.3.

.84»?

Hr.“

r... ,h%.

a ..
:Ii hrh
2.. .

m-‘I ~ .

. ‘0. v ftzl l\ t... 5’33
IAN «(fill
i I: I." 1. I: .v: 0. .35.).
y‘. . . . .tp : Ill-(.10
- . 51!... .
l. 2. it: ,\...P:. s. x!
r .L 110
- .PIJ rt 1 Iv
. u . .
VAL
‘ \nvi

 

 

mmafliflz.

anxmwfimnhwfinmg .

THEES

 

 

 

$593331; my“ 11/121 jflfl/fllj/Mflf/WI/Ifl
University

 

 

 

This is to certify that the

thesis entitled

LIGHT, NUTRIENT AVAILABILITY, AND DEFOLIATION EFFECTS
ON RESOURCE ACQUISITION, RESOURCE ALLOCATION,
AND HERBIVORE RESISTANCE OF PAPER BIRCH
AND SUGAR MAPLE
presented by

Heather L. Govenor

has been accepted towards fulfillment
of the requirements for

M. S. degree in Entomology_

sag/gag

Major professor

Date inguiy [T28

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINB return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE I DATE DUE DATE DUE
001 G 2 638%

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1!” WM“

LIGHT, NUTRIENT AVAILABILITY, AND DEFOLIATION EFFECTS ON
RESOURCE ACQUISITION, RESOURCE ALLOCATION, AND HERBIVOR-E
RESISTANCE OF PAPER BIRCH AND SUGAR MAPLE

By

Heather L. Govenor

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Entomology
and
Ecology, Evolutionary Biology and Behavior Program

1998

ABSTRACT
LIGHT, NUTRIENT AVAILABILITY, AND DEFOLIATION EFFECTS ON
RESOURCE ACQUISITION, RESOURCE ALLOCATION, AND HERBIVORE
RESISTANCE OF PAPER BIRCH AND SUGAR MAPLE
By

Heather L. Govenor

This research examined light, nutrient, and defoliation effects on resource
acquisition, resource allocation, and herbivore resistance in juvenile paper birch and
sugar maple. Tree physiological responses were consistent with theories of life history
and phenotypic plasticity. Fertilization did not consistently affect resource acquisition or
allocation of either tree species, but shade did. Treatment effects on host quality were
dependent on insect species. Effects of defoliation on birch physiology mimicked
nutrient deficiency. Defoliation of shaded birch elicited rapid induced resistance to .
lepidopteran larvae. There was no evidence of delayed induced resistance to any

herbivore species the year following defoliation.

This work is dedicated to Grace Agnes Bic/tel,
the most loving and giving woman 1 have ever known.

and

lo the memory of'Hether ( 'Iair Miller.
All who knew her are better/Or ii.

iii

ACKNOWLEDGEMENTS

This thesis represents my synthesis Of the contributions of many. First and
foremost, I thank my major advisor Dr. Dan Herms for entrusting me with the challenge
of working on this project, his brainchild. I am grateful for his support and approval
(sometimes enthusiastic, sometimes tentative) of my pursuit of projects outside the scope
of the research represented here, and for his sincerity and encouragement throughout. Dr.
Mark Scriber provided priceless information on conferences and funding opportunities.
and helpful comments for revision of the final draft. Dr. Bill Mattson graciously
provided lab space, technical support, advice, and a full week of his own time weighing
larvae and frass. Dr. Don Dickmann provided valuable insight into the interpretation of
results throughout the research phase and thoughtful comments on the manuscript.

I am indebted to those who assisted me in data collection. Dr. Matt Ayres added
an entire component to the research, providing lab equipment and weeks of his time to
monitor foliar light responses. I am thankful for his quick and thoughtful advice and
willingness to help however possible on all phases of this project. Laura (Haas) Comilla
provided non-stop energy at the lab, in the field, at the gym, hiking about, and at home. I
am thankful for her hard work and determination, and most of all, for her friendship.
Nate Krivitzky was always willing to do anything to get things done. His thought-
provoking questions (often completely unrelated to the research at hand) and lab
serenades made our first summer fun as well as productive. Cathy Papp Herms. Bruce
Birr, Carolyn Glynn, Brad Proper, Bill Styer. Terry Trier. James Nitao. Bryant Chambers.

Dave McCartney. and Merlee Nair and his graduate students provided crucial lab and

iv

technical support and/or advice. Thanks to the technical and secretarial staff ofthe
Michigan State University Department of Entomology who provided friendly and skilled
assistance in countless aspects of my degree program.

I would not have survived this degree without the moral support of my friends and
family who were always willing to listen to trials, tribulations. and success stories of the
work at hand. or were able to distract me completely from all thoughts of the matter. as
need be. I am especially indebted to Becky Durkin for her unwavering friendship and
willingness to listen at all hours ofthe day or night. .Ieff Feitl provided entertainment.
stress relief and a ready ear. Heather Rowe was ever-present with great ideas, editorial
skills, snacks. and extracurricular activities. Special thanks to Dylan Parry who, in
addition to being a good friend, served as a surrogate committee member of sorts.
providing insects for the 1997 study and valuable advice on the organization of the
manuscript. Piera Giroux provided muscles for the collection of soil samples, statistical
and editorial advice, and greatly enhanced my world knowledge and vocabulary. Chrissy
Jarzomski excelled at Thursday night “quality time” and enabled me to get much more
out of classes than I would have on my own. C orine Vriesendorp made the final phase of
this project fun. Jeremy Wojdak, while providing an excellent distraction from work.
challenged me to produce nothing less than the best thesis I could write and helped me
stay focused on my priorities and what really matters in this life. Grace Bickel provided
(in addition to two long days in the lab). nothing but encouragement. support. and love.
as she has unwaveringly for the last 25 years.

Finally. I would like to express my sincerest gratitude to the Environmental

Protection Agency. the Michigan State University College OfNatural Science and the

Department of Entomology who provided financial support for this project. and to the

Dow Gardens where this research was performed.

vi

TABLE OF CONTENTS

LIST OF TABLES ........................................................................
LIST OF FIGURES .......................................................................

CHAPTER 1 :
INTRODUCTION ........................................................................

CHAPTER 2:
EFFECTS OF LIGHT, NUTRIENT AVAILABILITY AND DEF OLIATION
ON RESOURCE ACQUISITION AND ALLOCATION IN PAPER BIRCH

(Betula papyrifera) AND SUGAR MAPLE (Acer saccharum) .....................
Introduction .................................................................................
Life History Differences in Resource Allocation .............................
Effects of Light, Nutrients, and Defoliation on Resource Acquisition and
Resource Allocation .....................................................
Light .......................................................................
Nutrients ..................................................................
Defoliation ...............................................................
Environmental Variation at the Scale of the Individual .....................
Theory on Mechanisms Driving Resource Allocation Patterns ............
Carbon/ Nutrient Balance Hypothesis ................................
Growth/ Differentiation Balance Hypothesis ........................
Objectives and Predictions ......................................................
Materials and Methods ....................................................................
Experimental Site and Design ...................................................
Experimental Treatments ........................................................
Light Manipulation ......................................................
Soil Nutrients ............................................................
Natural Defoliation ......................................................
Plant Physiological Responses ..................................................
Acclimation to Shade ....................................................
Photosynthetic Capacity (Light Response Curves) ........
Specific Leaf Mass and Leaf Area ...........................
Resource Acquisition ...................................................
Photosynthetic Assimilation at Ambient Conditions ......
F Oliar Carbon and Nitrogen ....................................
Resource Allocation .....................................................
Height and Diameter Growth .................................
F oliar Condensed Tannin and Total Phenolic
Concentrations .........................................
Statistical Analyses ..............................................................
Results ......................................................................................

vii

xiv

Acclimation to Shade ............................................................
Photosynthetic Capacity (Light Response Curves) .................
Specific Leaf Mass and Leaf Area ....................................

Intrinsic Differences in Specific Leaf Mass and Leaf Area
Light Effects on Specific Leaf Mass and Leaf Area ......
Nutrient Effects on Specific Leaf Mass and Leaf Area. ..
Defoliation Effects on Specific Leaf Mass and Leaf Area

Resource Acquisition ............................................................
Net Photosynthesis Under Ambient Conditions .....................
F oIiar Carbon and Nitrogen ............................................

Intrinsic Differences in Carbon and Nitrogen Concentration
and UN Ratios .......................................
Light Effects on F Oliar Carbon and Nitrogen Concentration
and ON Ratios ........................................
Nutrient Effects on F oliar Carbon and Nitrogen
Concentration and C/N Ratios .......................
Defoliation Effects on F oliar Carbon and Nitrogen
Concentration and C/N Ratios .......................
Resource Allocation ..............................................................
Height and Radial Growth .............................................
Intrinsic Differences in Birch and Maple Growth .........
Light Effects on Birch and Maple Growth ..................
Nutrient Effects on Birch and Maple Growth ...............
Defoliation Effects on Birch and Maple Growth ............
F oliar Condensed Tannin and Total Phenolic Concentrations. . . ..
Intrinsic Differences in Condensed Tannin and Total
Phenolic Content .......................................
Light Effects on Condensed Tannin and Total Phenolic
Content ..................................................
Nutrient Effects on Condensed Tannin and Total Phenolic
Content ..................................................
Defoliation Effects on Condensed Tannin and Total
Phenolic Content .......................................
Discussion ...................................................................................

Intrinsic Differences in Paper Birch and Sugar Maple Physiology ........

Seasonal and Leaf Age Related Changes in Birch F Oliar Chemistry ......

Effects of Light on Birch and Maple Physiology .............................

Effects OfNutrients on Birch and Maple Physiology ........................

Effects of Defoliation on Birch and Maple Physiology ......................

Conclusions ........................................................................

CHAPTER 3:

EFFECTS OF LIGHT, NUTRIENT AVAILABILITY. AND DEFOLIATION
ON THE RESISTANCE OF PAPER BIRCH (Beta/a papryrfera) AND SUGAR
MAPLE (Acer saccharum) TO CHEWING F OLIVORES ..........................

viii

b-b-btutubauawwtx)
\JOCOCOCIQIQIQOCOC

J;
\J

A
\l

48

50
53
53
53
56
56
58
6]

61

61

66

67
67
67
7O
7]
75
77
7‘)

Introduction ................................................................................. 82

Materials and Methods ..................................................................... 89

Experimental Site and Design and Treatments ................................ 89

Insect Bioassays ................................................................... 90

Gypsy Moth ............................................................... 90

Whitemarked Tussock Moth ............................................ 95

Forest Tent Caterpillar ................................................... 96

Insect Performance in Relation to F oliar Traits ............................... 97

Statistical Analyses ............................................................... 98

Results ....................................................................................... 105

Insect Bioassays ................................................................... 1 (15

Gypsy Moth ............................................................... 105

Whitemarked Tussock Moth ............................................ 122

Forest Tent Caterpillar ................................................... 130

Insect Performance in Relation to F oIiar Traits ................................ 132

Gypsy Moth ............................................................... 132

Whitemarked Tussock Moth ............................................ 132

Forest Tent Caterpillar ................................................... 134

Discussion .................................................................................... 1 34

Light Effects on Host Quality .................................................... 134

Nutrient Effects on Host Quality ................................................ 135

Defoliation Effects on Host Quality ............................................. 137

Relative Host Quality of Paper Birch and Sugar Maple ...................... 139

Compensatory Consumption ..................................................... 142

Conclusions ........................................................................ 1 44
CHAPTER 4:

CONCLUSIONS ........................................................................... 146

APPENDIX 1 Record of Deposition of Voucher Specimens ............... 154

APPENDIX 1.1 Voucher Specimen Data ....................................... 156

LITERATURE CITED .................................................................... 158

LIST OF TABLES

Table 2.1. Average photosynthetic assimilation (A, umoles C02 111'2 5-2) as a function of
irradiance (I, umol rn‘2 s") in non-defoliated trees. See text for measurement and
calculation details. N = 4-6 per treatment combination ....................................... 29

Table 2.2. F -values and degrees of freedom (dj) from ANOVA of blocking, light, species
and nutrient effects and their interactions on the initial “curve” (parameter a) and initial
slope (parameter b) of light response curves, and on assimilation at PAR = 0 umols m'2 s'
' (RESP), 100 umols m'2 s" (A100), 200 umols m'2 s" (A200), 400 umols m'2 s" (A400)
and 1000 umols m‘2 S’I (A1000). F-values for light were calculated with block*light as
the error term ......................................................................................... 30

Table 2.3. Foliar carbon and nitrogen concentration, carbon/nitrogen (C/N) ratio.
condensed tannin and total phenolic content. specific leaf mass (SLM). and leaf area of
paper birch and sugar maple foliage in 1995. 1996, and 1997 (least square means i
standard error) ........................................................................................ 33

Table 2.4. F-values and degrees of freedom (df) from ANOVA of blocking, light, and
nutrient effects and their interactions on specific leaf mass (A) and leaf area (B) of birch
and maple foliage sampled on 15 June 1995. There was no defoliation treatment in 1995.
F -values for light were calculated with block*light as the error term ........................ 34

Table 2.5. F-values and degrees of freedom (d/) from ANOVA of blocking. light.
nutrient, and defoliation effects and their interactions on specific leaf mass (A) and leaf
area (B) of birch and maple foliage sampled on 11 June 1996, birch foliage originating
from buds formed during the previous season (“Mature birch”) and from leaf primordia
formed in the current season (“Immature birch”) sampled on 1 August 1996, and from
birch and maple foliage sampled on 16 June 1997. F-values for light were calculated
with block*light as the error term ................................................................. 35

Table 2.6. F-values and degrees of freedom (d/) from ANOVA of blocking. light.
nutrient, and defoliation effects and their interactions on net photosynthesis and stomatal
conductance of paper birch and sugar maple under ambient conditions. Photosynthesis
was measured on two dates for each species in 1996. F -values for light were calculated
with block*light as the error term ................................................................. 41

Table 2.7. F—values and degrees of freedom (df) from ANOVA of blocking, light and
nutrient effects and their interactions on carbon content (%C), nitrogen content (%N) and
carbon/ nitrogen ratios (C/N ratio) of paper birch and sugar maple foliage collected in
June 1995. There was no defoliation treatment in 1995. F—values for light were
calculated with block*light as the error term ................................................... 43

Table 2.8. F-values and degrees of freedom (df) from ANOVA of blocking. light.
nutrient, and defoliation effects and their interactions on carbon content (%C), nitrogen

content (%N) and carbon/ nitrogen ratios (C/N ratio) of paper birch and sugar maple
foliage collected on 16 June 1996. F -values for light were calculated with block*light as
the error term ........................................................................................ 44

Table 2.9. F-values and degrees of freedom (df) from ANOVA of blocking, light.
nutrient, and defoliation effects and their interactions on carbon content (%C), nitrogen
content (%N) and carbon/ nitrogen ratios (C/N ratio) of paper birch foliage originating
from buds formed the previous season (“Mature birch”) and from the current season‘s
leaf primordia (“Immature birch”) collected 1 August 1996. Fvvalues for light were
calculated with block*light as the error term .................................................... 45

Table 2.10. F-values and degrees of freedom (df) from ANOVA of blocking. light.
nutrient, defoliation effects and their interactions on carbon content (%C), nitrogen
content (%N) and carbon/ nitrogen ratios (C/N ratio) of paper birch and sugar maple
foliage collected 16 June 1997. F -values for light were calculated with block*light as the
error term ......................................... i .................................................... 46

Table 2.11. F—values and degrees of freedom (df) from ANCOVA of blocking. light.
nutrient, defoliation effects, and their interactions on the trunk height and diameter
growth during the 1994, 1995, and 1996 seasons with height and diameter in May of
1994 as covariates, respectively; and over-all trunk height at the end of 1994, 1995, and
1996, with May 1994 height as covariate. Models for each year correspond to treatments
in effect during that year. F -values for light were calculated with block*light as the error
term ................................................................................................... 54

Table 2.12. F-values and degrees of freedom (df) from ANOVA of blocking. light.
nutrient, and defoliation effects and their interactions on total phenolic and condensed
tannin content (% foliar dry weight) of paper birch and sugar maple foliage collected 1 1
June 1996. F -values for light were calculated with block*light as the error term ......... 62

Table 2.13. F—values and degrees of freedom (d/) from ANOVA of blocking. light.
nutrient, defoliation effects and their interactions on total phenolic and condensed tannin
content (% foliar dry weight) of paper birch foliage originating from buds formed the
previous season (“Mature birch”) and from leaf primordia formed in the current season
(“Immature birch”) collected 1 August 1996. F-values for light were calculated with
block*light as the error term ....................................................................... 64

Table 2.14. F—values and degrees of freedom (d/) from ANOVA of blocking. light.
nutrient effects and their interactions on condensed tannin content (”/0 foliar dry weight)
of paper birch and sugar maple foliage collected 16 June 1997. F—values for light were
calculated with block*light as the error term ................................................... 65

Table 3.1. Predicted effects of treatments on the performance (growth. pupal mass.
and/or duration of larval period) of chewing herbivores based on their effect on foliar
traits in 1996 (percent change from control levels). Increases in foliar nitrogen and
decreases in condensed tannin and/ or total phenolic content are predicted to be beneficial

xi

for chewing herbivores. Direction of light, nutrient, and defoliation treatment effects on

foliar traits of paper birch and sugar maple relative to control levels are from Chapter
2 ....................................................................................................... 87

Table 3.2. Predicted effects of treatments on the performance (growth, pupal mass.
and/or duration of larval period) of chewing herbivores based on their effect on foliar
traits in 1997 (percent change from control levels). Increases in foliar nitrogen and
decreases in condensed tannin and/ or total phenolic content are predicted to be beneficial
for chewing herbivores. Direction of light, nutrient, and defoliation treatment effects on

foliar traits of paper birch and sugar maple relative to control levels are from Chapter
2 ....................................................................................................... 88

Table 3.3. Abbreviations and calculations of nutritional nutritional indices ................ 91

Table 3.4. Chi-square values, degrees of freedom (df), and p-values from nonparametric
analyses (SAS CATMOD) of light. species, nutrient, defoliation effects and their
interactions on gypsy moth 12 d survival and Whitemarked tussock moth 14 d survival in
1996. 60 5 N 5 73 insects on six trees per treatment combination ........................... 99

Table 3.5. F—values and degrees of freedom (df) from ANOVAS of light, Species,
nutrient, defoliation effects and their interactions on gypsy moth relative growth rates
measured from egg hatch to pupation (RGR total), and during the fourth stadium (RGR
L4), and relative consumption rate (RCR), approximate digestibility (AD), and efficiency
of conversion of digested food to biomass (ECD) measured during the fourth stadium. F -
values for light were calculated with block*light as the error term ......................... 100

Table 3.6. F—values and degrees of freedom (d/) from ANOVAS of light, Species.
nutrient, defoliation effects and their interactions on gypsy moth pupal mass and larval
duration in males and females. F -values for light were calculated with block*light as the
error term ........................................................................................... 102

Table 3.7. F-values and degrees of freedom (df) from ANOVAS of light, species.
nutrient, defoliation effects and their interactions on gypsy moth relative growth rates
over 48 h during the second (RGR L2) and fourth (RGR L4) stadii, and relative
consumption rate (RCR), approximate digestibility (AD), and efficiency of conversion of
digested food to biomass (ECD) measured during the fourth stadium. F -values for light
were calculated with block*light as the error term ........................................... 103

Table 3.8. Pearson coefficients for correlations between indices of gypsy moth (A),
Whitemarked tussock moth (B), and forest tent caterpillar (C) larval performance and host
defoliation and variables of quality of foliage sampled in June from paper birch (upper
value) and sugar maple (lower value) for 1996 and 1997 laboratory experiments (34 _<_ N
:I_ 48) ................................................................................................. 1 10

Table 3.9. F-values and degrees of freedom (df) from ANOVAS of light. species.
nutrient, defoliation effects and their interactions on Whitemarked tussock moth relative

xii

growth rates measured from egg hatch to pupation (RGR total), and during the fourth
Stadium (RGR L4), and relative consumption rate (RCR), approximate digestibility (AD),
and efficiency of conversion of digested food to biomass (ECD) measured during the
fourth stadium. F-values for light were calculated with block*light as the error
term .................................................................................................. 114

Table 3.10. F-values and degrees of freedom (df) from ANOVAS of light. species.
nutrient, defoliation effects and their interactions on Whitemarked tussock moth pupal
mass and larval durations in males and females. F—values for light were calculated with
block*light as the error term ..................................................................... l 16

Table 3.11. F—values and degrees of freedom (df) from ANOVAS of light. Species.
nutrient, defoliation effects and their interactions on Whitemarked tussock moth relative
growth rates over 48h during the second (RGR L2) and fourth (RGR L4) stadii, and
relative consumption rate (RCR), approximate digestibility (AD). and efficiency of
conversion of digested food to biomass (ECD) measured during the fourth stadium. F-
values for light were calculated with block*light as the error term .......................... l 17

Table 3.12. F—values and degrees of freedom (df) from ANOVAS of light, species,
nutrient, defoliation effects and their interactions on fifth instar forest tent caterpillar
relative growth rate (RGR L5), relative consumption rate (RCR), approximate
digestibility (AD), and efficiency of conversion of digested food to biomass (ECD). F -
values for light were calculated with block*light as the error term ......................... 126

xiii

LIST ()F FIGURES

Figure 2.1. Photosynthetic light response curves for sugar maple (A) and paper birch (B)
grown under factorial combinations of light and nutrient availability ....................... 31

Figure 2.2. Effects of light on the specific leaf mass (SLM) of paper birch and sugar
maple foliage in June (A) and August (B) of 1996 and June of 1997 (C). Asterisks
indicate significant effects of light treatment within Species or leaf age class (p 5 0.05).
See text for description of age classes ............................................................ 36

Figure 2.3. Effects of light on the leaf area of paper birch and sugar maple foliage in
June (A) and August (B) of 1996 and June of 1997 (C). Asterisks indicate significant
effects of light treatment within species or leaf age class (p _<_ 0.05). See text for
descriptionofage classes37

Figure 2.4. Light effects on net photosynthesis of paper birch and sugar maple. Asterisks
indicate significant effects of light within species (p g 0.05) ................................. 42

Figure 2.5. Effects of nutrient treatments on the foliar nitrogen content of paper birch
and sugar maple in June of 1995 (A), 1996 (B), and 1997 (C). Asterisks indicate
significant effects of nutrient within species (p 5 0.05) ....................................... 49

Figure 2.6. Correlations between percent defoliation by forest tent caterpillar in 1996 and
foliar nitrogen content in paper birch in 1997 (A) and sugar maple in 1996 (B) ........... 51

Figure 2.7. Effects of fertilization and defoliation on the foliar nitrogen content of paper
birch in full sun (A) and in shade (B) in June of 1996. Asterisks indicate significant
effects of defoliation within within fertilization treatment (p g 0.05) ........................ 52

Figure 2.8. Diameter growth (A), height growth (B), and overall height (C) of paper
birch and sugar maple in 1994, 1995. and 1996. Asterisks indicate significant differences
in species growth or overall height within years (p 5 0.05) ................................... 55

Figure 2.9. Effects of light availability on diameter growth (A) and height growth (B) in
1995 and on diameter growth (C) and height growth (D) of paper birch and sugar maple
in 1996. Asterisks indicate significant effects of light within species (p 5 0.05)..... . . . . .57

Figure 2.10. Effects of nutrient availability on diameter growth (A) and height growth
(B) in 1995 and on diameter growth (C) and height growth (D) of paper birch and sugar
maple in 1996. Asterisks indicate Significant effects of fertilization within Species (p <_;
0.05) ................................................................................................... 59

Figure 2.11. Effects of defoliation on height growth of paper birch and sugar maple in
1996 (A) and the influence of fertilization on the magnitude of the defoliation effect (B).
In (A) asterisks indicate significant effects of defoliation within species (p g 0.05). In (B)
bars with the same letter above them are not significantly different as or = 0.05 ........... 60

xiv

Figure 2.12. Effects of shade on the concentrations of carbon-based secondary
metabolites of paper birch and sugar maple: condensed tannin levels in June of 1996 (A),
total phenolic levels in June of 1996 (B), and condensed tannin levels in June of 1997
(C). Asterisks indicate significant effects of light treatment within species ([7 _<_ 0.05)...63

Figure 3.1. Gypsy moth pupal masses and duration of larval period in 1996. Data within
host species form distinct clusters of high and low pupal maSS/ development time
combinations. Insects with pupal masses < 0.80 g that completed larval development
within 34 d were considered male ................................................................ 92

Figure 3.2. Relative growth rates (RGR) of gypsy moth measured over 48 h during the
fourth instar and from egg hatch to pupation (“larval period”) on paper birch and sugar
maple in 1996 (A). Relative consumption rates (RCR) (B), approximate digestibility
(AD) (C), and efficiency of conversion of digested food to biomass (ECD) (D), of fourth
instar gypsy moth measured over 48 h. Asterisks indicate significant Species effects on
insect growth or nutritional indices (p 5 0.05) .................................................. 101

Figure 3.3. Relative growth rates (RGR) of gypsy moth measured over 48 h during the
second and fourth instar on paper birch and sugar maple in 1997 (A). Relative
consumption rates (RCR) (B), approximate digestibility (AD) (C), and efiicienCy of
conversion of digested food to biomass (ECD) (D), of fourth instar gypsy moth measured
over 48 h. Asterisks indicate significant species effects on insect growth or nutritional
indices (p _<_ 0.05) ................................................................................... 104

Figure 3.4. Effects of fertilization, influenced by shade, on gypsy moth relative growth
rates as measured over the total larval period in 1996 (A) and growth of second instar
gypsy moth measured over 48 h in 1997 (B). Asterisks indicate significant effects of
fertilization on insect growth (p 5 0.05) ........................................................ 106

Figure 3.5. Interactive effects of fertilization and defoliation on the relative growth rate
(RGR) of second instar gypsy moth on paper birch and sugar maple in 1997. Asterisks
indicate significant effects of fertilization within defoliation treatments of birch or maple
(p 5 0.05) ............................................................................................. 108

Figure 3.6. Interactive effects of defoliation and light treatment on digestive efficiencies
(ECD) (A) and consumption rate (RCR) (B) of fourth instar gypsy moth in 1996.
Asterisks indicate significant effects of defoliation within light treatments (p _<_. 0.05)
......................................................................................................... 109

Figure 3.7. Relationship of defoliation with gypsy moth relative growth rate (RGR) as
measured over the total larval period on paper birch (A) and sugar maple (B) in

1996 .................................................................................................. 112

Figure 3.8. Relative growth rates (RGR) of Whitemarked tussock moth measured over 48
h during the fourth instar and from egg hatch to pupation (“larval period”) on paper birch

X V

and sugar maple in 1996 (A). Relative consumption rate (RCR) (B). approximate
digestibility (AD) (C), and efficiency of conversion of digested food to biomass (ECD)
(D), Of fourth instar Whitemarked tussock moth measured over 48 h. Asterisks indicate
significant species effects on insect growth or nutritional indices (p 5 0.05) ............. l 15

Figure 3.9. Relative growth rates (RGR) of Whitemarked tussock moth measured over 48
h during the second and fourth instar on paper birch and sugar maple in 1997 (A).
Relative consumption rate (RCR) (B), approximate digestibility (AD) (C), and efficiency
of conversion of digested food to biomass (ECD) (D), of fourth instar Whitemarked
tussock moth measured over 48 h. Asterisks indicate significant species effects on insect
growth or nutritional indices (p 5 0.05) ......................................................... l 18

Figure 3.10. Interactive effects oflight and defoliation on relative growth rate (RGR) of
fourth instar Whitemarked tussock moth in 1996. Asterisks indicate significant effects of
shade on growth within defoliation treatments at or = 0.05 ................................... 120

Figure 3.11. Interactive effects of fertilization and shade on Whitemarked tussock moth
relative growth rates measured over the total larval period in 1996. Asterisks indicate
significant effects of fertilization within light treatments (p 5 0.05) ....................... 121

Figure 3.12. Interactive effects of light and defoliation on digestive efficiencies (ECD) of
fourth instar Whitemarked tussock moth feeding on paper birch (A) and sugar maple (B).
Asterisks indicate significant effects of defoliation on digestive efficiencies within light
treatments (p 5 0.05) ............................................................................... 123

Figure 3.13. Effects of defoliation on relative growth rate (RGR) of Whitemarked tussock
moth measured over the entire larval period on paper birch and sugar maple. Asterisks
indicate significant effects of defoliation within tree (p 5 0.05) ............................. 124

Figure 3.14. Relationship between percent defoliation and Whitemarked tussock moth
relative growth rate (RGR) as measured over the total larval period (A). and larval
development time on birch (B) in 1996 ......................................................... 125

Figure 3.15. Relative growth rate (RGR) of forest tent caterpillar measured over 48 h
during the fifth instar on paper birch and sugar maple in 1997 (A). Relative consumption
rate (RCR) (B), approximate digestibility (AD) (C). and efficiency of conversion of
digested food to biomass (ECD) (D), of fifth instar forest tent caterpillar measured over
48 h. Asterisks indicate Significant species effects on insect growth or nutritional indices
(p 5 0.05) ............................................................................................. 127

Figure 3.16. Relationship between foliar nitrogen content of paper birch (A) and sugar
maple (B) and the relative growth rate (RGR) of fourth instar gypsy moth in 1997.129

Figure 3.17. Relationship of maple condensed tannins (A) and total phenolic

concentrations (B) with Whitemarked tussock moth relative growth rate (RGR) measured
over the total larval period in 1996 .............................................................. 131

xvi

CHAPTER 1:

INTRODUCTION

Forest ecosystems vary in light and nutrient availability across a range of spatial
and temporal scales. This variability is driven by both natural and anthropogenic
perturbations which alter availability of resources to forest trees. For example. light
conditions are altered as a result of gap formation from tree fall, canopy closure over the
course of succession, logging, and road construction though forested areas. Global levels
of available light may decrease with increased cloud cover as atmospheric carbon dioxide
levels increase (Ayres 1993). Nutrient levels in forest soils vary seasonally with shifting
input and output of nutrients related to leaf-fall, leaching, and erosion. Tree fall releases
nutrients on a local level, while atmospheric nitrogen deposition resulting from
industrialization is increasing nutrient levels in soils on a global scale (Vitousek et al.
1997)

Environmental factors can have large impacts on plant physiology, influencing
both resource acquisition and resource allocation. A reduction in available light
decreases photosynthetic assimilation, thereby limiting carbon available for growth.
primary and secondary metabolism. and storage. Plants may acclimate to reduced light
by reallocating dry matter. and by changing leaf morphology. leafchlorophyll. and
protein content. decreasing respiration. decreasing enzyme activity. decreasing electron

transport activity. altering whole plant architecture (Holt 1995). and increasing,

decreasing, or maintaining photosynthetic rates (Turnbull 1991, Givnish 1988. Holt 1995.
Mulkey et al. 1993).

Nitrogen deficiencies can result in reduced levels of foliar nitrogen (Floater 1997).
in turn reducing photosynthetic rates (Reich et al. 1991). Fertilization can lead to
increased plant growth (Linder and Rook 1984. McDonald et al. 1986. Bryant et al. 1987.
Muzika 1993), and increased allocation to above ground biomass production. Stored
starch (Wargo et al. 1972). leaf morphology (Linder and Rook 1984). and levels of
defensive compounds (Bryant et al. 1992) also are responsive to nutrient availability.

In addition to abiotic stressors such as low levels of light and nutrients. trees
respond to biotic stressors such as herbivory. Plant susceptibility to herbivores is
influenced by the abiotic environment. Loss of leaf area resulting from herbivory may
result in increased photosynthetic rates in remaining foliage (Ovaska et al. 1992. Lovett
and Tobiessen 1993), and decreased (Schaffer and Mason 1990. Fox and Morrow 1992).
or in some cases compensatory growth (Houle and Simard I996). The location and
severity of herbivory can influence patterns of nitrogen distribution in the plant canopy
(Stockhoff 1994). F oliar defenses may increase or decrease following herbivory
(Haukioja et al. 1985, F aeth 1992, Bryant et al. 1992) with responses occurring in the
season of damage (Haukioja 1980) or in subsequent years (Parker and Houston 1971.
Haukioja 1990. Tuomi et al. 1990).

Physiological responses of forest trees to changes in their environments in turn
can affect the ability ofthese trees to defend their foliage against insect herbivores. In
most cases, insect responses have been found to correspond to associated changes in

foliar toughness, water content, nutrient content, or levels ofdefensive compounds.

Because different insect Species respond differently to particular defensive compounds,
and may differ in nutritive requirements, it is often inaccurate to make statements in
regards to plant defenses without examining specific responses of herbivores.

