rsosroyalsocietypublishingorg

ResearchCite this article Hopkins GR French SSBrodie Jr ED 2017 Interacting stressors and thepotential for adaptation in a changing worldresponses of populations and individualsR Soc open sci 4 161057httpdxdoiorg101098rsos161057

Received 17 December 2016Accepted 19 May 2017

Subject CategoryBiology (whole organism)

Subject Areasecologyevolutionenvironmental science

Keywordsamphibian newt salinity stresstemperature variation

Author for correspondenceGareth R Hopkinse-mail garethrhopkinsgmailcom

daggerPresent address School of BioSciencesUniversity of Melbourne MelbourneVictoria 3010 Australia

Electronic supplementary material is availableonline at httpsdxdoiorg106084m9figsharec3796144

Interacting stressors andthe potential for adaptationin a changing worldresponses of populationsand individualsGareth R Hopkinsdagger Susannah S French and

Edmund D Brodie JrDepartment of Biology and the Ecology Center Utah State University 5305 Old MainHill Logan UT 84322 USA

GRH 0000-0003-1427-5315

To accurately predict the impact of environmental change itis necessary to assay effects of key interacting stressors onvulnerable organisms and the potential resiliency of theirpopulations Yet for the most part these critical data aremissing We examined the effects of two common abioticstressors predicted to interact with climate change salinityand temperature on the embryonic survival and developmentof a model freshwater vertebrate the rough-skinned newt(Taricha granulosa) from different populations We found thatsalinity and temperature significantly interacted to affectnewt embryonic survival and development with the negativeeffects of salinity most pronounced at temperature extremesWe also found significant variation among and especiallywithin populations with different females varying in theperformance of their eggs at different salinityndashtemperaturecombinations possibly providing the raw material for futurenatural selection Our results highlight the complex nature ofpredicting responses to climate change in space and time andprovide critical data towards that aim

1 IntroductionFew if any environments are truly stable and free of stressorsthat disrupt homeostasis and threaten survival [1] Additionallyit is widely recognized that organisms rarely face a singlestressor in isolation in these environments and there is thepotential for multiple stressors to interact in complex ways (eg[2ndash6]) Climate change arguably increases this potential [278]

2017 The Authors Published by the Royal Society under the terms of the Creative CommonsAttribution License httpcreativecommonsorglicensesby40 which permits unrestricteduse provided the original author and source are credited

on May 20 2018httprsosroyalsocietypublishingorgDownloaded from

2

rsosroyalsocietypublishingorgRSocopensci4161057

To accurately predict the consequences of climate change on populations and species it is critical toassay the responses of sensitive organisms to key environmental stressors and yet this informationis lacking for most organisms [9] Understanding how species populations and individuals persistdevelop reproduce and regulate physiological processes in the face of interacting stressors is thereforea key priority area of research for biologists

Temperature and salinity have long been recognized as two of the most important abiotic factorsaffecting the lives of aquatic organisms and have the potential to interact significantly with climatechange [810ndash12] Temperature and salinity have individually been shown to influence demography[13ndash15] survival growth development reproduction (eg [16ndash19]) mobility [20] and even predatorndashprey interactions [21] of aquatic organisms Further temperature and salinity can interact in critical wayswith one stressor influencing the potency of the other (eg [11162223])

While the interactions of salinity and temperature have been well studied in marine and estuarineorganisms much less work has been conducted on freshwater taxa (but see [24ndash32] examining eachstressor in isolation) This is in spite of the fact that elevated salinity in freshwater habitats due toroad de-icing salt application (eg [3334]) changing land-use patterns (eg [3536]) and coastal salt-water intrusion from rising sea levels (eg [3738]) is of increasing conservation concern [39] Withclimate change-associated alteration of weather patterns global temperatures and sea-level rise thelikelihood of temperature and salinity interacting to influence the survival of organisms in freshwaterenvironments only increases [810ndash12] The interaction of salinity and temperature on metabolism andnitrogen exchange has been investigated in the model freshwater fishes Gambusia affinis and Daniorerio [40] but the majority of studies have been conducted on euryhaline fishes such as Cyprinodonmacularius [41] Menidia beryllina [42] and Fundulus species (principally Fundulus heteroclitus) [43ndash45]By contrast much less work has been conducted on freshwater organisms perceived to be especiallysensitive to salinity stress Owing to their permeable skin and eggs requirement of aquatic systems forreproduction and ectothermic nature amphibians have long been considered ideal sensitive freshwatermodels with which to examine the effects of salinity and temperature (albeit generally with eachstressor presented in isolation) (eg [313246ndash52]) Peabody amp Brodie [53] examined the effects ofboth salinity and temperature on embryonic development and vertebral number of the salamanderAmbystoma maculatum and an early study on the tailed frog Ascaphus truei indicated that salinity andtemperature in combination were much more lethal to adults and larvae than each stressor alone[54] More recently salinity and temperature were found to interact to affect larval survival anddevelopment in two species of European toads Epidalea calamita and Bufo viridis [5556] To the best of ourknowledge these studies represent the only examination of the interaction of salinity and temperaturein amphibians

Intriguingly these studies on European toads [5556] also highlight that different populationsof the same species may vary in their resiliency to stressor combinations which could haveimportant implications for conservation efforts [56] Populations may vary in their resiliency dueto different evolutionary histories of exposure to the stressors [57ndash60] and the importance ofevolutionary processes including local adaptation in this context is beginning to be realized[755565961] While population-level responses give insight to the current mean response of thepopulation individual-level responses (ie within populations) allow insights into the potentialfor populations to adapt to future changes in stressor exposure as intrapopulation variation iscritical for natural selection [9466263] Despite its potential importance for conservation [964] thevariability of responses to multiple stressors among and especially within populations remainsunderstudied [9]

We examined the interacting effects of temperature and salinity stress on the embryonic survivaland development of a North American amphibian the rough-skinned newt (Taricha granulosa SkiltonCaudata Salamandridae figure 1) from populations that varied in their evolutionary history ofexposure to salinity and temperature regimes Rough-skinned newts inhabit a variety of lenticand lotic habitats that naturally vary in both temperature and salinity ([67] GRH and Z MHopkins 2012ndash2014 unpublished data) making them excellent models to examine the interactionof these stressors on freshwater organisms We examined both inter- and intra-population variationin responses to salinity and temperature to investigate the potential for freshwater organisms toadapt to combinations of these stressors We hypothesized that the effects of each stressor wouldbe highly dependent on the levels of the other that newts from different populations would varyin their responses according to their differing evolutionary history with each stressor and that therewould be high individual variation in responses within each population which may allow forfuture adaptation

on May 20 2018httprsosroyalsocietypublishingorgDownloaded from

3

rsosroyalsocietypublishingorgRSocopensci4161057

rough-skinned newt(Taricha granulosa)

life cycle

Figure 1 Life cycle of the rough-skinnednewt (T granulosa) Adults live terrestrially (greenbackground) but return toponds and streams(blue background) to breed every spring Mating occurs in the water followed by the deposition of fertilized eggs on aquatic littoralvegetation Eggs develop and hatch over the next month and gilled aquatic larvae grow and develop over the next several monthsbefore metamorphosing and returning to land Photos by GRH B Gall and A Spence Adapted with permission from [65]

