A petiole-galling insect herbivore decelerates leaf lamina litter decomposition rates

A petiole-galling insect herbivore decelerates leaf

lamina litter decomposition rates

Christopher J. Frost†,1,2,3,4, Jennifer M. Dean1,4, Erica C. Smyers4, Mark C. Mescher1,4, John

E. Carlson1,2,3,5, Consuelo M. De Moraes1,4 and John F. Tooker*,1,4

1Center for Chemical Ecology, Pennsylvania State University, University Park, Pennsylvania 16802, USA; 2School of

Forest Resources, Pennsylvania State University, University Park, Pennsylvania 16802, USA; 3Schatz Center for Tree

Molecular Genetics, Pennsylvania State University, University Park, Pennsylvania 16802, USA; 4Department of Entomol-

ogy, Pennsylvania State University, University Park, Pennsylvania 16802, USA; and 5Department of Bioenergy Science

and Technology (WCU), Chonnam National University, 333 Yongbongro, Buk-gu, Gwangju 500-757, Korea

Summary

1. Herbivore-mediated changes in leaf-litter chemistry are often considered responsible for alter-

ing litter decomposition rates, but such chemical changes often co-occur with other factors such

as physical alteration of leaf material that also influence decomposition rates. We attempted to

disentangle these effects using the poplar petiole gall moth (Ectoedemia populella Brusk), which

forms galls on petioles at the base of the leaf lamina but does not alter leaf morphology. Thus,

differences in leaf decomposition rates between galled and ungalled leaves should be explained

by gall-mediated changes in leaf chemistry.

2. Petiole galling decelerated leaf lamina litter decomposition in two Populus host species, but in

temporally distinct ways. In Populus granidentata, galling decelerated decomposition by 7%

after 4 months. After 12 and 18 months, Populus tremuloides litter decomposition rates were

12% and 17% lower, respectively, in lamina tissue whose petiole had been galled relative to ung-

alled. On average, the petiole galler increased leaf lamina nitrogen concentrations by 17%,

decreased tannin concentrations from 37% to 53% and decreased tannin-binding capacity by

11% and 37% in P. grandidentata and P. tremuloides, respectively. These changes would be

expected to increase, rather than decrease, decomposition rates.

3. Unlike other insect herbivores guilds that have variable effects on litter decomposition in

direction and magnitude, all gall insects studied to date have decelerated leaf-litter decomposi-

tion. This consistent effect of galling on decomposition provides a framework for deciphering a

fundamental aspect of insect herbivory on a critical ecosystem process.

4. We used a gall-inducing moth with a distinctive natural history to confirm the role of herbi-

vore-mediated litter chemistry in leaf-litter decomposition dynamics. Moreover, we advance the

hypothesis that gall-induced defensive manipulations that protect a host plant from injury by

other herbivores lead to decelerated litter decomposition.

Key-words: Ectoedemia, leaf chemistry, leaf-litter decomposition, plant–herbivore interactions,

Populus

Introduction

The process of decomposing dead plant material facilitates

the recycling of mineral nutrients and organic matter essen-

tial for biological activity in most terrestrial ecosystems

(Parton et al. 2007). Senesced leaf litter is an abundant,

ubiquitous example of such material, and variation in leaf-

litter quality as a substrate can have demonstrable effects

on decomposition processes and thus terrestrial nutrient

availability. As many plant species presumably increase

their Darwinian fitness by altering foliar quality in response

to herbivores (Karban & Baldwin 1997), the potential for

herbivores to indirectly influence litter quality – and thus

decomposition rates – has long been considered plausible

(Choudhury 1988). While a number of studies have shown

clear effects of herbivores on litter decomposition (Findlay

et al. 1996; Belovsky & Slade 2000; Chapman et al. 2003;

Schweitzer et al. 2005b; Chapman, Schweitzer & Whitham

*Correspondence author. E-mail: [email protected]†Present address. Warnell School of Forest Resources, University of

Georgia, Athens, Georgia 30601, USA.

� 2012 The Authors. Functional Ecology � 2012 British Ecological Society

Functional Ecology 2012 doi: 10.1111/j.1365-2435.2012.01986.x

2006; Crutsinger et al. 2008; Frost & Hunter 2008; Kurokawa

& Nakashizuka 2008), constructing a more detailed theoret-

ical framework has been difficult because herbivory can

accelerate or decelerate decomposition depending in part on

whether secondary metabolites are repressed or augmented,

what type of physical damage occurs, the type of herbivory

suffered or the ecosystem in which the decomposition is

measured. Moreover, determining the importance of her-

bivory-induced changes in leaf chemistry for decomposition

rates is often complicated by the co-occurrence of other

herbivore-related factors that can also alter litter decompo-

sition. For example, herbivory may induce changes in leaf-

drop phenology that results in herbivore-affected litter

entering the detrital system in a different condition than

undamaged litter (Chapman et al. 2003). Of equal impor-

tance, most foliar herbivores physically damage leaf tissue

during feeding, and such damage can alter access to the lit-

ter substrate (Findlay et al. 1996; Cornelissen et al. 1999;

Perez-Harguindeguy et al. 2000; Ostertag, Scatena & Silver

2003). Insect herbivore species that influence host-plant

chemistry but only minimally alter other aspects of a host

plant, such as some species of gall insects, offer an ideal

opportunity to explore the effects of herbivore-induced leaf

chemistry on leaf-litter decomposition apart from other

co-occurring effects.

Gall-inducing insects are herbivores that force their host

plants to produce a tumour-like growth that provides the

insect with food and shelter, usually at the expense of plant

growth and ⁄or reproduction. These insects have evolved inti-

mate relationships with their host plants and an unparalleled

ability to influence host-plant morphology and physiology

(Larson 1998; Stone & Schonrogge 2003). Gall insects com-

monly modify host-plant chemistry, often altering concentra-

tions of plant secondary metabolites for their own purposes

(Weis & Abrahamson 1986; Nyman & Julkunen-Tiitto 2000;

Tooker, Koenig & Hanks 2002; Allison & Schultz 2005). For

example, high concentrations of secondary metabolites such

as phenolics tend to be localized in gall exteriors, where they

presumably provide protection against natural enemies, while

the inner nutritive tissue on which the gall insect feeds is lar-

gely or entirely devoid of such potentially toxic compounds

(Nyman & Julkunen-Tiitto 2000; Allison & Schultz 2005).

Moreover, gall insects also variably influence the chemistry of

plant tissue that is not part of the gall itself, although investi-

gations thus far have been confined to neighbouring tissues

(Tooker et al. 2008; Cooper & Rieske 2009); such systemic

effects on chemistry may also influence decomposition

dynamics of those tissues.

