v-
.

. , Nuns .3,
”‘Qéfi

.235.

fun-wt!

m.
“3:1.
Wk

'1‘?
...
on-

. 5“
1.127. 1.23:2-3 7.27.. .. ‘3;
u... .. «m .u
.q . u a .u. u .,

-
u».

Guvuvu~<¢~

runs. ..

:.,:.
:III;:"
,.
7“

. -un‘

. mu.

4::

’H'
:55":
film;

‘ifiiiiifii
‘an - ‘ .. . _
. giz§:-:’:§:¥‘.Jf;,m. «. ' ,. - - ' p ‘ , *3. 4‘“
mpgras Kiri-3; 7' “ ‘ " “- "f ‘
:iéztlg¥;k3q-b , , - , :ihi‘é
‘.-. ‘ ‘0, : ..
r “gag?! ‘ESH
. . t: “' ’= ifiufir
' 3:} ) 5. 21‘5 i; i;
. 15.1 $' ‘1‘": 4" "
AI: . V , ,1: I : l , g: h,
athfls xxg '1 M
Iii! ’ ‘1 ‘ , ~ ‘7 '

" “.1373: 33'

.._
E...-
‘35“,

uh I

' . 11;: a

1, is}; ‘1

Rim
’t

“.3 gm;

:hzfifiiuiwumu

.
.
9 a
..
J . o
11:. " L’ J»
. u L-
,. 'a .—
..
=39“. ?.

an...“-
,4.-
in;
v

c
,. N- .. n...“
.,

v 1‘35; :2 5}: ~
'3':

95:94...

3:13:53

 

L'
MICHIGAN STATE UNIVERSITY UBRARIES

H \\\\\\\\\\\\\\\\\\\\\\\\\\\\\W\\\\\\\\\\\\\\\\\\\i

3 1293 01020 5049

 

 

 

 

i

 

This is to certify that the
thesis entitled

INFLUENCE OF HOST PHENOLOGY ON VOLTINISM PATTERNS

OF ALDER, BIRCH AND HAWTHORN LEAFMINER

(HYMENOPTERA: TENTHREDINIDAE) , WITH EVIDENCE FOR

BET-HEDGING STRATEGIES IN BIRCH AND ALDER LEAFMINER.
presented by

JOseph Lee Chichester

has been accepted towards fulfillment
of the requirements for

_Ma81'.er_'s_ degree in 1mm

flM/flmg

Major professor

 

Date 171 W/qu

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

 

 

‘ LIBRARY
erchlgan State
4 Unlversity

 

 

 

PLACE ll RETURN BOXto roman this chockwt from your mood.
TO AVOID FINES return on or baton data duo.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU In An Manama WM Oppommlty lnutlttflon
Wanna-m

INFLUENCE OF HOST PHENOLOGY ON VOLTINISM PATTERNS OF
ALDER, BIRCH AND HAWTHORN LEAF MINER
(HYMENOPTERA: TENTHREDINIDAE), WITH EVIDENCE FOR BET-HEDGIN G
STRATEGIES IN BIRCH AND ALDER LEAF MINER.

By
Joseph Lee Chichester

A THESIS

Submitted to
Michil'gfin State Uninverstiy
in partial fu ment of the requirements
for the degree of

MASTER OF SCIENCE
Department of Entomology

1994

ABSTRACT

INFLUENCE OF HOST PHENOLOGY ON VOLTINISM PATTERNS OF
ALDER, BIRCH AND HAWTHORN LEAF MINER
(HY MENOPTERA: TENTHREDINIDAE), WITH EVIDENCE FOR
BET-HEDGING STRATEGIES IN BIRCH AND ALDER LEAF MIN ER

By

Joseph Lee Chichester

Alder, birch, and hawthorn leafrniner oviposited exclusively in immature foliage.
Larvae were aggregated even though no consistent physical and chemical differences were
detected between colonized and non-colonized leaves, and despite intra-specific
competition. Larval mortality increased as leaves matured, becoming tougher and less
nutritious. Pre-mature extra-long diapause occurred in first and second generation birch
and alder leafminers, with the number of emerging adults proportional to the quantity of
immature foliage available to their progeny.

These data support the hypothesis that larvae can develop only in immature foliage,
and that the indeterminate growth of birch and alder permitted the evolution of
multivoltinism in birch and alder leafminers, while determinate growth of hawthorn has

constrained hawthorn leafminer to one generation. Premature extra-long diapause of birch

and alder leafminers may be a bet-hedging strategy; diapausing individuals forego a
breeding opportunity, but avoid competition for the decreased supply of their obliate food

resource: immature leaves.

I dedicate this work to Marcy Ochs and my immediate family.
Their patience, support, and understanding

could never be surpassed.

iv

ACKNOWLEDGMENTS

I thank Daniel A. Herms, my major advisor and friend, for giving me the
opportunity to pursue a career in entomology. Dan spent countless hours reviewing my
work and making valuable suggestions which have helped me grow as a scientist. This work
would not of been possible without the financial support of The Dow Gardens and the
Herbert and Grace A. Dow Foundation of Midland, Michigan. I thank Douglas J.
Chapman, Director of The Dow Gardens, for his willingness to support me in this project.

I thank my committee, William Mattson, Guy Bush, James Miller, and Cathy
Bristow for their valuable suggestions throughout the planning of and interpretation of my
research. I would like to extend my gratitude to William Mattson for his valuable data
analysis assistance and suggestions.

This research would not of been possible without the extended help of Ted Bomba,
Tisha Walls, and Bruce Birr. Ted and Tisha spent many tedious hours counting leaves and
collecting other data. Their good sense of humor made long days something to look
forward to. Bruce spent numerous hours conducting chemical analysis on hundreds of leaf
samples. I also thank Bruce for his patience and friendly assistance with my countless
questions and interruptions. I would also like to recognize D.R. Smith, Research
Entomologist, Systematic Entomology Laboratory, USDA, for the proper identification of
the insects used in this study. '

I would like to acknowledge two colleagues: Joe Spencer and Cathy Papp. I am
grateful for their support, suggestions, and friendship, which I can’t thank them enough for.

My immediate family was very supportive in my quest to earn my degree. They
constantly reminded me of my success and how I made them very proud. Last, but not
least, I gratefully thank Marcy Ochs for her love, friendship, support, understanding, and

most of all her extreme patience. I am forever grateful!

vi

TABLE OF CONTENTS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

LIST OF FIGURES -- ......................................................................... x

INTRODUCTION -- - -- -- -- 1

Objectives -- - . ..... - -- 7

Theoretical Model -7

Specific Objectives _---8
CHAPTER 1- Abundance, Larval Distribution, and Voltinism Patterns of

Alder, Birch, and Hawthorn Leafminer

in Central Michigan - --9

Introduction ...... 9

Objectives ..... - -- - ............................................................ 11

Methods and Materials ................................................................................................... 12

Research Sites 12

Adult Abundance and Phenology ..................................................................... 12

Larval Distribution and Abundance A- - -- - 13

misfit figtfilfifiefififl‘fiEiSSEf ............................................................... 13

Data Analysis - .............................. 14

Adult Abundance and Phenology and Larval

Distribution and Abundance - - -- -_ .................. 14

fm fiflfi’l‘ffiit'ftfi‘fiiiSYEf ................ = ............................ 1.

Results ................................................................... 15

Adult Abundance and Phenology ..................................................................... 15

Larval Distribution and Abundance ................................................................ 18

vii

1mysical and Chemical Comparison of

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

sted and Uninfested Leaves 26
Discussion 41
Adult Abundance and Phenology _ - " ........ -- _ 41
Larval Distribution and Abundance - - -44
IPh‘jysical and Chemical Comparison of
ested and Uninfested Leaves 46
CHAPTER 2- Interspecific Variation in Voltinism Patterns
Results from Interspecific Differences 1n Host
Phenology ...... 48
Introduction . 48
Objectives so
Methods and Materials 50
Leaf Age Influences Oviposition Site Selection
of Second and Third Generation Alder and
Birch Leafminer _ 50
Number of Eggs Per Leaf in Relation to Physical
Leaf Characteristics - 52
Physical and Chemical Comparisons of Mature and
Immature Leaves 52
Age Specific Larval Mortality 1n Relation to
Leaf Maturity 53
Results 54
Leaf Age Influences Ovi sition Site Selection
During the Second and ‘ d Generations 54
Number of Eggs Per Leaf in Relation to Physical
Leaf Characteristics ...................... 54
Physical and Chemical Comparisons of Mature and
Immature Leaves 60
Age Specific Larval Mortality 1n Relation to
Leaf Maturity _ -74
Discussion -- 81

 

viii

CHAPTER 3- Host Phenology, Intraspecific Competition,
Climatic Variation and the Evolution of

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Intraspecific Variation 1n Voltinism Patterns --83
Introduction - ........ - ..... -- - 83
Objectives -- 86
Methods and Materials -86
There is Seasonal Variation in the Production
of Immature Leaves - 86
Intraspecific Variation 1n Voltinism Patterns
Corresponds to Phenological Patterns of New
Leaf Production 87
Intraspecific Competition Occurs Among Larvae ............... 88
Results .............. 89
Seasonal Variation in the Production
of Immature Leaves ........ - 89
Intraspecific Variation in Voltinism Patterns
Corresponds to Phenological Patterns of New
Leaf Production ........ _ 91
Intraspecific Competition Occurs Among Larvae 93
Relationship Between larval Density and
Pupal Mass ........... 93
Relationship Between Larval Density and
Survival, Aclult Emergence and Diapause ........................................... 93
Relationship Between Pupal Mass and
Survival, Aclult Emergence and Diapause - ............. 97
Discussion -- -- -- ...................... 106
SUMMARY 111
APPENDIX 1. Record of Deposition of Voucher Specimens - ..... 116
APPENDIX 1.1 Voucher Specimen Data - .-....- ......................... 117
LIST OF REFERENCES _ - - .............................................................. 118

 

ix

Figure 1:

Figure 2:

Figure 3:

Figure 4:

Figure 5:

Figure 6:

Figure 7:

Figure 8:

Figure 9:

Figure 10:

LIST OF FIGURES

Seasonalp phenologyB of adult alder leafminer in
1992 (A) and 199 (B) -

Seasonal phenolo OB)? of adult birch leafminer m
1992 (A) and 199 () - -

Seasonal phenolo of adult hawthorn leafminer

in 1992 (A) and 1 93 (B) ..................................................................

Mean versus variance for the number of first
geeneration larvae per leaf 1n 1993. (Agw alder
afminer, (B) birch leafminer, and (C hawthorn

...... 16

...... 17

...... 19

leafminer _ ................. 21

 

Distribution and abundance of alder leafminer
larvae per leaf: (A) first generation (2 June, 1993),
(B) second eneration (2 June 1993) and (C) second

generation 13 July, 1993) -- .......... 22

 

Distribution and abundance of birch leafminer
larvae per leaf: (A) first generation, (B) second
generation, and C) third generation 1993)

Distribution and abundance of hawthorn leafminer

larvae per leaf (1993) - - -- ................

 

Tou ness of infested and uninfested leaves during 1992:
(A) uropean alder, (B) paper birch, and (C). hawthorn.
Sampling dates correspond with peak oviposrtion and larval
development of first generation alder (A), birch (B), and

25

...... 27

hawthorn (C) leafminer .......................................................................... 28

Tou%hness of infested and uninfested leaves during 1993:

uropean alder, (B) paper birch, and (C) hawthorn.
ampling dates correspond with peak oviposrtion and larval

development of first generation alder (A), birch (B) and

hawthorn (C) leafminer _, _ ................ ............... 29

 

Area of infested and uninfested leaves durin 1992:

(A) European alder, (B) paper birch, and (C hawthorn.
Sampling dates correspond with peak ovr rosrtion and
larval development of first eneration a1 er (A),

birch (B), and hawthorn (C leafminer ................................................ 32

X

Figure 11:

Figure 12:

Figure 13:

Figure 14:

Figure 15:

Figure 16:

Figure 17:

Figure 18:

Figure 19:

Area of infested and uninfested leaves durin 1993:

(A) European alder, (B) paper birch, and (C hawthorn.
Sampling dates correspond with peak ovi osrtion and
larval development of first eneration al er (A),

birch (B), and hawthorn (CI leafminer ........................................

Specific leaf mass of infested and uninfested leaves

during 1992: (A) European alder, (B) paper birch,

and (C) hawthorn. Sampling dates correspond with

peak oviposition and larval development of first
eneration alder A), birch (B), and

 

awthorn (C) 1e 'ner _ - - ...............................

Specific leaf mass of infested and uninfested leaves

during 1993: (A) European alder, (B) paper birch,

and (C) hawthorn. Sampling dates correspond with

peak oviposition and larval development of first
eneration alder A), birch (B), and

awthorn (C) le miner ..................................................................

Nitrogen concentration of infested and uninfested
leaves during 1992: (A) European alder,

(B) paper birch, and (C) hawthorn. Sampling dates
correspond with eak oviposition and larval

develo ment of 'rst generation alder (A),

birch ( ), and hawthorn (C) leafminer ........................................

Phosphorous concentration of infested and uninfested

leaves during 1992: (A) European alder, (B) paper

birch, and ( ) hawthorn. Sampling dates corres ond

with peak oviposition and larval development 0 first
eneration alder A), birch (B), and

awthorn (C) lea 'ner - -- _ _- .......

 

Comparison of immature and mature European alder
leaves as ov1posrtron sites b alder leafminer:

(A) no-choice experiment, B) choice experiment ....................

Comparison of immature and mature paper birch leaves
as ovrposition sites by birch leafminer: (A) no-choice

experiment, (B) choice experiment ..............................................

Relationship between leaf toughness and the number of
e 5 per leaf: (A) European alder, (B) paper birch, and

( hawthorn. Sampling dates correspond with Ipeak
ov1 osition of first generation alder (A), birch (

 

an hawthorn (C) leafminer in 1993 -

Relationship of leaf area and the number of e 5 per leaf:

(A) European alder, (B) paper birch, and (C) awthorn.
Sampling dates corres ond with peak oviposrtion of first
eneration alder (A), irch (B), and hawthorn (C)

eafminer in 1993 .............................................................................

xi

........ 33

........ 35

........ 36

........ 38

........ 40

......... 55

........ 56

57

........ 58

Figure 20: Relationship between s ecific leaf mass and the number
of eggs laid per leaf: (A Euro ean alder, (B) aper birch,
and (C) hawthorn. Sampling ates correspond) with peak
oviposition of first generation alder (A), birch (B),
an hawthorn (C) leafminer in 1993 .................................................... 59

Figure 21: Ton ness of immature and mature leaves during 1992:
(A) uropean alder, (B) paper birch, and (C) hawthorn.
Sampling dates correspond with peak ovi osrtion and
larval development of each generation 0 alder (A),
birch (B), and hawthorn (C) leafminer ................................................ 61

Figure 22: Ton hness of immature and mature leaves during 1993:
(A) uropean alder, (B) paper birch, and (C) hawthorn.
Sampling dates correspond with peak ovi 051tion and
larval development of each generation 0 alder (A),
birch (B), and hawthorn (C) leafminer ................................................ 62

Figure 23: Area of immature and mature leaves during 1992:
A) European alder, (B) paper birch, and (C) hawthorn.
ampling dates correspond with peak ovi osrtion and
larval development of each generation 0 alder (A),
birch (B), and hawthorn (C) leafminer ................................................ 64

Figure 24: Area of immature and mature leaves during 1993:
(A) European alder, (B) paper birch, and (C) hawthorn.
Sampling dates correspond with peak ovi osrtion and
larval development of each generation 0 alder (A),
birch (B), and hawthorn (C) leafminer ................................................ 65

Figure 25: Specific leaf mass of immature and mature leaves
during 1992: (A) European alder, (B) paper birch,
and (C) hawthorn. Sampling dates correspond with
peak oviposition and larval development of each
lgleneration of alder (A), birch (B), and
awthorn (C) leafminer .......................................................................... 67

Figure 26: Specific leaf mass of immature and mature leaves
during 1993: (A) European alder, (B) paper birch,
and (C) hawthom. Sampling dates correspond with
peak oviposition and larval development of each
generation of alder (A), birch (B), and
awthorn (C) leafminer .......................................................................... 68

Figure 27: Nitrogen concentration of immature and mature leaves
during 1992: (A) European alder, (B) paper birch, and
(C) hawthorn. Sampling dates corres ond with peak
ov1position and larval development 0 each generation
of alder (A), birch (B), and hawthorn (C) leafminer ......................... 71

xii

Figure 28:

Figure 29:

Figure 30:

Figure 31:

Figure 32:

Figure 33:

Figure 34:

Figure 35:

Figure 36:

Figure 37:

Figure 38:

Figure 39:

Phosphorous concentration of immature and mature leaves
during 1992: (A) European alder, (B) paper birch, and

(C) hawthorn. Sampling dates corres ond with peak
ov1position and larval development 0 each generation

of alder (A), birch (B), and hawthorn (C) leafminer ......................... 73

The relationship between leaf toughness and instar-specific
larval mortality of first Igeneration alder leafminer
in 1992 (A) and 1993 (

The relationship between leaf toughness and instar-specific
larval mortality of second generatron alder leafminer
in 1992 (A) and 1993 (B) - -

Relationship between leaf toughness and instar-specific
larval mortality of first eneration birch leafminer
in 1992 (A) and 1993 ( )

Relationship between leaf toughness and instar-specific
larval mortality of second generation birch leafrrnner

) - ................. ..... -

in 1993 ...... --

 

Relationship between leaf toughness and instar-specific
larval mortality of hawthorn leafminer in 1992 (A)
and 1993 (B) ...... _ - - -

Seasonal production of immature leaves b European
alder (A), paper birch (B), and hawthorn C)

 

Relationship between the percent of leaves that are
immature and the size of each generation of adult
leafminers, relative to the first generation:

75

. ............ 77

-_ -- ........................ 78

79

80

-90

alder (A), birch (B), and hawthorn (C) leafminer (1993) ................. 92

Relationship between first generation larval density

and pupal mass in 1993: (A) alder leafminer, (B) birch

leafminer, and (C) hawthorn leafminer

 

Relationship between second generation larval density
and ‘pupal mass in 1993: (A) alder leafminer and

(B) '

Relationship between first generation alder leafminer
larval density and (A) percent of individuals surviving,

94

1rch leafminer .................................................................................. 95

- --96

 

EB) percent of individuals emerging as adults, and
C percent of individuals diapausing (1993)

Relationship between second generation alder leafminer
larval density and (A) percent of individuals surviving,
(13 percent of individuals emerging as adults, and

xiii

percent of individuals diapausing (1993) ...................................... 98

Figure 40:

Figure 41:

Figure 42:

Figure 43:

Figure 44:

Figure 45:

Relationship between first generation birch leafminer
larval density and (A) percent of individuals surviving,
EB) percent of individuals emerging as adults, and

C percent of individuals diapausing in 1993 -

 

Relationship between second generation birch leafminer
larval density and (A) percent of individuals surviving,
EB) percent of individuals emerging as adults, and

C

Relationship between first generation alder leafminer
pu a1 mass and (A) percent of individuals surviving,
ngpercent of individuals emerging as adults, and

Relationship between second eneration alder leafminer
pu al mass and (A) percent 0 individuals surviving,
(ngercent of individuals emerging as adults, and

C

Relationship between first generation birch leafminer
pu a1 mass and (A) percent of individuals surviving,
Eggpercent of individuals emerging as adults, and

percent of individuals diapausing (1993) .................................... 100

percent of individuals diapausing (1993) .................................... 101

percent of individuals diapausing (1993) .................................... 103

............. 104

 

percent of individuals diapausing (1993)
Relationship between second eneration birch leafminer
pu a1 mass and (A) percent 0 individuals surviving,
(ngercent of individuals emerging as adults, and
C

xiv

percent of individuals diapausing (1993) .................................... 105

INTRODUCTION

Three related hymenopteran leafminers (Tenthredinidae: F enusini) occur
sympatrically in Midland, Michigan: alder leafminer (F enusa dohmii Tischbein),
birch leafminer (Fenusa pusilla Lepeletier) and a leafminer of hawthorn (Profenusa
canadensis Marlett) (hereafter referred to as the hawthorn leafminer). They
specialize on tree species within the genera Aluus, Betula, and Crataegus,
respectively. The biology of all three species is similar, with the birch leafminer
being most thoroughly studied. Birch leafminer females deposit eggs singly between
the epidermal layers of an immature leaf (Friend 1933, DeClerck & Shorthouse
1985). Upon hatching, larvae begin mining the palisade tissue. Mining ceases when
the fifth instar (pre-pupal stage) is reached (Friend 1933). Mature larvae exit the
leaf, fall to the ground, and pupate a few centimeters below the soil surface (Jones
and Raske 1976, DeClerck & Shorthouse 1985). At this point, interspecific and
intraspecific differences in life history patterns occur. The hawthorn leafminer is
univoltine, remaining in the soil until the following spring (Wilson et al. 1982).
However, alder and birch leafminers are multivoltine, with first and second
generation individuals emerging as adults in early July and mid-August to initiate a
second and third generation, respectively (DeClerck & Shorthouse 1985; Hart et al.
1991).

