulating long-term data on terrestial UVradiation will allow us to have a baselinefor comparing current and future levels.

Carey et al. (1996) measured UV-B ir-radiance at midday in air and water atbreeding sites of three species of amphibi-ans in Colorado. They also conducted lab-oratory experiments in which amphibianeggs and larvae were exposed to variousdoses of UV-B radiation. Salamanders ex-perienced significant mortality at relativelylow levels of simulated UV-B radiation.Eggs and larvae of two toad species weremore tolerant to simulated UV-B radiationthan were salamanders. Based on these lab-oratory tolerance limits and field measure-ments, they concluded that field levels ofexposure were not sufficient to be a caus-ative factor in the decline of the species ofamphibians examined.

We regard this conclusion as premature.No field experiments were conducted. Un-der field conditions there are many factorsthat vary and which could potentially influ-ence how UV varies and how it impactsamphibians. Biotic parameters also vary.Performing an experiment in a dynamicnatural system such as a lake, pond orstream will be the most rigorous test of theeffects of ambient UV-B on amphibians.Without such a test, UV-B measurementswill not provide much information on UV-B tolerance levels of amphibians in nature.

Corn (1998) used remote satellite data inan attempt to estimate UV levels in Oregonand Colorado. However, in the absence ofcorroborative ground measurements, satel-lite estimates may not be very useful. More-over, even if satellite data showed the samelevels of UV-B at the precise microhabitatswhere experiments in Colorado and Oregonwere conducted (which was not possible),this would not influence the conclusions ofthe studies conducted in those regions.

Natural selectionWhy are the eggs of some species of am-

phibians affected by UV-B whereas theeggs of others are not? One possible reasonis that, over evolutionary time, there werestrong selection pressures for certain spe-cies to evolve mechanisms that counteract-ed the harmful effects of UV-B. When UVradiation penetrates a cell, photoproductsmay form that can lead to mutations or celldeath. Embryos of some amphibian speciesmay be more resistant to UV-B becausethey can repair UV damage more efficientlythan others. One important repair process isenzymatic photoreactivation. A single en-zyme, photolyase, uses visible light energyto remove the most frequent UV-induced le-sion in DNA, cyclobutane pyrimidine di-mers (CPDs) (Friedberg et al., 1995).Moreover, multi-protein broad specificityexcision repair processes can remove CPDsand other UV photoproducts. Both mecha-nisms may be used simultaneously. How-ever, photoreactivation appears to be thefirst level of defense for many organismsexposed to sunlight (Pang and Hays, 1991;Friedberg et al., 1995).

We suggest that greater photolyase activ-ity makes some amphibian species more re-sistant to UV-B radiation than others, by re-moving more of the primary cytotoxic/mu-tagenic photoproducts from their DNA. Ifeggs are damaged by UV-B in field exper-iments, obviously none of the repair mech-anisms is working efficiently enough to re-pair the damage. However, if eggs are re-sistant, it is difficult to determine whetherexcision repair, photolyase or both are re-pairing damage. Because photolyase repairis probably the most important repair mech-anism in amphibians, a parsimonious expla-


Dow

nloaded from



TABLE 2. Photolyase levels in North American amphibian eggs fromexperiments examining the effects of UV on hatching success have been

the Pacific Northwest for which fieldconducted.

SpeciesActivity of photolyase 10"

CPDs* per hr per |ig (±SE)

Hyla regillaPacific treefrog

Rana luteiventrisColumbia spotted frog

Rana pretiosaOregon spotted frog

Rana auroraRed-legged frog

Rana cascadaeCascades frog

Bufo boreasWestern toad

Ambystoma gracileNorthwestern salamander

Ambystoma macrodactylumLong-toed salamander

7.5 (0.35)

6.84 (0.3)

6.62 (0.01)

6.09 (0.22)

2.4 (0.23)

1.3 (0.08)

1.0 (0.10)

0.8 (0.72)

Blaustein et at., 19946

Blaustein and Hays,unpublished

Blaustein and Hays,unpublished





Blaustein et al, 19946

' Cyclobutane pyrimidine dimers.

nation is that those species with the highestphotolyase activity are the most resistant toUV damage. Indeed, the amount of photo-lyase in eggs is positively correlated withsurvival of embryos in field experiments(Table 2). Thus, eggs of the most resistantspecies in our field experiments (H. regilla,R. aurora, R. pretiosa, and R. luteiventris)had much higher levels of photolyase activ-ity than eggs of the more susceptible spe-cies (R. cascadae, B. boreas, A. macrodac-tylum, A. gracile).