2 Material and methods21 OrganismRough-skinned newts (T granulosa) (figure 1) are caudate amphibians commonly found living in forestsstreams and ponds along the Pacific coast of North America from southern California to Alaska [66]Newts spend most of the year in the forest but return to breed in aquatic systems every spring where theymate in the water and individually lay jelly-covered eggs among littoral vegetation Eggs take severalweeks to months to develop (depending on the temperature) with gilled aquatic larvae hatching andliving in the water for several months before metamorphosing as terrestrial juveniles (figure 1)

22 Populations and field collectionWe collected gravid female rough-skinned newts (11ndash12 per population) by dip-net minnow trap andhand from two locations in Oregon USA 14ndash21 May 2013 a collection of human-made freshwater ponds(Soap Creek ponds Benton Countymdashhereafter lsquoSoap CreekmdashSCrsquo approx 62 km inland from the PacificOcean 44deg40236prime N 123deg16613prime W see [4662] for more information on this habitat) and a coastal tidalstream Hunter Creek (Curry County 42deg23306prime N 124deg25228prime W see [6768] for more details on thishabitat) At Hunter Creek three lsquopopulationsrsquo were sampled newts found in the tidal area of the streamup to 500 m from the Pacific Ocean (hereafter lsquoHunter TidalmdashHTrsquo) newts found approximately 35 kmupstream from the tidal area in a freshwater non-tidal area of the stream (hereafter lsquoHunter FreshmdashHFrsquo)and newts found a further approximately 35 km upstream from that location (75 km from the oceanhereafter lsquoHunter Upper FreshmdashHUFrsquo) While newts occur throughout Hunter Creek for the purposes ofthis paper we are referring to these three locations as lsquopopulationsrsquo as related species of Taricha generallydo not disperse more than 13 km between breeding sites [69]

on May 20 2018httprsosroyalsocietypublishingorgDownloaded from

4

rsosroyalsocietypublishingorgRSocopensci4161057

Populations were chosen to represent potentially different histories of exposure to salt (decreasing

history of exposure travelling inland) as well as possibly different thermal regimes (pond versus stream)Salinity and temperature in all locations were measured using a handheld YSI EC300 multimeter (YSIYellow Springs OH USA) on multiple days and locations within the habitats and at both high andlow tide (for HT) Water temperature varied considerably at all sites depending on weather shadeand depth and ranged from 8 to 21degC at Hunter Creek (mean temperature at HT = 157degC HF = 14degCHUF = 136degC) and 12 to 23degC at the Soap Creek Ponds (mean = 166degC) during the newt breeding season(May) Salinity was consistently 00ndash01 ppt in all locations with the exception of HT which experiencedintermittently elevated salinity from the ocean Salinity at HT varied depending on depth distance fromthe ocean tides and microhabitat and ranged between 01 and 14 ppt While we did not measure highlyelevated salinity in HT this habitat is subjected to occasional storm events that may raise salinities tolevels much greater than those measured (see [6768] for more details on this habitat) By contrast newtsat the inland Soap Creek ponds are never subjected to increased salinity as they are located over 300 mfrom the nearest minor road which is not treated with road de-icing salts [46]

23 Female newt husbandry and ovipositionFemale newts were transported back to the laboratory and housed individually in non-leachingplastic containers (35 times 20 times 13 cm) with 20 l of aged chilled tap water filtered to remove chlorine(salinity = 02 ppt) and a styrofoam perch in a temperature-controlled room at 14degC with a 12 L 12 Dcycle After 3 days of acclimation newts were injected with 10 microl luteinizing hormone-releasing hormone([des-Gly10 D-His(Bzl)6] LH-RH ethylamide Sigma No L2761 Sigma Aldrich St Louis MO USA) toinduce oviposition perches were removed and replaced with submerged polyester fibre to serve asan oviposition substrate Newts began laying eggs within 24 h Individual eggs were removed fromthe substrate and randomly assigned to a temperaturendashsalinity treatment within 24 h of oviposition

24 Salinity and temperature treatmentsSalinity treatment solutions were made by mixing laboratory grade NaCl (Mallinckrodt Baker Inc ParisUSA) and filtered tap water to a concentration of either 20 or 50 ppt These salinities were chosen asconcentrations that could be found in roadside environments impacted by road de-icing salts [70] aswell as estuarine environments [71] that can seriously affect newt survival development and physiology[4667] The filtered tap water (02 ppt) was used as a freshwater control Temperature-controlled roomswere set to 7 14 or 21degC representative of the range of temperatures that newts may experience duringthe breeding season in the wild and which are known to significantly influence embryonic and larvalgrowth and development in this species [52]

25 Experimental procedureWithin 24 h of oviposition eggs (mean plusmn sd egg diameter = 267 plusmn 027 mm N = 481) were removedfrom their motherrsquos container and placed singly in 35 cm diameter 1 cm deep lidded round plasticPetri dishes with 4 ml of randomly assigned salt or control solution This Petri dish was then randomlyassigned to one of the three temperatures This procedure was continued until all females from allpopulations had approximately 10 eggs (replicates) assigned to each temperaturendashsalt combinationEggs were reared based on the methods of Hopkins et al [46] and were checked daily for mortalityor hatching Water levels in Petri dishes were checked daily for evaporation and replaced with smallamounts of distilled water to maintain constant salinity Upon hatching the total length of larvae wasmeasured using an ocular micrometer attached to an Olympus stereomicroscope and developmentalstage [72] determined Rough-skinned newts typically hatch at Harrison stages 39ndash42 depending onenvironmental conditions population of origin and maternal identity (figure 4c in Results) Thesedevelopmental stages are characterized by the full development of gills and the beginning developmentof the fore-limb bud (see [6272] for further details) The presence of any developmental deformities wasrecorded using the criteria of Hopkins et al [73]

26 Statistical analysesEgg survival () time to hatching length at hatching developmental stage at hatching and the presenceof developmental deformities () were analysed using generalized linear mixed model ANOVAs with

on May 20 2018httprsosroyalsocietypublishingorgDownloaded from

5

rsosroyalsocietypublishingorgRSocopensci4161057

Table 1 Effects of salinity temperature population and their interactions on newt (T granulosa) embryonic survival (a) Main effects(b) intrapopulation variance among females in offspring survival (random effects) Asterisks highlight statistical significance

