".2123
2 J.)
”a Eh “

1:

Iv
a

.—
Ma

0*
m

 

fiI'IESIS
1
01

 

7L‘003
Hgozmo‘f
2
£9
E w E
E 5 §
LCD .95
:l 5 3
.9
E

 

This is to certify that the
dissertation entitled

BEHAVIORAL ECOLOGY OF THE GRAPE BERRY MOTH,
ENDOPIZA VITEANA CLEMENS, IN MICHIGAN
AGROECOSYSTEMS

presented by

Natalia Botero-Garcés

has been accepted towards fulfillment
of the requirements for the

Ph.D. degree in Entomolgqy

 

 

1/456“.

Maon Professch Signature
<e\\22_\‘33

Date

 

MSU is an Affirmative Action/Equal Opportunity Institution

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

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 o'JClRC/DateDquBS-pJS

BEHAVIORAL ECOLOGY OF THE GRAPE BERRY MOTH,
ENDOPIZA VITEANA CLEMENS, IN MICHIGAN AGROECOSYSTEMS

By

Natalia Botero—Garcés

A DISSERTATION
Submitted to
Michigan State University
in partial fiilfillment of the requirements
for the degree of
DOCTORATE OF PHILOSOPHY

Department of Entomology

2003

ABSTRACT
BEHAVIORAL ECOLOGY OF THE GRAPE BERRY MOTH,
ENDOPIZA VITEANA CLEMENS, IN MICHIGAN AGROECOSYSTEMS
By
Natalia Botero-Garcés

Vitis spp. are the native hosts of the grape berry moth, Endopiza viteana Clemens
(Lepidoptera: Tortricidae), a primary pest of cultivated grapes throughout Eastern North
America. Distribution within and between habitats in the grape agroecosystem was
determined with pheromone traps placed at different heights in woods and vineyards.
Moth relative abundance increased dramatically with height in woods, whereas 90.0% of
moths caught in vineyards were at the canopy level. Relatively few moths (< 4%) were
trapped in the interface between these habitats, where grapevines are not present. Overall,
results indicate that earlier suggestions of greater numbers of adult moths in vineyards at
the end of the season were based on insufficient understanding of the vertical distribution
of E. viteana. The distribution of this specialist insect seems to be associated with the
distribution of its wild and cultivated host throughout the agroecosystem. The level of
larval infestation of grape clusters and the abundance of adult male grape berry moth
were compared in commercial vineyards bordered by four types of uncultivated habitat:
deciduous woods, coniferous woods, a single row of trees, and grasses. Wild grapevines
were surveyed at each uncultivated habitat site for their presence or absence, fi'uiting, and
abundance. Infestation was greatest in vineyard borders near deciduous woods, and
lowest near grasses. Moth abundance inside vineyards was similar irrespective of

bordering habitat, but fewer moths were captured in grass borders than in borders with

trees. Wild grapevines were likely to be present in any habitat, but their fruiting was most
common in deciduous woods. Larval infestation in vineyards and abundance of wild
grapevines and wild clusters in neighboring habitats were correlated, indicating that the
presence of wild grapevines adjacent to vineyards poses a risk of increased damage by
grape berry moth in vineyards. To determine whether E. viteana are moving from these
wild hosts into vineyards, a mark-release-recapture study was undertaken. A total of
3,505 marked moths was released at the center of a vineyard and 850 were released inside
a neighboring woodlot. From the vineyard releases, 222 moths were recaptured in the
vineyard using pheromone traps and passive interception traps. Of the moths released in
the woods, 24 were recaptured, 18 of them in the vineyard, indicating that inter-habitat
movement occurs in this species. This is the first such study to demonstrate active flight
of female and male grape berry moth, and the data suggest greater flight capacity by
males. To assess the potential for passive dispersal of pupae, leaves of vines at the
vineyard border and interior were painted afier harvest and sampled in the succeeding
spring. Recovery of leaves in adjacent woods indicated that they dispersed during the
winter season. This wind-driven movement may well be the means of transportation for

the overwintering population of E. viteana in this agroecosystem.

With all love and admiration,
to my parents Benjamin Botero Forero and Beatriz Garcés de Botero,

and my sister Silvia Botero K.

iv

ACKNOWLEDGEMENTS

I would like to thank the members of my committee: Rufus Isaacs, Charles Edson,
Doug Landis, and James Miller for their guidance and unrelenting attention during the
five years of this research. I would like to thank the great people in the Small Fruit
Entomology Laboratory; Keith Mason, Zsofia Szendrei, Rodrigo Mercader, Nikhil
Mallampalli, and Darek Gajek for their pertinent advice and help both in the laboratory
and in the field. The undergraduate students Katie Hohauser, Elly Maxwell, Kelly Bahns,
Jeff Eastman, Sergey Balyinski, Matt Wilson, and Scott Reynolds for their excellent
assistance, especially under hot and humid weather; John Wise, Matt Daly and the staff at
Trevor Nichols Research Complex, for the use of their facilities. I am very grateful to the
growers Rick Brown, Louie Bonamego, Bryan Cronenwett, Ed Oxley, Don Thornton,
Bob Blum, Bob Van Vleck, Bill Mihelich, Tim Seppala, and John Mann for giving me
access to their land and allowing me to place all sorts of bizarre paraphernalia in their
vineyards. Thanks to the people who helped with suggestions, tips and the occasional
field day; Mark Longstroth, Doug Murray, Deb McCullough, Murray Hannah, Peter
McGhee, Fabian Mellaned, Dave Epstein, Lina Patifio and Gary Parsons. Special thanks
to the Department of Entomology at Michigan State University, particularly Fred Stehr,
Moe Nielsen, Barb Stinnett, George Ayers, Karim Maredia, Ed Grafius, and the staff, Jill
Kolp, Heather Lenartson, Angela J ernstadt, and Linda Gallagher. I would like to
acknowledge the advice received for Chapter 2 from the Michigan State University
Statistical Consulting Center, particularly from Corina Sirbu. Funding for this research
was made available in part by National Grape Cooperative, Michigan State Horticultural

Society, Michigan Grape and Wine Council, Project GREEEN, and the Michigan

Agricultural Experiment Station. Funding was also provided by a Dissertation
Completion Fellowship awarded by the College of Agriculture and Natural Resources, at
Michigan State University, and by several grants from the Hutson Endowment,
Department of Entomology, that allowed this research to be divulged in national and
international meetings. I am indebted to John Brown, from the Systematic Entomology
Laboratory (USDA) in Washington DC, for the most pleasant telephone conversations
while he tried to unravel for me the taxonomic issue concerning the grape berry moth.
Special thanks to the kind people who also helped by being my friends in East
Lansing: Jim Miller, Zsofia Szendrei, Fred Stehr, Meghan French, Ron and Helen Priest,
Pat Bills, Evelyne Cudel, Gary Parsons, Heather Eisthen, Justine L. Cudel, Eric
Hoffrnann and Markus Ruchter, Keith and Julienne Mason. With deep gratitude for my
professors in Colombia, Gilberto Morales and Constantino Mantilla, two notable
scientists whom I credit for making a researcher out of me. Finally, the most heartfelt
thanks to my major professor Rufus Isaacs, for accepting me as his first student,
encouraging me every step of the way, and showing me what a great scientist and a most

excellent human being is.

vi

TABLE OF CONTENTS

LIST OF TABLES ............................................ ’ ...................................... xi

LIST OF FIGURES ............................................................................... xiii

CHAPTER 1: The grape berry moth, Endopiza viteana, in grape agroecosystem.. . .1

Introduction .................................................................................. 1
History ................................................................................ 1
Grape agroecosystems in Eastern North America .............................. 2
Biology of the grape berry moth .................................................. 3

Life cycle and description .................................................. 3
Reproductive biology ...................................................... 5
Ecologyofthe grape berrymoth ........................ 7
Role of surrounding habitats in pest insect abundance ........................ 9
Local dispersal of tortn'cid moths ............................................... 10
Economic impact of grape berry moth .......................................... 12
Historical importance in agriculture .................................... 12
Current agricultural importance ......................................... 13
Management of grape berry moth ...................................... 14

CHAPTER 2: Distribution of grape berry moth, Endopiza viteana, in natural and
cultivated habitats ................................................................................. 18

Introduction ................................................................................ 18

vii

Materials and methods ................................................................... 21

Vertical distribution ............................................................... 21
Horizontal distribution ............. _ .......... ' ..................................... 22
Vertical and horizontal distribution ............................................. 23
Data analysis ....................................................................... 23
Results ....................................................................................... 25
Vertical distribution ............................................................... 25
Horizontal Distribution ........................................................... 26
Vertical and Horizontal Distribution ........................................... 27
Discussion .................................................................................. 32