Shade stress has been associated with both increases (Bryant et al. 1987,
Bultzman and Faeth 1988, Nichols—Orians 1991, Dale and Causton 1992. Lowman 1992.
Shure and Wilson 1993, Johnson et al. 1997) and decreases (Coley 1983. Lincoln and
Mooney 1984, Bultzman and Faeth 1988, Bassett 1991, Koricheva et al.1998) in
herbivory and/ or herbivore densities. Nutrient availability also has variable effects on
herbivores. Fertilization has generally corresponded to increased performance of
individual herbivores, but can have negligible or negative effects on herbivores at the
population level (KytO et al. 1996). Defoliation-induced changes in host foliage may
increase or occasionally decrease the host‘s resistance to the defoliating herbivore ( T uomi
et al. 1988, Karban and Baldwin 1997) or co—occurring species in the same feeding guild
(Denno et al. 1985).

As indicated above, studies on single environmental variables, such as light.
nutrient availability, and herbivory have given insight into the independent effect of these
variants on plant physiology. However, it is rare that in natural settings plants will be
facing only one of these stresses. With this in mind, recent research has focused on plant
responses to multiple stresses (Chapin et a1. 1987, Chapin 1991). The number of studies
on interactions among multiple stresses remains relatively low, but is growing as
evidence builds that there is an interactive aspect of plant response to environmental

stress.

When examining stress effects on trees, it is important to keep in mind the life
history strategy of the tree species of interest. Different plant species have evolved
different strategies for dealing with environmental perturbations, and other selective -
forces in their habitats (e.g. competition and herbivory). Species adapted to low resource
environments are Often slower growing, more highly defended, and less responsive to
environmental perturbation than species adapted to high resource environments (Coley
1983, Bryant et al. 1983, Bryant et al. 1985. Coley et al. 1985).

Although at present no one theory sufficiently explains or predicts plant responses
to environmental stresses the carbon/ nutrient balance hypothesis (Bryant et al. 1983.
Tuomi et al. 1984) and the growth/ differentiation balance hypothesis (Loomis 1932.
Lorio 1986, Herms and Mattson 1992) are closely related theories that have been useful
in conceptualizing potential mechanisms of plant response to their environments. Both
theories address resource limitations that restrict and guide patterns of resource allocation

among life processes, especially between growth and defense.

The objectives of this thesis were two fold. First, I sought to determine the
simultaneous effects of light, nutrient availability. and natural defoliation on resource
acquisition and allocation patterns of sugar maple (Acer saccharum Marsh.) and paper
birch (Betula papyrifera Marsh). These Objectives are addressed in Chapter 2. Second. I
sought to determine corresponding effects of treatments on the lepidopteran defoliators
gypsy moth (Lymantria dispar (L.) Lymantriidae). forest tent caterpillar (ll/ulucosmmr

dis's'tria Hubner, Lasiocampidae), and Whitemarked tussock moth (Orgvia leucosligmu (.1.

Smith), Lymantriidae) as mediated through effects on the quality of host foliage. These
objectives are addressed in Chapter 3.

Light treatments were designed to mimic roughly successional canopy closure.
with the first three years since planting Spent in high light environments followed by two
years of shade which simulated overstory closure. In the final year of the experiment. the
shade was eliminated and once again all trees were exposed to full sun. mimicing gap
formation.

In contrast to most studies which have used artificial defoliation treatments. forest
tent caterpillar larvae were used to naturally defoliate randomly selected trees.
Defoliation effects on foliar traits were measured both the season of defoliation and the
subsequent season so that any delayed effects could be detected. Carbon acquisition was
measured as net photosynthetic rate. while nitrogen assimilation was indirectly reflected
by foliage nitrogen content. Treatment effects on resource allocation were measured as
changes in plant growth. specific leaf mass. and foliar concentrations of total phenolics
and condensed tannins. Treatment effects on herbivore resistance were determined by
quantifying larval survival, development time. relative growth rates and nutritional
indices, as well as pupal mass.

Paper birch and sugar maple trees have contrasting life history Strategies. with
birch being a fast-growing, shade intolerant. early-successional species and sugar maple
being a slow-growing. shade tolerant. climax species. To my knowledge. this is the first
study to examine simultaneous responses of both a pioneer and a climax species to
simultaneous Shade. nutrient enrichment, and defoliation. An improved understanding of

the interrelated nature ofplant responses to stress. coupled with an improved

understanding of the mechanistic links between environmental factors and insect
performance, is essential for developing accurate predictions of the effects of natural and
human disturbance on forested ecosystems, and on the herbivores residing within these
ecosystems. Understanding how tree stress in turn affects populations of the outbreak
defoliators gypsy moth and forest tent caterpillar is especially important, as these species
are the most important defoliators of deciduous forest trees in North America (Mattson et
al. 1991). Defoliation effects on host quality can illuminate how outbreaks of one species
may indirectly regulate the population dynamics of another through host-mediated
interspecific competition. A mechanistic understanding of environmental effects on tree
physiology and herbivore resistance is essential for understanding the dynamics of forest

ecosystems, and for their effective stewardship.

CHAPTER 2:
EFFECTS OF LIGHT, NUTRIENT AVAILABILITY AND DEFOLIATION ON
RESOURCE ACQUISITION AND ALLOCATION IN PAPER BIRCH

(Betula papyrifera) AND SUGAR MAPLE (Acer saccharum)

INTRODUCTION

Forest ecosystems are mosaics of light, nutrient, and water resources. The
distribution of tree species is determined in part by their differential responses to given
resource conditions. The resources available to an individual tree can change
dramatically over its lifetime as stochastic events alter the environment. Gap formation
after tree-fall can increase the availability of both light and nutrients. Temporal and
spatial distributions of resources also shift as community succession progresses. Canopy
closure reduces the amount of light available to understory trees; neighbours compete for
nutrients and water. Environmental factors can have large impacts on plant physiology.
influencing both resource acquisition and resource allocation. Patterns of resource
allocation among growth, reproduction, storage, maintenance, and defense in turn
determine future ability both to compete for resources and to retain resources previoUsly

acquired.

Life History Differences in Resource Allocation.
Allocation of resources to any one process (growth. reproduction. storage.

maintenance. defense) necessitates a reduction of resources allocated to others (Cody

1966, Mooney 1972, Bazzaz et al. 1987). In environments where resource limitations
constrain an individual’s ability to replace lost tissue (e.g. foliage loss resulting from
herbivory), resource retention may be especially important (Coley et al. 1985). Perhaps
for this reason, species found in low resource enviromnents, especially those in low light
areas (typically slow growing, shade tolerant species) often have higher levels of
constitutive defenses than Species typical of resource rich environments (Bryant et al.
1983, Bryant et al. 1985, Coley et al. 1985, Mole et al. 1988. Poorter 1989, Nichols-
Orians 1991, Coley 1993, Shure and Wilson 1993, Dudt and Shure 1994. Klepzig et al.
1995, Coley and Barone I996). The occurrence offast-growing, poorly defended plants
in high-resource areas and slow-growing, well-defended plants in low-resource areas has
been observed in many systems both among (Coley 1983, Bryant et al. 1985, Bryant et al.
1992, Walters et a1 1993, Ashton and Berlyn 1994) and within (Winn and Evans 1991.
Niesenbaum 1992, Dale and Causton 1992) species. Slow growth in species adapted to
low resource environments can be viewed as a consequence of selection for traits that
enable success in unfavorable conditions (Chapin 1991, Lambers and Poorter 1992).
intrinsic differences in abilities to acquire resources or hormonally stimulate growth
(Chapin 1991), or as a result of environmental constraints on growth (see Carbon/
Nutrient Balance Hypothesis, below).

The extent to which resource allocation patterns are genetically determined versus
the extent to which they are determined by the environment is uncertain. Phenotypic
plasticity in plant responses to environmental stress has been shown to be greater in
Species adapted to high-resource (especially high light) environments (Smith 1991.

Riddoch et al. 1991. Thompson, Huang and Kiedemann 1992. Thompson. Kriedemann

and Craig 1992, Chazdon and Kaufmann 1993. Ashton and Berlyn 1994, Lusk et al.

1997).

Effects of Light, Nutrients, and Defoliation on Resource Acquisition and Resource
Allocation.
L_igh_t.

Light is an essential resource for higher plants. Light levels to which forest trees
are exposed vary from deep shade, with less than 2% of incident photosynthetically active
radiation (PAR) penetrating discrete openings in dense forest canopies (Larcher 1995) to
levels near to those in full sun at the center of large gaps. Woody forest species are often
classified as shade tolerant or intolerant. Classically, this division was based upon
perceived abundance in the forest understory, with abundant understory species identified
as shade-tolerant (Baker 1949). More recent studies quantitatively base classifications on
relative growth rates and/ or low light survivorship (Kitajima 1994, Kobe et al. 1995.
Walters and Reich 1996). Shade intolerant species generally grow faster in high relative
to low light and/or have reduced survival in low light. Additional physiological
differences between shade tolerant and intolerant species are noted by Givnish (1988).

A reduction in available light limits photosynthetic assimilation, thereby limiting
carbon available for growth, primary and secondary metabolism, and storage. Plants may
acclimate to reduced light by reallocating dry matter, changing leaf morphology.
chlorophyll and protein content. decreasing respiration. decreasing enzyme activity.

decreasing electron transport activity. altering whole plant architecture (Holt 1995). and

9

increasing, decreasing, or maintaining photosynthetic rates (T urnbull 1991. Gil

Holt 1995, Mulkey et al. 1993).

Low light levels often result in reduced plant growth (Schaffer and Mason 1990.
Denslow et al. 1990, Niesenbaum 1992, McKee 1994, Groninger et al. 1996. Johnson et
al. 1997), especially in shade intolerant species. Decreased irradiance has been associated
with decreased nitrogen content in foliage and a range of other plant tissues (Hollinger
1996, Schaffer and Mason 1990, Dale and Causton 1992). although in some studies no
effect of irradiance on nitrogen content was observed (Cui and Caldwell 1997). Because

light is used for carbon-fixation, Shade may limit the carbon content of plant tissues.

Leaf morphology is often altered by changing light environments as plants
attempt to maintain adequate photosynthetic rates. Shaded individuals often have thinner
but larger leaves than conspecifics in the sun (Potter 1992). Leaf mass per unit area
within species correspondingly tends to increase with increased irradiance (Niinemets and
Kull 1994, Gottschalk 1994), and among species is higher in shade intolerant species
(Niinemets and Kull 1994).

While plant species adapted to shade may have higher levels of constitutive
defenses than Species adapted to sun due to differing life history strategies (discussed
above), the opposite pattern seems to hold within a species (Coley 1993). Shaded
individuals are often less defended than conspecifics in higher light environments (Bryant
et al. 1992. Coley 1993, Dudt and Shure 1994, Johnson et al. 1997). Shade-induced

reduction Ofcarbon-based foliar defenses may be especially pronounced in nutrient-rich

10

soils, as these soils promote growth and thereby enable increased use ofcarbon in

primary rather than secondary metabolism (Bryant et al. 1983).

Nutrients.

Although nitrogen is one ofthe most abundant elements in Earth‘s atmosphere. it
is one of the most biologically limiting nutrients because it cannot be used by plants in its
most common form (sz)~ Among its many uses. nitrogen is richly employed in
photosynthetic cells where it is an element of light-harvesting apparati and photosynthetic
enzymes (Chapin et al. 1987). Much research has documented a strong correlation
between foliar nitrogen and plant photosynthetic rates (Gulmon and Chu 1981. Mooney
et al 1981, DeJong 1982, Field et al. 1983, Field and Mooney 1986, Reich et al. 1991).
Alleviation of nitrogen deficiencies in plants by fertilization can lead to increased foliar
nitrogen (Floater 1997. Lawler et al. 1997). Therefore. individuals grown in nutrient rich
areas may have increased photosynthetic capacities relative to individuals in nutrient-
deficient regions.

Fertilization has been shown to increase plant growth (Linder and Rook 1984.
McDonald et al. 1986, Bryant et al. 1987. Muzika 1993), perhaps in association with
increased photosynthetic carbon acquisition. Fertilization can increase above ground
plant growth proportionally more than root growth. resulting in decreased root/ shoot
ratios.

Nutrient addition has been correlated with increased starch storage (Wargo et al.
1972, McDonald et al. 1986. Burke et al. 1991). and increased leaf size in hardwoods

(reviewed in Linder and Rook 1984). Increased foliar nitrogen levels have generally been

I l

associated with decreased polyphenolic levels (Bryant et al. 1992, Muzika 1993. Kinney
et al. 1997, Lawler et al. 1997), perhaps as a result of decreased carbon/ nitrogen ratios in

plant tissues (see Carbon/ Nutrient Balance Hypothesis, below).

Defoliation.

In addition to abiotic stresses such as low levels of light and nutrients. plants may
be exposed to herbivory. a key biotic stress. Loss of leaf area resulting from herbivory
may result in increased photosynthetic rates in remaining foliage (Ovaska et al. 1992.
Lovett and Tobiessen 1993). Often this compensatory photosynthesis is not sufficient to
maintain needed assimilation, and growth rates are reduced in the presence of herbivores
(Schaffer and Mason 1990, Fox and Morrow 1992, May and Killingbeck 1995), although
in some cases compensatory increases in growth have been observed (Houle and Simard
1996).

Modelling work has indicated that the location and severity of herbivory can
influence patterns of nitrogen distribution in the plant canopy (Stockhoff 1994).
Defoliation is often associated with reductions in levels of foliar nitrogen (Fox and
Morrow 1992). Perhaps for this reason, fertilization has been shown to mitigate
defoliation-induced resistance (Hunter and Schultz 1995). F oliar carbon content is less
effected by herbivory than foliar nitrogen content. as carbon can be mobilized from
storage in response to damage (Wargo et al. 1972).

Defoliation has been both positively and negatively correlated with levels of foliar
phenols and other foliar chemical defenses (Haukioja et al. 1985. F aeth 1992. Bryant et

al. 1992). Plant responses to defoliation may be a result of shifting carbon/ nitrogen

17

n—

balances in individuals (Bryant et al. 1983, see below), and/ or may be elicited by specific
herbivore cues (e. g. salivary chemicals) which trigger defenses, as natural defoliation
sometimes elicits a stronger response from the plant than artificial foliage removal
(Haukioja and Neuvonen 1985). Defoliation induced responses can occur in the season
of damage (Haukioja 1980) or in following years (Parker and Houston 1971. Haukioja

1990. Tuomi et al. 1990).

Environmental Variation at the Scale of the Individual.

While much research has focused on differences among individuals subjected to
different levels of environmental stresses, less attention has been given to the variation
that an individual will experience over its lifetime. As discussed above, both light and
nutrients vary over the course of succession and in response to disturbance. Defoliation.
too, varies annually. Although there is generally some base level of defoliation
experienced by an individual. the cyclic population dynamics of many outbreak
defoliators result in high herbivory stress in some years and in relatively negligible stress
in others years.

The response of plants to long-term successional changes in light availability
(declining with canopy closure and increasing with random gap formation. which was
found to contribute over 65% of the total seasonal photosynthetically active radiation
received in the understory of one northern United States hardwood forest. C anharn et al.
1990) may have important effects on successional trajectories. This would especially be
true if species that differ in life history strategies also differ in the nature of their response

to fluctuations in the light environment. Response time, duration. and strength of plant

responses may all be important variables to consider when examining the impact of long-
term responses on plant fitness and on interactions among individuals of the forest
community. For example, herbivore populations may be indirectly affected by
successional changes as they consume plant parts morphologically or chemically

responsive to changing environments.

Theory on Mechanisms Driving Resource Allocation Patterns.

Physiological processes underlying observed plant responses must be
examined to understand and predict interactions among stress factors. At present no
single theory sufficiently explains or predicts plant responses to environmental stresses.
The carbon/ nutrient balance hypothesis (Bryant et al. 1983, Tuomi et al. 1984) and the
growth/ differentiation balance hypothesis (Loomis 1932. Lorio 1986. Herms and
Mattson 1992) are closely related theories that have been useful in conceptualizing
potential mechanisms of plant response to their environments. Both theories address
resource limitations that restrict and guide patterns of resource allocation among life

processes including growth, storage, maintenance, and defense.

C arbon/ Nutrient Balance Hypothesis.

The carbon/ nutrient balance hypothesis (Bryant et al. 1983. Tuomi et al. 1984)
predicts allocation to growth and secondary metabolism based upon the carbon/ nutrient
(or nitrogen) ((‘/N) ratio ofthe plant. Specifically. carbon-based secondary metabolism is
predicted to vary directly with the UN ratio ofthe plant. Growth receives resource

priority in favorable environments. but can be limited by either carbon or nitrogen

14

availability. Because moderate nutrient deficiency (infertile soils, loss ofnutrients to
herbivores) limits growth more than photosynthetic capacity, C/N ratios will be increased
in nutrient deficient plants and excess carbon will be available for allocation to the
production Of secondary metabolites. In shaded environments. low levels of
photosynthetically active radiation will limit carbon fixation. reducing C/N ratios.
Limited carbon availability will decrease concentrations of carbon based secondary
metabolites. However, nutrients that accumulate in excess of growth requirements (e. g.
when other factors limit growth, such as low light) may be converted to nitrogen-based
defensive compounds such as alkaloids.

It is relevant to note that the carbon/ nutrient balance hypothesis assumes that
there is no “cost” of defense, as defense compounds are produced with resources acquired
in excess of those needed for growth (Bryant et al. 1983, Tuomi et al. 1988). However,
nutrient loss due to herbivory can induce higher levels of defensive compounds in
regrowth foliage in the same or the following season (Neuvonen and Haukioj a 1984.
Neuvonen et al. 1987, Haukioja 1990, Tuomi et al. 1990) by shifiing the C/N balance of
the plant upward. This “delayed induced resistance” Should be especially strong when
defoliation occurs early in the season when leaves contain proportionally more nitrogen
than carbon. Because they result from nutrient loss, defoliation-induced responses are
predicted to be weak or non-existent in fertilized environments. Empirical studies show
that induced resistance can also occur within damaged and undamaged foliage of the host
tree within hours or days ofdarnage (Haukioja 1980). Environmental effects on the
strength of “rapid induced resistance” are not as easily predicted by the carbon/ nutrient

balance hypothesis.

Growth/ Differentiation Balance Hypothesis.

The growth/ differentiation balance hypothesis (Loomis 1932, Lorio 1986. Herms
and Mattson 1992) subsumes the carbon/ nutrient balance hypothesis, placing it within an
expanded framework, which broadens the range of predictions. Unlike the carbon/
nutrient balance hypothesis, the growth/ differentiation balance hypothesis emphasizes
physiological processes over environmental factors and therefore predicts shifts in
resource allocation based on a wider range of environmental conditions including nutrient
availability, temperature, and drought (Herms and Mattson 1992). The growth/
differentiation balance hypothesis also addresses the importance of plant developmental
stage, which is not addressed by the carbon/ nutrient balance hypothesis.

The growth/ differentiation balance hypothesis distinguishes among resource
allocation to "processes and structures further enhancing resource acquisition“ (i.e.
growth; Herms and Mattson 1992, p.297) versus “non-growth processes and structures
required to retain and make efficient use of resources under adverse environmental
conditions and/ or intense herbivory” (i.e. differentiation; Herms and Mattson 1992.
p.297). While many growth and differentiation processes are co-dependent. some
differentiation processes (e.g.. manufacture of anti-herbivore secondary metabolites) may
limit growth by diverting resources from growth processes (Herms and Mattson 1992).
This implies a potential cost of defense not acknowledged by the carbon/ nutrient balance
hypothesis.

The growth] differentiation balance hypothesis holds that the effects of

environmental variation on allocation patterns of constitutive carbon-based secondary

16

metabolism can be predicted by the impact of the variant on plant growth relative to
photosynthesis (resource acquisition). Environmental factors that slow growth more than
photosynthesis (moderate nutrient deficiency. drought. temperature stress) are predicted
to increase carbon availability for secondary metabolism and storage reserves because
carbon will accumulate in excess of growth requirements. Growing meristems are strong
carbon sinks. Weak sink strength of slow growing meristems in resource-limited
environments will reduce phloem transport. thus increasing carbon accumulation in
strong source leaves which continue to maintain high rates of photosynthesis. This
carbon is often diverted to biosynthesis ofcarbon-based secondary metabolites.
Developmental stage of the plant can also affect this relationship (Herms and Mattson
1992), with young rapidly growing tissues being stronger sinks than older tissues. Active
meristems utilize carbon in the processes necessary for growth, such as protein synthesis
and respiration. Limited carbon constrains secondary metabolism in strong sinks. For
factors such as shade, which limit source strength. both growth and secondary
metabolism are predicted to be carbon limited. Effects ofenvironmental stresses on
defoliation induced resistance are not addressed by the growth/ differentiation balance

hypothesis.

Objectives and Predictions.

The objectives ofthis study were to detemiine the simultaneous effects of light.
nutrient availability. and natural defoliation on resource acquisition and allocation
patterns of sugar maple (Acer saccharum Marsh.) and paper birch (BCIU/u pupyri/eru

Marsh). Shade treatments were designed to mimic successional canopy closure. with all

17

trees grown in full sun for three years, after which half the trees were exposed to two
years of shade, thereby simulating natural overstory closure. In the final year of the
experiment, all trees were once again exposed to full sun, mimicking gap formation or
overstory defoliation. Defoliation effects on foliar traits (see below) were measured both
during the season of defoliation and during the following season so that any delayed
effects could be detected.

Treatment effects on plant acclimation to shade were measured as morphological
changes in foliage (Specific leaf mass and leaf area) and light response curves. Treatment
effects on acquisition of carbon were measured as net photosynthesis. while effects on
acquisition of nitrogen were estimated by foliage nitrogen content. Resource allocation
was determined by quantifying plant growth and levels of total phenolics and condensed
tannins in foliage. Phenolics (assayed in both the total phenolic measure and the
condensed tannin measure) are the predominant class of carbon-based secondary
metabolites in birch and maple and have been implicated in constitutive and induced
herbivore resistance in a wide range of systems (F eeny 1970, Haukioja et al. 1985; bUt
see Bemays 1981, Ayres et al. 1997).

I focused on paper birch and sugar maple because these Species have contrasting
life history strategies, and therefore are expected to respond differently to environmental
stresses. Additionally. both Species are economically important. have overlapping ranges
in Michigan. and are important hosts ofthe forest tent caterpillar. ll/Iulucosomu dissiriu
(Hubner, Lepidoptera: Lasiocampidae) (Goodman et al. 1990. SaffOrd et al. 1990). which

was used to implement the natural defoliation treatment (see below). To my knowledge.

18

this is the first study to examine simultaneous effects of defoliation, light. and nutrient
stress on species with contrasting life history strategies.

Paper birch is a fast-growing, early-successional, shade-intolerant species (Loehle
1988, Safford et al. 1990). This species develops best on cool moist Sites and is sensitive
to nutrient availability (Safford et al. 1990). It produces a diversity of foliar carbon-based
secondary metabolites (Palo 1984).

At the opposite end Of the life history continuum. sugar maple is a Slow-growing.
late-successional, shade-tolerant Species (Leohle 1988. Goodman et al. 1990). This
species is also sensitive to soil fertility and is common to cool, moist sites (Goodman et
al. 1990). It produces both carbon- and nitrogen-based foliar defenses (Barbosa et al.

1990).

Based upon life history theory, I predicted that: (1) light response curves of birch
would have higher light-saturated photosynthetic rates and higher light saturation points
than those of maple; (2) in high resource environments (high light, nutrients), birch
would have faster growth and higher ambient photosynthetic rates than maple; (3) birch
responses (growth, photosynthetic activity, foliar chemistry) to environmental variability
(light, nutrients, defoliation) would be more plastic than maple responses; (4) birch
foliage would contain lower concentrations of defensive compounds than maple foliage
(lower concentrations of condensed tannins and total phenolics than maple foliage). ‘

Based upon the growth/ differentiation hypothesis. I predicted that ( I) if shade
decreases photosynthesis of birch and maple. then they would have lower levels of C-

based secondary defense chemicals (condensed tannin and total phenolic concentrations)

19

than conspecifics grown in the sun; (2) if fertilization increases growth more than
photosynthesis, then carbon-based defenses would be lower in fertilized treatments; and
(3) if defoliation decreases foliar nitrogen, then levels ofcarbon-based defenses would be

higher in defoliated trees in 1996 and 1997.

MATERIALS AND METHODS
Experimental Site and Design.

Studies were conducted in established research plots at the Dow Gardens in
Midland, MI. Climatological information for Midland is provided by the US.
Department of Commerce National Oceanic and Atmospheric Administration (1971).
The research plots consisted of forty-eight l In2 cells arranged in 6 rows of 8. Rows were
spaced 2 m apart. Walls of cells were lined with 30 mil poly-vinyl chloride liners
extending 1 m into the ground and 10 cm above ground level. Overlapping edges were
heat-sealed, which effectively isolated the soil environment of each cell. Bottoms of cells
were open. Neighbouring rows were grouped in sets of 2 to comprise blocks (3 blocks
total).

One sugar maple and one paper birch tree were transplanted into each cell in April
of 1992, at which point maple were entering their fifth growing season and birch their
third. Trees within species are open-pollinated halfsiblings. The maternal source of
maple was native to Minnesota. close to the Wisconsin border. while the maternal source
of birch was native to eastern Wisconsin.

Light intensity. soil nutrients, and defoliation treatments were administered in

complete factorial fashion. The experiment was designed as a randomized. complete

20

block, split plot with light as the whole plot and nutrients and defoliation as subplots.
Each of the 16 treatment combinations (species*light*nutrient *defoliation) were

replicated twice in each of the three blocks for a total of 96 trees in the experiment.

Experimental Treatments.
Light Manipulation.

One row of each block was randomly selected to be shaded, while the other
experienced ambient light conditions. Shading was accomplished by using a wooden
frame to support an 80% black shade cloth (E. C. Geiger, Inc.) over the top of selected
rows and hanging 30% shade cloth from the top of the frame supports to the base of the
canopy. In both 1995 and 1996, plots were shaded from May (as leaves reached full size)
until initiation of leaf senescence in October.

Values of quantum flux recorded during photosynthetic measurements (see
below) verified that shade cloth reduced light in treated plots by 80%. Average
photosynthetically active radiation (PAR) measured in full sun on measurement days was
1285 i 48 pmol rn'2 SI and in shade treatments was 273 i 49 pmol m'2 5". Air circulation,
precipitation, insect herbivores. invertebrate predators and avian species were not

noticeably affected by the presence of shade structures.

Soil Nutrients.

Four cells per row were randomly selected to receive a 225 kg/ha/yr of actual

nitrogen (N) fertilizer with phosphorous (P) and potassium (K) applied in the ratio of

21

18:5:9 NPK. The remaining cells were not fertilized. Nitrogen in the fertilizer was in the
form Of urea N (16.05%, with 6.75% sulfur coated for slow-release) and ammonia N
(1.95%) with the balance as nitrate. Phosphorus was in the form of P30, and potassium
in the form of K20. Half of the annual amounts of fertilizer were applied in May and the

remainder in October in 1994, 1995, and 1996.

Natural Defoliation.

Forest tent caterpillar, one of the most important outbreak defoliators of
deciduous trees in North America (Mattson et al. 1991), was used to implement the
defoliation treatment. On 16 May 1996, half of the experimental trees were inoculated
with forest tent caterpillar egg bands, which were held in mesh packets attached to limbs
(10 to 15 egg bands per tree). Egg bands were field collected from quaking aspen
(Populus tremuloides) stands in Cochrane, Ontario on 7 May 1996. Prior to deployment.
egg bands were surface sterilized by rinsing with bleach for 3 minutes. followed by a 5
minute tap water rinse (Grisdale 1968). Egg hatch occurred on 16 May, corresponding
with bud break as it does in natural populations. Trees first broke bud between 8 and 19
May, and all buds had opened by 14-30 May. Larval feeding on control trees was
restricted by manual redistribution of larvae, restraining of overlapping treatment and
control tree branches with twine, and applying sticky bands to the trunks of control trees.
Estimates Of defoliation (percent leaf area removed) were determined as the average of
visual estimates by four independent observers. Defoliation levels were I i 3% and 8 i

3% (range 0-32.5%) in maple and 6 i 3% and 52 i 3% (range 0.5-98.5%) in birch.

22

Forest tent caterpillar avoided eating maple even when continually placed on this
Species. Although some herbivores, including gypsy moth, do not tend to prefer maple
and other hosts which contain alkaloids (Barbosa et al. 1990). this behavior in forest tent
caterpillar was unexpected as Dankert (1995) found forest tent caterpillar to prefer maple

over birch in a study using trees from the same cohort as those used in the present study.

Plant Physiological Responses.
Acclimation to Shade.

Photosynthetic Capacity (Light Response Curves). Between 26 and 31 August.
1996, photosynthetic responses of trees to varying levels of photosynthetically active
radiation were determined for all non-defoliated trees using a CI-301PS open
photosynthesis system (CID, Inc., Vancouver, Washington State USA). Carbon
assimilation was measured for 55-second intervals at 0. 50.100. 200. 300. 450. 600. 800.
and 1200 pmol rn‘I s'l at the leaf surface. To account for changing intensity of light with
distance from the light source to the leaf. the above light levels were achieved by dividing
these target values by 0.61 and then setting the light source intensity to adjusted values.
One leaf per tree was measured. Only fully-expanded leaves in the exterior of the canopy
with minimal or no herbivore or other damage were used. On birch. only short shoot
foliage was sampled. A preliminary study conducted earlier had indicated a possible
Species*nutrient interactive effect on carbon assimilation at maximal light intensity
(Amax). In order to increase power for detecting fertilization effects. a sampling protocol

was developed in which fertilized and non-fertilized trees of the same species and light

treatment were sampled consecutively. Light response curves were fit to an equation of
the general form:

A: ((b*I)/(1+a*l))-Resp.
where A: carbon assimilation rate (measured), I=irradiance (measured). and “a”. “b”. and
“Resp” are parameters fit with the Marquardt-Levenberg algorithm. Parameter “a”
(“curve”) is inversely related to photosynthetic carbon assimilation at light saturation. "b”
indicates the initial slope of the light response curve. and “Resp” is the dark respiration
rate (carbon assimilation at 1:0).

Specific Leaf Mass and LeafArea. Twelve leaves from each tree were collected
from throughout the canopy on 15 June 1995, l 1 June 1996 and 16 June 1997 (base 42°F
degree days at Saginaw, Michigan: 15 June 1995 = 1039, 11 June 1996 = 827. I August
1996 = 2268, and 16 June 1997 = 911; data from Jeff Andresen, Michigan State
University Geography Department). Leaf area was measured using a digital image
analysis system (AgVision Digital Image Analysis System. Decagon Devices. Inc. in
1995 and 1996, and CI400 Computer Image Analysis System, CID. Inc. in 1997). and
average leaf area was calculated. Samples were oven-dried for 48 h at 45°C, then
weighed. Specific leaf mass (SLM) was calculated with the equation [SLM = dry mass/

leaf area (g/m2)].

Resource Acquisition.
Photosynthetic Assimilation at Ambient Conditions. Net photosynthetic carbon

assimilation at ambient conditions was measured for all trees using a closed-system

LICOR 6200 infrared gas analyser (LICOR Inc., Lincoln, NE). Measurements were
made on sunny days with clear skies between 100011 and 1300 h EST. In Midland.
morning Skies are typically clear with patchy clouds approaching from the NW between
1230 h and 1300 h. One fully-expanded leaf in full sun with minimal or no damage was
measured per tree. Birch trees were measured on 10 July and 16 July. and maple on 1 1
July and 22 July 1996. Stomatal conductance measurements were made only on the
second measurement date per species. Sample sizes prohibited both species (96 trees)
from being measured on the same day without introducing excessive diurnal variation.
F oliar Carbon and Nitrogen. Dried leaf matter collected for specific leaf mass
measurements (see above) from each tree was ground to pass through #20 mesh in a
Wiley mill, and analyzed for percent carbon and nitrogen content by ignition in a Carbo
Erba model NA 1500 CHN Analyzer with methods described by Daun and DeClercq

(1994).

Resource Allocation.

Height and Diameter Growth. Trunk diameters at 50 cm above ground level were
measured at the beginning and end of 1994, 1995, and 1996 growing seasons (5 May
1994, 10 October 1994, 28 May 1996 and 26 October 1996). Diameters were measured
with callipers consistently oriented east to west to minimize year-to-year variation in
measurements due to trunk asymmetries. Trunk heights were measured with a
telescoping ruler on 12 May and 3 October 1994, 23 September 1995. and 26 October

1996. Increase in diameter and height within years was calculated as the difference

between height (or diameter) measured at the beginning and at the end of each growing

SCOSOD.

