Journal of Induced resistance of host tree foliage during and after a Ecology 0888\ 57\ 271Ð278 natural outbreak

PEKKA KAITANIEMI \ KAI RUOHOMAÝKI\ TOOMAS TAMMARU and ERKKI HAUKIOJA Section of Ecology\ Department of Biology\ and Kevo Subarctic Research Institute\ University of Turku\ FIN! 19903 Turku\ Finland

Summary 0[ Plant resistance against insect herbivores often increases after experimental damage to foliage\ but few studies have obtained _eld estimates of the e}ect of induced resistance on insect populations during and after a natural insect outbreak[ 1[ This study measured the e}ect of quality of the host tree\ mountain birch "Betula pubescens ssp[ czerepanovii#\ on the periodically ~uctuating folivore Epirrita autumnata "\ Geometridae# during peak and postpeak years of an outbreak in Finnish Lapland[ Comparisons were made both within and between study sites to assess host plant quality\ and thereby the e}ect of delayed induced resistance "DIR#[ 2[ In within!site comparisons\ a set of experimental trees was defoliated by wild larvae in the peak year of the outbreak\ whereas control trees were protected from defoliation by spraying with an insecticide[ The e}ect of host plant quality was quanti_ed in the following year by measuring the pupal mass of E[ autumnata larvae reared in enclos! ures on these trees[ 3[ In between!site comparisons\ the sizes of pheromone!trapped males were measured at both outbreak and low density sites during the progress of the outbreak[ The size of trapped males was subsequently used to estimate the corresponding fecundity of females at the same sites[ 4[ Pupal mass of E[ autumnata reared on trees defoliated in the previous year was 9Ð 09) lower than on those trees protected from defoliation by the insecticide[ Field! collected adults indicated a similar pattern] they were smaller at outbreak sites than at low!density sites\ and the size reached its minimum in the post!peak year[ However\ the estimated loss of reproductive capacity of females resulting from DIR was too small to be the sole explanation for the termination of the outbreak[ 5[ Whether the weak DIR response in this system was a characteristic of the 0889s outbreak alone remains unclear\ because di}erent terminating agents may be impor! tant for di}erent individual outbreak peaks[ During this outbreak\ larval parasitism and developmental asynchrony between larvae and birch were probably more impor! tant reasons for population collapse than DIR[ Key!words] Betula pubescens\ Epirrita autumnata\ foliage quality\ induced resistance\ insect outbreak\ phenology\ plantÐherbivore interactions\ population dynamics[ Journal of Animal Ecology "0888# 57\ 271Ð278

remained controversial "Myers 0877^ Berryman 0885#[ Introduction One of the possible driving factors behind population The relative role of di}erent factors behind cyclicity cycles is herbivore!induced\ density!dependent in population density of forest Lepidoptera has deterioration in plant quality[ If such induced resist! ance has delayed detrimental e}ects on survival or Correspondence] Pekka Kaitaniemi\ Section of Ecology\ Department of Biology\ University of Turku FIN!19903 fecundity of in years following large scale Þ 0888 British Turku\ Finland[ Tel[ ¦247 1222 4961[ Fax] ¦247 1222 5449[ defoliation\ it can suppress the population for some Ecological Society E!mail] Pekka[KaitaniemiÝutu[_ years\ until its e.ciency is mitigated and population