Behavioral and ecological attributes alsomay limit UV-B damage to amphibians. Forexample, many salamander species (andsome frogs and toads) lay their eggs underleaf litter, in crevices or under logs wherethey are not subjected to high levels of sun-light. Many amphibians lay their eggs inmuddy water where light penetration maynot be significant to damage eggs. Thus, inthe clear, high mountain lakes in Oregon,where we have conducted our research, thepotential for UV penetration is greater thanin relatively turbid sea-level ponds wheremany eastern and some western species laytheir eggs (Stebbins, 1954; Nussbaum et al.,1983).

Even if eggs are laid in the open at highaltitudes (where under certain conditionsUV levels may be high), take very long todevelop and are subjected to intense levelsof UV-B radiation, they may not be affected

by UV-B radiation if they have efficientDNA repair mechanisms. Conversely, spe-cies at sea level (where UV levels may below) with very low photolyase levels maybe quite sensitive to UV-B radiation as wehave shown (Blaustein et al., 1995, 1996).

Within a species it is possible that indi-viduals from one population may differfrom members of another population intheir sensitivity to UV-B radiation. Thismay be due to differences in their ability torepair DNA damage or differences in indi-vidual egg-laying behavior. Thus, the eggsof one species may differ between popula-tions in their exposure to solar radiation.For example, long toed-salamanders (A.macrodactylum) deposit their eggs in smalldiscrete clutches in shallow temporaryponds in the Willamette Valley of Oregon.However, in the Cascade Range of Oregonthey scatter eggs singly with some beinglaid in open shallow water and others indeeper water under rocks in temporary orpermanent ponds (personal observations).

Why is there mortality under UV-Bshields?

Hatching success in nature is usually lessthan 100% (Duellman and Trueb, 1986).Thus, in field experiments, there was somemortality even under UV-B shields (e.g.,Blaustein et al., 1994&; Anzalone et al.,1998; Lizana and Pedraza, 1998). Mortality


Dow

nloaded from



may occur because of species differences inhatching success, individual genetic vari-ability, pathogens, other biotic factors andnumerous abiotic factors including UV-A(315—400 nm) radiation. Importantly, how-ever, in several species described in this re-view, hatching success was greater underUV-B shielded regimes compared with re-gimes where embryos were exposed toUV-B radiation. Thus, UV-B was a signif-icant source of mortality in several species(Table 1).

How do we interpret negative results?It is difficult to interpret negative data.

Thus, if field experiments fail to find a "UVeffect," one must be cautious about inter-preting the results because it is not possibleto prove that something did not occur. Asdiscussed above, several studies have failedto find an effect of ambient UV-B radiationon the eggs of some amphibian species asassessed by hatching success. It is possible,however, that UV-B radiation did affect theembryos but the effects may not becomemanifest until later life-history stages {e.g.,larvae, metamorphs, adults). It is possiblethat the methods, materials and measure-ments used in certain studies were not sen-sitive enough to detect detrimental effectsof UV-B on eggs. It is also possible that theeggs found at some sites are UV-resistantremnants of a population whose other eggswere less resistant and that perished.

Studies of the same species at differentsites add some assurance that negative re-sults are real. For example, the effects ofambient UV-B radiation on H. regilla wereexamined at three sites in Oregon, and onesite each in Canada and California. It is dif-ficult to compare the results of different in-vestigations. However, based on the numberof locales used and the consistency of theresults, we believe that it is most appropri-ate to conclude that hatching success of H.regilla is unaffected by UV-B radiationacross the species' range.