CHAPTER 3: Influence of uncultivated habitats and natural host plants on cluster

infestation by grape berry moth, Endopiza viteana, in Michigan vineyards ........... 37
' Introduction ................................................................................ 37
Materials and methods ................................................................... 39
Vineyard infestation ............................................................... 39

Moth abundance ................................................................... 40

Wild grape survey ................................................................. 40

Data analysis ....................................................................... 42

Results ....................................................................................... 43
Vineyard infestation by E. viteana .............................................. 43

Abundance of male E. viteana ................................................... 45

viii

Survey of wild grape .............................................................. 47

Relationship between wild grape and grape berry moth infestation. . . .50

Discussion .............................................. ' .................................... 54

CHAPTER 4: Movement of grape berry moth in vineyard agroecosystems .......... 62

Introduction ................................................................................ 62

Laboratory methods for measuring flight capacity ........................... 65

Methods for marking insects ..................................................... 66
Methods for insect recapture under field conditions .......................... 67 ,

Movement by E. viteana ......................................................... 69

Methods ..................................................................................... 70

Mark-Release-Recapture study .................................................. 7 O

Traps used for recapture .......................... . ....................... 70

Marking moths ............................................................ 71

Moth releases ............................................................. 73

Moth recapture ............................................................ 74

Weather data .............................................................. 74

Data analysis .............................................................. 75

Malaise traps to measure movement in and out of vineyards ............... 76

Painted leaves study ............................................................... 78

Sampling .................................................................. 78

Data analysis .............................................................. 79

Results ....................................................................................... 81

ix

Mark-release-recapture study .................................................... 81

Vineyard releases ......................................................... 81

Woods releases ....................... ‘ ..................................... 91

Malaise trap study ................................................................. 92

Painted leaves study ............................................................... 93

Discussion .................................................................................. 95

Adult E. viteana movement ...................................................... 95

Malaise trapping ................................................................... 99

Passive movement of E. viteana ................................................ 99

CHAPTER 5: Conclusions and future directions .......................................... 102
REFERENCES CITED ......................................................................... 108
APPENDICES .................................................................................... 121
Appendix 1: Taxonomic note on Endopiza viteana Clemens ............................... 122
Appendix 2: Record of deposition of voucher specimens .................................... 123
Appendix 2.1: Voucher specimens data ........................................................ 124

LIST OF TABLES

Table 1.1. Spray table for conventionally-managed vineyards in Southwest Michigan,
indicating the phenological state of grapes at the time pests should be targeted. . . . . . . . ....17

Table 2.1. Average number of Endopiza viteana caught at four different heights at the
woods edge and the woods-vineyard interface in two different years. Means within a
column followed by the same letter are not significantly different (Tukey on = 0.05). . ....25

Table 2.2. Average number of Endopiza viteana caught per flight in traps placed at a
height of 1.5 m in four different positions within the habitats sampled during 2000.
Means within a row followed by the same letter are not significantly different (Tukey a =
0.05) .................................................................................................. 28

Table 2.3. Average number of Endopiza viteana caught in traps per position (1.5-9.0 m
heights pooled), during three flights sampled during 2001. Means within a row followed
by the same letter are not significantly different (Tukey a = 0.05) ........................... 28

Table 3.1. Infestation of grape clusters by Endopiza viteana larvae at border and interior
positions of vineyards adjacent to four types of uncultivated habitats. Samples were taken
at three times of the year during 2001 and 2002. For each year, means within a column
followed by the same letter are not significantly different (Tukey a = 0.05). . . . . . . . . . . .....44

Table 3.2. Relative abundance of male Endopiza viteana adults during 2001 and 2002 at
border and interior positions of four types of uncultivated habitats adjacent to commercial
vineyards. Samples were taken weekly during each season, and cumulative values are
reported for the three dates infestation was sampled on. For each year, means within a
column followed by the same letter are not significantly different (Tukey or = 0.05). . ....46

Table 3.3. Values obtained for regression analysis of cluster infestation in vineyards and
cumulative number of male Endopiza viteana adults captured in a) uncultivated habitats
and b) in vineyards. Significant regressions are displayed in bold ........................... 48

Table 3.4. Parameters measured in the 2002 survey of wild grape, Vitis spp., within four
types of uncultivated habitats adjacent to vineyards. For wild grape indices, means within
a colmnn followed by the same letter are not significantly different (Tukey a = 0.05)....49

Table 3.5. Correlation between vineyard percent cluster infestation by Endopiza viteana
larvae at vineyard borders and interiors, and the incidence of wild grapevines bearing
fruit clusters in the uncultivated habitat adjacent to the vineyard. Significant Kruskal-
Wallis-test CHI-square values are in bold ....................................................... 52

xi

Table 4.1. Total number of marked and recaptured Endopiza viteana moths released and
recaptured in two habitats. Percentages were calculated for each sex based on a 1:1 sex-
ratio of released moths .............................................................................. 84

Table 4.2. Maximum and mean distances flown by marked and recaptured Endopiza
viteana moths released in two habitats, and mean angle of direction flown. . . . . . . . . .....85

Table 4.3. Regression analysis of distance flown by vineyard-released marked Endopiza
viteana moths against time to recapture and average weather conditions for periods of
flight activity between release and recapture of each moth. Significant regressions are in
bold .................................................................................................... 86

xii

LIST OF FIGURES

Figure 2.1. Schematic representation (not to scale) of a study site in 2001. Pheromone
traps (triangles) were hung at different heights on poles or ropes in five positions across
the vineyard-woods habitats. Grey boxes represent the extent of the two habitats. . . . . ....24

Figure 2.2. Average number of male Endopiza viteana caught per trap during each flight
in 2001, trapped at four heights across five positions in the vineyard-woods
agroecosystem ........................................................................................ 29

Figure 2.3. Average number of male Endopiza viteana per trap during each flight,
trapped at six heights inside woods adjacent to vineyards during 2001. Within each flight,
bars with the same letters are not significantly different (Tukey or = 0.05). . . . . . . ...31

Figure 3.1. Schematic representation of an experimental site (not to scale), with
pheromone traps (triangles) at four positions across the vineyard-uncultivated habitat
transect. The vertical double line indicates the width of the tree row habitat; the interface
width ranged from 5-10 m depending on the vineyard. Wild grape surveys were done by
five observers (black ovals) following a 60 m transect (horizontal dotted lines). Counts of
wild grape were recorded every 6 m along these transects .................................... 42

Figure 3.2. Infestation levels by Endopiza viteana larvae in vineyard borders (dark) and
interiors (white) adjacent to habitats where wild grape clusters were present or not. a)
July 2001, b) July 2002. The asterisk indicates a significant difference between sites with
wild grape clusters absent or present .............................................................. 51

Figure 3.3. Relationship between percent cluster infestation in vineyards by Endopiza
viteana larvae in July 2001 (top) and July 2002 (bottom) and wild grape index (WGI) or
restricted wild grape index (RWGI) values of the adjacent habitats. Cluster infestation at
the borders is shown by dark circles and solid lines, and infestation at the interiors is
shown by open circles and dashed lines. The asterisk denotes significance ............... 53

Figure 4.1. Schematic representation (not to scale) of the experimental vineyard-woods
set up where marked moths releases took place. Pheromone and pane traps were arranged
around a central vineyard release point in concentric circles; a release point inside the
woods was also used to determine extent of movement between woods and vineyards...72

Figure 4.2. a) Detail of the Malaise trap insect collecting containers, made of two tops of
2 liter soda bottles encased in one another, with a lid fill] of a killing agent. b) Schematic
representation of the vineyard set up with two bi-directional Malaise traps placed at the
edge of woods and vineyard (gray boxes) at 1.5 m high, and one bi-directional Malaise
trap placed 9.0 m high near the woods canopy .................................................. 77

xiii

Figure 4.3. Schematic representation (not to scale) of the arrangement of painted vines in
the leaf dispersal experiment. Grapevines (circles) at the border and interior of the
vineyard in two positions per set (N-S and E-W) were spray-painted with four different
colors (lower). The sampling transects are shown by seven parallel transects from the
inside of the woods to 45 m inside the vineyard (upper). .‘ ..................................... 80

Figure 4.4. Weather conditions during the period between 1 July and 21 September of
2001 and 2002; wind, temperature, and relative humidity. Data were obtained by
averaging values between 1700 and 2200 h (moth active time) for each day. Vertical
arrows represent releases for each year ........................................................... 82

Figure 4.5. Weather conditions during the period between 1 July and 21 September of
2001 and 2002. Data were obtained by averaging values between 1700 and 2200 h (moth
active time) for each day. a) Circular histograms with mean angular vector and 95%
confidence limits for wind direction for each year. The arrows represent the average wind
direction. b) Daily precipitation during moth active time during the period of
releases ............................................................................................... .83