(a) fixed effects F df p-value

salinity 114278 282 lt00001

temperature 507 282 0008

population 427 341 0010

salinitytimes temperature 533 4164 00005

populationtimes temperature 098 682 045

populationtimes salinity 240 682 0035

populationtimes salinitytimes temperature 127 12164 024

(b) random effects (nested within population) variance

female minus0001

femaletimes salinity minus0001

femaletimes temperature 0010

femaletimes salinitytimes temperature 0014

plt 005

the Identity link function and restricted maximum-likelihood estimation of random effects using PROCGLIMMIX in SAS

regv 94 Significance was set at α = 005 The fixed effects in these models were

salinity temperature population and the interactions of all of these factors To investigate the variationamong females within populations female identity was incorporated as a random effect crossed withtemperature and salinity and nested within population Because of very high mortality at 5 ppt thesample size of surviving embryos was greatly reduced to examine sublethal effects These effects (timeto hatching length at hatching developmental stage at hatching and the presence of developmentaldeformities) were therefore analysed both including and excluding survivors at 5 ppt We present thefull model results in the main text and provide the models excluding 5 ppt in electronic supplementarymaterial table S1

3 Results31 Egg survivalSalinity temperature and the interaction of salinity and temperature all significantly affected eggsurvival (table 1a) Increased salinity generally led to decreased survival with extremely low survivalat 5 ppt (less than 10 except for the Hunter Creek populations at 14degC) (figure 2) The negativeeffects of salinity were greatest at 7degC and less at 14degC (figure 2 table 1a) Soap Creek newts ingeneral had lower average survival than Hunter Creek newts (significant effect of population table 1aTukeyndashKramer post hoc multiple comparisons among populations p lt 005) particularly at increasedsalinities (significant interaction of population and salinity table 1a and figure 2) There was not asignificant three-way interaction of population with salinity and temperature (table 1a) Rather withinpopulations individual females varied in the survival of their offspring in different salinity andtemperature combinations (table 1b and figure 3 electronic supplementary material figures S1 andS2) Intrapopulation variation among females was generally high among all populations and in alltreatments (except for 5 ppt table 2) Different females in each population appeared to have differentspecific salinityndashtemperature combinations best suited for the survival of her eggs (variance estimatefor female times salinity times temperature table 2 and figure 3 electronic supplementary material figures S1and S2) For example female SC22 experienced 100 survival of her offspring in 2 ppt in 14degC but 0survival in the same salt concentration at 7degC (figure 3a) while female HUF11 experienced almost exactlythe opposite pattern (figure 3d)

32 Time to hatchingTemperature had the greatest effect on time to hatching (table 3a electronic supplementary materialtable S1a) with eggs hatching in approximately 16 days at 21degC 33 days at 14degC and 124 days

on May 20 2018httprsosroyalsocietypublishingorgDownloaded from

6

rsosroyalsocietypublishingorgRSocopensci4161057

temperature and population

e

gg s

urvi

val

SC HT HF HUF SC HT HF HUF SC HT HF HUF

21deg14deg7deg

0

20

40

60

80

100

0 ppt2 ppt5 ppt

salt treatment

Figure 2 Per cent survival in salinity (0 2 5 ppt) and temperature (7 14 21degC) treatments of eggs from rough-skinned newts(T granulosa) from four different populations

SC10SC12

SC14SC15

SC19SC21

SC22SC24

SC27 SC4SC8

0

20

40

60

80

100

female ID

e

gg s

urvi

val

71421

temperature

HF1HF17

HF18HF22

HF23HF25

HF27 HF3HF31

HF33 HF60

20

40

60

80

100

e

gg s

urvi

val

female ID

HT11HT13

HT14HT16

HT17HT18

HT20HT21

HT24HT25

HT27HT7

0

20

40

60

80

100

(b)(a)

(d)(c)

female ID

HUF1

HUF10

HUF11

HUF14

HUF15

HUF18

HUF20

HUF23

HUF25

HUF30

HUF50

20

40

60

80

100

female ID

Soap Creek pond (2 ppt) Hunter Creek tidal (2 ppt)

Hunter Creek fresh (2 ppt) Hunter Creek upper fresh (2 ppt)

Figure 3 Interfamily variation in per cent survival of eggs from 11 to 12 different female newts (T granulosa) from each of four differentpopulations in 2 ppt salinity in different temperature treatments (figures showing survival in other salinity treatments can be found in theelectronic supplemental material)

at 7degC (figure 4a electronic supplementary material table S2) Salinity significantly interacted withtemperature (table 3a electronic supplementary material table S1a) such that the effects of increasedsalinity delaying hatching were most significant at 7degC (figure 3a) Population played little role inaffecting time to hatching (table 3a electronic supplementary material table S1a) although there waslarge intrapopulation variation in the effects of salinity times temperature interactions on hatching timing(variance estimate of female times temperature times salinity table 3a electronic supplementary material tableS2 and figures S3ndashS5)

on May 20 2018httprsosroyalsocietypublishingorgDownloaded from

7

rsosroyalsocietypublishingorgRSocopensci4161057

(d)(c)

(b)(a)

time

to h

atch

ing

(day

s)

SC HT HF HUF SC HT HF HUF SC HT HF HUF0

20406080

100120140160 0 ppt

2 ppt5 ppt

salt treatment

temparture and population

stag

e at

hat

chin

g

SC HT7deg 14deg 21deg 7deg 14deg 21deg

7deg 14deg 21deg 7deg 14deg 21deg

HF HUF SC HT HF HUF SC HT HF HUF34

36

38

40

42

44

leng

th a

t hat

chin

g (m

m)

SC HT HF HUF SC HT HF HUF SC HT HF HUF4

6

8

10

12

temparture and population

d

efor

med

at h

atch

ing

SC HT HF HUF SC HT HF HUF SC HT HF HUF0

20

40

60

80

100

Figure 4 Sublethal effects of salinity and temperature embryonic developmental stress on hatchling newts (T granulosa) from fourdifferent populations Meansplusmn se (a) Time to hatching (days) (b) length at hatching (mm) and (c) developmental stage at hatching(Harrison stages) [72] (d) Percentage of offspring born with a developmental deformity