271 272 density can start to rise again "Haukioja 0879\ 0871#[ pheromone!trapped E[ autumnata males caught from P[ Kaitaniemi et al[ The reduction in fecundity of Lepidopteran species study plots located both at outbreak and low!density has been observed during outbreaks in the _eld sites[ Second\ in 0883\ the year following the _rst "Mason\ Beckwith + Paul 0866^ Carter\ Ravlin + peak year\ foliage quality was bioassayed directly at McManus 0880#\ and such changes have been attri! outbreak sites by rearing larvae on trees which had buted to changes in foliage quality "Baltensweiler + either been defoliated by wild larvae or been protected Fischlin 0877#\ but\ in general\ data concerning the from defoliation by insecticide spraying in the pre! e}ect of induced changes in host quality on natural vious year[ insect populations are lacking "Karban + Baldwin 0886#[ Materials and methods Increased plant resistance\ interpreted from decreased performance of herbivores\ has often been STUDY ORGANISMS AND SITES demonstrated after experimental damage to foliage Epirrita autumnata is a spring folivore which over! "for reviews see Haukioja + Neuvonen 0876^ Karban winters in the egg stage[ The larvae hatch at the time of + Myers 0878^ Haukioja 0880^ Tallamy + Raupp bud burst of birch\ and complete their growth within 0880#^ for example\ in numerous experiments with about 5 weeks[ Adults emerge in August or Septem! birches "Betula spp[#\ defoliation has been shown to ber[ Populations of E[ autumnata ~uctuate period! trigger both delayed and rapid forms of induced resist! ically in northern Fennoscandia\ reaching outbreak ance ðdelayed induced resistance "DIR# and rapid densities at 8! to 09!year intervals and sweeping across induced resistance "RIR#\ respectivelyŁ in foliage mature birch forests "Tenow 0861^ Haukioja et al[ "Neuvonen + Haukioja 0880#[ Because RIR a}ects 0877a^ Bylund 0884^ RuohomaÃki et al[ 0886#[ Because the current generation of herbivores\ it tends to sta! of the extensive defoliation it causes during such out! bilize insect population dynamics\ whereas DIR\ breaks\ E[ autumnata is the most destructive herbivore because of its delayed density dependence on suc! of its staple food plant\ mountain birch ðBetula pubes! cessive generations\ can contribute to cyclic ~uc! cens ssp[ czerepanovii "Orlova# HaÃmet!ahti\ syn[ tuations in insect density "Haukioja 0871#[ B[ pubescens ssp[ tortuosa "Ledeb[# Nyman^ Tenow However\ the possible role of insectÐhost plant 0861^ Kallio + Lehtonen 0862#[ interactions in insect population dynamics is more The performance of E[ autumnata is largely deter! complex than RIR or DIR alone^ for example\ insects mined by the quality of the foliage it consumes\ and may be able to avoid resistant plants in the _eld "e[g[ is re~ected in the _nal larval mass or pupal mass Sork\ Stowe + Hochwender 0882#\ insect performance "Neuvonen + Haukioja 0880^ Tammaru 0887#[ Pupal may be a}ected by the e}ect of defoliation on syn! mass is often used as the measure of individual _tness\ chrony between leaf ~ush and eclosion of larvae "Kai! because it is the main determinant of potential fec! taniemi\ RuohomaÃki + Haukioja 0886b#\ and insects undity in females "Haukioja + Neuvonen 0874a^ Tam! may have feeding habits which help them to overcome maru\ Kaitaniemi + RuohomaÃki 0885#[ The strong plant resistance partially or completely "Tallamy 0874^ correlation between pupal mass and fecundity appar! Dussourd + Eisner 0876^ Roland + Myers 0876^ Hau! ently persists also in the _eld conditions as a result of kioja et al[ 0877b^ Sagers 0881^ Kaitaniemi\ Ruo! restricted importance of adult feeding and short adult homaÃki + Haukioja 0886a#[ Kaitaniemi et al[ "0886a# lifespan in this species "Tammaru et al[ 0885#[ have recently shown that the larvae of Epirrita aut! This study was conducted in the central part of umnata "Bkh[# "Lepidoptera\ Geometridae# were able Finnish Lapland at 09 outbreak sites and 01 low den! to alleviate the DIR response of birch by consuming\ sity sites of about 0 ha each "Fig[ 0#[ The sites were simultaneously with defoliation\ the primordial apical characterized by mixed Norway spruce ðPicea abies buds developing for the next year|s growth[ Accord! "L[# KarstenŁ\ Scots pine "Pinus sylvestris L[# and birch ingly\ experimental studies must be accompanied by forest[ The area belongs to the interzone of two _eld observations to estimate the total e}ect of di}er! hybridizing subspecies of white birch\ Betula pubes! ent factors on insect populations[ Understanding the cens ssp[ czerepanovii and pubescens "Ehrh[# "Kallio\ contribution of di}erent factors is increasingly impor! Niemi + Sulkinoja 0872#[ The ground layer of the sites tant as the climate change may render wider areas was typically covered by the dwarf shrubs Vaccinium susceptible to outbreaks of E[ autumnata "RuohomaÃki myrtillus "L[#\ V[ vitis!idaea "L[#\ V[ uliginosum "L[# et al[ 0886^ Virtanen\ Neuvonen + Nikula 0887# and and Empetrum nigrum "L[#[ Other deciduous woody other species[ plants\ dwarf birch ðB[ nana "L[#Ł\ and Salix spp[\ were The present study used two complementary present occasionally[ methods to evaluate the importance of DIR and other food plant!related traits on the performance of E[ ESTIMATION OF LARVAL DENSITY AND autumnata during a natural population peak\ during Þ 0888 British DAMAGE TO FOLIAGE Ecological Society the onset of population decline\ and during a low! Journal of Animal density phase in Finnish Lapland between 0881 and A density index "see RuohomaÃki + Haukioja 0881# Ecology\ 57\ 271Ð278 0886[ First\ in 0882Ð86 the sizes were recorded of was used as a quick method to describe larval density 273 Induced tree resistance during insect outbreak