A potential consequence of gall insect-mediated alterations

in host-plant chemistry is the modification of rates of decom-

position of senesced or dead plant parts. Such effects have

been demonstrated using the aphid Pemphigus betae and the

midge Rhopalomyia solidaginis, two disparate gall-inducing

taxa; these insect species induced changes that altered rates of

leaf-litter decomposition (Schweitzer et al. 2005b; Crutsinger

et al. 2008). Yet, both of these species alter leaf morphology

to some degree by forming their gall on the leaf lamina

(P. betae) or by radically altering leaf growth patterns

(R. solidaginis) (Fig. S1, Supporting information). As a

result, these systems have shown clearly that gall insects have

ecosystem-level effects, but neither system necessarily disen-

tangles chemical and morphological factors influencing litter

decomposition.

Here, we describe how herbivory by larvae of the poplar

petiole gall moth (Ectoedemia populella Busck.) influences

leaf-litter chemistry and decomposition in two species of Pop-

ulus. Ectoedemia are monotrysian, nepticulid moths that

include some of the smallest known lepidopterans; adults of

E. populella have c. 6-mm wingspans (Wilkinson & Scoble

1979). The family consists primarily of leaf miners; only a few

Ectoedemia species consume bark, buds, or – in the case of

E. populella – induce galls (Wilkinson & Scoble 1979). Leaf-

mining Ectoedemia trace at least to the mid-Cretaceous era

97 million years ago (Labandeira et al. 1994); gall-inducing

Ectoedemia on Populus trace to the Miocene era, some 5–

17 million years ago (Madler 1936). Importantly,E. populella

form galls at the junction of the petiole and the lamina caus-

ing no observable morphological difference in the lamina tis-

sue itself (Fig. S2, Supporting information). Moreover, the

phenology of the moth is offset with its hosts such that the

adults emerge and oviposit in May after leaves of their hosts

have fully expanded foliage (Wilkinson& Scoble 1979). Thus,

E. populella galls have limited or no influence on lamina

growth and development, although they may influence lam-

ina quality. Larvae emerge and pupate in October and over-

winter in the soil, so larvae do not reside in decomposing

galled leaf litter. Little else is known about the ecology of

E. populella, including the direct effects of the gall on leaf

lamina metabolite profiles associated with the galled petiole.

However, there are at least two reasons to hypothesize that

E. populella will influence decomposition possibly by altering

laminar chemistry. First, the gall envelopes the petiole, forc-

ing all assimilates and other transported materials through

vasculature that has been altered by the gall. Second, E. pop-

ulella deposits its frass inside the gall in a tight packet that

remains in the gall after leaf senescence; frass, which is nutri-

ent rich, is known to influence ecosystem processes in other

systems (Frost & Hunter 2007; Madritch, Donaldson &

Lindroth 2007).

Throughout its range, E. populella forms galls on big-

toothed aspen (Populus grandidentata) and quaking aspen

(Populus tremuloides) (Wilkinson & Scoble 1979). This

makes the Populus ⁄Ectoedemia system well suited to explore

the independent effects of host species and galling. As plant

diversity at different levels can affect decomposition rates

(Madritch & Hunter 2002; Ball et al. 2008) – and the effects

of herbivores on these rates (Schweitzer et al. 2005b) – we

tested the hypothesis that E. populella galling influences leaf-

litter decomposition dynamics using naturally senesced

leaves from P. grandidentata and P. tremuloides separately.

This allowed us to examine variation of E. populella-

mediated changes to leaf-litter decomposition rates, chemis-

try and nutrient release dynamics between host species over

the course of 18 months.

� 2012 The Authors. Functional Ecology � 2012 British Ecological Society, Functional Ecology

2 C. J. Frost et al.

Materials and methods

Leaf litter from P. grandidentata and P. tremuloides was collected

separately from the ground during natural leaf senescence (October

2006) from three separate locations in Centre County, PA and pooled

by plant species. The two aspen species tend to drop their leaves over

a window of generally <10 days (Fig. S3, Supporting information).

We did not collect leaves until they naturally dehisced from the trees

to ensure natural levels of leaf senescence. However, we also used

leaves that were freshly fallen and sitting on top of other leaf litter to

ensure that they had no contact with soil, to minimize any potential

‘preprocessing’ of litters prior to the experiment. We observed no

apparent difference in leaf-drop phenology; we confirmed this by

marking a number of leaves and monitoring their drop phenology,

and there was no difference in leaf-drop timing between galled and

ungalled leaves (Fig. S3, Supporting information). Litter from each

species was sorted based on the presence of galls. Leaves were selected

that had little or no visible damage from chewing herbivores to the

leaf tissue itself, as such damage has been shown to affect decomposi-

tion dynamics in Populus and other woody plant species (Findlay

et al. 1996; Belovsky & Slade 2000; Chapman et al. 2003; Chapman,

Schweitzer & Whitham 2006; Frost & Hunter 2008). However, we

cannot exclude the possibility that piercing ⁄ sucking herbivores such

as aphids may have fed on the foliage during the growing season,

although nonewas observed during sampling.

Litters were dried separately in the laboratory at room tempera-

ture, and then, subsamples were selected haphazardly, weighed and

sealed in 15 · 15 cm screen bags (1 · 1-mm mesh-size) as previously

described (Frost & Hunter 2008). We used c. 3 g of litter for each lit-

terbag. Based on the amount of material available, we filled 55–57 lit-

terbags with each of the P. grandidentata galled and ungalled litters

and P. tremuloides ungalled litters; 20 litterbags were made with

P. tremuloides galled litter owing to a limited amount of material.

Following previous convention (Schweitzer et al. 2005b; Crutsinger

et al. 2008), we removed petioles from all galled and ungalled laminas

prior to filling the litterbags but did not remove galls from the lamina

litter, and the Ectoedemia gall body was c. 8% of the total litter mass

in galled samples. Importantly, different decomposition patterns

between the two Populus species and the magnitude of the observed

effects (see Results) suggest that the differences between the mass of

galled and ungalled litters were not artifacts of the gall body itself.

We then established a field plot in January 2007 in the Scotia

Range near State College, PA (40�47¢0Æ56¢¢N; 77�57¢20Æ56¢¢W). The

site was a mixed-hardwood stand containingmatureP. grandidentata

and P. tremuloides. The field site consisted of 15 replicated plots

(0Æ5 · 2 m per plot). We assigned randomly one litterbag per treat-

ment per collection date to each plot. Thus, fifteen replicates per date

were established for each of the P. grandidentata galled and ungalled

and P. tremuloides ungalled litters. Five replicates per date were

established for the P. tremuloides galled litters. For the initial sam-

ples, there were 9–12 bags forP. grandidentata galled and ungalled lit-

ters and P. tremuloides ungalled litters and five bags for the

P. tremuloides galled litter.

One set of litterbags was collected immediately to measure initial

litter quality, while the remaining litterbags were collected after 4, 12

and 18 months in the field. Litterbags containing initial litter samples

were brought to the field, established at the site and then immediately

collected and returned to the laboratory. During collections, the stak-

ing flag was carefully removed, debris on the outside of each collected

bag was brushed from the surface and care was taken not to disturb

remaining litterbags. Litterbags were transported to the laboratory at

ambient temperatures and then dried at 65 �C for 48 h. Samples were

then weighed, and the lamina litter was separated from the gall body

in the samples that were galled. The lamina litter tissue only was then

ground to a fine powder in a ball mill, and the resulting powder was

analysed for total C (mg g)1), total N (mg g)1), C ⁄N ratios, tannins

(mg g)1) and fibre content (lignin, cellulose, hemicellulose; mg g)1).