Since females oviposit almost exclusively in immature foliage (Friend 1933,
Wilson et al. 1982, Hart et al. 1991), the evolution of voltinism patterns of the alder,
birch, and hawthorn leafminers may have been influenced by the phenology of their

respective host plants. Alder and birch display indeterminate growth, producing
1

new foliage throughout the summer (personal observation, Friend 1931). Hawthorn
trees do not produce new foliage after the initial flush at bud-break, thus displaying
determinate growth (personal observation). The continuous production of leaves by
alder and birch may have allowed alder and birch leafminer to evolve a multivoltine
life cycle, whereas hawthorn leafminer may be restricted to a univoltine life cycle
due to the determinate growth pattern of its host. Such phenological constraints on
life history have been reported by Auerbach and Simberloff (1984) for two
leafmining species, Eriocraniella sp and Dysen'ocranina sp (Lepidoptera:
Eriocraniidae), which also feed only on first-flush leaves.

Qualitative differences between immature and mature leaves may explain
why females oviposit in immature foliage. Food quality has been shown to play a
role in insect seasonality in many cases (Wolda 1988). Ayres and MacLean (1987)
identified several leaf characteristics that change with age, affecting the growth of a
geometrid larvae (Epim’ta autumata). As leaves age, water, nitrogen, and
phosphorous concentrations decline and larval performance decreases while
toughness and specific mass increase. They proposed that phenological constraints
imposed on the larvae by leaf maturation should influence the evolution of larval
phenology, selecting for early season development. A similar relationship between
leaf and insect phenologies was documented by Merle (1988) for the green oak
leafroller, Tortrix vin'dana (Lepidoptera: Tortricidae) and various species of oak.

In central Michigan, the number of first generation adult birch leafminers
(overwintering individuals) always greatly
xceed numbers of second and third generation adults (Herms unpublished data). In
the case of the alder leafminer, large first and second generations are followed by a
small third generation (Hart et al. 1991). This intrapopulation variability in adult
emergence may be related to host plant phenology. If the number of immature

leaves present at any one time decreases after the first flush, then the quantity of

immature leaves available to second and third generation alder and birch leafminer
larvae would be only a fraction of that available to the first.

Intrapopulation variability in the timing of adult emergence which coincides
with temporal variability in resource availability has been reported in two
Lepidopteran (Gracillariidae) leafmining species, Acrocercops sp. and Neurobathra
strigifinitella (Auerbach & Simberloff 1984). The voltinism patterns of these two
species are similar to that of the birch leafminer, with a large first generation
followed by a small second and third generation.

The decreased availability of immature leaves over the course of the season
could create the potential for competition among larvae. The observed intraspecific
variation in voltinism patterns of the birch leafminer may be a density-dependent
evolutionary response to intraspecific competition among larvae for the limiting
food resource (immature leaves). Intraspecific competition has been observed in
several leafminer species. Tuomi et al. (1981) documented intraspecific
competition in Scolioneura betuleti, a hymenopteran leafminer colonizing birch
leaves. Intraspecific competition occurred as population density increased, resulting
in decreased larval survival. They also found a negative correlation between mean
larval weight and larval density. Furthermore, mean larval weight was positively
correlated with leaf area available per larva. These findings coincide with those of
Quiring and McNeil (1984), who documented competition in alfalfa blotch
leafminer (Agromyza fmntella Rondani) populations. Stiling ( 1984) reported that
larval mortality of Hydrellia valida Loew on salt marsh cordgrass increased when two
or more mines coalesced, and that intraspecific competition accounted for 1 1.4% of
total mortality. Intraspecific competition has also been documented in populations
of leafminers on oak, aspen, and lima beans (Bultrnan & F aeth 1986a, Auerbach
1991, and Petitt & Wietlisbach 1992, consecutively). It appears that intraspecific

competition occurs at high larval densities due to food limitation.

Intraspecific competition may select for density-dependent premature extra-
long diapause (sensu Hanski 1988). Diapause is a state of low metabolic activity
which is hormonally controlled. During diapause, morphogenesis is reduced,
resistance to environmental extremes is increased, and behavioral activity is altered
or reduce ('Tauber et al. 1984). Hanski (1988) reviewed the different kinds of extra-
long diapause. Extra-long diapause occurs when individuals miss one or more
breeding opportunities experienced by other members of the population by entering
diapause early (premature extra-long diapause) or by exiting diapause late
(prolonged extra-long diapause). A

Premature and prolonged extra-long diapause may be subdivided into
density-independent and density-dependent diapause. Density-independent
diapause induction is not affected by density and all individuals face the same
environmental factors (Hanski 1988). Density-dependent diapause, on the other
hand, is affected by the number of individuals in which a target individual interacts
and therefore, a function of population size or density. As larval density increases,
the frequency of negative interactions (intraspecific competition) between
individuals increases. Thus, intraspecific competition among larvae may select for
density-dependent premature extra-long diapause. Premature extra-long diapause
may be an evolutionary response to larval competition by a portion of the second
and third generation of alder and birch leafminer population. If there is no
intrapopulation variability in emergence (i.e. all individuals emerge as adults) and if
fewer leaves are available for oviposition as the season progresses, then larval
competition could potentially be very high during the second and third generations,
resulting in high larval mortality. The pitcher-plant mosquito, Meomyia smithii, is
one of the few documented examples of density-dependent premature extra-long

diapause. According to Istock et al. (1975), when populations are high, some

individuals may shift to a univoltine life cycle by entering diapause until resources
are more plentiful.

The small size of the second generation birch leafminer, relative to the first
generation, may result because most individuals do not emerge from pupation for a
second generation. Instead, these individuals may enter density-dependent
premature extra-long diapause (sensu Hanski 1988) to avoid intraspecific
competition among offspring in a resource-limited habitat (limited availability of
immature leaves). These individuals would loose a breeding opportunity and
emerge the following spring when a full flush of leaves is once again available.
Thus, a "bet-hedging" (Philippi and Seger 1989) diapause strategy may have evolved
in birch leafminer. Philippi and Seger (1989) defined bet-hedging as an
evolutionary trade off between mean and variance of fitness. Individuals with a
lower mean fitness may have a selective advantage over highly fit individuals under
certain conditions. In years where conditions are favorable for larval development
(warm summers, high quality food, and low levels of competition), the small portion
of the population that emerges for a second generation would have a selective
advantage over the portion of the population that diapaused. In years when
conditions are not favorable, the portion of the population entering diapause would
have a selective advantage.

A evolutionary bet-hedging strategy linking winter diapause termination
with resource availability has been identified in two Lepidoptera (Eriocraniidae)
leafmining species. Acrocercops sp. and Neurobathra sp. have staggered termination
of winter diapause which allows them to respond to atypically large leaf flushes at
certain times of the growing season (Auerbach & Simberloff 1984).

Climatic conditions such as photoperiod, temperature, and resource
availability and quality may play a significant role in insect phenology (Tauber and

Tauber 1973). Climatic conditions influence the termination date of leaf production

and larval development rates (Scriber & Lederhouse 1992). Thus, climatic variation
has the potential to influence the evolution of a density-independent premature
extra-long diapause in second generation alder and birch leafminers. Decreased
numbers of adults giving rise to third generation larvae suggest that some second
generation alder and birch leafminer individuals may hedge their bets by entering
density-independent premature extra-long diapause (sensu Hanski 1988). This
portion of the population would loose a breeding opportunity but emerge the
following spring when the development of their offspring would not be constrained
by climatic conditions. The offspring of the smaller portion of the population giving
rise to the third generation, may have to "race" to finish development before
unfavorable fall climate conditions prevail (cessation of immature leaf production,
low temperatures, and leaf abscission). These individuals would have a selective
advantage over diapausing individuals in years when development can be
completed.

Bet-hedging may also take the form of risk-spreading through phenotypic
variability in offspring. Offspring from a single alder or birch leafminer female may
be phenotypically different from one another to ensure that at least some
individuals will be adapted to a uncertain future. Thus, it is an adaptive strategy
that may increase fitness in a temporally variable environment (Bradford & Roff
1993). Bradford and Roff (1993) suggested that a bet-hedging diapause strategy
occurs in a partially bivoltine population of cricket (Allanemobius fasciatus). As first
generation females aged, the proportion of diapausing eggs increased, reflecting a
decrease in the probability of second generation offspring being able to complete

development before the end of the growing season.

OBJECTIVES
Theoretical Hypothesis

The primary objective of this research was to test the following theoretical

hypothesis:

The phenological pattern of immature leaf production by their respective
host plants has directed the evolution of the voltinism patterns of alder,
birch, and hawthorn leafminer. Specifically, the determinate growth pattern
of hawthorn has constrained the hawthorn leafminer to a univoltine life
history, while the indeterminate growth pattern of alder and birch has
allowed the evolution of multivoltinism in alder and birch leafminer.
Furthermore, intraspecific temporal variation in the pattern of new leaf
production by alder and birch has resulted in a bet-hedging diapause
strategy in alder and birch leafminer. This strategy takes the form of
density-dependent premature extra-long diapause (sensu Hanski 1988)
selected for by intraspecific competition for the immature leaf food

resource.

Specific Objectives

The following experimental objectives were used to test the theoretical

hypothesis:

(1)

(2)
(3)

(4)

(5)

(6)

(7)

to characterize the seasonal presence and abundance of alder, birch,
and hawthorn leafminer;

to characterize the spatial distribution of larvae of all three species;
in order to explain observed distribution patterns, to compare
physical and chemical traits of colonized and noncolonized leaves;
to test the hypothesis that immature leaves are the only suitable
resource for the survival and development of all three species by:

(a) in choice and no-choice experiments, comparing
oviposition preference for immature relative to mature
foliage,

(b) comparing chemical and physical differences of
immature and mature leaves,

(c) quantifying larval survival of all three species in
relation to leaf maturity;

to quantify seasonal variation in the availability of the immature leaf
resource of all three host species;

to test for intraspecific competition among larvae of all three species
for their respective immature leaf resource; and

to test for correlation between variation in the relative seasonal
abundance of the three leafminer species and temporal variation in

the production of immature leaves by their respective host plants.

CHAPTER ONE

ABUNDANCE, LARVAL DISTRIBUTION, AND VOLTINISM
PATTERNS OF ALDER, BIRCH, AND HAWTHORN LEAFMIN ER IN
CENTRAL MICHIGAN

Alder, birch, and hawthorn leafminer are closely related hymenopteran
sawflies (Tenthredinidae: F enusini) occurring sympatrically in Midland, Michigan.
They specialize on tree species within the genera Alnus, Betula, and Crataegus,
respectively. Alnus glutinosa L., Betula papyrifera Marsh., and Crataegus crusgalli L.
were the respective hosts of the leafminer populations studied in this project. The
hawthorn leafminer is native to North America, while the alder and birch leafminers
were introduced from Eurasia (Smith 1979).

The biology of the three leafminer species are similar. Each overwinters as
pupae in the soil directly below its host. Adult emergence begins in late April to
mid-May as bud break occurs (Slingerland 1905, Parrott & Fulton 1915, Friend
1933, Hart et al. 1991). Females of all three species oviposit almost exclusively in
immature leaves (Slingerland 1905, Parrott & Fulton 1915, Friend 1933, Wilson et
al. 1982, Hart et al. 1991). Eggs of the birch leafminer do not hatch if deposited in
mature or nearly mature leaves (Friend 1933).

Females deposit eggs singly between the epidermal layers of a leaf
(Slingerland 1905, Parrott & Fulton 1915, Friend 1933, DeClerck & Shorthouse
1985). Upon hatching, larvae begin mining the palisade tissue. Mining ceases when
the fifth instar (pre-pupal stage) is reached (Friend 1933). Mature larvae exit the

leaf, fall to the ground, and pupate a few centimeters below the soil surface
9

10

(Slingerland 1905, Parrott & Fulton 1915, DeClerck & Shorthouse 1985). At this
point, interspecific and intraspecific differences in life history patterns occur. The
hawthorn leafininer is univoltine, remaining in the soil until the following spring
(Parrott & Fulton 1915, Wilson et al. 1982). However, alder and birch leafminers
are multivoltine with first and second generation individuals emerging as adults in
early July and mid-August to initiate respective second and third generation
(DeClerck & Shorthouse 1985; Hart et al. 1991).

The distribution of organisms within their environment may be random or
non-random. A random distribution implies that the environment is homogenous
and/ or individuals do not selectively choose their habitat. Organisms may also be
non-randomly distributed due to some constraints. Non-randomly distributed
organisms can be either clumped (aggregated) or uniform (evenly distributed)
(Ludwig & Reynolds, 1988). Such factors as wind and water currents, light intensity,
reproductive strategy (asexual, sexual), territoriality, and intraspecific competition
may influence the distribution of organisms (Ludwig & Reynolds, 1988).

A result of oviposition site selection by females, the larval distribution of
several leafminer species has been found to be clumped. Stiling et a1. (1987)
reported that the larvae of Stilbosis quadricustatella, a leafmining moth, are
distributed non-randomly within leaves of sand live oak (Quercus geminata) and
water oak (Quercus nigra). When leaves contained two or more larvae, fewer larvae
were found together on the same side of the mid-vein than expected by chance.

Bultman and Faeth (1986b) reported that four species of leafminers~ on
Quemus emoryi selected leaves based on leaf size. Stigmella sp. and T ischen'a sp.,
which produce small mines, selected small leaves. Stilbosr's juvantis produces
intermediate sized mines and selected average sized leaves while Cameran'a sp.
selected large leaves and produced large mines. In addition, Cameran’a sp. densities

were lower and dispersion was more aggregated on small-leaved branches and, in

11

contrast to the birch leafminer, densities were greater in the shaded regions of trees
(Faeth 1991).

Oviposition and larval distribution may be influenced by such factors as leaf
location (sun leaves vs shade leaves) (Hamilton 1943, Faeth 1991), leaf size
(Bultman & Faeth 1986b, Faeth 1991), leaf abscission (Stiling et al. 1991), and
resource or interference competition (Stiling et al. 1987, F aeth 1991).

Birch leafininer distribution is concentrated on the tip and periphery of the
crown (Jones and Raske 1976). Friend (1931) reported that birch leafminer females
only oviposited in immature leaves, resulting in only the tips of the branches being
infested. Hawthorn leafminer females prefer to oviposit in leaves that first appear
and which are still in the unfolded state (Parrott & Fulton 1915) and which are in a

warm sunny location (Hamilton 1943).

OBJECTIVES

The objectives of this research were to characterize larval and adult
phenology and abundance of alder, birch, and hawthorn leafminers, as well as the
spatial distribution of larvae within the host plant. Physical and chemical
differences between infested and uninfested leaves were also characterized in an

attempt to understand the mechanisms underlying the observed distributions.

METHODS AND MATERIALS

Research Sites

Populations of all three species were identified in Midland, Michigan. The
alder leafininer population had colonized an experimental planting of European
alder (Alnus glutinosa), and the birch leafminer population was located in an
experimental paper birch planting (Betrda papyn'fera) at the Dow Gardens.
Hawthorn leafminer had established a population in an ornamental hawthorn
(Crateagus crusgalli) planting on the estate of H. H. Dow. These populations and

host plants were used for all studies.

Adult Abundance and Phenology

To monitor the presence and abundance of each species, one yellow sticky
trap was deployed in each of eight trees for all three species. Traps were hung near
the terminal portion of branches. Therefore, during the second and third generation
of alder and birch leafminer, traps were near immature leaves (oviposition sites). In
1993, four additional alder trees were used to monitor the population. Each of the
four alder trees received two yellow sticky traps, while each of the original eight
trees received one yellow sticky trap, for a total of 16 traps. In 1992 and 1993, from
May to August, the number of adults per trap were counted weekly and traps were

changed as needed. The mean number of adults per trap was plotted against time.

12

Larval Distribution and Abundance

In 1993, larval abundance and distribution were quantified by examining
each individual leaf on each of four branches on six alder, eight birch, and eight
hawthorn trees. Branches of approximately equal size that exhibited long extension
growth the previous year were sampled. The number of larvae in each leaf of each
branch was recorded. In 1992, the larval density and distribution of birch, alder, and
hawthorn leafminers was attempted, but complete data sets were not acquired. The

density and distribution of all generations of each species were quantified in 1993.

Physical and Chemical Comparisons of Infested and Uninfested Leaves

In 1992 and 1993, during the first generation, ten infested and ten
uninfested leaves of the same age class were randomly sampled on a weekly basis
from each of eight trees of all three species. While in the field, leaves were placed
in labelled plastic bags and stored on ice. Leaves were promptly transported to the
laboratory.

For all leaves the following variables were measured: (1) toughness (g) with
a penetrometer, (2) leaf area (cm?) with a digital image analyzer, (3) specific leaf
mass (g/mz), and (4) nitrogen and phosphorous concentration (% dry wt.) using
standard micro-kjeldahl techniques. Chemical analyses were conducted in the
laboratory of WJ. Mattson by Bruce Birr, USDA Forest Service, at Michigan State
University, East Lansing, MI.

13

DATA ANALYSIS

Adult Abundance and Phenology

and Larval Distribution and Abundance

To test the hypothesis that larvae were distributed randomly among leaves,
the observed larval distribution was compared with the theoretical Poisson
distribution using Chi-Square Analysis. If the hypothesis of a Poisson distribution
was rejected (indicating a non-random distribution), and the variance-to-mean ratio
was greater than one, indicating clumping, then the hypothesis that larvae were
distributed according to the negative binomial model was tested using Chi-Square
analysis and by calculating the k parameter. As k approaches zero, the degree of
clumping increases. Evidence of clumping was further evaluated by calculating the
index of dispersion (ID) (variancezmean ratio) and d-statistic. An ID value of less
than one indicates a random distribution, a value equal than one indicates a uniform
distribution, and a value of greater than one indicates a clumped distribution
(Ludwig & Reynolds, 1988). An absolute d-statistic value of less than 1.96 indicates
a random distribution, a d-statistic value less than -1.96 indicates a uniform
distribution, and a d-statistic value greater than 1.96 indicates a clumped
distribution (Ludwig & Reynolds, 1988). The degree of clumping was also
estimated using Green’s index. Green’s index ranges from zero to one, with zero
indicating a random distribution and one indicating a clumped distribution (Ludwig
& Reynolds, 1988). In addition, larval abundance and distribution were also visually

examined by plotting a frequency histogram of the number of larvae per leaf.

14

Physical and Chemical Comparisons of Infested and Uninfested Leaves

Means of the following variables of infested and uninfested leaves were
compared using a t-test: (1) leaf toughness, (2) leaf area, (3) specific leaf mass, and

(4) chemical composition (percent nitrogen and phosphorous).

RESULTS

Adult Abundance And Phenology

Alder leafminer

In 1992, adults emerged before sampling began on 11 May (Figure 1A).
Distinct generations were difficult to distinguish, but at least two population peaks
were observed. The first peak occurred on 11 May (16:3 adults/trap), and the
second on 27 July (28: 6 adults/ trap). No adults were observed after 13 September.

In 1993 traps were deployed on 28 April. Adults were first collected on 11
May (Figure 1B). As in 1992, distinct generations were difficult to distinguish, but
at least two population peaks were observed with the second being larger (205 z 63
adults/ trap on 26 July) than the first (80:25 adults/ trap on 10 May). N 0 adults
were observed after 26 August.

Birch leafminer
In 1992, adults emerged before sampling began on 11 May; the first
population peak was large (122324 adults on 11 May) (Figure 2A) while the second

15

16

 

50
40 '

30'

 

20’

Adults / Trap

10

 

 

300
250 ' I B
200 ' .

150 ~ 1
100 '
50 '

may all... 1.1. 1...

Adults I Trap

 

 

 

Figure 1: Seasonal phenology of adult alder leafminer in 1992 (A) and 1993 (B).

