the growthdefense tradeoff and habitat specialization by plants in amazonian forests

13 Pages·2006·0.29 MB·English

Ecology,87(7)Supplement,2006,pp.S150–S162
(cid:1)2006bytheEcologicalSocietyofAmerica
THE GROWTH–DEFENSE TRADE-OFF AND HABITAT SPECIALIZATION
BY PLANTS IN AMAZONIAN FORESTS
PAUL V.A. FINE,1,2,3,9 ZACHARIAHJ. MILLER,3 ITALOMESONES,4SEBASTIANIRAZUZTA,5 HEIDIM.APPEL,6
M.HENRYH.STEVENS,7ILARI SA¨A¨KSJA¨RVI,8JACKC. SCHULTZ,6ANDPHYLLISD. COLEY1
1DepartmentofBiology,UniversityofUtah,SaltLakeCity,Utah84112USA
2EnvironmentalandConservationProgramsandDepartmentofBotany,FieldMuseumofNaturalHistory,Chicago,Illinois60605USA
3DepartmentofEcologyandEvolutionaryBiology,UniversityofMichigan,AnnArbor,Michigan48109-1048USA
4DepartmentofForestry,UniversidadNacionaldelaAmazonı´aPeruana,PlazaSerafı´nFilomeno246,Iquitos,Peru
5DepartmentofBiology,McMasterUniversity,Hamilton,OntarioL8S4K1Canada
6PesticideResearchLaboratory,PennsylvaniaStateUniversity,UniversityPark,Pennsylvania16802USA
7DepartmentofBotany,MiamiUniversity,Oxford,Ohio45056USA
8ZoologicalMuseum,CentreforBiodiversity,FIN-20014,UniversityofTurku,Finland
Abstract. Tropicalforestsincludeadiversityofhabitats,whichhasledtospecializationin
plants.NearIquitos,inthePeruvianAmazon,nutrient-richclayforestssurroundnutrient-poor
white-sandforests,eachharboringauniquecompositionofhabitatspecialisttrees.Wetested
thehypothesisthatthecombinationofimpoverishedsoilsandherbivorycreatesstrongnatural
selectionforplantdefensesinwhite-sandforest,whilerapidgrowthisfavoredinclayforests.
Recently,wereportedevidencefromareciprocal-transplantexperimentthatmanipulatedthe
presence of herbivores and involved 20 species from six genera, including phylogenetically
independent pairs of closely related white-sand and clay specialists. When protected from
herbivores, clay specialists exhibited faster growth rates than white-sand specialists in both
habitats.But,whenunprotected,white-sandspecialistsoutperformedclayspecialistsinwhite-
sandhabitat,andclayspecialistsoutperformedwhite-sandspecialistsinclayhabitat.
Herewetestfurtherthehypothesisthatthegrowth–defensetrade-offcontributestohabitat
specializationbycomparingpatternsofgrowth,herbivory,anddefensivetraitsinthesesamesix
genera of white-sand and clay specialists. While the probability of herbivore attack did not
differbetweenthetwohabitats,anartificialdefoliationexperimentshowedthattheimpactof
herbivoryonplantmortalitywassignificantlygreaterinwhite-sandforests.Wequantifiedthe
amountofterpenes,phenolics,leaftoughness,andavailablefoliarproteinfortheplantsinthe
experiment. Different genera invested in different defensive strategies, and we found strong
evidenceforphylogeneticconstraintindefensetype.Overall,however,wefoundsignificantly
higher total defense investment for white-sand specialists, relative to their clay specialist
congeners. Furthermore, herbivore resistance consistently exhibited a significant trade-off
againstgrowthrateineachofthesixphylogeneticallyindependentspecies-pairs.
Theseresultsconfirmtheoreticalpredictionsthatatrade-offexistsbetweengrowthrateand
defenseinvestment,causingwhite-sandandclayspecialiststoevolvedivergentstrategies.We
proposethatthegrowth–defensetrade-offisuniversalandprovidesanimportantmechanism
bywhichherbivoresgovernplantdistributionpatternsacrossresourcegradients.
Keywords: Amazon;ecologicalgradient;growth–defense trade-off; habitatspecialization;herbivory;
phenolics;phylogeneticcontrol;rainforest;reciprocal-transplantexperiment;terpenes;tropicaltrees.
INTRODUCTION ment and out-compete other plants that are not so
closely suited to the local conditions (Ashton 1969,
The regional diversity of plant species arises, in part,
because a given species is restricted to a subset of Cody 1978, Bunce et al. 1979). However, herbivore–
environmental conditions. But how and why does this plantinteractionscanalsocontributetotheevolutionof
habitat specialization occur? The most common expla- habitatspecialization.Theoreticalworkhasdemonstra-
nation is that habitat specialists are physiologically ted that herbivores can alter competitive relationships
adapted to growing in their particular abiotic environ- among plants, especially when there is spatial hetero-
geneityofresources(Loudaetal.1990,GroverandHolt
1998). Empirical studies at the population and com-
Manuscriptreceived8February2005;revised29April2005;
accepted 3 May 2005; final version received 11 July 2005. munity levels have documented that herbivores can
Corresponding Editor: A. A. Agrawal. For reprints of this reduceplants’potentialdistributions,restrictingthemto
SpecialIssue,seefootnote1,p.S1. a subset of the habitats that they might physiologically
9Present address: Department of Ecology and Evolu-
tolerate (Parker and Root 1981, Louda 1982, 1983,
tionary Biology, University of Michigan, Ann Arbor,
Michigan48109-1048USA.E-mail:paulfi[email protected] Louda and Rodman 1996, Olff and Ritchie 1998,
S150
July2006 PLANTTRADE-OFFSANDSPECIALIZATION S151
Carson and Root 2000, Harley 2003). Thus, herbivores cally plastic as opposed to genetically controlled
can play a major role in determining which species of adaptations toa particular habitat.