Fig[ 0[ Location of study sites in the outbreak area in northern Finland[

at study sites[ The density index was calculated as the ation degrees for 0881 from the leaf litter samples number of larvae found per 09!min search at each site[ collected in spring 0882[ At least 09 haphazardly selected birch trees\ both short trees and the lowest branches of tall trees\ were WITHIN!SITE EXPERIMENT] EFFECT OF inspected for larvae for up to 29 s by the same person FOLIAGE QUALITY ON PUPAL MASS OF at all sites and in all years[ A majority of larvae were E[ AUTUMNATA at the third instar stage when densities were estimated[ The applicability of this density index in estimating The direct e}ect of defoliation!induced changes in absolute larval densities was assessed at 04 sites\ where foliage quality on E[ autumnata larvae was measured absolute larval densities "number of larvae per amount by rearing larvae on trees at four outbreak sites in of foliage in birches# were also determined[ The results 0883 "sites 0\ 3\ 4 and 09^ Fig[ 0#\ the year following of these density measurements were plotted against the _rst peak year[ At each site\ 29 mature 0=4Ð2=9 m the density index\ and a signi_cant correlation was high trees\ located 4Ð49 m apart\ were selected on 18 found "r  9=50\ P  9=91#[ However\ instead of using May in the previous year\ 0882[ At that time E[ aut! the density index as a continuous variable\ it was used umnata larvae were hatching but had not yet caused to divide the study sites into outbreak and low density signi_cant damage to ~ushing leaves[ At each site\ 04 classes\ because at very high densities the number of randomly assigned trees out of the 29 trees were simultaneously found larvae often exceeded the num! sprayed with insecticide to prevent current!year ber that could be instantly counted[ A minimum of 89 defoliation[ The insecticide used was a pyrethroid larvae:09 min was considered as su.cient for a site to "Folition\ Bayer\ Germany#\ containing fenitrothion have outbreak density\ whereas sites having densities as the killing agent "and xylene as a solvent#[ The less than 49 larvae:09 min had a distinguishably lower spraying concentration used was 4 mL in 09 L water[ density and degree of defoliation[ The other 04 trees were sprayed with water only[ During the peak years 0882 and 0883\ the degree of Defoliation of the insecticide!sprayed trees remained defoliation was visually estimated from 04 trees at six below 4) in the year when the spraying was conduc! outbreak sites[ To obtain defoliation estimates for ted\ and was caused by late!season herbivores[ Hence 0881\ feeding marks in leaf litter were used^ samples the trees fell into two treatment groups] natural defoli! of leaf litter were collected haphazardly from an area ation vs[ insecticide!sprayed control trees[ of a few thousand square meters at six outbreak sites The performance of E[ autumnata larvae was mea! in the springs of 0882Ð84 "09 or more plots mixed in sured in the following year\ 0883\ by rearing 19 larvae one sample each year#[ Thus\ for 0882 and 0883 _gures "belonging to 19 di}erent broods# from egg to pupa were available for the degree of defoliation both in a mesh bag in each tree "for more methodological directly estimated from the trees and taken from the details\ see Haukioja + Neuvonen 0874b#[ On 10 May\ Þ 0888 British Ecological Society leaf litter samples in the next spring "49 leaves: a single 59!cm!long top branch of each tree was Journal of Animal sample#[ These data were used to construct a enclosed in a mesh bag with a similar mixture of Ecology\ 57\ 271Ð278 regression equation to estimate site!speci_c defoli! E[ autumnata eggs[ Parents of the test larvae had been 274 collected as third or fourth instar larvae from both P[ Kaitaniemi et al[ outbreak and low density sites within the area of this study[ An equal proportion of eggs from both origins were used for rearings at each site[ The birch leaves enclosed in the mesh bags were in excess of con! sumption to prevent starvation[ At the end of the larval period\ the mesh bags were checked daily for larvae approaching pupation^ these were removed and allowed to pupate in 37!mL plastic vials "0Ð1 larvae per vial# half!_lled with moist Sphag! num moss[ Pupal mass\ measured a fortnight after pupation\ was used to assess the e}ect of foliage qual! ity on E[ autumnata performance[ A separate study Fig[ 1[ Relationship between pupal mass of male and female E[ autumnata reared on same foliage in identical conditions[ showed that the previous year insecticide spraying per Each dot represents average pupal mass of 4Ð04 males and se had no signi_cant e}ect on the pupal mass "or 4Ð04 females[ survival# of larvae on sprayed trees in 0883 "Kai! taniemi et al[ 0886a#[ to use the size of males trapped at the study sites to estimate the corresponding size and maximum fec! BETWEEN!SITE EXPERIMENT] ADULT SIZE OF undity of females at the same sites\ and to evaluate E[ AUTUMNATA DURING AND AFTER THE the possible reduction in fecundity of females at the OUTBREAK outbreak sites[ To do this\ male femur length "in mm# was _rst transformed into male pupal mass "in mg# Sizes of adult E[ autumnata males were used to esti! by a regression equation based on laboratory!reared mate the e}ect of foliage quality on wild larvae[ Pher! ] male pupal mass  35=9 × "male femur omone traps with synthesized E[ autumnata pher! length# − 30=0 "R1  9=60\ N  16#[ The resulting omone "Zhu et al[ 0884# were used to catch male male pupal mass was transformed into female pupal moths[ The trappings were carried out in the autumns mass by the equation given in Fig[ 1\ and female pupal of 0882Ð86\ scheduling the trapping