17

 

 

 

 

 

 

150
O.
2
1— 100
\
.9
3 .
.0 so
<

o

150 B
O.
S .
,_ 100
1?.
.3 50+
<

 

 

 

 

 

May June July Aug

Figure 2: Seasonal phenology of adult birch leafminer in 1992 (A) and 1993 (B).

18

and third population peaks were small relative to that of the first (< 1 adult/ trap).
Adults were last observed on 14 July (Figure 2A).

In 1993, traps were deployed on 28 April and adult emergence was first
observed on 11 May. Again a large first generation (112116 adults/ trap on 18 May)
was followed by a relatively small second and third generation (05 adults / trap)
(Figure 2B). Peak emergence of the second and third generations were observed on
13 July and 10 August, respectively and no adults were observed after 17 August
(Figure 28).

Hawthorn leafminer

In 1992, adults emerged before sampling began on 11 May, which was also
when the largest adult population count of 1071 :84 adults per trap was observed.
No adults were trapped after 22 June (Figure 3A).

In 1993, traps were deployed on 28 April and adults were first captured on
11 May. A single generation was observed with peak emergence occurring on 18
May (1100150 adults/ trap) and no adults were trapped after 29 June (Figure 3B),

resulting in one generation.

Larval Distribution and Abundance

Alder leafminer

During the first generation of 1993, larval distribution was found to be non-
random as the hypothesis of a Poisson distribution was rejected (P< 0.001; X2 = 1857;
d. f. =3). The hypothesis of a negative binomial distribution was also rejected
(P=0.01; X2: 15; d. f. =10; k=0.082). The overall variance of the number of larvae

per leaf increased at a faster rate than the mean number of larvae per leaf

19

 

1400
1200
1000
800 '
600 '
400 '
200 ’

Adults I Trap

 

 

 

 

1400
1200 ' B
1000 '
800 '
600 '
400 '
200 '

May June July Aug

Adults I Trap

     

 

 

 

Figure 3: Seasonal henolo of d lt h ' '
and 1993 (B). p gy a u awthorn leafminer 1n 1992 (A)

20

(Y= 0.4e16x; r2 =0.76) (Figure 4A) and the variance-to-mean ratio was 15. Both
results suggest a clumped distribution. The d-statistic (d = 171) also indicated a
highly clumped larval distribution. However, Green’s index (GI =0.008) suggested
that larval distribution approached random.

During the first generation, 81% of all leaves were uninfested (n = 1768).
One larvae per leaf was the most frequently observed class among infested leaves
(n = 109) and up to 80 larvae per leaf were observed (Figure 5A). There was 1: 14
larvae per leaf (Figure 5A).

On the second sample date (29 June) larval distribution, over all leaves
(mature and immature combined) and over immature leaves only, was non-random
and the hypothesis of a Poisson distribution was rejected (P< 0.001; X2 =3954; d. fi =3
and P< 0.001; X2=2264; d. fi =4, respectively). The hypothesis of a negative
binomial distribution was also rejected (P< 0.001; X2 =36; d. )5 = 9; k =0.0364 and
P< 0.001; X2 =29; (Lfi =9; k=0.199, respectively). Both the variance-to-mean ratio
and the d-statistic, however, indicated a highly clumped larval distribution for all
leaves (ratio =10; d = 156), as well as for immature leaves only (ratio = 9; d = 89). As
for the first sample date, Green’s index indicated that the larval distribution was
near random for mature and immature leaves combined (GI =0.006), as well as for
immature leaves only (GI = 0.005).

Considering immature and mature leaves combined, only 10% of 2374 were
infested; 25% of 1000 immature leaves contained larvae. Among the infested
leaves, one larvae per leaf was the most frequently observed class (n =44), and as
many as 29 larvae per leaf was observed (Figure 5B). There were 1 i 6 larvae per
leaf for all leaves considered (mature and immature combined), but 2.4.- 13 larvae
per leaf for immature leaves only.

Similar trends in larval distribution occurred on the third sample date (13

July). The hypothesis of a Poisson distribution was again rejected using Chi-Square

21

 

Variance

 

 

 

 

  
  

 

 

 

 

 

 

 

60
50 ' B
a)
o 40 .
C
,9 30-
3 .
20 '
> . r2=0.62
1" ' v-2.se°-5X
o .
0 2 4 6 8 1O
3 c
o
o 2’
C
to
E
> 1 ' r2-0.45
0 Y-O.Be°-°"
O . .
O 1 2 3

Mean Number of Larvae/Leaf

Figure 4: Mean versus variance for the number of first generation larvae per leaf in
1993: (A) alder leafminer, (B) birch leafminer, and (C) hawthorn leafminer.

22

Mature Leaves
- Immature Leaves

2000 A

 

1500 '

1000 P

500 ’

Frequency

 

 

012345678910>10

 

3000
2500 - B
2000 -

8001:
640 -
480 -
320
160 -

Frequency

 

 

 

012345678910>10

C

 

3000
2500 r
2000 '

3001
240 .
180 ~
120 ~
so

W

Frequency

 

 

 

012345678910>10

Larvae / Leaf

Figure 3: Distribution and abundance of alder leafminer larvae er leaf: (A) first
generation (2 June, 1993), (B) second generation (29 June, 1993 and (C) second
generation 13 July, 1993).

23

analysis when tested against the observed larval distribution over all leaves

(P< 0.001; X2=9709; d. f. =5), and over immature leaves only (P< 0.001; X2=7186;

d. f. = 12). The negative binomial distribution was also rejected for all leaves and for
immature leaves only (P< 0.001; X2=67; d. f. =7; k=0.0178 and P <0.001; X2=49;

d. f. = 8; k= 0.1604, respectively). However, the variance-to-mean ratio indicated that
larvae were highly clumped over all leaves (ratio =35) and over immature leaves
only (ratio =28). In addition, the d-statistic indicated that larvae were highly
clumped over all leaves (d =374) and over immature leaves only (d = 128). Again,
Green’s index indicated that larvae were distributed near randomly over all leaves
(GI = 0.009), as well as over immature leaves only (GI = 0.007).

Uninfested leaves were the most frequently observed class when
considering immature and mature leaves together (n =2657, 93%). When
considering immature leaves only (n =451), forty seven percent were infested.
Three and four larvae per leaf were most frequently observed class (n = 17 each),
and 98 larvae per leaf was the maximum (Figure 5C). The mean number of larvae
per leaf for all leaves combined was 1 :49 and for immature leaves only, the mean

was 9:245.

Birch leaflniner

During the first generation of 1993, larval distribution was not random
(P< 0.001; X7=4862, d. f. =6). The negative binomial distribution was also rejected
(P< 0.001; X2 = 126, d. f. = 15 ; k =0.3178). However, the variance-to-mean ratio was
5.3, with the variance increasing at a faster rate than the mean number of larvae per
leaf (Y= 2.6e05x; r2=0.62;) (Figure 4B), and the d-statistic was 72. Each of these
results indicates a highly clumped larval distribution. However, Green’s index
suggested that larvae were distributed randomly (GI = 0.001).

24

About half (49%) of all leaves were infested (n= 1548), with one larvae per
leaf being the most frequently observed class. The mean number of larvae per leaf
was 2: 12, but up to 23 larvae per leaf were observed in a leaf.

Larvae were also not distributed randomly during the second generation
(P< 0.001; X2 =3243, d. )5 =3) when considering mature and immature leaves
combined, and (P<0.001; X2= 1787, dfi =11) when considering immature leaves
only. The negative binomial distribution was also rejected for all leaves (P< 0.001;
X2=60, (Lfi =4), and for immature leaves only (P<0.001; X2=39, (if =4). The
variance-to-mean ratio indicated a clumped distribution over all leaves (25) and
over immature leaves only (16). In addition, the d-statistic indicated a clumped
distribution for all leaves (d =285) and for immature leaves only (d =65). In
contrast, Green’s index indicated a more random larval distribution over all leaves
(0.011), and over immature leaves only (0.007).

During the second generation, 95% of the leaves, immature and mature
combined, were not colonized (n =2458). However, when considering immature
leaves only, 46% were infested, and as many as 56 larvae were observed in one leaf
(Figure 6B). When considering all leaves, there were I: 20 larvae per leaf, but
9: 151 when considering immature leaves only. Even though third generation

adults were present in August, no colonized leaves were observed.

Hawthorn leafminer

In 1993, larval distribution was not random as the hypothesis of a Poisson
distribution was rejected (P< 0.001; X2=384; ctfi =5). The hypothesis that larvae
were distributed according to the negative binomial model was also rejected
(P< 0.001; X2 =51; d. f. =6; k= 1.68). The variance-to-mean ratio was 1.49, with the
overall variance increasing faster than the mean (Y= 0.8e0-6X; r2=0.45;), suggesting a

degree of clumping (Figure 4C). The d-distribution statistic equalled 18, also

25

Mature Leaves
- Immature Leaves

1000
900 - A
800 -

700 -
eoo -
soo -
4oo -
300
200 .

100 '
0

 

Frequency

 

 

 

012345678910>10

 

3000
2500 ~ , B
2000 -
1501:
120
so -
60 ~
so ~

Frequency

 

 

 

012345678910>10

C

 

3000
- 2500 '
2000 ’

50 x
40
30
20 '
10 '

Frequency

 

 

 

 

012345678910>10

Larvae / Leaf

Figure 6: Distribution and abundance of birch leafminer larvae er leaf: (A) first
generation, (B) second generation, and (C) third generation (1 3).

26

suggesting clumping. However, Green’s index was near zero (0.0001), suggesting
that larval distribution was near random.

More infested (n = 1872) than uninfested (n = 1241) leaves were observed,
with the number of larvae per leaf ranging from 1 to 8. One larvae per leaf was the
most frequently observed class among infested leaves (n =793) (Figure 7). The

mean number of larvae per leaf was 1: 2.

Physical and Chemical Comparisons of Infested and Uninfested Leaves

Leaf Toughness

Alder: On the first sample date in 1992 (2 June), during peak oviposition,
by first generation alder leafminer, there was no difference in the toughness of
infested and uninfested leaves (t=0.36; d. jE =6,6; P=O.72) (Figure 8A). Weekly
samples revealed that leaf toughness increased over time, with infested and
uninfested leaves remaining similar. Between 2 June and 17 June, the toughness of
infested alder leaves increased from 217: 9 to 247 z 13 g, while toughness of
uninfested leaves increased from 213 i 8 to 253 t 16 g (Figure 8A).

In 1993, there were no differences in the toughness of infested and
uninfested leaves (t=-0.31; dfi = 5,5; P=O.76) (Figure 9A). Toughness of infested
and uninfested leaves again increased over time with no significant differences
recorded between them. Between 12 May and 2 July, the toughness of infested
leaves increased from 172: 10 to 239132 g, while the toughness of uninfested leaves

increased from 176: 7 to 229127 g (Figure 9A).

27

 

2000

1500 F

1 000

500

Frequency

 

 

 

012345678910

Larvae / Leaf

Figure 7: Distribution and abundance of hawthorn leafminer larvae per leaf (1993).

28

Uniniested

 

 

 

 

 

 

Amy mwOCEDDOF h“0|_

   
 

 

3v mmoccmsom. ado..—

 

/////////

C

 

 

 

w o

300
00

3v wmoccmnaF 30.—

J

29 May

Date

With peak

.ng 1992: 3A) EuroPean
Eon
iICh (B). and

5 dun

tlon alder (A),

“5

re 8: Toughness of infested and uninfested leave
B) paper birch, and (C) hawthorn. Sampli

$5.“,
OVI

»(

'position and larval development of first genera

hawthorn (C) 16mm“.

 

Infested

 

 

 

 

500
....A

300'

200'
100‘
0

5-12 5-20 5-26 6-9 6-16 7-2

Leaf Toughness (g)

 

 

 

 

 

 

 

 

 

 

500
400 r B

300 '

200'
100- %l%r%
/ /

5-13 5-21 5-27 6-10 6-18

 

 

Leaf Toughness (g)

 

 

500
400 ' C
300 '
200 ' .
we a

0
5-19 5-28 6-11

Date

 

 

Leaf Toughness (g)

 

 

 

 

 

Figure 9: Toughness of infested and uninfested leaves during 1993: 3A) European
al er, (B) paper birch, and (C) hawthorn. Sampling dates corresgon with peak
oviposition and larval development of first generation alder (A), irch (B), and

hawthorn (C) leafminer.

30

Birch: Based on the first sample date of 1992 (22 May), during the period
of peak oviposition by first generation birch leafminer, infested and uninfested
leaves were not different in toughness (t=0.77; d.fi =7,7; P=0.45) (Figure 8B). Leaf
toughness increased over time with infested and uninfested leaves remaining
similar. Between 22 May and 10 June, the toughness of infested birch leaves
increased from 141 : 6 to 229: 8 g, while uninfested leaves increased from 134: 7 to
243:5 g (Figure 8B).

In 1993, there was also no difference in the toughness of infested and
uninfested leaves (t =-0.83; d. f. =3,3; P=0.44) (Figure 9B). Toughness of infested
and uninfested leaves again increased over time with no significant differences being
recorded. Between 13 May and 18 June, the toughness of infested and uninfested
leaves increased from 126:5 to 238:22 g, and from 135 :9 to 243:23 g,
respectively (Figure 9B).

Hawthorn: Based on the first sample date of 1992 (29 May), during the
period of peak oviposition by hawthorn leafminer, there was no difference in
toughness of infested and uninfested leaves (t=-1.70; d. f. =7,7; P=0.11) (Figure 8C).
Weekly samples revealed that leaf toughness increased over time with infested and
uninfested leaves remaining similar. Between 29 May and 8 June, the toughness of
infested hawthorn leaves increased from 154: 8 to 217: 12 g while uninfested leaves
increased from 167:2 to 206:6 g (Figure 8C).

In 1993, there was also no difference in toughness of infested and uninfested
leaves (t=-0.04; €1.15 =7,7; P=0.97) (Figure 9C). Toughness of infested and
uninfested leaves again increased over the sampling period with no significant
differences recorded. Toughness of infested and uninfested leaves was equal (171 g)
on 19 May (first sample date). By 21 June, the toughness of infested leaves had

increased to 364:5 g, while uninfested leaves increased to 354: 5 g (Figure 9C).

31

Leaf Area

Alder: Leaves chosen as oviposition sites by first generation alder leafminer
were smaller than non-selected leaves (t=-5.53; d. f =6,6; P= < 0.001) (Figure 10A).
On the first sampling date of 1992 (2 June), corresponding with peak oviposition,
the area of infested leaves was 14.06: 0.78 cmz, while the area of uninfested leaves
was 19.78:0.68 cmz. By 17 June, infested and uninfested leaves were of equal size
(21.57: 1.60 and 21.33 : 0.84 cmz, respectively) (Figure 10A).

In contrast to 1992, leaves selected for oviposition in 1993 by first
generation females were larger than those not selected (t =3.37; d. f =5,5; P= 0.0071)
(Figure 11A). On 12 May (first sample date), the area of infested leaves was
17.25 : 0.94 cmz, nearly 4 cm2 larger than that of uninfested leaves (13.01 : 0.83 cmz)
(Figure 11A). Also in contrast to 1992, the area of fully-expanded infested leaves
was greater than that of uninfested leaves. By 2 July, infested leaves had reached an
area of 40.83 : 3.35 cmz, approximately 12 cm2 larger than that of uninfested leaves

(28.55 :2.68 cm?) (t=2.89, d. fi =5,5; P=0.0162) (Figure 11A).

Birch: In 1992 and 1993, the size of leaves chosen as oviposition sites by
first generation birch leafminer were not different than those not selected (Figure

1o&11).

Hawthorn: On the first sample date (29 May), leaves infested and
uninfested by hawthorn leafminer were equal in size (4.14 : 0.16, 4.14 : 0.08 cmz,
respectively) (Figure 10C). However, on the last sample date (8 June), infested
leaves were slightly larger than uninfested leaves (6.14 : 0.23 and 5.53 : 0.21 cmz,
respectively). However, differences were marginally significant (t = 1.9601; d. f. =7,7;

P=0.0702) (Figure 10C).

32

Ullinfested
A

 

 

   

 

AwEOV G¢h< hMOJ

 

 

  

 

AwEov ao..< 3m.—

 

        

   

 

 

000000

ANEOV fi¢h< ~50:—

ropean alder,

ak
(B), and

ue
E c
)m.fl
“Mb
\h
«AMA
90(
90....
15¢
gem
nna
.non
UCO
dad.
Sta
ear
Vde
a n
knee
dumna
mph
am
.mS..m
o t
mmm
dom
mmw
dad
whv
30¢»...
mfldh
Mmmm
mhdh
Nun)
Iwnmw
mem
namo
unmm
fimom

 

Infested Uninfested

 

 

 

 

so

...A

30b

20'

10’
0

Leaf Area (cmz)

 

 

 

 

 

 

 

 

 

 

 

 

 

5-12 5-20 5-26 8-9 6-16 7-2
50
.. . B
30 '

Leaf Area (cmz)

 

 

 

 

 

 

5-27

   

 

 

 

 

50
«A C
40-
E
O
V
a 30
0
h
< 20-
h
8
10' .
_l y y V
o I%% l% . /,, z
5-19 5-28 6~11 6-21
Date

F1 11: Area of infested and uninfested leaves during 1993: (A) European alder,
(B paper birch, and (C) hawthorn. Sampling dates correspond with peak
ov1posrtion and larval development of first generation alder (A), birc (B), and
hawthorn (C) leafminer.

34

As in 1992, infested and uninfested leaves were similar in size (4.44: 0.10
and 4.11 :01] cm2, respectively) (Figure 11C). By 21 June (the last sample date),
infested leaves were larger than uninfested leaves (9.46: 0.15 and 7.62: 0.37 cmz,

respectively) (t= 4.59; if =7,7; P< 0.001) (Figure 11C).

Specific Leaf Mass

Alder: In 1992, leaves chosen as oviposition sites by first generation alder
leafminer had a slightly higher specific mass than leaves not selected (t = 1.62;
d. f =6,6; P=O.13) (Figure 12).

In 1993, during peak oviposition on 12 May, no differences in specific leaf
mass were recorded between infested and uninfested leaves (I =0.24; d. f. =5,5;
P=0.8137) (Figure 13A). Over time however, infested leaves acheived a higher

specific leaf mass than uninfested leaves.

Birch: In 1992, during peak oviposition by the first generation birch
leafminer, infested leaves had a slightly higher specific mass than non-infested
leaves (t =3.34; d. f. =7,7; P=0.0049) (Figure 12B). The difference was magnified by
June 1 (t=4.78; d. f. =7,7; P<0.001), but disappeared by 10 June (t=0.16; d. f. =7,7;
P=0.8716) (Figure 12B).

During peak oviposition by the first generation birch leafminer in 1993 (13
May), there was no difference in the specific leaf mass of infested and uninfested
leaves (t= 1.88; d. f. =3,3; P=O.1096) (Figure 13B). Both infested and uninfested
leaves increased in specific leaf mass over time, with no consistent differences

observed between them.

 

 

Specific Leaf Mass (g/m’)

 

 

 

 

 

Specific Leaf Mass (glm’)

 

Leaf Mass (g/m’)

 

 

Specific

 

Figure 12: Specific leaf mass of infested andWth uninfested leav ves during1992: (A)
Eur uropean alder ,i(B) paper rbirch, and (C)ha wth.orn Sampling datesc orrs esonp
withp eako ovipost nna ndlarval dev velopmen ntof first ge er rioat nralde (A),

bir rchp(B) an hawthorn (C) leafnune

 

 

 

 

 

uring 1993: (A)
awthorn.
lopment of first

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

€53 new: 36.. £20on

 

 

 

 

 

 

 

 

 

 

£53 «an: 36.. 05025

 

   
 

 

  

 

 

 

 

 

€53 new: 36.. 2:0on

fig“ m %///////////////////////m
A mB émC
mmmmmmmmmawmmmmwam

Date

sition and larval deve

pyn'fera Marsh), and (C) h
orn (C) leafminer.

0

ch (B), and ha

nd w1th peak

ES}

cific leaf mass of infested and uninfested leaves d
pa
3‘38

, (B) paper birch (Betula

33

r
dates corres
alder (A),

Sampling
generation

S

European al

Figure 13:

37

Hawthorn: In 1992 during peak oviposition by hawthorn leafminer, infested
leaves were slightly higher in specific mass than uninfested leaves (t =2.96; d. f =7,7;
P=0.0103) (Figure 12C). But by the last sample date (8 July) infested and
uninfested leaves were similar in specific leaf mass (t = -0.3265; d. f = 7,7; P= 0.7489)
(Figure 12C).