plants dominate in a community, as well as in which Thus, to test whether the growth–defense trade-off
habitatsa specieswill besuccessful. contributes to habitat specialization in white-sand and
ThelowlandAmazonianrainforestnearIquitos,Peru clay forests, we combined field observations and a
provides an ideal system to study habitat specialization reciprocal-transplant experiment to ask the following
and the role of herbivores. Forests in the Iquitos area questions: (1) Are there differences in herbivore abun-
grow on a mosaic of soil types; including red clay soils danceinthetwohabitats?(2)Isthereadifferenceinthe
and extremely infertile white-sand soils (Kauffmann et impact of herbivory in the two habitats, suggesting
al.1998).Thetwosoiltypeslieimmediatelyadjacentto selection for greater defense investment in white-sand
eachother,theboundariesarewelldefined,andeachsoil habitats?(3)Dowhite-sandandclayspecialistsdifferin
their type of defensive strategy or in their amount of
type is associated with a distinctive flora (Gentry 1986,
defense investment? Are these differences phylogeneti-
Va´squez1997,Fine2004).White-sandforestsaremuch
cally constrained or repeatedly and independently
moreresourcelimitedthanclaysoilforests(Medinaand
evolved? (4) Are defensive traits in white-sand and clay
Cuevas 1989, Coomes and Grubb 1998, Moran et al.
specialists affected by resource-driven phenotypic plas-
2000). Resource availability theory proposes that
ticity? (5) Do white-sand and clay specialists follow the
resource-limited species will have slower growth rates
predictions of thegrowth–defense trade-off?
and higher optimal levels of defense, reflecting the
decreasedabilityofaresource-limitedplanttocompen- MATERIALSANDMETHODS
satefortissueslostduetoherbivory(Janzen1974,Coley
Studysiteandstudy species
et al. 1985, Coley 1987b). Thus we predict that species
growing in white-sand forests should evolve to allocate We conducted this research in the Allpahuayo-
relatively more resources to defense than species Mishana National Reserve near Iquitos, Peru (38570 S,
growing inclay forests (Fineetal. 2004). 738240W).This57600-hareserveisat;130melevation
Recently, we reported the results of a reciprocal- andreceivesmorethan3000mmofprecipitationduring
transplantexperimentof20speciesofseedlingsfromsix the year, with no distinct dry season (Marengo 1998).
genera of phylogenetically independent pairs of white- Many white-sand specialist trees belong to the same
generaasneighboringclayforestspecialists,allowingfor
sand and clay specialist plants (Fine et al. 2004). We
a phylogenetically controlled experiment using edaphic
manipulated the presence of herbivores and found that
specialist species. For a reciprocal-transplant experi-
clay specialists grew significantly faster than did white-
ment, we chose 20 common white-sand and clay
sand specialists in both habitats when protected from
specialists from six genera from five families (see Fine
herbivores. But when herbivores were not excluded,
et al. [2004] for a phylogeny). The genera were Mabea
white-sand specialists out-performed clay specialists in
(Euphorbiaceae),Oxandra(Annonaceae),Pachira(Mal-
white-sand forests, and clay specialists grew faster than
vaceaesensulato),Parkia(Fabaceae),Protium(Burser-
white-sand specialists in clay forests. These results
aceae), and Swartzia (Fabaceae). Each genus was
strongly supported the existence of a growth–defense
represented by one white-sand specialist and one clay
trade-off, with habitat specialization being enforced by
specialist,exceptforProtium,whichwasrepresentedby
herbivores(Fine etal.2004).
six clay specialists and four white-sand specialists.
Here, we test further the predictions of the growth–
Designation of habitat for each species was accom-
defensetrade-offbycomparingspecies-levelpatternsof
plished by extensive inventories (Fine 2004, Fine et al.
growth,herbivory,anddefenseinthissamephylogeneti-
2005) as well as consultation of local floras and other
cally diverse group of tree species. We predicted that
published species lists from the western Amazon
closely related species specialized to contrasting soil
(Va´squez 1997, Ruokolainen and Tuomisto 1998,
types should diverge in traits that confer defense vs.
Jørgensenand Le´on-Ya´nez 1999, Garcı´aet al.2003).
those that confer growth. We investigated the evidence
for such differential investment while controlling for Nitrogenavailability
phylogeny.Therefore,anydifferencesindefensealloca- Totestfordifferencesinnitrogenavailabilitybetween
tion found between closely related white-sand and clay white-sandandclayhabitats,wefilled27nylonstocking
specialistscanbeinferredtobetraitsderivedforhabitat bags filled with 8 g of Rexyn 300 (H-OH) analytical
specialization. This phylogenetically controlled ap- grade resin beads. In May 2002, we placed the ion-
proachenabledustoinvestigatethedegreeofconstraint exchangeresinbagsbeneaththelitterlayerandrootmat
involved in the type and amount of defense, and to at the organic material–mineral soil interface at our
separate this from the repeated and independent white-sand and clay sites (Binkley and Matson 1983).
evolution of defensive traits due to selection from Thebagswerecollectedafterfiveweeks,extracted with
similarecologicalconditions.Second,examiningdefense KCl, and measured by standard techniques with an
investment with a reciprocal-transplant experiment autoanalyzer (University of Wisconsin Soils Labora-
allowedustoidentifywhichtraits(ifany)arephenotypi- tory). Nitrate, ammonium, and root mat depth differ-
S152 PAULV.A.FINEETAL. EcologySpecialIssue
encesweretestedforsignificancebetweensoiltypeswith common in both white sand and clay forests: Protium
a Wilcoxon signed-rankstest. (Burseraceae), Hevea and Mabea (Euphorbiaceae),
Pachira (Malvaceae s.l.), and Macrolobium (Fabaceae).