Subsamples were ashed (550 �C, 5 h), and data are reported relative

to ash-free dry mass (AFDM). Thus, the chemical analyses represent

only the ungalled lamina portions of the leaf-litter tissue.

Total C and N were measured with a Carlo Erba 1500N total

CHN analyzer (Carlo Erba Instrumentazione, Milan, Italy). Fibre

concentrations were determined by sequential acid digestions using

an ANKOM A200 Fiber Analyzer according to manufacturer’s

instructions (ANKOM Technology, Macedon, NY, USA). Tannin

concentrations and protein-binding capacity were assayed colorimet-

rically following well-established methods (Bate-Smith 1977a,b) after

three rounds of initial extraction with 70 ⁄ 30% acetone ⁄water with1 mM ascorbic acid and subsequent removal of the acetone under vac-

uum evaporation (Frost & Hunter 2008). Condensed tannin concen-

trations were assayed following acid depolymerization in a polar

solvent (N-butanol) at 100 �C (Hagerman & Butler 1980; Hagerman,

Rice & Ritchard 1998), and hydrolyzable tannin concentrations were

measured using the standard potassium iodate assay (Rossiter,

Schultz & Baldwin 1988). We also measured protein-binding capacity

of the leaf tannin extracts using the protein precipitatable phenolics

assay, which provides a realistic assessment of the biological activity

of the tannins. Briefly, protein-binding tannins form a precipitate

complex with the substrate protein bovine serum albumin; the precip-

itate is re-dissolved in triethanolamine ⁄ sodium dodecyl sulphate and

tannins quantified (mg protein-binding tannin per g dry weight) spec-

trophotometrically after adding ferric chloride (Hagerman & Butler

1978). Standards for all phenolic assays were generated from a pooled

mix of undecomposed leaf litter from all the individual experimental

trees that was extracted exhaustively with 70 ⁄ 30% acetone ⁄waterwith 1 mM ascorbic acid similar to the individual samples (Hagerman

& Butler 1989; Hall et al. 2006; Ball et al. 2008; Frost & Hunter

2008). This pooled aqueous extract was then frozen and lyophilized,

and an weighed aliquot of the resulting powder was resuspended in

water and used to generate standard curves for all tannin assays

(Madritch & Hunter 2002, 2004; Hunter & Forkner 2004; Hall et al.

2006; Ball et al. 2008; Frost &Hunter 2008).

Data were analysed using PROC MIXED in SAS (Littell, Stroup &

Freund 2002) using tree species and gall presence as main effects and

plot as a random effect. Data reported on litter AFDM remaining

were arcsine square root transformed, and all other data were log

transformed prior to analysis to satisfy assumptions of ANOVA.

Results

Presence of the petiole gall reduced decomposition rates of

the big-toothed aspen and quaking aspen leaf litters, but in

temporally distinct ways (Fig. 1). Both galling and tree spe-

cies had significant main effects on rates of AFDM loss (spe-

cies F1,163 = 12Æ90, P < 0Æ001; gall F1,163 = 7Æ62,P = 0Æ006); however, only galling influenced AFDM loss

over time (date*gall F3,149 = 3Æ05, P = 0Æ030; date*speciesF3,150 = 2Æ19, P = 0Æ091). Importantly, galling affected

decomposition differently between tree species (date*spe-

cies*gall F3,149 = 2Æ89, P = 0Æ038): galling decelerated the

‘early’ stages of mass loss in P. grandidentata litter and the

‘late’ stages in P. tremuloides litter. After 4 months, AFDM


Galling effects on leaf-litter decomposition 3

loss of the P. grandidentata galled leaf litter was 6% lower

than that of ungalled P. grandidentata litter (F1,26 = 5Æ40,P = 0Æ028); after 12 and 18 months, AFDM loss was 15%

and 17% lower, respectively, in the galledP. tremuloides litter

relative to the ungalled P. tremuloides litter (F1,20 = 4Æ99,P = 0Æ038; F1,18 = 8Æ08, P = 0Æ011; Fig. 1a). There were nosignificant effects of galling on AFDM loss of P. tremuloides

litter at 4 months or P. grandidentata litter at 12 or

18 months. The changes to decomposition mediated by gall-

ing and species altered the decomposition constant k over the

full-time course of the experiment (Fig. 1b).

Galling had strong, species-specific effects on tannin con-

centrations and protein-binding activity in the initial

senesced, undecomposed litter. Condensed tannin concentra-

tions were 46% lower in the P. tremuloides galled litter rela-

tive to ungalled P. tremuloides litter (F1,14 = 12Æ30,P = 0Æ004), while galling had no such effect on P. grandiden-

tata-condensed tannin concentrations (P = 0Æ379; Fig. 2a).

Conversely, hydrolyzable tannins were 37% and 53% lower

in galled P. grandidentata and P. tremuloides litters (gall:

F1,24Æ8 = 19Æ41, P < 0Æ001), respectively, and P. grandiden-

tata litter had an overall greater hydrolysable tannin concen-

tration than did P. tremuloides litter (species: F1,23Æ7 = 55Æ40,P < 0Æ001; Fig. 2b). The protein-binding capacity of the ini-

tial litters was strongly correlated with hydrolysable tannin

concentrations; galled P. grandidentata and P. tremuloides

litters had 11% and 35% lower protein-binding capacity than

did their respective controls (gall: F1,30 = 10Æ13, P = 0Æ003;Fig. 2c). Moreover, the protein-binding capacity of ungalled

P. grandidentata litters was 96% greater than that of ungalled

P. tremuloides litters (species: F1,30 = 155Æ85,P < 0Æ001).Time in the field dramatically reduced litter tannin concen-

trations and protein-binding capacity (Fig. 2). After

Ash

-free

dry

mas

s re

mai

ning

(%)

40

50

60

70

80

90

100

110

Month0 2 4 6 8 10 12 14 16 18 20

40

60

80

100

(a)

(b)

Fig. 1. Mass remaining (% ash-free dry mass) ofPopulus grandiden-

tata L. and Populus tremuloides L. leaf litters that had been galled by

the petiole galler Ectoedemia populella Brusk. In both panels, circles

represent P. grandidentata, triangles represent P. tremuloides sam-

ples; shaded symbols are leaf samples that were ungalled, open sym-

bols are leaf samples that had been galled by E. populella.