In 1993, during oviposition (19 May), no differences in specific leaf mass
were observed between infested and uninfested leaves (t=0.00; d. j =7,7; P= 1.00)
(0.50: 0.01 g/mz) (Figure 13C). However, as leaves matured, infested leaves

acheived a higher specific leaf mass.

Nitrogen

Alder: During peak oviposition (2 June) by the alder leafminer, no
significant differences were observed in the nitrogen concentration of infested and
uninfested leaves (t=0.75; d.fi = 15,11; P=0.4586) (Figure 14A). By the second
sample date (9 June), infested leaves had a lower nitrogen concentration than
uninfested leaves (t=-4.12; d. fi = 13,15; P= < 0.001) (Figure 14A), a difference that

was maintained through the final sample date (17 June).

Birch: During peak oviposition by first generation birch leafminer (22
May), no differences were observed between infested and uninfested leaves (t = 1.26;
d. f. = 15,15; P=0.2180) (Figure 14B). Infested and uninfested leaves decreased
equally in nitrogen content between 1 June and 10 June (t =0.19; d. f. = 15,15 ;
P=0.8523 and t=0.1448; d. f =15,15; P=0.89, respectively) (Figure 14B).

Hawthorn: No differences in nitrogen concentration were observed

between leaves infested and those not infested by hawthorn leafminer (29 May)

38

 

 

 

              

m m % //,, , 2
WA // NB a%/////m C ////////////
.... ,I-

   

/

  
 
  

I

n n -
555555555555

coooEz .x. comoEz .x.

V.
a
M
2
2

 

 

 

 

 

E] Infested
2 J

 

 

 

555555

coaoiz .x.

 

ng dates

) hawthorn. Sampli

D ate
(C
peak oviposition and larval deve opment of first generation alder

pean alder, (B) paper birch, and

nd with

P0

Figure 14: Percent nitrogen concentration of infested and uninfested leaves during
(A), b1rch (B), and hawthorn (C) leafminer.

1992: (A) Euro

corres

39

(t= 1.01; d.fi =9,7; P=0.3294) (Figure 14C). Nitrogen content of hawthorn leaves
also decreased overtime, with infested leaves having a lower nitrogen concentration

than uninfested leaves (t = -3.84; df = 15,14; P< 0.001) (Figure 14C).

Phosphorous

Alder: During peak oviposition (2 June) by the first generation alder
leafminer, infested leaves were lower in percent phosphorous than uninfested leaves
(0.16: 0.01 and 0.18:0.01%, respectively) (t=-3.15; d.fi =15,11; P< 0.01) (Figure
15A). On 9 June and 17 June, percent phosphorous concentration of infested leaves
remained lower than uninfested leaves (0.13 : 0.01, 0.17 : 0.01% and 0.14 : 0.01,
0.15 : 0.01%, respectively), (t = -5.96, d. )5 =13,15;P< 0.001 and t = -1.80, d. f =15,15;
P =0.0821) (Figure 15A).

Birch: Infested birch leaves had a higher phosphorous concentration than
uninfested leaves during peak oviposition by first generation birch leafminer (22
May) (0.39: 0.01 and 0.35 : 0.01%, respectively) (t =2.58; (if = 15,15 ; P< 0.05)
(Figure 15B). Infested leaves maintained a slightly higher phosphorous
concentration than uninfested leaves on each sample date: 1 June (0.28 : 0.00 and
0.25 : 0.01%, respectively) (t =3.72; d. fi = 15,15; P< 0.001), and 10 June (0.22: 0.01
and 0.21: 0.01%, respectively) (t= 1.63; d. f =15,15; P=O.1137) (Figure 15B).

Hawthorn: No differences in phosphorous concentration were observed
between infested and uninfested hawthorn leaves during peak oviposition by
hawthorn leafminer (29 May) (0.28:0.01 and 0.28:0.01%, respectively) (t=-0.47;
d. f. =9,7; P=0.6474) (Figure 15C). Infested and uninfested leaves remained similar

40
:1 Infested Uninfested
0.50
. A

0.30 '

 

0.20

0.10 '

% Phosphorous

 

 

0.00

 

 

0.50
0.40 ' B
0.30 r
0.20 '

0.10 r

% Phosphorous

 

 

 

 

 

 

0.00 ' ‘
22 May 1 June 10 June

 

0.50

o... . C

% Phosphorous

 

 

 

29 May 8 June

Date

Figure 15: Percent phosphorous concentration of infested and uninfested leaves
during 1992: (A) European alder, (B) paper birch, and (C) hawthorn. Sampling
dates correspond with peak oviposition and larval development of first generation
alder (A), b1rch (B), and hawthorn (C) leafminer.

41

on the second sample date (0.25 :0.01 and 0.26:0.01, respectively) (t=-1.01;
d. 15 = 14,15; P=0.3225) (Figure 15C).

DISCUSSION

Adult Abundance and Phenology

First generation adults of each species typically begin emergence in late
April to mid-May as bud break occurs (Slingerland 1905, Parrott & Fulton 1915,
Friend 1933, Hart et al. 1991, Herms unpublished). In 1992 and 1993, a large first
generation of birch leafminer, begining around 11 May, was followed by a small
second generation. In 1992, a few second generation adults were observed, but no
adults were trapped. The relatively cool temperatures experienced throughout the
summer of 1992 may have been a reason why no birch leafminer adults were
trapped after 14 July. Adult emergence typically occurs until mid-August to early
September (Herms, unpublished data). Instead of emerging as adults, individuals
may have diapaused, remaining in the soil until the following spring. The relatively
cool temperatures may have prevented second and third generation individuals from
completing development before cooler fall temperatures set in and leaf abscission
occurred.

In 1993, a large first generation in May was followed by a smaller second
and third generation in July and August respectively. Friend (1931) reported that
during late summer and early fall, an increasing proportion of pupae enter diapause.
In Newfoundland in 1972, two generations of birch leafminer was reported with part

of the population completing a third generation, a pattern similar to that observed

42

in Europe (Jones and Raske 1976). Three generations were also observed in
Connecticut (Friend 1931) and Quebec (Jones and Raske 1976).

The first generation of birch leafminer is largest possibly because
emergence occurs when all leaves on the host plant are immature, which is the
preferred oviposition site (Friend 1933), and possibly the required food source for
larvae. Larvae may not be able to complete development in mature leaves.

If, as the season progresses, the availability of immature leaves decreases,
some individuals may enter premature extra-long diapause (sensu Hanski 1988) to
avoid placing their offspring in a resource-limited environment. If all first
generation individuals emerged as second generation adults, and had fewer leaves
available as oviposition sites, then resource competition among larvae could be
intense, resulting in the evolution of a density-dependent "bet hedging" diapause
strategy. Under this scenario, the majority of first generation individuals would
enter diapause and by-pass a breeding opportunity, only to emerge the following
spring when a full flush of immature leaves is once again available, therefore placing
their offspring in a potentially less competitive environment. On the other hand,
some individuals emerge as second generation adults, take advantage of the
breeding opportunity, but risk placing their offspring in a potentially more
competitive environment.

In 1992 and 1993, alder leafminer adults also began emerging around 11
May. Relatively cool spring temperatures in 1992 and 1993 may have prolonged
adult emergence of some alder leafminer individuals, causing each generation to
become less distinct from one another. The cool spring temperatures may have
been a reason why the abundance of alder leafminer adults emerging in May and
June was less than that which emerged in July and August, causing the generations
to overlap and resulting in the second and/or third generation to appear larger than

the first. Head capsule measurements of larvae, taken weekly in order to determine

43

instar for other experiments, revealed the presence of a small third generation in
August. After second generation larvae began reaching the fourth and fifth instar,
first instar individuals were again more commonly observed.

Slingerland (1905) reported that alder leafminer adults were present
throughout the summer, but was unable to distinguish the number of generations
because the broods overlapped. Slingerland suggested that two to three generations
were present. In 1991, Hart et al. reported three distinct generations of alder
leafminer (late May, early July, and mid-to-late August) in Iowa and Michigan. The
Michigan population studied by Hart et al. (1991) was the same used for this
research. This pattern would be consistent with second generation alder leafminer
expressing a density-independent "bet-hedging" diapause strategy as described for
the second generation of birch leafminer. The large first and second generation
followed by a small third generation suggests that only the second generation may
"hedge their bet". Some individuals may enter diapause, by-passing a breeding
opportunity only to emerge the following spring. Individuals that emerge as third
generation adults take advantage of the breeding opportunity but their offspring run
the risk of not being able to complete development before the onset of cool fall
temperatures and leaf abscission.

In both 1992 and 1993, the hawthorn leafminer was observed to have only
one generation in the spring. Each year, adult emergence began on approximately
11 May and adults were not observed after the end of June. The population was
large relative to that of the alder and birch leafminer, with over 1000 adults
collected per trap per week during peak emergence in 1992 and 1993. These results
are consistent with those of Hamilton (1950) who reported Profenusa canadensis as
being a univoltine insect with emergence beginning in late April to early May,
shortly after bud-break, and ending in early to mid June. If immature leaves are

required for oviposition and for larval development, as seems to be the case for the

44

alder and birch leafminer, then hawthorn leafminer may be univoltine because its

host plant does not produce immature leaves throughout the summer.

Larval Distribution and Abundance

During both 1992 and 1993, for each generation of all three species, larvae
were not distributed randomly or according to the negative binomial distribution.
However, the variance-to-mean ratio, and the d-statistic suggested that larval
distribution was clumped in all cases. Meanwhile, Green’s index suggested that
larval distribution was near random in all cases.

The majority of the alder and birch leaves sampled were uninfested even
though numerous larvae were observed per leaf, and intraspecific competition
occurs among larvae (see chapter 3). During the first generation, the distributions
of both the birch and alder leafminer were highly clumped. Slingerland (1905)
reported 15-20 alder leafminer larvae per leaf. In this study, as many as 80 alder.
leafminer larvae and 23 birch leafminer larvae were observed in a single leaf during
the first generation. In later generations, up to 98 alder leafminer larvae and 56
birch leafminer larvae per leaf were observed. The mean and variance for the
number of larvae per leaf also increased with each generation. The number of alder
leafminer larvae per leaf increased from 1:14 during the first generation, to 2:13
during the second generation, and finally to 9: 245 larvae per leaf during the third
generation. As for birch leafminer, the number of larvae per leaf during the first
generation was 2: 12 and increased to 9:151 during the second generation; to few
larvae were observed during the third generation for analysis. The number of larvae
per leaf may have increased due to a decrease in the number of immature leaves

available as the season progressed.

45

Bultman and Faeth (1986b) reported the distribution of five species of
leafminers on oak (Quercus emmyi) to be clumped. The five species studied were
Bucculatn'x cerina, Cameraria sp. nov., Stigmella sp., and Stilboszlr quadricustatella.
The non-random distribution of Cameran'a sp. larvae resulted because fewer leaves
were mined and because more leaves had multiple mines than expected by chance
(Faeth 1991). Faeth (1991) concluded that the clumped distribution of larvae was
due to leaf size with larvae being more clumped on small leaves. This preference
may be the result of females preferring to oviposit in sun leaves which tend to be
smaller than shade leaves in Emory oak trees (Faeth 1991). However, higher rates
of premature leaf abscission of sun leaves may contribute to lower over-all densities
of larvae in sun leaves relative to shade leaves (Faeth 1991).

Hawthorn leafminer larvae were more randomly distributed than were
alder and birch leafminer larvae. This may be because the smaller size of hawthorn
leaves accommodates the successful development of only one or two larvae,
compared to five to ten commonly accommodated by birch and alder (personal
observation). Fully expanded hawthorn leaves are approximately 10 cm2 in size
while birch and alder leaves are typically 20 and 40 cm2 in size, respectively. The
average leaf area consumed by a single hawthorn leafminer per leaf was
approximately 4 cm2, which is approximately fifty percent of the average leaf. On
the other hand, a single alder leafminer per leaf consumed approximately 3 cm2
(about 8% of the leaf) and a single birch leafminer consumed approximately 2 cm2
(10% of the leaf) before exiting the leaf to pupate in the soil. Thus, the number of
alder and birch leafminer larvae per leaf often exceeded that which could be
supported by the leaf, while hawthorn leafminer larvae tended not to exceed their
carrying capacity.

The over abundance of alder and birch leafminer per leaf may result in

intraspecific competition among larvae for the available food resource. This creates

46

the potential for selection for density dependent premature extra-long diapause
within the population (sensu Hanski 1988). The smaller population size of the
second and third generation of birch leafminer observed in this study and the
smaller size of the third generation of alder leafminer, reported by Hart et al. (1991)

are consistent with the occurrence of such a bet-hedging strategy.

Physical and Chemical Comparisons of Infested and Uninfested Leaves

During the first generation there were no consistent differences in the
toughness, size, and specific mass of immature alder, birch, and hawthorn leaves
selected for oviposition by alder, birch, and hawthorn leafminer and those which did
not receive any eggs. Differences in the size and specific mass of selected and non-
selected alder leaves varied from year to year. In 1992, infested leaves were smaller
in size and were higher in specific leaf mass. However, in 1993 infested leaves were
larger in size and there was no difference in specific leaf mass. In 1992, birch leaves
of a slightly higher specific mass were preferred, but this was not the case in 1993.

For all three species, the nitrogen concentration of leaves selected by
females were similar to leaves not receiving eggs. Since larval distribution tends to
be aggregated, this suggest that females are selecting leaves as oviposition sites
based on some other variable and not nitrogen concentration, even though nitrogen
is generally a limiting nutrient for larval development (Mattson 1980).

Differences in the phosphorous concentration of infested and uninfested
leaves were also inconsistent between species during the first generation. Infested
alder leaves were lower in phosphorous than leaves not receiving eggs. However,

birch leaves selected for oviposition had a higher phosphorous concentration than

47

leaves which did not receive eggs. Hawthorn leaves receiving eggs and those not
receiving eggs were similar in phosphorous content.

In summary, the distribution of larvae of all three species were highly
aggregated. However, there were no consistent differences detected between
infested and uninfested leaves in the physical (toughness, area, and specific leaf
mass) and chemical (nitrogen and phosphorous) variables measured in this study.
Due to these inconsistencies, it can not be concluded that alder, birch, and hawthorn
leafminers selecte leaves for oviposition based on the traits measured in this study.

Hamilton (1943), suggested that these leafminer species seemed to prefer to
oviposit in sun leaves rather than shade leaves. Adults of the red oak leafminer
(Profenusa alumna MacGillivray) also deposit their eggs in the upper sunlit portion
of the trees (Labonte and Lipovsky 1984). Sun leaves may be preferred for
oviposition because larvae may be able to develop faster in warmer temperatures.
In addition, sun leaves are thicker (Kozlowski et al. 1991) therefore, they may
provide more tissue per unit area for consumption by larvae. However, oviposition

in relation to light intensity was not quantified in this study.

CHAPTER TWO ,

INTERSPECIFIC VARIATION IN VOLTINISM PATTERNS
RESULTS FROM INTERSPECIFIC DIFFERENCES IN HOST
PHENOLOGY

Feeny (1970) reported that the number of spring feeding Lepidoptera on
oak trees, such as the winter moth (Operophtera brumata L.), exceeded the number
of late season feeders. Feeny proposed that this difference was due to seasonal
changes in the texture and chemical composition of the leaves. Immature leaves are
less tough, and contain lower tannin concentrations than older leaves (Feeny 1970).
Furthermore, concentrations of nitrogen, a limiting nutrient to herbivores, is highest
in young tissues (Mattson 1980). Seasonal changes in leaves may result in mature
leaves becoming a less suitable food resource for larvae, and may select for early
season feeding (Feeny 1970).

Differences in leaf characteristics of mature and immature leaves may also
play a role in the voltinism patterns of leafminers. Harbaugh et al. (1983) reported
that damaged caused by Lin'omyza trifolii Burgess, a leafminer on Chrysanthemum,
increased linearly as leaf nitrogen increased from 2.2 to 4.0%. Mature leaves may
also posses noxious or digestibility reducing compounds such as alkaloids and
tannins (Faeth et al. 1981). Ayres and MacLean (1987) identified several leaf
characteristics that change with age, which affect the growth of a geometrid larvae
(Epim'ta autumata). As leaves aged, water, nitrogen and phosphorous
concentrations declined, toughness and specific mass increased and larval
performance decreased. They proposed that phenological constraints imposed on

the larvae by leaf maturation should influence the evolution of larval phenology,
4 8

49

selecting for early season development. A similar relationship between leaf and
insect phenologies was documented by Merle (1988) for the green oak leafroller,
Tomix vin'dana (Lepidoptera: Tortricidae) and various species of oak.

Such phenological constraints have also been reported by Auerbach and
Simberloff (1984) for two leafmining species, Eriocraniella sp. and Dysen'ocranina sp.
(Lepidoptera: Eriocraniidae). Both are phenologically restricted, feeding only on
first-flush leaves. In addition, Faeth et al. (1981) stated that the densities of active
mines of several leafmining species vary seasonally because of their voltinism
patterns and because feeding is restricted to a particular leaf stage. The density of
two other leafmining species, Acrocerwps sp. and Neurobathra strigifinitelia, were
highly correlated with new leaf production (Auerbach and Simberloff 1984). They
suggested that the abundance of both species is often limited by the availability of
immature leaves. F aeth et al. (1981) suggested that both the number of
phytophagous species present and their density would decreases seasonally due to
an increase in leaf toughness and tannins, and a decrease in nitrogen.

Since alder, birch, and hawthorn leafminer females prefer immature leaves
as oviposition sites (Parrott & Fulton 1915, Friend 1933, Wilson et al. 1982, Hart et
al. 1991) the evolution of their voltinism patterns may have been influenced by the
phenological pattern of new leaf production by their respective host plants. Alder
and birch trees display indeterminate growth, producing new foliage throughout the
summer, while hawthorn trees do not produce new foliage after the initial flush at
bud-break, and thus display determinate growth (personal observation). The
continued production of immature leaves, throughout the season may have allowed
alder and birch leafminer to evolve a multivoltine life cycle. The lack of immature
leaves after bud break may have constrained the hawthorn leafminer to one

generation.

OBJECTIVES

This research tested the following predictions relating host phenology
(availability and quality of immature foliage) to the evolution of voltinism patterns:
(1) adults oviposit only in immature leaves, (2) immature leaves are less tough and
more nutritous than mature leaves, and (3) age specific larval mortality increases as

leave mature.

METHODS AND MATERIALS

Leange Influences Oviposition Site Selection of Second and Third

Generation Alder and Birch Leafminer

Birch Leafminer

Infested leaves containing fifth instars were collected and brought to the
laboratory. These leaves were placed in a petri-dish with a plaster filled base
saturated with water to maintain humidity. They were then placed in an incubator
at 25 °C (18:6 lightzdark) until adults emerged. Ten female and five male adults
were confined to a potted birch seedling within a mesh bag, where females were
allowed to oviposit freely. The availability of immature and mature leaves was
manipulated manually by removing either mature or immature leaves resulting in
the following three treatments: (1) seedlings with only mature leaves, (2) seedlings
with only immature leaves, and (3) unmanipulated seedlings (mature and immature
leaves present). The use of these three treatments resulted in two oviposition

preference experiments. Comparison of treatments 1 and 2 represented a no-choice
50

51

experiment. Comparison of the number of eggs laid in mature and immature
foliage in treatment 3 represented a choice experiment.

Seedlings were arranged in a randomized complete block design with four
blocks, each containing one seedling from each treatment. The number of eggs per
leaf, and the number of immature and mature leaves per seedling were quantified.
For each experiment (no-choice and choice) the total number of eggs laid on
mature and immature leaves were compared by AN OVA. Data were square root
transformed (Y+ 1/2) because eggs were distributed non-randomly among leaves
(Little and Hills 1978). Non-transformed data are presented in all figures. SAS was

used for all analyses.

Alder Leafminer

Second generation adults were obtained as above. Ten adult females were
confined to a branch inside a mesh bag. Three branches were selected from each of
five tress. Branches similar in size and in the number of immature and mature
leaves were selected for this experiment. The availability of immature and mature
leaves were manually manipulated by removing either mature or immature leaves,
resulting in three treatments: 1) branches with mature leaves only, (2) immature
leaves only, and (3) both immature and mature leaves (unmanipulated branches).

The experiment was designed as a randomized complete block with trees
serving as blocks. The number of eggs per leaf, and the number of leaves per
branch were quanitified. The selection of these treatments resulted in both a choice
and a no-choice oviposition preference experiment as described for the birch

leafminer. Data were analyzed as described for birch leafminer.