Table 2 Intrapopulation variation (minimum maximum mean) in egg survival () among females (N numbers) from four newt(T granulosa) populations under different salinityndashtemperature (degC) treatment combinations

treatment population

SC (N= 11) HT (N= 12) HF (N= 11) HUF (N= 11)

temperature min max mean min max mean min max mean min max mean

0 ppt

7 1111 100 7879 5714 100 9076 5556 100 7981 7500 100 9242

14 5556 100 8788 8785 100 9352 6667 100 8687 4444 100 8674

21 5556 100 8081 7778 100 9167 2222 100 8687 6667 100 8990

2 ppt

7 000 100 4270 2000 100 7066 4118 100 7216 3333 100 7922

14 1111 100 8081 4444 100 3146 7778 100 9091 1111 100 7576

21 2222 100 6869 5556 100 9082 3333 100 8182 2222 100 8283

5 ppt

7 0 1111 303 0 3333 713 0 2500 837 0 2500 559

14 0 2222 606 0 5556 1389 0 4444 1818 0 3333 1443

21 0 3333 404 0 0 0 0 1111 101 0 2222 404

33 Length and developmental stage at hatchingSalinity and temperature both influenced length at hatching and developmental stage at hatching(table 3bc electronic supplementary material table S1bc) with salinity and temperature interacting toaffect stage (table 3c electronic supplementary material table S1c) but not length (table 3b electronicsupplementary material table S1b) at hatching Embryos reared in saline solutions hatched significantlysmaller (figure 4b) and less developed (figure 4c) than those in the control Effects on developmentwere most dramatically seen at 7degC and 5 ppt (figure 4c) and Soap Creek animals were consistentlyless developed (though not smaller) than animals from any of the Hunter Creek populations (table 3bc

on May 20 2018httprsosroyalsocietypublishingorgDownloaded from

8

rsosroyalsocietypublishingorgRSocopensci4161057

Table3Sublethaleffects

ofsalinitytemperaturepopulation

andtheirinteractionsonnewt

(Tgranulosa)embryonic

growthanddevelopm

entMain

effectsaswellasthevarianceamongfemalesw

ithinpopulations(random

effects)arepresentedResultsforstatisticalmodelsexcludingindividualswhosurvivedexposureto5ppt(duetoverylow

survivalat5pptmdash

seeResults)arefoundinelectronicsupplem

entarymaterialtable

S1Asteriskshighlight

statisticalsignifi

cance

(a)timetohatching

(b)lengthathatching

(c)stageathatching

(d)deformitiesathatching

fixedeffects

Fdf

p-value

Fdf

p-value

Fdf

p-value

Fdf

p-value

salinity

743

265

00012lowast

9565

265

lt00001lowast

999

265

00002lowast

2147

265

lt00001lowast

temperature

265615

282

lt00001lowast

588

282

00041lowast

2181

282

lt00001lowast

115

282

032

population

137

341

026

263

341

006

506

341

00045lowast

238

341

008

salinity

timestemperature

549

491

00005lowast

144

490

023

1027

489

lt00001lowast

1061

490

lt00001lowast

population

timestemperature

195

682

008

210

682

006

155

682

017

899

682

lt00001lowast

population

timessalinity

083

682

055

059

665

074

202

665

008

210

665

006

population

timessalinity

timestemperature

086

1191

058

127

1190

025

146

1189

016

485

1190

lt00001lowast

random

effects(nestedw

ithinpopulation)

variance

variance

variance

variance

female

107

005

004

minus0002

female

timessalinity

minus027

005

002

014

female

timestemperature

253

minus0008

001

minus0001

female

timessalinity

timestemperature

4144

0041

016

minus008

lowast plt005

on May 20 2018httprsosroyalsocietypublishingorgDownloaded from

9

rsosroyalsocietypublishingorgRSocopensci4161057

TukeyndashKramer post hoc multiple comparisons among populations p lt 005 figure 4bc) There was amarginal (population times salinity p = 008) trend of the eggs of Soap Creek females hatching especiallyless developed in 5 ppt (figure 4c) Females within populations varied in both length (table 3b electronicsupplementary material table S3 and figures S6ndashS8) and developmental stage (table 3c electronicsupplementary material table S4 and figures S9ndashS11) of their offspring at hatching under differentsalinity times temperature combinations

34 Developmental deformitiesThe interactions of salinity temperature and population all significantly affected the percentage ofembryos hatching deformed (table 3d electronic supplementary material table S1d) Increasing salinityleads to increased rates of deformities with up to 100 of hatchlings being deformed at 5 ppt (figure 4d)This pattern was most extreme at 7degC and dampened significantly at 14degC with Soap Creek pondanimals generally more prone to deformities than Hunter Creek animals in increased salinity (althoughHUF also tended to have more deformities than HF at various salinities TukeyndashKramer post hoc multiplecomparisons among populations p lt 005) This pattern of SC newts suffering from deformities wasespecially obvious at 7degC where over 30 of SC embryos hatched deformed even in the freshwatercontrol (significant effect of population times temperature table 3d figure 4d) In addition while the effectsof salinity (especially 2 ppt) on deformities were dampened significantly at 14degC compared with 7degCover 40 of SC hatchlings in 2 ppt and 14degC still hatched deformed (far higher than any of the HunterCreek populations significant effect of population times salinity times temperature table 3d and figure 4d)Intrapopulation variation in the presence of deformities was low across salinity and temperaturecombinations but higher when examining the interacting effects of salinity and female alone (ieaveraged across temperatures table 3d electronic supplementary material table S5 and figures S12ndashS14)Developmental deformities ranged from abnormal spinal bending to reduced gill or limb developmentand abdominal cystsoedemas (ie as seen in [73])

4 DiscussionSalinity and temperature strongly interacted to affect the survival and development of newt embryosin this study with salinity generally having the greatest effect at temperature extremes We alsodetected variation in responses both between and within populations with different females ineach population having different temperature and salinity combinations best suited to the survivaland development of their eggs These results highlight the complex interacting nature of commonabiotic stressors facing organisms and provide important information for the potential resiliency ofpopulations

Increased salinity led to decreased survival increased time to hatching decreased length anddevelopmental stage at hatching and a high percentage of developmental deformities as seen beforefor amphibians (eg [26284673ndash75]) and other freshwater organisms (eg [252976]) Survival at 5 pptsalinity was especially low (less than 20 across temperatures and populations and over 90 ofsurvivors being small and deformed) perhaps representing an upper limit of salinity tolerance of newteggs While these results were not unexpected due to the high permeability of amphibian eggs and theirlack of effective osmoregulatory mechanisms [77] we found that the effects of salinity were stronglydependent on temperature with most lethal and sublethal salinity effects being strongest at temperatureextremes (7 and 21degC) Even if eggs escaped mortality at these stressor combinations hatching smallerless well developed and more deformed has important fitness consequences [7879] Similar synergisticpatterns of combined salinity and temperature stress have also been shown in fish (eg [4144]) andanuran tadpoles [54ndash56] where animals were able to tolerate some degree of temperature or salinitystress in isolation but not in interaction The toxicity of other stressors such as pesticides [4280]polychlorinated biphenyls [8] and dissolved oxygen [41] has also been shown to be amplified at marginaltemperatures andor salinities