Figure 4.6. Circular histograms of the frequency and distribution of vectors of moths and
wind direction for the period that moths were flying between release and recapture in a)
pheromone traps and b) pane traps. The mean angular vector with 95% confidence limits
runs beyond the outermost circle; the 0° corresponds to the vineyard North and the center
of the histogram the vineyard release point, while the arrows represent the average
directions taken by the moths and the wind ...................................................... 88

Figure 4.7. Circular histograms of the frequency and distribution of vectors of a) female
moths and wind direction and b) male moths and wind direction for the period that moths
were flying between release and recapture in pane traps. The mean angular vector with
95% confidence limits runs beyond the outermost circle with 0° corresponding to the
vineyard North. The center of the histogram is the vineyard release point, and the arrows
represent the directions taken by the moths and the wind ...................................... 90

Figure 4.8. Circular histograms of the frequency and distribution of vectors of moths
released from the woods and wind direction for the period that moths were flying
between release and recapture in vineyard traps. The mean angular vector with 95%
confidence limits runs beyond the outermost circle. The 0° corresponds to the vineyard
North and the center of the histogram is the vineyard release point. The arrows represent
the average direction taken by the moths and the wind ......................................... 92

Figure 4.9. Mean :l: SE number of leaves counted per transect (45 m2) from inside the

woods habitat to 45 m inside the vineyard, for leaves that were painted a) on border vines
and b) on interior vines. The arrows indicate the location of painted vines . .. . . ..94

xiv

CHAPTER 1:
THE GRAPE BERRY MOTH, ENDOPIZA VITEANA,

IN GRAPE AGROECOSYSTEMS

INTRODUCTION
History

For more than 30 years, the North American species of grape berry moth,
Endopiza viteana Clemens (or Paralobesia viteana (Clemens) (see Appendix 1)) was
considered the same species as the European vineyard pest, Lobesia botrana (Denis and
Schiffermiiller). This was a consequence of misidentifications by early taxonomists and
inaccurate reports that it fed on a wide range of host plants (Johnson and Hammar 1912).
To further complicate the issue, E. viteana had historically been placed in several other
genera, including Polychrosis and Lobesia, and was described under the synonym of
Penthina vitivorana by Packard in 1869 (J. Smith, pers. comm). There is consensus
today, however, that these last three synonyms are old and out of use. Early in the 20“1
century, Slingerland (1904) demonstrated that the grape berry moth was an exclusively
American species, pointing out differences in habits that permitted separation of the two.
Unlike the European L. botrana, larvae of the grape berry moth are monophagous and
feed only on native and cultivated grapes (Vitis spp., Vitaceae), while pupation and
overwintering occur in a rolled or folded leaf in the leaf litter rather than on posts or
grape canes as is the case for L. botrana (Slingerland 1904). Today, the name used by
most authors is Endopiza viteana Clemens (but see Appendix 1 for taxonomic

discussion).

Grape agroecosystems in Eastern North America

The grape berry moth is a native tortricid, found only in Eastern North America. Its
territory extends fi'om Southern Canada and New England in the north to Florida in the
south, and to Texas in the west, coinciding with the range of its ancestral host, wild
gapevines (Vitis spp., Vitaceae) (Johnson and Hammar 1912, Isely 1917). Johnson and
Hammar (1912) provided a list of states most affected by the grape berry moth (in order
of importance): Ohio, New York, Pennsylvania, Indiana, Illinois, Michigan, Missouri,
New Jersey, Virginia, Maryland, West Virginia, Iowa, Delaware, Arkansas, and less
affected due to limited grape culture, Massachusetts, Connecticut, Kentucky, Kansas,
Texas, Nebraska, and Wisconsin. More than 50 species of wild grape occur in N.
America (F ergusson-Kolmes and Dennehy 1993) and these are commonly found in
stands of young woods, perturbed habitats, and on the borders of mature forests (Morano
and Walker 1995). Galet (1979) lists 10 principal American species, among which the
most common are V. aestivalis, the common blue grape or winter grape, V. labrusca,
known as the Fox grape, or Northern Muscadine, and V. riparia, called Bermuda vine,
Riverbank grape and other names. Voss lists these, and one more rare species, V. vulpina
present but scarce in southern Michigan (Voss 1985).

In Michigan agoecosystems, woods frequently border vineyards, and they typically
contain wild gape. When clusters from these wild gapevines were sampled during
August 1999, infestation by E. viteana was on average 84.9% in woods adjacent to
vineyards (range 64.5-96.3) (N. Botero-Garcés, unpublished), similar to results from New

York where 50-80% of wild berries have been reported to be infested (Dennehy et a1.

1990a). The presence of woods is positively correlated with larval infestation in adjacent
vineyards (Hoffman and Dennehy 1989, Hoffman et al. 1992), particularly at the borders
where high levels of infestation are found. However, infestation by this pest can also be
clumped in the vineyard interiors. Regional variation in vineyard risk to gape berry moth
has been noted between Ontario (Canada) and New York, and this has been attributed to
differences in absolute temperatures in winter (Dennehy et al. 1990a). Grape berry moth
pupae exposed to low temperatures die after a single exposure to —24 °C, and this was
found to be less common in vineyards with higher infestation than in less infested ones
(Dennehy et al. 1990a). High levels of snow accumulation in winter in gape gowing
regions near the Great Lakes favor pupal survival (Dennehy et al. 1990a, Martinson et al.

1991).

Biology of the grape berry moth

Life cycle and description. This species overwinters as diapausing pupae.
Emergence in spring usually spans six weeks and is determined largely by genetic factors
and temperature (Tobin et al. 2002). Some overwintering individuals can emerge as early
as April, while others may delay emergence until July or August, making generations
difficult to distinguish (Tobin et al. 2002). In Michigan, adult moths start appearing at the
end of April or beginning of May, having emerged from the leaf layer on the gound.
This species is protandrous; the first males begin flight activity before females are present
(Tobin et al. 2002). The adult moth varies in length between 4—6 mm, and is brown with
slate-gay paintings on its dorsal wing surface, shaped like a saddle. The female can be

distinguished from the male because she has a wider, rounder, and more blunt abdomen

(see Appendix 2). The male has a slim abdomen that ends in a scaly tip, where two
feathery claspers can be observed.

Grape berry moth females of the first flight emerge from pupation during gapevine
bloom, and thus do not have access to gape clusters, but rather flowers and minute
forming berries. Thus, first-generation eggs are laid on the rachis and stems of the
clusters of flowers or tiny developing berries, where they are difficult to see. They are
circular, flattened, but slightly convex, 0.8 mm in diameter (Tobin et al. 2002), and a
pearly, translucent white. The dark head capsule can be observed through the chorion
from the 3rd day of egg development.

After four days under normal temperature and light conditions (25°C, 16L:8D), a
single 1 mm long larva hatches, crawling out from a slit of the chorion near the edge of
the egg. First instars of this first generation crawl on stems and produce webbing in the
forming cluster. After feeding externally for 14-16 days, during which the larvae are
relatively unprotected against biotic and abiotic factors, mature larvae (4th instars) find
leaves in which to pupate. They do this by pulling a corner of the leaf over themselves, or
cutting a crescent-shaped section on the middle of the leaf body, and wrapping
themselves into it to spin their cocoon. The pupa is blue geen soon after the molt, but
turns to a dark brown until the adult emerges some six to nine days later.

Insects in the second flight, or first generation, behave in the same manner as the
previous (overwintering) generation, except that females lay eggs on top of the small
berries, sometimes several per berry under high infestation cases. Upon hatching, lst
instars direct themselves toward the stem to enter the berry. Other preferred entrance

points include contact spots where two berries touch, and skin folds where molds have

developed. Once inside, the larva feeds under the berry skin, causing discoloration and
premature ripening of the area. Full-gown larvae emerge from the berry in the 4th stage
and move to nearby leaves to pupate. After pupating, second generation adult moths
(called the 3rd flight) will fly, mate, and lay eggs again on the ripe gapes, but usually this
last generation will not fully develop as adults until the next year. Instead, the pupae will
enter diapause and spend the winter on fallen leaves.