(a)Means ± SE of each treatment group, with n = 5 forP. tremulo-

ides galled samples and n = 15 for all others. (b) Regression analysis

to determine the decomposition constant (k: slope of regression line)

for each group. Symbol identity are the same as panel a, and regres-

sion lines vary to distinguish the treatments: dash = P. tremuloides

ungalled; solid = P. tremuloides galled; dot-dash = P. grandidenta-

ta ungalled; dot = P. grandidentata galled.

Months0 2 4 6 8 10 12 14 16 18 20

Pro

tein

bin

ding

tann

ins

(mg

g–1

DW

)

0

50

100

150

200

250

300

350

Con

dens

ed ta

nnin

s(m

g g–

1 D

W)

0

50

100

150

200

250

300

Hyd

roly

zabl

e ta

nnin

s(m

g g–

1 D

W)

0

50

100

150

200

(a)

(b)

(c)

0

2

4

6

4 12 18

4 12 1802468

10

4 12 18

05

10152025

Fig. 2. (a) Condensed and (b) hydrolysable tannin concentrations

and (c) protein-binding capacity of tannin extracts in initial and

decomposing Populus tremuloides L. and P. grandidentata L. leaf lit-

ters. Circles represent P. grandidentata, triangles represent P. tremu-

loides samples; shaded symbols are leaf samples that were ungalled,

open symbols are leaf samples that had been galled by the petiole gal-

ler Ectoedemia populella Brusk. Inset graphs show details of time

points at 4, 12 and 18 months not evident in the larger figures. See text

forMaterials andmethods.



4 months, condensed tannin concentrations were higher in

P. grandidentata litters relative to P. tremuloides litter

(F1,36Æ5 = 42Æ0, P < 0Æ001), but both had dropped precipi-

tously (c. 4% and 0Æ2% of initial concentrations in the

P. grandidentata and P. tremuloides litters, respectively).

Consistently, tannin concentrations were very low in the 12-

and 18-month litters. Not surprisingly, the protein-binding

capacity of litter extracts that had been decomposed in the

field was essentially non-existent after 4 months. While there

were some statistically significant interactions over time (not

shown), they are likely biologically meaningless considering

how little tannin remained in the litters. More importantly,

the minimal tannin concentrations after 12 and 18 months in

the field (Fig. 2) suggest that tannins exerted little if any influ-

ence on decomposition after the first 4 months in the field.

Gall effects on N and C concentrations and dynamics

also differed between Populus species. Leaf N was higher in

the galled litter before decomposition independent of spe-

cies (F1,28 = 5Æ75, P = 0Æ023; Fig. 3a), and P. grandidentata

had overall lower initial N concentrations than did P. tre-

muloides litters (F1,28 = 9Æ98, P = 0Æ004). Tree species and

galling independently influenced litter N dynamics. Populus

tremuloides accumulated N at a greater rate over time than

did P. grandidentata (date*species F3,153 = 3Æ69,P = 0Æ013). In addition, galling exerted a significant influ-

ence on N concentrations over time independent of plant

species (date*gall F3,152 = 3Æ49, P = 0Æ017). In contrast to

N dynamics, litter total C was marginally influenced by

galling (F1,156 = 3Æ77, P = 0Æ084; Fig. 3b). Total C was

consistently higher in P. grandidentata (F1,155 = 11Æ75,P < 0Æ001). At 18 months, galled samples had higher total

C in P. tremuloides litter and lower total C in P. grand-

identata relative to the ungalled controls (species x gall

F1,43 = 5Æ16, P = 0Æ028); as a result, galling affected total

C dynamics over time uniquely between species (date*spe-

cies*gall F3,150 = 2Æ88, P = 0Æ038).

Nitr

ogen

(mg

g–1

DW

)

68

101214161820222426

Car

bon

(mg

g–1

DW

)

440

450

460

470

480

490

500

(a)

(b)

Cel

lulo

se (m

g g–

1 D

W)

60

80

100

120

140

160

Month

Lign

in (m

g g–

1 D

W)

50100150200250300350400

Hem

icel

lulo

se (m

g g–

1 D

W)

80

90

100

110

120

130

140 (e)

Month0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20

Lign

in:N

ratio

10

12

14

16

18

20

22

C:N

ratio

10203040506070

P. grandidentata ungalledP. grandidentata galledP. tremuloides ungalledP. tremuloides galled

(g)(d)

(c) (f)

Fig. 3. (a) Nitrogen [N], (b) carbon [C], (c) cellulose, (d) lignin and (e) hemicellulose concentrations, and (f) C ⁄N and (g) lignin ⁄N ratios in ini-

tial and decomposing Populus tremuloides L. and P. grandidentata L. leaf litters. Circles represent P. grandidentata, triangles represent P. tremu-

loides samples; shaded symbols are leaf samples that were ungalled, open symbols are leaf samples that had been galled by the petiole galler

Ectoedemia populella Brusk. Only leaf lamina tissue is represented; that is, the physical gall was removed prior to analysis. Symbols represent

means ± SE, with n = 5 forP. tremuloides galled samples and n = 15 for all others. See text forMaterials andmethods.



Concentrations of cell wall constituents were influenced by

galling and plant species. Cellulose and lignin concentrations

in senesced, undecomposed leaf tissue were 12% and 29%

higher in galled litters relative to ungalled litters, respectively,

independent of plant species (cellulose: F1,28 = 21Æ51,P < 0Æ001; lignin: F1,28 = 22Æ97, P < 0Æ001; Fig. 3c,d).

During decomposition, cellulose concentrations segregated

by plant species (species: F1,158 = 39Æ95, P < 0Æ001) and the

initial effect of galling were lost (gall: F1,159 = 0Æ85,P = 0Æ359). The two species also varied in the rate of cellu-

lose loss (date*species: F3,149 = 4Æ90, P = 0Æ003). In con-

trast, lignin concentrations and their rate of change in the

decomposing litter were influenced by galling (gall:

F1,158 = 8Æ27, P = 0Æ005; date*gall: F3,149 = 6Æ61,P < 0Æ001). The date*gall interactive effect seemed driven by

a significant gall effect at 18 months (F1,32Æ6 = 4Æ27,P = 0Æ047) that was not present at 4 or 12 months. Hemicel-

lulose concentrations were not affected by either tree species

or galling (Fig. 3e).

Nutrient ratios can play an important role in shaping litter

decomposition dynamics. The C ⁄N ratios were driven by N

concentrations and, P. tremuloides had lower C ⁄N ratios

(F1,161 = 99Æ45, P < 0Æ001) and a more rapid reduction in

C ⁄N ratios over time (date*species F3,152 = 3Æ78,P = 0Æ012)than did P. grandidentata (Fig. 3f). In other words, assuming

that C ⁄N ratios are predictive of litter quality, P. tremuloides

was of higher quality to decomposers than was P. grandentata.