F oliar Condensed Tannin and Total Phenolic Concentrations. Concentrations Of
foliar condensed tannins were determined for samples of 12 birch or maple leaves per tree
collected from throughout the canopy on 11 .Iune 1996 and 16 .lune 1997. and for 12
“mature” (Short shoots) and I2 “immature” (axial growth) leaves from each birch tree on
1 August of 1996. Mature leaves had expanded earlier in the season and were noticeably
a darker green color than immature leaves which expanded relatively later in the season.
Total phenolic concentrations were determined for all leaves sampled in 1996 but not in
1997. Leaves were submerged in liquid nitrogen within two minutes of sampling to
prevent enzymatic degradation and phenolic oxidation. Leaves were transported on dry
ice to the laboratory. and stored frozen until they could be lyophilized. Leaves were
ground to pass through #40 mesh in a Wiley mill, and 70-80 mg aliquots were weighed
out for polyphenolic extractions. Except during grinding and weighing. leaves were kept
refrigerated at all times after lyophilization.

Total phenolic activity was determined by the F Olin-Denis assay (Swain and Hillis
1959) using an REA-300 Rapid Flow Analyzer (Astoria-Pacific International. C lackamas.
OR). Phenolics were extracted by 30 minute agitation in 50% methanol. Samples were
run with tannins from Alaska paper birch (Betula resini/era. purified from Alaska
populations by Tom Clausen, University ofAlaska. Fairbanks) and sugar maple (purilied
from the study population by James Nitao. Michigan State University) as standards to

obtain values for percent dry weight total phenolics on a species-specific basis.

26

Condensed tannins concentrations (CT) were determined by the n-butanol
proanthocyanadin assay (Hagerman 1995). Phenolics were extracted from 70—80 mg
foliage samples in 8 ml of 70% acetone at 5°C for 18 h on a shaker table. Tannins from 4
g of ground leaves from bulk samples of each species (collected the same day as foliage
sampled for chemistry) were extracted in 70% acetone and purified for use as standards.
Tannin purification was accomplished by removing acetone from bulk extracts by
rotoevaporation. packing aqueous extract on to a Sephadex LH-20 column. eluting
phenolic monomers by flushing 95% ethanol though the column until a negative Prussian
Blue spot test (Hagerman 1995) was obtained. and collecting larger molecular weight
phenolics by washing the column with 70% acetone until a negative Prussian Blue spot
test indicated all phenols had eluted. Acetone was then again removed by

rotoevaporation and samples were frozen and lyophilized.

Statistical Analyses.

Main and interactive effects of treatments on the initial slope (parameter "b”) and
curve (parameter “a”) of light response curves. stomatal conductance. net photosynthesis
rates, foliar carbon, nitrogen. condensed tannin and total phenolic concentrations and UN
ratios. Specific leafmass. and leafarea were analysed using univariate analysis of
variance (PROC GLM; SAS Institute. Inc. 1990). Tree height. annual height growth. and
annual diameter growth were analyzed using analysis of covariance (ANCOVA: SAS
Institute. Inc. 1990). with May 1994 height and diameter measures as covariates.

respectively. Assumptions of normality and equal variances required for ANOVA were

27

verified with Shapiro-Wilkes normality tests (SAS PROC UNIVARIATE) and visual
examination of normality plots. respectively. Extreme outliers (greater than 3 standard
deviations from the mean) were removed from analyses only if they previously had been
noted as unusual, or ifthe datum was biologically infeasible. Block was treated as a
random effect and species. light, nutrient, and defoliation were treated as fixed effects.
Models were adjusted to account for the split-plot design by testing for shade effects
(whole plot factor) with the block*light mean square error (MSE) in the denominator of
the F statistic. as opposed to the complete model MSE. Data are reported as least square
means i standard errors, unless noted. Standard errors of shade effects were calculated
using block*light as the error term.

Within a species. the relationship between percent defoliation and tree height and
diameter growth. absolute tree height. foliar carbon and nitrogen content. C/N ratios.
condensed tannin, total phenolic content. and specific leaf mass were also tested using
Pearson correlation coefficients (SAS PROC GLM) and treating defoliation as a

continuous variable.

RESULTS
Acclimation to Shade.
Photosynthetic Capacity (Light Response Curves).
Light response curves generated in 1996 for each light*fertilization treatment
combination for birch and maple are reported in Table 2.1. Birch had higher

photosynthetic rates than did maple for all intensities of photosynthetically active

28

Table 2.1. Average photosynthetic assimilation (A. pmoles CO; m'2 s") as a function of
irradiance (1, pmol m'2 s") in non-defoliated trees. See text for measurement and
calculation details. N = 4-6 per treatment combination.

 

Tflment Equiition
Maple
Shaded
Fertilized A = (0.0279 I/ (l + 00193 I)) - 0.6013
Non-fertilized A = (0.0258 1/ (l + 0.0060 1)) — 0.4567

Non-shaded

Fertilized A = (0.0375 1/ (l + 0.01 12 I)) - 0.6299
Non-fertilized A = (0.0207 1/ (1 + 0.0048 1)) — 0.6901
Birch
Shaded
Fertilized A = (0.0327 I/ (1 + 0.0044 1)) — 0.4633
Non-fertilized A = (0.0340 1/ (l + 0.0038 I)) — 0.7217
Non-Shaded
Fertilized A = (0.0295 1/ (1 + 0.0034 1)) — 0.6817
Non-fertilized A = (0.0312 1/ (l + 0.0032 1)) — 0.8150

 

(l

.585qu itxoewqu. gown”?

 

 

2 E 55
2.0 o _ .o of a saw 2: 3o moo _ E2:..z..zw:.aeam
8.0 go So NS 8.0 8.0 moo _ Easing?
85 8d 8.0 SE a: 8a a; _ EaEszsacam
one as N _ .0 3o :3 we; mom _ 25:52
woo 33 8.0 3o 86 Sc woo _ ijaaaam
.11.... mmém 5.1.... wN.wv :3; 050‘s .31.... _NdN Owd «no; a. hvé _ museum
2: who ”2 a SN 23 :3 _2 N S5285
as 85 8.0 ”no ”3 :2 .2 _ gas
a com a :.m a EN 3.. : mom 2: of m .85
802 83. 8.9.. 8:. Emma 8% 0:8 E 8:25 .6 850m

_ .ES cote 2: mm EwrxooE 5:5 358.8 203 Ew: Lo..— moagi Soc—«Q 7m “.5 £9:1 82 ecu 803$ 7m “-8 30E: can .835 7.0. we £05:
com .82 $ 7m N.E 22:: oo. Emmy: 7m NE 29:: o u mad on :ozszBmm so new .833 8:88.. Em: .8 3 5.2533 32m REE was Am 56589:
$23.. 23:: of .5 228825 :2: new nooto 25:5... new 380% .Ew: .wcioos .3 <>OZ< :5: SUV Eonooc .8 momemen Ea 82m>i .N.N o3:

 

 

 

 

 

 

 

 

Maple
Q _,
g .. 0 shade, fert
7 ~ I shade. no fert
7’; 2’ ‘ 0 sun, fert
c. .
g '7' :1 sun, no fert
a E 4 ‘ i
1 6 3 ..
§ 0
m __ 2 -
{‘2’ o
E ' ‘
3 O I t l I
-1 Y 500 1000 1500 2000
-2
PAR (pmol m'2 s")

 

 

 

 

Birch

 

0 shade. fert
I shade, no fert

 

 

 

 

 

Assrmrlatron
(umol C02 m'2 s")

 

0 sun, fert

D sun, no fert
0 I I I 1
-| 500 r 000 r 500 2000

PAR (pmol m2 s")

 

 

 

Figure 2.1. Photosynthetic light response curves for sugar maple (A) and
paper birch (B) grown under factorial combinations of light and nutrient

availability.

radiation (PAR) examined, but the species did not differ in respiration rates (Table ._.2
Figure 2.1). Although both species adjusted their leaf morphology in shaded
environments (see Specific Leaf Mass, below). neither improved their capacity to

assimilate carbon in low light or high nutrient environments (Table 2.2; Figure 2.1 ).

Specific Leanass and LeafArea.

Intrinsic Differences in Specific Leanass and Lea/Area. Birch had greater
specific leaf mass (SLM) than maple in 1995. 1996 and 1997 (Table 2.3). Within a
species, 1997 values tended to be greater than 1996 values. In maple this may reflect the
fact that Shade decreased foliage mass per unit area in 1996 (see below), with the lack of
Shading in 1997 increasing the average mass per unit area of maple foliage. The SLM of
birch increased over the 1996 season. Both mature and immature leaves harvested in
August weighed more per unit area than leaves sampled in June. This was likely a result
oftree-wide variation rather than a function of leaf age. as immature foliage sampled in
August had greater SLM than mature foliage (Table 2.3).

Birch had smaller leafareas than maple in 1995. 1996. and 1997 (Table 2.3 ).

Leaf areas of both species were highest in 1995 before treatment implementation.

Light Effects on Specific Leaf/Mass and Leaf A rea. Shading had no effect on the
specific leaf mass ofbirch or maple in June of 1995 (Table 2.4a). In .Iune of 1996. the
specific leaf mass Of maple was 20% lower in Shaded trees than in trees grown in full sun
(Table 2.5a. Figure 2.2). Light effects on the SLM of birch were not significant in June;

although. as in maple. specific leaf mass tended to be lower in shaded birch. In August.

to)
IJ

Table 2.3. Foliar carbon and nitrogen concentration. carbon/nitrogen (C/N) ratio. condensed tannin and
total phenolic content, specific leaf mass (SLM) and leaf area of paper birch and sugar maple foliage in
1995, 1996, and 1997 (least square means 1‘ standard error).

M

 

Paper Birch Sugar Maple
Carbon (% dry leaf) 49.3 i 0.3 49.4 .+_ 0.3
Nitrogen (% dry leaf) 3.03 i 0.05 2.28 i 0.05
C/N ratio 18.7 i 0.4 19.6 i 0.4
Specific leaf mass (g ml) 59.9 i 1.2 49.6 i 1.2
LeafArea (cm!) 28.7 i 1.0 100.7 x 1.0

Ju_nc_l_9£§

Paper Birch Sugar Maple
Carbon (% dry leaf) 51.1 i 0.3 49.4 i 0.3
Nitrogen ("/0 dry leaf) 4.06 i 0.05 2.96 i 0.05
C/N ratio 12.8 i 0.4 16.9 i 0.4
Condensed tannins (% dry leaf) 1.12 i 0.08 1.32 _+_ 0.09
Total phenolics (% dry leaf) 12.4 i 0.5 21.3 i 0.5
Specific leafmass (g ml) 36.4 :t 1.0 31.1 i 1.0
Leaf Area (cmz) 19.8 :t 0.7 81.4 i 2.5

 

August 1996

 

Mature Birch Immature Birch
Carbon (% dry leaf) 48.8 i 0.3 48.3 i 0.3
Nitrogen (% d1y leaf) 2.60 i 0.05 2.59 i 0.05
C/N ratio 19.3 t 0.4 18.9 i 0.4
Condensed tannins (% dry leaf) 0.78 i 0.08 0.88 i 0.08
Total phenolics (% dry leaf) 1 1.8 i 0.5 12.5 i 0.5
Specific leafmass (g ml) 55.3 i 1.0 58.9 i 1.0
Leaf Area (cmz) 15.5 : 0.9 25.0 i 1.0

Ju_m|_997.

Paper Birch Sugar Maple
Carbon (% dry leaf) 47.1 _+. 0.3 45.9 :r 0.3
Nitrogen (% dry leaf) 2.89 i 0.05 2.19 i 0.05
C/N ratio 16.6 i 0.4 21.3 i 0.4
Condensed tannins (% dry leaf) 1.70 i 0.08 1.05 :r 0.09
Specific leaf mass (g mi) 43.4 i 1.0 36.3 i 1.0
LeafArea (cmj) 19.9 i 0.8 78.4 i 2.1

 

Table 2.4. F—values and degrees of freedom (elf) from ANOVA of blocking, Iight.and nutrient effects and
their interactions on specific leaf mass (A) and leafarea (B) ofbirch and maple foliage sampled on 15 June
1995. There was no defoliation treatment in 1995. F—values for light were calculated with block‘light as
the error term.I

Specific Leaf Mass

 

 

 

 

June 1995
A. Source ofVariation (If Birch Maple
Block 2 2.48 2.61 ft
Light 1 0.02 1.20
Block‘Light 2 0.18 1.02
Nutrient I 0.00 0.00
Light‘Nutrient 1 0.84 0.99
Error df 40 37
Leaf Area
June 1995
B. Source of Variation dl' Birch Maple
Block 2 0.12 2.34
Light 1 1.44 2.94
Block*Light 2 0.10 2.22
Nutrient 1 1.12 2.57
Light*Nutrient 1 0.01 1.19
Error df 40 37

'#=p50.10.*=p50.05.*“*=/)50.0001.

2000.0!” 1...... «0.02“ .. 0.02””;

 

 

 

 

mm mm 0... mm mm mm ,0. 8:0
0m.0 _0.0 N _ .0 :0 N: 0 _ .0 _ 8802.51882.20:
2.0 E EN 00.0 _00 _00 _ 00800898282
20 :0 .. A: 00.0 NS 02 _ 888898:
00.0 00.0 8.0 _2 000 N_ .0 _ 88:08:
00.. mN.0 NN.0 00.0 _00 0N0 _ 28:82.20:
00.0 0N0 NN.0 :0 00.0 00.0 _ 28:82
8.0 20 2.0 00.0 $0 SN N 541.5185 1
00.0 0N8 a E: 00.0 SN 20 _ 08:

a 92 2.0 .. $0 00: a 00.m NN.0 N .85
2am: Lotm :85 2:80:05 :85 Sam—Z 2sz £0.2m \h cocatw> .fio ouLzom
80. 2:... 000. .803. 000. 2:;
wou< baud

mm 0m 0m 00 N0 0... 0. 8:0
00.0 NN.0 .00 00.: N0 a N0.N _ 8888a18:82.28:
8.0 N _ .0 X0 0N0 N00 00.0 _ 888898282
_00 8.0 00.0 N00 :0 8.0 _ 8880...?ij
00.0 03 0. .0 .. 0N0 N00 :0 _ 8888a
02 00.0 00.0 000 N0 .03 _ 8082.28:
8.0 00.0 0N0 N00 80 SN _ 08:82
0: N0: .. :2 :..0 .2 :.N N 85.820
00.. 0N0 a 00.: .. N000 .. NN._N a 03 _ Em:
00.N t... 8.: 00.0 03 N00 .2 N 85

flaw—Z zutm :85 0.585;: 29.5 0.532 2922 cohm xx» COCMCS/ mo venom
80. 250 000. .803. 002 2:...

032 .83 21:0on

_.::2 808 2: m0 Em:t_oo_o :23 8823.8 0.53 Em: .8 0029i
.32 2:: o. co BEES 090:8 3qu 8:0 :85 :8: new. 63. mama}. _ co 829:3 A285 230E025 .8008 89:3 2: E 888.: 088058 82 E0: new

A225 8202:: 8080 0:285 05 wESn 3E8: 0.0.5 :8: wccmfiwto 88:8 :85 .002 2:: _ _ co BEES om0:o.. 080:. 8:0 :25 8 Am: 88 :8. use
Also 30E .82 ”0:809. 20 228825 .85 8:0 380.8 88:08.0 new .8283: .8»: .wExooE .8 <>OZ< :6: Q3 8088: 8 82?: 8:0 02:0? 1.\ .m.~ 0.8...

D.
«.3.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

   

 

 

A June 1996
A 80 '1
0"E 60 - *
3 40 - En: sdhade
s a e
"’ J
0 _—
birch maple
3 August 1996
80 -
.E 60 .
3 4o . Elno shade
:1 20 . Ishade
O .
mature birch immature birch
C June 1997
A 80 -
E 60 .. Elno shade
8’ 40 . I
E shade
_l 20 '1
"’ J
0 _
birch maple

 

 

 

Figure 2.2. Efiects of light on the specific leaf mass (SLM) of paper birch

and sugar maple foliage in June (A) and August (B) of 1996 and June of

1997 (C). Asterisks indicate significant effects of light treatment within species
or leaf age class (p _<_0.05). See text for description of age classes.

 

 

 

 

 

 

 

 

 

 

 

 

 

   

 

 

 

 

 

 

 

 

 

 

 

A June 1996
«A 100 -
E 80 -
g 60 - Elno shade
< 40 - Ishade
*5 20 -
3 0 .__.[:-_r_
birch maple
B Mature birch August 1996 Immature birch August 1996
El no shade D no shade
35 l I shade 35 . * I shade
a?" 30 ..
E E 25 -
m 20 -
a:
2 15 -
3; 1o -
3 s-
0 u
no defol defol no defol defol
C June 1997
N" 100 -
g 80 -
3 60 J Elno shade
E 40 ‘ lshade
‘3; 20 -
3 0 _ -
birch maple

 

 

 

Figure 2.3. Effects of light on the leaf area of paper birch and sugar maple
foliage in June (A) and August (B) of 1996 and June of 1997 (C). Asterisks
indicate significant effects of the light treatment within species or leaf age classes
(p g 0.05). See text for description of leaf age classes.

the specific leaf mass of both mature and immature foliage from birch trees was lower
when grown in shade relative to full sun (p g 0.05 and p 5 0.10. respectively) (Figure
2.2). In 1997 there was no evidence of prior season‘s shading on birch or maple SLM
(Table 2.5a, Figure 2.2).

Shade did not affect the leaf area of birch or maple in June of 1995 or 1996. but
did have an effect on birch leaf area measured in August of 1996 (Tables 2.4b and 2.5b).
Shade increased the area of immature birch foliage. but only on trees that had not been
defoliated (significant lighti‘defoliation interaction. Table 2.5b. Figure 2.3). Shade also
influenced the effect ofdefoliation on the leaf area of mature birch foliage (see below ).

Prior seasons’ shading did not affect the leaf area of birch or maple in 1997 (Table 2.5b).

Nutrient Eflecls on Specific Leqf Mass and LeafArea. The nutrient treatment did
not affect the SLM ofmaple in June of 1995. 1996 or 1997. or the SLM ofold or new
birch foliage in August of 1996 (Tables 2.4a and 2.5a). In June of 1996. the SLM of
birch foliage was greatest for non-shaded trees that had not been fertilized (significant
light*nutrient interaction, Table 2.5a).

Fertilization did not affect the leaf area ofbirch or maple in 1995. 1996. or 1997

(Tables 2.4b and 2.5b).

Defhliatirm Effects on Specific Lea/"Aims and Lea/Area. Both ANOVA (Table
2.5a) and correlation analyses indicated that defoliation had no effect on the SLM of

maple in 1996 or 1997. perhaps due to the low levels ofdefoliation achieved.

ln birch, defoliation did not affect the SLM of foliage sampled in June 1990.
ANOVA indicated that the SLM of mature, but not immature foliage of defoliated birch
collected in August was 8% higher (57.15 i 1.19 g in?) than foliage from non-defoliated
trees (53.15 i 1.19 g ml, Table 2.5a). This was most likely an artefact of methodology
and not indicative of true differences in SLM. Many ofthe defoliated birch were
defoliated to the extent that the only foliage available for analyses was predominantly
mid-rib with bits of leaf blade attached. Because the mid-vein in denser than the leaf
blade, samples from such trees would therefore contain a higher proportion of mid-rib

than samples of intact leaves.

There were no significant correlations between percent defoliation and the SLM
ofbirch foliage sampled in June and August. There were also no delayed effects of
defoliation on the SLM of birch in 1997. as indicated by either ANOVA or correlation
analyses.

Defoliation in 1996 did not affect the leaf area of maple or birch foliage collected
in June of that year or in June of 1997. However, the leaf area of mature but not
immature birch foliage collected in August of 1996 was lower on defoliated trees in the
shade (significant light*defoliation interaction, Table 2.5b, Figure 2.3). This result may
have been observed because in many cases mature leaves remaining on defoliated trees
had some leaf area removed by herbivores and therefore had lower leaf areas than leaves
on non-defoliated trees. The restriction ofthis defoliation effect to shaded trees may

reflect a preference for the defoliators to feed in shaded areas. Although there was no

statistically significant effect of shade on the defoliation treatment, there was a trend (p =

0.08) for defoliation to be greater in the shade.

Resource Acquisition.
Net Photosynthesis Under Ambient Conditions.

Statistical comparisons of relative photosynthetic rates ofbirch and maple could
not be made as species effects were confounded by dates of measurement. However. in
agreement with light response curves. the photosynthetic rate of birch (7.49 i 0.27 pmol
m"2 s") was higher than that of maple (4.89 i 0.30 umol m'2 s") as measured under
ambient conditions in June 1996. Date of measurement did not significantly affect
photosynthetic rates and did not interact with any treatment (data not shown).

In birch, shade reduced photosynthetic rates ca. 50% relative to trees in full sun on
both measurement dates, although this effect was only statistically significant on the
second date of measurement (for the first date. p = 0.057; Table 2.6, Figure 2.4). In
maple, the light treatment did not have a significant effect on photosynthetic rates.
although trends of decreased photosynthesis in the shade were apparent on both dates of
measurement (Table 2.6. Figure 2.4). There was no effect of fertilization or defoliation
on the photosynthetic rates of birch or maple (Table 2.6).

The stomatal conductance of maple (0.09 i 0.0] umol CO3 cm'l s") tended to be
lower than that of birch (0.22 i 0.01 pmol CO2 cm" 5"). corresponding with lower

photosynthetic rates. Although again no statistical comparisons can be made due to

40

«cowat.deowqua_

 

 

 

 

mm mm mm gscosm.
a. 3N N... E... . 5.5.0.2.usesztzm:
NI. N... a... . ceieaaiearfiz
3.. 3.0 N... . 5.5.0.0923...
Neo em. Nee . coze=eao
.3... ...o 92. . Eacazéa...
. ..o S... .3 . 22.52
8.. mN. .N... N Ewing...
Om. Non mm. . .gww.
N2. a .NN as N V.83.
>2... _ _ 33:86:00 _ngOHW .13. mm :o_~m__E_mm< >2... : coca—75mg. \h :EECmN/mo venom
3.2.).
em 2 mm a 5E.
.N... N3 N2 . 8......_eaoté..sz.:.w...
Nee aNeN sec . coce=eaotzaceea
N . .o 8.. we. . 8.5.0.0923:
cod N2. is . 8:288
.5 m... OWN . 5.23%:
NS 8.. . N. . . 22.32
a omN 2... 8.. N £91.28...
8... _. @on a N. e. . EN:
8.. :3 NN... N V.85
33—. o_ ..G_>_~u:_ucoU _SmEoum .13. ©_ coca—7534‘. 33. o_ :o_.m__E_mm< NB 22:23.2 no wot—sow
:omz.

.55. cote 2: we 23:38.9 5.3 38.328 v.63 Em: .8 8...?
.u. boo. E 3.8% some 8.. mean 95 :o 3.38:. was $855885 42222.8 255:8 Sec: 292: .&3 new :85 5me do 8:826:09 .8an3 was
28593993 8: :0 22688:. :2: new $8th cocfléon 9:... .2832: .Ew: @5283 do <>OZ< Eoc Q3 E88: .6 889.0 use 8...? i .o.u «Ba...

41

 

>

BIRCH

15-

 

10-1

 

El sun
lshade

 

 

 

54

 

 

 

   

 

 

Assumllatlon
(pMOI CO, [11‘2 s")

10-Jul-96 16-Jul-96

Measurement date

 

 

 

 

cu

MAPLE

15-

 

10‘ Dsun

5- - [Li— .Shade
o-——— . ,

1 1-Jul-96 22-Jul-96

 

 

 

 

 

Assimilation
(pmol C02 rn'2 5")

measurement date

 

 

 

Figure 2.4. Light effects on net photosynthesis of paper birch and sugar maple.
Asterisks indicate significant effects of light within species ([7 _<_ 0.05).

42

 

 

.Qovqn._
R R R 9. 3 9. \n .2...

8... 2.... SN .mN 2... Ne... . 25.2228:
8... ._ .3. 8N 2... o... .v... . - 26:52

3.. 2... 8... v... 3... no... N .55....8...
ON... . .NNN E... 8... 8n 8... . 5w:
3... .N... .2... N... 8... 8... N .8...
2...: 28 z... o: 0...: 26 z... 0.x. 8. 5.5.; .0 850m

89.2 :25

.58 Ste 05 m: 88:28... 5.3 5828.8 88> Em: .8 mos—8....

.32 :. 8:858: 8.8.6.8: o: 83 80...... .32 83. :. 588:8 8:58 28:. .85.. :8 5:... 88.. :o 3.8.. ZS. 858 885.5858 :8 .ZQ... 58:8
885.: Aug... 8:88.. 858 :0 82888:. :05 :5. acute 88:5: 5:: .Ew: F828... :0 <>OZ< 82.: $3 588: .8 88qu 55. 328% 6d «3.2.

a).

«Qov.u*d:ovqun_

 

 

mm mm mm .m 0. gm .w.p:m
mm... om... E... . 8.. . 8... 8... . €52.2885:32.8:
8.. E... 3... .2 h . .. z... . 8.3.6.092232
3... z... 8... a. R... 8.. _ 5.3.6.8...st
.N... 2... E. u .2 8.. 8... . 8.3.6.8
8... z... 2... a .2 .3 8.. . 52:52.2»:

.. 3.. u 8.. a... m. a} n .3 SN . 22.52
a and u «2 n .3 «.3 S. 2... N 3.528.:
a... :.N 8... .. RN. .. a... u 8.2 . .82 SE a 28.2.83. 2....
8.. 3... .. 8m 2... 2... _.. :8 N V.8...
2.2 28 z: 0.... 2.2 28 2.... 0.... .n 58.530 8.58
«En—z :85

_.:..8 .88 8.. mm :.w.......uo.n_ 5.3

88.8.8 803 Ew: .8 828$ .03. 0:... : 8.828 08:... 28:. .85... E... 8.... .88 .«o 3.8. 2.0. 8.8. :owo...:\:8.8 8m .28. .:8:8 88...:
.09... .:8:8 83.8 :0 80.888... .6... :5. 88.0 8.8.3.8. :5. 80.8: .Ew: “8.80... .o <>OZ< :5... 83 E88... .0 888: gm 82.3.... .w.~ ~33.

44

._o.ovmu:.mo.ovqufoeovqnn.

 

 

3 cm 2. cm 8 cm Kn 55
z 2.: . can 3o 35 NS 2: _ 8:23th35:52:33
v2 mac :3 3m _. 23 n _X _ 83.309255;
«2 So So 8o 3. who _ 85.9392»:
86 86 u 3m . v.3 . 23 3d _ 83.85
5.. ”no as . o3 .. .3 a 3m _ 232.2%:
N3 3. a3 : S.» n 8% SN _ 20:52
a? a ”3 35 Sam u ME 23 N 53285
u 5.: 32 3o . SS 3, 8.2 . 8.3 _ Em:
«no 26 8o :8 02 one N VEa
2:2 26 2.x. 0.x. 2E 26 z: 0.x. \3 55:5 3 850m

:35 9.5.3::—

gflfi 9:522

3E“: 8t» 2t 8 33283 65 333.8 89$ Em: 5.. 839i .002 umzw=< _ 380:8 A285 £2255.» £3255 (82 @583. 22:5 2:
Eat can 285 2322.9 .538 3038a 05 waste v.25 Eat wESEwto owmzfl :83 893 $0 3:8 ZDV 8:8 cowegscopfio Ea Azoxcv 53:8 comobE
AUX; 52:8 5&3 :0 228825 :2: 23 acute 532.8% can 35:5: .Ew: .3383 Mo <>OZ< Eo¢ Q3 8033 ‘8 803% 25 muggi d.” 03.;

._o.ovau._._.8.oqu foeovquf

 

 

_m _m _m 8 0m cm :0 5:0
go :.o 8.0 n 03 3: 3o _ 83.8505:52:50:
m _ ._ 3: m2 :3 3o .3 _ 502300905252
:3 8.0 :3 two .2 3o _ 8:23:93»:

0 m; as 2.0 0. 0:. o: m; _ 502305
8.0 So 02 0 3m .. Re 8: _ 05:52.20:

... 3.0 r. 2.2 u 03 0 Cum mom 23 _ .5532
2: So 1, 3.0 . 85 a: . 2; m 205.0320
Rd Fm 03 2.0 00.. N3 _ Em:
who 23 z 02 2.0 mg 8.0 N 0.8.0

2:2 2b zoxc 0:. 2E 26 2.x. 0.x. :0 5:35, 00 8.8m
0302 :85

2E8 5:0 05 mm Ew__.v_00_n 5.3 35:28

0003 Em: .8 00291 .32 0:3. 2 3.00:8 0.00:8 032: Swan .05 :23 00am: :0 3:8 25: 8:8 33.252808 30 2.5 20:50 Swot... AUX:
E0280 8980 :0 2288005 :05 .03 300:0 83:80: :3 .2053: .Ew: .wciooE mo <>OZ< E0: 83 8030: :0 0003.0: 30 0.0303: .:_.N 030...

46

confounding date effects. Stomatal conductivity in both species was not affected by light,

nutrient, or defoliation treatments (Table 2.6).

Foliar Carbon and Nitrogen.

Intrinsic Differences in Carbon and Nitrogen Concentration and ( '/N Ratios.
Birch foliage had greater carbon (C) and nitrogen (N) concentrations, and lower C/N
ratios than did maple foliage in 1995, 1996, and 1997 (Tables 2.3, 2.7. 2.8. 2.10), with the
exception that in 1995 there was no difference in carbon concentration.

In August of 1996, immature and mature birch leaves did not differ from each
other in carbon or nitrogen composition, or in the ratio of C to N (Tables 2.3 and 2.9).
Both nitrogen and carbon content declined and ON ratios increased in birch over the
growing season, as indicated by comparison ofbirch foliage sampled in June and in

August.

Light Effects on F oliar Carbon and Nitrogen Concentration and (.‘/N Ratios. In
the first year of the light treatment (1995), light had no effect on the foliar carbon content
or C/N ratios of either species (Table 2.7). In maple, but not birch, shade was associated
with an 8% increase in foliar nitrogen content (Table 2.7).

In June of 1996. the second season of shading, there was again no main effect of
light on foliar carbon levels or C/N ratios. nor was there an effect of light on the nitrogen
content of birch or maple foliage sampled in June (Table 2.8). Although effects were not
statistically significant, shade tended to increase carbon content (p = 0.0088) and nitrogen

content (p = 0.0593) and to decrease C/N ratios (p = 0.0702) ofbirch foliage sampled in

47

June. Light effects on birch foliage did become statistically significant as the season
progressed and leaves aged. In August. shade decreased foliar carbon content by 2%.
increased C/N ratios by 20%. and tended (p = 0.0559) to increase the nitrogen content of
mature but not immature birch foliage. Seasonal effects of light treatments on maple
were not monitored.

Light availability had no long-term effects on leafchemistry. In 1997. all trees
were exposed to ambient light conditions as may occur in previously shaded trees
following gap formation. There was no evidence ofdifferences in foliar carbon. nitrogen.
or C/N ratios between birch of maple trees that had been shaded and those which had
always existed in full sun (Table 2.10).

Nutrient Effects on F oliar Carbon and Nitrogen Content and ('/N Ratios. In
1995. the first year of the nutrient treatment. fertilization of maple increased foliar
nitrogen 9% (Table 2.7. Figure 2.5). but had no effect on foliar carbon or C/N ratio. In
birch. however, fertilization had no effect on any of these foliar traits.

In June of 1996, the effect of fertilization on foliar nitrogen content of maple was
no longer statistically significant. There also was no effect of fertilization on foliar
carbon content or C/N ratio of birch and maple, or on foliar nitrogen content of birch. As
the season progressed, however. mature foliage ofbirch grown in full sun responded to
fertilization with a 16% increase in foliar nitrogen. and a corresponding 14% reduction in
UN ratio. In contrast. mature foliage ofshaded birch did not respond to fertilization
(light*nutrient interaction. Table 2.9). Fertilization had no effect on the carbon or

nitrogen content or C/N ratio of immature birch foliage (Table 2.9).

48

 

>

1995

El no fert
Ifert

nitrogen
(% dry weight)

 

 

birch maple

 

 

 

 

B 1996

El no fert
Ifert

nitrogen
(% dry weight)

 

 

birch maple

 

 

 

 

C 1997

 

Elno fert‘
Ifert ‘

nitrogen
("/0 dry weight)

 

 

 

 

 

Figure 2.5. Effects of nutrient treatment on the foliar nitrogen content of
paper birch and sugar maple in June of 1995 (A), 1996 (B), and 1997 (C).
Asterisks indicate significant effects of fertilization within species (1) E 0.05).