Although relatively little work has been conducted on freshwater organisms salinity and temperaturehave repeatedly been shown to interact to affect survival growth and development of marine andestuarine fishes (eg [172281]) and invertebrates (eg [12168283]) with the most favourable conditionsfor survival and development often found at median temperatures [16172282] Indeed our resultsin concordance with these studies appear to support the general assertion that the interaction ofenvironmental stressors is most impactful at the marginal limits of tolerance of a species to anygiven stressor [124383] as organisms may be more vulnerable to further stressors when already

on May 20 2018httprsosroyalsocietypublishingorgDownloaded from

10

rsosroyalsocietypublishingorgRSocopensci4161057

stressed As temperature extremes continue to become the norm with climate change the toxicityof other interacting stressors to aquatic organisms both marine and freshwater may thereforealso increase

The exact optima of salinity and temperature for marine organisms often varies substantially fordifferent populations or closely related species (eg [1645588485]) Similar patterns have been foundfor different populations of the euryhaline European toads E calamita [55] and B viridis [56] (but not forthe frog A truei [54]) where the interaction of salinity and temperature impacted some populationsmore than others We found in general that population was often a significant predictor of survivaland development in our study with the three Hunter Creek populations grouping together in theirresponses compared with the Soap Creek pond newts Soap Creek newts had particularly high eggmortality in salt solutions and high rates of developmental deformities in 7degC and 2 ppt comparedwith Hunter Creek newts Soap Creek pond newts are found furthest from the ocean and are notnaturally or anthropogenically exposed to salinity (see [46]) In addition newts from Soap Creek pondvery rarely experience water temperatures as cold as 7degC (GRH and EDB 2012ndash2104 unpublisheddata) while newts in Hunter Creek experience a wide range of temperatures ranging from 7 to 21degClargely dependent on microhabitat depth flow-rate and weather (Z M Hopkins and GRH 2012ndash2104unpublished data) Indeed lentic habitats in general typically experience more stable (and warmer)temperatures than those in lotic habitats Thus thermal stress at this temperature may synergizewith salinity stress in particular for this lentic population This combined lack of history with lowtemperatures and even moderate salinities may thus help explain the relatively high mortality andhigh occurrence of sublethal effects the embryos of these animals experienced on average at 2 ppt and7degC relative to animals from coastal lotic Hunter Creek Interestingly the embryos of animals in thetidal environment of Hunter Creek (HT) were generally no more saline tolerant than other upstreampopulations contrary to our hypothesis and recent results on adult physiology of newts from HT versusHF populations [67] This may be due to the fact that amphibian eggs cannot osmoregulate as effectivelyas later life stages [7778] and so subtle differences among closely grouped populations in any onearea (ie within Hunter Creek) do not appear at the egg stage Our ability to extrapolate these findingsmore generally is hampered by only having one lentic inland population to compare with several loticpopulations from the same coastal stream Further work is needed sampling across a wider varietyof populations in different habitat types in order to gather the evidence needed to fully explain thepopulation-level variation in salinity times temperature tolerance we observe

While we did not always find significant differences among populations for most traits examinedin the interaction of temperature and salinity we did find a consistent pattern of variation in theresponses of different femalesrsquo eggs within populations to different salinity times temperature combinationswhich provides important information on the capacity for the population to evolve in the future[9] Within any given population it appears that the eggs of some females persist best in certainsalinities at low temperatures while others do much better at higher temperatures and the eggsof other females all react similarly to temperature Although almost all individuals do poorly at5 ppt (except one HT female in 14degC table 2 electronic supplementary material figure S2) at leastat lower salinities at any given temperature combination there are individuals within populationswhose offspring will survive significantly better than others and thus shift the mean response valueof the population in response to a selection event If this variation is heritable (which has yet to bedefinitively determined) then the population can evolve and evolutionary rescue [86] may occur infuture climate change scenarios for this species It will be important however for further work tocompare a larger sample size of females from each population and control for other possible sourcesof variance before definitive interpretations of population adaptability are made In addition theimportance of this adaptive possibility for conservation must be balanced with the dangers of bottlenecksand inbreeding depression inherent with reduced population sizes resulting from selection events[7586] The severity of selection will also be dependent on the severity of the stressor applied (iethere may be more potential for population persistence in response to 2 versus 5 ppt salinity) and itis possible that simultaneous selection from multiple stressors may impact the ability for adaptation[55] Despite these limitations our results combined with those of previous studies [4662] clearlyshow an intriguing pattern of strong differences in the survival of individual femalesrsquo offspring inresponse to salinity and temperature stress which have the potential to be important for adaptationWhile responses of populations and individuals to multiple environmental stressors are clearly complextaking into account intrapopulation variation in response to multiple interacting stressors has thepotential to provide important insights into the abilities of populations and species to persist in thefuture [9]

on May 20 2018httprsosroyalsocietypublishingorgDownloaded from

11

rsosroyalsocietypublishingorgRSocopensci4161057

By examining the effects of two of the most common abiotic stressors predicted to interact with

climate change on a model freshwater organism our study has emphasized the importance as wellas complexities therein of predicting responses of vulnerable organisms to environmental changeSuch complexities will undoubtedly only increase in a more natural setting where additional stressorswill also interact and be applied in unpredictable ways that are difficult to mimic in the laboratoryTaking into account the interacting nature of stressors in both the laboratory and the field as wellas the variation in responses of organisms both between and within populations will allow us betterpredictive power in striving to understand the ability of organisms to survive develop and reproducein increasingly stressful environments now and in the future

Ethics Newts were used in accordance with the Utah State University Institutional Animal Care and Use committeepermit 1524 and collected under Oregon Department of Fish and Wildlife permit 009-13Data accessibility Original data for this experiment are found in the electronic supplementary materialAuthorsrsquo contributions GRH SSF and EDB devised the study GRH conducted field and laboratory work analysedthe data and drafted the manuscript SSF and EDB contributed laboratory space and supplies All authorscontributed to writing of the manuscript and gave their final approval for publicationCompeting interests We declare we have no competing interestsFunding This research was funded by the USU Ecology Center a graduate research award from the Society forNorthwestern Vertebrate Biology and was supported by the USU Department of Biology GRH was additionallysupported by a postgraduate fellowship from the Natural Sciences and Engineering Research Council of CanadaAcknowledgements We thank Brittany Chamberlain for assistance in the laboratory and Zoeuml Hopkins for help in thefield We thank Turtle Rock Resort and Oregon State University for access to our field sites Thanks to Susan Durham(Utah State University) for help with data analysis and Zoeuml Hopkins Jessi Henneken (University of Melbourne) andHieu Pham (University of Melbourne) for reviewing a previous draft of the manuscript GRH acknowledges thesupport of Dr Thereacutesa Jones and the University of Melbourne during the drafting of this manuscript