Early studies in New York claimed there was only one full and one incomplete brood
of larvae per year (Johnson and Hammar 1912), but further data collection has
determined that, depending on weather conditions, the number of generations ranges
fi'om two to three per year (Hoffman et al. 1992) and up to four (Tobin et al. 2002).
Phenological studies undertaken by several authors (Clark and Dennehy 1988, Biever and
Hostetter 1989, Hoffman and Dennehy 1989, Alrn et al. 1989) have attributed variation to
differences in soil types (Pfeiffer et al. 1990), in degee-day accumulation (Hoffman et al.
1992, Tobin et al. 2001), and in possible moth races, as proposed by Tobin et al. (2002).
It is not known whether lSt and 2nd generation adults show the same protandry exhibited
by the overwintering population, but preliminary data suggest that males develop faster
than females (Tobin et al. 2002). Observations in the laboratory with a colony of this
species under constant temperature conditions (N. Botero-Garcés, unpublished) indicate
that males emerge at least a day earlier than females of the same age.

Reproductive biology. There is little data on reproductive biology of this species.
Comparisons are usually made with the European gape berry moth, L. botrana, but it
must be remembered that L. botrana is a generalist herbivore whose survival therefore

depends less on critical choices made by adult moths. Data on phenology of female E.

viteana are not available, primarily due to the lack of trapping studies and an effective
female attractant. The presence of a sex pheromone and identification of its main
component as (Z)-9-dodecenyl acetate (Z9-12Ac), was reported by Roelofs et a1. (1971).
The pheromone was later used by Taschenberg and others to develop mating disruption
strategies (Taschenberg et al. 1974, Taschenberg and Roelofs 1977). In recent studies,
Witzgall et al. (2000) extracted pheromone glands from female gape berry moths and
found that each contained approximately 1.2 ng of pheromone on the 2-3rd day of life.
Apart from the main component, eight other compounds were identified for this moth;
(Z) and (E)-11 tetradecenyl acetate; dodecyl acetate, dodecanol, (Z)-9-dodecenol,
tetradecyl acetate, tetradecanol and hexadecanol, some serving as synergists and others as
precursors of the main sex pheromone (Witzgall et al. 2000).

Though no studies have been carried out to determine onset of pheromone release in
gape berry moth, one-day-old females are believed to attain maximum pheromone
release on the second and third days of calling (Witzgall et al. 2000). Mating occurs
within the first three days of adult life for both sexes and most egglaying occurs on the 4-
7th day of adult life (G. English-Loch, pers. comm.)

In the laboratory, one female may lay from a few to several tens of eggs in one
night (N. Botero-Garcés, pers. obs.); most egg laying occurs one-hour either before or
after scotophase (Clark and Dennehy 1988). Studies on the European gape berry moth,
L. botrana, showed an oviposition capacity of approximately 30 eggs per female per day
(F eldhege et al. 1995); 90% of eggs were laid during the first ten days of the female’s life
(Savopoulou-Soultani et al. 1998). The adult lifespan of E. viteana ranges from 4 to 23

days (Johnson and Hammar 1912, N. Botero-Garcés, pers. obs.), depending mainly on

water availability. It is not known whether E. viteana mates once or multiple times, but
monandry has been demonstrated in L. botrana and two other important tortricid pests,
the oriental fruit moth, Cydia molesta, and the codling moth, C. pomonella (Torres-Vila
et al. 1997). Clark and Dennehy (1988) showed that, E. viteana oviposit exclusively on
gape berries in the field. There was a positive correlation between the number of berries
available and the number of eggs laid. In the laboratory, some eggs were found on
petioles and gape leaves, particularly when females were given the limited alternate
choice of an artificial gape substrate, but again most oviposition was on the clusters

(Clark and Dennehy 1988).

E cology of the grape berry moth

Insects are rarely distributed evenly in space and time (Schowalter 2000). For
phytophagous insects, the distribution of host plants is often a major determinant of the
insect's distribution. E. viteana is a specialist, and would be expected to be closely linked
to the host plants on which it lays eggs and feeds as larvae. Most gape agoecosystems in
Eastern North America consist of vineyards bordered by woodlots of several ages and
stages of succession, riparian areas unsuited for agriculture, windbreaks made of an
individual or two rows of trees, fallow and cultivated fields, or gasses. Grape berry
moths have been trapped mainly in vineyards (Biever and Hostetter 1989, Dennehy et al.
1990a, Trimble 1993), but surveys of wild gapevines invariably found eggs or larvae
present in gape clusters (N agarkatti et al. 2002a).

Pheromone traps are a useful tool for identifying the start and peak of moth flight

activity, as was observed by Johnson et al. (1988). Traps placed at the edge of woods

adjacent to a vineyard captured moths six days earlier than traps placed at the edge of the
vineyard. It has been observed early in the season that more E. viteana are caught in traps
placed at or near the edge of vineyards than in traps placed inside vineyards (Biever and
Hostetter 1989, Hoffman and Dennehy 1989, Trimble et al. 1991). Johnson et al. (1988)
described how moth abundance at the woods and vineyard edge shifted towards the
center of the vineyard as the season progessed. Hoffman and Dennehy (1989) trapped
gape berry moths at 15 different positions along a transect from a vineyard, into woods,
an alfalfa field, another woods, and another vineyard. They showed that a geater
proportion of moths were caught in the woods at the beginning of the season, but that this
changed near harvest when more moths were caught inside the vineyard than anywhere
else (Hoffman and Dennehy 1989). These previous studies, however, used data from
three traps removed from the woods edge position and relocated to the vineyard as the
season progessed (Johnson et al. 1988) or on only one transect of 15 traps in a single
vineyard/adj acent habitats setting was used (Hoffman and Dennehy 1989). A more robust
sampling system is needed to determine the patterns of male relative abundance within
and among habitats in Michigan vineyard ecosystems, in which vineyards are smaller in
area (0.4-2 ha) and most often bordered by woods on at least one side.

The importance of trap placement in relation to crop height, particularly in fruit
crops, has been demonstrated for many economically important pests (Peterson 1926,
Rothschild and Minks 1977, Riedl et al. 1979, Gabe] and Renczés 1985, Howell et al.
1990). The vertical distribution of insects has been studied with Lepidoptera in woods
(Boiteau et al. 2000a) and the rainforest (DeVries et al. 1999), with parasitoid

communities in forest stands (Nyrop and Simmons 1986), and with Carabidae and

Elateridae beetles for flying profiles in deciduous woods (Boiteau et al. 2000b). These
studies usually found that different species of insects maintained a particular height of
flight that often dependended on the structure of the habitat they were in.

Pheromone traps for E. viteana are commonly hung from the trellis at 1.5 m
above the gound. This recommendation stems from two considerations: first, it is the
average height of the gape canopy, and second, this is convenient when checking the
traps. However, this species inhabits a gape agoecosystem with considerable differences
in canopy heights. The vineyard and woods habitats of gape berry moth vary markedly
in their structure and complexity. In the former, gapes are tied on to trellises typically
less than 2 m high, whereas in the latter, gape vines extend vertically up to 25 m or
more, by climbing on trees. A response to habitat structure may be an important
component of the ecology of this insect in this agoecosystem, however, no study of

optimal trap height has been conducted for the gape berry moth.

Role of surrounding habitats in pest insect abundance

Habitat heterogeneity is a common feature of many gape-gowing regions of the
Eastern United States. This heterogeneity is high in Michigan because vineyards are
smaller and agricultural diversity is high, second after California. This diversity in crops
and uncultivated habitats is not only important in insect conservation, but can be an asset
to agriculture by aiding in maintaining low insect pest populations (Altieri 1983, Welch
1990, Macdonald and Smith 1990, Ekbom 2000), while at the same time increasing the
effectiveness of natural enemies (Wratten and Thomas 1990). The other side of the issue

is that uncultivated habitats may be so good at conserving some insects, that they

facilitate pest survival and disease transmission from wild or alternate hosts, as well as
acting as reservoirs of certain pest populations (Barrett 2000, J eanneret 2000).

In general, studies on the impact that adjacent uncultivated habitats have on crops
have focused on negative effects such as pest immigation, or positive effects such as
conservation of natural enemies (Seaman et al. 1990, Dorn et al. 1999). Few studies have
concentrated on native hosts within these wild habitats and their relationship with the
crop pest. Dennehy et al. (1990b) showed that woods were a risk factor for infestation by
E. viteana, but the importance of wild gape near vineyards for vineyard infestation by E.
viteana has not been investigated. It is unclear to what extent wild host abundance varies

with habitat type and how this is related to the levels of vineyard infestation.

Local dispersal of tortricid moths

It is generally accepted that insect populations within the agricultural landscape
are dynamic and able to flow from one habitat to the next in space and time (Kareiva
1983, Landis 1994, Ekbom 2000). Traditional trapping data (at 1.5 m) and comparisons
on relative abundance at different positions in the vineyard suggest that gape berry moth
populations may be moving from wild habitats into vineyards during the gowing season
(Hoffman and Dennehy 1989). Researchers have implied that discrepancy in trap data
and cluster infestation levels may be due to female immigation from woods to lay their
eggs in vineyard gapes (Taschenberg et al. 1974, Hoffman and Dennehy 1989, Trimble
et al. 1991, Trimble 1993). Such movement between wooded habitats and vineyards has
not yet been demonstrated categorically for E. viteana, but rather the reports assume

movement has occurred from the changing distribution of adults and larval damage.