49

In 1997, fertilization increased the foliar nitrogen content of maple by 149%
(Figure 2.5). and decreased C/N ratios by 10%, but had no effect on carbon content
(Table 2.10). Although fertilization had no effect on the nitrogen content ofbirch in I995
and 1996, in 1997 fertilization increased the nitrogen content of birch by l5%. Although
trees were not shaded in 1997, fertilizer only affected birch trees that had not been shaded
the previous two seasons (1995 and 1996) (light*nutrient interaction. Table 2. I 0).
Nitrogen content of previously shaded birch was not affected by fertilization (Table 2.10).

Fertilization did not affect C/N ratios or the carbon content of birch in 1997.

Defbliation Effects on F oliar Carbon and Nitrogen Concentration and ( 7N
Ratios. Defoliation did not affect foliar carbon content, but did decrease foliar nitrogen
and C/N ratios. Fertilization often mitigated the effects ofdefoliation on leaf chemistry.

Defoliation during spring of 1996 had no effect on foliar carbon or nitrogen. or on
UN ratios of maple foliage sampled in June (Table 2.8). However. in maple there was a
weak positive correlation between percent defoliation and levels of foliar nitrogen (r =
0.33, p = 0.026. n = 45; Figure 2.6). and a corresponding weak negative correlation
between percent defoliation and ON ratio (r = -O.29. p = 0.047, n = 48).

In 1996, birch foliage from defoliated trees generally had a lower nitrogen content
and a greater C/N ratio than foliage from non-defoliated trees. This effect was present
both in June and August. was inlluenced by light treatments. and was negated by
fertilization. Defoliation had no effect on the foliar nitrogen content or (‘x’N ratio of
fertilized birch in June or August of 1996. However. in non-fertilized birch trees grown

in full sun. defoliation decreased foliar nitrogen by 17% and increased the UN ratio by

50

 

 

 

 

 

 

 

Birch 1997

E
m
'8
3
E
'o
E: 1 _‘ f: -0.29
2 p = 0.047

0 i l l I j

0 20 4O 60 80 100
defoliation (%)
Maple 1996

6
g 5
.5 4 , .
3 3
5 2
i 1 r = 0.33
2 O l l l I 1 p = 00261

O 20 4O 60 80 100
defoliation (%)

 

 

 

Figure. 2.6. Correlations between percent defoliation by forest tent caterpillar in 1996
and foliar nitrogen content in paper birch in 1997 (A) and sugar maple in I996 (B).

51’

 

A Birch June 1996 U "0 defol
* Idefol

 

 

 

 

6i
Z”: 4.
'0:
5032.3 l
2 3

0- , -

nofeit fen
Noshade

 

 

 

B Birch June 1996 000 defol
5 - Idefol

i] .

"° '9“ Shade

A
4

l

 

 

 

 

weight)

O—‘NOO
1

N(°/o dry

 

fert

 

 

 

Figure 2.7. Effects of fertilization and defoliation on the foliar nitrogen content of
paper birch in full sun (A) and in shade (B) in June of 1996. Asterisks indicate
significant effects of defoliation within fertilization treatment (p 5 0.05).

52

20% in the foliage sampled in June (light*nutrient*defoliation interaction, Table 2.8.
Figure 2.7). There was no effect of defoliation on the nitrogen content of shaded birch.

Similar patterns were seen later in the season. As in June. the nitrogen content in
mature birch foliage harvested in August 1996 was reduced 15% on defoliated hosts. but
only on those trees that had not been fertilized (significant nutrient*defoliation
interaction. Table 2.9). Defoliation increased the UN ratio of mature foliage in both
fertilized and non-fertilized birch by 8%. Defoliation also decreased the nitrogen content
of immature leaves from non-fertilized birch trees by 13%. but only for those trees grown
in full sun (light*nutrient*defoliation interaction, Table 2.9).

In 1997, both ANOVA (Table 2.10) and correlation analyses indicated that
defoliation had no effect on carbon or nitrogen content, or C/N ratios of maple foliage. 1n
birch, previous season’s defoliation had no effect on C or N content. but did increase
foliar C/N ratio by 7% (Table 2.10). Correlation analyses did not reveal any relationship
between percent defoliation and carbon content. nitrogen content. or C/N ratios in birch

in 1997.

Resource Allocation.
Height and Radial Growth.

Intrinsic Differences in Birch and Maple Growth. Birch growth prior to
implementation of treatments (1992-1993) was greater than that of maple (data not
shown). At the beginning of 1994. although birch were three years younger. they were an

average of 32 cm (12%) taller than maple. During the 1994 growing season. birch

(.11
DJ

Table 2.11. F-values and degrees of freedom (d/) from ANCOVA of blocking. light. species. nutrient.
defoliation effects, and their interactions on the trunk height and diameter growth during the 1994. 1995.
and 1996 seasons with height and diameter in May of 1994 as covariates, respectively; and over-all trunk
height at the end of 1994, 1995, and 1996, with May 1994 height as a covariate. Models for each year
correspond to treatments in effect during that year. F -va1ues for light were calculated with block*light as

the error term.I

 

 

 

 

 

 

 

1994
Diameter Height Over-all
Source of Vflation df growth growth height
Block 2 0.13 0.41 0.41
Species 1 9.46 ** 17.68 **** 17.82 "**
Nutrient 1 0.03 0.01 0.01
Species*Nutrient l 0.07 0.29 0.28
May 1994 diameter or height (covariate) 1 0.33 0.89 66.56 ****
Error df 86 86 86
1995
Diameter Height Over-all
Source of Variation df growth growth height
Block 2 3.13 * 8.01 *** 4.82 "
Light 1 32.35 * 0.04 0.74
Block*Light 2 0.29 1.59 1.32
Species 1 1.44 1.92 14.36 ***
Species*Light I 0.16 0.60 0.57
Nutrient I 1.15 4.66 " 3.65 #
Species‘Nutrient 1 0.10 1.10 0.34
Light*Nutrient 1 0.69 0. 12 0.01
Species*Light*Nutrient 1 0.78 0.27 0.82
May 1994 diameter or height (covariate) 1 0.07 0.23 28.78 ****
Error df 80 77 77
1996
Diameter Height Over-all
Source of Vgiation df growth growth height
Block 2 6.93 ** 0.85 5.39 **
Light | 11.74 # 79.70 * 1.98
Block‘Light 2 2.30 0.13 1.54
Species 1 0.01 1.15 14.91 ***
Species*Light 1 1.04 0.24 0.00
Nutrient 1 0.00 0.30 2.56
Species‘Nutrient 1 0.61 0.25 0.1 I
Light‘Nutrient 1 1.21 2.44 1.12
Species*Light*Nutrient 1 4.69 * 0.02 1.55
Defoliation 1 1.69 6.45 * 3.10 #1
Species*Defoliation 1 0.08 3.24 I .77
Light‘Defoliation 1 0.03 0.48 5.84
Species’Light"Defoliation 1 1.92 0.01 0.37
Nutrient*Defoliation 1 0.74 4.82 * 1.72
Species*Nutrient*Defoliation 1 0.42 0.01 0 3}
Light*Nutrient*Defoliation 1 5.22 * 2.63 0.70
Species“Light*Nutrient* Defoliation l 1.70 1.09 0.93
May 1994 diameter or height (covariate) I 2.25 0.38 38.70 ****
Error d/ 72 69 72

'# =p<0.10, * =p<0.05. ** =p<0.01. *** =/)‘-’10.001.**“‘ =p<0.0001.

54

 

 

 

 

A ‘2 ‘ I Birch
‘0 - lElMaple

 

 

 

 

 

Diameter growth (mm)
0 N b 0') on

I.
l _{

 

 

 

 

 

 

 

 

 

   

 

 

 

 

 

 

 

 

   

 

 

 

 

 

 

1994 1995 1995
Year
B 10°“ IBirch
*
so. DMaple
g .—
2 A 60‘ ‘ ‘
35
g v
3 20-
I
o.
1994 1995 1996
Year
0 400 «——~ *
350: IBirch i.
A 300- EJMaple *
g 250‘
v 200- *
E 150
a:
“a 100-
.r:
a, 50 ‘fl
2 0" I I —'1
1‘ 1993 1994 1995 1996
End of Year

 

 

 

Figure 2.8. Diameter growth (A). height growth (B). and absolute height (C)
of paper birch and sugar maple in 1994. 1995. and 1996. Asterisks indicate
significant differences in species growth or height within years (p _<_ 0.05).

growth in both height and diameter was greater than that of maple. as predicted based
upon comparative life history strategies (Table 2.1 1. Figure 2.8).

From 1995-1996. height growth of both species was equal. but because of faster
growth prior to 1994. birch remained significantly taller than maple. Both diameter and
height growth of birch and maple were equivalent in 1995 and 1996 (Table 2.1 1. Figure
2.8). On average. birch was 46 cm (14%) taller than maple at the end of the 1995

growing season. and 38 cm (10%) taller than maple following 1996.

Light Effects on Birch and Maple Growth. Shade reduced both height and
diameter growth of both species, but effects differed by year and were affected by
interactions with other treatments. Effects of shade were not strong enough to affect the
overall height ofeither birch or maple in 1995 or 1996. From 1992-1994. all trees grew
in full sun. Shade treatments were first implemented in 1995. mimicking canopy closure
with succession. That year. shade decreased diameter growth of birch by 44% and
diameter growth of maple by 30% (Table 2.11. Figure 2.9). Height growth of both birch
and maple was unaffected by shade in 1995.

In contrast, in the second year of the light treatment (1996), shade led to strong
reductions in height growth of both species (41% reductions in birch and 45% in maple;
Table 2.1 1. Figure 2.9). Shade reduced diameter growth in fertilized birch and in non-
fertilized maple (significant species‘i‘light*nutrient interaction. Table 2.1 1. Figure 2.9).
Shade interacted with nutrient and defoliation treatments (significant
light”nutrient*defoliation interaction. Table 2.1 l). Birch and maple in full sun generally

showed the greatest diameter growth in the absence of fertilization and defoliation. In

56

.Amod .vi 5 $6.3m Ears Em: .«o muuofiu Emoc_:w_m 2865 3389a. 63— E oEmE “swam use :25 Emma go An: £322.11. 532 ES
A 8 5.503 85:86 so can mam: E Am: 538w Ems: was As 5sz 53:86 so 3328.6 Ew: .6 flowcm .a.~ 35w...—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mamE :85 wins. :85
e - o 9 t3 0: t9 to». o: to“.
, . ow m .. . w . o

:30 . cm Wt 36ch .. v m.
.. . 8 M can . 0 1m,

1.. . m m

. 2: A
a u or J...
* (

. NF
82 5.53 29¢: a $2 5305 33520 0 w.
mam
_ E . :85 mamE :95
O n

. om m o m

. N o

82m- . ow M 25an . v M
can. . om m :80 u. 1 o 1w:
. 8 M .. t . m m

1h... 1 or Mr

. oor . NV (

3.: 522m 222.. m 32 532m 335% <

 

 

 

 

 

 

contrast. fertilization and defoliation had no effect on diameter growth of shaded trees

(Figure 2.9).

Nutrient Effects on Birch and Maple Growth. Contrary to expectations.
fertilization did not generally increase tree growth. and in some cases actually was
associated with reductions in growth. During the first year of the treatment (1994).
fertilizer had no effect on height or diameter growth. or on the over-all height of birch or
maple (Table 2.1 1).

In 1995. fertilization did not affect diameter growth ofeither species. but
unexpectedly reduced height growth by an average of 31%. (39% reduction in birch
height growth and 19% reductions in maple height growth) (Table 2.1 1. Figure 2.10).
The effect of fertilization on height growth was not substantial enough to result in
treatment effects on over-all height.

In 1996. the third season of treatment. fertilization had no effect on height or
diameter growth. or on over-all height of birch or maple (Table 2.1 1). However.
fertilization did influence the effect of the light (see above) and defoliation (see below) on

plant growth.

DefO/iation Effects on Birch and Maple Growth. Forest tent caterpillar defoliation
in 1996 reduced the height growth of birch and maple by 57% and 10%. respectively
(Table 2.1 1. Figure 2.1 l ). The effect ofdefoliation on height growth was strongest in
fertilized trees, where defoliation reduced height growth 55% (nutrient*defoliation

interaction. Table 2.1 1. Figure 2.1 1). Defoliation influenced the effect of shade on

58

 

 

 

 

:30 w 3 $68...

5:23 :o_§.~:_t£ mo flout“. 258.:ch 88:8: 85:83. .052 E 292: 595 can :25 $me (:0 Rd 538w Ewan 25 CV
538w 58:86 so can 33 E Amv 538w Emma: use 2: £38m 5on85 :0 325252». 805:: go 88km .c_.~ 95w:—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mmfl :595 29¢:

 

mine :85
P I O m
. om 0
t8- .. ow m.
t8 05.... . cm W
. 8 M.
. 09 1
82 .235 :55: o
mEmE :85
- ll o m
. om o
. w
t9- ov u.
:9 8n. . 8 1m,
n K
H ow U.
. cor

 

 

 

 

 

 

 

game :85
_ 9
m
_ thI_ N
w
:8 0:0 H w
M.
- 8 L.
82 £265 35855 o
mamE :85
1 L1 9
- W
_ twai_ ,1 N
m w
t 80:0 w
m.
82 £265 58555 <

 

 

59

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A Height growth 1996
A 100 -
's. 80 - .. _
g 50 a 121 no defol
g 40 4 Idefol ‘
g 20 J
(.9 0 .1
birch maple
B Height growth 1996
A 100 1 b
';. 80 - b I
g 50 J a. a a 1:1 no defol
g 40 - Idefol
g 20 -
o 0 __
no fert

 

 

 

 

Figure 2.11. Effects of defoliation on height growth of paper birch and sugar
maple in 1996 (A) and the influence of fertilization on the magnitude of the
defoliation effect (B). In (A) asterisks indicate significant effects of defoliation
within species (p _<_ 0.05). In (B) bars with the same letter above them are not

significantly different at a = 0.05.

60

diameter growth of birch and maple in 1996. but did not influence over-all height of

either species.

Foliar Condensed Tannin and Total Phenolic Concentrations.

Intrinsic Differences in Condensed Tannin and Total Phenolic Content. In 1996.
birch and maple foliage sampled in June did not differ from each other in condensed
tannin content. but in June of 1997 birch foliage had higher condensed tannin
concentrations than did maple foliage (Table 2.3). Mature and immature birch foliage
sampled in August of 1996 did not differ from each other in condensed tannin content.
but had lower condensed tannin levels than birch foliage sampled in June ofthat year.
reflecting a seasonal decline in condensed tannins in birch. Birch condensed tannin levels
were lower in 1996 than in 1997 while maple condensed tannin levels were higher in
1996.

Total phenolic concentrations also varied between birch and maple. Birch
appeared to have lower total phenolic concentrations than maple in 1996 (Table 2.3 ). As
for condensed tannin content. mature and immature birch foliage sampled in August did
not differ from each other in total phenolic content. There was no evidence of seasonal
variation in birch total phenolic levels. as foliage sampled in June and foliage sampled in
August had equivalent total phenolic contents. 'l‘otal phenolic content was not measured

in 1997. so no comparisons can be made across years.

Light Effects on ('onclensed Tannin and Total Phenolic ('ontent. ln .lune of 1996.

shade was associated with a 37% reduction in birch condensed tannin levels. but did not

61

aqovqutdeovqun_

 

 

mm 2. 9.. 3 \n .25
moo 3.: $3 a can _ 83.8.5tssszEm:
:5 :3 88 8.0 _ 85.85.255.52
w _ .o as :8 42 _ 85385.25:
Po 3o :2 om: _ 8:288
3 .m m _ ._ mm. 3.0 _ 25:52.25:
8.0 8.: o: E: _ 2852

a... :m $5 8.: 8.0 N 5.1.5285
o3 .. 8.2 .. 2.8 8.” _ am:
5.. 3o 8.: 8.: N 8.85
2:3:de Uomcuccou mozocoza _39 mamas—19 town—0250 muzocosa _99 \h cocmtw>mo uohsom

2:22

:25

_.E._8 :85 0:“ mm Ew:_._xoo_:

8:: 35.3.8 803 Ew: :8 335?“: .000. 2:: _ _ 380:8 08:8 2ng Swzm can :85 :38 .8 C:w53 .96 5:8 8.: 58:8 :55: 535850
can 835:: .88 :0 888885 :55 85 8085 55:83 85 .5883: gm: .3552: :o <>OZ< ES: $3 8038 .6 88mm: 85 839i .N_.n «BE.

()3

 

>

Condensed Tannins 1996

 

 

 

 

 

e 3-

'D

33352" * Dnoshadel
c a

.E 31. I I Ishade

C

m

l-

O
l

birch maple

 

 

 

 

to

Total phenolics 1996

*
20- I

birch maple

00
O
I

 

[:1 no shade
Ishade

 

 

Phenolics (% dry
weight)

 

 

_L
O O
I 1

 

 

 

 

 

 

 

C Condensed Tannins 1997
E‘ 3 -
'D
‘3 E 2 ' Elno shade
2 42” Ishade
"E 3 1 ‘ -
5
i- 0 '

 

birch maple

 

 

 

Figure 2.12. Effects of shade on the concentrations of carbon-based

secondary metabolites of paper birch and sugar maple: condensed

tannin levels in June of 1996 (A). total phenolic levels in June of 1996 (B). and
condensed tannin levels in June of 1997 (C). Asterisks indicate significant effects
of light treatment within species (,1 g 0.05).

()3

._O.OVQ.H**.M.O.OVQ.H a.

 

 

:m :m :m :m x: 5:0
2.: :m: :8 ::.: _ 8:26:05£582.20:
N:.: 3.: e _ .: we: : 8:80:09852
No: :8 8.: 8.: : 88:88.20:
8.: N: N:.: 8.: _ 8:26:00
8.: Ne: N:.: N:.: _ 20:52.30:
8.: .. :3. R: 2.. _ 25:52
:3 :8 N:.: N: N 20% .285

a _:.N_ «N: t 3.2: 8: _ 205
N: K: 8.: :2 N :85

mECCS Vomcoficoo muzocosm :30: 2:55: tom—50:00 max—0:25:30: \B :O_:m_:m> mo 00::OW

8:: 0.5225:—

:0:E 0:322

.88: :88 0:: 8 £87052: 5:: 88.8.8 0:03 Ew: :8 83:3: :8: :m:w=< _ 880:8 FEE: 83:88:: 880: E85: 0:: E 88:8
8885:: .52 Eat cc: A225 05:22:: 880: 8030:: 0:: 888 8:: E8: wEBEwto 08:8 :85 :08: :0 AEwBB :6 8:8 :5: E058 5:8: 88088
8:: 0:808 5:0: :0 88:00:02: :_0:: 8:: 808:0 882.80: 2:: .8055: .::w: 8:28.: :o <>OZ< 88.: 88 E8008 :0 000:8: 8:: 828...: .m_.N 0.50.:

()4

Table 2.14. F-values and degrees of freedom (4]) from ANOVA of blocking, light. nutrient, and
defoliation effects and their interactions on condensed tannin content ("/0 foliar dry weight) of paper birch
and sugar maple foliage collected 16 June I997. F-values for light were calculated with block*light as the

error term.'

1997 condensed tannins (% dry weight)

 

 

Source of Vafizgion (If Birch Maple
Block 2 l.04 l.50
Light I L6] 0.7]
Block‘Light 2 L36 0.32
Nutrient l I .53 0.06
Light“ Nutrient l 0.13 3.27 1,3
Defoliation I 0.0l I. I4
Light*Defoliation I 0.07 0. I6
Nutrient*Defoliation I I .63 0.07
Light*Nutrient* Defoliation l 0.! l O.()|
Error 4f 36 32

|#=/)<0.10

65

influence total phenolic concentrations (Table 2.12, Figure 2.12). Conversely, in maple,
shade reduced total phenolic concentrations by 17%, but did not influence condensed
tannin concentrations (Table 2.12, Figure 2.12).

In late season birch, shade reduced the condensed tannin concentration of foliage
from the mature age class by 35% and tended to reduce the condensed tannin
concentration ofimmature birch foliage by 41% (p = 0.0700) (Table 2.13). As in June,
shade did not affect total phenolic levels of August birch foliage.

In 1997, there was no evidence of an effect of prior season‘s shading on
condensed tannin concentrations of birch or maple foliage (Table 2.14). Total phenolics

were not measured in 1997.

Nutrient Effects on Condensed Tannin and Total Phenolic Content. Fertilization
did not affect condensed tannin or total phenolic concentrations of birch or maple foliage
in June of 1996 (Table 2.12) or condensed tannin concentrations of either species in .lune
of 1997 (Table 2.14). However, in August of 1996. total phenolic concentrations of
immature birch foliage were decreased 17% in fertilized treatments (fertilized: l 1.37 i
0.78%, non-fertilized: 13.69 i 0.78%, Table 2.13). Condensed tannin levels of immature
birch were not effected by fertilization. Neither condensed tannin nor total phenolic

content of mature birch foliage was affected by fertilization (Table 2.13).

DefOliution Effects on Condensed Tannin and Total Phenolic ('m'zlcm. ANOVA

indicated no significant rapid or delayed effects of defoliation on condensed tannin or

66

total phenolic content of foliage from either species in either 19% or 1997 (Tables 2.12.
2.13, and 2.14). Correlation analyses. however, revealed weak negative relationships
between defoliation and levels of condensed tannins in June sampled birch (r = -0.2‘). p =

0.047, n = 48) and immature (but not mature) leaves of birch sampled in August (r =

-030. p = 0.038. n = 48).

DISCUSSION
Birch and maple differed in many physiological characteristics. Patterns of
resource acquisition and allocation in both species were influenced by light. nutrient, and
defoliation treatments, acting alone or in combination. While some generalizations can
be made regarding the influences of these treatments in the present study. often each tree

species responded differently to environmental stressors.

Intrinsic Differences in Paper Birch and Sugar Maple Physiology.

As predicted based on life history, shade-intolerant paper birch had higher
photosynthetic capacities (i.e. higher light-saturated photosynthetic rates. indicated by
light response curves). and correspondingly higher rates of carbon acquisition in both sun
and shade than did maple. Although prior work has suggested that shade-tolerant species
are more successful in low-light areas because of relatively higher photosynthetic rates at
low irradiances but lower maximum assimilation rates in full sun than shade-intolerant
species (Bazzaz 1979). this study affirms more recent work indicating that this is not the
determinant of shade tolerance in all cases (Kobe ct al. 19%). Shade tolerance in sugar

maple cannot be explained by a lower respiration rate than birch (which would result in

67

more efficient utilization of acquired carbon), as birch and maple had equivalent
respiration rates. However, this study only measured respiration over a short time frame
(days) in mid July. There is likely to be diurnal and seasonal variation in respiration. as
well as possible differences in whole plant (including root, trunk, and branch) respiration.
First year paper birch have been shown to have higher respiration rates than first year
sugar maple (Walters et al. 1993) and first year red maple (Ac-er rubrwn. Kubiske and
Pregitzer 1996). Phenological variation in allocation and metabolic processes may be
very important in differential shade tolerance and successional mechanisms.

Birch had a greater specific leaf mass than maple in 1995. 1996 and 1997. This is
likely related to the presence of higher nitrogen and photosynthetic apparati in the leaves.
Specific leaf masses appeared to be higher within species in 1997 than in 1996. This
most likely reflects release of birch and maple from shade, which depressed specific leaf
mass during the second year of shading (1996).

Corresponding to its higher photosynthetic rates. birch contained higher levels of
foliar nitrogen than maple in all years examined. This may indicate higher nitrogen
acquisition rates in addition to higher carbon acquisition rates, or may reflect intrinsic
differences in N-allocation patterns between these species.

Also in agreement with life history predictions. paper birch exhibited greater
growth (both in height and diameter) than sugar maple in the first year ofthe study.
However. in subsequent years there were no differences between the species in height or
diameter growth. This convergence in growth appeared to be due to decreased growth of
birch over time. while growth of maple held constant (Figure 2.8). Although birch

remained taller than maple throughout the experiment. if these trends were to continue

()8

over time. maple would soon surpass birch in height. This would be expected to occur
naturally, as the persistent maple replaces the pioneering birch. Even in cases where the
species were able to coexist, maple would be expected to surpass birch in height. as
maple trees reach 27-37 m (90-120 ft) in height while birch will reach only 21 m (70 ft)
on average (Goodman et al. 1990. Safford et al. 1990). Similarly, birch can be expected
to reach 25-30cm (10-12”dbh) and live 140 —- 200 years (Safford et al. 1990). while maple

can reach 79-91 cm (30-36” )dbh and live 300 - 400 years (Goodman et al. 1990).

Reduction in height growth in birch over time is expected as individuals age.
However, the slowed growth observed early in the trees’ lives in this study should not
necessarily be expected considering the highly competitive environment in which this
species commonly occurs. and considering that birch do not reach reproductive maturity
until 15 years of age (Safford et al. 1990). Birch responded more strongly to treatments
than did maple, and reductions in birch height growth likely reflect these treatment
effects. In 1995, birch height growth was unexpectedly reduced 31% by fertilization. In
1996, birch height growth was reduced 41% by shade and 5 7% by defoliation. These
results correspond to the progressively reduced average height growth in birch over the

course of the study.

Birch had lower concentrations of total foliar phenolics than maple. in agreement
with other studies showing that pioneer species often allocate fewer resources to defense
than late successional species (Bryant et al. 1985. (‘oley et al.1985. Dudt and Share
1994). In contrast to total phenolic trends. maple had higher levels ofcondensed tannins

than birch. Condensed tannins are components ofthe total phenolics measure. and total

69

phenolics are only one of the array of defenses available to birch and maple. Birch also
contains papyricific acid (in stems), a deterrent to browsing mammals and other
herbivores (Bryant et al. 1992). Maple contains alkaloids, which are associated with
deterrence of gypsy moth (Barbosa et al. 1990) and other forest Lepidopterans. Because
plants contain a diversity of secondary metabolites which have defensive attributes. and
because herbivores are not equally deterred by all defenses. it is not possible to make
accurate statements of the relative strength of plant defense without reference to a specific
herbivore or class of herbivores. Defensive quality may not be directly correlated with

the quantity of resources allocated to particular secondary metabolites (see Chapter 3 ).

Seasonal and Age Related Changes in Birch F oliar Chemistry.

There were no strong differences in nutritive, chemical. and morphological traits
of mature (short shoot) and immature (axial growth) birch foliage collected in August of

1996 although the age classes did differ in their sensitivity to treatments.

Carbon and nitrogen levels increased over the season as did SLM, while
condensed tannin content decreased. Baldwin et al. (1987) found that hydrolyzable
tannin content in yellow birch (BU/Illa allegiaeniensts) decreased as the season progressed.
while condensed tannin content increased toward midseason and then declined again as

the season progressed.

70

It is possible that there were not stronger differences between old and new foliage
because although the foliage differed in age. all foliage was fully expanded when
sampled. Many differences in young and old foliage may be attributed to their
differential source/ sink status, which in turn can affect the susceptibility of individual
leaves to environmental stresses (Coleman 1986). The source/ sink transition normally
occurs when leaves are two-thirds expanded. therefore the leaves sampled were probably
acting as photosynthetic sources. Even so, differences do exist (apparently in traits not
measured here) between birch mature and immature foliage. Swallowtail butterfly larvae
(Papillio sp.) showed a preference for older foliage both in the lab (Herms unpublished
data) and the field (personal observation). In other systems, younger foliage has been

preferred by herbivores (Stamp and Bowers 1990, F aeth 1992).

Effects of Light on Birch and Maple Physiology.

The light treatment under which leaves developed did not affect light-saturated
photosynthetic rates (Amax) in either birch or maple, indicating that there was no
effective photosynthetic acclimation to shade, even though leaf morphology did respond
to the light treatment (discussed below). In contrast, Riddoch et al. (1991) found higher
initial slopes of light response curves and higher light-saturated photosynthetic rates in
one pioneer and one climax tropical tree species under high light. Likewise. 'l‘hompson.
Huang. and Kriedemann (1992) found a strong positive correlation between light-

saturated photosynthesis and environmental light availability in both shade-tolerant and

71

shade-intolerant rainforest trees. Similar results have been found by other researchers
(Turnbull 1991, discussed in Givnish 1988). Tropical species may be more responsive to
light conditions when developing. as leaves in these species are generally longer-lived
than those of deciduous temperate trees. Alternatively, low-light levels in the present
study (20% ambient), while extreme enough to influence other aspects of plant
physiology (discussed below), may not have been low enough to influence the efficiency
of photosynthetic apparati. Finally. shade supports in the present study were
implemented shortly after budbreak. such that initial foliage expansion of all plants
occurred in full sun as is the case in nature as expanding leaves provide little shade. 11 is
possible that light-availability affects development of photosynthetic apparati during
differentiation of foliar tissues, and that acclimation potential is decreased after foliar

expansion.

Although there was no effect of the light treatment on photosynthetic capacity.
specific leaf mass and leaf area did respond to shading. Trees acclimated to temporal
shifts in light availability in 1996 by decreasing leaf mass per unit area ratios (specific
leaf mass). In maple, this response was evident early in the season (with June-sampled
foliage) and was not associated with an increase in leaf area. In birch the response was
not evident until August. and in immature foliage. corresponded to an increase in leaf
area in non-defoliated but not defoliated trees. Decreased specific leaf mass may result
from increased leaf size. from compositional differences in sun versus shade-grown
foliage. or from a combination of these and other factors. By spreading out

photosynthetic apparati over a larger area by producing larger but thinner leaves. plants

72

may more efficiently harvest the limited incident radiation in shaded environments.
Shade led to increased foliar nitrogen per unit mass in maple (discussed below).
suggesting that decreases in other foliar components may have occurred. In 1997 the
shade cloth was not erected, thus previously shaded trees experienced ambient light
conditions as they might in nature after gap formation. Foliage on trees that were
previously shaded had specific leaf masses equivalent to foliage on trees which had
always grown in full-sun. This indicates that birch and maple trees may quickly adjust to

gap formation.

Shade reduced photosynthetic rates of birch, but did not have a significant effect
on maple photosynthesis, reflecting the greater tolerance of maple to shade and greater
plasticity of pioneer species to environmental perturbation (Riddoch et al. 1991. Ashton

and Berlyn 1994).

As predicted, shade reduced plant growth in both species. Height growth was
equal in sun and shade during the first year of shading (1995), while diameter growth
suffered. Equivalent height growth in sun and shade at a cost to diameter growth may be
expected as plants attempt to compete for light. In 1996 height growth was not
maintained but was decreased in shade, while there was no influence of light on diameter
growth. Additionally, primordia for height growth are formed in the previous season.
which may explain the lagging effect of shade observed in this study. Alternatively. trees
may need to maintain optimal height/diameter ratios to ensure structural integrity. If

carbon reserves from the previous year are used to drive height growth in the following

year. then shade in 1995 may have resulted in an insufficient carbon supply to maintain

height growth in 1996.

Shade also increased foliar nitrogen content of maple during the first year of the
treatment (1995) but did not influence the nitrogen content of birch foliage. Similarly,
Kubiske and Pregitzer (1996) found that 84% shade increased foliar nitrogen in A (‘t’l‘
rubrum (80% shade the present study) but had no effect on foliar nitrogen content of

paper birch.

Although shade did not have a strong effect on foliar carbon or (‘/N ratios. it was
associated with reductions in carbon-based defenses in both birch and maple in 1996. as
predicted by the carbon/ nutrient balance hypothesis. However, birch and maple did not
respond the same way. Total foliar phenolics of birch were not affected by shade. but the
composition of phenolics changed, as condensed tannin content was reduced. Shade also
induced compositional changes in maple phenolics, but in this case shade decreased total
phenols and had no effect on condensed tannin levels. Ellsworth and Reich (1993) found
that shade increased foliar phenolics in sugar maple. It is very likely that the
compositional structure of plant defenses is more important than total levels of groups of
compounds. as specific herbivores will respond to only some subset ofthe chemicals
present (Kinney et al. 1997). Therefore. while it is clear that shade altered the
composition of foliar defenses in birch and maple in this study. it is not clear how these

alterations will affect their resistance to herbivores (see Chapter 3).

74

The effects oflight on foliar defensive compounds disappeared in 1997 when
trees were again exposed to full sun. This reinforces the conclusion that birch and maple
respond quickly to long-term shifts in light availability such as may be experience over

the course of succession.