Galling had a relatively small effect on C ⁄N ratios. In con-

trast, the influence of galling on lignin ⁄N ratios was pro-

nounced (date*gall F3,147 = 3Æ17, P = 0Æ026; Fig. 3g). Eventhough there was no statistically significant date*spe-

cies*gall interaction (P = 0Æ278), galling clearly affected lig-

nin ⁄N ratios in P. tremuloides more than in

P. grandidentata: there were significant species*gall interac-

tions at 4 and 12 months (4 months F1,45 = 6Æ91,P = 0Æ011; 12 months F1,35Æ4 = 5Æ56, P = 0Æ024). In fact,

lignin ⁄N ratios were unchanged over time in galledP. tremu-

loides litters (Fig. 3g).

Discussion

A long-standing hypothesis in plant–herbivore ecology is that

herbivores influence rates of leaf-litter decomposition via

changes in litter chemical quality (Grace 1986; Choudhury

1988).While abiotic factors such as temperature andmoisture

clearly drive decomposition dynamics across biomes (Trofy-

mow et al. 2002; Wall et al. 2008), the influence of substrate

quality is most pronounced on spatial scales where tempera-

ture and moisture conditions have relatively minimal varia-

tion (Swift, Heal & Anderson 1979). That is, substrate quality

likely matters in most mixed-stand terrestrial forests. Yet,

herbivores can influence litter quality in a number of ways. In

all cases of chewing or piercing ⁄ sucking herbivores, physical

damage to the lamina tissue occurs regardless of herbivore-

induced changes to lamina chemistry and may complicate the

effects of herbivore-induced chemistry per se. As a drastic

example of physical damage, experimental crushing of leaf lit-

ter dramatically increases decomposition rates by facilitating

access to the substrate (Madritch & Hunter 2003). Such

substrate access increases the more a leaf is consumed by a

chewing herbivore, which causes physical damage to the leaf

tissue (Cornelissen et al. 1999; Perez-Harguindeguy et al.

2000). The Ectoedemia ⁄Populus system described here pro-

vides an opportunity to examine the role of herbivore-altered

plant chemistry apart from other factors associated with her-

bivory because the galler does not physically manipulate the

lamina. Our results indicate that an insect herbivore that has

little influence on leaf lamina development and does not

physically damage the lamina can nonetheless clearly

influence litter decomposition and nutrient release rates.

Herbivore-mediated changes in litter chemistry alone are suf-

ficient to alter leaf-litter decomposition rates and, presum-

ably, other ecosystem processes responsive to those rates.

Although decomposition dynamics differed between tree

species, E. populella petiole galling consistently decelerated

leaf lamina litter decomposition. Deceleration of decomposi-

tion theoretically results from low substrate quality; this typi-

cally implies relatively lower concentrations of beneficial

components such as N and relatively higher concentrations of

detrimental components such as tannins and lignin, each of

which has been shown to influence decomposition via effects

on decomposer organisms (Northup, Dahlgren & McColl

1998; Maie et al. 2003; Hattenschwiler & Gasser 2005). This

general model held for the species main effect in our system:

P. grandidentata litter had lower total N, higher tannin con-

centrations and binding activity (but no difference in initial

litter lignin) and decomposed more slowly than did P. tremu-

loides litter. However, the presence of the gall confounded this

relationship. The relatively higher N concentrations in galled

P. tremuloides and relatively lower tannin concentrations and

binding capacity in galled tissues would typically be predicted

to accelerate, not decelerate, decomposition. Evidently some

other metabolite(s) associated with galled lamina trumped

the higher N concentrations. Moreover, the gall-mediated

reduction in initial tannin concentrations and binding capac-

ity suggests that tannins are not involved. One possibility is

that galling altered leaf toughness, which is has been shown to

affect decomposition rates (Cornelissen et al. 1999; Perez-

Harguindeguy et al. 2000). We did not measure leaf tough-

ness directly, but P. grandidentata leaves were obviously

tougher than were P. tremuloides and, as expected, P. grand-

identata decomposed more slowly. There were no such obvi-

ous phenotypic differences in leaf toughness in galled vs.

ungalled litter within either species that would appear to

account for the differential decomposition rates.

The result that Ectoedemia galling decelerated lamina litter

decomposition is consistent with effects from other gall

insects. In fact, gall insects from diverse lineages (i.e. a social

hemipteran, a cecidomyiid dipteran and the microlepidopter-

an studied here) independently decelerate leaf-litter decompo-

sition under field conditions (Schweitzer et al. 2005b;

Crutsinger et al. 2008). These effects presumably result from

modification of foliar chemical composition as our data sug-

gest, but may also include morphological changes in the other



systems. Indeed, gall insects may have a larger relative influ-

ence on decomposition dynamics than other insect–herbivore

guilds because of their intimate host-plant associations and

their ability to manipulate the molecular physiology of their

host (Schweitzer et al. 2005b). However, other herbivore

guilds can also influence litter decomposition. In some cases,

chewing and piercing-sucking herbivores have accelerated

decomposition (Ritchie, Tilman & Knops 1998; Chapman

et al. 2003); in other cases, herbivory did not influence mass

loss despite significant differences in initial litter quality (Hall

et al. 2006; Frost & Hunter 2008). Based on the three galling

systems studied, the magnitude of galling-herbivore effects on

decomposition relative to other insect–herbivore guilds may

not be as important as their direction: galling modifies leaf

quality in a manner that consistently decelerates leaf-litter

decomposition independent of the identity of the gall insect

or host-plant species.

Attributes that are common to gall insects may provide a

theoretical framework for the consistently observed gall-

mediated deceleration of litter decomposition that is based on

modified litter chemistry. In forcing the host plant to con-

struct an enclosed domicile with a fortified exterior and nutri-

tive interior, many gall insects are less vulnerable to predators

yet have access to an ideal food source during development.

One limitation of galling is that the insect depends on the suc-

cess of the host plant in a manner that is distinct from other

herbivore guilds. While chewing and piercing-sucking herbi-

vores can often search for new suitable host plants, the sessile

nature of gall insects mandates that they have a vested interest

in protecting their host plants. This suggests that gall insects

may manipulate their host plants to be less suitable to other

herbivores while evading host defenses themselves (Tooker &

De Moraes 2007; Stireman & Cipollini 2008; Tooker et al.

2008). Indeed, some gall insect activity can increase concen-

trations of defensive metabolites systemically in ungalled

parts of the plant, which presumably protects the resource

from other forms of herbivory (Pascual-Alvarado et al. 2008;

Cooper & Rieske 2009; Tooker & De Moraes 2009). This

leads to the hypothesis that gall-induced defensive manipula-

tions that protect a host plant from injury by other herbivores

result in decelerated litter decomposition. That is, the same

chemical compounds in foliage that confer resistance to her-

bivores also decelerate decomposition in leaf litter.