The reciprocal-transplant experiment
InSeptember2000,inthesamewhite-sandandclaysites
We used a reciprocal-transplant experiment to test where the wasps were collected, we sampled 355
whether white-sand and clay specialists had different individuals in the field from .20 species of Protium,
growthratesanddefenseinvestmentsaspredictedbythe twospeciesofHevea,twospeciesofMabea,twospecies
growth–defense trade-off hypothesis. In addition, the of Pachira, and three species of Macrolobium. Most of
reciprocal-transplant experiment allowed us to test for thesespecieswerefoundinonlyoneofthetwohabitats.
phenotypic plasticity of defense investments under Plantswere1–3mtall(juveniletrees).Wemarkednewly
different edaphicand herbivoretreatments. expanding leaves (or leaves that had already expanded
In May 2001, we built 22 control and 22 herbivore butwerenottoughened)withsmallcoloredwires,from
exclosures(33332m);halfwerelocatedinclayforest 1–10 leaves or leaflets per plant. After five to seven
and half in white-sand forest. We transplanted 880 weeksweestimatedtheamountofleafareamissingfrom
seedlings from the six genera into the controls and the marked leaflet (0–100%). Average amount of leaf
exclosures(seeFineetal.2004).Usingtheresultsofthe area missing was divided by number of days between
reciprocal-transplant experiment reported in Fine et al. marking and the census (damage per day). These data
(2004), we compared the amount of leaf and height were arcsine square-root transformed to improve
growth of the plants grown in herbivore exclosures to normality, and a mixed-model ANOVA (including the
the unprotected controls, and estimated the effect random factor genus and the fixed factor habitat) was
herbivory had on growth rates for each white-sand performed on the data to test for differences in
andclayspecialist.Thismeasureisreferredtothrough- herbivoryratebetweenwhite-sandandclayhabitats.
out as ‘‘protection effect.’’ The experiment lasted until
Impactof herbivory (defoliation experiment)
February 2003 (21 mo after transplanting, 18 mo after
firstdatacollection),atwhichpointleaveswerecollected In February 2003, after collecting leaf material for
to measuredefensivetraits. chemical analyses from all of the seedlings in the
reciprocal-transplant experiment, we removed 100% of
Insect abundanceandspecies richness
the remaining leaves to test the effect of defoliation on
To evaluate differences in insect abundance and white-sandandclayspecialistsinthetwohabitats.After
composition across habitats, we used a portable black three months, we counted the number of seedlings that
lightattachedtoabatterytoattractinsectsinfivewhite- survived defoliation. To compare mortality rates, we
sandandfiveclaysites.During8–20December2002,on averaged mortality for white-sand specialists and clay
rain-free evenings between 1900–2000, the black-light specialists in each of the 44 controls and exclosures
was illuminated and suspended above white sheets. We (Protium species in each control and exclosure were
collectedallinsectsfromthefamilies/ordersBlattoideae, weighted to give each genus equal importance in the
Coleoptera, Hemiptera, Homoptera, and Orthoptera. analyses). A fixed-factor ANOVA including the terms
We excluded all obvious predators and collected all habitat (white-sand or clay), origin (white-sand or clay
herbivorous insects from these five groups and counted specialist),andtheorigin3habitatinteractionwasused
and identified them to order and family and then to assess the effects of origin and habitat on mortality
separated them into morphospecies. Parasitoid wasps duetodefoliation.Posthoctestsontheindividualgroup
werecollectedwithmalaisetrapsoveratwo-yearperiod means were performed using the studentized t distribu-
in 15 white-sand and nonwhite-sand forest sites in the tion (appropriate for equal sample sizes; Sokal and
Allpahuayo-Mishana National Reserve (from 15 of the Rohlf1995).
samesitesdescribedinFine[2004])asapartofamuch
Defensive characteristics of white-sand
larger study on ichneumonid wasps (for detailed
andclay specialists
methods see Sa¨a¨ksja¨rvi [2003]). Since these parasitoid
waspsattackeitherherbivorousinsects(orpredatorsof Comparingdifferencesinherbivoryandgrowthisthe
herbivorous insects), we would expect parasitoid diver- best method of comparing defense investment in white-
sityandabundancetotrackherbivorousinsectdiversity sandandclayspecialists, sincethisapproachtakesinto
and abundance in white-sand and clay forests. To test account the entire arsenal of plant defenses as experi-
for differences between white-sand and clay habitats enced by the actual herbivores (cf. Simms and Rausher
(both the black light trap data and the wasp data), 1987). However, to investigate which particular defen-
Wilcoxon signed-ranks tests were conducted on the sive traits are deterring herbivores, we measured two
ranked abundancesand numbersof species. classesofchemicaldefenses,aphysicaldefense,andthe
nutritional quality of white-sand and clay specialists.
Fieldherbivory
After the transplant experiment was completed, we
Forherbivorycomparisonsinadditiontothosefrom collected leaves from all surviving plants to compare
thetransplantexperiment,wechosesixgenerathatwere defense investment in white-sand and clay specialists,
July2006 PLANTTRADE-OFFSANDSPECIALIZATION S153
andtoassessthe effectofhabitatandtreatmentonthe ourtotalphenolicsobtainedasdescribedwithavailable
plasticity of defense investment for each species. We foliar protein data to create a phenolic :protein ratio
collectedmarkedmatureleavesthatwereproducedafter (Nichols-Orians 1991).