Number of Eggs Per Leaf in Relation to Physical Leaf Characteristics

For each species, infested leaves were randomly sampled on a weekly basis
until eggs hatched. Ten infested leaves per tree were sampled from eight trees of all
three species. While in the field, leaves were placed in zip-lock plastic bags and
stored on ice in a cooler. Leaves were promptly transported to the laboratory.

The following variables were measured for all leaves: (1) number of eggs
per leaf were counted using a disecting microscope, (2) leaf toughness (g) using a
penetrometer, (3) leaf area (cmz) using a image analyzer, and (4) specific leaf mass
(g/mz). The relationships between the number of eggs per leaf and (1) leaf
toughness, (2) leaf area and (3) specific leaf mass were quantified using correlation

analysis.

Physical and Chemical Comparisons of Mature and Immature Leaves

In 1992 and 1993, during the first generation, fully expanded immature and
mature leaves (ten of each per tree) were randomly sampled from each of eight
alder (six in 1993), eight birch, and eight hawthorn trees for physical and chemical
analysis. Mature leaves were distinguished from immature leaves by their position
on a branch, size, color (darker green), and texture (tougher, thicker).
Corresponding with the period of peak oviposition of each generation, foliage was
sampled during mid-May, mid-July, and mid-August in order to determine the
difference in immature and mature leaves available to ovipositing females. The first
sample contained only immature leaves. For birch and alder, the second and third
samples contained both immature and mature leaves (except on the third sample

date in 1992 when only mature leaves were available). Because its determinate
52

53

growth pattern, on the last two sample dates only mature hawthorn leaves were
available for sampling.

While in the field, sampled leaves were placed in zip-lock plastic bags and
stored on ice in a cooler. leaves were promptly transported to the laboratory. For
all leaves, the following variables were measured using methods described in
Chapter One: (1) toughness, (2) leaf area, (3) dry weight, and (4) nitrogen and
phosphorous concentration. Results for each sample date were analyzed separately.

Means for mature and immature leaves were compared using a t-test.

Age Specific Larval Mortality In Relation to Leaf Maturity

Ten infested leaves were sampled weekly from eight trees of each species
throughout the first and second generation. The following variables were measured:
(1) number of larvae (or eggs) per leaf, (2) head capsule width of each insect to
determine instar, (3) larval status (dead or alive), (4) number of mature larvae
previously emerged from the leaf (determined by exit holes), (5) leaf toughness, (6)
leaf area and (7) area of leaf mined.

Data from all leaves were combined to determine percent mortality by
instar for both the first and second generations. Mortality rates in relation to leaf

toughness were examined quantitatively.

RESULTS

Leaf Age Influences Oviposition Site Selection
During the Second and Third Generations

Both alder and birch leafminer preferred to oviposit in immature leaves. In
both the no-choice and choice experiments, alder leafminers oviposited exclusively
in immature leaves (F =56.63; d. f. = 1,9; P<0.01 and F =2.63; d. fi = 1,9; P =0.18,
respectively) (Figure 16). Birch leafminer females also laid substantially more eggs
in immature than mature leaves in the no-choice experiment. In the choice
experiment, females oviposited exclusively in immature leaves (F = 13.39; df = 1,7;
P=0.04 and F =8.05; d. f =1,7; P=0.07, respectively) (Figure 17).

Alder and birch leafminer larval development in immature and mature
leaves could not be compared because females did not oviposit in mature leaves,

even when offered only mature leaves.

Number of Eggs Per Leaf In Relation to Physical Leaf Characteristics

For all three species, there was no relationship between the number of eggs
laid per leaf and leaf toughness (Figure 18) and leaf area (Figure 19).

For the alder leafminer, there was a significant positive, but weak,
correlation between the number of eggs per leaf and specific leaf mass (n =60;
Y= 0.33e520x; r2=0.18; P<0.01) (Figure 20A). For the birch and hawthorn
leafminers, there was no correlation between the number of eggs per leaf and

specific leaf mass (Figure 20B and 20C).
54

55

 

 

 

       

 

 

 

 

 

 

 

 

 

 

 

 

100
.c h- A
O 80
C
s eo- ——
(D

/
3 4° ' //
U)
U)
[U 20 -
o / / //
Immature Mature

100
f, eor B
t:
3 so -
m
\
a, 40 "
O)
5’ 2o- ‘1—

’ / //
Immature Mature
Leaf Type
F 16: Comparison of immature and mature European alder leaves as

ov1position sites by alder leafminer: (A) no-choice experiment, (B) choice
experiment.

56

 

200

Eggs / Seedling

 

 

 

 

 

200

 

 

Eggs / Seedling

 

 

Immature Mature

Leaf Type

Figure 17: Comparison of immature and mature paper birch leaves as oviposition
sites by birch leafminer: (A) no-choice experiment, (B) choice experiment.

Eggs I Leaf Eggs / Leaf

Eggs l Leaf

40

30'

20'

10

40

30'

20’

10'

30

20'

10’

57

 

 

 

300

 

 

 

 

200

300

 

 

 

Leaf Toughness (g)

 

 

 

300

{Eng 18: Relationshi between leaf toughness and the number of eggs per leaf:
u

ropean alder, (
with peak ovi sition o

leafminer in 993.

BE

aper birch, and (C) hawthorn. Sampling dates correspond
t generation alder (A), birch (B), and awthorn (C)

Eggs I Leaf Eggs / Leaf

Eggs l Leaf

58

 

 

 

 

 

 

 

 

 

 

 

 

 

 

40
30 '
o
20- ..
o
. o
10 ' 0 e .o o.
cove .00.: .
0 any" -
10 20 30 40
40
B
20 ’
o’ o
..’~°°..’.
10 00 . . :
.0 . . o
o 9 .
0 10 20 30 40
10
.. C
e o
6 - o e
e o o
4- coco e o
co... co
2- coco-coco
“so 0
o l j
0 2 4 6 8 10

Leaf Area (cm?)

Figure 19: Relationship of leaf area and the number of eggs per leaf: (A) Euro
r birch, and (C) hawthorn. Sam ling dates correspond with

alder, (B)a pa
1V1 3.

ition 0 first generation alder (A), birch

), and hawthorn (C) leafmmer 1n

59

 

 

 

 

 

 

 

 

 

 

 

 

 

40
A .
“5 30 r
0 r2- .18 o
_J 0
fl)
0)
U)
UJ 10
0
0.00 0.20 0.40 0.80 1.00
(5'2)
40
”a 30 *
0
_.I
\ 20 I.
g o ’o
O) 3 e
w .. a
. o
’ o
o ' F! ‘
0.00 0.20 0.40 0.60 1.00
(52)
10
- .. C
M
0 o
_l g L o.
\ o o oo
‘0 .
a 4 o coco
U) o...
“J 2 - a...
‘ on...
o I; l
0.00 0.20 0.40 0.60 1.00
(5'2)

Specific Leaf Mass (g/cmz)

re 20: Relationship between specific leaf mass and the number of efigs laid per

ltfiaf: (A) European alder
c(Irrespond with peak ov1positlon of first generation alder (A), birc (

hawthorn (C) leafminer in 1993.

, (B) paper birch, and (C) hawthorn. SamKlirlifi

ates
, and

Physical and Chemical Comparisons of Mature and Immature Leaves

Leaf Toughness

Alder: Mature leaves were always substantially tougher than immature
leaves. During peak oviposition of first generation females (22 May 1992), when
only immature leaves were available, leaf toughness was 187: 9 g (Figure 21A).
During the same period in 1993 (12 May), toughness of immature leaves was 173 : 6
g (Figure 22A).

In 1992 (20 July) and 1993 (14 July), when second generation leafminers
were ovipositing, mature leaves were tougher than immature leaves: 253 : 8 and
138:7 g, respectively in 1992 (t=-10.46; d. f. =7,7; P < 0.001) (Figure 21A), and
246: 11 and 141:9 g, respectively in 1993 (t=-7.41; d. j =5,5; P<0.001) (Figure
22A).

In 1992, the third sample was taken after leaf production had ceased (26
September), therefore only mature leaves were sampled. Mean toughness of these
leaves was 244: 11 g (Figure 21A). On 18 August 1993, which corresponded with
the third generation of the alder leafminer, mature leaves were much tougher than
immature leaves (241 : 11 and 150: 8 g, respectively) (t =-6.83; d. )3 =5,5; P< 0.001)
(Figure 22A).

Birch: During the first generation when only immature leaves were
available for ovipositing females, leaf toughness of immature leaves was 134 : 7 g in
1992 (22 May) (Figure 21B), and 136:6 g in 1993 ( 13 May) (Figure 22B).

During the second generation, immature leaves were much less tough than

mature leaves: 160:3 and 256:8 g, respectively on 21 July 1992 (t=-11.13; d. fi =7,7;

60

 

 

 

 

 

 

 

 

 

 

 

 

55555

3v mmoccoaoh Eon

 

 

   

  

 
     

 

 

 

 

o o o o 0 o
o o 0 o o
5 4 3 2 1

ADV mmOCLODOF awar—

21 July 26 Sept

 

   

 

 

 

 

55555

3v awe—.330... 3a..

6..
D.

27 July

Date

21 May

ak
Si and

ghness of immature and mature leaves during 1992: (A) European
ition and larval development of each generation of alder (A), birch (

21: Ton
al er, (B) paper birch, and (C) hawthorn. Sampling dates correspond with

hawthorn (C) leafminer,

OViPOS

 

e
r
u
t
a
m
m

 

 

 

 

 

 

 

 

 

 

mmmmo

A8 36:59.0... 30..

18 A09

14 June

12 May

 

 

  

 

 

 

 

 

mmmmmo
55555

§ 888:3 ES

15 Jun.

13 May

 

 
   

  

 

 

 

 

 

 

ADV meCSDDOF *GOJ

9
u
A
5
2

O
n
U

14,)

19 May

Date

ak
Bi...

ghness of immature and mature leaves during 1993: (A) European

B) paper birch, and (C) hawthorn. Sampling dates correspond with

re 22; T011

(

ovipos1tion and larval development of each generation of alder (A), birch (

hawthorn (C) leafminen

1',

3f:

63

P < 0.001) (Figure 21B), and 170: 12 g and 250:25 g, respectively, on 15 July 1993
(t=-2.97; d. f. =6,6; P<0.05) (Figure 22B).

Hawthorn: Immature leaves were first sampled when females were
ovipositing (21 May 1992 and 19 May 1993). These leaves were compared to
mature leaves sampled at times that corresponded with the presence of second and
third generation alder and birch leafminer (27 June and 18 September in 1992 and
14 June and 25 August in 1993, respectively). The toughness of immature leaves
was 148:5 g in 1992 (Figure 21C) and 171 :3 g in 1993 (Figure 22C). These
differences were highly significant.

LeafArea

Alder: During the first generation of alder leafminer in 1992 (22 May) and
1993 (15 May), respective means of immature leaves were 17:1 (Figure 23A) and
13 : 1 cm2 (Figure 24A). On the second sample date in 1992 (20 July), which
corresponded with peak oviposition of second generation alder leafminer, immature
leaves were not significantly larger than mature leaves (31 : 2 and 28: 2 cm2,
respectively) (t=0.97, d.jE =7,7; P=0.3496) (Figure 23A). However, when leaves
were sampled during peak oviposition by second generation adults in 1993 (14 July),
immature leaves were smaller than mature leaves (25 : 2 and 30:1 g, respectively)
(t=-2.22; d. f =5,5; P=0.05) (Figure 24A).

On the third sample date in 1992 (26 September) mature leaves (the only
leaf type available for sampling) had a mean area of 29:2 cm2 (Figure 23A). On
the third sample date in 1993 (18 August), which corresponded to peak oviposition
by third generation adults, immature leaves were smaller than mature leaves (18 : 2

and 37:3 cmz, respectively) (t=-4.84; d. f. =5,5; P<0.001) (Figure 24A).

 

 

 

 

 

 

 

 

m
u
m
m
m

 

 

 

$.53 e24. 36..

20 July

22 May

 

26 Se

W.
J

 

 

 

 

 

$.83 o2< Eon

0|
0.

21u

22 May

 

1880

 

 

 

$88 «92. 36..

t
p

27 July

21 May

Date
23: Area of immature and mature leaves during 1992: (A) European alder,

per birch, and (C) hawthorn. Sampling dates correspond with peak

pment of each generation of alder (A), birch (B), and-

051tion and larval develo

hawthorn (C) leafminer.

$3155
(Mp

0
pl
U
A...
a
M

 

 

 

 

 

 
 

    
 

 

 

 

 

 

 

A W

 

000000
5 4 3 2 1

14 July 18 Aug,

12 ”8y

 

 

%////////////%

  

 

 

 

B m

00000
5 4 3 2 1 0

 

15 July 17 Aug,

13 May

 

 

 

 

 

000000
55555

ANEoV moi. 56..

14 July 25 Aug.

19 May

g 1993: (A) European alder,

(C) hawthorn. Sampling dates correspond with peak

1' birCh, and

P?“

a
osmon and larval development of each generation of alder (A), birch (B), and

re 24: Area of immature and mature leaves durin

5'2)
0v1p

Fl
(B
hawthorn (C) 1m“.

66

Birch: During the first generation of birch leafminer in 1992 (22 May) and
1993 (13 May), the mean area of immature leaves was 11: 1 (Figure 23B) and 9: 1
cm2, respectively (Figure 24B). On the second sample dates (21 July 1992 and 15
July 1993), which corresponded to peak oviposition by second generation birch
leafminer, immature leaves were smaller than mature leaves: 1992 (15 : 1 and 21 : 2
cm?) (t=-2.58; d.f =7,7; P<0.05) (Figure 23B) and 1993 (15 : 1 and 27:4 cmz) (t=-
3.14; d. f. =6,6; P< 0.01) (Figure 24B).

On the third sample date in 1992 (26 September), only mature leaves were
available for sampling, and their area was was 23 :1 cm2 (Figure 23B). On the third
sample date in 1993 (17 August), corresponding with third generation birch
leafminer oviposition, immature leaves were not significantly smaller than mature
leaves (27:4 and 31 :2 cm2, respectively) (t=-0.70; d. f. =4,4; P=0.5057) (Figure
24B).

Hawthorn: Immature leaves sampled during the first generation in 1992 (21
May) were smaller in size than mature leaves collected on the second (27 July) and
third (18 September) sample dates (t=-20.27; d. f =7,7; P< 0.001 and t=-14.78;
d. fi =7,7; P< 0.001, respectively) (Figure 23C). The leaf area increased from 4: 1
cm2 to 10: 1 cm2 as leaves matured (Figure 23C). The same patterns was observed

in 1993 (Figure 24C).
Specific Leaf Mass

Alder: On the first sample date in 1992 (22 May) and 1993 (12 May), when
only immature leaves were available, their respective specific masses were 46: 03
g/m2 (Figure 25A) and 46:02 g/cm2 (Figure 26A). On the second sample date in
1992 (27 July) and 1993 (14 July), which corresponded with peak oviposition by

67
D Immature Mature

 

 

 

 

 

 

 

 

 

 

 

 

 

a“ 1.00

E

3 0.80 -

(I)

8

z 0.60 '

'2': .

_' 0.40

.2

3'3 0.20 -

0

O.

a) 0.00

22 May 20 July 26 Sept

‘5‘ 1.00

E

’

3 0.80 -
U)

8

2 0.60 '

7:3, .

.1 0.40

.2

7, 0.20 ‘

e

O.

<0 0.00 ,
r 1.40
E .
3 1.20
g 1.00 -
a
2 0.80 -
3 0.60 -
.J
g 0.40 r-
E 0.20 r
Q
m 0.00

Figure 25: Specific leaf mass of immature and mature leaves during 1992: (A)
European alder, (B) paper birch, and (C) hawthorn. Sampling dates correspond

with peak oviposition and larval development of each generatlon of alder (A),

birch (B), an

hawthorn (C) leafminer.

68
C; Immature Mature

 

 

 

Specific Leaf Mass (g\m2)

0.40 // /
% /
fl Z2 %

12May July 18Au.g

0.40 '
0.20 '

13 May

 

 

 

 

Specific Leaf Mass (g\m’)

 

1.20

0.80 *

0.60 '

0
:::::fl

0.00
19 May 14 July 25 Aug.

Date

 

Specific Leaf Mass (g\m"’)

 

 

Figure 26. Specific leaf mass of' 1mmature and mature leaves during 1993: (A)
European alder, (B) paper birch, and (C) hawthorn. Sampling dates correspond
with peak oviposition and larval development of each generation of alder (A),
birchpc (B), an hawthorn (C) leafminer.

69

second generation alder leafminer, specific mass of immature leaves was less than
that of mature leaves: 41 :03 and 51:03 g/mz, respectively in 1992 (t=-2.27;

d.fE =7,7; P<0.05) (Figure 25A), and 39:02 and 52:03 g/mz, respectively in 1993
(t=-4.16; d. f. =5,5; P<0.01) (Figure 26A).

On the third sample date in 1992 (26 September) only mature leaves were
available, the specific mass of which was 61: 2 g/m2 (Figure 25A). On 17 August
1993, corresponding with oviposition by third generation females, specific mass of
immature leaves was less than that of mature leaves (38: 02 and 52: 03 g/m2,

respectively) (t=-3.70; d. )5 =5,5; P<0.005) (Figure 26A).

Birch: On the first sample date in both 1992 (22 May) and 1993 (13 May),
when only immature leaves were available to ovipositing females, their respective
specific leaf mass was 46: 01 g/m2 (Figure 25B and 26B). On the second sample
date in 1992 (21 July) and 1993 (15 July), corresponding with oviposition by second
generation birch leafminer, the specific mass of immature leaves was less than that
of mature leaves: 66:03 and 77:03 g/mz, respectively in 1992 (t=-2.98; d. f. =7,7;
P< 0.01) (Figure 25B), and 57:06 and 64:06 g/mz, respectively in 1993 (t=-0.87;
d. f. =6,6; P=0.40) (Figure 26B).

On the third sample date in 1992 (26 September), when only mature leaves
were available, their specific mass had increased to 84: 03 g/ m2 (Figure 25B). On
the third sample date in 1993 (17 August), the specific mass of immature leaves was
again less than that of mature leaves (43 : 04 and 75 : 03 g/m2, respectively) (t = -
6.16; d. f. =4,4; P<0.001) (Figure 26B).

Hawthorn: In 1992, immature leaves collected during peak oviposition (21
May) had a lower specific mass than mature leaves collected on the second (27 July)

and third (18 September) sampling dates, which corresponded to presence of second

70

and third generation alder and birch leafminers (t =-18.63; d. f. =7,7; P< 0.001 and
t=-14.55; d. fi =7,7; P<0.001, respectively) (Figure 25C). Specific mass of immature
leaves was 50: 01 g/m2, while that of mature leaves was 103 : 03 (27 July) and
115 : 04 g/m2 (18 September) (Figure 25C).

The same trend was observed in 1993, with immature leaves sampled on 19
May being lower in specific mass than mature leaves sampled on 14 July and 25
August (t = -11.57; d. f. = 7,7; P < 0.001 and t= -9.82; ctfi = 7,7; P < 0.001, respectively)
(Figure 26C). The specific mass of immature leaves was 50: 01 g/m2, increasing to

99:04 g/m2 on 14 July, and 105 :05 g/m2 on 25 August (Figure 25C).

Nitrogen

Alder: In 1992, on the first sample date (22 May) which corresponded with
peak oviposition of alder leafminer, only immature leaves were available for
analysis, and their nitrogen concentration was 3.29: 0.01% (Figure 27A). On the
second sample date (20 July), which corresponded with the second generation
leafminers, immature leaves were higher in percent nitrogen than mature leaves
(3.18: 0.09 and 2.91 : 0.06%, respectively) (t =-2.36; d. f. = 15,15 ; P< 0.05) (Figure
27A). On the third sample date (26 September), only mature leaves were available
for analysis and their nitrogen concentration was 2.91: 0.06% (Figure 27A).