References1 Wingfield JC 2013 Ecological processes and the

ecology of stress the impacts of abioticenvironmental factors Funct Ecol 27 37ndash44(doi1011111365-243512039)

2 Gunderson AR Armstrong EJ Stillman JH 2016Multiple stressors in a changing world the need foran improved perspective on physiological responsesto the dynamic marine environment Annu RevMar Sci 8 357ndash378 (doi101146annurev-marine-122414-033953)

3 Coors A De Meester L 2008 Synergisticantagonistic and additive effects of multiplestressors predation threat parasitism and pesticideexposure in Daphnia magna J Appl Ecol 451820ndash1828 (doi101111j1365-2664200801566x)

4 Rohr JR Elskus AA Shepherd BS Crowley PHMcCarthy TM Niedzwiecki JH Sager T Sih A PalmerBD 2004 Multiple stressors and salamanderseffects of an herbicide food limitation andhydroperiod Ecol Appl 14 1028ndash1040

5 Ban SS Graham NAJ Connolly SR 2014 Evidencefor multiple stressor interactions and effects oncoral reefs Glob Change Biol 20 681ndash697(doi101111gcb12453)

6 Todgham AE Stillman JH 2013 Physiologicalresponses to shifts in multiple environmentalstressors relevance in a changing world IntegrComp Biol 53 539ndash544 (doi101093icbict086)

7 Kimberly DA Salice CJ 2015 Evolutionary responsesto climate change and contaminants evidence andexperimental approaches Curr Zool 61 690ndash701(doi101093czoolo614690)

8 Hooper MJ Ankley GT Cristol DA Maryoung LANoyes PD Pinkerton KE 2013 Interactions betweenchemical and climate stressors a role formechanistic toxicology in assessing climate change

risks Environ Toxicol Chem 32 32ndash48(doi101002etc2043)

9 Urban MC et al 2016 Improving the forecast forbiodiversity under climate change Science 3531113ndash1125 (doi101126scienceaaf4802)

10 Schiedek D Sundelin B Readman JW MacdonaldRW 2007 Interactions between climate change andcontaminantsMar Pollut Bull 54 1845ndash1856(doi101016jmarpolbul200709020)

11 Cole VJ Parker LM OrsquoConnor SJ OrsquoConnor WAScanes E Byrne M Ross PM 2016 Effects of multipleclimate change stressors ocean acidificationinteracts with warming hyposalinity and low foodsupply on the larvae of the brooding flat oysterOstrea angasiMar Biol 163 1ndash17 (doi101007s00227-015-2782-x)

12 Chen X Stillman JH 2012 Multigenerational analysisof temperature and salinity variability effects onmetabolic rate generation time and acute thermaland salinity tolerance in Daphnia pulex J ThermBiol 37 185ndash194 (doi101016jjtherbio201112010)

13 Somero GN 2005 Linking biogeography tophysiology evolutionary and acclimatoryadjustments of thermal limits Front Zool 2 1ndash9(doi1011861742-9994-2-1)

14 Karraker NE Gibbs JP Vonesh JR 2008 Impacts ofroad deicing salt on the demography of vernalpool-breeding amphibians Ecol Appl 18 724ndash734(doi10189007-16441)

15 Gunter G 1956 Some relations of faunaldistributions to salinity in estuarine waters Ecology37 616ndash619 (doi1023071930196)

16 Browne RA Wanigasekera G 2000 Combinedeffects of salinity and temperature on survival andreproduction of five species of Artemia J Exp Mar

Biol Ecol 244 29ndash44 (doi101016S0022-0981(99)00125-2)

17 Alderdice DF Forrester CR 1968 Some effects ofsalinity and temperature on early development andsurvival of the English sole (Parophrys vetulus)J Fish Res Board Can 25 495ndash521 (doi101139f68-043)

18 Costlow Jr JD 1967 The effect of salinity andtemperature on survival and metamorphosis ofmegalops of the blue crab Callinectes sapidusHelgolander Wiss Meeresunters 15 84ndash97(doi101007BF01618611)

19 Gutieacuterrez J Soriano-Redondo A Dekinga A VillegasA Masero J Piersma T 2015 How salinity andtemperature combine to affect physiological stateand performance in red knots with contrastingnon-breeding environments Oecologia 1781077ndash1091 (doi101007s00442-015-3308-4)

20 Glova GJ McInerney JE 1977 Critical swimmingspeeds of coho salmon (Oncorhynchus kisutch) fry tosmolt stages in relation to salinity and temperatureJ Fish Res Board Can 34 151ndash154 (doi101139f77-021)

21 Anderson MT Kiesecker JM Chivers DP BlausteinAR 2001 The direct and indirect effects oftemperature on a predatorndashprey relationshipCan J Zool 79 1834ndash1841 (doi101139z01-158)

22 Huang Z-H Ma A-J Wang X-A Lei J-L 2014 Theinteraction of temperature salinity and bodyweight on growth rate and feed conversion rate inturbot (Scophthalmus maximus) Aquaculture 432237ndash242 (doi101016jaquaculture201404013)

23 Dimock Jr RV Groves KH 1975 Interaction oftemperature and salinity on oxygen consumption of

on May 20 2018httprsosroyalsocietypublishingorgDownloaded from

12

rsosroyalsocietypublishingorgRSocopensci4161057

the estuarine crab Panopeus herbstiiMar Biol 33301ndash308 (doi101007BF00390568)

24 Van Meter R Swan CM 2014 Road salts asenvironmental constraints in urban pond foodwebs PLoS ONE 9 1ndash12 (doi101371journalpone0090168)

25 Benbow EM 2004 Road-salt toxicity of selectMichigan wetland macroinvertebrates underdifferent testing conditionsWetlands 24 68ndash76(doi1016720277-5212(2004)024[0068RTOSMW]20CO2)

26 Dougherty CK Smith GR 2006 Acute effects of roadde-icers on the tadpoles of three anurans ApplHerpetol 3 87ndash93 (doi101163157075406776984266)

27 van Doorslaer W Stoks R 2005 Thermal reactionnorms in two Coenagrion damselfly speciescontrasting embryonic and larval life-history traitsFreshw Biol 50 1982ndash1990 (doi101111j1365-2427200501443x)