period each year mass was transformed into egg number by a regression to match the estimated peak of the ~ight period[ For equation taken from Tammaru et al[ "0885#] egg num! practical reasons\ a slightly di}erent combination of ber of a female  1=82 × "female pupal mass# − 090=8 sites "Fig[ 0# was used in di}erent years[ The outbreak "R1  9=66\ N  15#[ sites 0\ 3\ 4\ 5 and 09 were sampled in all years and To elucidate the typical low!density phase fecundity the additional outbreak sites were] 1 and 7 in 0882^ 1 of E[ autumnata\ the average egg number of E[ aut! in 0883^ and 1\ 6\ 7 and 8 in all other years[ The low umnata females originating from _ve low density density sites were] 00\ 04Ð08\ 10 and 11 in 0882^ 00 phase sites\ some 149 km north of the outbreak area and 11 in 0883^ 00Ð04\ 07Ð19 and 11 in both 0884 and "Table 0#\ was measured[ These sites have not experi! 0885^ 01Ð03 and 07 in 0886[ enced outbreaks during the last 29 years[ All _ve data To assess the magnitude of normal year!to!year sets\ including three di}erent years\ yielded fairly simi! variation in male size\ which could confound the potential e}ects of host plant quality\ pheromone trappings were conducted in 0884Ð86 also at eight 1 sites in an area of 049 km in south!western Finland Table 0[ Average egg numbers of _eld!grown Epirrita aut! near Turku[ This area has no E[ autumnata outbreaks\ umnata females originating from low!density!phase popu! and is located several hundred kilometres away from lations in Finnish Lapland the current outbreak area[ Egg number The wings and body were often damaged in the Origin N "84) CL# pheromone traps\ so the hind!leg femur length of the males was used as a measure of size[ The femur lengths Wild adult females collected from 02 027 "004Ð050# of samples of 29 males from each site were measured a _eld site in 0882 Copulating females collected from 06 024 "013Ð035# by microscope to the nearest 9=92 mm[ a _eld site in 0882 Wild larvae collected from a _eld 07 008 "099Ð027# site 1 days before pupation in 0881\ fed with host tree leaves CONVERSION OF MALE SIZE TO FEMALE Wild larvae collected from a _eld 11 017 "009Ð035# FECUNDITY site 1 days before pupation in Þ 0888 British 0881\ fed with host tree leaves E[ autumnata Ecological Society The pupal masses of males and females Singly grown larvae from mesh bag 02 015 "095Ð035# Journal of Animal are similarly a}ected by environmental conditions rearings in 0880 Ecology\ 57\ 271Ð278 "Fig[ 1\ see also Tammaru 0887#[ It was thus possible 275 lar egg numbers\ and averaged 018 eggs with 84) index was 045 larvae found:09 min at outbreak sites\ Induced tree con_dence limits being 008Ð028 eggs[ and 14 larvae:09 min at low density sites[ In 0883 the resistance during values were 79 and 21 larvae:09 min\ respectively[ In insect outbreak 0884 densities had clearly declined at both outbreak STATISTICAL ANALYSES "07 larvae:09 min# and at low!density sites "4 larvae: The data were analysed using the SAS statistical pack! 09 min#\ and remained below 3 larvae:09 min in both age[ Because of missing cells in the data\ Proc Mixed 0885 and 0886[ was used to test the e}ect of larval density on size of pheromone!trapped males "Littell et al[ 0885#[ In the EFFECTS OF PREVIOUS!YEAR DEFOLIATION ON analyses\ year and density were considered as _xed PUPAL MASS OF E[ AUTUMNATA e}ects\ and the tests were based on the _rst!order autoregressive covariance structure of data "Littell At three out of four outbreak sites "3\ 4 and 09#\ the et al[ 0885#[ pupal mass of E[ autumnata was lower in previously In the within!site experiment\ the e}ect of treatment defoliated trees than in insecticide!sprayed trees\ on pupal mass of E[ autumnata was analysed using an although signi_cantly so only at site 3 "Table 1#[ The analysis of variance "ANOVA#[ One water!sprayed and true di}erences may be slightly underestimated\ two insecticide!sprayed trees at site 09 were excluded because defoliation occurring in 0881 could have because they had been damaged by a by!passing log! a}ected foliage quality of insecticide!sprayed trees[ ging machine[ As a result of the low number of sites However\ observations suggest that the DIR response in the experiment\ the nested ANOVA was used as sug! loses most of its e.cacy by the second summer after gested by McKone + Lively "0882# for experiments defoliation "Kaitaniemi\ Neuvonen + NyyssoÃnen conducted at multiple\ but few randomly chosen sites 0888#\ implying that\ if DIR was present\ it should "i[e[ sites that represent a sample from a larger popu! have been weak in the insecticide!sprayed trees[ There! lation of sites#[ This method allowed the e}ect of the fore\ trees with a {strong| DIR may have been com! treatment itself to be better detected[ In this method\ pared with trees with a {weak| DIR[ the e}ect of treatment was tested separately for each At the fourth site "0# pupal mass was 54 mg in both site by nesting the treatment within each site[ Nor! defoliated and sprayed trees "reported in Kaitaniemi mality of the data and homoscedasticity of the vari! et al[ 0886a#\ possibly as a result of the ameliorating ances were con_rmed with ShapiroÐWilk|s tests and e}ect on foliage quality of bud consumption by larvae Cochran|s tests\ respectively[ at this site "see Kaitaniemi et al[ 0886a#[ Bud con! sumption was not observed in the other outbreak sites[ Results ADULT SIZE OF WILD E[ AUTUMNATA THE COURSE OF THE OUTBREAK Measurements of the femur length of pheromone! High larval densities were observed in the study area trapped males gave a similar picture of the e}ect of for the _rst time in the summer of 0881 "reported by larval growth conditions] males were signi_cantly