Birch: On the first sample date in 1992 (22 May), which corresponded with
first generation birch leafminer, only immature leaves were available for analysis.
The nitrogen concentration of these leaves was 3.60: 0.09% (Figure 27B).
Immature leaves were higher in nitrogen than mature leaves on the second sample
date (21 July) (1.94:0.08 and 1.82:0.04%, respectively) (t=-1.41; d. fl =14,15;
P=O.1700) (Figure 27B). On the third sample date in 1992 (26 Sept), only mature

 

 

 

 

 

 

 

 

 

 

 

555555

.5352 .x.

oh
0.
9

26s

20 July

22 May

 

 

B

 

 

 

 

 

 

555555

couoazz $

o."
p
O

268

21 me

22 May

 

EC ///////////

 

 

 

 

555555

couofz as

on
D.
O

immature and mature leaves during

hawthorn. Sampling dates
opment of each generation of

)

C

eVe(l

Paper birch, and
sit1)on and larval d
hom (C) leafininer.

akald?” (B
0v]
Ham

Figure 27 : Percent nitrogen concentration of
). and

19921 (A) Enfope
COl'l'CS n d with
alder 8:). birch g

72

leaves were available for analysis, their nitrogen concentration had declined to

1.84:0.06% (Figure 27B).

Hawthorn: On the first sample date in 1992 (21 May), which corresponded
with peak oviposition, only immature leaves were available for analysis, and their
nitrogen concentration was 4.09 :‘0.01% (Figure 27C). Their nitrogen concentration

declined to 2.47: 0.04% by 27 July, where it remained unchanged (Figure 27C).

Phosphorous

Alder: The phosphorous concentration of immature leaves sampled on 22
May was 0.24: 0.01% (Figure 28A). On the second sample date (20 July),
immature leaves were higher in phosphorous than mature leaves (0.22: 0.01 and
0.14: 0.01%, respectively) (t =-6.68; d. f. =15,15; P< 0.001) (Figure 28A). On the
third sample date in 1992 (26 September), only mature leaves were available for
analysis, and their phosphorous concentration had declined to 0.14:0.01% (Figure
28A).

Birch: The phosphorous concentration of immature birch leaves sampled
on 22 May was 0.35 : 0.01% (Figure 28B). Immature and mature leaves were equal
in phosphorous concentration on the second sample date (21 July) (0.22: 0.01%)
(t=0.19; d. fl =14,15;P=0.8493) (Figure 28B). On the third sample date in 1992 (26
September), only mature leaves were available for analysis, the phosphorous

concentration of which increased to 0.32: 0.02% (Figure 28B).

O
f
U
t
8
M

 

Immature

 

 

 

 

 

 

 

A /////2

 

 

 

 

 

 

maoaocamocn xx.

20 JulY 25 so“.

22 May

 

 

B///////////////%

  
 

26 S°Pt,

 

 

 

 

 

 

 

_

machosamosa axe

 

 

C

 

 

 

 

 

O
5.
O

0
4

0.

0
3.
O

O
2

0.

O

1

0.

0,00

macaochoza axe

27 July 18 s.”

21 May

Date

pean alder, (B) paper birch, and (C) hawthorn. Sampling

(A) Euro
pond with peak oviposition and larval development of each generation

gure 28: Percent phosphorous concentration of immature and mature leaves
of alder (A), birch (B), and hawthorn (C) leafminer.

dates cones

Fl
riuri11g 1992:

74

Hawthorn: The phosphorous concentration of immature hawthorn leaves
sampled on 21 May was 0.43 : 0.01% (Figure 28C). On the second sample date (27
July) only mature leaves were available for analysis, and these leaves were 1
compared to immature leaves collected on the first sample date. The immature
leaves had a phosphorous concentration twice that of mature leaves (0.44 : 0.02 and
0.20:0.01%, respectively) (t=-13.26; d.fl =15,15;P<0.001) (Figure 28C). On 18
September, the phosphorous concentration of mature leaves was also substantially
less than that of immature leaves (0.44:0.02 to 0.17:0.01%) (t=-15.85; d. )3 = 15,15;
P<0.001) (Figure 28C).

Age Specific Larval Mortality In Relation to Leaf Maturity

Leaf toughness

Alder leafminer: For first generation alder leafminer in 1992, first instar
mortality was 36% in leaves with a toughness of 200 g or less, while mortality was
40% in leaves greater than 200 g (Figure 29A). N o fourth and fifth instar mortality
was observed. However, these conclusions are limited by the number of
observations. In 1993, first generation first instar mortality increased with leaf
toughness. Mortality was 24% when leaf toughness was less than 200 g, and
increased to 53% for leaves within the toughness range of 250-300 g (Figure 29B).
However, mortality of second and third instars decreased as toughness increased
and leaves matured (Figure 29B).

For the second generation in 1992 and 1993, first and second instar
mortality tended to increase as leaf toughness increased. In 1993, first and second

instar mortality was 46% and 22%, respectively, when leaf toughness was less than

Instar I 1st E] 2nd % 3rd 4th

100

% Larval Mortality

% Larval Mortality

100

75

51h

 

.. A

60
40

20

 

<200 >200

 

CD

80

60

 

40"

20'

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

<200 200-250 250-300 >300

Leaf Toughness (9)

Figure 29: The relationship between leaf toughness and instar-specific larval
mortality of first generation alder leafminer 1n: 1992 (A) and 1993 (B).

76

200 g and increased to 92% and 80%, respectively, when leaf toughness exceeded
250 g (Figure 30B).

Birch leafminer: In 1992 and 1993, first generation mortality of first and
second instar birch leafminer increased as leaf toughness increased (Figure 31A).
In 1992, first and second instar mortality was 21% and 14%, respectively, when leaf
toughness was less than 150 g, and increased to 50% and 29%, respectively, when
leaf toughness exceeded 200 g (Figure 31A). In 1993, when leaf toughness was less
than 150 g, first and second instar mortality was 5% and 0%, respectively, and
increased to 67% and 60%, respectively, when leaf toughness exceeded 250 g
(Figure 31B).

Second generation morality was not analyzed in 1992 due to limited
observations. However, in 1993 first instar mortality increased from 58%, when leaf

toughness was less than 150 g, to 80% when leaf toughness exceed 200 g (Figure 32).

Hawthorn leafminer: In 1992, first, second and third instar hawthorn
leafminer mortality increased from 50%, 6%, and 0%, respectively, when leaf
toughness was less than 150 g to 67%, 11%, and 8%, respectively, when leaf
toughness exceeded 200 g (Figure 33A). To few observations were recorded for
fourth and fifth instars to draw conclusions.

In 1993, second and third instars and fourth and fifth instars were grouped
to increase the number of observations. For each grouping (lst, 2nd + 3rd, and
4th +5th instars), mortality increased as leaf toughness increased. First instar
mortality ranged from 34% to 99% and second instar mortality ranged from 16 to
75% for leaves with toughness less than 200 and greater than 250 g, respectively
(Figure 33B).

77

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Instar I 1st [II 2nd @ 3rd 4th 5th
100
>‘ A
r:
7“ so -
t
o
2 so
3
> 40
I.
«i
-' 20
S’
o . ..
<1so 150-200 201-250 >250
100
>‘ B
3::
75 so -
4—0
3
2 60 7
E .
> 40
h
m
-' 20 -
o x'l 7 ....
<2oo 200-250 >250

Leaf Toughness (9)

Figure 30: The relationship between leaf toughness and instar-s cific larval
mortality of second generation alder leafminer in 1992 (A) and 993 (B).

78

Instar I 1st [:1 2nd @ 3rd 4th 51h

 

 

 

roo
" A
.1:
7“ so
.0-0
‘5
2 so
'25
> 40
S-
is
-' 20
S"
o
<1so 151-200 201-250
100
>‘ B
r:
B so .
.0—0
h
o
2
E
>
L—
to
.1
S"

 

 

 

 

<150 150-200 200-250 >250

LeafToughness(g)

Figure 31: Relationship between leaf toughness and instar-specific larval mortality
of first generation birch leafminer in 1992 (A) and 1993 (B).

79

"1813' -1st [:3 2nd W 3rd W 4th

100
80
60
40

20

 

% Larval Mortality

<150 150-200 >200
LeafToughness(g)

Figure 32: Relationship between leaf tougémess and instar-specific larval mortality
of second generation b1rch leafminer in 1 3.

80

 

INStar - fat [:1 2nd W 3rd
100
>' A
J:
a so
g—e
h
g so
‘3 40
L-
(0
—l 20
c,\"
0
.<150 150-200 >200
INStar - 1st :3 2nd+3rd m 4th+5th
100
7 B
.".'_'
'5 so
4—0
L—
g 60
g 40
h
(0
_.| 20
5°

 

<200 200-250 >250

LeafToughness(g)

Figure 33: Relationship between leaf toughness and instar-specific larval mortality
of hawthorn leafminer 1992 (A) and 1993 (B).

DISCUSSION

Alder leafminers oviposited exclusively on immature leaves and birch
leafminers highly preferred immature leaves as oviposition sites, even when only
mature leaves were available. When birch leafminer females were offered mature
leaves for oviposition, only two leaves were selected as oviposition sites in the entire
experiment; these two leaves were the youngest leaves on a single seedling and
received a total of 8 eggs. This provides strong evidence that immature leaves are
the only leaves utilized by alder and birch leafminers as oviposition sites. Friend
(1931) reported that birch coppice growth, which produces new leaves in
abundance, were heavily infested by birch leafminer compared to foliage on
established shoots. These data, combined with high first instar mortality in mature
leaves, strongly suggest these species are dependent on immature leaves for survival.
This conclusion is consistent with the hypothesis that the indeterminate growth
pattern of European alder and paper birch may have allowed the alder and birch
leafminers to evolve a multivoltine life cycle by providing a continuous supply of
immature leaves.

Experiments were not conducted to determine if hawthorn leafminers
would oviposit in mature leaves. However, given the fact of nearly 100% first instar
mortality occurs in leaves with toughness exceeding 250 g, and that mature leaves
present in July and August approach 400 g in toughness, it is speculated that early
instar larvae would not be able to survive in the tougher mature leaves.

Feeny (1970) found one of the most obvious changes in oak leaves as they
mature to be an increase in toughness. He suggested that increased toughness may
have a significant negative effect on the ability of larvae to feed. Consistent with
this, as alder, birch, and hawthorn leaves matured, their toughness increased and

early instar survival declined. However, as larvae matured, leaf toughness had a
8 1

82

much less pronounced effect on mortality. Tanton (1962) reported that leaf
toughness retarded feeding and growth of Phaedon cochleariae mustard beetle)
larvae. As toughness increased the area consumed by larvae decreased, larval and
pupal mortality increased, and growth rate decreased. As observed in this research
with late instars, Tanton found third instar larvae to be less influenced by toughness.
He proposed that this was due to an increase in mandibular strength.

Physical and chemcial differences between immature and mature alder and
birch leaves available to the second and third generations were large. Differences
between immature hawthorn leaves sampled during peak emergence of the
hawthorn leafminer, and mature leaves sampled in correspondence with the second
generation of alder and birch leafminer were also large. For each species, mature
leaves were tougher, larger, higher in specific leaf mass, and were lower in
phosphorous and nitrogen concentrations, which may have a negative influence on
developing larvae.

Feeny (1968) reported that tannin concentrations of oak leaves increased
with leaf age, and that larval growth and pupal weight decreased as tannin
concentrations increased. Feeny suggested that this was due to a reduction in the
availability of protein. In addtion, Feeny suggested that the seasonal increase in
tannins concentration of oak leaves may be a factor contributing to the selection for
early feeding larvae on oak.

All of these characteristics of mature leaves may help to explain why all
three species oviposit in immature leaves, and first instar larval mortality of all three
species increases as leaves mature. These results are consistent with the hypothesis
that the hawthorn leafminer is constrained to a univoltine life-history by the
determinate growth habit of it host plant, while the indeterminate growth habit of
birch and alder has permitted the evolution of a multivoltine life history of the birch

and alder leafminer.

CHAPTER THREE

HOST PHENOLOGY, INTRASPECIFIC COMPETITION, CLIMATIC
VARIATION AND THE EVOLUTION OF INTRASPECIFIC
VARIATION IN VOLTINISM PATTERNS

Although alder and birch trees display indeterminate growth, the number of
immature leaves present at any one time may decrease rapidly after the first flush.
Consequently, the quantity of immature leaves available to second and third
generation alder and birch leafminers may only be a fraction of that available to the
first generation. Corresponding to this pattern of resource availability, numbers of
first generation adult birch leafminers (overwintering individuals) always exceed
numbers of second and third generation adults (Herms, unpublished). In the case of
alder leafminer, large first and second generations are followed by a small third
generation (Hart et al. 1991). In addition, decreased availability of immature leaves
available for females to oviposit in could create the potential for competition among
larvae as larvae are concentrated within the few available leaves.

Intraspecific competition has been documented in several leafminer species.
Tuomi et al. (1981) documented intraspecific competition in Scolioneura betuleti, a
hymenopteran leafininer infesting birch leaves. lntraspecific competition occurred
as the population density increased, decreasing larval survival. They also found a
negative correlation between mean larval weight and larval density. Furthermore,
mean larval weight was positively correlated with leaf area available per larva.
These findings coincide with those of Quiring and McNeil (1984a), who documented

competition in alfalfa blotch leafminer (Agromyza frontella) populations. Stiling
8 3

84

(1984) reported that larval mortality of Hydrellia valida, on salt marsh cordgrass,
increased when two or more mines coalesced, and that intraspecific competition
accounted for 11.4% of total mortality. lntraspecific competition has also been
documented in populations of leafminers on oak, aspen, and lima beans (Bultman &
Faeth 1986a, Auerbach 1991, and Petitt & Wietlisbach 1992, consecutively). It
appears that in leafminers intraspecific competition frequently occurs at high larval
densities due to food limitation.

lntraspecific competition may select for density-dependent premature extra-
long diapause (sensu Hanski 1988), which occurs when some individuals enter
diapause, avoiding competition, but missing one or more breeding opportunities
experienced by individuals within the population that do not diapause. The pitcher-
plant mosquito, Wyeomyia smithii, is one of the best documented examples of
density-dependent premature extra-long diapause. According to Istock et al. (1975 ),
at high densities some individuals may enter diapause, foregoing breeding
opportunities experienced by others, but avoiding competition for limited food.

The observed small population size of the second generation of birch
leafminer, relative to the first generation, suggest that most individuals may not
emerge from pupation as second generation adults. Instead, these individuals may
enter density-dependent premature extra-long diapause to avoid intraspecific
competition among offspring in a resource limited habitat. These individuals loose
breeding opportunities experienced by non-diapausing individuals, but emerge the
following spring when a full flush of immature leaves is once again available. This
may represent a bet-hedging strategy (Philippi and Seger 1989). Philippi and Seger
(1989) defined bet-hedging as an evolutionary trade-off between mean and fitness of
individuals; individuals with a lower mean fitness may have a selective advantage

over highly fit (high mean fitness) individuals under certain conditions.

85

Climatic conditions such as photoperiod, temperature, and resource
availability and quality may play a significant role in insect phenology (Tauber and
Tauber 1973). Climatic conditions influence the termination date of leaf production
and larval development rates (Scriber & Lederhouse 1992), and thus have the
potential to select for evolution of a density-independent premature extra-long
diapause (sensu Hanski 1988). Decreased numbers of adults giving rise to third
generation larvae suggest that some second generation birch and alder leafminer
individuals may enter density-independent premature extra-long diapause during
the pupal stage. The second generation of each species may be "hedging their bet"
(Philippi and Seger 1989) as to whether to give rise to third generation individuals
or to enter diapause. Third generation individuals may have to "race" to finish
development before unfavorable fall climate conditions prevail, and may not be able
to complete development in some years. However, diapausing individuals loose a
breeding opportunity, but will emerge the following spring. However, they may also
face elevated risks of mortality prior to reproduction.

Bet-hedging may also take the form of risk spreading through phenotypic
variability in offspring. Offspring from a single female may be phenotypically
different to ensure that at least some will be adapted to a uncertain future. Thus, it
may be an adaptive strategy that may increase fitness in a temporally variable
environment (Bradford & Roff 1993). Bradford and Roff (1993) suggested that a
bet-hedging diapause strategy occurs in a partially bivoltine population of cricket
(Allanemobius fasciatus). As first generation females aged, the proportion of
diapausing eggs increased, reflecting a decrease in the probability of second
generation offspring being able to complete development before the end of the
growing season. Therefore, it is possible that observed variability in voltinism
patterns can be controlled by the ovipositing female. A small portion of the

offspring of second generation birch and alder leafminer may be conditioned by

86

their mother to emerge as second or third generation adults, while the majority

would overwinter.

OBJECTIVES

The objective of this research was to test the following predictions: (1)
there is intraspecific seasonal variation in the availability of immature leaves, (2)
phenotypic patterns of voltinism within and among the three species correspond to
seasonal patterns of resource availability (immature leaves), and (3) intraspecific
competition occurs among larvae which may give rise to density dependent

premature extra-long diapause.

METHODS AND MATERIALS

There is Seasonal Variation in the Production of Immature Leaves

In 1992, four branches per tree from each of eight trees of all three species,
which showed long extensional growth the previous year were flagged at bud break.
The following variables were measured three times throughout the summer (late
May, late July, and early September): (1) total number of leaves per branch, (2)
number of immature leaves per branch, (3) number of leaves on the terminal shoot,
and (4) terminal shoot length. Seasonal variation in foliage production on birch
trees was documented. However, due to severe deer browsing, insufficient data

were collected for alder and hawthorn trees.

87

In 1993, the following modifications to methods were made: (1) alder (only
6 alder trees was used) and hawthorn branches above the deer browse line were
sampled, (2) in order to increase the accuracy of counting the total number of
hawthorn leaves per branch, the terminal leaf on each shoot was tagged and (3)
measurements were made at the time of peak adult emergence of each generation
(early June, mid-July, and mid-August) to estimate the number of immature leaves
available for the ovipositing adults. Measurements ceased when terminal bud
formation occurred.

Total number of immature leaves, total number of leaves, mean terminal
shoot length, and total number of leaves on terminal shoots were compared in
relation to Julian date by using Repeated Measures AN OVA. The proportion of
immature leaves available to the second and third generations relative to that of the

first generation were compared using AN OVA.

Intraspecific Variation in Voltinism Patterns Corresponds to
Phenological Patterns of New Leaf Production

The number of individuals emerging as adults (not entering premature
extra-long diapause) in relation to the rate of production of new leaves by the host
plant (resource availability) was determined. See Chapter One for adult sampling
protocols.

The average number of adults per trap at peak emergence, average number
of larvae per leaf and percent of leaves infested were correlated with the number of

available immature leaves using correlation analysis.

Intraspecific Competition Among Larvae

In 1992, leaves with second generation birch and alder leafminer larvae, and
in 1993, leaves with first and second generation alder and birch larvae were
collected within two days of emergence of mature larvae and brought into the
laboratory for analysis. Leaf area (cmz) was recorded, the petiole was inserted into
a Water Pik, and each leaf was then placed in a petri-dish with a plaster filled base
saturated with water to maintain humidity. The leaves were stored in a incubator
(18:6) at 25 0C until larva(e) completed development. After larvae emerged from
the leaf, they were placed individually in a sand filled vial in order to better simulate
natural conditions. After all larvae had exited the leaf, total mined area (cm2) of the
leaf was determined. After the larvae pupated, each pupa was removed from its
case and weighed (mg). After weighing, each individual was returned to its
designated vial. Each vial was examined daily to determine if adult emergence
occurred. Vials were examined for two weeks after the last adult emerged.
Individuals that did not emerge as an adult were examined to determine whether
they remained as pupae in diapause or if they died.

The relationship between larval density and (1) pupal mass, (2) percent
emerging as adults, (3) percent diapausing and (4) percent survival (percent

emerging as adults + percent diapausing) were quantified using correlation analysis.

88

 

RESULTS

Seasonal Variation in the Production of Immature Leaves

Alder

The production of immature leaves by alder decreased over the course of
the season in 1993 (P<0.001) (Figure 34A). During peak emergence of first
generation alder leafminer (1 June), there were 74: 9 immature leaves per branch.
By 29 June, the number of immature leaves per branch had decreased to 42: 5. By
13 July, during peak oviposition by second generation alder leafminer, the number
of leaves per branch had decreased to 20:3. Even fewer immature leaves were
- available to third generation females (20 August), when each branch had only 7: 1

immature leaves per branch (Figure 34A).

Birch

Seasonal production of immature leaves by birch trees also decreased in
1993 (P=0.01) (Figure 34B). For first generation (1 June) birch leafminer, when all
leaves were immature, there were 49 : 3 leaves per branch. By the second
generation (15 July), there was a seven-fold decrease in leaf production to 7:1
immature leaves per branch. Finally, during the third generation (12 August), there

were only 0.1 : 1 immature leaves per branch (Figure 34B).