Although the life history of E. populella allows us to con-

clude that changes in leaf chemistry play a central role influ-

encing leaf-litter decomposition, gall-mediated differences in

leaf quality that result in the observed decomposition

changes remain unclear. To complicate matters, the effects

of E. populella on leaf-litter decomposition in our study var-

ied between the two host Populus species (Fig. 1), which sug-

gests that the gall-induced chemistry that altered

decomposition dynamics may differ between tree species.

Unfortunately, the gall-induced changes in litter chemistry

in these species are not well known; we therefore focused on

leaf chemistry parameters that are known to affect decompo-

sition (and had demonstrated effects between the two tree

species). Lignin concentrations in our system were 29%

higher initially in galled tissues, and this difference may have

affected the slower decomposition after 4 months in the

P. grandidentata litters. Yet, lignin dynamics over time did

not appear to explain the observed galling effects on decom-

position rates in P. tremuloides litters. In Solidago, lower

total N concentration in the midge-galled litters related to

decelerated decomposition (Crutsinger et al. 2008); Ectoede-

mia-galled litters had higher N concentrations but nonethe-

less exhibited decelerated decomposition. It is possible that

the form of N was different between the galled and ungalled

litters and that this difference was important. In the cotton-

wood system, higher condensed tannin concentrations in

galled litters contributed to decelerated decomposition

(Schweitzer et al. 2005b). In our case, tannin concentrations

and binding capacity were lower in galled tissue and were

essentially lost after 4 months. Although these are com-

monly measured aspects of litter quality, they do not neces-

sarily confer resistance against insect herbivores and do not

appear to explain the decelerating effects of gall-mediated lit-

ter chemistry on litter decomposition. A future line of

enquiry will obviously be needed to explore in both Populus

species the gall-induced changes in chemistry that may be

associated with delayed decomposition.

Other aspects of E. populella life history might also influ-

ence leaf chemistry in ways that alter litter decomposition.

For example, insect frass, which is N rich and remains in the

gall after leaf senescence (See Fig. S1, Supporting informa-

tion) and the higher initial concentration ofN in the galled lit-

ter may have resulted from N absorbed by the plant from the

frass (Frost &Hunter 2007). This frass may therefore provide

a nutrient resource to decomposer microfauna and accelerate

decomposition (Reynolds &Hunter 2001; Fonte & Schowalter

2005; Madritch, Donaldson & Lindroth 2007; Frost & Hun-

ter 2008), although the consistent deceleration of decomposi-

tion in galled tissue does not support this. Nevertheless, our

results do not rule out a role of frass in mediating lamina

chemistry because frass can also contain high levels of plant

defensive compounds (Kopper et al. 2002; Chen et al. 2007).

With E. populella in particular, frass remains within the gall

body and we assume has antimicrobial metabolites to pre-

vent rampant bacterial or fungal growth in the enclosed gall

environment (Tooker and De Moraes 2006). Such metabo-

lites may have extended, indirect effects inhibiting microbial

colonization of leaf litter that presumably would also decel-

erate decomposition.

As a final point, our experimental design isolated litters

from each tree species separately to investigate independent

effects of galling on the two main host tree species, but litters

in natural mixed-stand forests obviously contain a mix of spe-

cies. Litters from different tree species and genotypes decom-

pose at different rates (Fig. 1a), but mixed-species litters can

decompose in an non-additive manner that is not predicted

by the decomposition rates of individual species (Madritch &

Hunter 2003, 2005; Schweitzer et al. 2005a; Ball et al. 2008).

Herbivory may impose an additional factor on mixed-species

litter decomposition dynamics. Herbivory and geno-

typic ⁄ genetic variation are known to interact to affect litter



decomposition (Schweitzer et al. 2005b; Chapman, Schweit-

zer &Whitham 2006), and our results thatEctoedemia galling

differentially affected decomposition dynamics in the two

host Populus species provide further evidence for this. How-

ever, it is unknown how herbivory interacts specifically with

non-additive effects of mixed-species litter decomposition.

So, while the different decomposition dynamics between the

two Populus species in our study is not surprising, the differ-

ential effects of galling on decomposition may be altered

when the two leaf litters are mixed in in ways that may not be

predicted by the galling effects on either species individually.

In summary, we have used a model system with a unique

natural history to demonstrate that insect herbivore-mediated

changes in leaf chemistry can alter leaf-litter decomposition

rates. Further, based on the results from this system and the

other two gall systems, we develop the hypothesis that gall-

mediated deceleration in leaf-litter decomposition results

from induced defensive changes that protect a host plant from

damage from other herbivores, thus increasing gall insect fit-

ness. This hypothesis provides a basis for generalizing the

influence of a large guild of herbivores on a fundamental eco-

system process and should contribute to the theoretical

framework relating species interactions with ecosystem

ecology.

Acknowledgements

We thank Jennifer Schweitzer for insightful comments, Janet Saunders for labo-

ratory assistance, Peter Wilf and Conrad Labandeira for discussions on Ec-

toedemia evolutionary history, Patrick Abbot for Pemphigus photographs

and discussion, Megan Marshall and Tom Richard for use of an ANKOM

fiber analyzer, and Gary Felton and Michelle Peiffer for use of a SpectraMax

plate reader. TomMaddox (Odum School of Ecology Analytical Laboratory)

processed the C ⁄ N samples. The National Research Initiative of the USDA

(#2007-35302-18087 [CJF] and #2006-01823 [JFT]) funded this work.

References

Allison, S.D. & Schultz, J.C. (2005) Biochemical responses of chestnut oak to a

galling cynipid. Journal of Chemical Ecology, 31, 151–166.

Ball, B.A., Hunter, M.D., Kominoski, J.S., Swan, C.M. & Bradford, M.A.

(2008) Consequences of non-random species loss for decomposition dynam-

ics: experimental evidence for additive and non-additive effects. Journal of

Ecology, 96, 303–313.

Bate-Smith, E.C. (1977a) Astringent tannins of Acer species. Phytochemistry,

16, 1421–1426.

Bate-Smith, E.C. (1977b) Detection and determination of ellagitannins. Phyto-

chemistry, 11, 1153–1156.

Belovsky,G.E.& Slade, J.B. (2000) Insect herbivory accelerates nutrient cycling

and increases plant production. Proceedings of the National Academy of Sci-

ences of theUnited States of America, 97, 14412–14417.

Chapman, S.K., Schweitzer, J.A. & Whitham, T.G. (2006) Herbivory differen-

tially alters plant litter dynamics of evergreen and deciduous trees. Oikos,

114, 566–574.