Effects of Nutrients on Birch and Maple Physiology.

Fertilization did not affect photosynthetic rates of birch or maple. llowever.
fertilization did increase the foliar nitrogen content of sun-grown birch in August but not
June of 1996. Increased foliar nitrogen levels are often associated with increased
photosynthetic rates (Reich et al. 1991), therefore the lack of response of photosynthesis
to fertilization in the present case is a bit surprising. This lack of response in birch could
reflect the fact that photosynthesis was measured in July of 1996, which may have been
before fertilization had its effect of increasing levels of nitrogen in birch foliage that year.
Also, although foliar nitrogen content ofbirch was increased in August. this was
measured on a per mass basis. Ellsworth and Reich (1992) found a significant
relationship between light-saturated photosynthetic rates and nitrogen on a per leaf area
but not per leaf mass basis. The distribution of nitrogen per unit area but not per unit
mass may reflect optimal distribution of photosynthetic apparati in foliage.

Surprisingly. fertilization reduced height growth ofbirch and maple in 1995. In
contrast. increased nutrient levels have been shown to increase the growth rate of small
birch (Betula pendula. Linder and Rook 1984. McDonald et al. 1986) and juvenile paper

birch (B. papyri/era, Bryant et al. 1987). Observed reductions in height growth in 1995

75

may be the result of an interaction of fertilization with shade that was undetected that
year. Although there was no significant effect of shade on height growth in 1995. a
reduction of height growth in shaded treatments in 1996 was observed only in fertilized
trees. Both shade and fertilization are expected to decrease plant C/N ratios. which might
explain the interdependence of these two stressors. Height growth may be a function of
C/N ratios in the plant. However. this speculation is not supported by the observation
that although shade decreased height growth in 1996. there was no effect of shade on UN
in any year of this study.

In addition to fertilization being associated with reduced height growth. diameter
growth was greatest in non-fertilized trees (with fertilization effects influenced by light
and defoliation). Even so, for the most part fertilization had no effect on birch or maple
growth rates. This indicates that nitrogen was not limiting in the non-fertilized
treatments. In fact, nitrogen levels in fertilized treatments may have been sufficiently
high such that some other micronutrient became growth-limiting. Subsequent effects of
fertilization may be due to shifts in concentrations of one or more micronutrients. not in
concentrations of nitrogen. This is supported by field observations of chlorotic leaves on
both species, which may be indicative of manganese deficiency (data not shown).

Although nitrogen was not growth-limiting. increases in available nitrogen and
micronutrients with fertilization did decrease defensive allocation in late season birch.
Levels of total phenolics were decreased in young. fully expanded birch foliage of
fertilized trees in August. This response was in agreement with carbon/ nutrient balance
predictions although it was not accompanied by significant increases in C/N ratios in

foliage of fertilized birch. In contrast to the August response. fertilization did not

76

influence C-based defenses of foliage sampled in June in either species in either year.
These results suggest that subtle differences in mineral and/ or nutrient availability can
influence defensive allocation and therefore herbivore resistance (Chapter 3). even when
nutrient availability is not at levels low enough to depress growth. These effects may
vary over the season. and are likely to depend on the plant species. Dudt and Share
(1994) found no influence of fertilization on phenolic levels or herbivory ofdogwood or

poplar trees.

Effects of Defoliation on Birch and Maple Physiology.

Defoliation was not associated with compensatory increases in photosynthesis in
this study, although others have found increased photosynthesis in the remaining foliage
of defoliated trees (Ovaska et al. 1992, Lovett and Tobiessen 1993). However. this study
is not a strong test of defoliation effects on average photosynthetic rates as there is much
canopy-wide variation in photosynthetic rates (Ellsworth and Reich 1993) and in this

study photosynthesis was measured on only two leaves per tree.

Defoliation effects often mimicked those of nutrient deficiency in that fertilization
often mitigated the effects of defoliation and vice versa. The depression of birch
photosynthetic rates associated with fertilization was not seen in defoliated trees (see
above). Defoliation decreased foliar nitrogen levels in non-fertilized birch.
(.‘orrespondingly. defoliation was associated with increased C/N ratios. but this was true

for both fertilized and non—fertilized trees. Hunter and Schultz ( 1995) also found

77

fertilization to mitigate defoliation effects in oak. Defoliation may cause nutrient
deficiency because foliage contains the majority ofthe plant‘s nitrogen supply (Chapin et
al. 1987). while other resources such as carbon are available from stored reserves in other

plant parts.

Defoliation did reduce height growth in both species and there was no evidence
that fertilization mitigated this effect. Because defoliation levels differed between birch
and maple. no comparisons of relative species response can be made. However. Schaffer
and Mason (1990) found herbivory to have a greater effect on growth in sun-grown rather

than in shade-grown tropical trees.

Defoliation had no strong effects on levels of carbon-based defenses in maple
foliage. This was most likely due to the low levels of defoliation experienced by this
species. In birch, defoliation levels were more severe and were weakly negatively
correlated with condensed tannin levels of June-sampled birch foliage in 1996. but no
delayed effects of defoliation were observed in 1997. For insects sensitive to total
phenols, this effect could lead to increased resistance of birch in the season of defoliation
(see Chapter 3 for examination of foliage feeding Lepidopteran responses). Related to
these observations. Dudt and Share (1994) found herbivory and total phenolics to be
inversely related on shade tolerant dogwood trees. but they found no relationship between
herbivory and total phenolics in the shade intolerant tulip poplar. F aeth ( 1992) found
increased hydrolyzable tannins. but lower condensed tannins in rcflush foliage of

defoliated oak relative to levels found in mature primary leaves.

78

Conclusions.

Paper birch and sugar maple trees differed in intrinsic physiological traits as
predicted based upon life history considerations. Shade-intolerant paper birch had greater
photosynthetic rates than shade tolerant maple. The higher level of shade tolerance in
maple could not be explained by superior low light photosynthesis or respiration rates. as
birch had greater photosynthetic rates at all light levels and respiration rates did not differ
between species. Rather shade tolerance may be more dependent upon low light survival
(Kobe et al. 1995) which may be influenced by patterns ofcarbohydrate storage (Kobe
1997). Paper birch grew faster than sugar maple, but was more likely to suffer reduced
growth in response to environmental stress, especially shade. Birch also exhibited more
plastic responses to shading and fertilization treatments. Although leaf morphology
differed in sun and shade treatments, two years of shading did not affect photosynthetic

light response curves in birch or maple, suggesting that neither species effectively

acclimated to shade.

Resource acquisition in both species was influenced by shade. fertilization.
defoliation, and the interaction of these stressors. Shade limited carbon acquisition of
both birch and maple, and decreased foliar C/N ratios as predicted by the carbon/ nutrient
balance hypothesis. However, this decrease in the C/N ratio was driven by increases in
foliar nitrogen content. not by decreased carbon content which was not affected.
Fertilization did not affect carbon acquisition. and there was only limited evidence of
fertilization influencing nitrogen acquisition. It is not possible from these data to say if

defoliation directly influenced nitrogen acquisition: however, defoliation was associated

79

with decreases in foliar nitrogen levels. This may indicate that nitrogen uptake was
reduced in defoliated trees, perhaps as a result of decreased leaf area available for
transpiration to drive root uptake. Alternatively, shifts in the allocation of available

nitrogen may have occurred as nitrogen in the foliage was removed by herbivores.

Resource allocation patterns in both birch and maple were also influenced by
shade, fertilization, and defoliation. Shade reduced growth of both tree species and
altered the composition of foliar defenses. The carbon-based defense cornponnds of
shaded trees differed compositionally from those of conspecifics grown in full sun. In
birch, total phenolic levels remained constant although compositional changes occurred.
while in maple total phenolic levels decreased. Shade decreased specific leaf mass.
presumably resulting from the plant’s attempts to maximize photosynthetic capacities
under limiting light conditions. Fertilization also influenced resource allocation. but did
not have effects as strong as those of shade. Fertilization decreased height growth in
1995 in both birch and maple, but overall did not have a strong effect on tree growth.
suggesting that non-fertilized treatments were not strongly nitrogen limiting. In fact.
fertilized treatments may have had a surplus of nitrogen resulting in other nutrient
limitations. Fertilization did have subtle effects on defensive allocation in birch but did
not influence the defensive allocation of maple. The effects of defoliation were often
dependent upon the fertilization treatment of the plant even though fertilization was not
generally growth-limiting. Defoliation resulted in reductions in allocation to height

growth. but did not influence defensive allocation.

80

Shade, fertilization, and defoliation had influential effects on both resource
acquisition and allocation in paper birch and sugar maple. Tree responses to these
stresses were influenced by intrinsic characteristics, and therefore often differed among
the two species. Moreover, the effects of one stress on tree health was often dependent
upon simultaneous exposure to other stresses. The current understanding of plant
responses to their environments will benefit greatly from more mechanistic studies aimed
at determining the physiological steps involved in these responses. Only with a
mechanistic understanding of plant responses will we be able to accurately predict the

impact of interacting environmental stresses on individuals and populations.

81

CHAPTER 3:
EFFECTS OF LIGHT, NUTRIENT AVAILABILITY, AND DEF OLIATION ON
THE RESISTANCE OF PAPER BIRCH (Betula papyrifera) AND SUGAR MAPLE

(Acer Saccharum) TO CHEWING F OLIVORES

INTRODUCTION

Light and nutrient levels vary in forests as a result of natural events. such as gap
formation when trees fall, and anthropogenic perturbations, such as logging. road
construction, and nutrient loading from atmospheric nitrogen deposition (Vitousek et al.
1997). Herbivorous forest insects can be affected both directly and indirectly by abiotic
factors such as light and nutrient availability, and by biotic factors such as defoliation of
their host plants.

Shade stress has been associated with both increases (Bryant et al. 1987,
Bultzman and F aeth 1988, Nichols-Orians 1991, Dale and Causton 1992, Lowman 1992,
Shure and Wilson 1993, Johnson et al. 1997) and decreases (Coley 1983, Lincoln and
Mooney 1984, Bultzman and F aeth 1988, Bassett 1991, Koricheva et al.1998) in
herbivory or herbivore densities. Changes in herbivore abundance and in observed
amounts of herbivory (reflected in lost leaf area) can result from behavioural responses of
herbivores to light. These can include increased abundance in areas of high light for
purposes of thermoregulation or higher abundances in shaded areas where insects may be

less visible to predators. Differences in total herbivory observed between sunny and

82

shaded areas may reflect differences in herbivore abundance, or indirect effects of light
on herbivores mediated though changes in the quality of host foliage. Increased feeding.
observed as increased amounts of herbivory, may reflect increased preference for a host
or, alternatively, may reflect increased feeding to compensate for low nutritive quality of
the host plant. To determine how host quality affects herbivore performance.
physiological and fitness parameters such as growth rate, nutritional indices, survival. and
fecundity must be examined. However. surprisingly few studies linking light
environment to host quality have directly measured insect performance (but see Lawler et
a1. 1997).

Nutrient availability, like light availability, has been shown to have variable .
effects on herbivores. Fertilization most frequently has been implicated in increased
performance of individual herbivores (reviewed in Kyto et al. 1996). Population-level
effects are more variable, but for populations of leaf-chewing insects fertilization more
frequently had negative effects than positive effects (Kyto et al. 1996). On some
occasions, host nitrogen availability had no effect on herbivores (Floater 1997, Dudt and
Shure 1994). It is unlikely that nutrient levels directly affect herbivore populations.
Rather, effects are linked to the nutrient content of host foliage. nutrient-induced changes
in concentrations of defensive chemicals, and to the effects of fertilization on community
structure (e. g. increased ground cover supporting increased predator and parasitoid
populations) (Scriber 1984, Kyto et al. 1996).

Defoliation-induced changes in host foliage often increase. and occasionally
decrease. the host’s resistance to the defoliating herbivore (Tuomi et a1. 1988. Karban and

Baldwin 1997) or other species with which it shares the host plant (reviewed in Denno et

al. 1995). Effects of defoliation on insect performance (growth, survival, fecundity) and
herbivory may be observed during the same season (Haukioja 1980, Roland and Myers
1987, Hanhimaki 1989, Hunter 1987, Dankert 1995, Wold and Marquis 1997), or in
following years (Roland and Myers 1987, Haukioja 1982, Haukioja 1990. Tuomi et al.
1990), and can be environmentally dependent (Dankert et al. 1997). Effects of severe
defoliation on herbivores feeding concurrent to the defoliation event may reflect direct
density-dependent behavioral or physiological changes in individuals in response to
competition, in addition to changes in host quality. For example, individuals in hi gh-
density populations may disperse to escape competition. regardless of defoliation-induced
changes in host quality.

Indirect effects of abiotic and biotic environmental stresses, including light.
nutrients, and defoliation, on herbivores are mediated through changes in the quality of
host foliage. This “quality” is often associated with levels of foliar nitrogen (Tuomi et al.
1984. Nichols-Orians 1991) and water content, which can be major limiting resources for
insects (Scriber 1977, Mattson 1980, Scriber and Slansky 1981). Host quality is also a
function of morphological traits of foliage such as the presence of trichomes and leaf
thickness, and can be dependent on temporal synchronization of insect and plant
phenology (Kolb and Teulon 1991). Host quality is often inversely related to levels of
secondary metabolites (such as carbon-based polyphenols). which may play a defensive
role (Feeny 1970. Haukioja et al. 1985. Johnson et al. 1997; but see Bernays 1981. Ayres
et al. 1997). Stressed hosts may also protect themselves by attracting predators or
parasitoids of their herbivores (Turlings et al. 1995). Nutrients, defensive chemicals. and

morphological changes in foliage can slow the development time of insects and thereby

84

increase their time of exposure to predators and pathogens (Coley and Barone 1996). In
Chapter 2, I discuss theoretical predictions regarding the effects of light, nutrient. and
defoliation stress on these foliar traits and report observations on the effects of these
stresses on paper birch and sugar maple. In the current chapter, I examine the
performance of insects feeding on foliage of these trees to determine how treatment

effects on host traits translate into effects on herbivore resistance.

The objectives ofthis study were to determine the indirect effects ofdefoliation.
light, and nutrient availability on the lepidopteran defoliators gypsy moth (Lymantria
dispar, (L.) Lymantriidae), forest tent caterpillar (Malacosoma disstria, Hubner.
Lasiocampidae), and Whitemarked tussock moth (Orgyia leucostigma, (J. Smith),
Lymantriidae), as mediated through effects on the quality of sugar maple and paper birch.

Gypsy moth is an exotic species which was introduced into eastern North America
from Europe in the 18605 (Michigan Department of Agriculture et al. 1984) and has since
spread rapidly westward across the eastern United States. Because of its recent
establishment, gypsy moth and forest tree populations in North America have not had an
opportunity to co-evolve. A polyphagous feeder, gypsy moth eats a wide range of hosts
including paper birch and sugar maple. which are abundant in the forests of the Great
Lakes region. Gypsy moth is an outbreak species which undergoes cyclic changes in
population sizes. It is univoltine in the United States and feeds early in the season as
hatch occurs concurrent with bud break.

Whitemarked tussock moth is a native polyphagous folivore, and therefore has an

evolutionary history with its North American hosts. including paper birch and sugar

85

maple. Whitemarked tussock moth is bivoltine in central Michigan. with the first
generation feeding early in the season and the second generation feeding toward the
middle to the end of the growing season. Second generation larvae hatch approximately
six weeks after gypsy moth.

Forest tent caterpillar, like Whitemarked tussock moth, is native to North America.
ranging from northern Canada to northern Mexico (Fitzgerald 1995). Like gypsy moth.
forest tent caterpillar is a univoltine, highly polyphagous outbreak species with a host
range that includes paper birch and sugar maple. As early instars, forest tent caterpillar
larvae are gregarious, but become solitary and wander upon reaching the fourth instar.

As indicators of insect performance, I measured survival, larval development
time, pupal mass, relative growth rate, and nutritional indices of the insect species in
1996 and 1997. Insect bioassays were conducted concurrent with studies reported in
Chapter 2. in which I reported effects of light. nutrient. and defoliation stress on a suite of
foliar traits of paper birch and sugar maple (1996 data are summarized in Table 3.1 and
1997 data are summarized in Table 3.2). To eludicate potential mechanistic links
between insect responses and foliar characteristics, I examined correlations between
insect responses and foliar carbon, nitrogen, specific leaf mass, and polyphenols
including total phenolic and condensed tannin concentrations. Insects were predicted to
perform better (i.e. grow faster. develop more quickly. or have higher survival or pupal
mass) on hosts with relatively high levels of foliar nitrogen and relatively low levels of
total phenolics and condensed tannins. Predictions of how the responses ofthese
variables to environmental treatments will affect insect performance are presented in

Tables 3.1 and 3.2.

86

Table 3.]. Predicted effects of treatments on the performance (growth, pupal mass, and/ or duration of
larval period) of chewing herbivores based on their effects on foliar traits in 1996 (percent change from
control levels). Increases in foliar nitrogen and decreases in condensed tannin and/ or total phenolic
content are predicted to be beneficial for chewing herbivores. Direction of light, nutrient, and defoliation
treatment effects on foliar traits of paper birch and sugar maple relative to control levels are from Chapter
2. ‘

 

 

 

 

Treatment ianer birch Sggg Maple
June August June
SHADE N N
C --- -2% C ---
C/N --- -20% UN ---
CT -3 7% -3 5% CT ---
FD --- --- FD - l 7%
SLM --- -15% SLM -200/0
Predicted insect performance: improved improved improved
FERTILIZATION N N
C --- --- C «-
C/N --- -9% C/N «-
CT --- --- CT ---
FD --- --- FD ---
SLM --— --- SLM «-
Predicted insect performance: no effect no effect no effect
DEF OLIATION N -7% N
C --- --- C ---
C/N --- +80/o C/N ---
CT CT ---
FD ~-- FD ---
SLM --- +8% SLM ---
Predicted insect performance: no effect reduced no effect

 

IBirch foliage was sampled in both June and August; maple was sampled only in June. Birch foliage
sampled in August has flushed at budbreak (“mature foliage” sensu Chapter 2). Tussock moth fed on
foliage closer in phenology to that sampled in August while gypsy moth fed on foliage closer in plrenolog ,
to that sampled in June.

7 . . , . , . . _, .
’Abbrevratrons: C: % carbon. N: % nitrogen, UN: carbon/nitrogen ratio. LT: % condensed tannrns. FD:

% Folin- Denis reactive total phenolics, SLM: specific leaf mass (g/mz). CT, FD. C. and N were all
measured in units of % foliar dry weight.

87

Table 3.1. Predicted effects of treatments on the performance (growth, pupal mass, and/or duration of
larval period) of chewing herbivores based on their effects on foliar traits in 1997 (percent change from
control levels). Increases in foliar nitrogen and decreases in condensed tannin and / or total phenolic
content are predicted to be beneficial for chewing herbivores. Direction of light, nutrient, and defoliation
treatment effects on foliar traits of paper birch and sugar maple relative to control levels are from Chapter

 

 

 

2.l
Treatment Paper birch Sugar Maple
PRIOR YEARS’ SHADE N --- N ---
C --- C m
C/N --- C/N ---
CT --- CT —--
SLM --- SLM ---
Predicted insect performance: no effect no effect
FERTILIZATION N N + r 4%
C --- C ---
C/N --- C/N - 10%
CT --- CT ---
SLM --- SLM ---
Predicted insect performance: no effect improved
PAST YEAR’S N N
DEFOLIATION C C
C/N +7% C/N ---
CT --- CT ---
SLM --- SLM ---
Predicted insect performance: no effect no effect

 

' Abbreviations: C= % carbon, N= % nitrogen, C/N= carbon/nitrogen ratio, CT= % condensed tannins.
SLM= specific leaf mass (g/mz), --- = no change. CT. FD, C, and N were all measured in units of "/0 foliar

dry weight.

88

An improved understanding of the mechanistic links between environmental
factors and insect performance is essential for accurately predicting the effects of natural
and human disturbance on herbivory in forested ecosystems. Understanding stress effects
on gypsy moth and forest tent caterpillar is especially important as these species are the
most important defoliators of deciduous forests in North America (Mattson et al. 1991).
Furthermore, these environmental factors may mediate indirect interactions among
herbivores. Ifdefoliation alters host quality. then outbreaks ofone species may naturally
regulate the population dynamics of another through host-mediated interspecific
competition. Data on stand susceptibility, strengthened by a mechanistic understanding
of stress effects on tree resistance to herbivores, is essential for the effective stewardship

of forested ecosystems.

MATERIALS AND METHODS
Experimental Site and Design and Treatments.

Studies were conducted with the same trees and research plots described in
Chapter 2. Details of implementation of light. fertilization. and defoliation treatments are
also provided in Chapter 2. To review. I examined two levels of light availability: 20%
and 100% of ambient. Shading was implemented in 1995 and 1996 with the use of shade
cloth. Fertilization levels were 0 and 225 kg ha" yr'I of 182529 NPK. with half
administered at the beginning and the remainder at the end ofthe 1994. 1995. and 1996

growing seasons. In 1996. halfofthe trees were naturally defoliated by forest tent

89

caterpillar. Defoliation reached levels of (mean i SE) 1 i 3% and 8 i 3% (range 0-

32.5%) in sugar maple, and 6 i 3% and 52 i 3% (range 0.5-98.5%) in paper birch.

Insect Bioassays.

Bioassays with gypsy moth, Whitemarked tussock moth. and forest tent caterpillar
were conducted in laboratories at the Dow Gardens in Midland. M I in 1996 and at the
Ohio Agricultural Research and Development Center, The Ohio State University. in

Wooster, OH in 1997.

vaSY moth.

In 1996, gypsy moth egg masses were obtained from the Canadian Forest Service.
Insect Rearing Facility. Sault Ste. Marie, ON. Upon hatch (22 May), groups of 12
neonates were confined in arenas with detached foliage from one ofeach of the
experimental trees. Egg hatch of lab-reared gypsy moth coincided with that of naturally
occurring populations. Insects were reared in a growth chamber at 25°C with an 18L:6D
photoperiod. Rearing containers were 150 x 25 mm diameter petri dishes with moistened
charcoal-amended plaster bases that provided ample humidity for leaves to remain turgid
between feedings. Foliage was changed three times per week, ensuring a constant and
fresh food supply. F rass and uneaten leaves were removed from dishes as needed. at least
once weekly and more often for later instars. To minimize growth chamber effects on
insect growth. the position of individual dishes in the incubator were shifted on each

feeding day.

90

meg coweoo was M33 533235 80¢ a
.822: 63 mm noun—:28 225 meSmE 9508 Mo WMOM .832: be 8m 8065 53.0.5 SE was 5.58 “8 wow: 8332 o

Emmet,» 63.3 3.25 I Emmy,» 835 ES» u 653% r.

mmefi mo $9: I coca—63:8 n cocmzfimmmmv

:mmmE REE \mme 35.: :2 \AmeE FEE I 39: 3:5 n $88 _mccucoaxo 525 “3 m
mmmE e08 _mch mme wood BEE n serge—ago N

$.33 roam do :2:an II. P _

 

32:05 9 e08 woumowme

 

 

02 ... Acocmzemmmw Ere/83 n Dom gov Gum Lo 56568.3 xocomuEm

2: .. 888328 Eofizéae u o< ea 3. ézcawa Business?
C. ... 03v \cosmEzmcoU n mom 3% ME we: mum 88 538338 96.2mm

H \QmmmE 3:55-338 EEEVGG u mwm 9b _-wE wEv mwm 8E 532w 9523*
SEN—dw—IMU 3:5 :c=w_>o._nn< 5?: Quote—flit

N.-— .moomfifim _NCOCEHSC two SCUM—5076 USN mCOCmTzvunflxx :MM o—DNr—t

91

 

 

pupal mass (9)

2.0

 

1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2

 

.
-
OC’
0 O
O.
0
OO

O
00. CD000.

00C)

0.
O

0 00000 C

O O
00.0 .0.
O

0.0

O

 

 

0.0
20

25 30

duration of larval period (d)

35

40

45

O Birch
0 Maple

 

 

 

Figure 3.1. Gypsy moth pupal masses and duration of larval period in 1996.
Data within host species form distinct clusters of high and low pupal mass/

development time combinations. Insects with pupal masses < 0.80 g that complete

larval development in 5 34 days were considered male.

92

 

Survival was measured after 12 d, at which point one larva per group, which by
visual estimation appeared representative of the average size, was retained for continued
rearing for later nutritional studies. Remaining surviving larvae in each group were
disposed of.

Nutritional indices, including relative growth rate (RGR), relative consumption
rate (RCR), approximate digestibility (AD), and efficiency of conversion ofdigested food
to biomass (ECD) were calculated gravimetrically for one larvae per experimental tree
using the equations listed in Table 3.3 (Waldbauer 1968. Gordon 1968). Growth.
consumption, and frass production were quantified during 48 h feeding bioassays
conducted within one day following molt to the fourth stadium (4-5 June). Larvae were
weighed prior to being confined in plaster-filled petri dishes (as described above) with
90-120 cm2 of fresh foliage (3-4 leaves) from the tree upon with they had been reared.
and were allowed to feed undisturbed for 48 h in the growth chamber. After 48 h. larvae
were again weighed, and area of unconsumed foliage was determined. Frass and the
remaining foliage were then dried at 60°C for 2 d, and weighed.

Following the 48 h bioassay, larvae were reared until pupation. Larval
development time (days from hatch to pupation) and pupal mass (without final instar
exuviae) were recorded 2-3 (I after pupation. Relative growth rate over the total larval
period was calculated using 0.0006 g as the average initial mass of a neonate gypsy moth
and pupal mass as the final mass of individuals.

Gypsy moth males generally have five instars while females have six. Hence. on

average males develop faster and are smaller than females; therefore, insects were sexed

by examining the distribution of development times and pupal masses. Pupal mass was
plotted against larval development time separately for both birch and maple-reared larvae
(Figure 3.1). Data formed two distinct clusters for each tree species: one with quicker
development times and smaller pupal masses and the other with relatively longer larval
development times and larger pupal masses. Insects with larval development times under
34 d and pupal masses under 0.8 g were defined as male.

In 1997. fourth instar gypsy moth were field collected from hybrid poplars
(Populus x euroamericuna c.v. “Eugeneii”) in Kalamazoo County. M1 on 16 .lune. These
insects were part of a separate defoliation experiment, and were eclosed from eggs
collected in Huron Township, Wayne County MI in late March 1997. Eggs were surface
sterilized (10% formalin for 1 h) and then placed in the field prior to budbreak. When
they reached the fourth instar, approximately 100 larvae were transported to Wooster. OH
in petri dishes without food. Nutritional indices were determined with 48 h bioassays
conducted on larvae 1-2 d into the fourth stadium. These bioassays were conducted as in
1996 (described above). Foliage from experimental trees (growing in Midland. MI) was
collected on 16 June, stored in air-tight plastic bags, and kept on ice during transport to
Ohio where it was refrigerated until bioassays were initiated on 17 June.

In 1997, second instar bioassays were conducted at the same time as the fourth
instar bioassays with a separate cohort of insects. Larvae used in this bioassay were
obtained as eggs from the Canadian Forest Service. Insect Rearing F acility. Sault Ste.
Marie. ON. They were reared at 25°C with an 18L26D photoperiod in groups of50 - 150

on a constant supply of river birch and paper birch foliage (from Wooster. Ohio

94

populations). Beginning on 18 June, two second-instar larvae were reared for 48 h on
detached foliage from each of the experimental trees. Relative growth rate was calculated

gravimetrically as RGR=[(ln(final group mass)-ln(initial group mass))/2].

Whitemarked Tussock Moth.

In 1996, Whitemarked tussock moth egg masses were obtained from the Canadian
Forest Service, Insect Rearing Facility, Sault Ste. Marie, ON. Hatch occurred on 19 .luly.
Groups ofeight neonates were confined in a petri dish with detached foliage from one of
the experimental trees. (one group per tree). Larvae were reared in growth chambers at
25°C with an 18L:6D photoperiod. Leaves were kept fresh and in ample supply. dishes
were cleaned as needed, and positions of dishes in the chamber were changed regularly as
described above.

Survival was measured after 14 d. at which point one larva per tree. which was
deemed representative of the average group size, was retained for continued rearing for
later nutritional studies. Additional larvae were discarded. Fourth instar nutritional
indices, larval development time, pupal mass, and relative growth rate from egg to pupa
were determined for one larva per experimental tree using methods identical to those
described for gypsy moth in 1996 (above and Table 3.3). Relative growth rate over the
entire larval period was calculated using 0.0004 g as the initial mass (estimated average
mass of neonates) and pupal mass as the final mass. Adults were sexed following

CliTCl‘gCDCB.

95

In 1997, Whitemarked tussock moth eggs were obtained from the Canadian Forest
Service, Insect Rearing Facility, Sault Ste. Marie, ON. Insects were reared from egg
hatch in growth chambers at 25 °C with an 18L:6D photoperiod. Hatch ofa second
cohort was delayed by refrigeration of eggs. and rearing temperature was adjusted as
needed such that fourth instars were available for bioassays on 17 June and second instars
on 18 June. Larvae were reared in groups of 50 — 150 on a constant supply of fresh paper
birch and river birch foliage (Wooster, Ohio populations). Nutritional indices were
calculated for fourth instars from gravimetric measures of larvae, foliage. and frass as
described above for gypsy moth. Relative growth rate of groups of second instars was
measured over 2 (1. Second instar growth was determined using the equation
RGR=[(ln(final group mass)-ln(initial group mass))l2]. Group sizes ranged from two to

three.

Forest tent caterpillar.

In 1997, fifth instar forest tent caterpillar were field collected from hybrid poplars
(Populus x euroamericana c.v. “Eugeneii”) in Kalamazoo County, M1 on 16 June (the
same trees from which gypsy moth were collected), and transported to Wooster, Ohio in
empty petri dishes. Nutritional indices were determined over 48 h (starting 17 .1 une) for
larvae feeding on detached foliage from experimental trees with methods identical to

those used for gypsy moth in 1997.

96

Insect Performance in Relation to F oliar Traits.

Relationships between insect performance and foliar traits were examined during
1996 and 1997. In 1996, Whitemarked tussock moth and gypsy moth relative growth rate.
relative consumption rates, approximate digestibility. efficiency of conversion ofdigested
food to biomass, pupal mass, and duration of the larval period were examined for
correlations with foliar carbon, nitrogen, carbon/nitrogen ratios, specific leaf mass.
condensed tannin, and total phenolic levels ofbirch and maple foliage collected 1 1 .lune
1996. Additionally, Whitemarked tussock moth traits were tested for correlations with
these same foliar traits of birch foliage sampled 1 August of 1996. Gypsy moth fourth
instar 48 h feeding trials were run 4-7 June. one week prior to .lune sampling of foliage
for chemical analyses. Whitemarked tussock moth fourth instar 48 h feeding trials were
run 19-21 July, 5% weeks after June sampling. and 2 weeks prior to August sampling of
birch.

In 1997, relative growth rate, relative consumption rates, approximate
digestibility, and efficiency of conversion of digested food to biomass in fourth instar
gypsy moth, fourth instar Whitemarked tussock moth. and fifth instar forest tent
caterpillar were examined for correlations with foliar carbon. nitrogen. carbon/nitrogen
ratios. specific leaf mass, and condensed tannin levels ofbirch and maple foliage
collected 16 .lune (the same date as foliage was collected for bioassays). Details of
foliage processing. chemical analyses. and effects of light. nutrient. and defoliation stress

on plant physiology in both years are provided in Chapter 2.

97

Statistical Analyses.

Treatment effects on relative growth rate, nutritional indices (ECD, RCR, AD).
pupal mass, and larval development time were analysed by analysis of variance
(ANOVA) using SAS PROC GLM (SAS Institute, Inc. 1990; version 6.1 1 ). Shapiro-
Wilkes tests (SAS PROC UNIVARIATE) indicated that data met assumptions of
normality required for ANOVA analyses, while visual examination of box plots (SAS
PROC UNIVARIATE) indicated equal variances. Extreme outliers (greater than 3
standard deviations from mean) were removed from analyses only if they were associated
with larvae that had molted during the 48 11 study or had otherwise been noted to be
unusual, or if the datum was obviously biologically unrealistic (ECD or AD < 0 or > 100;
RCR or RGR excessively negative).

Blocks were treated as random effects, while species, light, nutrient, and
defoliation were treated as fixed effects. Models were adjusted to account for the split-
plot design with the addition of a separate test for shade effects (the whole plot factor).
Effects of the light treatment were tested for with the block*light interaction mean square
error (MSE) as the F -test denominator. All bar graphs report least square means :
standard error; standard errors of shade least square means were calculated by SAS with
block*light as the error term. Relationships between percent defoliation and relative
growth rate, ECD, AD. and RCR were also examined using correlation analysis (Pearson
correlation coefficients. SAS PROC GLM) with defoliation treated as a continuous

variable.