10

The flight capacity of another tortricid fruit pest, C. pomonella, has been
examined according to the insect's physiological state and sex (Schumacher et al. 1997).
Mated or virgin adults of both sexes can have equal flight capacity, with maximum
dispersal measurements of 11 km. Flight ability was geatest during the 2“d to 7th day of
life, and females were not likely to engage in long (between-habitat) flights before having
oviposited, in disageement with the oogenesis-flight syndrome described by Johnson
(1969). In contrast, the oriental fi'uit moth, C. molesta, has much lower flight capacity
(approximately 1 km) and tends to be more sedentary; a small proportion of moths
ventured into long flights and gavid females were more likely to be the colonizers
(Hughes and Dom 2002). One of Davis’ (1980) explanations for the preponderance of
short-fliers in dispersal studies, is that, rather than a strategy, this is a limitation of flight-
range experiments. Nonetheless, this type of study yields important knowledge that may
help improve the efficiency of pest management, i.e., chemical controls, pheromone
disruption, and cultural controls aiming for removal of abandoned crop lands (orchards)
and wild hosts (Dom et al. 1999). For example, mating disruption can fail when gavid
females irnrrrigate from areas where they have mated into crop areas to lay eggs. Wild
host removal also becomes ineffective if insect mobility exceeds the distance between the
original and new habitat. Finally, insecticide resistance may develop more slowly if gene
flow is high between insect populations (Dorn et al. 1999).

In the case of E. viteana, there is no information on how far individual moths can
fly and whether differences between sexes and mating status exist. Understanding the
dispersal potential of adult moths is an important component of unraveling the behavior

and ecology of a key pest, such as the insect’s ability to adapt to changing and/or

11

deteriorating environments and colonizing new habitats, all of which may undermine the

effectiveness of management strategies (Keil et al. 2001, Hughes and Dom 2002).

Economic impact of grape berry moth

Historical importance in agriculture. There has historically been such a profusion
of wild gapes in North America, that in 1000 the Norse Viking “Leif the Lucky", upon
exploring New England, called it “Vineland” (Hendrickson 1913). The gape berry moth
evolved in this ecosystem exploiting the gapes and supporting its suite of natural
enemies (Slingerland 1904, Seaman et al. 1990). With the advent of commercial gapes,
first European and then improved American varieties, the gape berry moth had the
opportunity to exploit this new food source with its particular conditions of light,
humidity, and inter- and intra- specific competition. In 1852, the Concord gape was
introduced in Massachusetts, marking the beginning of successful large-scale gape
gowing in Eastern North America. By the early 18608, more than 2,428 ha (6,000 acres)
of gapes were gown east of the Mississippi, and 40 years later, the area had increased to
97,128 ha (240,000 acres) (Hendrickson 1913). Coincident with the increase in gape
hectareage, problems with gape berry moth infestation became more significant and the
object of study by entomologists (and concern by gowers).

Reports from the gape gowing regions of Eastern United States and Southern
Canada show this insect was present in vineyards without causing much damage until
1869. It was first reported as an injurious pest that year, because of yield losses estimated
to be up to 50% in Cleveland and Hudson, Ohio, as well as Missomi, and Southern

Illinois (Johnson and Hammar 1912). Isely (1917) states that the exact pest status of the

12

gape berry moth was unclear since there were few economic estimates of gape loss, and
the pest showed erratic patterns of infestation not only within vineyards but around gape
gowing regions, being particularly limiting in certain areas of Pennsylvania and New
York, and very little in Michigan from where few complaints had arisen. After visiting
Lawton and Paw Paw production regions in Michigan, Isely (1917) confirmed that
although the pest was present everywhere, the small numbers probably had caused its
damage to be confused with black rot in the past. R.H. Pettit first described the
destructiveness of the gape berry moth in Michigan in 1922, though there is no
quantified yield loss (Pettit 1932).

Current agricultural importance. As of 1990, the total area of vineyards in
North-Eastern America, comprising New York, Pennsylvania and the province of
Ontario, Canada, was 32,376 ha (80,000 acres) (Dennehy et al. 1990b). Michigan gape
area is increasing and is currently 6,070 ha (15,000 acres), including gapes gown for
wine in the Northwest of the Lower Peninsula. During recent years, damage from gape
berry moth in some vineyards has exceeded 50% cluster infestation (reaching 99% at
some sites in 2002), a level far above an economic threshold for juice gape production.
Isaacs and others (unpublished) found that yield loss from gape berry moth larval
infestation could amount to two tons per hectare in vineyards when no treatment was
applied. In 2002, levels of infestation by grape berry moth were higher than usual in New
York, Pennsylvania, Michigan and Ohio, and unprecedented load rejections occurred in
the first three of these states due to direct gape berry moth damage or decay associated
with it. In Michigan, this amounted to more than 1,000 tons of gapes worth $300,000

(T. Davenport, pers. com.)

13

Management of grape berry moth. The gape berry moth causes extensive
damage to gapes, not only by directly feeding on the berries, but because larval feeding
infects berries with Botrytis cinerea as has been reported for the European berry moth L.
botrana (F errnaud and Le Menn 1989, 1992), and predisposes them to other pathogens
such as sour rots. Management of this species in the 19205 consisted of application of the
few available products and cultural methods that followed the general wisdom brought
about by behavioral and phenological studies of the pest. Pettit (1932, 1933)
recommended applications of Bordeaux mixture and arsenical poison sprays, including
the “Kedzie Mixture” invented by Dr. R.C. Kedzie, chemist of the Michigan Agicultural
College (Anonymous 1911). Cultural control measures included mounding of earth
around the vines and sanitation of vineyards by removal or destruction of trash and leaf
litter to reduce pupal densities (Smythe 1913). Pettit (1933) pointed out that neither
neglected vineyards nor wild gapes ought to be allowed near vineyards because it could
make gape berry moth pest pressure very intense. This prompted the removal of
abandoned vineyards undertaken by the Michigan Department of Agiculture through the
signing of Act No. 72, Public Acts of 1945, with the help of gowers (Ball and Lovitt
1968). This eradication progam removed 2,966 ha (7,329 acres) of gapes in 12 years
(Ball and Lovitt 1969), and 134.4 ha (332 acres) more between 1968, when the progam
was reinstated, and 1971 (Ball and Lovitt 1971).

The discovery of E. viteana sex pheromone components and the subsequent
synthesis of these components for a lure to bait traps precipitated research aimed at
mating disruption, using the delivery of high and constant quantities of pheromone into a

vineyard (Taschenberg et a1. 1974, Taschenberg and Roelofs 1977). The application of

14

pheromone to crops, successful in many fi'uit systems such as European vineyards with L.
botrana (Sauer and Karg 1998) and other tortricids and leafrollers in a number of
countries (Deland et al. 1994), proved to be a safe alternative because the materials are
innocuous to non-target organisms and the environment at large. Overall, varying results
were obtained, from success in impeding male location of baited traps and reducing fi'uit
infestation in treated vineyards (Taschenberg and Roelofs 1977), to unsuccessful in
reducing pest pressure in treated vineyard edges (Trimble et al. 1991, Trimble 1993).
Several reasons could account for the failure of mating disruption at the edge of
vineyards: improved survival of gape berry moth along borders, adjacent habitats
serving as refuges during pesticide applications, higher wind velocity near the vineyard
edges lowering pheromone concentration in these areas, and the influx of gavid females
from wild areas (Trimble et al. 1991) as was also proposed for L. botrana (Cardé and
Minks 1995, Karg and Sauer 1997). Schmitz et al. (1995) proposed that higher L. botrana
larval densities at edges corresponded with higher mating success due to lower mobility
in searching males. Lewis and Johnson (1999) concluded from their pheromone
disruption experiments that this practice needs to be accompanied by insecticide
applications and must only be used in low to moderate-risk vineyards (Dennehy et al.
1990a, b) exceeding 2 ha (5 acres) in area. In vineyards with high pest pressure and small
area (more border area than interior) pheromone disruption is likely to fail due to the
reasons stated above (Trimble et al. 1991).

Biever and Hostetter (1989) used pheromone-baited traps in a study carried out in
Missouri, and concluded it was a potential aid in the improvement of pest management

progams, when these were tailored to specific vineyards. However, it became obvious

15

before long that although gape berry moth phenology might be well described by trap
captures early in the season, the same did not occur later on during the 2nd and 3rd flights
(Dennehy et al. 1990b). These researchers noted that moth captures did not correlate with
egg-laying in vineyards, possibly because of competition from calling females.