plants were transplanted. We measured terpenes, total
Leaftoughness
phenolics, toughness, and available protein for all
seedlings in the reciprocal-transplant experiment. Ter- A ‘‘penetrometer’’ (Chatillon Universal Tension and
penes and phenolics are carbon-based secondary com- CompressionTester,Largo,Florida,USA)wasusedto
pounds common to many families of plants, including puncture holes through the leaf (or leaflet) lamina to
thoseinourresearch(MabryandGill1979,Bernayset giveameasureoftoughness.Itwasimpossibletotestthe
al.1989,SchultesandRaffauf1990).Althoughphenolics pair of species from the genus Parkia, since both have
andterpeneshavealternativefunctions,theycommonly bipinnately compound leaves, with leaflets not much
function in herbivore deterrence (Mabry and Gill1979, larger than the 3 mm diameter of the testing machine’s
Bernaysetal.1989,HermsandMattson1992,Langen- rod. We standardized the punch position to midway
heim1994;butseeHarborne1991,CloseandMcArthur between leaf tip and base, between the midrib and the
2002). Increased toughness of leaves (sclerophylly) is a leafmargin,avoidingthemainveinswherepossible.The
mechanical antiherbivore defense that is commonly punch test measures a combination of shear and
found worldwide in plants that live in resource-limited compressive strength and resistance to crack propaga-
environments (Coley 1983, 1987a, Grubb 1986, Turner tion. For these reasons, it has been criticized as not
1994).Finally,availablefoliarproteinisagoodmeasure specifically measuring leaf toughness (Choong et al.
of a plant’s nutritional quality. Moran and Hamilton 1992).Nevertheless,itiseasytoperforminthefieldand
(1980) hypothesized that plant nutrition can be consid- highlycorrelatedwithmorespecificteststomeasurethe
eredadefensivetraitifitcanbeselectedforbyherbivore physical properties of leaf toughness (shearing and
attack. This can result if herbivores detect nutritional tearing parameters) (Edwards et al.2000).
differences and prefer plants with higher nutrition (cf.
Soluble proteinassays
Scheirs et al. 2003). A second mechanism is if slow
growth by herbivores due to low nutrition results in Theamountofavailablefoliarproteinwasmeasured
higherpredationrates(cf.Dennoetal.2002). at the Appel/Schultz laboratory using the same dried-
leaf samples collected for the phenolics analyses. (See
Chemical defenses
Appendix A fordetailed methods.)
To compare terpene investment among the species,
Statistical analyses of growthanddefensive traits
;500 mg (fresh mass) leaves from the experimental
seedlingswerecollectedattheexperimentalsitesin2-mL Clay and white-sand specialists in each of the six
glassvialsandfilledwithdichloromethane(DCM).This genera were the experimental unit. Because there were
leaf–DCM mixture was used for qualitative and four white-sand specialists and six clay specialists from
quantitative analyses with gas chromatography (GC) the genus Protium, the responses for all Protium
and gas chromatography–mass spectrometry (GCMS). individuals were weighted to give each genus equal
(See Appendix A for detailed methods of terpene importanceintheanalysis.Thefourwhite-sandspecial-
extraction andanalysis.) ist Protium species were weighted at 0.25, the six clay
For comparisons of total phenolics, ;2 g fresh mass specialist Protium species were weighted at 0.167, and
of mature leaves of 16 individuals (8 protected and 8 species from all other genera were weighted at 1. We
unprotected) from each species in the reciprocal-trans- usedfixedfactorANOVAstotestforgenus,origin(the
plantexperimentwerecollectedandimmediatelyplaced difference between white-sand specialists and clay
in plastic tubes containing silica gel desiccant. Leaves specialists), habitat (whether species responded differ-
were later analyzed for phenolic compounds in the ently depending on where they were planted), and
Appel/Schultz laboratory at Penn State University. treatment(whetherdefenseinvestmentdiffereddepend-
Whenever possible, bulk tannins were prepared to ing on whether the plants were exposed to herbivores).
provide standards for the phenolic assays of individual Since we had a priori knowledge that different genera
samples. This is a crude purification, and although would have different defensive strategies (i.e., some
nonphenolicmaterialsareunlikelytobepresent(Hager- specieshaveterpeneinvestment,othersdonot),weused
manandKlucher1986;H.M.AppelandJ.C.Schultz, fixed-factor ANOVAs for defensive traits (genus was
unpublished data), the product is merely a more treated as a fixed factor), since our ability to generalize
representative sample of extractable polyphenols found our results in these analyses to unsampled genera is
in the actual plant than is a commercial standard from limited.Subsequenttotheoveralltest,individualgroup
some other source. (See Appendix A for detailed means werecomparedwith Tukeyhsd posthoctests.
information on all methodology of phenolic extraction,
Defense index
purification,andanalysis.)Becausetotalphenolicslikely
function as an antiherbivore defense by precipitating Becausedifferentspeciesofplantsarelikelytoemploy
availableprotein(HermsandMattson1992),wedivided different defensive strategies, we therefore devised a
S154 PAULV.A.FINEETAL. EcologySpecialIssue
simple method to combine all measures of chemical plots to test our predictions that (1) growth and
defense, leaf toughness, and available protein to inves- herbivory would be positively correlated, (2) growth
tigate whether, for each genus, white-sand specialists and defense would be negatively correlated, and (3)
were more defended than clay specialists. Values for herbivory and defense investment would be negatively
phenolics, terpenes, and leaf toughness were averaged correlated. Hypotheses about the correlations of traits
across both habitats and Z-transformed to give the were tested by the difference scores of the slopes and
defensetraitsamongthesixpairsofwhite-sandandclay wereevaluatedforsignificancewithone-tailedWilcoxon
specialists a mean of zero and a standard deviation of pairedsign-ranktests(Zar 1999).