Hawthorn
Immature leaves were present on hawthorn only during the first generation
of hawthorn leafminer (at bud break). No leaf production occurred after the initial

flush at bud break (Figure 34C).
89

90

 

 

 

 

 

 

 

 

 

 

 

 

 

.5 100
c
2 1: A
m 80‘
\
3 so»
>
a
0
.1 40-
o
‘5
s 20'
E 1
g o L l
1 June 29 June 13 July 20 Au
.1: 100
o
c
. - B
m 80
\
8 so-
>
a
o
-l 40
o
h-
3 .
a 20
E
1 June 15 July 12 Au
5 100
c
2
1:1 50’ C
\
3 so-
>
(I!
o
.J 40-
o
5
iii 2°”
E
.5. o
2 June 9 July 16 Au

Figure 34: Seasonal production of immature leaves by European alder (A), paper
birch (B), and hawthorn (C).

91

Intraspecific Variation in Voltinism Patterns Corresponds to
Phenological Patterns of New Leaf Production

Alder Leafminer

The relative population size of each generation of adult alder leafminer did
not correspond with leaf production in 1993 (Figure 35A). The second generation
was only eight percent smaller than the first generation, even though the number of
immature leaves available to the second generation had declined to 84 percent. The
availability of leaves to the third generation was only five percent of that available
to the first generation, corresponding with the size of the third generation of alder

leafminer relative to that of the first (Figure 35A).

Birch leafminer

Changes in adult population size from one generation to the next was highly
correlated with changes in immature leaf production. The second generation of
adults was 76% smaller than the first generation, and the third was 98% smaller.
Meanwhile, the number of immature leaves available to the second and third
generations was 86% and 99% less than that available to the first generation,

respectively (Figure 35B).

Hawthorn leafminer
After the first generation was completed no adults emerged for a second

generation (univoltine) and no leaf production occurred (Figure 35C).

 

92

— Adults
--- Immature Leaves

 

100

80'

60

40'

Percentage

 

 

 

 

80

O
0

Percentage
h
0

20

 

 

 

 

80'

O
0

Percentage
8

N
O

 

 

 

 

O
016

2

Generation

‘

Figure 35: Relationship between the percent of leaves that are immature and the
size of each eneration of adult leafminers, relative to the first generation: alder
(A), birch ( , and hawthorn (C) leafminer (1993).

 

 

Intraspecific Competition Occurs Among Larvae

Relationship Between Larval Density and Pupal Mass

Alder leafminer: During the first generation in 1993, pupal mass decreased
as larval density increased (Y= 0.0041‘0-46"; r2=-0.41; P<0.01) (Figure 36A). This
trend was also observed during the second generation (Y= 0.0047964"; r2 =-0.40;

P<0.01) (Figure 37A).

Birch leafminer: During the first generation, pupal mass decreased
exponentially as larval density increased (Y= 0.0027'036"; r2 =-0.22; P< 0.01) (Figure
36B). Pupal mass also decreased as larval density increased during the second

generation (Y= 0.0035‘057"; r2=-0.52; P< 0.01) (Figure 37B).

Hawthorn leafininer: There was not a significant relationship between

larval density and pupal mass.

Relationship Between Larval Density and Survival, Adult Emergence, and Diapause

Alder leafminer: For first generation alder leafminer, percent of individuals
per leaf surviving, emerging as adults, and diapausing decreased as larval density
increased (Figure 38) (Y=10550e'145x; r2=-0.77; P< 0.01, Y= 94.19133"; r2 = -0.69;
P<0.01 and Y=22.18'2-14X; r2=-0.88; P< 0.01, respectively). The percentage of
individuals surviving, emerging as adults, and diapausing was 52, 49, and 2%,

respectively.
9 3

 

0.60

Pupal Mass (mg)

0.00

0.00

0.60

Pupal Mass (mg)

0.00

0.00

1.20

Pupal Mass (mg)

0.00

0.00

94

 

0.50 ' o
. ... 0 Y- 0.0041.o.46x
0.40 ’
0.30 '
0.20 '

0.10 '

 

r2' -O.41

 

 

 

0.25 0.50 0.75 1.00 1.25 1.50

 

0.50 '

0.40 ’

0.30 ’

0.20 '

0.10 ’

 

 

r2- -0.22
Y- 00027-0-86X

B

 

 

0.25 0.50 0.75 1.00 1.25 1.50

 

1.00 ’

0.80 ’

0.60 '

0.40 '

0.20 '

 

C

.e?" "

5'.
o

 

j l l j

 

0.25 0.50 0.75 1.00 1.25 1.50

Larvae I cm2

Figure 36: Relationship between first generation larval density and pupal mass in
1993: (A) alder leafminer, (B) birch leafminer, and (C) hawthorn leafinmer.

 

95

 

 

 

 

 

0.80
a r2= -0.40
E 0-60 4 Y- 0.0047-0-64X
(I)
8
0.40
2
711'
O. 0.20 ‘
3 .
O.
0.00 ‘ ‘ ’
0.00 0.25 0.50 0.75 1.00 1.25 1.50
0.80

 

r2= -0.52
0.60 - y- 0003505“ B

 

0.20 '

Pupal Mass (mg)

 

 

 

0.00 J l l 1 l
0.00 0.25 0.50 0.75 1.00 1.25 1.50

Larvae I cm2

Figure 37: Relationship between second eneration larval density and pupal mass
in 1993: (A) alder leafminer and (B) birc leafmmer.

96

 

 

 

 

 

 

 

 

 

 

 

100
' A
__ so - . ’
g - 2
"5 so . . 0. l' " -O.77
O
5 - , \(-=1os.502-45X
CD 40 - . .0
$3 .’ e .
20 ’
0* . , . . . o .
0.00 0.25 0.50 0.75 1.00 1.25 1.50
g 100
° ' B
“O _ o
< 80 o
to o
‘5 60 ’ 9. ..e r2- '0.69
O
.5 4o '. Y- 9449-2-33x
O . ’ o o
E 20 -
LU o
g 0 fl ' ' ’ '
0.00 0.25 0.50 0.75 1.00 1.25 1.50
100
a r2--0.88 ( >
.5 8° v-22.1si'-14x
3
(6 60
O.
.2 40
0
5° 20 o
O
0

0.00 0.30 0.60 0.90 1.20 1.50
Larvae / cm2

Fi 38: Relationship between first eneration alder leafminer larval density and
(A percent of individuals surviving, (B3) percent of individuals emerging as adults,
and (C) percent of individuals diapausing (1993).

 

97

The same trend occurred in the second generation, with percent survival,
percent emerging as adults, and percent diapausing decreasing as larval density
increased (Figure 39) (Y= 93.40109"; r2 = -0.61; P < 0.01, Y= 96.96253"; r2 = -O.61;
P< 0.01 and Y= 37.91'1-797‘; r2 = -0.50; P< 0.01, respectively).

Birch leafminer: During the first generation, there was no correlation
between larval density and percent survival (P>0.10), percent emerging as adults
(P> 0.10), and percent of individuals diapausing (P> 0.10) (Figure 40C). The
percent surviving, emerging, and diapausing were 58, 39, and 19%, respectively.

During the second generation of birch leafminer, there was no correlation
between percent survival and larval density (n =56; P> 0.10) (Figure 41A).
However, as larval density increased, percent emerging as adults and percent
diapausing decreased (Y=61.70‘°58X; r2=-0.14; P<0.05 and Y= 26.63994"; r2=-0.13;
P< 0.05, respectively) (Figure 41B&C, respectively). As was the case for alder
leafminer, the percentage surviving and emerging increased from the first
generation to 67 and 51%, respectively. Meanwhile, the percent diapausing
decreased slightly to 16%.

Relationship Between Pupal Mass and Survival, Adult Emergence, and Diapause

Alder leafminer: For first generation alder leafminer, percent of individuals
per leaf surviving (individuals emerging as an adult + diapausing individuals),
emerging as adults, and diapausing increased as pupal mass increased (Figure 42)
(Y= 7.06m"; r2=0.23; P< 0.01, Y= 6.7251“; r2=0.22; P< 0.05 and Y= 2.7243“; r2 =0.46,

respectively).

 

 

 

 

 

 

 

 

 

 

98
100 W
o
,0. .. A
3 ’ o
.2 ,0. r2: -0.61
> oo o .
s -0 O Y: 93.40-2.09X
(I) 40 - ’
a e . oo
o\ .. .
20 ' e o
e
O r f . ¢ . L
0.00 0.25 0.50 0.75 1.00 1.25 1.50
g 100 - ¢
3 o
2 so - o B
to ’ 2
(U 60 ' Q. Q . r = '0.61
g o oo o '
'- 4o - o Y' 96.96'2'53x
U)
h o o
0 O .
E 20 ' . o
LU ' p ’
g o -.—d-o—e—l-o—o—r ' +
0.00 0.25 0.50 0.75 1.00 1.25 1.50
100
r2- -0.50
g 80 ’
,7, Y- 3191-1-79X
3 60 '
a o o
.9 4o -
D 0
oo o
$ 20 ' e o
o
" o

 

 

 

0W ' '
0.00 0.25 0.50 0.75 1.00 1.25 1.50

Larvae I cm2

Figure 39: Relationshi between second eneration alder leafminer larval density
and (A) percent of ind1viduals survivin ) percent of individuals emergmg as
adults, and (C) percent of individuals 1apausing (1993).

99

 

 

 

 

 

 

 

 

 

 

 

100 r—o—eo———o—e-e
80 - 0‘ e , , A
a e e e .
> o. e e
; so - . 00 0.3.0. ° .
5 e 000 e e
(D 40 ' 0
o e co .
°\ 20 O .
9 e
o r—o—%
0.00 0.20 0.40 0.60 0.80
:2 100 —¢
° B
2 8° ' .
to no 0. e
00 oo o .’ ’ ,
g 00 ee 0 e one o
"' 4O ' . 0 0 .
9 e e
o . ..
00 e
E 20 r ‘0
LU ° .
g o 'W
0.00 0.20 0.40 0.60 O 80
100 +
g .. . C
.5 0 e
e
g 60 ' .
a. O O O
.9. 4o » 0 °
Q 00 e e . e
e.
3‘3 20 - o
O O ...
o W
0.00 0.20 0.40 0.60 0.80

Larvae / cm2

figure 40: Relationship between first eneration birch leafminer larval density and
(A percent of individuals surviving, (B percent of individuals emerging as adults,
and (C) percent of individuals diapausing (1993).

 

100

 

 

 

 

 

 

 
 

 

 

 

 

 

 

100 PH 9
' ’ O’ ’ A
_ e
'5 8° '° " .’ ' ’ ..
h 0 Q 0
u e
3 g .
(D 40 ’ .9 9
o e e e e
5 e
20 ' 9
o l I l I
0.00 0.25 0.50 0.75 1.00 1.25 1.50
(D 100 V
g . 2= o 14
u r - .
< 80 ' .°“ Y- 61.7O'°-5°x
e
(0 . . O
(B so 9 . e .
2° -- .1:
°- 40 - 3
9 e .‘ .09 e
O o e
LIJ
g o ho-e—e ‘ ‘ '
0.00 0.25 0.50 0.75 1.00 1.25 1.50
100
0 r2- -0.13 C
,5 8° Y- 26.-63'0““
‘9 o
‘36 60 r .
Q 9 9 e
.9! 40 b 0
Q 9 9 e
o ‘ .09 O 9
o\ 20 - g. . 0
O . o
o W ’ '1
0.00 0.25 0.50 0.75 1.00 1.25 1.50

Figure 41: Relationship between second
and (A) percent of individuals survivin

Larvae / cm2

adults, and (C) percent of individuals iapausing (1993).

 

eneration birch leafminer larval density
63) percent of individuals emerging as

 

    
    

 

 

 

 

  

 

 

 

 

 

 

 

101
100 2 c
r = 0.23 '
. O
__ 30 y- 7 06529.36X
a O
.2 .
> so
5
a
to 40- .
o\°
20 -
o 1 1 . 4
0.00 0.10 0.20 0.30 0.40 0.50 0.60
g 100 2
g r = 0.22
so- ’
< y- 6_7251a.2ax
rn
‘6 eo -
o:
c
'- 4o -
9 e
(D
E 20 -
LIJ
°\° o l L . ‘
0.00 0.10 0.20 0.30 0.40 0.50 0.60
100
r2-0.46 C
a so -
.E Y-2.72‘-3"‘
(I)
3 60 L
as
a.
.9 4o-
0
$ 20 - o
O
o ‘ ‘ W

0.00 0.10 0.20 0.30 0.40 0.50 0.60
Pupal Mass (mg)

Fi re 42: Relationship between first eneration alder leafminer pupal mass and
(Agapercent of individuals surviving, (8% percent of individuals emerging as adults,
and (C) percent of individuals diapausing (1993).

 

102

For second generation alder leafminer, the percent of individuals surviving,
emerging as adults, or diapausing increased as pupal mass increased (Figure 43)
(Y= 13.9831”; r2 = 0.18; P < 0.05, Y= 6.25”“; r2 = 0.22; P < 0.01 and Y= 6.372813";
r2 = 0.22; P < 0.01, respectively).

Birch leafminer: For first generation birch leafminer pupal mass was not
related to the percent of individuals surviving (P> 0.10) (Figure 44A) nor the
percent of individuals per leaf emerging as adults (P> 0.10) (Figure 448). However,
the percent of individuals diapausing per leaf was positively correlated with mean

pupal mass (Figure 44C) (Y=11.78495X; r2=0.24;P<0.01).

 

As with the first generation, percent survival per leaf was not related to !
pupal mass (n =56; P> 0.10) for the second generation of birch leafminer. However,
the percent emerging as adults and percent diapausing increased as pupal mass
increased (Y= 22.5427“; r2 = 0.14; P < 0.05 and Y= 5.93338"; r2 = 0. 10; P < 0.10,
respectively) (Figure 45).

103

 

 

 

 

 

 

 

  
  

 

 

 

 

 

    

 

 

 

100 2 4v 3 “*—
r = 0.18 ., A
_ 3° ’Y- 13.983‘2-36" . ' .
g e e
E so ~
3 O
U) 40 r
o\°
20 -
o .
0.00 0.10 0.20 0.30 0.40 0.50 0.60
g 100 r
e B
< 80 . 2 .
g r = 0.22
so . .,
O y- 6. 25494.5” .fi ,
r:
'- 4o - 0
8’ .
0
E 20 -
“J 0
g o 1 W ‘ J
0.00 0.10 0.20 0.30 0.40 0.50 0.60
100
C r2= 0.22
2° 8° ’ y- 6_37288.1ox
E so
a . l
.2 4o -
D o
o\° 20 -
e 0 e.
o ' a; W

0.00 0.10 0.20 0.30 0.40 0.50 0.60

Pupal Mass (mg)

Fi re 43: Relationship between second generation alder leafminer pupal mass and
(A percent of individuals surviving, (B) percent of individuals emerging as adults,
and (C) percent of individuals diapausing (1993).

104

 

 

 

 

 

 

 

 

 

 

    

 

 

100 e m H e
0
so - A g 3
_ e . e e
g 3' e 0'. . m
E so - ,0 o .,
:3 co 00 e co
(0 40 ~ 0
o e e 0
°\ 0 O o
20 "
. e
O L ’ I: L
0.00 0.10 0.20 0.30 0.40
(I) 100 F
2
a B
- e
< 80 . .
m e c on
_ e e
‘3 eo . .0
53 mm m e
‘3 o
a 40 ’ 9 9
u. . e . e
O u e
E 20 - e e
LU O
o\° 0 L——Lo—o—e u H+
0.00 0.10 0.20 0.30 0.40
100 2 9
r - 0.24 C
2’ 8° Y- 11.784953”: . °
(0
3 60 r
(U
D.
.2 40
0
5° 20
O o
o h—I—oor—moomo—A
0.00 0.10 0.20 0.30 0.40

Pupal Mass (mg)

Fi 44: Relationship between first eneration birch leafminer pupal mass and
(A percent of individuals surviving, (B percent of individuals emerging as adults,
and (C) percent of individuals diapausing (1993).

 

 

 

 

 

 

 

 

 

 

 

105
100 c : :uu—eo————
O
A ° °’ ' '
30 ’ .0 ~ '0
w 0 O ...
E 60. . - ., '
h . 9
3 O O Q.
(I) 40- 3 . o
O O O O
o\° .
20 - 0
o k g
0.00 0.10 0.20 0.30 0.40 0.50
3 100 2 : ecu
g r = 0.14 .
< 80 -y_ 225427431): . 9
«n
or so -
o
.E
a 40 h
h
m
E 20 P
LIJ
$ 0 ‘ A :L: c
0.00 0.10 0.20 0.30 0.40 0.50
100
r2- 0.10 C
2’ 80 ' y- 5.9338808X
.5 .
a so- .
a
.9 40
D
o\° 20 -
o

 

 

 

 

 

0.00 0.10 0.20 0.30 0.40 0.50

Pupal Mass (mg)

Fi 45: Relationship between second generation birch leafminer pupal mass and
(Al)1 percent of individuals surviving, (B) percent of individuals emerging as adults,
and (C) percent of individuals diapausing (1993).

 

DISCUSSION

The results of these studies combined with previous results showing that
larvae can develop only in immature foliage (Chapter 2), support the hypothesis that
the indeterminate growth of birch and alder has permitted the evolution of
multivoltinism in birch and alder leafminers, while the determinate growth pattern
of hawthorn has constrained hawthorn leafminer to one generation. Thus, the data

are consistent with the idea that the patterns of production of immature leaves by

 

their host plants has directed the evolution of voltinism patterns in the three species

if

and also selected for premature extra-long diapause in birch and alder leafminer.
Alder leafminers oviposit almost exclusively in immature leaves.
Furthermore, survival of first and second instars is much greater in immature leaves
(Chapter 2). Thus, immature leaves seem to be an obligate resource for alder
leafminer. Alder trees continued to produce immature leaves through mid-August.
However, the quantity of immature leaves available to each successive generation of
alder leafminer decreased greatly as the season progressed, resulting in fewer leaves
being available for ovipositing females and less food for larvae. However, sampling
showed that the second generation alder leafminer adult population size only
declined eight percent compared to the first generation, even though the availability
of immature leaves declined dramatically. Therefore, the small change in the
second generation adult population size, relative to the first, did not correspond to
the dramatic decrease in the availability of immature leaves over the same period.
However, the size of the third generation was greatly reduced relative to the first
generation. This suggests that a large proportion of the second generation
population entered premature extra-long diapause, corresponding with the large

reduction in the availability of immature leaves. Laboratory experiments also
106 x

107

confirmed the presence of extra-long diapause in first and second generation
individuals.

The clumped distribution of alder leafminer larvae (Chapter 1), coupled
with the fact that larvae are restricted for life to a single leaf, creates the potential
for intraspecific competition for the available food resource. This seems to be
especially true during the second generation when the availability of immature
leaves is greatly limited, relative to the population size. In fact, as second
generation alder leafminer larval density increased, survival and pupal mass
decreased. The greatly reduced population size of third generation alder leafminer
could be the result of selection for a bet-hedging strategy based on extra-long
diapause. Competition among second generation larvae may have selected for
density-dependent extra-long diapause (weak competition among first generation
larvae would help explain the lack of premature extra-long diapause in first
generation individuals). Individuals entering diapause would fore-go a breeding
opportunity and avoid placing their offspring in a more competitive environment
(relative to that experienced by first generation individuals). They would emerge
the following spring as first generation adults, when immature leaves are once again
most plentiful.

Alternatively, year-to-year variation in degree-day accumulation could
select for density-independent extra-long diapause. Third generation larvae may
have to "race" climatic conditions to complete development before leaf maturation
and/ or abscission occurred. Increased degree-day accumulation during a warm
summer would favor individuals emerging since third generation larvae would have
an increased probability of completing development (because third generation
adults would emerge sooner and/ or larvae would be able to complete development
faster). Decreased degree-day accumulation in cooler summers would favor

diapausing individuals. Individuals attempting a third generation would appear

 

108

later in the summer (due to slower development of the first and second generation
individuals), decreasing the probability of their offspring completing development
before cessation of immature leaf production and leaf abscission. The result may be
a bet-hedging diapause strategy in which most individuals diapause and some
emerge as third generation adults. The relative proportion of each group in the
population would be dependent upon the probability of their being sufficient time
for their offspring to complete development before unfavorable conditions arrive.