Dennehy et al. (1990b) reported that before the 19808, only 40-70% of New York
vineyards required insecticide applications, which were mainly prophylactic owning to
the low cost of chemical products and high prices paid for gapes. Methods to control the
gape berry moth had focused on the use of insecticides such as methyl-parathion,
azinphosmethyl, and carbaryl (Dennehy et al. 1990b, Nagarkatti et al. 2002b). However,
the Food Quality Protection Act of 1996 brought about the loss of registrations for some
of the most important insecticides. In Michigan, current (2003) recommended sprays
against gape berry moth for conventionally-managed vineyards (Table 1.1) include
azinphosmethyl, phosmet, methomyl, carbaryl, methoxyfenozide. (Wise et al. 2003).
Higher prices of the newer insecticide products have prompted gape gowers to
increasingly consider the integated pest management (IPM) approach which focuses on a
combination of strategies, including pest monitoring, biological and cultural control
(sanitation), new chemical products with different modes of action, less harmful to the
environment and directed more selectively to the pest. Promising reduced-risk

insecticides include methoxyfenozide, spinosad, and acetamiprid.

16

Table 1.1. Typical spray progam for insect management in conventionally-managed
vineyards in Southwest Michigan, indicating the phenological state of gapes at the time

 

 

pests should be targeted.
Grape Phenology Pests Controlled Insecticide
Bud Break Climbing Cutworm Chlorpyrifos or
If scouting reveals Grape Flea Beetle Fenpropathrin.
sigrificant damage.
Pre-bloom Rose Chafer Carbaryl, methomyl,
Grape Berry Moth fenpropathrin or
azinphosmethyl.
Post-bloom Grape Berry Moth Carbaryl, methomyl,
(1“ Cover) - 10—14 days fenpropathrin or
after pre-bloom spray. azinphosmethyl.
Grape Leafhopper Methomyl, imidacloprid or
fenpropathrin.
Bunch closing Grape berry moth Carbaryl, methomyl,
(3"! Cover) - 4 weeks after methoxyfenozide or
post-bloom spray, if phosmet.
scouting reveals a need for Grape Leafhopper Methomyl, imidacloprid or
treatment. fenpropathrin.
Post-veraison Grape berry moth Carbaryl, methomyl or

(5" Cover) if scouting
reveals a need for
treatment.

17

phosmet.

CHAPTER 2:
DISTRIBUTION OF MALE GRAPE BERRY MOTH, ENDOPIZA VITEANA,

IN NATURAL AND CULTIVATED HABITATS

INTRODUCTION

Studies of gape berry moth distribution have usually centered on determining
moth abundance in the vineyard with the help of pheromone traps, and counting infested
gape clusters. Hoffman and Dennehy (1989) found that infestation by Endopiza viteana
was unpredictable from year to year, from vineyard to vineyard, and within vineyards.
Biever and Hostetter (1989) suggested this was due to variation in winter survival and/or
the presence of wild gapes in surrounding woods, from which moths could irnmigate
into the vineyard. The presence of woods near the vineyard is a sigrificant factor in
vineyard risk assessment, and this is assumed to be because they harbor wild gape and
are associated with geater vineyard infestations than at sites where no woods are found
(Dennehy et al. 1990b). However, woods containing Vitis spp. may also provide shelter
and food sources for parasitoids of E. viteana (Seaman et al. 1990, Dennehy et al. 1990b).

By placing traps at the edge of the woods and in adjacent vineyards, Johnson et al.
(1988) found that E. viteana emergence in woods is about a week earlier than at the edge
of the vineyard. In addition to the differential timing of emergence, authors have implied
that more E. viteana are at the edge of vineyards than inside vineyards on the basis of
damage assessments (Dennehy et al. 1990b, Trimble 1993). This pattern suggests geater
abundance of E. viteana in natural habitats than managed habitats (Hoffman and

Dennehy 1989), as occurs with the redbanded leafroller, Argyrotaenia velutinana, in

18

gapes (Biever and Hostetter 1989). Johnson et al. (1988) described early season
abundance at the woods and vineyard edge, with a subsequent shift towards the center of
the vineyard as the season progessed. This pattern was supported by Hoffman and
Dennehy (1989), who trapped gape berry moths at 15 different positions along a transect
from a vineyard, into woods, an alfalfa field, another woods, and another vineyard. They
caught a geater proportion of moths in the woods at the beginning of the season but near
the time of harvest; more moths were caught inside the vineyard than anywhere else
(Hoffman and Dennehy 1989). Similar patterns have been observed by Biever and
Hostetter (1989) and Trimble et al. (1991).

Height is an important consideration for understanding insect distribution and
abundance, particularly when there is considerable difference in canopy height between
habitats in which the insect is distributed (Derrick et al. 1992, Humphrey et al. 1999,
Boiteau et al. 2000a). The woods and vineyard habitats of gape berry moth vary
markedly in their structure and complexity. In vineyards, gapes are on trellises typically
less than 2 m high whereas in woods they climb primarily on deciduous trees up to 25 m.
Thus, a response to habitat structure may be an important component of this insect’s
ecology in these two adjacent habitats.

The relationship between vertical distribution of Lepidoptera and host distribution
has been reported for a few species. Derrick et al. (1992) placed traps at l-1.5 to 3.0 m
high for monitoring European corn borer, Ostrinia nubilalis, in potatoes and corn, and
found that traps placed in the crop canopy caught the highest number of moths. In apple
orchards, captures of the oriental fruit moth, Cydia molesta, increased with trap height

(Peterson 1926); the geatest captures were obtained whenever traps were placed in the

19

fruit zone, irrespective of height (Rothschild and Minks 1977). Riedl et al. (1979) and
Howell et al. (1990) examined vertical variation in captures of codling moth C.
pomonella, though these studies differ in their conclusions. Howell et al. (1990) did not
find sigrificant variation in moth captures with height, but concluded the tree canopy had
the geatest effect because captures depended on whether traps were hung inside the
canopy or on its periphery. Riedl et al. (1979), however, recorded maximum captures of
C. pomonella at geater heights in the canopy. Studies of the effects of trap desigr and
height on captures of the European vine moth (or European gape berry moth) Lobesia
botrana, and the vine moth, Eupoecilia ambiguella led Gabel and Renczés (1985) to
emphasize the importance of adapting sampling to “the ethological and physiological
characteristics of the particular pest”. They compared catches of traps placed at 0.4 and
1.2 m, heights in accordance to vine trunk height. No sigrificant difference was found
between these two heights, but for both species, more moths were captured at 1.2 m
(Gabel and Renczés 1985).

In Eastern North America, the gape berry moth exists in habitats of different
structure, environmental conditions, and host distribution. The study described herein
aimed to determine the relative distribution of this highly specialist herbivore in natural
and cultivated habitats. The specific objectives of this study were to determine: 1) the
vertical distribution of gape berry moth, 2) the horizontal distribution of gape berry
moth across the vineyard-woods ecosystem, 3) the simultaneous vertical and horizontal

distribution of gape berry moth through the season.

20

MATERIALS AND METHODS

This study was conducted in juice gape (Vitis Iabrusca, var. Concord and
Niagara) vineyards in Van Buren County, Michigan. All sites had a history of gape berry
moth and were bordered on at least one side by deciduous woods. Relative moth
abundance was measured using pheromone traps (large plastic delta trap, Suterra LLC,
Bend, Oregon) each baited with a lure containing 0.1 pg of synthetic sex pheromone of
E. viteana (90:10 ratio of (Z)-9-12Ac and (Z)-11-14Ac). Trap inserts were replaced as
needed and pheromone lures were changed monthly, using the same batch of lures for all
traps at each change. All traps were checked weekly and the number of gape berry moth
males recorded. Vineyard management was conventional, following recommended

practices for management of insect pests (Table 1.1).

Vertical distribution

To determine variation in E. viteana abundance with height, two vertical transects
of traps were placed 80-100 m apart on the edge of woods bordering four vineyards, in a
complete block desigr. Traps were suspended by a loop of rope hung from a tree branch
at least 10 m above the gound. Four pheromone traps were hung on each rope at 1.5, 3.0,
6.0 and 9.0 m above the gound. By using a rope at least 27 m long, the highest trap could
be easily lowered, checked, and pulled back to position. The number of male gape berry
moths trapped was recorded weekly fi'om 1 July to 21 October 1999.