one. Missing data was scored as zero. For available
protein, we standardized the inverse of the species RESULTS
averages, because a larger amount of available protein Differences innutrient availabilities
corresponds to lower defense. All four standardized
Clayforestsitescontainedsignificantlymoreavailable
defensevariableswerethensummedtocreateadefense
nitrogen(Z¼3.53,P,0.0004)thanwhite-sandforests,
index(DI).Foreachgenus,theDIfortheclayspecialist
more than twice as much available ammonium (Z ¼
wassubtractedfromtheDIofthewhite-sandspecialist. 2.71, P , 0.0061), more than an order of magnitude
Thismethodhastheassumptionthateachofthesefour more available nitrate (Z ¼ 3.59, P , 0.0003), and a
measures has equal weight, which is undoubtedly muchthinnerrootmat(Z¼4.89,P,0.0001;Table1).
incorrect, but preferable than subjectively assigning
different weights to defense types. These difference Habitat differencesin herbivore abundance
scores (DIWS – DICL) were used to test the prediction We found no significant differences in herbivore
that white-sand specialists are more defended than clay abundance or species richness between habitats for all
specialists with a one-tailed Wilcoxon paired signed- herbivoresoranyofthesixordersofherbivorousinsects
rankstest(Zar 1999). that we collected (P . 0.05, Wilcoxon signed-ranks
tests;Table1).Ofthe311morphospeciescollected,208
Phylogeneticindependence of growth, herbivory,
were collected only once (67%). Of the morphospecies
anddefense traits
collected more than once, 41 were collected only in
In order to evaluate whether growth, herbivory, and white-sand forest, 28 were collected only in clay forest,
defensetraitsweremoresimilarincloselyrelatedgenera, and 34 were collected in both forests (33% of the
we mapped each of the indices listed above, as well as morphospecies collected more than once). For para-
each individual defensive trait onto a phylogeny sitoid wasps, no statistical differences in abundance or
representing the relationships among the six genera morphospeciesdiversitywerefoundbetweenwhite-sand
and20species(seeFineetal.2004).Usingtheprogram and nonwhite-sand forest sites (Table 1). Moreover, in
PhylogeneticIndependence2.0,wetestedwhethertraits the reciprocal-transplant experiment, mean effect of
exhibited significant phylogenetic independence by protection for white-sand and clay specialists did not
comparing the average contrast values (C-stat) among changebetween habitats(Fig. 1a,b).
the actual trait values for the plant species to the
distribution of contrast values created by randomly Differences inthe magnitude ofherbivore attack
placing the trait values at the tips within the topology Clayspecialistsshowedanaverageincreaseingrowth
2000 times and testing for serial independence (TFSI) of0.25cm2/dinleafarea(pairedttest,df¼5,t¼(cid:1)2.91,P
(Abouheif 1999). If a trait is significantly phylogeneti- ,0.05)and0.0018cm/dinheight(pairedttest,df¼5,t¼
callyconstrained,thentheaverageC-statfortheactual (cid:1)2.59,P,0.05)whenprotectedfromherbivores,while
value will be greater than 95% of the average contrast white-sand specialists grew just as well or better in the
valuesgenerated bythe randomization. unprotectedvs.protectedtreatments.Whentheeffectof
herbivoreprotectiononleafareaandheightgrowthare
Correlations ofgrowth, defense, andherbivory data
Z-transformed and summed, all genera show the same
for thesix genera
patternthatclayspecialistsreceivedagreaterbenefitfrom
Species averages for growth (leaf area and height, herbivoreprotectionthandidwhite-sandspecialists.
averaged across habitats), the effect of herbivore During our study of field herbivory rates in the two
protection on growth (arithmetic difference between habitats, plants in clay forest sites suffered more than
the average leaf area and height with and without twice the herbivory rates on their new leaves than did
protection,foreachwhite-sandandclaygenusaveraged plantsinwhite-sandsites(mixedmodelANOVA,F
1,349
across habitats) and defenses, as described, were Z- ¼6.69, P , 0.01). Clay plants lost almost 23% of their
transformed and analyzed by a method analogous to new leaves per month, while white-sand plants lost
phylogenetically independent contrasts (Harvey and slightly.10% (Table1).
Pagel1991).Totestfortrade-offs,weplottedthevalues
Habitatdifferences inthe impactof herbivory
for the six species pairs and analyzed the six slopes, to
seeiftherelationshipbetweentraitswasconsistentwhen Aspredicted,seedlingsoverallsufferedhighermortal-
controllingforphylogeneticrelationship.Weusedthese ity due to total defoliation in white-sand habitat than
July2006 PLANTTRADE-OFFSANDSPECIALIZATION S155
TABLE1. Comparisonsofwhite-sandandclayforestsforleaflitterdepth,nitrogenavailability,young-leafherbivory,andinsect
abundanceandmorphospeciesrichness(means6SEreported).
Variable Clayforestsites White-sandforestsites
Rootmat(cm)(N¼44plots) 0.9161.0a 8.4860.6b
Nitrogenavailability(ppm)fromion-exchangeresinbags(N¼27resinbags)
NO– 349.2625.7b 25.6613.8a
3
NHþ 135.2632.7b 62.1617.5a
4
Totalnitrogen 484.4643.0b 87.7623.0a
Herbivory(%leafeaten/mo)(N¼355individuals) 22.864.3b 10.363.3a
Insectherbivoreabundance((no.individuals)(cid:2)(lighttrap)(cid:1)1(cid:2)h(cid:1)1)
Totalinsectherbivoreabundance 87.2612.6a 74.8618.1a
Blattoidabundance 3.060.7a 2.660.9a
Coleopteranabundance 20.069.0a 22.467.8a
Hemipteranabundance 7.664.9a 13.4611.7a
Homopteranabundance 20.064.5a 17.061.9a
Orthopteranabundance 36.668.2a 19.263.6a
Insectherbivorespeciesrichness((no.morphospecies)(cid:2)(lighttrap)(cid:1)1(cid:2)h(cid:1)1)
Totalinsectherbivoremorphospecies 45.064.3a 34.863.9a
Blattoidmorphospecies 2.660.5a 2.460.8a
Coleopteranmorphospecies 7.662.1a 8.062.0a
Hemipteranmorphospecies 3.261.0a 2.060.4a
Homopteranmorphospecies 15.862.8a 11.462.0a
Orthopteranmorphospecies 15.861.7a 10.862.1a
Parasitoidwasp((no.individuals)(cid:2)site(cid:1)1(cid:2)(malaisetrap)(cid:1)1for2yr)
Totalparasitoidwaspabundance 67.7628.5a 59.9610.8a
Totalparasitoidspeciesandmorphospecies 25.566.4a 22.063.3a
Note:Significantdifferencesbetweenforestsareindicatedbydifferentsuperscriptletterswithinarow(mixed-modelANOVA,
effect of habitat for herbivory, Wilcoxon signed-ranks tests between habitats for litter depth, nitrogen availability, insect
abundance,andspeciesrichness).