Chapman, S.K., Hart, S.C., Cobb, N.S., Whitham, T.G. & Koch, G.W. (2003)

Insect herbivory increases litter quality and decomposition: an extension of

the acceleration hypothesis.Ecology, 84, 2867–2876.

Chen, H., Gonzales-Vigil, E., Wilkerson, C.G. & Howe, G.A. (2007) Stability

of plant defense proteins in the gut of insect herbivores. Plant Physiology,

143, 1954–1967.

Choudhury, D. (1988) Herbivore induced changes in leaf-litter resource quality:

a neglected aspect of herbivory in ecosystem nutrient dynamics. Oikos, 51,

389–393.

Cooper, W.R. & Rieske, L.K. (2009) Woody stem galls interact with foliage to

affect community associations.Environmental Entomology, 38, 417–424.

Cornelissen, J.H.C., Perez-Harguindeguy, N., Diaz, S., Grime, J.P., Marzano, B.,

Cabido, M., Vendramini, F. & Cerabolini, B. (1999) Leaf structure and

defence control litter decomposition rate across species and life forms in

regional floras on two continents.NewPhytologist, 143, 191–200.

Crutsinger, G.M., Habenicht,M.N., Classen, A.T., Schweitzer, J.A. & Sanders,

N.J. (2008) Galling by Rhopalomyia solidaginis alters Solidago altissima

architecture and litter nutrient dynamics in an old-field ecosystem. Plant and

Soil, 303, 95–103.

Findlay, S., Carreiro, M., Krischik, V. & Jones, C.G. (1996) Effects of

damage to living plants on leaf litter quality. Ecological Applications,

6, 269–275.

Fonte, S.J. & Schowalter, T.D. (2005) The influence of a neotropical herbivore

(Lamponius portoricensis) on nutrient cycling and soil processes.Oecologia,

146, 423–431.

Frost, C.J. & Hunter, M.D. (2007) Recycling of nitrogen in herbivore feces:

plant recovery, herbivore assimilation, soil retention, and leaching losses.

Oecologia, 151, 42–53.

Frost, C.J. &Hunter, M.D. (2008) Insect herbivores and their frass affect Quer-

cus rubra leaf quality and initial stages of subsequent litter decomposition.

Oikos, 117, 13–22.

Grace, J.R. (1986) The influence of gypsymoth on the composition and nutrient

contentof litter fall in aPennsylvania oak forest.ForestScience, 32, 855–870.

Hagerman, A.E. & Butler, L.G. (1978) Protein precipitationmethod for quanti-

tative determination of tannins. Journal of Agricultural and Food Chemistry,

26, 809–812.

Hagerman, A.E. & Butler, L.G. (1980) Condensed tannin purification and

characterization of tannin-associated proteins. Journal of Agricultural and

Food Chemistry, 28, 947–952.

Hagerman, A.E. & Butler, L.G. (1989) Choosing appropriate methods and

standards for assaying tannin. Journal of Chemical Ecology, 15, 1795–1810.

Hagerman, A.E., Rice, M.E. & Ritchard, N.T. (1998) Mechanisms of protein

precipitation for two tannins, pentagalloyl glucose and epicatechin 16 (4–

>8) catechin (procyanidin). Journal of Agricultural and Food Chemistry, 46,

2590–2595.

Hall, M.C., Stiling, P., Moon, D.C., Drake, B.G. & Hunter, M.D. (2006) Ele-

vated CO2 increases the long-term decomposition rate of Quercus myrtifolia

leaf litter.Global Change Biology, 12, 568–577.

Hattenschwiler, S. & Gasser, P. (2005) Soil animals alter plant litter diversity

effects on decomposition. Proceedings of the National Academy of Sciences

of the United States of America, 102, 1519–1524.

Hunter,M.D. &Forkner, R.E. (2004)Hurricane damage influences foliar poly-

phenolics and subsequent herbivory on surviving trees. Ecology, 80, 2676–

2682.

Karban, R. & Baldwin, I.T. (1997) Induced Responses to Herbivory. University

of Chicago Press, Chicago, IL.

Kopper, B.J., Jakobi, V.N., Osier, T.L. & Lindroth, R.L. (2002) Effects of

paper birch condensed tannin on whitemarked tussock moth (Lepidoptera:

Lymantriidae) performance.Environmental Entomology, 31, 10–14.

Kurokawa, H. & Nakashizuka, T. (2008) Leaf herbivory and decomposability

in aMalaysian tropical rain forest.Ecology, 89, 2645–2656.

Labandeira, C.C., Dilcher, D.L., Davis, D.R. & Wagner, D.L. (1994) Ninety-

seven million years of angiosperm-insect association: paleobiological

insights into the meaning of coevolution. Proceedings of the National Acad-

emy of Sciences of the United States of America, 91, 12278–12282.

Larson, K.C. (1998) The impact of two gall-forming arthropods on the photo-

synthetic rates of their hosts.Oecologia, 115, 161–166.

Littell, R.C., Stroup,W.W. & Freund, R.J. (2002) SAS for LinearModels. SAS

Institute, Inc., Cary, NC.

Madler, V.A.K. (1936) Eine Blattgalle an einem vorweltlichn Pappel-Blatt.

Natur und Volk, 66, 271–274.

Madritch, M.D., Donaldson, J.R. & Lindroth, R.L. (2007) Canopy herbivory

can mediate the influence of plant genotype on soil processes through frass

deposition.Soil Biology & Biochemistry, 39, 1192–1201.

Madritch, M.D. & Hunter, M.D. (2002) Phenotypic diversity influences

ecosystem functioning in an oak sandhills community. Ecology, 83,

2084–2090.

Madritch, M.D. & Hunter, M.D. (2003) Intraspecific litter diversity and nitro-

gen deposition affect nutrient dynamics and soil respiration.Oecologia, 136,

124–128.

Madritch,M.D. &Hunter, M.D. (2004) Phenotypic diversity and litter chemis-

try affect nutrient dynamics during litter decomposition in a two species mix.

Oikos, 105, 125–131.

Madritch,M.D.&Hunter,M.D. (2005) Phenotypic variation in oak litter influ-

ences short- and long-term nutrient cycling through litter chemistry. Soil

Biology & Biochemistry, 37, 319–327.



Maie, N., Behrens, A., Knicker, H. & Kogel-Knabner, I. (2003) Changes in the

structure and protein binding capacity of condensed tannins during decom-

position. Soil Biology& Biochemistry, 35, 577–589.

Northup, R.R., Dahlgren, R.A. & McColl, J.G. (1998) Polyphenols as regula-

tors of plant-litter-soil interactions in northern California’s pygmy forest: a

positive feedback?Biogeochemistry, 42, 189–220.