The role of UV-B radiation in populationdeclines

UV-B radiation is affecting the eggs andembryos of several species of amphibiansin widely scattered locales around the world

(Table 1). Continued mortality at the eggstage may eventually lead to a populationdecline. However, the effects of mortality atthe egg or embryo stage may not be ob-served at the population level for manyyears, especially in relatively long-livedspecies (Blaustein et ai, 1994a). For ex-ample, well documented and continuingmortality at the egg stage in western toads{B. boreas) in Oregon {e.g., Kiesecker andBlaustein, 1997) may eventually lead to apopulation decline in adults. However, adultB. boreas may live for at least 16 years (un-published data, A.R.B.) and the effects ofmortality in eggs and embryos may not beobserved in adult populations until adultseventually disappear and recruitment is in-sufficient to maintain the population. More-over, if a population has been recruitingpoorly for several years it may reboundquickly with one good bout of recruitment.Populations of those species whose eggsand embryos are susceptible to UV-B ra-diation are at risk, but the extent of this riskis unknown.

UV-B radiation obviously is not the onlyagent that may contribute to a populationdecline. It would be an unlikely factor inthe declines of species that lay their eggsunder logs, in crevices, in deep water orunder dense forest canopy. As we haveshown in this review, several species of am-phibians are unaffected by UV-B in earlylife stages but still are in decline. Neverthe-less, the hatching success of many amphib-ian species is affected by UV-B radiation.Several factors, such as pathogens, low pH,and pesticides may act synergistically withUV-B radiation to enhance mortality in ear-ly life stages. These synergistic interactionsmay eventually contribute to a populationdecline.

The available evidence suggests thatmore than one agent is contributing to am-phibian population declines. A number ofinvestigators have employed field experi-ments to examine the effects of UV-B ra-diation on amphibians. UV-B radiation isone agent that is significantly hamperingthe hatching success of some amphibianspecies at the egg and embryo stage. We donot have enough information on how UV-induced mortality in early life stages is af-


Dow

nloaded from



fecting adult amphibian populations. How-ever, we suggest that field experiments arean excellent method for examining agents,such as UV, that may be involved in am-phibian population declines.

ACKNOWLEDGMENTS

We thank Jim Hanken for inviting us towrite this review. Our colleagues John Haysand Pete Hoffman have been collaboratorson the DNA repair portion of our research.We thank Nareny Sengsavanh for help withpreparation of the manuscript and Mo Don-nelly, Jim Hanken and an anonymous re-viewer for critical comments that greatlyimproved the paper. Special thanks to SamLoomis, Laurie Strode, Annie Wallace, andMichael Myers for their help. Our researchwas supported by NSF grant DEB-9423333.

REFERENCES

Anzalone, C. R., L. B. Kats, and M. S. Gordon. 1998.Effects of solar UV-B radiation on embryonic de-velopment in three species of lower latitude andlower elevation amphibians. Conserv. Biol. 12:646-653.

Blaustein, A. R. 1994. Chicken Little or Nero's fiddle?A perspective on declining Amphibian popula-tions. Herpetologica 50:85-97.

Blaustein, A. R., B. Edmond, J. M. Kiesecker, J. J.Beatty, and D. G. Hokit. 1995. Ambient ultravi-olet radiation causes mortality in salamander eggs.Ecol. App. 5:740-743.

Blaustein, A. R., P. D. Hoffman, D. G. Hokit, J. M.Kiesecker, S. C. Walls, and J. B. Hays. 19946.UV-repair and resistance to solar UV-B in am-phibian eggs: A link to amphibian declines? Proc.Natl. Acad. Sci. U.S.A. 91:1791-1795.

Blaustein, A. R., P. D. Hoffman, J. M. Kiesecker, andJ. B. Hays. 1996. DNA repair and resistance tosolar UV-B radiation in eggs of the red-leggedfrog. Conserv. Biol. 10:1398-1402.

Blaustein, A. R., D. G. Hokit, R. K. O'Hara, and R.A. Holt. 1994c. Pathogenic fungus contributes toamphibian losses in the Pacific Northwest. Biol.Conserv. 67:251-254.

Blaustein, A. R., J. M. Kiesecker, D. P. Chivers, andR. G. Anthony. 1997a. Ambient UV-B radiationcauses deformities in amphibian embryos. Proc.Natl. Acad. Sci. U.S.A. 94:13735-13737.

Blaustein, A. R., J. M. Kiesecker, P. D. Hoffman, andJ. B. Hays. 19976. The significance of ultraviolet-B radiation to amphibian population declines.Rev. Toxicol. 1:309-327.