28 Chinathamby K Reina RD Bailey PCE Lees BK 2006Effects of salinity on the survival growth anddevelopment of tadpoles of the brown tree frogLitoria ewingii Aust J Zool 54 97ndash105(doi101071ZO06006)

29 Gonccedilalves AMM Castro BB Pardal MA Gonccedilalves F2007 Salinity effects on survival and life history oftwo freshwater cladocerans (Daphnia magna andDaphnia longispina) Ann LimnolInt J Limnol 4313ndash20 (doi101051limn2007022)

30 Sweeney BW Vannote RL Dodds PJ 1986 Effects oftemperature and food quality on growth anddevelopment of a Mayfly Leptophlebia intermediaCan J Fish Aquat Sci 43 12ndash18 (doi101139f86-002)

31 Komaki S Igawa T Lin S-M Sumida M 2016 Salinityand thermal tolerance of Japanese stream tree frog(Buergeria japonica) tadpoles from islandpopulations Herpetol J 26 209ndash213

32 Burraco P Gomez-Mestre I 2016 Physiologicalstress responses in aphibian larvae to multiplestressors reveal marked anthropogenic effects evenbelow lethal levels Physiol Biochem Zool 89462ndash472 (doi101086688737)

33 Kaushal SS Groffman PM Likens GE Belt KT StackWP Kelly VR Band LE Gisher GT 2005 Increasedsalinization of fresh water in the northeasternUnited States Proc Natl Acad Sci USA 10213 517ndash13 520 (doi101073pnas0506414102)

34 Petranka JW Francis RA 2013 Effects of road saltson seasonal wetlands poor prey performance maycompromise growth of predatory salamandersWetlands 33 707ndash715 (doi101007s13157-013-0428-7)

35 Williams WD 2001 Anthropogenic salinisationof inland waters Hydrobiologia 466 329ndash337(doi101023A1014598509028)

36 Kearney BD Byrne PG Reina RD 2012 Larvaltolerance to salinity in three species of Australiananuran an indication of saline specialisation inLitoria aurea PLoS ONE 7 e43427 (doi101371journalpone0043427)

37 Gornitz V 1995 Sea-level rise a review of recentpast and near-future trends Earth Surf ProcessesLandforms 20 7ndash20 (doi101002esp3290200103)

38 Rios-Loacutepez N 2008 Effects of increased salinity ontadpoles of two anurans from a Caribbean coastalwetland in relation to their natural abundance

Amphib-Reptil 29 7ndash18 (doi101163156853808783431451)

39 Cantildeedo-Arguumllles M Kefford BJ Piscart C Prat NSchaumlfer RB Schulz C-J 2013 Salinization of rivers anurgent ecological issue Environ Pollut 173157ndash167 (doi101016jenvpol201210011)

40 Uliano E Cataldi M Carella F Migliaccio O IaccarinoD Agnisola C 2010 Effects of acute changes insalinity and temperature on routine metabolismand nitrogen excretion in gambusia (Gambusiaaffinis) and zebrafish (Danio rerio) Comp BiochemPhysiol A 157 283ndash290 (doi101016jcbpa201007019)

41 Kinne O Kinne EM 1962 Rates of development inembryos of a cyprinodont fish exposed to differenttemperaturendashsalinityndashoxygen combinations CanJ Zool 40 231ndash253 (doi101139z62-025)

42 Hemmer MJ Middaugh DP Moore JC 1990 Effectsof temperature and salinity onMenidia beryllinaembryos exposed to terbufos Dis Aquat Organ 8127ndash136 (doi103354dao008127)

43 Garside ET Chin-Yuen-Kee ZK 1972 Influence ofosmotic stress on upper lethal temperatures in thecyprinodontid fish Fundulus heteroclitus (L) Can JZool 50 787ndash791 (doi101139z72-107)

44 Tay KL Garside ET 1974 Some embryonic responsesof mummichog Fundulus heteroclitus (L)(Cyprinodontidae) to continuous incubation invarious combinations of temperature and salinityCan J Zool 53 920ndash933 (doi101139z75-107)

45 Garside ET Morrison GC 1977 Thermal preferencesof mummichog Fundulus heteroclitus L and bandedkillifish F diaphanus (LeSueur) (Cyprinodontidae)in relation to thermal acclimation and salinityCan J Zool 55 1190ndash1194 (doi101139z77-154)

46 Hopkins GR French SS Brodie J 2013 Potential forlocal adaptation in response to an anthropogenicagent of selection effects of road deicing salts onamphibian embryonic survival and developmentEvol Appl 6 384ndash392 (doi101111eva12016)

47 Brown HA 1975 The timendashtemperature relation ofembryonic development in the northwesternsalamander Ambystoma gracile Can J Zool 54552ndash558 (doi101139z76-063)

48 Arrighi JM Lencer ES Jukar A Park D Phillips PCKaplan RH 2013 Daily temperature fluctuationsunpredictably influence developmental rate andmorphology at a critical early larval stage in a frogBMC Ecol 13 1ndash11 (doi1011861472-6785-13-18)

49 Bernabograve I Bonacci A Coscarelli F Tripepi MBrunelli E 2013 Effects of salinity stress on Bufobalearicus and Bufo bufo tadpoles tolerancemorphological gill alterations and Na+K+-ATPaselocalization Aquat Toxicol 132ndash133 119ndash133(doi101016jaquatox201301019)

50 Katz U 1973 Studies on the adaptation of the toadBufo viridis to high salinities oxygen consumptionplasma concentration and water content of thetissues J Exp Biol 58 785ndash796

51 Dunson WA 1977 Tolerance to high temperatureand salinity by tadpoles of the Philippine frog Ranacancrivora Copeia 1977 375ndash378 (doi1023071443921)

52 Smith GD Hopkins GR Mohammadi S Skinner HMHansen T Brodie ED French SS 2015 Effects oftemperature on embyronic and early larval growthand development in the rough-skinned newt(Taricha granulosa) J Therm Biol 51 89ndash95(doi101016jjtherbio201503010)

53 Peabody RB Brodie Jr ED 1975 Effect oftemperature salinity and photoperiod on thenumber of trunk vertebrae in Ambystomamaculatum Copeia 1975 741ndash746 (doi1023071443326)

54 Metter DE 1966 Some temperature and salinitytolerances of Ascaphus truei Stejneger J IdahoAcad Sci 4 44ndash47

55 Rogell B Hofman M Eklund M Laurila A HoumlglundJ 2009 The interaction of multiple environmentalstressors affects adaptation to a novel habitat in thenatterjack toad Bufo calamita J Evol Biol 222267ndash2277 (doi101111j1420-9101200901842x)