98

Table 3.4. Chi-square values, degrees of freedom (d/), and p-values from nonparametric analyses (SAS
CATMOD) of light, species, nutrient, defoliation effects and their interactions on gypsy moth 12 d survival
and Whitemarked tussock moth 14 d survival in 1996. 60 i N <_ 73 insects on six trees per treatment
combination. '

 

Survival
Source of Variation df gypsy moth wlritemarked tussock moth
Intercept 1 62.63 “” 10104.34 *“*
Species l 3.67 # 527‘
Light 1 3.19# 1.26
Species‘Light l 0.30 3.21 #
Nutrient l 3.88“ 4.17“
Species*Nutrient l 0.70 8.16 * *
Light*Nutrient 1 0.07 0.12
Species*Light*Nutrient l 0.91 4.05“
Defoliation l 2.86# 2.18
Species*Defoliation l 0.39 3.36::
Light*Defoliation l 1.75 3.10%:
Species* Light*Defoliation I 0.17 0.43
Nutrierrt*Defoliation l 0.00 0.00
Species‘Nutrient“Defoliation I 0.22 0.31
Light*Nutrient* Defoliation l 0.26 0.04
Species*Light*Nutrient" Defoliation l 0.13 0.92

 

'#=p_<_o.ro, * =p50.05,**=p50.01.*m =pgo.000I.

99

.5862” ......o.owqu:.8.owqu .52.???

 

 

S S E K E a 55
mm... 3.. NM... 8... a 3m . 8.5.6.09.22.52.23.aeam
2:. So a... a... one . SEESassazaas
m . .. 8... S... R... one . 8.3.6.8....E.§z.aeam
8.. S... Sm mom a... . 822.88.22.52
”.3 0...... .3 mm... 3... . 8.5.85tawsaaaafi

.. woe we. . mo... 3.. S... . 83.85....5...
8... one 8... mm. x... . 8.5.0.09393
.2 8... EN 26 3o . 8.5.88
we. N2 :3 8.0 a... . 5:52.55 .3625
a; w . .. a... $3 _. mm... . 35:52.5...

3. 8.... No. 3.... 6.. cm... . 5:52.393
m3 mm. 3... SN m3 . 22.52

.. a... a new a .3 3.. a... . 5.4.3.83

r. N; r. 3... r: 8.8 So .. m2 . 8.03m
x... a... NE. and 9.... N Ewing...

8.0 S... mac 3... a No. . . 2w:

.. m2 o..~ 3.. 3.. mi N .85

com. 9. mom 3 mom .92 m9. ts 8:22.30 858

.EB .83 05 3
£375.83 :23 38328 803 Ew: .8 829% .Eaefim 558 05 waist 3582: 50m: .68 enamowfi do 5.83:8 do CECE... new row: b=350w6
Baggage .955 8a.. cozafizmcoo 3:29. ecu A3 momv 5:68... 5.58 as mats—o can .288 MOM: cognac o. :22 wwo Eo¢ 3.38:. 8:: £38m
3:22 50.: .393 .8 228835 .65 can flout» :23280n 6.85:: .86on .Ew: .8 m<>OZ< Eat $3 Eoeooc .8 82woe ecu mosaik .m.m ~35.

100

.80 w. a. 8%...

3.8.88: 8 £38m 88:. :0 float“. 8.8% E8....:m.m 886:. 332?. .; M3 .95 .5532: 50:. $3.0. .32.
588.8 ADV Sum. £285 0. woo. 888%: 8 8.23:8 .8 556$» 8:: .CV 54; 5.3.8338 32.9089. Am:
EU”: 8:: 8:95.88 8.2.3.3. $9 coo. E 2qu Ema can :85 8%.. :0 5.80.8: RES... 8:82. o. :39. $0
Eoc 28 .82. £88 2.: wcgu .. wv 8>o 82:30:. 50E >83 .8 3.05 3:: crazm 82.23. .~.m Ezwr—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

285. :25 was: :85 m
u:
r p _ . )
co m
a. .
M ed .r
x. 0/0 D.
( I ., 3” (L
0 ed m
mam—>— cohm 02th 8.33 :39... :3 W
. _ O a
_
FL 0 )
E F.° w
on . 6
8 W .. Ne w
.. 8 w no p.»
8 e... (_L
<
cow 0 md

 

 

 

 

 

l0]

38¢qu::.s.owqu:.8.oqu #2.???

 

 

5 mm om R i \n 55
3o 2 .o _ 8:353.3222...sz .8303
Ed go coo o3 _ c2585as.._sz.zw_,_
£1 ”no 2: NS _ 8:26thL=2._sz.§uam
So So So :3 _ 836.39.20.22
mi Ed So NS _ 8:25523.3;on
3c 86 3.0 So _ ceifiotcw:
SN :3 $.N SS _ 53.853283

. :5 So of o: _ 8:385
a; go 3 .o E .c _ Eocsztsyfscam
:5 So 2.. So _ 29:52.23
86 86 Rd 3.0 _ .Ehszausam
3o 5.. $6 a 3m _ 50:32
2: 23 Rd 5.0 _ 23.36%
86 :: tam : 33 5a _ 8:25
mg 8.9 m3 m3 N 294285
”3 a 3.” . £3 2... _ am:
one as So of N v.85

EMS—om ENE 39:0“ ‘ 22.: \h cocmtm> go uuhzom

2:: Eviasgon man—2 .59:—

_.E._8 hobo 05 mm szkooB 2:3 35:28 22> Em: he mos—gi .8325 25 33:. E 5:83 :35 EB
32: .33 £9: 39%. no 228225 :2: new 98:» cozmzeon J5EE. .86an .Ew: .8 <>OZ< Eoc QB 8809: .8 moohwun new 32m>i 66 «Bah

102

38¢qu ::.S.owqu: .modwau fledwquf

 

 

8 we S _N 2 xx 55
moo NN.o a: :5 Nno _ ceéoxaE23229;328%
98 N _ .o m; :3 a 3m _ SEEELscszEw:
Rd N _ ._ a: a 3N : w: _ ceazeootcuszz.8525
m _ .o ”to 3. 8d a: _ 55:85 15:52
a: NN.o 3o Nod n 3N _ 53:8823.39%
oNN Ed 8.. 2.0 3N _ 8:23:9wa
as 23 0; So u .3 _ ceipsoneuam
2; 36 NS 85 3d _ 522.85
23 ”mo NN.o 33 NvN _ zaészijneoam
Bo mod 8.0 cod .. 03 _ 52.52.52;
:.o 38 m; _N._ : z: _ Ecsznmsam

.. a: one a: N 3m N3 _ 2252

u 3N a 3m :N n 32 o; _ 23.88%

t N; ”Nd MN 2:... NS. :1. 3.8 _ 3625
3o 36 Po 02 «NN N 33385
N2 RN ”3 Ed o3 _ Em:
NN.N 3o .2 2.. 2d N v.85
mom o< 5m 3 mom S «cm kn SEES ho 85%

_._E2 hobo 05 3 £32003 5:» 33.3.8 203 Em: 5% 82?»; .Eansm 5.58 05 waist 3530:. 50m: 88 nonmowu “0 5382.8
“0 5:3qu 28 .55 bzficmowE osEioaam AMONG 8a.. :ozneacoo 3:22 98 £37. :5 10”: 5.58 can ANA mo”: “E83 2: wctsc N. 3 .53 8::
£38» 3:22 52: 393 so 223825 :2: 28 acute coca—eon .2532. .86on .Ew: ho <>OZ< ENE RS 8033 .3 88mg 28 8233* .N...” «E:

.306 w 8 806.: 72555:: 5 539m 635 :o Saute 860% EmucEwG 3865

mxmcfim< .; wv 8>o 3532: £08 3.9% .835 5.58 mo ADV Sum: mmmEofi 8 boom Baumfi mo 555.68 .6 mocofifio
was A8 at; xtzpcmowu BchNoaqa Am: 205 3:5 cocaEsmcoo mags—ox A<V .30— .: 23E 5?; new :85

Emma co E35 5.38 van 2508 2: wctzn : wv macaw cuEmmoE 50E 39$ («0 E05 88 539m mafiom .m.m whar—

 

 

 

mams =25 2%: :25

E _ N _ 1 9o

 

 

 

 

 

(%) 003
ll
:1: (VP L.5111 5w) 838

 

D

 

 

 

 

 

 

 

flaws. :25 :32: :3 RES ucu

 

 

 

 

 

 

 

5
(kn t15w! 6w) 393

   

 

 

 

:85 I * w... <

 

 

 

 

 

W4

Relative growth rates and relative consumption rates are ratios. Although some
controversy exists over the use ofANOVA to analyze ratios (Packard and Boardman
1988. Raubenheimer 1995. Beaupre and Dunham 1995). I could not use the
recommended analysis of covariance to analyze absolute growth and consumption
because assumptions of equal response slopes across treatments were violated
(Raubenheimer and Simpson 1992). Violation of assumptions was indicated by
significant interaction terms between covariates and main treatment effects in ANOVA
analyses (data not shown), indicating that the relationship between the covariate and the
variable of interest was not independent of the experimental treatments.

Survival was analyzed by SAS PROC CATMOD (SAS Institute. Inc. 1990) for
nonparametric analyses of categorical data. Data on foliar traits (foliar carbon. nitrogen.
condensed tannin, and total phenolic content. carbon/nitrogen ratios. and specific leaf
mass; details in Chapter 2) were correlated to insect performance by analysis of Pearson
correlation coefficients (SAS PROC GLM). All results reported are statistically

significant at or = 0.05 unless noted.

RESULTS
Insect Bioassays.

Gypsy moth.

 

Birch was a better host for gypsy moth than was maple at all stages of insect
development. In 1996. 12 d survival on maple (93.4 i 1.0%) was 3% lower than that on

birch (96.6 i 1.0%) (p = 0.056. Table 3.4). Approximate digestibility of maple to fourth

105

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

   

 

 

 

 

 

A 0.20 -
1996 [Inc fen
0.16 - Ifertilization
JP 0.12 . *
in
E 0.08 -
a:
‘E' 0 04
a: ' "
0
K 0.00 ,
sun shade
1.0 - ,
B 1 997 One fert
A 0.8 - Ifertilization
g. ,
.5
‘7 0.6 - *
a:
E 0.4 -
a:
.5. 0.2 .
5
a: 0.0 ,
sun pfiorshade

 

 

 

Figure 3.4. Effects of fertilization. influenced by shade. on gypsy moth relative
growth rates as measured over the total larval period in 1996 (A) and growth of
second instar gypsy moth measured over 48 h in 1997 (B). Asterisks indicate
significant effects of fertilization on insect growth (p S 0.05).

106

instars was reduced 28% relative to that of birch. and digestive efficiencies (ECD) of
fourth instars were decreased on maple by 21% (Table 3.5. Figure 3.2). However. larvae
were able to compensate for decreased ECD on maple by increasing consumption rates
37% relative to those on birch. resulting in maintenance of growth rates. Growth rates
over the duration of the total larval period were 5% lower on maple than birch (Table 3.5.
Figure 3.2). Female pupal mass was 21% lower on maple (1.06 i 0.05 g) than on birch
(1.35 i 0.06 g) (Table 3.6). while pupal masses of larvae defined as males (see Materials
and Methods) averaged 0.59 i 0.05 g for both host species. Male larval period was four
days longer on maple (30 d) than on birch (26 (I. Table 3.6).

In 1997. birch again was a better host for gypsy moth than was maple. Second
instars grew 32% slower on maple than on birch (Table 3.7. Figure 3.3). Although
approximate digestibility was equivalent on birch and maple (unlike in 1996). digestive
efficiencies (ECD) of fourth instars were again 28% lower on maple (Table 3.7. Figure
3.3). Unlike in 1996. larvae were not able to compensate for poor host quality through
increased consumption. and consequently growth was reduced 32% on maple (Table 3.7.
Figure 3.3).

The light treatment did not have a strong effect on gypsy moth performance. but
when effects were significant. shaded trees were lower quality hosts. Shade did not
influence survival in 1996 (Table 3.4); however. digestive efficiencies (ECD) of fourth
instar larvae fed from shaded birch (but not maple) were reduced 21% (Table 3.5). liCDs
of gypsy moth fed birch grown in full sun were 56 i 4%. while ECDs of larvae fed

foliage from shaded birch were 44 i 4%. Even so. shade did not decrease fourth instar

107

 

0.7 -

'50 0.6 -
To) 0.5 - Dno fert
E 0'4 ' Ifert
g 0.3 J
I: 0.2 -
o 0.1 -
a: 0 __

 

 

 

 

no defol defol no defol defol

birch maple

 

 

 

Figure 3.5. Interactive effects of fertilization and defoliation on the relative growth

rate (RGR) of second instar gypsy moth on paper birch and sugar maple in 1997.
Asterisks indicate significant effects of fertilization within defoliation treatments

of birch or maple (p 5 0.05).

108

 

50-
40-
30-
20-
10‘

0'1—

 

ECD (%)

 

 

 

 

[I] no defol
Idefol

no shade shade

El no defol
I defol

no shade shade

 

 

 

 

 

-‘ N
-l 01 N 01 (a)
l 1 n n l

0.5 -

 

 

 

 

-_.l

RCR (mg mg'1 d")
o

 

 

 

Figure 3.6. Interactive effects ofdefoliation and light treatment on digestive
efficiency (ECD) (A) and consumption rate (RCR) (B) of fourth instar gypsy
moth in 1996. Asterisks indicate significant effects of defoliation within light

treatments (p 5 0.05).

109

Table 3.8. Pearson coefficients for correlations between indices of gypsy moth (A). Whitemarked tussock
moth (B). and forest tent caterpillar (C) larval performance and host defoliation and variables of quality of
foliage sampled in June from paper birch (upper value) and sugar maple (lower value) for 1996 and 1997

laboratory experiments (34 5 N 5 48)." 2

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A, Foliage Variable
Gypsy moth carbon/
performance ”/o nitrogen condensed total specific leaf
variable defoliation nitrogen ratio tannins phenolics mass
1996
RGR L4 NS NS NS NS NS NS
NS NS NS NS NS NS
RGRtotal -0.4l ** NS NS NS NS NS
0.34 * NS NS NS NS NS
RCR NS NS NS NS NS NS
NS NS NS NS NS NS
ECD NS NS 0.30 * NS 0.37 * NS
NS NS NS NS NS NS
Pupal mass NS NS NS NS NS NS
NS NS NS NS NS 0.31 “
Larval NS NS NS NS NS NS
development NS NS NS NS NS NS
time
1997
RGR L2 NS NS NS NS NA NS
NS NS NS —0.30 " NS
RGR L4 NS 0.44 ** -0.45 ** NS NA NS
NS 0.32 “ NS NS NS
RCR NS NS NS NS NA NS
NS NS NS NS NS
ECD NS NS NS NS NA -0.32 *
NS NS NS NS NS

 

'#=pgo.io.* =p50.05.** =p50.01, *m =p50.0001.

2 Abbreviations and units: nitrogen (% dry weight). condensed tannins (°/o dry weight -— birch or maple
tannin equivalents). total phenolics (% dry weight — birch or maple tannin equivalents) measured only in
1996. specific leaf mass (g/mz). RGR L4 = relative growth rate during 48 h ofthe fourth stadium (mg mg‘I
d"). RGRtotal = relative growth rate during entire larval stage (mg mg‘I d"), RCR = relative consumption
rate (mg mg'| d"). ECD = efficiency of conversion of digested food to biomass (%). pupal mass (g). larval
development time (d). RGR L2 = relative growth rate during 48 h of the second stadium (mg mg'| d" ).
RGR L5 = relative growth rate during 48 h ofthe fifth stadium (mg mg'I d").

110

Table 3.8 (continued).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

B, Foliage Variable
Whitemarked
tussock moth carbon/
performance "/0 nitrogen condensed total specific leaf
variable defoliation nitrogen ratio tannins phenolics mass
1996
RGR L4 NS NS NS NS NS NS
NS NS NS NS NS NS
RGRtotal -0.6l **** NS NS NS NS NS
NS NS NS 0.39 “ -0.36 * .32 "
RCR NS NS NS NS NS NS
NS NS NS NS NS NS
ECD NS NS NS NS NS NS
NS NS NS NS NS NS
Pupal mass NS NS NS NS NS NS
NS NS NS NS NS NS
Larval 0.54 **** NS NS NS NS NS
development NS NS NS -0.40 ** 0.32 * NS
time
1997
RGR L2 NS NS NS NS NA NS
NS NS NS NS NS
RGR L4 NS 0.37 * -0.33 * NS NA NS
NS NS NS NS NS
RCR NS NS NS NS NA NS
NS NS NS NS NS
ECD NS 0.33 * -0.35 "‘ NS NA NS
NS NS NS NS NS
C, Foliage Variable
Forest tent
caterpillar carbon/
Performance nitrogen Condensed specific leaf
variable o/u defoliation nitrogen ratio tannins mass
1997
RGR L5 NS NS NS NS NS
NS NS NS NS NS
RCR NS NS 0.32 * NS NS
NS NS NS NS NS
ECD 0.33 * NS NS NS NS
NS 0.35 * NS NS NS

 

111

 

>

RGR (mg mg'1 d")

 

0.14
0.12
0.10
0.08 -
0.06 -
0.04 -
0.02 -

 

 

Birch 1996

O

’99

r = -0.41
p = 0.005

 

 

0.00

I I I I

20 4o 60 so
defoliation (%)

100

 

 

 

w

RGR (mg mg'1 d")

 

0.14
0.12
0.10
0.08
0.06 -
0.04

0.02 1

 

Maple 1996

r = 0.34
p = 0.026

 

 

0.00

80 100

20 4o, , 60
defoliation (%)

 

Figure 3.7. Relationship of defoliation with gypsy moth relative growth rate
(RGR) as measured over the total larval period on paper birch (A) and sugar maple

(B) in 1996.

 

growth rate (Tables 3.5). F urthermore. the light treatment did not influence the duration
of the larval period of males or females. but did reduce the pupal mass of female (but not
male) gypsy moth by 17% (Table 3.6). Pupal mass of females averaged 1.09 i 0.03 g on
shaded and 1.32 i 0.03 g on non-shaded hosts. There were no effects ofprior season's
shading on second or fourth instar growth or nutritional indices in 1997 (Table 3.7).

The effects of nutrient treatments on gypsy moth performance were dependent
upon host species. light. and defoliation. In 1996. fertilization was associated with a 2%
increase in the survival of early instars (Table 3.4). Fourth instar growth and nutritional
indices were not influenced by fertilization. but fertilization of both birch and maple
grown in shade (not sun) reduced the growth rates of larvae as measured over the entire
larval stage by 9% (significant light*nutrient interaction. Table 3.5. Figure 3.4).
However. fertilization did not influence larval development time or pupal mass (Table
3.7).

In 1997, fertilization of maple (but not birch) increased relative growth rates of
second instars by 23% (Table 3.7. Figure 3.5). Growth rates of second instars on
fertilized maple were greatest if trees had not been previously defoliated (significant
species*nutrient* defoliation interaction, Table 3.7, Figure 3.5). Fertilization also
increased growth of second instars by 15% on previously shaded hosts. but not on hosts
which had always grown in full sun (significant light*nutrient interaction. Table 3.7.
Figure 3.4). Unlike in 1996. fertilization increased efficiency ofconversion of digested

birch and maple foliage to biomass by 37% in 1997 (31 i 2% and 22 i 2% for fertilized

113

.socowqu t: .53qu 1... .Scqu 2.18.26wa . .332” a.

 

 

3 S S S 3 s 55
a _ .o a... as coo So _ 8:2.85.22.52.293.3;on
86 3o 93 3o :3 _ 8.338.Eacszzzws
:._ a a: 23 3o «2 _ =25_oaotsezzkeam
N2 moo 86 N3 N3 _ SEBLDPEUEE
_. Se _. 2:. 2.. 93 2d _ 83.8.5.zwssguam
.. New .3 5o .. 86 as _ 85.85.23
:3 is go mom .. w; _ 8.3.3398;an
:3 one a; 8.0 : a; _ 858.5
a a...” as m: a 9.2m m2 _ 22.52.23.35an
cod so NE m3 3.. 8.5 _ 25:52.33
2.. OS o; 43 3o _ zecszaguam
:3 cg 93 Mom 33 _ 22.52
_3 _3 3o :3 a a: _ 23.332;
1...... 2.2 Ca a Sn .2. 42 .1... was _ scam
one SO WE .. .2 a Sum N 23:85
So as as «no oi. _ 2w:
:2 2.. :2 :02 23 N 3.85
mom 3 mom 5 Mom _ee «9. a Eugenio 85%

258 echo 2: mm waxooE 53> “0223.8

225 Em: .8 3283* .8233 5.58 05 wage 3582: Sum: .8255 9 week nowmowG Co 55828 do 5:3qu use .55 b__5:mow_n osmEioaam
.355 2a.. coiEBEfi 3522 can .95 «OM: E233 £58 2: merge can :99 ~55 5:33 8 :82 wwo Eat 3582: 8:: 5.58» 3:22

50E xoomma “0338823 no 228885 :05 can saute cozfléon 45:5: .362? .Ew: do <>OZ< E0: $3 Eouooc .«o 38an can $267k d6. «Zn...

114

.8... w 3 was... .22....2.

.o 530% 68:. :o 28.8 8.8% .:8c.:w.m 88.8. mxm..8m< .: WV .03 8.38:. 50:. .3033 81:28:43 .82.
5.20.. mo An: Sum. 8255 o. 80.. 8.8m... mo :o.m.o>:8 mo .68.qu :5 AU. 27.. 3:5..8m6 B..E.xo.&:

Am. 20M: 8:. 5.38238 3.3.3. .2. com. :. 2an 2%.; 8w 8.5 .092. :0 6.8.8: HES... 5.89:. o. :82 wwo
E9. 8m .82. 5.30,. of w:..:: F. wv .o>o 8.38:. EOE .089: 8.88%.? .o 9.05 88. 530% 8.2.8.3. .w.m Ezw...

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  

 

   

2%.... sea 2%.... 5...... w
.1 _ O _ h o H
FL W
cm 3 N 5
8 m w
.. s w v m
an (
._. w
2:. D m
max—>— cohm norm: its. .33. :3 M
. _ c o... a
a M. m
V .
ow w
o W vd P
8 2%....n. m... h
2:0 no.5. ad 4.

 

 

 

 

 

 

 

 

 

acowuu.deowunn_

 

 

S v 2 v \n 55

_ 53:85ascszfiwgmocam

1e 23 _ 8:23th.3052ka

. . _ £2320on{22.532.503on

So 8d NS 8.0 _ ceéotozficsz

m3 . N3 . _ cocaéuo224.39%

30 2d 23 moo _ 8585.23

x o _ e EN 86 2 .o _ 83.85.3303
n. Fm N3 cod 3o _ SEES

. 8a . omo _ E_._sz.._sm_f8_oam

«3 «3 3o .2 _ 5222ka

3o 86 30 :.o _ 22523283
85 8o o3 38 _ 22.32

:5 8a one 2d _ $53283
m3 Ed 93 NS _ 8525

m3 .3 2.~ So N 23385
23 . on? ”3 u 2 .2 _ :3:
N2 25 23 n; N x85
29:3 2&5 EMEUM ENE \h :O_Hm_._m> mo oogsom

05:. ~=oEa2o>oQ _atax— mas—Z 315n—

ZEB Ste 2.: 8 E3383 2:3 vows—:28 22> Em: he 829i .mo_mE£ new 33E 5 20:33 _mtfl 28 $9: .83 52:

x833 3:36:82? :0 328882 :2: new manta cosm=80c .2533: .86on .Ew: we <>OZ< Eat Q3 Eonog mo mquon .25 mo:_m>¢ .3“. «Bah

116

4o¢wqu..«Qoqu.deowq"n_

 

 

so 00 _5 mm m» 5w.o+mw
a a: one _2 2: 3o _ 85.8825:52qu.8525
tam 3o Pm ._ 8m moo _ c228515:32.ij
So 3.. 3.. a _ .o 2: _ 83.88.Eosszaauam
owe ”mo 3.. So u «3. _ 8363922532
80 8o 86 h _ .o .2 _ 83.85233283
.3 v3 85 ”no N. .o _ 8:263?sz
NE 8d 3: 3o «2 _ ceéeooaocam
mg a: s _ .o 85 8o _ 856.5
3.0 m _ .N 3o :3 3. SM _ ,5_._sz..zws.8_uam
8d 8.0 3. O3. mod :3 _ 22.52.23
23 mod «3 mac 2d _ Ecszkcam
3 .o .3 as :3 NZ _ 22.52
8d 3.0 so :5 8o _ $53283
._ man So . ”3 t 85 3o _ 360%
:5 Po mod as _3 N Ewszogm
u 2.0 2.0 one am a 3.2 _ 5w:
:3 n _ ._ w _ .N 85 oi m 5%
sum. o< 5x 3 mom 3 mom E 8:35 go 858

i=8
Stu 2.: mm “3:383 5.3 8830.8 983 Em: .8 mos—gi .8233. 558 05 waist 3532: SUB coo; n28w6 b.o 553:8 mo 55650 new .53
bzszmowgn Bus—_xoaam .205 88 cozaEsmcoU «>322 can £95 35 ~55 558 was Am: x05 988m of macaw ; mv B>o WES £38m 3:22 50E

x833 coy—558E; co mcocUEBE :2: new auoto :ocfléou .2253: 360% .Ew: ho <>OZ< Eot Q3 E808» ‘6 823v can $297k ._ _.m «Bah

117

30.0 W 3 32?: 72825:: 5 532m 835 so 38.?“ 862W EnocEmfi 286E mxmcam< .; wv

$>o 32:23:: EOE V6033 umanBEB SE: 5.58 we ADV Sum: 3255 9 v08 83%? go 55528.8 _.,.u:2qu cam
.AUV AQ<V \Ezocmowfi 295x053 Am: 205 BE cocaEzmcoo gasox A<V 33 E oEmE Ema Ea :85 Bang :0

:32: 558 was 9.88 of mESv ; wv $>o 3:58.: EOE *5me cog—EEBEBE E05 29 £39m 9:on .9». Esz

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2%.). sea 2%.). :25 w

[ _ O F _ ed M.
_JL m

8 o m

y m h o.~ .l

ow % P

ow * on (..

can. 9v m

mass. :85 .29: :3. Ems .25 w

. . O _ . 0.0 a
a v .1- m m

ow \Gm .., nd W»

cm W ed P

8 2%20 .3 (t

85 semi w... <

 

 

 

 

 

 

118

and non-fertilized, respectively; Table 3.7). There was no effect of fertilization on fourth
instar growth rates.

Defoliation generally appeared to improve host quality for gypsy moth, although
some responses indicated poorer host quality of defoliated trees. In 1996, defoliation did
not affect early instar survival on either host species (Table 3.4). In sun (but not shade),
defoliation increased quality of both species. as indicated by 25% improvement in
efficiency of conversion of digested food to biomass (ECD) and 14% lower relative
consumption rates (RCR) of insects on defoliated hosts (significant light*defoliation
interaction, Table 3.5, Figure 3.6). Results of ANOVA showed no significant effect of
defoliation or species*defoliation interactions on gypsy moth growth as measured over
the entire larval period. However, for insects feeding on maple, correlation analyses
indicated a positive correlation between defoliation and growth during the total larval
duration and a negative correlation between these same variables for insects feeding on
birch (Table 3.8 A, Figure 3.7). Female gypsy moth took 2 (1 longer to pupate when
feeding on defoliated hosts, but pupal mass was not affected by defoliation (Table 3.6).

There was no evidence of delayed induced resistance of birch or maple to gypsy
moth. In 1997, prior season’s defoliation was not correlated with growth rates of second
instar gypsy moth, or with growth, relative consumption rate, or efficiency of conversion

of digested food to biomass in fourth instar gypsy moth (Table 3.8 A).

119

 

 

 

 

 

0.5 -
‘7. 0.4 -
'o
E! 0'3 ' Elno shade
a) 0.2 . Ishade
E s
O! 0.1 '
8
0.0 -—

 

 

 

 

 

no defol defol

 

 

 

Figure 3.10. lnteractive effects of light and defolation on relative growth rate
(RGR) of fourth instar Whitemarked tussock moth in 1996. Asterisks indicate
significant effects of shade on growth within defoliation treatments (p _<_ 0.05).

120

 

1996

 

Dnomn
I fertilized

0.14 d
0.12 4 *
0.10 .
0.08 -
0.06 4
0.04 -
0.02 1
0.00 1

 

 

RGR (mg mg'1 d")

 

   

 

 

 

 

 

sun shade

 

 

 

Figure 3.11. Interactive effects of fertilization and shade on Whitemarked tussock
moth relative growth rates measured over the total larval period in 1996. Asterisks
indicate significant effects of fertilization within light treatments (p g 0.05).

121

Whitemarked tussock moth.

As with gypsy moth, birch was a better host than maple for Whitemarked tussock
moth. In 1996, survival of early instars was reduced 6% on maple (87.5 i 2%) relative to
birch (93.2 i 2%) (Table 3.4). Fourth instars grew 26% more slowly on maple (Table
3.9, Figure 3.8). resulting from a 39% reduction in efficiency of conversion of maple
foliage to biomass relative to birch (Table 3.9. Figure 3.8). Unlike gypsy moth.
Whitemarked tussock moth were unable to sufficiently increase consumption to
compensate for reduced host quality, and growth rates measured over the total larval
period were reduced 9% on maple relative to birch in 1996 (Table 3.9, Figure 3.8). Host
species did not influence pupal mass (mean i SD: female 0.31 i 0.08 g; male 0.15 i 0.04
g) or larval development time (mean i SD: female 34 i 7 d; male 28 i 4 (I, Table 3.10).

In 1997, host species did not affect second instar growth, but fourth instars grew
45% more slowly on maple (Table 3.1 I. Figure 3.9). As in 1996, reduced growth
resulted from a 36% reduction in efficiency of conversion of maple foliage to biomass
relative to birch (Table 3.1 1, Figure 3.9). Larval consumption rates on maple were
reduced 34% relative to birch (Table 3.1 1. Figure 3.9).

Like gypsy moth, Whitemarked tussock moth did not respond strongly to the light
treatment. When effects were significant. shade decreased host quality. but these effects
were generally dependent upon the defoliation treatment. In 1996. early instar survival
was equivalent on shaded and sun-grown trees (Table 3.4). However. fourth instars grew
slower on shaded hosts relative to sun-grown hosts. but only when trees were defoliated.

Growth of fourth instars was reduced 30% on foliage from shaded hosts that had been

 

 

 

 

 

 

 

 

 

 

 

A BIRCH
*
25 -
7 20 " D no defol
i 15 - Idefol
8 1O 1
w 5 .
0 -
no shade shade
B MAPLE
25 -
A 20 "
°\°
v 15 - Dno defol
a
LIJ 5 ..
0 .l
no shade shade

 

 

 

Figure 3.12. lnteractive effects of light and defoliation on digestive
efficiency (ECD) of fourth instar Whitemarked tussock moth feeding on
paper birch (A) and sugar maple (B). Asterisks indicate significant effects
of defoliation on digestive efficiencies within light treatments (p g 0.05).

l23

 

0.12 -
0.10 d
0.08 a
0.06 a
0.04 -
0.02 -
0.00

 

C] no defol
I defol

 

 

RGR (mg mg‘1 d")

 

   

 

 

 

 

birch maple

 

 

 

Figure 3.13. Effects ofdefoliation on relative growth rate (RGR) ofwhitemarked

tussock moth measured over the entire larval period on paper birch and sugar
maple. Asterisks indicate significant effects of defoliation within host tree (p g 0.05).

124

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A
o
o
5‘ 9
‘76, . .
E
D)
g r = -0.61
m p < 0.0001
0
m 0.00 . . . u
0 20 4O 60 80 100
defoliation of birch (%)
B 60
o
50 - 0
E 40 r 9 o o 4..
9
.§ 30 W , 3 0’)
H
.- 0. t o ’9. "
g 20 -
E r = 0.54
g 10 ‘ p < 0.0001
g 0 I I l I
g 0 20 40 60 80 100
defoliation of birch (%)

 

 

 

Figure 3.14. Relationship between percent defoliation and Whitemarked tussock
moth relative growth rate as measured over the total larval period (RGR) (A). and
larval development time on birch (B) in 1996.

 

.sooowqu :.:._oo.owqu z; gown” _. Sowqu?