Blaustein, A. R. and D. H. Olson. 1991. Declining am-phibians. Science 253:1467.

Blaustein, A. R. and D. B. Wake. 1995. The puzzle of

declining amphibian populations. Sci. American272:52-57.

Blaustein, A. R., D. B. Wake, and W. P. Sousa. 1994a.Amphibian declines: Judging stability, persis-tence, and susceptibility of populations to localand global extinctions. Conserv. Biol. 8:60-71.

Broomhall, S. D., W. Osborne, and R. Cunningham.1999. Comparative effects of ambient ultraviolet-B (UV-B) radiation on two sympatric species ofAustralian frogs. Conserv. Biol. (In Press)

Carey, C. 1993. Hypothesis concerning the causes ofdisappearance of boreal toads from the mountainsof Colorado. Conserv. Biol. 7:355-362.

Carey, C, E. Little, Hurtubise, and D. Fabacher. 1996.Field exposures to and tolerances of UVB light inColorado amphibians. Amer. Zool. 36:18A.

Corn, P. S. 1998. Effects of ultraviolet radiation onboreal toads in Colorado. Ecol. Applications 8:18-26.

Crump, M. L., F. R. Hensley, and K. L. Clark. 1992.Apparent decline of the golden toad: Undergroundor extinct? Copeia 1992:413-420.

Drost, C. A. and G. M. Fellers. 1996. Collapse of aregional frog fauna in the Yosemite area of theCalifornia Sierra Nevada, USA. Conserv. Biol.10:414-425.

Duellman, W. E. and L. Trueb. 1986. Biology of am-phibians. McGraw-Hill, New York.

Dunson, W. A., R. L. Wyman, and E. S. Corbett. 1992.A symposium on amphibian declines and habitatacidification. J. Herpetol. 26:349-352.

Ehrlich, P. R. 1997. A world of wounds: Ecologists andthe human dilemma. Ecology Institute, Oldendorf/Luhe, Germany.

Fisher, R. N. and H. B. Shaffer. 1996. The decline ofamphibians in California's Great Central Valley.Conserv. Biol. 10:1387-1397.

Friedberg, E. C, G. C. Walker., and W. Siede. 1995.DNA repair and mutagenesis. ASM Press, Wash-ington.

Gasc, J.-P. (ed.) 1997. Atlas of amphibians and reptilesof Europe. Museum National D'Histoire Naturel-le, Paris.

Goebel, A. M. 1996. Systematics and conservation ofbufomds in North America and in the Bufo boreasspecies group. Unpubl. Ph. D. Diss. University ofColorado, Boulder.

Hairston, N. G., Sr. 1989. Ecological experiments:Purpose, design and execution. Cambridge Univ.Press, Cambridge.

Hairston, N. G., Sr. 1994. Vertebrate zoology: An ex-perimental field approach. Cambridge Univ.Press, Cambridge.

Hatch, A. and G. A. Burton, Jr. 1998. Effects of pho-toinduced toxicity of fluoranthene on amphibianembryos and larvae. Environmental Tox. Chem.(In press)

Jaeger, R. G. and S. C. Walls. 1989. On salamanderguilds and ecological methodology. Herpetologica45:111-119.

Kiesecker, J. M. 1996. pH-mediated predator-prey in-teractions between Ambystoma tigrinum and Pseu-dacris triseriata. Ecol. App. 6:1325—1331.

Kiesecker, J. M. and A. R. Blaustein. 1995. Synergism


Dow

nloaded from



between UV-radiation and a pathogen magnifiesamphibian mortality in nature. Proc. Natl. Acad.Sci. U.S.A. 92:11049-11052.

Kiesecker, J. M. and A. R. Blaustein. 1997. Influencesof egg laying behavior on pathogenic infection ofamphibian eggs. Conserv. Biol. 11:214-220.

Kiesecker, J. M. and A. R. Blaustein. 1998. Experi-mental analysis of the effects of introduced bull-frogs and small mouth bass on native red-leggedfrogs (Rana aurora). Conserv. Biol. 12:776-787.

Krebs, C. J. 1994. Ecology: Experimental analysis ofdistribution and abundance. Harper Collins, NewYork.