56 Rogell B Berglund A Laurila A Houmlglund J 2011Population divergence of life history traits in theendangered green toad implications for a supportrelease programme J Zool 285 46ndash55(doi101111j1469-7998201100843x)

57 Kefford BJ Buchwalter D Cantildeedo-Arguumllles M DavisJ Duncan RP Hoffman A Thompson R 2016Salinized rivers degraded systems or new habitatsfor salt-tolerant faunas Biol Lett 12 1ndash7(doi101098rsbl20151072)

58 Gonzalez A Bell G 2013 Evolutionary rescue andadaptation to abrupt environmental changedepends upon the history of stress Phil Trans RSoc B 368 1ndash6

59 Schoville SD Barreto FS Moy GW Wolff A BurtonRS 2012 Investigating the molecular basis of localadaptation to thermal stress population differencesin gene expression across the transcriptome of thecopepod Tigriopus californicus BMC Evol Biol 12170ndash186 (doi1011861471-2148-12-170)

60 Sih A Ferrari MCO Harris DJ 2011 Evolution andbehavioural responses to human-induced rapidenvironmental change Evol Appl 4 367ndash387(doi101111j1752-4571201000166x)

61 Schulte PM 2007 Responses to environmentalstressors in an estuarine fish interacting stressorsand the impacts of local adaptation J Therm Biol32 152ndash161 (doi101016jjtherbio200701012)

62 Hopkins GR Gall BG French SS Brodie Jr ED 2012Interfamily variation in amphibian early life-historytraits rawmaterial for natural selection Ecol Evol2 1637ndash1643 (doi101002ece3287)

63 Barrett RDH Schluter D 2007 Adaptation fromstanding genetic variation Trends Ecol Evol 2338ndash44 (doi101016jtree200709008)

64 Carroll SP Hendry AP Reznick DN Fox CW 2007Evolution on ecological time-scales Funct Ecol21 387ndash393 (doi101111j1365-2435200701289x)

65 Hopkins GR 2015 Salt and the rough-skinned newt(Taricha granulosa) evolutionary investigations oflocal adaptation to an anthropogenic and naturalstressor PhD dissertation Utah State UniversityLogan UT USA

66 Nussbaum RA Brodie Jr ED Storm RM 1983Amphibians and reptiles of the Pacific northwestMoscow ID University of Idaho Press

67 Hopkins GR Brodie J Mohammadi S Neuman-LeeLA Hopkins ZM French SS 2016 Physiologicalresponses to salinity vary with distance to the oceanin a coastal amphibian Physiol Biochem Zool 89322ndash330 (doi101086687292)

68 Hopkins GR Hopkins ZM 2015 Salty salamanderoccurrence of Dicamptodon tenebrosus in a tidalstream Northwest Nat 96 147ndash149 (doi101898nwnt-96-02-147-1491)

on May 20 2018httprsosroyalsocietypublishingorgDownloaded from

13

rsosroyalsocietypublishingorgRSocopensci4161057

69 Trenham PC 1998 Demography migration and

metapopulation structure of pond breedingsalamanders PhD dissertation University ofCalifornia Davis Davis CA USA

70 Environment Canada 2001 Priority substanceslist assessment report road salts p 283 OttawaCanada Environment Canada

71 Vernberg FJ Vernberg WB 2001 The coastal zonepast present and future Columbia SC University ofSouth Carolina Press

72 Harrison RG 1969 Harrison stages and description ofthe normal development of the spottedsalamander Amblystoma punctatum (Linn) InOrganization and development of the embryo (ed RGHarrison) pp 44ndash66 New Haven CT YaleUniversity Press

73 Hopkins GR French SS Brodie J 2013 Increasedfrequency and severity of developmentaldeformities in rough-skinned newt (Tarichagranulosa) embryos exposed to road deicing salts(NaCl amp MgCl2) Environ Pollut 173 264ndash269(doi101016jenvpol201210002)

74 Sanzo D Hecnar SJ 2006 Effects of road de-icingsalt (NaCl) on larval wood frogs (Rana sylvatica)Environ Pollut 140 247ndash256 (doi101016jenvpol200507013)

75 Brady SP 2013 Microgeographic maladaptiveperformance and deme depression in response to

roads and runoff PeerJ 1 e163 (doi107717peerj163)

76 Blasius BJ Merritt RW 2002 Field and laboratoryinvestigations on the effects of road salt (NaCl) onstreammacroinvertebrate communities EnvironPollut 120 219ndash231 (doi101016S0269-7491(02)00142-2)

77 Karraker NE Gibbs JP 2011 Road deicing saltirreversibly disrupts osmoregulation of salamanderegg clutches Environ Pollut 159 833ndash835(doi101016jenvpol201011019)

78 Hopkins GR Brodie J ED French SS 2014Developmental and evolutionary history affectsurvival in stressful environments PLoS ONE 9e95174 (doi101371journalpone0095174)

79 Gall BG Stokes AN French SS Schlepphorst EABrodie III ED Brodie Jr ED 2011 Tetrodotoxin levelsin larval and metamorphosed newts (Tarichagranulosa) and palatability to predatorydragonflies Toxicon 57 978ndash983 (doi101016jtoxicon201103020)

80 Lau ETC Karraker NE Leung KMY 2015Temperature-dependent acute toxicity ofmethomyl pesticide on larvae of three Asianamphibian species Environ Toxicol Chem 342322ndash2327 (doi101002etc3061)

81 Alderdice DF Forrester CR 1971 Effects of salinitytemperature and dissolved oxygen on earlydevelopment of the Pacific Cod (Gadusmacrocephalus) J Fish Res Board Can 28883ndash902 (doi101139f71-130)

82 Cain TD 1973 The combined effects of temperatureand salinity on embryos and larvae of the clamRangia cuneataMar Biol 21 1ndash6 (doi101007BF00351185)

83 Coles SL Jokiel PL 1978 Synergistic effects oftemperature salinity and light on the hermatypiccoralMontipora verrucosaMar Biol 49 187ndash195(doi101007BF00391130)

84 Dorgelo J 1974 Comparative ecophysiology ofgammarids (Crustacea Amphipoda) frommarinebrackish and fresh-water habitats exposed to theinfluence of salinityndashtemperature combinationsHydrobiol Bull 8 90ndash108 (doi101007BF02254909)

85 Yoshimatsu T Yamaguchi H Iwamoto HNishimura T Adachi M 2014 Effects oftemperature salinity and their interaction ongrowth of Japanese Gambierdiscus spp(Dinophyceae) Harmful Algae 35 29ndash37(doi101016jhal201403007)

86 Bell G 2013 Evolutionary rescue and the limitsof adaptation Phil Trans R Soc B 368 1ndash6

on May 20 2018httprsosroyalsocietypublishingorgDownloaded from