theydidinclayhabitat(effectofhabitat,F ¼4.96,P, the white-sand congenerthan in the claycongener, and
1,84
0.05).Inaddition,white-sandspecialistssufferedsignifi- thatourpredictionofhigherdefenseinthewhite-sandis
cantly more mortality than did clay specialists in both supported (one-tailed Wilcoxon paired signed-ranks
habitats(effectoforigin,F1,84¼22.8,P,0.0001;Fig.2). test, T0.05(1),6¼1, P,0.05, Fig. 3).
Forphenoliccompounds,white-sandspecialistsover-
Differencesin defense investment
all had significantly higher values for both total
Type of defense.—We found strong evidence for phenolics (effect of origin, F ¼ 50.3, P , 0.0001)
1,292
phylogenetic constraint for type of defense. The main and phenolic : protein ratios (F ¼ 128.2, P ,
1,292
effect of genus was always significant for differences in
0.0001) with, respectively, three-sixths and four-sixths
terpenes, phenolics, leaf toughness, and available foliar
of the genera exhibiting significant relationships in the
protein. Moreover, it is clear that different genera are
predicted direction (Fig. 3, see Appendix D). The two
relyingondifferentdefensestrategies,aseachofthesix
genera that invested in terpenes, Protium and Oxandra,
genera had a distinct defense investment pattern (see
exhibitedverydifferentpatternsofterpeneinvestmentin
Appendix C). For example, only two genera, Oxandra
their white-sand and clay specialists (see Appendix D).
and Protium, contained measurable terpenes identified
Oxandraxylopiodies,theclayspecialist,hadsignificantly
by GCMS (Appendix C). Similarly, only two genera,
higher sesquiterpene and total terpene concentrations
Pachira and Parkia had white-sand specialists with
than O. euneura, the white-sand specialist (P , 0.05,
obviously tougher leaves than clay specialists. The
Tukey tests; see Appendix D). In contrast, Protium
pattern of high phenolic investment and low available
white-sandspecialistshadhighermonoterpeneandtotal
foliar protein in white-sand specialists was more
terpeneconcentrationsthandidProtiumclayspecialists
consistent across the six genera, but still there were
(P,0.05,Tukeytests;seeAppendixD).BothOxandra
exceptions(OxandraandProtium forphenolics, Mabea
foravailable protein; Fig.3). and Protium white-sandspecies hadsignificantly higher
Whereas different genera invest in different defensive concentrations of diterpenes and other resins than did
strategies, we found no consistent relationship between their respective clay specialists (see Appendix D).
anyparticulardefensivetraitsthatwouldsuggesteither Therewasnooveralleffectoforiginonleaftoughness
anegativetrade-offorasynergisticrelationshipbetween (see Appendix D). In contrast, white-sand species had
defensetypes (Fig. 3). loweravailableproteinintheirleavesthanclayspecial-
Amount of defense investment.—We found that five- ists (significant effect of origin, F ¼ 393.5, P ,
1,292
sixthsofthegenerahaveahigherdefenseindex(DI)in 0.0001; seeAppendix D).
S156 PAULV.A.FINEETAL. EcologySpecialIssue
FIG.1. Theeffectoforiginandhabitatinthereciprocal-transplantexperimentfor(a)theeffectofherbivoreprotectiononleaf
growth(cm2/d),(b)theeffectofherbivoreprotectiononheightgrowth(cm/d),(c)totalterpenesandresins(mLterpenes/mgofdry
leaf material), (d) total phenolics (g phenolics/g dry leaf material), (e) phenolic:protein ratio (phenolics divided by available
protein,aunitlessratio),(f)leaftoughness(gramsofmasstopuncha3-mmrodthroughaleaf;1.0g¼1.38kPa),and(g)available
protein (g soluble protein/g dry leaf material). Histograms show means 6 SE. Different letters above bars indicate significant
differencesamongthedifferentgroups(Tukeytests).
Defensive traits andphenotypicplasticity The effect of habitat on leaf toughness was highly
significant(F ¼51.6,P,0.0001;Fig.1f).Sixteenof
There was no significant overall effect of habitat for 1,388
terpenes(Fig.1c).Asidefromtheoutlierbehaviorbyone 17speciesmeasuredhadgreaterleaftoughnessinwhite-
species,therewasnoevidenceofphenotypicplasticityin sandthanclayhabitat;threeofthoseweresignificant(see
phenolic investment (Fig. 1d). Swartzia cardiosperma is AppendixC).Incontrast,eventhoughnitrogenavailabil-
the only species of 20 in the experiment that showed a itiesdifferedbymorethanfivetimesinthetwohabitats,
significant effect of habitat for phenolic: protein ratios there was no significant effect of habitat on available
(seeAppendixC). proteinforeitherwhite-sandorclayspecialists(Fig.1g).