In 2000, gape berry moth vertical distribution was sampled next to the woods, on
the gassy 7-14 m wide interface surrounding each vineyard. This area is used by gowers

to manoeuver machinery. Two 10 m tall PVC poles were placed at least 3 m from the

21

woods edge at each of four vineyard-woods interfaces. Each pole was constructed from
PVC pipe, using a 3.3 m x 10.4 cm inside diameter (i.d.) piece connected to two 3.3 m x
9.1 cm i.d. pieces that were joined by an overlapping 70 cm piece of 10.4 cm i.d. PVC.
All pole connections were secured with steel bolts. A horizontal 2 m piece of PVC was
attached to the top of the pole using a T-shaped PVC connector with a 3.3 cm eye-bolt at
one end. This was used to hold the rope carrying pheromone traps, as described above.
The base of each pole was buried 10 cm into the gound, and stabilized with four guy
ropes attached 3 m below the top and tied to 1.2 m reinforced steel bars inserted into the

gound. Traps were checked weekly from 3 June until 3 October 2000.

Horizontal distribution

To determine the relative abundance of gape berry moth in different parts of the
vineyard-woods ecosystem, pheromone traps were placed 1.5 m above the gound
(spaced 80-10 m apart) at four positions between the woods interior and vineyard
interior, at six commercial vineyards bordered by deciduous woods. At each vineyard, a
transect of traps was established at four positions from the woods interior to the vineyard
interior. Three traps were placed 30 m inside the woods, five traps along the edge of
those woods, five traps directly across the interface on the first row of vines, and three
traps 30 m inside the vineyard. Traps were checked weekly fiom 15 April to 21 October
2000. The average number of male E. viteana captured per trap was compared among the
four positions for each of the three flights to determine the temporal change in relative

distribution between woods and vineyards.

22

Vertical and horizontal distribution

During 2001, the vertical and horizontal distribution of E. viteana was
simultaneously compared across the three flights. This was done in two vineyards at each
of four farms, at sites where vineyards were bordered by deciduous woods on at least one
side. At each vineyard, 9.2 m tall steel telescoping poles (Channel Master, Srnithfield,
North Carolina) were placed in each of four positions: at the edge of the woods, on the
interface, at the edge of the vineyard, and 30 m inside the vineyard, each with four
pheromone traps hung at 1.5, 3.0, 6.0 and 9.0 m above the gound (Figure 2.1). Inside the
woods, where the poles could not be erected, loops of rope were hung from tree branches,
at 1.5, 3.0, 6.0, 9.0 m. Traps were also hung further up, at 12.2, and 15.2 m above the
gound, using loops of rope passed over tall tree branches using a bow with an adapted
arrow and a string attached. Traps were installed during the spring, when few obstacles
hindered the arrow’s path and total visibility of the canopy was possible. All traps were

deployed by 19 April and checked weekly until 15 October 2001.

Data analysis

Shapiro-Wilkinson and Kolmogorov-Smirnov tests revealed that raw data were
non-normal, and so all data were transformed (log n+1) to meet the criteria of normality
and homogenous variance among treatments. Main factors tested included height,
position, and flight. All analyses were performed with the SAS progam (SAS Institute
1996). For all sigrificant factors, Tukey’s test was performed to determine differences
between means at the 5% probability level. In vertical distribution experiments, data were

analyzed with a one-way ANOVA using PROC GLM (SAS, Version 8.0). In the

23

horizontal distribution experiment, data were analyzed with a one-way ANOVA using
PROC MD(ED (SAS, Version 8.0), taking flights as repeated measures. Data from the
three-dimensional (vertical and horizontal and time) study were analyzed using an
ANOVA with a two-way treatment structure (height and position) with repeated
measures (flights) using PROC MIXED (SAS, Version 8.0). To test the sigrificance of
differences among positions only, data from the four heights were pooled within

positions and analyzed for each flight.

woods interface vineyard

6%f—LTV A \

 

 

15.2- 1 m If: .

     
    

 

 

 

 

.. ,. r
. ,‘iiilnli‘illll‘ “W ‘iil lilllllllll 11""
12.2- a“ 3111111le '1‘, Wmlmtrn
E H mull Ill ‘::“:irii-‘lllll'i fill‘illi
E) 9.0" af‘lntim'ii“viiii 5‘ l3 s
E. J. H ‘ V , hm .uL s
E ‘ i "iLr.n.mi“"""“‘
3.0- . , “\‘LH {MW M A s _ ‘

 

 

 

Figure 2.1. Schematic representation (not to scale) of a study site in 2001. Pheromone
traps (triangles) were hung at different heights on poles or ropes in five positions across
the vineyard-woods habitats. Grey boxes represent the extent of the two habitats.

24

RESULTS
Vertical distribution

Throughout this experiment, 3,434 moths were trapped at the edge of the woods
moreover moth captures varied sigrificantly with height (F335 = 31.37, P < 0.0001)
(Table 2.1). Traps at 9.0 m caught sigiificantly more moths than traps at lower heights (P
< 0.0001); more than 61% of all moths caught at 9.0 m. This compared to only 9.8%
caught at the typical trap deployment height of 1.5 m. There was no significant difference
in captures between traps at 1.5 and 3.0 m (P = 0.74). When traps were placed at the
vineyard-woods interface, moth captures were low (556 males) and did not vary
sigiificantly with height (F335 = 1.93, P = 0.15) (Table 2.1). The variability in these data

reflects differences in populations among the four farms.

Table 2.1. Average number of Endopiza viteana caught at four different heights at the
woods edge and the woods-vineyard interface in two different years. Means within a
column followed by the same letter are not sigiificantly different (Tukey or = 0.05).

 

Mean t SE moths caught per trap

 

 

Trap height Woods edge Vineyard-woods interface
(In) (1999) (2000)
1.5 42.0 i 13.6c 4.5 i 7.3a
3.0 43.5 i 14.9c 3.7 21: 3.3a
6.0 79.1 4.: 26.0b 3.2 i 4.3a
9.0 264.6 i 93.8a 1.7 21: 7.0a

 

25

Horizontal Distribution

The number of male moths captured varied according to the position of the traps
in the vineyard-woods system and time of season. Almost three times more moths
(23,275) were caught in the woods than in the vineyard (8,453). During Flight 1, 84% of
moths were caught in the woods, whereas 52% and 49% were trapped in the woods in
flights 2 and 3, respectively, indicating variation in distribution between habitats over
time. Twelve thousand and fifty eight moths (38% of the moths trapped all season) were
caught in F arm 2, and contrary to the trend observed in the other five farms, consistently
fewer moths were trapped inside the vineyard during all three flights at this farm.
Analysis that included Farm 2 showed similar trends in abundance among positions to
analysis excluding it, but because of the numerical difference between Farm 2 and the
other five farms, the normality assmnption could not be met. Therefore, F arm 2 was
excluded from further analysis.

In the five remaining farms, total moth captures were sigrificantly different
among the three flights of 2000 (F2,109 = 132.16, P < 0.0001) and there was a sigrificant
interaction between positions and flights (Fares = 8.76, P < 0.0001) (Table 2.2). During
Flight 1, 71.5% of moths were caught in the woods habitat, a proportion that decreased to
46.8% during Flight 2, and to 27.4% in Flight 3. Although the difference in relative
abundance of gape berry moth was not sigrificant among vineyard positions, more
moths were captured inside the vineyard than at the vineyard edge for Flights 2 and 3

(Table 2.2).

26

Vertical and Horizontal Distribution

Captures of gape berry moth varied sigrificantly across habitats (positions) (F497
= 63.94, P < 0.0001) and during each individual flight (F2306 = 56.82, P < 0.0001)
(Figure 2.2), while the interaction between position and flight was sigrificant (F 33 ,206 =
4.04, P < 0.0001). Combining captures of E. viteana between 1.5 and 9.0 m, I observed a
1.9-fold decrease in moth capture from Flight 1 (total of 2,994 moths) to Flight 2 (1,561)
and a subsequent 2.6-fold decrease to Flight 3 (603) (Table 2.3). During Flight 1, 86% of
the moths were caught in the woods habitat (both inside and edge) compared to only 10%
caught in the vineyard habitat (both edge and inside). However, during Flight 2, the
proportion of moths trapped in the woods habitat decreased to 57% and increased in the
vineyard habitat to 40%. Captures were similar in both habitats during Flight 3, with 45%
of captures in the woods and 49% in the vineyards. Simultaneous sampling of E. viteana
adults across horizontal and vertical gadients confirmed the pattern observed in 2000.
Captures in the interface were low throughout the season (Figure 2.2), and more moths
were trapped inside the vineyard than at the edges. This difference was sigrificant in

Flights 2 and 3 (Table 2.3).

27

Table 2.2. Average number of Endopiza viteana caught per flight in traps placed at a
height of 1.5 m in four different positions within the habitats sampled during 2000.