local newspapers#^ the estimates of defoliation degree\ smaller "F0\04  6=05\ P  9=91# at outbreak than at based on feeding marks in leaf litter\ likewise sug! low!density sites during all the study years "Table 2#[ gested a relatively high degree of defoliation already The smallest size was observed in 0883\ the year after in 0881 "Fig[ 2#[ The highest larval densities were peak density\ and it corresponded to females having observed in 0882 when the average of larval density 61 mg pupal mass or 097 eggs[ The di}erences between

years were also signi_cant "F3\30  8=95\ P  9=9990#] the largest size was achieved at low density sites 1 years after the outbreak in 0885\ corresponding to 67 mg female pupal mass or 017 eggs[ This maximum value is very close to values observed during the low density phase before the outbreak years "Table 0#[ However\ DIR was not strong enough to cause

a signi_cant {density × year| interaction "F3\30  9=83\ P  9=34#\ which would have resulted from a par! ticularly large size reduction after the peak year[ The e}ect of DIR may have been partly confounded by a natural size di}erence between outbreak and low density sites] during 0885 and 0886 there still was a 9=91 mm di}erence\ although otherwise the year!to! Þ 0888 British Ecological Society Fig[ 2[ Correspondence between defoliation estimates based year _gures seemed to be settling[ An alternative Journal of Animal on leaf litter and those obtained directly from trees[ Numbers explanation could be that DIR was still active\ but Ecology\ 57\ 271Ð278 indicate study sites according to Fig[ 0[ this may be unlikely because the e}ect of DIR has 276 Table 1[ ANOVA table for di}erences in the pupal mass of Epirrita autumnata reared on birch which were either defoliated by P[ Kaitaniemi et al[ larvae or protected from defoliation by an insecticide in the previous year[ Hierarchical ANOVA for three sites "3\ 4\ 09# according to McKone + Lively "0882#

Pupal mass$ 2 SE ANOVA

Source "site# Insecticide Defoliation d[f[ MS FP

Def"3# 69=4 2 1=7 58=1 2 1=4 0 01=36 9=09 9=640 Def"09# 63=4 2 2=5 57=7 2 2=9 0 102=96 0=62 9=081 Def"4# 62=2 2 2=0 54=9 2 1=4 0 381=74 2=88 9=938 Site 1 34=64 9=26 9=580 Error 79 012=30

$ Note that these values have been corrected for the e}ect of sex\ and represent intermediate values compared to true male or female pupal mass[

Table 2[ Durations of pheromone trappings\ average femur lengths of Epirrita autumnata males\ 84) con_dence limits "CL#\ and average numbers of males measured "N# at each site] "a# in outbreak and low density sites within an outbreak area in northern Finland "0882\ 0883 were outbreak years#^ and "b# in low density area in south!western Finland "outside outbreak area#

Outbreak Low density

Year Trapping period$ Femur "mm# 84) CL N Femur "mm# 84) CL N

"a# Outbreak area 0882 11 AugÐ97 Sep 1=39 1=26Ð1=32 29 1=34 1=30Ð1=37 29 0883 20 AugÐ98 Sep 1=27 1=23Ð1=30 29 1=33 1=28Ð1=38 29 0884 12 AugÐ93 Sep 1=35 1=32Ð1=37 13 1=36 1=33Ð1=38 13 0885 14 AugÐ09 Sep 1=38 1=34Ð1=40 13 1=40 1=38Ð1=43 11 0886 12 AugÐ97 Sep 1=37 1=34Ð1=40 12 1=49 1=35Ð1=43 20

"b# Low density area 0884 03 SepÐ17 Sep 1=48 1=47Ð1=50 16 0885 08 SepÐ96 Oct 1=48 1=46Ð1=50 29 0886 07 SepÐ09 Oct 1=48 1=44Ð1=51 17

$ The actual periods may have been shorter because of saturation of traps by the moth[