 

 

 

2 8 we 3 . I \m 55
2.0 Ed _2 :.o _ ceieuoLscszkwsbeam
”do 36 3o 23 _ gassesescszagws
8.. 0: 2; mac _ Sugars ..E_._sz.aeam
e _ .o of E o3 _ ceazotcéazsz
So :3 So So _ 85.85.:_.n._._..3_8am
3o 23 So 2.. _ 83.8.4923
SN m; 8.0 8.0 _ geezeaobcam
:3 :3 one So _ 856:9
cod who 2.. 23 _ E...E_.z.zw_f3_§m
a _ .o NS :3 3o _ 25:32.33
23 _ _.o 23 OS _ Eacszbeam
3o 03 3o Ed _ 20:52
So .3 :3 3o _ 23.39%

i... 5.: NS ._ Sn 3.: 8.2 _ 832$
:3 28 Rd 23 N Emjzugm
3.0 8.0 a 5.2 w: _ em:
33 an. 9.3 3. m 605
com o< mom 3 mom be SEES a 85%

_.E._8 cote 2: mm waxooz :23 88328 83> Ew: S; 329% .58: 8285
8 woe unauwzs .3 53328 yo 5:205“. 28 .53 bzficmumfi BameoEqm .205 22 cocaEsmcou 3:22 .34 mo”: 2.: 5.52% 3:32 5:928
E2 380.. :32: cc: :0 228885 :05 new 98:0 5:25.26 .5233: 528% .Ew: do <>OZ< Eoc Q3 Eonoé .8 $03.26 Ea BEEQ .~_.m «Ea...

126

.306 w 8 806.: 23225:: co 539m 885 :o mocha 86on Ewociwfi 8865

mxmcem< .: wv $>o @2338 5:928 E2 $28 :32. at: mo An: 50m: 3255 8 n08 33qu mo 2230250 ho
xocufifio new ADV AQ<V 5:550ch onmeoEam Am: 354: 0:: cosafismcoo gum—om A3 .39 E 2qu Ema use
:85 59a so .835 5E of wage ; wv $>o coSmmuE 5:888 E8 $28 mo 20.”: 22 539w 9533* .m_.m 2:!”—

 

 

mams. 55

_Il_

 
 

 

mass.

 

 

 

:25

m (.10 ,.6w 6w) sou

 

 

 

 

 

2%.). 5.5

 

 

 

 

 

(%) (JV

 

cor U

 

2922

 

 

 

 

 

 

:25

md <

 

 

 

defoliated by forest tent caterpillar (significant light*defoliation interaction. Table 3.9.
Figure 3.10). Shading had no effect on RCR, AD, or ECD of Whitemarked tussock moth
(Table 3.9). Shading did increase the development time of male but not female
Whitemarked tussock moth by 4 d. but did not influence pupal mass of either sex (Table
3.10).

In 1997, prior seasons’ shading did not affect growth rates of second instars
(Table 3.11), or fourth instar growth, relative consumption rates, approximate

digestibilities, or efficiency of conversion of digested food to biomass (Table 3.1 1).

Fertilization interacted with the light treatment to improve Whitemarked tussock
moth performance. In 1996, fertilization of maple increased survival by 14%, but the
effect was only seen in the shade (significant species*light*nutrient interaction) (Table
3.4). Fertilization did not affect fourth instar growth or nutritional indices (Tables 3.9).
However, fertilization did increase relative growth rate over the total larval period by
10%, but only when trees were grown in full sun (significant light*nutrient interaction,
Table 3.9, Figure 3.11). Fertilization did not influence Whitemarked tussock moth pupal
mass or larval development time (Table 3.10). In 1997, fertilization did not affect second
instar growth or fourth instar growth or nutritional indices (Table 3.1 1).

Early season defoliation ofbirch. but not maple, induced resistance to
Whitemarked tussock moth later during the same season. Defoliation did not have an
effect on the survival of early instars in 1996 (Table 3.4). Fourth instar growth was not
affected by defoliation, although in shaded birch defoliation was associated with a 40%

reduction in digestive efficiency (ECD) (significant species*light*defoliation interaction.

128

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A
0.7
A 0.6 -
1: 0.5-
1,, 0.4 -
E 0.3J
m . o ’ _
E, 3.? : ~ r: 0.44
t: . - . , p - 0.002
0 00 I 14 I l I
CK 0100 1.0 2.0 3.0 4.0 5.0 6 0
-0.2
foliar nitrogen (% dry weight)
B
Maple
1:
.. o
'm 04 - 00 :3,
E o
5’ 0.2 - 9 =
E O. . .. r 0.32
V ‘9‘ p = 0.046
g 0.0 4 . . f
t: I0 1.0 2.0 3.0 4.0 5.0 6 0
-0.2
foliar nitrogen (% dry weight)

 

 

 

Figure 3.16. Relationship between foliar nitrogen content of paper birch (A) and
sugar maple (B) and the relative growth rate (RGR) of fourth instar gypsy moth
in 1997.

129

Table 3.5, Figure 3.12). Defoliation of birch reduced growth over the total larval period
by 13%, bringing growth rates down to levels equal to those for maple-feeding larvae
(Table 3.5, Figure 3.13). Correlation analyses also showed a negative relationship
between percent defoliation and growth over the total larval period. and a positive
relationship between percent defoliation and duration of the larval period (Table 3.8 B.
Figure 3.14).

As with gypsy moth, there was no evidence of delayed induced resistance of birch
to Whitemarked tussock moth (Table 3.1 1). Prior season’s defoliation was not correlated

with second or fourth instar larval growth or nutritional indices in 1997 (Table 3.8 B).

 

Forest tent caterpillar.

As with gypsy moth and Whitemarked tussock moth, birch was a better host for
forest tent caterpillar than was maple. In 1997. digestive efficiency was 52% lower and
consumption rate was 28% higher on maple relative to birch (Table 3.12. Figure 3.15).
Increases in consumption were not adequate to compensate for poor host quality. and
growth of fifth instar forest tent caterpillar was 35% slower on foliage of maple than
birch (Table 3.12, Figure 3.15).

Light and fertilization treatments did not significantly influence the growth or
nutritional indices of fifth instar forest tent caterpillar (Table 3.12). However. percent
defoliation was positively correlated with digestive efficiency (ECD) of this species
(Table 3.8 C). Consumption and relative growth rate were not correlated with percent

defoliation of either host.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A
0.12
.7" 0.10 . ’ : ° ’
.7." 0.08 ’ a o o
O)
a 32:
g, 0.02 r = 0.39
E 0'00 p = 0.011
0 .
a: 0 1 2 3 4
tannins (% dry weight)
B
0.12
F." 0.10 ’ ° 0 ‘0
'5 o
.7 0.08 o ’
U) 9 O
E 0.06
E 0'04 r = -0.36
E 0-02 p = 0.026
0 0.00
a: 0 10 20 30
phenolics (% dry leaf)

 

 

 

Figure 3.17. Relationship of maple condensed tannins (A) and total phenolic
concentrations (B) with Whitemarked tussock moth relative growth rate
measured over the total larval period in 1996 (RGR).

Insect Performance in Relation to F oliar Traits.
Gypsy moth.

There was no significant correlation between foliar nitrogen and gypsy moth
performance in 1996, but the quality of birch foliage was improved as total phenolic and
condensed tannin concentrations increased. In 1996. efficiency of conversion of digested
birch foliage to biomass (ECD) was positively correlated with total phenolic and
condensed tannin content of foliage (Table 3.8 A). In maple, gypsy moth pupal mass was
positively correlated with specific leaf mass in 1996 (Table 3.8 A).

In 1997, second instar relative growth rates were negatively correlated with foliar
condensed tannin content of maple, but not birch (Table 3.8 A). Fourth instar growth rate
was positively correlated with foliar nitrogen content in both birch and maple (Table 3.8
A, Figure 3.16), and negatively correlated with UN ratios in birch foliage (Table 3.83 A).
Digestive efficiencies (ECD) were negatively correlated with specific leaf mass of birch

(Table 3.8 A).

Whitemarked Tussog Moth.

In 1996, there were no significant correlations between birch foliar traits and
Whitemarked tussock moth performance. However, in 1997. as foliar nitrogen content
increased. the quality of birch to Whitemarked tussock moth also increased as indicated
by positive correlations between foliar nitrogen and efficiency of conversion ofdigested
food to biomass and relative growth rate of fourth instars (Table 3.8 B).

Correspondingly, C/N ratios in birch were negatively correlated with both fourth instar

relative growth rate and efficiency of conversion of digested food to biomass (Table 3.8
B).

In maple in 1996, larval growth as measured from egg hatch to pupation was
positively correlated with condensed tannin levels (Table 3.8 B, Figure 3.17) and specific
leaf mass, and negatively correlated with total phenolic levels of foliage (Table 3.8 B.
Figure 3.17). Duration of the larval period was positively correlated with both condensed
tannin and total phenolic levels of maple (Table 3.8 B). There were no significant

correlations between maple foliar traits and Whitemarked tussock moth performance in

1997.

Forest Tent Caterpillar.

In birch, neither nitrogen content nor concentrations of foliar phenolics were
correlated with forest tent caterpillar performance; however, larval consumption of birch
foliage was positively correlated with foliar C/N ratios in 1997 (Table 3.8 C). In maple.
increases in foliar nitrogen improved host quality; efficiency of conversion of digested
foliage to biomass was positively correlated with foliar nitrogen content (Table 3.8 C).
Neither fifth instar growth nor approximate digestibilities were correlated with foliar

traits.

DISCUSSION

Light Effects on Host Quality.

The effects of shade on host quality were dependent upon host species. insect
species, and other environmental conditions. Shade was important in influencing the
effect of other environmental stresses on host quality (see below). Because shading
reduced levels of total phenolics and condensed tannins in hosts, it was predicted to
improve the quality of both birch and maple foliage for folivores in 1996. However,
predictions for birch were based solely on reductions of condensed tannins in the shade.
Considering that condensed tannins were not correlated with insect performance in this
study (see Relative Host Quality of Paper Birch and Sugar Maple. below), the prediction
is refuted. Even so, shade did in some cases have an effect on host quality. Shading
reduced digestive efficiencies of gypsy moth fed birch in 1996, and decreased female
pupal masses of gypsy moth reared on both birch and maple. In Whitemarked tussock

moth. growth was reduced and duration of larval period lengthened by shade in 1996.

In agreement with predictions based upon treatment effects on foliage chemistry
(Tables 3.1 and 3.2), there was no effect of prior shading of either host species on insect
performance in 1997.

It is important to note that in the present study insects were fed detached foliage
under uniform laboratory conditions such that the effects of shade on insects are due only
to effects of shade on foliage quality. and do not include direct physiological and/ or

behavioral effects of light on the insects. In natural environments these direct and

 

indirect effects would be operating simultaneously. Light may affect herbivores strongly
through direct effects on behavior rather than indirect effects mediated by the host plant.
Both forest tent caterpillar and early instar gypsy moth are strongly photopositive and
feed predominately in areas of the canopy with maximum light (Dylan Parry. personal
communication). This study provides evidence that low-light can negatively affect insect
performance through changes in host-quality, but these effects were not overwhelmingly
strong or consistent. More severe levels of shading can be expected to lead to more
extreme changes in foliar traits. and therefore potentially in insect performance. The
mechanisms of observed effects remain unclear, but lack of significant correlations
indicate that they do not appear to be linked to nitrogen, carbon, condensed tannins, total
phenolics, or specific mass of leaves. Some foliar traits other than those measured.
perhaps water content, alkaloids or extent of lignification. may be responsible for the
indirect effects of shade observed on gypsy moth and Whitemarked tussock moth.
Carbon/ nutrient balance hypothesis predictions for nitrogen-based compounds
(increasing in foliage of shaded trees) are consistent with observed reductions of insect

performance in shade.

Nutrient Effects on Host Quality.

Host and herbivore responses to fertilization were species-specific and were
affected by light environment and defoliation. Based on effects on foliar nitrogen.
fertilization was predicted to have no effect in 1996 and to correspond to improved insect
performance in maple but not birch in 1997 (Tables 3.1 and 3.2). However. fertilization

decreased growth of gypsy moth in shaded treatments in 1996. also consistent with

predictions of the carbon/ nutrient balance hypothesis for nitrogen-based defenses. In full
sun. though, fertilization increased gypsy moth performance (survival. digestive
efficiencies) in 1996. In 1997, fertilization increased second instar gypsy moth growth on
maple. Growth of second instars was greatest on fertilized, non-defoliated hosts.
indicating that defoliation may influence host quality by depleting foliar nutrients (see
below), consistent with carbon/ nutrient balance hypothesis predictions for delayed
induced resistance. The effect of fertilization was greatest in shaded treatments, further
suggesting the potential for environmental factors to interact.

Because second but not fourth instar gypsy moth larvae were affected by
fertilization in 1997, I cannot say for certain that fertilization has a biologically
significant effect on this insect species. As discussed above, second instar gypsy moth
would not naturally feed on foliage of the age used in the 1997 study.

The effects of fertilization on Whitemarked tussock moth were also dependent
upon light environment. Fertilization increased tussock moth performance in 1996 by
increasing growth in sun treatments and increasing survival on shaded maple.
Fertilization effects on foliage nitrogen content were also dependent on the shade
treatment (Chapter 2). Fertilization increased foliar nitrogen in birch in full sun but not in
shade (Chapter 2). Correspondingly, fertilization increased Whitemarked tussock moth
performance in 1996 as reflected in increased relative growth rates. Although
correlations between Whitemarked tussock moth growth and birch nitrogen content were

not significant in 1996. there was a positive correlation between these traits in 1997.

I36

 

As with shade. fertilization did not affect forest tent caterpillar performance in
1997. Fifth instars may be impervious to subtle changes in foliar traits that would

influence performance of younger instars.

Defoliation Effects on Host Quality.

As with shade and fertilization. effects of forest tent caterpillar defoliation on host
quality were dependent upon host species. insect species. and interactions with other
treatments. Because maple received such low levels of defoliation (8 i 3% in the
defoliated treatment), any observed effects are likely not representative of those that
would be observed during natural outbreaks when trees are severely defoliated.

Defoliation generally increased the quality of birch and maple to gypsy moth in
the season of defoliation as indicated by improved digestive efficiencies. However. '
defoliation reduced growth of both males and females over the total larval period. and
increased larval duration of females. These observations cannot be attributed to changes
in foliar nitrogen. carbon, C/N ratios, condensed tannins, total phenolics, or specific leaf
mass as there were no defoliation-induced effects on these traits (Chapter 2. Tables 3.1
and 3.2). Ruohomaki et al. (1996). examining defoliation-induced resistance of mountain
birch to a geometrid larvae, were also unable to attribute insect response to defoliation to
corresponding changes in foliar defensive compormds. They examined resistance in three
groups of trees: control. shaded. or fertilized. Insect performance was the same in all
three groups. although shaded trees and fertilized trees had lower levels of foliar

condensed tannins and total phenolics than controls.

 

Defoliation by forest tent caterpillar did induce resistance of birch to Whitemarked
tussock moth. Percent defoliation of birch early in the season was correlated with
decreased growth and increased larval duration of Whitemarked tussock moth feeding
later in the season of defoliation. This adds to the growing body of literature supporting
the observation that early season defoliation can affect the performance of species feeding
later in the same season (Hunter 1987. Hanhimaki 1989. Dankert et al. 1997. Denno et al.
1995, Wold and Marquis 1997).

Effects of defoliation are important to consider when attempting to predict
susceptibility of forest stands to insect outbreaks. Many researchers have hypothesized
that stressed plants are more susceptible to herbivores than healthy plants (reviewed in
Waring and Cobb 1989, Koricheva et al. 1998). The results of the present study indicated
that. in general, defoliation-stressed birch are lower quality hosts for gypsy moth and
Whitemarked tussock moth than non-defoliated birch, but only when trees are shaded.
When growing in full sun. as they likely would be in nature, birch quality increased with
defoliation. Gypsy moth had increased digestive efficiency and decreased consumption
rates on defoliated birch in the sun and Whitemarked tussock moth were not effected by
defoliation of birch grown in the sun. Therefore, defoliated birch may be more
susceptible to attack by gypsy moth and Whitemarked tussock moth during outbreaks of
forest tent caterpillar. This study addresses only indirect effects of defoliation. mediated
through changes in host quality. Density dependent effects of defoliation on behavior
might substantially alter the performance of gypsy moth and Whitemarked tussock moth

on defoliated birch in natural settings.

Neither gypsy moth nor Whitemarked tussock moth were affected by prior
season’s defoliation in 1997. indicating no evidence for delayed induced resistance at the
levels of defoliation achieved in this experiment. Moreover, forest tent caterpillar
performance in 1997 was not correlated with prior season’s defoliation, indicating little
potential for self-regulation of this insect population through delayed density dependent
effects of defoliation on host quality (Haukioja 1980), at least at defoliation levels
observed here. Dankert et al. (1997) did observe delayed induced resistance in another
group of birch trees belonging to the same cohort used in this experiment when

defoliation levels were more severe (80-90%).

Relative host quality of paper birch and sugar maple.

Sugar maple was a lower quality host for gypsy moth, Whitemarked tussock moth.
and forest tent caterpillar than was paper birch. These findings are in agreement with
previous observations that slow-growing, shade-tolerant plants such as sugar maple are
more resistant to herbivory than fast-growing. shade-intolerant species such as paper
birch (Coley 1983, Dudt and Shure 1994). This is consistent with the hypothesis that
slow-growing species. limited in their ability to replace lost tissues. have higher levels of

constitutive defenses (Coley et al. 1985).

Total phenolics and condensed tannins were not consistently implicated as
determinants ofhost quality. Maple did have higher levels oftotal foliar phenolics (but
lower levels ofcondensed tannins) than birch (Chapter 2). which were negatively

correlated with growth rate and positively correlated with larval development time for

 

Whitemarked tussock moth in 1996 but not 1997. Conversely, condensed tannin
concentrations of maple, one component of total phenolic measures, were positively
correlated with Whitemarked tussock moth relative growth rates measured over the total
larval duration in 1996. suggesting that maple condensed tannins do not play a defensive
role against this species. Prior research has indicated that Whitemarked tussock moth is
not affected by hydrolyzable tannins in the diet (Karowe 1989) due to the protective role

of the peritrophic membrane of the midgut (Barbehenn and Martin 1992).

The negative correlation between Whitemarked tussock moth growth rate and total

phenolics, and the positive correlation with maple condensed tannins observed in this

 

study generally agrees with the idea that Whitemarked tussock moth is a tannin tolerant
species. However other compounds active in the total phenolic assay are apparently
having a detrimental effect on Whitemarked tussock moth. If one such compound is a
hydrolyzable tannin, it must not be any of those present in the tannic acid mixture used in
the diet studies of Karowe or Barbehenn and Martin. Alternatively. it may be a
hydrolyzable tannin present in both diet studies and in the maple foliage. but may remain
inert unless in the presence of some foliar enzymes or other compounds found in maple
leaves but not in artificial diet. It is also possible that the detrimental compound is not a
phenolic at all. Other defensive compounds (e. g. alkaloids) or foliar traits (e. g. water
content) may actually be the causal agents of maple resistance and may covary with total
phenolic content. There was also no correlation between birch condensed tannins or total
phenolics and larval growth rates. providing further evidence that Whitemarked tussock

moth is a tannin-tolerant species.

140

Gypsy moth also performed worse on maple than on birch. However. there is no
indication that condensed tannin or total phenolic concentrations are responsible for this
variation in host quality. In fact. gypsy moth digestive efficiency was positively
correlated with total phenolic concentrations. This is not altogether surprising. for
although gypsy moth can exhibit reduced growth on high tannin diets. it is not
uncommon to find a positive relationship between phenolic levels and larval feeding.
larval mass. and pupal masses (Kleiner and Montgomery 1994). at least in laboratory
studies. In the field. there is a limit to this tolerance. as strong negative correlations

between poplar condensed tannin levels and female pupal masses have been observed

 

(Dylan Parry, unpublished data). Similarly, Rossiter et al. (1988) found defoliation-

in

induced increases in foliar phenolics of natural red oak populations to be negatively

correlated to gypsy moth pupal mass. fecundity, and egg mass.

Forest tent caterpillar is a tannin-sensitive species (Karowe 1989, Barbehenn and
Martin 1994). However, I found no relationship between condensed tannins or total
phenolics. The reduced performance on maple relative to birch observed here is contrary
to findings of Dankert (1995) who found lower survival and growth on paper birch
relative to sugar maple in a study using trees from the same cohort as those used in this
experiment. These conflicting findings are difficult to explain as insects were reared
under identical environmental conditions. However. Dankert’s population of tent
caterpillars were collected in New Foundland. while those in the present study were
collected in Ontario. While sugar maple grows in New Foundland. it does not grow in

the region of Ontario where the forest tent caterpillar eggs were collected for this study.

141

It is possible that the relatively higher resistance of maple to the forest tent caterpillar
used in this study reflect the lack of an evolutionary history of this insect population with

maple.

Maple resistance to gypsy moth and forest tent caterpillar also may be linked to
lower levels ofnitrogen in maple than in birch foliage (Chapter 2). In 1997. growth rate
of fourth instar gypsy moth on birch and maple was positively correlated with foliar

nitrogen content. as was growth of fifth instar forest tent caterpillar on maple.

Compensatory Consumption.

Although relative growth rates of gypsy moth measured from egg hatch to
pupation were reduced on maple in both 1996 and 1997, in 1996 fourth instars were able
to compensate for decreased host quality by increasing consumption. Failure of fourth
instars to exhibit compensatory feeding in 1997 may have resulted from the fact that in
1996 larvae were reared from egg hatch on experimental trees and in 1997 they were field
collected from poplars, trees dominated by secondary metabolites (phenolic glycosides)
not found in birch and maple. Dietary switching can reduce the performance of some
herbivores on novel hosts (Joseph and Kelsey 1994). The timing of a host switch in the
course of the insects lifecycle is important for determining the effect ofthe switch on the
insect (Stoyenoff et al. 1994). Early instars. although more sensitive to host quality than
later instars (Table 3.8) may have been able to adapt to foliage quality in 1996 so that by

the fourth instar they had made behavioral (e.g.. consumption) and /or physiological (e.g.,

142

induction of digestive enzymes) adjustments enabling them to maintain relatively high
growth rates even on non-preferred hosts. Alternatively, it may be that in 1997. fourth
instars switched from poplar to maple were not sufficiently able to adjust their
consumption rates within the 2 d period of growth measurement to compensate for

decreased host quality.

Second instar gypsy moth and Whitemarked tussock moth were more sensitive to

 

I?”
foliage quality than were conspecific fourth instars in 1997 (as indicated by higher 1
numbers of significant treatment effects on second relative to fourth instars). These
results were expected as 1) second instars naturally feed on younger foliage than do i
r.

fourth instars, and young foliage is expected to have higher water and nutrient contents
relative to older foliage (Scriber and Slansky 1981), and 2) larvae experimentally tested
on maple underwent a host switch. While it is not unusual for lepidopteran larvae to
switch hosts in natural settings, this switch does not occur until larvae have completed
early development. In the case of gypsy moth, dispersal from the original host occurs
after larvae reach the fourth stadium (Lance and Barbosa 1979, Leonard 1981. as cited in
Stoyenoff et al. 1994). Therefore, younger larvae may be more affected by a change in
host than older instars. These findings emphasize the importance of mimicking natural

conditions as closely as possible in laboratory studies.

Second instar Whitemarked tussock moth were less sensitive to foliage quality
than were second instar gypsy moth in 1997. even though both species exhibited reduced
growth on maple. This is indicated by a higher number ofsignificant treatment effects on

second instar gypsy moth relative to second instar Whitemarked tussock moth. In 1997.

all feeding studies were conducted in mid-June. This corresponded with natural host-
herbivore phenology of fourth instar gypsy moth and fifth instar forest tent caterpillar.
and with the second instars of the second Whitemarked tussock moth generation.
However. second instar gypsy moth would not naturally be feeding on foliage ofsuch an
advanced phenological stage. This phenological asynchronicity may explain the greater
sensitivity of second instar gypsy moth to foliage quality. Thus. the results should be

interpreted with some caution.

Conclusions.

Host species was the most important factor affecting the performance of all three
folivores studied. Shade, fertilization, and defoliation also were influential in affecting
insect performance. However, effects of these factors never influenced all insect species
equally. and effects were often interactive. The mechanistic basis of the effects of shade.
fertilization. and defoliation were not made clear by my results. Previous work has
demonstrated that shade and fertilization can lower C/N ratios and carbon-based
defensive compounds in foliage. thereby improving insect performance (Bryant et al.
1983). However. neither condensed tannin nor total phenolic concentrations were good
predictors of insect performance in this study. Presence of nitrogen-based defensive
compounds, (which occur in maple but not birch), may be driving observed reductions of
insect performance in shaded trees, as these compounds are predicted to be increased by
shade and fertilization. Nutritional value of foliage also affects host quality (Mattson
I980, Scriber and Slansky I981, Tuomi et al. 1984). While there was limited evidence of

an association between foliar nitrogen content and insect performance in some cases. the

144

majority ofeffects oftreatments on insect performance could not be explained by changes
in foliar nitrogen levels. This does not negate the importance of foliar defensive
compounds or nitrogen content on insect performance; rather it implies that effects of
light, nutrient, and defoliation treatments in this study are not acting by a simple
mechanism which influences foliar nitrogen or carbon-based defensive compounds. The
interactive nature of environmental stresses on host quality and differential responses of
various insect species to host foliage complicate the process of devising models with
strong power to predict forest susceptibility to pest species.

Understanding of mechanisms of forest susceptibility will be aided by future
research examining foliar and insect responses across a range of nutrient, light. and
defoliation levels. Such experimental designs will enable regression analysis. which can
better expose the relative importance of individual foliar traits affected by environmental

stress that in turn influence herbivore performance.

145

CHAPTER 4:

CONCLUSIONS

Light availability, nutrients, and defoliation all influenced patterns of resource

acquisition, resource allocation and herbivore resistance in paper birch and sugar maple.

 

F“
Tree responses to these stresses were influenced by species-specific characteristics. and
therefore were not often the same in both species. Moreover, the effects of one stress on
tree responses were often dependent upon simultaneous exposure to other stresses.
Herbivore responses to environmentally stressed host-foliage were also species specific. L

The mechanistic basis of insect responses to host quality were not clear, but did not
appear to be strictly a function of foliar nutrient content or levels of carbon based
defenses.

Consistent with life history theory, paper birch was generally faster-growing. had
higher light-saturated photosynthetic rates, and was more responsive to environmental
variation than was sugar maple. Birch assimilated carbon more rapidly at all light levels.
even in shaded environments. Therefore, although maple is considered to be a more
shade tolerant species than birch. this tolerance could not be attributed to a superior
photosynthetic capacity of maple over birch in low light. Furthermore. the species had
equivalent respiration rates implying that their basic requirements for photosynthate were
similar. at least at the leaf level. However, the shade tolerance of maple was evidenced
by its photosynthetic rates being less responsive to decreased light levels than birch‘s

photosynthetic rates.

146

Corresponding to its higher photosynthetic rates and adaptation to high light
environments, birch had higher foliar nitrogen content and greater specific leaf mass than
maple. Total phenolic concentrations were also greater in birch, although the
compositional structure of those phenolics differed between the species as indicated by
maple having higher concentrations of condensed tannins than birch.

In agreement with Life History Theory, paper birch was a superior host for gypsy
moth, Whitemarked tussock moth, and forest tent caterpillar than was sugar maple.
Although levels of foliar nutrients and defensive compounds did influence the
performance of folivores, each species of insect responded differently to these traits. In
maple, total phenolics were correlated with reduced host quality for Whitemarked tussock
moth in 1996 but not 1997. Conversely, condensed tannin concentrations of maple were
positively correlated with Whitemarked tussock moth performance, indicating that
condensed tannins of maple do not play a defensive role against this species. Gypsy
moth growth on birch and maple in 1997 was positively correlated to foliar nitrogen

content, as was forest tent caterpillar growth on maple.

Shading had a large impact on tree growth and herbivore resistance. Trees did not
appear to acclimate photosynthetically to two years of shade, as development under
reduced light conditions did not affect light-saturated photosynthetic rates in either
species. However. the photosynthetic rate of birch in low light was significantly less than
that in full sun. Trees did respond to temporal shifts in light availability in 1996 with
decreased specific leaf mass. Decreased specific leaf mass may result from increased leaf
size or from compositional differences in sun versus shade-grown foliage. By spreading

out photosynthetic apparati over a larger area leaves may more efficiently harvest the

147

limited incident radiation in shaded environments. However, in this study, increased leaf
areas were evidenced only in immature foliage collected in August from non-defoliated
birch trees. Shade increased foliar nitrogen per unit mass in maple. suggesting that
decreases in other foliar components may have occurred. The shade treatment was not
implemented in 1997; therefore, previously shaded trees experienced ambient light
conditions mimicking gap formation. Foliage on trees that were previously shaded had a
specific leaf mass equivalent to that of foliage on trees which had developed in full-sun.
This indicates that birch and maple trees may quickly adjust to gap formation.

Both birch and maple exhibited reduced growth in shade. Height growth of both
species was not affected by light availability during the first year of shading (1995).
while diameter growth was decreased by shade. In the following year (1996). height
growth was decreased by shade. while there was no influence of light on diameter
growth.

Although shading had no strong effect on foliar carbon or C/N ratios. it reduced
carbon-based defenses in both birch and maple in 1996 as predicted by the carbon/
nutrient balance hypothesis. However. birch and maple carbon-based secondary
metabolites did not respond to shade in the same way. Total phenolics of birch remained
the same. but the composition of these phenolics changed, as condensed tannin content
was reduced. In maple. shade also induced compositional changes in phenolics but in
this case shade was associated with a reduction in total phenols and maintenance of
condensed tannin levels.

The light treatment did not generally have a strong effect on insect performance.

but was important in influencing the effect ofother environmental stresses on host

I48

quality. When shade did affect insect performance. it reduced host quality. Shading
reduced digestive efficiencies of gypsy moth fed birch in 1996. and decreased female
pupal masses of gypsy moth reared on both birch and maple. The growth of whiternarked
tussock moth was reduced and duration of larval period lengthened by shade in 1996.
However, there was no effect of prior shading of either host species on insect
performance in 1997. This study provides evidence that low-light can negatively affect
insect performance through changes in host-quality, but that these effects are not strong
or consistent. More severe levels of shading can be expected to lead to more extreme
changes in foliar traits and therefore potentially in insect performance. The mechanisms
of observed effects remain unclear, but they are linked in some cases to nitrogen. carbon.
condensed tannins, total phenolics, and specific leaf mass of leaves.

Effects of fertilization on tree physiology and herbivore resistance were not as
strong as expected. Fertilization did not affect the photosynthetic rates of birch or maple.
Fertilization also had no effect on tree growth rates with the exception that it did reduce
the height growth of both species in 1995. These findings were unexpected, and indicate
that nitrogen was not limiting in non-fertilized treatments. Reductions in growth in 1995
may indicate that a nitrogen surplus was reached in fertilized cells such that some other
nutrient became limiting. It would be necessary to analyse soil samples to determine if
this was indeed the case.

Fertilization decreased levels of total phenolics in immature birch foliage
harvested in August of 1996. This response was in agreement with predictions of the

carbon/ nutrient balance hypothesis. although it was not accompanied by significant

I49

increases in foliar C/N ratios. In contrast to the August response. fertilization had no
effect on C-based defenses of either species when sampled in June.

Although fertility had no effect on foliar defenses of birch or maple in June of
1996. herbivores did respond to fertilization. These responses were host and insect
species-specific and were affected by light environment and defoliation. Fertilization
decreased the growth of gypsy moth on shaded trees in 1996. but increased growth on
trees in full sun. Fertilization also increased second instar gypsy moth growth on maple
in 1997. The effect of fertilization was greatest in shaded treatments, further indicating
the potential for environmental factors to interact. Fertilization increased Whitemarked
tussock moth performance in 1996 by increasing growth in sun treatments and increasing
survival on shaded maple. Although correlations between Whitemarked tussock moth
growth and birch nitrogen content were not significant in 1996, there was a positive
correlation between these traits in 1997. These results suggest that increased soil nutrient
levels do not necessarily need to stimulate plant growth in order to influence herbivore
resistance.

As with shade and fertilization, effects of defoliation on foliar quality were
dependent upon host species. insect species, and the other treatments. Because maple
received such low levels of defoliation. any observed effects are not likely to be
representative of those that would be observed during natural outbreaks.