Laurance, W. E, K. R. McDonald., and R. Speare.1996. Epidemic disease and the catastrophic de-cline of Australian rain forest frogs. Conserv.Biol. 10:406-413.

Leonard, W. P., H. A. Brown., L. L. C. Jones., K. R.McAllister, and R. M. Storm. 1993. Amphibiansof Washington and Oregon. Seattle Audubon So-ciety, Seattle.

Licht. L. E. 1996. Amphibian decline still a puzzle.BioScience 46:172-173.

Lips, K. R. 1998. Decline of a tropical montane am-phibian fauna. Conserv. Biol. 12:106-117.

Lizana, M. and E. M. Pedraza. 1998. Different mor-tality of toad embryos (Bufo bufo and Bufo cal-amita) caused by UV-B radiation in high moun-tain areas of the Spanish Central System. Conserv.Biol. 12:703-707.

Long, L. E., L. S. Saylor, and M. E. Soule. 1995. ApH/UV-B synergism in amphibians. Conserv.Biol. 9:1301-1303.

McAllister, K. R., W. P. Leonard, and R. M. Storm.1993. Spotted frog (Rana pretiosa) surveys in thePuget Trough of Washington, 1989-1991. North-west Nat. 74:10-15.

van de Mortel, T. E and W. A. Buttemer. 1996. AreLitoria aurea eggs more sensitive to ultraviolet-Bradiation than eggs of sympatric L. peronii or Ldentata? Australian Zool. 30:150-157.

Nussbaum, R. A., E. D. Brodie, Jr., and R. M. Storm.1983. Amphibians & reptiles of the Pacific North-west. Univ. Idaho Press, Moscow.

Ovaska, K., T. M. Davis, and I. N. Flamarique. 1997.Hatching success and larval survival of the frogs

Hyla regilla and Rana aurora under ambient andartificially enhanced solar ultraviolet radiation.Can. J. Zool. 75:1081-1088.

Pang, Q. and J. B. Hays. 1991. UV-inducible and tem-perature-sensitive photoreactivation of cyclobutanepyrimidine dimers in Arabidopsis thaliana. PlantPhysiol. 95:536-543.

Pierce, B. J. 1985. Acid tolerance in amphibians.BioScience 35:239-243.

Pounds, J. A. and M. L. Crump. 1994. Amphibian de-clines and climate disturbance: The case of thegolden toad and the harlequin frog. Conserv. Biol.8:72-85.

Pounds, J. A., M. P. Fogden, J. M. Savage, and G. C.Gorman. 1998. Tests of null models for amphibiandeclines on a tropical mountain. Conserv. Biol.11:1307-1322.

Ramsey, F. L. and D. W. Shafer. 1997. The statisticalsleuth; A course in methods and data analysis.Duxbury Press, New York.

Real, L. A. and J. H. Brown (eds.), 1991. Foundationsof ecology. University of Chicago Press, Chicago.

Reaser, J. K. 1996. The elucidation of amphibian de-clines. Amph. Rept. Conserv. 1:4-9.

Richards, S. J., K. R. McDonald, and R. A. Alford.1993. Declines in populations of Australia's en-demic tropical rainforest frogs. Pacific Conserv.Biol. 1:66-77.

Semb-Johansson, A. 1989. Padden (Bufo bufo)—et ste-barn inorsk zoologi. Fauna 42:174—179.

Stebbins, R. C. 1954. Amphibians and reptiles of west-ern North America. McGraw-Hill Book Company,New York, USA.

Stebbins, R. C. and N. W. Cohen. 1995. A naturalhistory of amphibians. Princeton Univ. Press,Princeton, New Jersey.

Wake, D. B. 1991. Declining amphibian populations.Science 253:860.

Walls, S. C. 1991. Mechanisms of coexistence in aguild of ambystomatid salamanders: The impor-tance of stage-dependent regulation in specieswith a complex life cycle. Ph. D. Diss. Univ.Southwest. Louisiana, Lafayette.

Wilbur, H. M. In defense of tanks. 1989. Herpetologica45:122-123.

Wilson, E. O. (ed.). 1988. Biodiversity. Nat. Acad.Press. Washington, D. C.


Dow

nloaded from


effects of ultraviolet radiation on amphibians: field experiments

Documents