Similar trends were observed for birch trees and birch leafminer. Birch
leafminer first and second instar larval mortality increased as leaves matured and
became tougher. Previous experiments revealed that females require immature
leaves for oviposition (Chapter 2), indicating the importance of immature leaves for
leafminer survival. Thus, there was a strong correspondence between the large
reduction in immature leaf availability over the course of the season and the large
reduction in the number of second and third generation birch leafminer adults,
relative to the first generation. Trapping data revealed the size of the second
generation adult population was only 24% of the size of the first generation. The
rest presumably entered premature extra-long diapause or died. In addition, as leaf
production continued to decline, an even smaller proportion (2%) of birch
leafminer individuals, relative to the first generation, were trapped during the third
generation. The reduction in population size is not the result of mortality because
the large population size of the first generation is consistently preceded by a
relatively small third generation of the previous year. These data and laboratory
experiments confirmed the presence of premature extra-long diapause in this birch
leafminer population.

As for alder leafminer, the distribution of birch leafminer larvae was
clumped (Chapter 1), resulting in the potential for intraspecific competition among

larvae for the available food resource. In accordance, during the second and third

 

109

generations, when immature leaves are less available, survival and pupal mass
decreased, as birch leafminer larval density increased.

These data are consistent with the hypothesis that the premature extra-long
diapause is density-dependent, selected for by competition. In other words, the
birch leafminer population hedged their collective bet that there would be a
sufficient food resource (immature leaves) for their offspring, with the proportion of
individuals of each generation entering premature extra-long diapause (thus by-
passing a breeding opportunity) corresponding to the relative proportion of
immature leaves available to the larvae of that generation.

Interestingly, individuals experiencing higher levels of competition (lower
pupal mass) were less likely to enter premature extra-long diapause. Possibly, these
individuals are less likely to survive a long diapause period due to relatively small
amounts of stored resources. Therefore, these individuals may have no viable
choice but to emerge as a second or third generation adults.

The greatly reduced population size of third generation birch leafminer
could also be due to a density-independent bet-hedging diapause strategy in the
population. Density-independent extra-long diapause could be selected for by year-
to-year variation in degree-day accumulation, as discussed above for the alder
leafminer.

In contrast to alder and birch, hawthorn displayed determinate growth with
immature leaf production being terminated following first leaf flush. Consequently,
no immature leaves were present later in the season that would be available to
successive generations. This appears to be a fundamental constraint restricting
hawthorn leafminer to a univoltine life history. First instar mortality approaches
100% in leaves tougher than 200 g, and leaves reach a toughness of 400 g as they
mature. Thus, the dependency upon immature foliage for larval development may

have restricted the hawthorn leafminer to a univoltine life cycle.

 

110

Compared to alder and birch leafminer, intraspecific competition was less
significant among hawthorn leafminer larvae, with no relationship between larval
density and pupal mass documented. This may have resulted because larval
distribution was less aggregated, possibly due to the smaller leaves of hawthorn,
which support one or two larvae versus five to ten commonly observed in birch and

alder leaves (Chapter 1).

 

SUMMARY

Alder, birch, and hawthorn leafminer specialize on tree species within the
genera Alnus, Berula, and Crataegus, respectively; and all three species oviposit
almost exclusively in immature leaves. Alder and birch trees display a
indeterminate growth pattern while hawthorn displays a determinate growth
pattern. Alder and birch leafminer are multivoltine with second and third
generation adults emerging in early July and mid-August, respectively (DeClerk &
Shorthouse 1985; Hart et al. 1991). The hawthorn leafminer is univoltine (Wilson et
al. 1992). The population size of first and second generation of alder leafminer is
large relative to the third generation, and the size of the first generation of birch
leafminer is large relative to its second and third generation, conSistent with the
presence of premature extra-long diapause in a few first, and many second
generation alder leafminer individuals and in most first generation birch leafminer
individuals.

The primary objective of this research was to test the theoretical hypothesis
that the evolution of the voltinism patterns of alder, birch, and hawthorn leafminer
have been directed by the phenological pattern of immature leaf production by their
respective host plants. Specifically, the hypothesis states that the determinate
growth pattern of hawthorn has constrained the hawthorn leafminer to a univoltine
life history, while the indeterminate grth pattern of alder and birch has allowed
the evolution of a multivoltine life history by alder and birch leafminer.
Furthermore, the hypothesis was tested that the reduced production of immature

leaves after bud-break by alder and birch has resulted in a bet-hedging diapause
1 1 1

112

strategy in alder and birch leafminer. This strategy takes the form of density-
dependent premature extra-long diapause (sensu Hanski 1988) within the
population of first and second generation pupae selected for by intraspecific
competition for the immature leaf food resource. Thus, individuals emerging as
third generation adults take advantage of a breeding opportunity even though they
risk placing their offspring in a highly competitive environment.

The following experimental objectives were used to test this theoretical
hypothesis: (1) to characterize the seasonal presence and abundance of adults; (2)
to characterize the spatial distribution of larvae within and among leaves; (3) to
compare physical and chemical traits of colonized and non-colonized leaves; (4) to
test the hypothesis that immature leaves are the only suitable resource for larval
development and survival by: (a) choice and no-choice experiments comparing
oviposition preference for immature relative to mature foliage, (b) comparing
chemical and physical differences of immature and mature leaves, and (c)
quantifying larval survival in relation to leaf maturity; (5) to quantify seasonal
variation in the presence of immature leaves; (6) to test for intraspecific
competition among larvae for their immature leaf resource; and (7) to test for
correspondence between seasonal variation in the abundance of the three leafminer
species and seasonal variation in the production of immature leaves by their
respective host plants. ~

In 1992 and 1993, first generation alder leafminer adult emergence began in
early May with peak emergence in mid-May. In 1992 and 1993, distinct generations
were difficult to distinguish, but at least two population peaks were observed with
the July peak being larger than the May peak. In 1992, no adults were observed
after mid-September and in 1993 adults were not observed after late August.
During both years, a large first generation of birch leafminer was followed by much

smaller second and third generations. Peak emergence of the second and third

113

generations were observed in mid-July and early August, respectively. A single
generation of hawthorn leafminer was observed during both years, with peak adult
emergence in mid-May.

In 1993, larval abundance and distribution were quantified for each species
by counting the number of larvae in sampled leaves. For each generation of each
species, larval distribution was found to be non-random as the hypothesis of a
Poisson distribution was rejected. larval distribution also did not fit the negative
binomial distribution. The variance-to-mean ratio, which was always greater than
one and increased at a faster rate than the mean number of larvae per leaf, as well
as the d—statistic, supported strongly a highly clumped larval distribution for each
species (however, Green’s index always suggested that larval distributions were
more randomly distributed). None of the measured physical and chemical traits of
infested and uninfested leaves (toughness, size, specific mass, nitrogen, and
phosphorous) explained the clumped pattern of larvae which resulted from
oviposition site selection. Furthermore, for all three species, there was also no
relationship between the number of eggs laid per leaf and leaf toughness and size.
Neither was there a correlation between the number of eggs per leaf and specific
leaf mass for birch and hawthorn leafminers. But, there was a positive correlation
between the number of alder leafminer eggs per leaf and specific leaf mass.

Both choice and no-choice oviposition preference experiments did show
that leaf age influences oviposition site selection of second and third generation
alder and birch leafminer, both of which oviposited almost exclusively in immature
leaves.

Leaf maturation also had a negative effect on larval survival. For all three
species, first instar mortality increased as leaf toughness increased. Mortality
reached 80 to 90% when leaves exceeded 200-250 g. Leaf toughness had little

influence on fourth and fifth instar larvae.

 

114

As expected, mature leaves of each species were tougher, larger, and higher
in specific mass than immature leaves. In addition, mature leaves were lower in
percent nitrogen and phosphorous (birch leaves were equal) than immature leaves.

Immature leaves were present on hawthorn only during the single
generation of hawthorn leafminer. N 0 leaf production occurred after the initial
flush at bud break. The production of immature leaves by alder and birch trees
decreased over the course of the season. Therefore, the availability of immature
leaves for ovipositing adults also decreased with each generation.

To determine if intraspecific variation in voltinism patterns corresponded to
phenological patterns of new leaf production, the number of adults emerging each
generation was compared to the relative availability of leaves to each generation. In
1993, adult alder leafminer population size did not correspond with the availability
of immature leaves. The second generation was slightly smaller than the first
generation, even though the number of immature leaves available to the second
generation greatly decreased. However, the greatly reduced third generation was
proportional to the greatly reduced availability of immature leaves. Sampling and
laboratory experiments suggested that most second generation larvae entered
premature extra-long diapause with a few individuals emerging as third generation
adults, an observation consistent with the idea that a bet-hedging diapause strategy
was utilized by the second generation. 9

Conversely, changes in birch leafminer adult population size from one
generation to the next were highly correlated with changes in immature leaf
production. Most first generation birch leafminer larvae entered premature extra-
long diapause with only a few individuals emerging as second generation adults, and
even fewer for a third generation. This is consistent with the hypothesis that the

first generation utilizing a bet-hedging diapause strategy.

115

For hawthorn leafminer, after the first generation was completed, no adults
emerged for a second generation (univoltine) and no leaf production occurred.
These results, coupled with high larval mortality in mature leaves, are consistent
with the hypothesis that hawthorn leafminer is constrained to a univoltine life
history by the determinate growth pattern of its host.

There was strong evidence of intra-specific competition among alder and
birch leafminer larvae, but not among hawthorn leafminer larvae. For both first and
second generation alder and birch leafminer, pupal mass decreased as larval density
increased. Survival decreased as larval density increased for both generations alder
leafminer, but not for birch leafminer. For second generation birch leafminer, as
larval density and thus competition increased, the percent emerging as adults and
the percent diapausing decreased due to increased mortality.

In summary, the results support the hypothesis that larvae of all three
species can develop only in immature foliage, therefore the indeterminate growth of
birch and alder permitted the evolution of multivoltinism in alder and birch
leafminers, while the determinate growth pattern of hawthorn has constrained
hawthorn leafminer to one generation. Premature extra-long diapause of birch and
alder leafminers may be a bet-hedging strategy; diapausing individuals by-pass a
breeding opportunity, but avoid competition for the decreased supply of their
obligate food resource (immature leaves). Thus, these data support the theoretical
hypothesis that host plant phenology has influenced the voltinism patterns of alder,
birch, and hawthorn leafminers, and resulted in bet-hedging diapause strategies in

alder and birch leafminers.

APPENDIX

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.: 1994-2

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

Influence of Host Phenology on Voltinism Patterns of Alder, Birch and
Hawthorn Leafminer (HYmenoptera: Tenthredinidae), with Evidence for
Bet-Hedging Strategies in Birch and Alder Leafminer.

Museum(s) where deposited and abbreviations for table on following sheets:
Entomology Museum, Michigan State University (MSU)

Other Museums: U.S. National Collection
Systematic Entomology Laboratory, USDA

Investigator's Name (3) (typed)
Joseph Lee Chichester

Daniel A. Herms

 

Date 9-28-94

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

Deposit as follows:

Original: Include as Appendix 1 in ribbon copy of thesis or

dissertation.
Copies: Included 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 Curator,
Michigan State University Entomology Museum.

116

APPENDIX 1.1

Voucher Specimen Data

1 of 1 Pages

Page

 

hufimuo>fiCD wumum cmwwnofiz mzu :H ufimoamv

 

vmlmmlm mama

 

 

 

aw HoEoucm

11flafifiMHlJQJduflduu.

 

“0w mcwEHooam vmumfia m>onw ozu vm>wooom

 

Ntmmma .oz pososo>

Avoazuv

noumozowno .a common
Amvmamz m.uoumwfiumm>cH

Ahummmwoo: ma mumonm Hmcowufivvm mmav

 

3m:

3m:

OH

OH

OH

OH

mood was NH

Hz ..oo .cfiz
banana:

mumumm son .m.=

mama >mz NH

H2 ‘.00 .sz
OCMHUwZ
mcooumu zoo och

mmmH was NH

HZ ~.OU .Oflz
DCMHOflZ
mcmcumo son was

 

uuoaumz mwmcoomcoo mascououm

 

fimfiofla 35:8 82%

 

umwumnmmma mHHfimnm mmscmm

 

Museum
where

depos-
ited

 

Other

 

 

 

 

 

Pupae
Nymphs

Larvae

 

Eggs

umufimomow cam pom: no vouooaaoo
mamafiomaw you wont Hogan

 

:oxmu umsuo so mmfiooam

 

‘3 Adults 9

ponaaz

 

 

 

117

LIST OF REFERENCES

 

LIST OF REFERENCES

Ayres, M.P. and S.F. MacLean, Jr. 1987. Develo ment of birch leaves and the
ggréwth energetics of Epim'ta autumnata Geometridae). Ecology 68:558-

Auerbach, M. 1991. Relative impact of interactions within and between trophic
levels during an insect outbreak. Ecology 72: 1599-1608.

Auerbach, M. and D. Simberloff. 1984. Responses of leafminers to atypical
leaf production patterns. Ecol. Entomol. 9:361-367.

Auerbach, M. and D. Simberloff. 1989. Oviposition site reference and larval
mortality in a leaf-mining moth. Ecol. Entomol. 4:131-140.

Bradford, MJ. and DA. Roff. 1993. Bet-hedging and the diapause
strategies of the cricket Allonemobius fasciatus. Ecology 74:1129-1135.

Bultman, TL. and SH. Faeth. 1986a. erimental evidence for intraspecific
competition in a lepidopteran 1e miner. Ecology 67:442-448.

Bultman, TJ... and SH. Faeth. 1986b. Leaf size selection by leaf-mining insects on
Quercus emoryi (Fagaceac). Oikos 46:311-316.

DeClerck, RA. and J .D. Shorthouse. 1985. Tissue preference and damage by
Fenusa pusilla and Messa nana (Hymeno tera: Tenthredinidae), eafmining
sawflies on white birch (Betula papyn’fera . Can. Entomol. 117:351-362.

Faeth, SH. 1991. Effect of oak leaf size on abundance, dispersion, and survival of
the leafminer Cameran‘a sp. (Lepidoptera: Gracillariidae). Environ.
Entomol. 20:196-204.

Faeth, SH, 8. Mop r, and D. Simberloff. 1981. Abundances and
diversity 0 e -mining insects on three oak host species: effects of
host-plant phenology and nitrogen content of leaves. Oikos 37:238-351.

F eeny, Paul. 1968. Effect of oak leaf tannins on larval owth of the winter moth
Operophtera brumata. J. Insect PhysioL, 14:805- 17

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

Friend, RB. 1931. The life history and control of the birch leaf-mining sawfly,
Fenusa pusilla Klug. J. Econ. Entomol. 24:171-177.

118

119

Friend, R.B. £33. The birch leaf-mining sawfly. Conn. Agric. Exper. Stat.
Bull. .

Hamilton, D.W. 1943. Notes on the cherry leafminer. N.Y. State Agric.
Exper. Stat, Bull. 703: 59-61.

Hamilton, D.W. 1950. Profenusa canadensts, a pest of english morello
cherries in eastern New York. Ecol. Entomol. 43:694-696.

Hanski, I. 1988. F our kinds of extra long diapause in insects: A review of theory
and observations. Ann. 2001. Fennicz 25:37-53.

Harbaugh, B.K., J.F. Price, and CD. Stanely. 1983. Influence of leaf nitrogen on
eafmineé' damage and yield of spray Chrysanthemum. HortScience
18:880—8 1.

Hart, E.R., D.G. Petty, R.B. Hall, D.A. Herms, R.D. Hanna and J .N. Kean. 1991.
Activity of Fenusa dohmii (H enoptera: Tenthredinidae) on Alnus in the
Northcentral United States. nviron. Entomol. 20:534-539.

Istock, CA, 8.8. Wasserman, and H. Zimmer. 1975. Ecolo and evolution of the
pitcher plant mosquito: 1. Population dynamics and aboratory responses to
00d and population density. Evolution 29:296-312.

Jones, J .M. and AG. Raske. 1976. Notes on the biology of the birch leafminer,
Fenusa pusilla (Lep.), in NewFoundland (Hymenoptera: Tenthredinidae).
Phytoprotection 57:69-76.

Kozlowski, T.T., P.J. Kramer, and 8.6. Pallardy. 1991. The
Physiological Ecology of Woody Plants. Academic Press, San Diego.

Labonte, GA. and LJ. Lipovs . 1984. Life histo of a red oak
leaf-mining sawfly, Pro enusa alumna Ma illivray (Hymenoptera:
Tenthredinidae) J. Econ. Entomol. 77:350356.

Little, T.M. and FJ. Hills. 1978. cultural Experimentation:
Design and Analysis. John iley & Sons, New York.

Ludwig, J .A. and J .F. Reynolds. 1988. Statistical Ecology: A Primer on Methods
and Computing. J. Wiley & Sons, New York.

Mattson, WJ. Jr. 1980. Herbivory in relation to plant nitrogen content.
Annu. Rev. Ecol. Syst. 11:119-61.

Merle, RD. 1988. Phenological resistance of oaks to the green oak leafroller,
Tortrix vin'dana (Lepidoptera: Tortricidae). In Mattson, WJ.,
Levieux, and Bernard-Ba an, C., eds. Mechanisms of Woody Plant
Defenses Against Insects: earch for Pattern. Springer-Verlag, New York.

Parrott, PJ. and BB. Fulton. 1915. The che and hawthorn sawfly leaf-miner.
New York Agriv. Exper. Stat. Bull. 411: 51-580.

120

Petitt, FL. and DD. Wietlisbach. 1992. Intras ecific competition among
same-aged larvae of Liriomyza sativae 8X tera: Agromyzidae) in lima bean
primary leaves. Environ. Entomol. 21:13 -140.

Philippi, T. and J. Seger. 1989. Hedgiilg one’s evolutionary bets, revisited. Trends
in Ecology and Evolution 4:4 .

Quiring, D.T. and J .N. McNeil 1984a. Exploitation and interference
intras cific larval competition in the dipteran leaf miner, Agromyza
fronte la (Rondani). Can. .I. 2001. 62:42 -427.

Quiring, D.T. and J .N. McNeil. 1984b. Intraspecific competition between different
aged larvae of Agromyza fiontella (Rondani) (Diptera: Agromyzidae):
ggvzangtggzesgcgf an oviposition-deterring pheromone. Can. J. Zool.

: 1 - 1 .

Scriber, J .M. and RC. Lederhouse. 1992. The thermal environment as a resource
dictating gelggraphic patterns of feeding specialization of insect herbivores,
In Hunter, .D., Ohgushi, T., and Price, .W., eds. Efi‘ects of
Resource Distribution on Animal-Plant Interactions. Acadenuc Press,
San Diego.

Slingerland, M.V. 1905. Two new shade-tree pests: sawfly leaf-miners on
European elms and alder. Cornell Agric. Exper. Stat. Bull. 233:47-62.

Smith, DR. 1979. Symphyta. Vol. 1, K.V. Krombein, P.D. Hurd, jr., D.R. Smith,
and B.D. Burks, eds. In Catalog of Hymenoptera in America North of
Mexico. Smithsonian Instit. Press, Wash. D. C.

Stiling, RD, and B.V. Brodbeck, and DR. Strong. 1984. Intraspecific competition
in Hydrella valida (Di tera: E hydridae), a leaf miner of Spartina
altemiflora. Ecology 5:660-6 2.

Stiling, P.D., D. Simberloff, and LC. Anderson. 1987. N on-random distribution
patterns of leaf miners on oak trees. 0ecologia 74:102-105.

Stiling, P., D. Simberloff, and B. Brodbeck. 1991. Variation in rates of
leaf abscission between lants may affect the distribution patterns of sessile
insects. 0ecologia 88:36 -370.

Tanton, M.T. 1962. The effect of leaf "toughness” on the feeding of larvae of the
mustard beetle Phaedon cochleariae Fab. Entomol. Exp. Appl. 5:74-78.

Tauber, MJ. and CA. Tauber. 1973. Insect phenolo : criteria for analyzin
dormancy and for forecasting post diapause evelopment and repro uction
in the fie d. Search (Agriculture) 3(12):1-16.

Tauber, MJ., C.A. Tauber, and S. Masaki. 1984. Ada tations to hazardous
seasonal conditions: Dorman , migration, an polyphenism. In
Huffaker, Carl B. and Robert Rabb. eds. Ecological Entomology. John
Wiley & Sons, New York.

121

Tuomi, J ., P. Niemela, and R. Mannila. 1981. Leaves as islands: interactions of
ggolioételgrza betuleti (Hymenoptera) miners in birch leaves. 01ko' s
:1 -1 .

Wilson, M.C., D.L. Schuder, and A.V. Provonsha. 1982. Insects of Ornamental
Plants. Waveland Press. Prospect Heights, Illinois.

nICHIan STATE UNIV. LIBRARIES
lllllllllllmllllllllllllllflllllWillIIIIIIIUIHIHH
31293010205049