July2006 PLANTTRADE-OFFSANDSPECIALIZATION S157
Phylogenetic independence ofgrowthanddefensive traits
There was evidence for significant phylogenetic
dependence for total phenolics (C-stat ¼ 0.34, P ,
0.002), terpenes (C-stat ¼ 0.34, P , 0.002), and leaf
toughness(C-stat¼0.32,P,0.012).Thedefenseindex
(C-stat¼0.23,P,0.11)andavailableprotein(C-stat¼
0.11, P , 0.148) exhibited a trend toward phylogenetic
constraint. We found no evidence for phylogenetic
constraint in GI (C-stat ¼ 0.07, P , 0.35) and the
protectioneffectindex(C-stat¼0.01,P,0.399),results
that in part might reflect an artifact of our design
FIG. 2. Mortality results of the 100% defoliation experi-
ment. Bars show average mortality (6SE) for each origin and because our sampling within each genus was limited to
habitat combination. Different letters above bars indicate pairedwhite-sandandclayspecialists,whichmaximized
significantdifferences(posthoctests,studentizedtdistribution). the variation betweenclosely related species.
DISCUSSION
Evaluatingthetrade-off:growthvs.defensevs.herbivory
Habitatdifferences inherbivore populations
The growth index (GI) and the herbivory index (HI)
showed a significant positive relationship (all six of the Two separate measures of herbivorous insect com-
genera with positive slope, T ¼ 0, P , 0.025; munities found statistically similar diversity and abun-
0.05(1),6
Fig. 4a). There was a significant negative trade-off danceinthetwoforesttypes.Inaddition,afullthirdof
between GI and total DI, with five-sixths of the genera the morphospecies that were collected more than once
showing a negative slope (T ¼ 1, P , 0.05, occurred in both habitats. These results are likely
0.05(1),6
Wilcoxonpairedsigned-rankstest;Fig.4b).Finally,DI explained by the large home range and dispersal
showed a significant negative relationship with HI capabilities of many herbivorous insects (Stork 1988),
(T ¼1, P,0.05; Fig.4c). coupled with the fact that most white-sand forest
0.05(1),6
FIG.3. Thedefenseindex(DI)scoresforeachgenusareplotted,showingthedifferencebetweenclay(CL)andwhite-sand(WS)
specialists.Thethree-letterlabelsof thelinescorrespondto the genustablebelowthe plot.Black boxesin the tableindicate a
significantly higher defensive trait for that genus in the white-sand specialist, and shaded boxes indicate a significantly higher
defensivetraitfortheclayspecialist(contrarytopredictions).ThefinalcolumnshowstheDIscoresforeachgenus,withanegative
numbersignifyingascoreinthedirectioncontrarytopredictions.
S158 PAULV.A.FINEETAL. EcologySpecialIssue
insect herbivores indeed range into white-sand forests.
Moreover, these patterns are consistent with our
herbivory data from the reciprocal-transplant experi-
mentshowingthatclayspecialistseedlingswereattacked
at similar frequencies whether they were transplanted
intoclay orwhite-sand forests(Fig. 1a,b).
Habitatdifferences inthe impactof herbivory
We predicted that the impact of herbivory would be
greaterinawhite-sandforest,becauseitismoredifficult
for plants to replace the nitrogen lost to herbivores
(Coley 1987b, Craine et al. 2003). This prediction was
supported by the fact that all plants transplanted into
white-sand forest had significantly higher mortality
whendefoliatedthanthosetransplantedintoclayforest
(Fig.2).
In the defoliation experiment, white-sand specialists
suffered a significantly higher mortality rate than did
clay specialists (Fig. 2), confirming a key prediction of
thegrowth–defense trade-offthatwhite-sand specialists
ought to have more difficulty replacing tissue lost to
herbivores(Coleyetal.1985).Thisdifferentialresponse
todefoliationbyspeciesadaptedtolow-fertilitysoilsvs.
speciesadaptedtohigh-fertilitysoilswasalsofoundina
studyinSingapore(LimandTurner1996).Thus,when
heavilydefendedwhite-sandspeciesaredefoliated,they
lose costly leaves that represent a high percentage of
theirenergybudget.Duetotheirslowgrowthrate,they
arethenunabletocompensate,andthisinturnincreases
theirmortalityrate(Coleyetal.1987b).Forthisreason
the impact of herbivory appears to be substantially
greaterfor plants adaptedto low-resource conditions.
Differencesin defense investment
Type of defense.—Different genera adopted dramat-
ically different defensive strategies. There was a con-
sistent signal of phylogenetic constraint in our analyses
of plant defenses, as genus was a significant factor in
each defense variable (see Appendices B and D), and
tests for phylogenetic independence confirmed this. In
terms of terpenes, phenolics, toughness, and low
nutrition, there was no consistent ‘‘syndrome’’ of
defensive investment in the six genera; instead, each
genus allocated to different combinations of these (and
presumably other unmeasured) traits. Indeed, there is
little theoretical or empirical support for the idea of a
FIG. 4. Plots of the six species pairs for (a) growth rate
index vs. protection effect index, (b) growth rate index vs. general negative trade-off between types of defensive
defense index, and (c) herbivory vs. defense index. The strategies(Korichevaetal.2004,AgrawalandFishbein
consistency and magnitude of these slopes were used to test 2006).
thepredictionsofthegrowth–defensetrade-offhypotheses.The
Amount of defensive investment.—For Protium, we
three-letterlabelscorrespondtothesixgeneralistedinFig.3.