Effects ofdefoliation on birch physiology mimicked those of nutrient deficiency
in that fertilization often mitigated the effects of defoliation. Defoliation decreased foliar
nitrogen levels in non-fertilized birch. Correspondingly. defoliation also increased C/N

ratios, but this was true for both fertilized and non-fertilized trees. Defoliation can

150

 

increase the UN ratio because the majority ofthe plant‘s nitrogen supply is contained in
the canopy and is lost when the tree is defoliated. However, carbon is available from
stored reserves in other plant parts. Defoliation reduced height growth in both species;
however. there was no evidence of fertilization mitigating this effect.

The effects of forest tent caterpillar defoliation on the quality of birch as a host for
Whitemarked tussock moth and gypsy moth were environmentally dependent. Although
foliar condensed tannin levels were lower in all defoliated birch trees in June of 1996.
defoliated trees were lower quality hosts for gypsy moth and Whitemarked tussock moth
than non-defoliated birch, but only when trees were shaded. There were no delayed
effects of defoliation on birch resistance to gypsy moth or Whitemarked tussock moth.
indicating that forest tent caterpillar defoliation at the observed levels will not mediate
interspecific competition among these species. Moreover, forest tent caterpillar
performance the year following defoliation was not affected by the defoliation treatment.

indicating no evidence for delayed density dependent regulation of this insect population.

151

APPENDICES

152

APPENDIX 1

Record of Deposition of Voucher Specimens

153

 

APPENDIX 1
Record of Deposition of Voucher Specimens*

The specimens listed on the following sheet(s) have been deposited in the named
museum(s) as samples of those species or other taxa which were used in this research.
Voucher recognition labels bearing the Voucher No. have been attached or included in
fluid-preserved specimens.

Voucher No.: 1998-10

 

Title of thesis or dissertation (or other research projects):

Light. Nutrient Availability, and Defoliation Effects on Resource Acquisition,
Resource Allocation, and Herbivore Resistance of Paper Birch and Sugar Maple

Museum(s) where deposited and abbreviations for table on following sheets:

Entomology Museum, Michigan State University (MSU)

Other Museums:

Investigator’s Name(s) (typed)
Heather L. Govenor

 

 

Date ll/16/l998

* Reference: Yoshimoto, C. M. 1978. Voucher Specimens for Entomology in North
America. Bull. Entomol. Soc. Amer. 24:141-42.

Deposit as follows:
Original: Include as Appendix 1 in ribbon copy of thesis or dissertation.
Copies: Include as Appendix 1 in copies of thesis or dissertation.
Museum(s) files.
Research project files.

This form is available from and the Voucher No. is assigned by the C urator, Michigan
State University Entomology Museum.

154

APPENDIX 1.1

Voucher Specimen Data

155

 

 

 

&\v\l§m~m/a\m\ @5923 28

 

 

 

 

 

Baguio
t. §§
- map 2 .89
D Swan “205:: v 8.98.
Banana—393%: 26.3 ofivozooom ._ .358:
$3.52 £03385.

 

cfiimmma .02 .83>
3.8308: a 3023 3:363 one

 

agégpigfiaafioflaaepm Egan;

 

156

 

 

 

 

8m“ .42. D
355 .35. .95 flaw flea
g m 3%“. 2.85... E .3 E 5.688 2mg mambo
88 E 8
GEES .mflmz .Bm flaw
8’. m B§§§.8§§5§ 43%.»).
.w d 0028on can no»: .8 633:3 :85 .850 .3 3825
t e .m. e «858330.883...—

 

 

 

 

 

 

 

 

 

M
:11
Other

'3 Adults
Adults 0
Pupa
Nym
Larva
Eggs

 

 

LITERATURE CITED

157

 

LITERATURE CITED

Ashton, P. M. S., and G. P. Berlyn. 1992. Leaf adaptations of some .S‘hurea species to
sun and shade. New Phytology 1211587-596.

Ashton, P. M. S. and G. P. Berlyn. 1994. A comparison of leaf physiology and anatomy
of Quercus (section Erythrobalanus-Fagaceae) species in different light environments.
American Journal of Botany 81:589-597.

Ayres. M. P. 1993. Plant defense, herbivory, and climate change. Pages 75-94 in P. M. |
Kareiva. J. G. Kingsolver, and R. B. Huey, editors. Biotic interactions and global i
change. Sinauer Associates. Inc., Sunderland, Massachusettes, USA.

Ayres, M. P., T. P. Clausen, S. F. MacLean, Jr., A. M. Redman, and P. B. Reichardt.
1997. Diversity of structure and antiherbivore activity in condensed tannins. Ecology
78:1696-1712. ~—

 

Baker, F. S. 1949. A revised tolerance table. Journal of Forestry 47:179-181.

Baldwin, 1. T., J. C. Schultz, and D. Ward. 1987. Patterns and sources of leaf tannin
variation in yellow birch (Betula allegheniensis) and sugar maple (Acer .s'accharum).
Journal of Chemical Ecology 13:1069-1078.

Barbehenn, R. V. and M. M. Martin. 1992. The protective role of the peritrophic
membrane in the tannin-tolerant larvae of Orgyia leucostigma (Lepidoptera). Journal of
Insect Physiology 38:973-980.

Barbehenn, R. V. and M. M. Martin. 1994. Tannin sensitivity in larvae of Mulacosoma
disstria (Lepidoptera): roles of the peritrophic envelope and midgut oxidation. Journal of
Chemical Ecology 20: 1 985-2001.

Barbosa, P.. P. Gross, G. J. Provan, D. Y. Pacheco. and F. R. Stermitz. 1990.
Allelochemicals in foliage of unfavored tree hosts of the gypsy moth, Lymantria dispar
L. l. Alkaloids and other components of Lirioa’endron Iulipifera L. (Magnoliaceae). Accr
rubrum L. (Aceraceae). and Cornusflorida L. (Cornaceae). Journal of Chemical Ecology
16:1719-1730.

Bassett, Y. 1991. The spatial distribution of herbivory. mines and galls within an
Australian rain forest tree. Biotropica 23:271-281.

Bazzaz. F. A. 1979. The physiological ecology ofplant succession. Annual Review of
Ecology and Systematics 10:351-371.

158

Bazzaz, F. A., N. R. Chiariello, P. D. Coley, and L. F. Pitelka. 1987. Allocating
resources to reproduction and defense. BioScience 37:58-67.

Beaupre, S. J. and A. E. Dunham. 1995. A comparison of ratio-based and covariance
analyses of a nutritional data set Functional Ecology 92876-880.

Bernays, E. A. 1981. Plant tannins and insect herbivores: an appraisal. Ecological
Entomology 62353-360.

Bryant, J. P., F. S. Chapin, III, and D. R. Klein. 1983. Carbon/ nutrient balance of boreal
plants in relation to vertebrate herbivory. Oikos 40:357-368.

Bryant, J. P., F. S. Chapin III, P. B. Reichardt, and T. P. Clausen. 1985. Adaptation to P}.

resource availability as a determinant of chemical defense strategies in woody plants.
Recent Advances in Phytochemistry 19: 219-237.

Bryant, J. P., F. S. Chapin, III, P. B. Reichardt, and T. P. Clausen. 1987. Response of ,
winter chemical defense in Alaska paper birch and green alder to manipulation of plant I;
carbon/nutrient balance. Oecologia 72:510-514. 4- "

 

Bryant, J. P., R. B. Reichardt, and T. P. Clausen. 1992. Chemically mediated
interactions between woody plants and browsing mammals. Journal of Range
Management 45:18-24.

Bultman, T. L. and S. H. Faeth. 1988. Abundance and mortality of leaf miners on
artificially shaded Emory oak. Ecological Entomology 13:131-142.

Burke, M. K., D. J. Raynal, and M. J. Mitchell. 1991. Soil nitrogen availability
influences seasonal carbon allocation patterns in sugar maple (Acer saccharum).
Canadian Journal of Forest Research 22:447-456.

Canham, C. D., J. S. Denslow, W. J. Platt, J. R. Runkle, T. A. Spies, and P. S. White.
1990. Light regimes beneath closed canopies and tree-fall gaps in temperate and tropical
forests. Canadian Journal of Forest Research 20:620-631.

Chapin, F. S., III. 1991. Integrated responses of plants to stress. BioScience 41:29-36.

Chapin, F. S., A. J. Bloom, C. B. Field, and R. H. Waring. 1987. Plant responses to
multiple environmental factors. BioScience 37:49-57.

Chazdon. R. L. and S. Kaufmann. 1993. Plasticity of leaf anatomy oftwo rain forest
shrubs in relation to photosynthetic light acclimation. Functional Ecology 72385-394.

Cody, M. L. 1966. A general theory of clutch size. Evolution 20:174-184.

159

Coleman, J. S. 1986. Leaf development and leaf stress: increased susceptibility
associated with sink-source transition. Tree Physiology 22289-299.

Coley, P. D. 1983. Herbivory and defensive characteristics of tree species in a lowland
tropical forest. Ecological Monographs 53:209-233.

Coley, P. D. 1993. Gap size and plant defenses. Trends in Ecology and Evolution 8: 1-2.

Coley, P. D. and J. A. Barone. 1996. Herbivory and plant defenses in tropical forests.
Annual Review of Ecology and Systematics 27:305-335.

Coley, P. D., J. P. Bryant, and F. S. Chapin. III. 1985. Resource availability and plant
antiherbivore defense. Science 230:395-399.

Cui, M. and M. M. Caldwell. 1997. Shading reduces exploitation of soil nitrate and
phosphate by A gropyron desertorum and Artemisia tridentata from soils with patchy and
uniform nutrient distributions. Oecologia 1092177-183.

Dale, M. P. and D. R. Causton. 1992. The ecophysiology of Veronica chamaedrys; V.
montana, and V. oflicinalis. 11. Effects of shading on the phenology of biomass
allocations—a field experiment. Journal of Ecology 802505-515.

Dankert, B. A. 1995. Paper birch and sugar maple resistance to insect folivores as
influenced by defoliation, drought, and fertilization. M. S. thesis, Michigan State
University.

Dankert, B. A., D. A. Herms. D. Parry, J. M. Scriber, and L. A. Haas. 1997. Indirect
interspecific competition among folivores mediated by defoliation-induced changes in
host quality. Pages 71-88 in A. Raman, editor. Ecology and evolution of plant-feeding
insects in natural and man-made environments. International Scientific Publications.
New Delhi.

Denno, R. F ., M. S. McClure, and J. R. Ott. 1995. Interspecific interactions in
phytophagous insects: competition reexamined and resurrected. Annual Review of
Entomology 40:297-331.

Daun, J. K., and D. R. DeClerq. 1994. Comparison of combustion and Kjeldahl methods
for determination of nitrogen in oilseeds. Journal of the American Oil Chemistry Society
71:1069-1072.

DeJong, T. M. 1982. Leaf nitrogen content and C02 assimilation capacity in peach.
Journal of the American Society of Horticultural Science 1072955-959.

Denslow, J. S., J. C. Schultz, P. M. Vitousek. and B. R. Strain. 1990. Growth responses
of tropical shrubs to treefall gap environments. Ecology 712165-179.

160

Dudt, J. F. and D. J. Shure. 1994. The influence of light and nutrients on foliar phenolics
and insect herbivory. Ecology 75:86-98.

Ellsworth, D. S. and P. B. Reich. 1992. Leaf mass per area, nitrogen content and
photosynthetic carbon gain in Acer saccharum seedlings in contrasting forest light
environments. Functional Ecology 62423-435.

Ellsworth, D. S. and P. B. Reich. 1993. Canopy structure and vertical patterns of
photosynthesis and related leaf traits in a deciduous forest. Oecologia 96: 169-1 78.

Faeth, S. H. 1992. Do defoliation and subsequent phytochemical responses reduce
future herbivory on oak trees? Journal of Chemical Ecology 18:915-925.

Feeny, P. 1970. Seasonal changes in oak leaf tannins and nutrients as a cause of spring
feeding by winter moth caterpillars. Ecology 51:565-581.

_._.,'

Field, C., J. Merino, and H. A. Mooney. 1983. Compromises between water-use
efficiency and nitrogen-use efficiency in five species of California evergreens. Oecologia
60:84-89.

 

I?
1.

Field, C. and H. A. Mooney. 1986. The photosynthesis-nitrogen relationship in wild
plants. Pages 25-55 in T. Givnish, editor. On the economy of plant form and function.
Cambridge University Press, New York, USA.

Fitzgerald, T. D. 1995. The tent caterpillars. Cornell University Press, London.

Floater, G. J. 1997. Rainfall, nitrogen, and host plant condition: consequences for the
processionary caterpillar, Ochrogaster lung/er. Ecological Entomology 22:247-255.

Fox, L. R. and P. A. Morrow. 1992. Eucalypt responses to fertilization and reduced
herbivory. Oecologia 89:214-222.

Givnish, T. J. 1988. Adaptation to sun and shade: a whole-plant perspective. Australian
Journal of Plant Physiology 15:63-92.

Goodman, R. M.. H. W. Yawney, and C. H. Tubbs. 1990. Acer saccharum. Pages 78-
91 in R. M. Burns and B. H. Honkala, editors. Silvics of North America V.2 hardwoods.
Agriculture Handbook 654. Forest Service. United States Department ongriculturc.
Washington. DC. USA.

Gordon, H. T. 1968. Quantitative aspects of insect nutrition. Zoology 8:131-138.
Gottschalk. K. W. 1994. Shade. leaf growth and crown development of Qucrcus rubru.

Quercus velutina. Przmus serotina and Acer rubrum seedlings. Tree Physiology 14:735-
749.

161

Grisdale, D. G. 1968. A method for reducing incidence ofvirus infection in insect
rearings. Journal of Invertebrate Pathology 10:425.

Groninger, J. W., J. R. Seiler, J. A. Peterson, and R. E. Kreh. 1996. Growth and
photosynthetic responses of four Virginia Piedment tree species to shade. Tree
Physiology 162773-778.

Gulmon, S. L. and C. C. Chu. 1981. The effects of light and nitrogen on photosynthesis,
leaf characteristics, and dry matter allocation in the Chaparral shrub, Diplaucus
aurantiacus. Oecologia 49:207-212.

Hagerman, A. E. 1995. Tannin analysis. ((3 Ann E. Hagerman.

Hanhimaki, S. 1989. Induced resistance in mountain birch: defence against leaf-chewing
insect guild and herbivore competition. Oecologia 81:242-248.

Haukioja, E. 1980. On the role ofplant defenses in the fluctuation of herbivore
populations. Oikos 35 2202-2 1 3.

Haukioja, E. 1982. Inducible defenses of white birch to a geometrid defoliatior, Epirrita
autumnata. Proceedings of the 5th international Symposium on insect-plant relationships.
Wageningen 199-203.

Haukioja, E. 1990. Induction of defenses in trees. Annual Review of Entomology
36:25-42.

Haukioja, E. and S. Neuvonen. 1985. Induced long-term resistance of birch foliage
against defoliators: defensive or incidental? Ecology 66:1303-1308.

Haukioja, E., J. Suomela, S. Neuvonen. 1985. Long-term inducible resistance in birch
foliage: triggering cues and efficacy on a defoliator. Oecologia 65:363-369.

Herms, D. A. and W. J. Mattson. 1992. The dilemma of plants: to grow or defend.
Quarterly Review of Biology 67:283-335.

Hollinger, D. Y. 1996. Optimality and nitrogen allocation in a tree canopy. Tree
Physiology 16:627-634.

Holt, J. S. 1995. Plant responses to light: a potential tool for weed management. Weed
Science 43:474-482.

Houle, G. and G. Simard. 1996. Additive effects of genotype, nutrient availability and

type of tissue damage on the compensatory response of Salix pluni/blia spp. plcmifb/iu to
simulated herbivory. Oecologia 1072373-378.

162

, n“-
-.
'r

 

Hunter, M. D. 1987. Opposing effects of spring defoliation on late season oak
caterpillars. Ecological Entomology 12:373-382.

Hunter, M. D. and J. C. Schultz. 1995. Fertilization mitigates chemical induction and
herbivore responses within damaged oak trees. Ecology 76: 1226-1232.

Johnson, J. D., R. Tognetti, M. Michelozzi, S. Pinzauti, G. Minotta, and M. Borghetti.
1997. Ecophysiological responses of Fagus sylvatica seedlings to changing light
conditions. 11. The interaction of light environment and soil fertility on seedling
physiology. Physiologia Plantarum 101:124-134.

Joseph, G. and R. G. Kelsey. 1994. Acceptability and suitability of Douglas-Fir as a .
secondary host for gypsy moth (Lepidoptera: Lymantriidae). Environmental Entomology 1'1
232396-405.

Karban, R. and I. T. Baldwin. 1997. Induced responses to herbivory. University of
Chicago Press, Chicago, Illinois, USA.

 

Karowe, D. N. 1989. Differential effect of tannic acid on two tree-feeding Lepidoptera:
implications for theories of plant anti-herbivore chemistry. Oecologia 80:507-512.

Kinney, K. K., R. L. Lindroth, L. M. Jung, and E. V. Nordheim. 1997. Effects of CO;
and N03 availability on deciduous trees: phytochemistry and insect performance.
Ecology 78:215-230.

Kitajima, K. 1994. Relative importance of photosynthetic traits and allocation patterns as
correlates of seedling shade tolerance of 13 tropical trees. Oecologia 98:419-428.

Kleiner, K. W. and M. E. Montgomery. 1994. Forest stand susceptibility to the gypsy
moth (Lepidoptera: Lymantriidae): species and site effects on foliage quality to larvae.
Environmental Entomology 232699-71 1.

Klepzig, K. D., E. L. Kruger, E. B. Smalley, and K. F. Raffa. 1995. Effects of biotic and
abiotic stress on induced accumulation of terpenes and phenolics in red pines inoculated

with bark beetle-vectored fungus. Journal of Chemical Ecology 21:601-626.

Kobe. R. K. 1997. Carbohydrate allocation to storage as a basis of interspecific variation
in sapling survivorship and growth. Oikos 80:226-233.

Kobe. R. K., S. W. Pacala. J. A. Silander. C. D. Canham. 1995. Juvenile tree
survivorship as a component of shade tolerance. Ecological Applications 52517-532.

Kolb. T. E. and D. A. J. Teulon. 1991. Relationship between sugar maple budburst
phenology and pear thrips damage. Canadian Journal of Forest Research 21 : 1 043- 1048.

Koricheva, J ., S. Larsson, and E. Haukioja. 1998. Insect performance on experimentally
stressed woody plants: a meta-analysis. Annual Review of Entomology 43: 195-216.

Kubiske, M. E. and K. S. Pregitzer. 1996. Effects of elevated CO; and light availability
on the photosynthetic light response of trees of contrasting shade tolerance. Tree
Physiology 162351-358.

Kyto, M., P. Niemela, and S. Larsson. 1996. Insects on trees: population and individual
response to fertilzation. Oikos 75: 148-159.

Lance P. and P. Barbosa. 1979. Dispersal of larval Lepidoptera with special reference to
forest defoliators. Biologist 61290-1 10.

Lambers, H. and H. Poorter. 1992. Inherent variation in growth rates between higher
plants: a search for physiological causes and ecological consequences. Advances in
Ecological Research 23:187-261.

Larcher, W. 1995. Physiological plant ecology 3rd edition. Springer, New York, USA.

Lawler, I. R., W. J. Foley, 1. E. Woodrow, and S. J. Cork. 1997. The effects of elevated
C02 atmospheres on the nutritional quality of Eucalyptus foliage and its interaction with
soil nutrient and light availability. Oecologia 109259-68.

Leonard, D. E. 1981. Bioecology of the gypsy moth. Pages 8-29 in C. C. Doane and M.
M. McManus, editors. The gypsy moth: research toward integrated pest management.
Forest Service technical bulletin 1584, USDA, Washington, DC, USA.

Lincoln D. E. and H. Mooney. 1984. Herbivory of Diplacus aurantiacus shrubs in sun
and shade. Oecologia, 642173-176.

Linder, S. and D. A. Rook. 1984. Effects of mineral nutrition on carbon dioxide
exchange and partitioning of carbon in trees. Pages 21 1-236 in G. D. Bowen and E. K. S.
Nambiar, editors. Nutrition of plantation forests. Academic Press, New York, New
York, USA.

Loehle, C. 1988. Tree life history strategies: the role of defenses. Canadian Journal of
Forest Research 182209-222.

Loomis, W. E. 1932. Growth-differentiation balance vs. carbohydrate-nitrogen ratio.
Proceedings of the American Society of Horticultural Science 29:240-245.

Lorio, P. L.. Jr. 1986. Growth-differentiation balance: a basis for understanding
southern pine beetle-tree interactions. Forest Ecology and Management 14:259-273.

164

Lovett, G. M. and P. Tobiessen. 1993. Carbon and nitrogen assimilation in red oaks
(Quercus rubra L.) subject to defoliation and nitrogen stress. Tree Physiology 12:259-
269.

Lowman, M. D. 1992. Leaf growth dynamics and herbivory in five species of Australian
rain-forest canopy trees. Journal of Ecology 802433-447.

Lusk, C. H., O. Contreras. and J. Figueroa. 1997. Growth, biomass allocation and plant
nitrogen concentration in Chilean temperate rainforest tree seedlings: effects of nutrient
availability. Oecologia 109249-58.

Mattson, W. J. 1980. Herbivory in relation to plant nitrogen content. Annual Review of
Ecology and Systematics 11:119-161.

Mattson, W. J ., D. A. Herms, J. A. Witter, and D. C. Allen. 1991. Woody plant grazing
systems: North American outbreak folivores and their host plants. Pages 53-84 in Y. N.
Baranchikov, W. J. Mattson, F. P. Hain, and T. L. Payne, editors. Forest insect guilds:
patterns of interaction with host trees. General technical report NE-153. USDA
Northeast Forest Experiment Station, Randor, Pennsylvania, USA.

May, J. D. and K. T. Killingbeck. 1995. Effects of herbivore-induced nutrient stress on

correlates of fitness and on nutrient resorption in scrub oak (Quercus ilicifolia).
Canadian Journal of Forest Research 2521858-1864.

McDonald, A. J. S., T. Lohammar, and A. Ericsson. 1986. Uptake of carbon and
nitrogen at decreased nutrient availability in small birch (Betula pendula Roth.) plants.
Tree Physiology 2261-71.

McKee, K. L. 1994. Interspecific variation in growth, biomass partitioning, and
defensive characteristics of neotropical mangrove seedlings: response to light and
nutrient availability. American Journal of Botany 822299-307.

Michigan Department of Agriculture, Animal and Plant Health Inspection Service.
USDA, and Forest Service, USDA. 1984. State of Michigan 1984 gypsy moth
management program.

Mole. S., J. A. M. Ross, and P. G. Waterman. 1988. Light-induced variation in phenolic
levels in foliage of rain-forest plants. 1. chemical changes. Journal ofChemical Ecology
1421-21.

Mooney. H. A. 1972. The carbon balance ofplants. Annual Review of Ecology and
Systematics 32315-346.

Mooney, H. A., C. Field, S. L. Gulmon, and F. A. Bazzaz. 1981. Photosynthetic

capacity in relation to leaf position in desert versus old-field annuals. Oecologia 50:109-
1 12.

165

 

Mulkey, S. S., S. J. Wright, and A. P. Smith. 1993. Comparative physiology and
demography of three neotropical forest shrubs: alternative shade-adaptive character
syndromes. Oecologia 96:526-536.

Muzika, R. 1993. Terpenes and phenolics in response to nitrogen fertilization: a test of
the carbon/nutrient balance hypothesis. Chemoecology 423-7.

Neuvonen, S. and E. Haukioja. 1984. Low nutritive quality as defense against
herbivores: induced responses in birch. Oecologia 63:71-74.

Neuvonen, S., E. Haukioja. and A. Molarius. 1987. Delayed inducible resistance against
a leaf-chewing insect in four deciduous tree species. Oecologia 74: 363-369.

Nichols-Orians, C. M. 1991. The effects of light on foliar chemistry, growth and
susceptibility of seedlings ofa canopy tree to an attine ant. Oecologia 86:552-560.

Niesenbaum, R. A. 1992. The effects of light environment on herbivory and growth in
the dioecious shrub Lindera benzoin (Lauraceae). American Midland Naturalist 1282270-
275.

Niinemets, U. and K. Kull. 1994. Leaf weight per area and leaf size of 85 Estonian ’
woody species in relation to shade tolerance and light availability. Forest Ecology and
Management 7021-10.

Ovaska. J ., M. Walls, and P. Mutikainen. 1992. Changes in leaf gas exchange properties
of cloned Betula pendula saplings after partial defoliation. Journal of Experimental
Botany 43:1301-1307.

Packard, G. C. and T. J. Boardman. 1988. The misuse of ratios, indices, and percentages
in ecophysiological research. Physiological Zoology 6121-9.

Palo, R. T. 1984. Distribution of birch (Betula spp.), willow (Salix spp.), and poplar
(Populus spp.) secondary metabolites and their potential role as chemical defense against
herbivores. Journal of Chemical Ecology 10:499-520.

Parker, J. and D. R. Houston. 1971. Effects of repeated defoliation on root and root
collar extractives of sugar maple trees. Forest Science 17:91-95.

Poorter, H. 1989. Interspecific variation in relative growth rate: on ecological causes and
physiological consequences. Pages 45-68 in H. Lambers. editor. Causes and
consequences of variation in growth rate and productivity of higher plants. Academic
Publishing, The Hague, The Netherlands.

Potter, D. A. 1992. Abundance and mortality of a specialist leafminer in response to
experimental shading and fertilization of American holly. Oecologia 91:14-22.

166

 

Raubenheimer, D. 1995. Problems with ratio analysis in nutritional studies. Functional
Ecology 9:21-29.

Raubenheimer, D. and S. J. Simpson. 1992. Analysis of covariance: an alternative to
nutritional indices. Entomological Experiments and Applications 62:221-231.

Reich, P. B., M. B. Walters, and D. S. Ellsworth. 1991. Leaf age and season influence
the relationships between leaf nitrogen, leaf mass per area and photosynthesis in maple
and oak trees. Plant, Cell and Environment 14:251-259.

Riddoch, I., T. Lehto, and J. Grace. 1991. Photosynthesis of tropical tree seedlings in
relation to light and nutrient supply. New Phytology 1192137-147.

Roland, J. and J. H. Myers. 1987. Improved insect performance from host-plant
defoliation: winter moth on oak and apple. Ecological Entomology 12:409-414.

Rossiter, M., J.C. Schultz, and I. T. Baldwin. 1988. Relationships among defoliation.
red oak phenolics, and gypsy moth growth and reproduction. Ecology 69:267-277.

Ruohomaki, K., F. S. Chapin, III, E. Haukioja, S. Neuvonen, and J. Suomela. 1996.
Delayed inducible resistance in mountain birch in response to fertilization and shade.
Ecology 7722302-2311.

Safford, L. O., J. C. Bjorkbom, and J. C. Zasada. 1990. Betula papyrifera. Pages 158-
171 in R. M. Burns and B. H. Honkala, editors. Silvics of North America V.2
hardwoods. Agriculture handbook 654, Forest Service, United States Department of
Agriculture, Washington, DC, USA.

SAS Institute, Inc. 1990. SAS/STAT® User’s Guide, Version 6, Fourth Edition, Volume
2. SAS Institute Inc., Cary, North Carolina, USA.

Schaffer, B. and L. J. Mason. 1990. Effects of scale insect herbivory and shading on net
gas exchange and growth of a subtropical tree species (Guaiacum sanctum L.).
Oecologia 84:468-473.

Scriber. J. M. 1977. Limiting effects of low leaf-water content on the nitrogen
utilization, energy budget, and larval growth of Hyalophora cecropia (Lepidoptera:
Saturniidae). Oecologia 28:269-287.

Scriber, J. M. 1984. Nitrogen nutrition ofplants and insect invasion. Pages 441-460 in
R. D. Hauck. editor. Nitrogen in crop production. ASA-CSSA-SSSA. Madison.
Wisconsin, USA.

Scriber, J. M. and F. Slansky, Jr. 1981. The nutritional ecology of immature insects.
Annual Review of Entomology 262183-21 1.

167

 

 

Shure, D. J. and L .A. Wilson. 1993. Patch-size effects on plant phenolics in
successional openings of the southern Appalachians. Ecology 74:55-67.

Smith, N. J. 1991. Sun and shade leaves: clues to how salal (Gaultheria shallon)
responds to overstory stand density. Canadian Journal of Forest Research 21:300-305.

Stamp, N. E. and M. D. Bowers. 1990. Phenology of nutritional differences between
new and mature leaves and its effect on caterpillar growth. Ecological Entomology
15:447-454.

Stockhoff, B. A. 1994. Maximization of daily canopy photosynthesis: effects of
herbivory on optimal nitrogen distribution. Journal of Theoretical Biology 169:209-220. 7‘5

Stoyenoff, J. L., J. A. Witter, M. E. Montgomery, and C. A. Chilcote. 1994. Effects of -_.
host switching on gypsy moth (Lymantria dispar (L.)) under field conditions. Oecologia ;
97: 143-157. ‘.

 

Swain, T. and W. E. Hillis. 1959. The phenolic constituents of Prunus domestica. I. The

quantitative analysis of phenolic compounds. Journal of the Science of Food Agriculture
10:63-68.

1.. . _ Ian-(.-
“'1

Thompson, W. A., P. E. Kriedman, and I. E. Craig. 1992. Photosynthetic response to

light and nutrients in sun-tolerant and shade-tolerant rainforest trees. 1. Growth, leaf

anatomy and nutrient content. Australian Journal of Plant Physiology 19219-42.

Thompson, W. A., L. K. Huang, and P. E. Kriedemann. 1992. Photosynthetic response

to light and nutrients in sun-tolerant and shade-tolerant rainforest trees. 11. Leaf gas
3

exchange and component processes of photosynthesis. Australian Journal of Plant
Physiology 19219-42.

Tuomi. J ., P. Niemela, E. Haukioja. S. Siren, and S. Neuvonen. 1984. Nutrient stress: an
explanation for plant anti-herbivore responses to defoliation. Oecologia 61:208—210.

Tuomi, J., P. Niemela, F. S. Chapin, 111. J. P. Bryant, S. Siren. 1988. Defensive
responses of trees in relation to their carbon/nutrient balance. Pages 57-72 in W. J.
Mattson, J. Levieux, and C. Bernard-Dagan, editors. Mechanisms of woody plant
defenses against insects: search for pattern Springer-Verlag, New York. New York.

USA.

 

Tuomi, J ., P. Niemela. and S. Siren. 1990. The Panglossian paradigm and delayed
inducible accumulation of foliar phenolics in mountain birch. Oikos 59:399-410.

Turlings. T. C., J. H. Loughrin. P. J. McCall, U. S. R. Rose. W. J. Lewis. and .l. H.
Tumlinson. 1995. How caterpillar-damaged plants protect themselves by attracting
parasitic wasps. Proceedings of the National Academy of Science 9224169-4174.

168

Tumbull, M. H. 1991. The effect of light quantity and quality during development on
the photosynthetic characteristics of six Australian rainforest tree species. Oecologia
87:110-117.

U. S. Department of Commerce, National Oceanic and Atmospheric Administration.
1971. Climatography of the United States No. 10-20: Climatological summary for
Midland, Michigan 1940- l 969.

Vitousek. P. M., J. D. Aber, R. W. Howarth, G. E. Likens, P. A. Matson, D. W.
Schindler, W. H. Schlesinger, D. G. Tilman. 1997. Human alteration of the global
nitrogen cycle: sources and consequences. Ecological Applications 72737-750.

Waldbauer, G. P. 1968. The consumption and utilization of food by insects. Advances in
Insect Physiology 52229-289.

Walters, M. B., E. L. Kruger. and P. B. Reich. 1993. Growth, biomass distriubution and
CO; exchange of northern hardwood seedlings in high and low light: relationships with
successional status and shade tolerance. Oecologia 9627-16.

Walters, M. B. and P. B. Reich. 1996. Are shade tolerance, survival, and growth linked?
Low light and nitrogen effects on hardwood seedlings. Ecology 77 2841-853.

Wargo, P. M., J. Parker, and D. R. Houston. 1972. Starch content in roots ofdefoliated
sugar maple. Forest Science 18: 203-204.

Waring, G. L. and N. S. Cobb. 1989. The impact of plant stress on herbivore population
dynamics. Pages 167-226 in E. Bernays, editor. Insect-plant interactions. CRC Press.
Boca Raton, F lorida, USA.

Winn, A. A., and A. S. Evans. 1991. Variation among populations of Prime/la vulgaris
L. in plastic response to light. Functional Ecology 52562-571.

Wold, E. N. and R. J. Marquis. 1997. Induced defense in white oak: effects on
herbivores and consequences for the plant. Ecology 78:1356-1369.

1 ()9