been found largely to disappear within 1 years after erally defoliation degrees approaching 49) of the even severe manual defoliation "Kaitaniemi et al[ 0888#[ canopy have been found to be su.cient "Haukioja + In the area with no outbreaks in south!western Fin! Neuvonen 0874b#[ land there was no detectable year!to!year size vari! Both the approaches adopted by the present study\ ation\ and the average femur length was 1=48 mm in rearing larvae on trees in the _eld and monitoring the all 2 years "Table 2#\ suggesting that the average size size of adults in the _eld\ gave similar results regarding remains largely constant during the low density phase[ the suitability of foliage and other growth conditions for E[ autumnata at outbreak sites[ They suggested that quality of foliage decreased after defoliation*a Discussion result that was also re~ected in the biochemical traits This study was the _rst to assess the role of foliage of leaves "Kaitaniemi et al[ 0887#[ However\ the quality of B[ pubescens during a natural E[ autumnata reduction of adult size at outbreak sites was minor outbreak\ and one of few _eld studies assessing food compared to size at low density sites\ trees or years[ quality during an outbreak of any forest lepidopteran[ During and after the outbreak\ the estimated fec! Previous studies in our system had suggested that undity of females was maximally only 12 eggs "07)# DIR can have a strong e}ect on the performance below the average 018 eggs observed in the low density of E[ autumnata in the year following experimental phase[ The observed size reduction also had no conse! defoliation\ although the e.cacy of DIR has\ in quences for the ability of males to fertilize the eggs of general\ varied considerably among individual years females "Haukioja + Neuvonen 0874a#[ Þ 0888 British Ecological Society and experiments "RuohomaÃki et al[ 0881#[ Exper! Consequently\ the observed reduction in fecundity Journal of Animal imentally induced larval defoliation\ in particular\ has resulting from previous defoliation was presumably Ecology\ 57\ 271Ð278 seemed to induce a strong DIR response\ and gen! not the reason why the present peak in density ter! 277 minated in 0883[ This suggests other causes for the of Finnish Lapland[ Delayed induced resistance was Induced tree low survival rate of E[ autumnata in 0883[ That e}ect not e.cient enough to be the main reason for the resistance during may have been a general one\ because populations at termination of the outbreak\ and it seemed unlikely insect outbreak both high and low density sites decreased simul! that presence of DIR was required at all to terminate taneously[ The e}ects of parasitism will be treated the outbreak[ Whether this was a characteristic of the separately in the future "K[ RuohomaÃki + P[ Kait! 0889s outbreak alone remains unclear\ but Bylund|s aniemi\ unpublished information#\ but it is worth "0884# work with E[ autumnata and Baltensweiler|s mentioning here that the parasitism rates of larvae in "0882# work with Zeiraphera diniana indicate that 0882 "RuohomaÃki et al[ 0885# and in 0883 were not di}erent terminating agents may be important for very high "below 49) at all sites^ this does not include di}erent individual outbreak peaks\ even within the parasitism that occurred after the third larval instar#[ same system[ Even if plant!related traits are important Neither were any signs of disease observed\ which for the outbreak\ the weather or other environmental is in accordance with Bylund|s "0884# observations\ conditions may largely modify their extent and intro! suggesting that diseases seldom infect E[ autumnata duce variation into the behaviour of the system[ during outbreaks[ As discussed below\ this leaves open two possible reasons related to food quality] the poor foliage quality in 0883 as a result of to the cold summer Acknowledgements of 0882\ and the poor synchrony of larval hatch and We thank Erkki KorpimaÃki\ Seppo Neuvonen and birch budbreak in the spring of 0883[ Tero Klemola for their comments on earlier versions In the spring after a cold summer\ the mountain of this manuscript[ Susanna Haapala\ Lauri Kapari\ birch leaves remain smaller than normal\ and their Tero Klemola and Sari Mikkonen assisted with the quality as food for the larvae of E[ autumnata "as well _eld and laboratory work[ Miia TanhuanpaÃaà helped as for several saw~y species# is lower than after a with the statistical analyses[ Special thanks to Ellen warm summer "Senn\ HanhimaÃki + Haukioja 0881^ Valle for checking our English[ This study was _nanced HanhimaÃki\ Senn + Haukioja 0884#[ During the pre! by the Academy of Finland "e[g[ grant no[ 27314#\ and sent study\ summer temperatures were close to aver! by the Jenny and Antti Wihuri Foundation[ age\ except for the summer of 0882 "Kuukausikatsaus Suomen ilmastoon\ JuneÐSeptember 0882Ð84#\ which was cold and may thus have contributed to the low References foliage quality for E[ autumnata in 0883[ However\ the di}erences in adult size between 0883 and other years Ayres\ M[ "0882# Plant defense\ herbivory\ and climate change[ Biotic Interactions and Global Change "eds P[ M[ do not indicate any strong e}ects of weather[ Kareiva\ J[ G[ Kingsolver + R[ B[ Huey#\ pp[ 64Ð83[ A more likely explanation is asynchrony between Sinauer Associates\ Sunderland\ MA[ phenologies of larvae and birches in spring 0883[ Ayres\ M[P[ + MacLean\S[F[ Jr "0876# Development of birch Di}erences in average annual temperature and in the leaves and the growth energetics of Epirrita autumnata temporal accumulation of the temperature sum may "Geometridae#[ Ecology\ 57\ 447Ð457[ Baltensweiler\ W[ "0882# Why the larch bud!moth cycle col! a}ect the developmental rates of leaves and larvae lapsed in the subalpine larch!cembran pine forests in the di}erently\ thus either fostering or deterring devel! year 0889 for the _rst time since 0749[ Oecologia\ 82\ 51Ð55[ opmental synchrony between larvae and leaves "Ayres Baltensweiler\ W[ + Fischlin\ A[ "0877# The larch budmoth in 0882^ but see Neuvonen et al[ 0885#[ Field obser! the Alps[ Dynamics of Forest Insect Populations[ Patterns\ vations in the present study indicated poor synchrony Causes\ and Implications "ed[ A[ A[ Berryman#\ pp[ 220Ð 240[ Plenum Press\ New York[ of larval hatching with lea_ng in 0883 "Kaitaniemi Berryman\ A[A[ "0885# What causes population cycles of et al[ 0886b#\ but its real e}ects are hard to prove[ forest Lepidoptera< Trends in Ecology and Evolution\ 00\ Trees at four outbreak sites were inspected at the 17Ð21[ same phenological stage in both 0882 and 0883^ in the Bylund\ H[ "0884# Long!term interactions between the autum! former year numerous larvae were present in all trees nal moth and mountain birch] the roles of resources\ com! petitors\ natural enemies\ and weather[ PhD thesis\ Swedish at the time of inspection\ but in the latter year larvae University of Agricultural Sciences\ Uppsala[ were observed only occasionally\ even at site no[ 0\ Carter\ M[R[\ Ravlin\ F[W[ + McManus\ M[L[ "0880# Chan! which had its peak density in 0883[ Delayed hatching ges in gypsy moth "Lepidoptera] Lymantriidae# fecundity in relation to leaf development results in poor per! and male wing length resulting from defoliation[ Environ! formance and small size of E[ autumnata "Ayres + mental Entomology\ 19\ 0931Ð0936[ Dussourd\ E[E[ + Eisner\ T[T[ "0876# Vein!cutting behavior] et al MacLean 0876^ Kaitaniemi [ 0886b#[ Further! insect counterploy to the latex defense of plants[ Science\ more\ asynchrony may have increased larval para! 126\ 787Ð890[ sitism in 0883\ because it prolongs the period during HanhimaÃki\ S[\ Senn\ J[ + Haukioja\ E[ "0884# The con! which the larvae are vulnerable to parasitoids "Kai! vergence in growth of foliage chewing insect species on individual mountain birch trees[ Journal of Animal Ecol! taniemi + RushomaÃki 0888#[ Þ 0888 British ogy\ 53\ 432Ð441[ Ecological Society In conclusion\ the study reported here indicated Haukioja\ E[ "0879# On the role of plant defences in the Journal of Animal only a weak DIR response in birches during and after ~uctuation of herbivore populations[ Oikos\ 24\ 191Ð102[ Ecology\ 57\ 271Ð278 the recent outbreak of E[ autumnata in the central part Haukioja\ E[ "0871# Inducible defences of white birch to a 278 geometrid defoliator\ Epirrita autumnata[ Proceedings of lation cycles of forest Lepidoptera[ Advances in Ecological P[ Kaitaniemi et al[ the 4th International Symposium on Insect!Plant Relation! Research\ 07\ 068Ð131[ ships\ Wageningen\ The Netherlands\ pp[ 088Ð193[ Neuvonen\ S[ + Haukioja\ E[ "0880# The e}ects of inducible Haukioja\ E[ "0880# Induction of defenses in trees[ Annual resistance in host foliage on birch!feeding insects[ Phyto! Review of Entomology\ 25\ 14Ð31[ chemical Induction by Herbivores "eds D[ Tallamy + M[ Haukioja\ E[ + Neuvonen\ S[ "0874a# The relationship Raupp#\ pp[ 166Ð180[ John Wiley\ New York[ between size and reproductive potential in male and female Neuvonen\ S[\ Virtanen\ T[\ Nikula\ A[ + Varama\ M[ "0885# Epirrita autumnata "Lep[\ Geometridae#[ Ecological Ento! Climate change and the risk of forest insect outbreaks[ mology\ 09\ 156Ð169[ The Finnish Research Programme on Climate Change[ Final Haukioja\ E[ + Neuvonen\ S[ "0874b# Induced long!term Report "ed[ J[ Roos#\ pp[ 164Ð179[ Publications of the Acad! resistance of birch foliage against defoliators] defensive or emy of Finland 3:85[ Edita Ltd[\ Helsinki[ incidental[ Ecology\ 55\ 0292Ð0297[ Roland\ J[ + Myers\ J[H[ "0876# Improved insect per! Haukioja\ E[ + Neuvonen\ S[ "0876# Insect population formance from host!plant defoliation] winter moth on oak dynamics and induction of plant resistance] the testing and apple[ Ecological Entomology\ 01\ 398Ð303[ of hypotheses[ Insect Outbreaks "eds P[ Barbosa + J[ C[ RuohomaÃki\ K[ + Haukioja\ E[ "0881# Interpopulation Schultz#\ pp[ 300Ð321[ Academic Press\ San Diego[ di}erences in pupal size and fecundity are not associated Haukioja\ E[\ Neuvonen\ S[\ HanhimaÃki\ S[ + NiemelaÃ\P[ with occurrence of outbreaks in Epirrita autumnata "Lep[\ "0877a# The in Fennoscandia[ Dynamics Geometridae#[ Ecological Entomology\ 06\ 58Ð64[ of Forest Insect Populations[ Patterns\ Causes\ and Impli! RuohomaÃki\ K[\ HanhimaÃki\ S[\ Haukioja\ E[\ Iso!Iivari\ L[\ cations "ed[ A[ A[ Berryman#\ pp[ 052Ð067[ Plenum Press\ Neuvonen\ S[\ NiemelaÃ\ P[ + Suomela\ J[ "0881# Varia! New York[ bility in the e.cacy of delayed inducible resistance in Haukioja\ E[\ Pakarinen\ E[\ NiemelaÃ\ P[ + Iso!Iivari\ L[ mountain birch[ Entomologia Experimentalis et Applicata\ "0877b# Crowding!triggered phenotypic responses allevi! 