found higher concentrations of terpenes in white-sand
habitats in the Iquitos area are only a few square specialists as predicted, but for Oxandra the reverse
kilometers. It is important to recognize that our pattern was found (Fig. 3). The terpene profile of
herbivoresamplingwasextremelylimitedandprecludes Oxandra is driven by sesquiterpenes, which could
us from drawing definitive conclusions concerning the possibly be serving a function other than defense, or
relative abundance of herbivore populations in white- do not function in a dosage-dependent fashion (Ger-
sand and clay habitats. Nevertheless, our herbivore shenzon and Croteau 1991, Langenheim 1994). In
estimatesrepresenttwoindependentcorroborationsthat contrast to sesquiterpenes, both Protium and Oxandra
July2006 PLANTTRADE-OFFSANDSPECIALIZATION S159
white-sand specialists were found to have higher When the protection effect of each species is plotted
diterpenesandotherresinscomparedtoclay specialists against the overall growth rate (Fig. 4a), all six genera
(see Appendix B). Diterpenes are not volatile and are exhibited a positive relationship. In each genus, herbi-
thoughttobeeithertoxic(LerdauandPenuelas1993)or vores selectively attacked the faster growing species
a type of physical defense against herbivores or more than the slower growing species. This is evidence
pathogens (Langenheim 1994). that faster growing plants have lower resistance to
Total phenolics and phenolic : protein ratios were herbivores, consistent with the predictions of the
significantly higher overall for white-sand specialists growth–defense trade-off. Coley (1983, 1987b) found
thanforclayspecialists(seeAppendixB).Inourstudy, thesamerelationshipinPanamawherethegrowthrates
percentagedrymassoftotalphenolicsranged3–37%,a of40speciesoftreeswerepositivelycorrelatedwiththeir
large range that is certainly an overestimate and rates of herbivory.
highlights the difficulty of precise phenolic quantifica- In the graphs of Fig. 4a, the lengths of the lines
tion in the laboratory (Appel et al. 2001). Finally, we correspond to the amount of variation in growth rate
found significantly less available protein in white-sand and antiherbivore traits within the species (white-sand
specialists.Thiswasthemostconsistenttrait,withfive- andclay)ofeachgenus.Forexample,somegeneralike
sixthsofthespeciespairsshowingthesamepattern(see Parkia are represented by longer lines in the horizontal
Appendix D). direction (Fig. 4a), because this genus includes both
shade-tolerantspeciesandthosethatthriveinhigh-light
Defensive traits andphenotypic plasticity
conditions. Therefore, the clay specialist in Parkia is a
Wedidnotfindmanycasesofphenotypicplasticityin very fast grower relative to the Protium and Swartzia
the seedlings’ allocation to chemical defenses. Very few species,allofwhichareshade-tolerantspeciesandnever
specieshadsignificantincreasesordecreasesinterpenes found in tree-fall gaps (P. V. A. Fine, personal
or phenolics due to habitat (Fig. 1c, d). Similarly, observation).Yetthefactthattheslopesofthesixlines
available foliar protein did not change depending on in Fig. 4a are so similar suggests the existence of a
wheretheseedlingswereplanted(Fig.1g),eventhough universal trade-off, even among species with such
nutrient levels were significantly different between the disparate growth rates anddefensivestrategies.
two habitats. We conclude that, for the genera in our Whenthedefenseindex(DI)scoresforthesixgenera
study,herbivoreresistanceduetochemicaldefensesand are plotted against their growth rate index (GI) (Fig.
available protein content is due to genetically based, 4b), we found a significant negative relationship, with
fixed traits (but see Boege and Dirzo 2004). Thus, five of the six genera having higher DI scores in the
defense differences result from natural selection by slower vs. faster growing species. The slopes in this
herbivores and are not just passive responses to differ- graph exhibit much more variation than the growth vs.
encesof available nutrients inthe soils. herbivory graph (Fig. 4a), likely due to the coarse
In contrast to our results with chemical and nutri- method by which we attempted to quantify defense
tional defenses, we found a strong overall effect of investment in these species. The one outlier genus,
habitatonleaftoughness,whichwassignificantforthree Mabea, shows the opposite relationship than the other
species(Fig.1f;seeAppendixC).Overall,wefoundthat fivegenera,withahigherDIscoreinitsfastergrowing,
leaf toughness was significantly higher for white-sand clay specialist. Because the slower growing (white-sand
species in only two genera, Parkia (which we were not specialist) Mabea received the least amount of attack
able to measure with our penetrometer, but for which from herbivores in the experiment (see Appendix B), it
the pattern was obvious) and Pachira. In contrast, two seemslikelythatitactuallyisverywelldefendedandwe
previous studies found that white-sand plants had failed to accurately quantify its defensive investment.
significantly tougher leaves than clay plants (Coley OnereasonforthismaybethatMabeaistheonlygenus
1987a, Choong et al. 1992). These studies did not take ofthesixthatproducescopiouswhitelatex,andwedid
phylogenyintoaccount,buttheirresultsforwhite-sand not quantify this trait in our comparisons. The
and clay species averages were much more divergent herbivory vs. defense graph (Fig. 4c) echoes this point,
than ours. One possibility for the discrepancy is that with Mabea the only genus whose DI score does not
toughnessinthesetwostudieswasonlymeasuredinthe match itsherbivoryindex score.
plants’ home habitats. While our results in no way
Phylogeneticapproachto studying
negate the potentially strong selective effect of herbi-
thegrowth–defense trade-off
vores on sclerophylly, they do suggest that future
comparisons of white-sand and clay species should not Ourapproachusingmultiplepairsofcongenersfrom
only be controlled for phylogenetic relationships, but ecologically divergent habitats differs from some other
also forresource availability. more quantitative approaches that have used data on
branch lengths from a phylogenetic tree to test for
Evaluating thegrowth–defense trade-off
correlations between particular traits and habitat
The evolutionary trade-off between growth and association (Cavender-Bares et al. 2004). In our
defense is illustrated by the data graphed in Fig. 4. approach, we ignore branch lengths by design, since