Nyman, T. & Julkunen-Tiitto, R. (2000) Manipulation of the phenolic chemis-

try of willows by gall-inducing sawflies.Proceedings of the National Academy

of Sciences of the United States of America, 97, 13184–13187.

Ostertag, R., Scatena, F.N. & Silver, W.L. (2003) Forest floor decomposition

following hurricane litter inputs in several Puerto Rican forests. Ecosystems,

6, 261–273.

Parton,W., Silver,W.L., Burke, I.C., Grassens, L., Harmon,M.E., Currie,W.S.,

King, J.Y., Adair, E.C., Brandt, L.A., Hart, S.C. & Fasth, B. (2007) Global-

scale similarities in nitrogen release patterns during long-term decomposi-

tion.Science, 315, 361–364.

Pascual-Alvarado, E., Cuevas-Reyes, P., Quesada, M. & Oyama, K. (2008)

Interactions between galling insects and leaf-feeding insects: the role of plant

phenolic compounds and their possible interference with herbivores. Journal

of Tropical Ecology, 24, 329–336.

Perez-Harguindeguy, N., Dıaz, S., Cornelissen, J., Vendramini, F., Cabido, M.

& Castellanos, A. (2000) Chemistry and toughness predict leaf litter decom-

position rates over a wide spectrum of functional types and taxa in central

Argentina.Plant and Soil, 218, 21–30.

Reynolds, B.C. &Hunter, M.D. (2001) Responses of soil respiration, soil nutri-

ents, and litter decomposition to inputs form canopy herbivores.Soil Biology

& Biochemistry, 33, 1641–1652.

Ritchie, M.E., Tilman, D. & Knops, J.M.H. (1998) Herbivore effects on plant

and nitrogen dynamics in oak savanna.Ecology, 79, 165–177.

Rossiter,M., Schultz, J.C. & Baldwin, I.T. (1988) Relationships among defolia-

tion, red oak phenolics, and gypsy moth growth and reproduction. Ecology,

69, 267–277.

Schweitzer, J.A., Bailey, J.K., Hart, S.C. & Whitham, T.G. (2005a) Nonaddi-

tive effects of mixing cottonwood genotypes on litter decomposition and

nutrient dynamics.Ecology, 86, 2834–2840.

Schweitzer, J.A., Bailey, J.K., Hart, S.C., Wimp, G.M., Chapman, S.K. &

Whitham, T.G. (2005b) The interaction of plant genotype and herbivory

decelerate leaf litter decomposition and alter nutrient dynamics. Oikos, 110,

133–145.

Stireman, J.O. & Cipollini, D. (2008) Stealth tactics of galling parasites and

their potential indirect effects.New Phytologist, 178, 462–465.

Stone, G.N. & Schonrogge, K. (2003) The adaptive significance of insect gall

morphology.Trends in Ecology & Evolution, 18, 512–522.

Swift, M.J., Heal, O.W. & Anderson, J.M. (1979) Decomposition in Terrestrial

Ecosystems. University of California Press, Berkeley, CA.

Tooker, J.F. & De Moraes, C.M. (2006) Jasmonate in lepidopteran larvae.

Journal of Chemical Ecology, 32, 2321–2326.

Tooker, J.F. & De Moraes, C.M. (2007) Feeding by Hessian fly [Mayetiola

destructor (Say)] larvae does not induce plant indirect defences. Ecological

Entomology, 32, 153–161.

Tooker, J.F. & De Moraes, C.M. (2009) A gall-inducing caterpillar species

increases essential fatty acid content of its host plant without concomitant

increases in phytohormone levels.Molecular Plant-Microbe Interactions, 22,

551–559.

Tooker, J.F., Koenig, W.A. & Hanks, L.M. (2002) Altered host plant volatiles

are proxies for sex pheromones in the gall waspAntistrophus rufus.Proceed-

ings of the National Academy of Sciences of the United States of America, 99,

15486–15491.

Tooker, J.F., Rohr, J.R., Abrahamson, W.G. & De Moraes, C.M. (2008) Gall

insects can avoid and alter indirect plant defenses. New Phytologist, 178,

657–671.

Trofymow, J.A., Moore, T.R., Titus, B.D., Prescott, C., Morrison, I., Siltanen,

M., Smith, S.M., Fyles, J., Wein, R., Camir, C., Duschene, L., Kozak, L.M.,

Kranabetter, M. & Visser, S. (2002) Rates of litter decomposition over

6 years in Canadian forests: influence of litter quality and climate. Canadian

Journal of Forest Research, 32, 789–804.

Wall, D.H., Bradford,M.A., John,M.G.S., Trofymow, J.A., Behan-Pelletier, V.,

Bignell, D.D.E., Dangerfield, J.M., Parton,W.J., Rusek, J., Voigt, W., Wol-

ters, V., Gardel, H.Z., Ayuke, F.O., Bashford, R., Beljakova, O.I., Bohlen,

P.J., Brauman, A., Flemming, S., Henschel, J.R., Johnson, D.L., Jones,

T.H., Kovarova, M., Kranabetter, J.M., Kutny, L., Lin, K.C., Maryati, M.,

Masse, D., Pokarzhevskii, A., Rahman, H., Sabara, M.G., Salamon, J.A.,

Swift, M.J., Varela, A., Vasconcelos, H.L., White, D. & Zou, X.M. (2008)

Global decomposition experiment shows soil animal impacts on decomposi-

tion are climate-dependent.Global Change Biology, 14, 2661–2677.

Weis, A.E. & Abrahamson, W.G. (1986) Evolution of host-plant manipulation

by gall makers – ecological and genetic-factors in the Solidago-Eurosta sys-

tem.AmericanNaturalist, 127, 681–695.

Wilkinson, C. & Scoble, M.J. (1979) The Nepticulidae (Lepidoptera) of

Canada.Memoirs of the Entomological Society of Canada, 111, 1–129.

Received 17 July 2011; accepted 15 February 2012

Handling Editor: DanHare

Supporting Information

Additional supporting information may be found in the online ver-

sion of this article.

Figure S1. Photographs showing morphological and structural

changes to the leaf lamina by (A–B) Rhopalomyia solidaginis and (C)

Pemphigus betae.

Figure S2. Gall induced on the petiole by Ectoedemia populella Brusk

(A) in the field showing no apparent morphological changes to the

leaf lamina, which was fully expanded before galling. (B–C) Close

view of the gall exterior and interior. (D–E) E. populella larvae and

frass pack. Photo credits: Christopher Frost.

Figure S3. Leaf drop timing in Populus grandidentata during active

litter fall.

As a service to our authors and readers, this journal provides Sup-

porting Information supplied by the authors. Such materials may be

re-organized for online delivery, but are not copy-edited or typeset.

Technical support issues arising from Supporting Information (other

thanmissing files) should be addressed to the authors.



Documents

A petiole-galling insect herbivore decelerates leaf lamina litter decomposition rates