51\ 096Ð004[ ate consequences of crowding in Epirrita autumnata "Lep[\ RuohomaÃki\ K[\ Kaitaniemi\ P[\ Kozlov\ M[\ Tammaru\ T[ + Geometridae#[ Oecologia\ 64\ 438Ð447[ Haukioja\ E[ "0885# Density and performance of Epirrita Kaitaniemi\ P[ + RuohomaÃki\ K[ "0888# E}ects of autumn autumnata "Lep[\ Geometridae# along three air pollution temperature and oviposition date on timing of larval devel! gradients in northern Europe[ Journal of Applied Ecology\ opment and risk of parasitism in a spring folivore[ Oikos\ 22\ 662Ð674[ 73\ in press[ RuohomaÃki\ K[\ Virtanen\ T[\ Kaitaniemi\ P[ + Tammaru\ Kaitaniemi\ P[\ RuohomaÃki\ K[ + Haukioja\ E[ "0886a# Con! T[ "0886# Old mountain birch forests at high altitudes are sumption of apical buds as a mechanism of alleviating host prone to Epirrita autumnata "Lep[\ Geometridae# plant resistance for Epirrita autumnata larvae[ Oikos\ 67\ outbreaks[ Environmental Entomology\ 15\ 0985Ð0093[ 129Ð127[ Sagers\ C[L[ "0881# Manipulation of host plant quality] her! Functional Ecology 5 Kaitaniemi\ P[\ RuohomaÃki\ K[ + Haukioja\ E[ "0886b# bivores keep leaves in the dark[ \ \ Consequences of defoliation on phenological interaction 630Ð632[ between Epirrita autumnata and its host plant\ mountain Senn\ J[\ HanhimaÃki\ S[ + Haukioja\ E[ "0881# Among!tree birch[ Functional Ecology\ 00\ 088Ð197[ variation in leaf phenology and morphology and its cor! relation with insect performance in the mountain birch[ Kaitaniemi\ P[\ RuohomaÃki\ K[\ Ossipov\ V[\ Haukioja\ E[ + Pihlaja\ K[ "0887# Delayed induced changes in the bio! Oikos\ 52\ 104Ð111[ chemical composition of host plant leaves during an insect Sork\ V[L[\ Stowe\ K[A[ + Hochwender\ C[ "0882# Evidence outbreak[ Oecologia\ 005\ 071Ð089[ for local adaptation in closely adjacent subpopulations of northern red oak "Quercus rubra L[# expressed as resistance Kaitaniemi\ P[\ Neuvonen\ S[ + NyyssoÃnen\ T[ "0888# E}ects to leaf herbivores[ American Naturalist\ 031\ 817Ð825[ of cumulative defoliations on growth\ reproduction\ and Tallamy\ D[W[ "0874# Squash beetle feeding behavior] an insect resistance in mountain birch[ Ecology\ in press[ adaptation against induced cucurbit defenses[ Ecology\ 55\ Kallio\ P[ + Lehtonen\ J[ "0862# Birch forest damage caused 0463Ð0468[ by Oporinia autumnata "Bkh[# in 0854Ð55 in Utsjoki\ N Tallamy\ D[ + Raupp\ M[\ eds[ "0880# Phytochemical Induc! Finland[ Reports from the Kevo Subarctic Research tion by Herbivores[ John Wiley\ New York[ Station\ 09\ 44Ð58[ Tammaru\ T[ "0887# Determination of adult size in a foli! Kallio\ P[\ Niemi\ S[ + Sulkinoja\ M[ "0872# The Fen! vorous moth] constraints at instar level< Ecological Ento! noscandian birch and its evolution in the marginal forest mology\ 12\ 79Ð78[ zone[ Nordicana\ 36\ 090Ð009[ Tammaru\ T[\ Kaitaniemi\ P[ + RuohomaÃki\ K[ "0885# Real! Karban\ R[ + Baldwin\ I[T[ "0886# Induced Responses to ized fecundity in Epirrita autumnata "Lepidoptera\ Geom! Herbivory[ The University of Chigaco Press\ Chicago[ etridae#*relation to body size and population dynamics Karban\ R[ + Myers\ J[H[ "0878# Induced plant responses consequences[ Oikos\ 66\ 396Ð305[ to herbivory[ Annual Review of Ecology and Systematics\ Tenow\ O[ "0861# The outbreaks of Oporinia autumnata Bkh[ 19\ 220Ð237[ + Operophtera spp[ "Lep[\ Geometridae# in the Scan! Kuukausikatsaus Suomen ilmastoon "in Finnish#[ "0889Ð84# dinavian mountain chain and northern Finland 0751Ð JuneÐSeptember "published monthly#[ Finnish Meteoro! 0857[ Zoologiska Bidrag FraÄn Uppsala\ Suppl[ 1\ 0Ð096[ logical Institute\ Helsinki\ Finland[ Virtanen\ T[\ Neuvonen\ S[ + Nikula\ A[ "0887# Modelling Littell\ R[C[\ Milliken\ G[A[\ Stroup\ W[W[ + Wol_nger\ topoclimatic patterns of egg mortality of Epirrita aut! R[D[ "0885# SAS System for Mixed Models\ SAS Institute umnata "Lep] Geometridae# with Geographical Infor! Inc[\ Cary\ NC[ mation System] predictions for current climate and McKone\ M[J[ + Lively\ C[M[ "0882# Statistical analysis of warmer climate scenarios[ Journal of Applied Ecology\ 24\ experiments conducted at multiple sites[ Oikos\ 56\ 073Ð 200Ð211[ 075[ Zhu\ J[W[\ LoÃfstedt\ C[\ Philipp\ P[\ Francke\ W[\ Tammaru\ Mason\ R[R[\ Beckwith\ R[C[ + Paul\ H[G[ "0866# Fecundity T[ + Haukioja\ E[ "0884# A sex pheromone novel to the Þ 0888 British reduction during collapse of a Douglas!_r tussock moth Geometridae identi_ed from Epirrita autumnata[ Ento! Ecological Society outbreak in northeast Oregon[ Environmental Entomology\ mologia Experimentalis et Applicata\ 64\ 048Ð053[ Journal of Animal 5\ 512Ð515[ Ecology\ 57\ 271Ð278 Myers\ J[H[ "0877# Can a general hypothesis explain popu! Received 11 December 0886^ revision received 2 July 0887