Means within a row followed by the same letter are not sigrificantly different (Tukey or =
0.05).

 

Mean 5: SE moths caught per trap

 

WOODS INTERFACE VINEYARD
Flight Inside Edge Edge Inside

 

1 254.5 :1: 44.2a 68.3 i 31.0b 14.0 i 4.4b 7.6 i 2.1b 30.1 i 7.1b
2 81.5 i13.1a 28.9 i 4.2b 6.5 :t 1.3b 13.0 i 3.4b 65.3 i 11.02!
3 17.8 i- 6.lab 16.4 i 3.4ab 4.1 :l: 1.8b 10.0 i 2.2b 27.1 :1: 4.4a

 

Table 2.3. Average number of Endopiza viteana caught in traps per position (1.5-9.0 m
heights pooled), during three flights sampled druing 2001. Means within a row followed
by the same letter are not sigrificantly different (Tukey or = 0.05).

 

Mean i SE moths caught per trap

 

 

WOODS VINEYARD
thht Inside Edge Edge Inside
1 211.9 i 61.3a 184.4 1 32.9ab 79.1 :t 19.3b 75.1 i 25.2b
2 68.9 i 22.4a 102.2 i 29.7a 77.5 i 18.9a 142.5 1 27.8a
3 3.7 i 1.0b 10.0 i 2.3ab 13.3 i 2.9ab 31.9 i 9.8a

 

28

120

     
   

100 Flight 1

.n
,i

80
60
40
20 1.5 m
a. O 3.0 m
(B 6.0 m
5 9.0 m
h 120
8
m 100 Flight 2
5 80
C)
£5 60
5 40 a
w . V‘ .23
:5 20 # - t5n1
0 D 3.0 m
- 6.0 m
120 0 m

  

§

Flight 3

8

Figure 2.2. Average number of male Endopiza viteana caught per trap during each flight

in 2001, trapped at four heights across five positions in the vineyard-woods
agoecosystem.

29

When relative abundance was compared among heights, some clear patterns were
seen (Figure 2.2). Moth captures varied sigiificantly according to height (F 3,97 = 25.12, P
< 0.0001), and this variation was sigrificantly influenced by flight and position (F 33 ,206 =
4.04, P < 0.0001). The capture of moths at different heights depended on where (position
and habitat) the trap was placed (at 1.5 m F437 = 19.89, at 3.0 m F497 = 7.96, at 6.0 m
F497 = 40.35, at 9.0 m F437 = 75.79; P < 0.0001). At all positions except the interface
(F397 = 1.63, P = 0.19), there was a sigrificant variation in moth relative abundance
among heights (inside woods F397 = 14.64, at the woods edge F 3,97 = 10.20, at the
vineyard edge F397 = 31.40, and inside the vineyard F397 = 73.96; P < 0.0001); the
geatest number of moths was caught in the higher traps in the woods habitat and in the
lowest traps in the vineyard habitat (Figure 2.2).

When captures at 1.5 m were considered separately because of their relevance to
monitoring for this insect, more moths were always captured inside the vineyard than
inside the woods (Figure 2.2), though this difference was not sigrificant during the first
flight. At the edge and interior of the vineyard, sigiificantly more moths were trapped at
1.5 m than at any other height (Tukey P = 0.0068 between 1.5 and 6.0 m, P = 0.0013
between 1.5 and 9.0 m at the vineyard edge, and P < 0.0001 for the same comparisons
inside the vineyard), with very few moths found in traps placed above the canopy (Figure
2.2). When the firll height of the tree canopy in wild habitats was taken into
consideration, the geatest captures of moths were made in traps at 12.2 and 15.2 m high.
Indeed, moth captures increased sigrificantly with height of trap inside the woods (F 5 ,29 =
21.76, P < 0.0001), so that 76.1% of moths captured in the vertical transects within the

woods were caught at or above 9.0 m (Figure 2.3). This change in relative abundance

30

with height in the woods was consistent across flights (Figure 2.3), and the change among

flights was consistent with results during 2000 (Table 2.2).

 

200 -

El 1.5m
E] 3.0m
6.0m
I 9.0m
I12.2m
a I15.2m

a

 

 

 

8

 

Mean :I: SE moths per trap
e

 

 

 

 

 

 

 

Fligm Fligit2 Flights

Figure 2.3. Average nrunber of male Endopiza viteana per trap during each flight,
trapped at six heights inside woods adjacent to vineyards during 2001. Within each flight,
bars with the same letters are not sigrificantly different (Tukey or = 0.05).

31

DISCUSSION

This study found differential spatial and temporal distribution of gape berry moth
across the vineyard and woods ecosystems. Relative abundance of gape berry moth was
found to vary sigiificantly with the type of habitat in which this insect was sampled, and
with the height at which samples were taken. The geatest captures of E. viteana males
were made during Flight 1 and captures decreased for successive flights. There are
several possible explanations for this reduction. Harnstead et al. (1972) found a similar
pattern in A. velutz'nana, and suggested that early in the season lower temperatures
favored traps over sexually mature females, whose release of pheromone was reduced.
Another explanation could be that diapause frequency in E. viteana increases after June
25 (Flight 2 and 3) as day length shortens (Nagarkatti et al. 2001). Tobin et al. (2002)
suggested that E. viteana is protandrous, which would explain abundant captures of males
early in the season when only the lures inside traps are releasing pheromone. The
decreasing captures of males in successive flights have been explained by increasing
abundance of virgin female moths, which increasingly compete with the pheromone traps
as the population gows through the season (Howell 1974, Hoffman and Dennehy 1989,
Dennehy et al. 1990b, Aslarn et al. 1990).

Sampling across vineyard-woods habitats throughout the season showed that
captures of E. viteana vary sigrificantly with sampling position and trap height, a trend
that is likely due to the response of this species to the structure and composition of its
habitat. My results obtained fi'om traps at 1.5 m agee with previous findings (Hoffinan
and Dennehy 1989, Lewis and Johnson 1999), but by placing traps between 1.5 and 9.0

m above the gound in these different habitats, I have shown that a majority (90.2% in

32

1999) of E. viteana males in woods are consistently distributed above the typical height
for trap placement (Table 1; Figures 2.2 and 2.3). Hoffman and Dennehy (1989) found
that pheromone traps placed at wooded edges captured few moths even though high
numbers of eggs were deposited in the same area on wild gapes, suggesting that male E.
viteana emigate from areas of oviposition activity. The number of times female E.
viteana mate is not known, but if they usually mate only once as found for L. botrana
(Torres-Vila et al. 1997), emigation from areas of oviposition would improve a male
moth's chance of locating virgin females. However, in view of my results, previous
studies using traps within easy reach have probably missed moths flying high in the
woods. Traps placed at different heights in the woods canopy caught more moths high in
the woods than did traps placed at 1.5 m in the vineyard. This suggests that rather than a
shift in abundance of this species from the woods to the vineyard as the season
progesses, abundance is geatest in the higher canopy. Relative abundance between
habitats is geatest in the woods throughout the season, but because of the typical 1.5 m
monitoring position used, this has remained unnoticed.

Low captures of E. viteana in traps placed in the vineyard-woods interface in
2000 and 2001, coupled with the similar captures at different heights (Table 1; Figure
2.2.) indicate that the lack of host plants in this position provided no host substrate to
which the moths could respond. In vineyards, the geatest moth captures were
consistently within the canopy; few moths were captured above 1.5 m. Taken together
with the results in woods described above, these results agee strongly with the
suggestion of Hoffman and Dennehy (1989), that E. viteana distribution is tightly

coupled to the structure of the habitat where its host is present.

33

Four species of wild Vitis are commonly found throughout wild and perturbed
habitats in the Eastern United States (Morano and Walker 1995). I took samples from the
sites studied and followed a simple key by Voss (1985) which showed the wild gape
species present were principally V. riparia, V. labnrsca and V. aestr'valis. At the edge of
woods, vines gow on border trees, sometimes covering them from the gound to the
canopy top. Inside the woods, they gow on trees, developing into the canopy where light
intensity is geatest. The majority of fruiting occurs at this height, typically 16-18 m high
in the deciduous woods surrounding Michigan vineyards (N. Botero-Garcés, unpublished
data). The variation in captures of male moths with height may be a response to canopy
height, fruit distribution, or virgin female distribution. Correlations between fi'uit moth
abundance and canopy height have been described before for the two gape pests E.
ambiguella and L. botrana (Gabel and Renczés 1985), for C. pomonella (Riedl et al.
1979, Howell et al. 1990) and for C. molesta (Rothschild and Minks 1977). Vertical
distribution of foraging insects can be tightly linked to resource vertical distribution
(Muirhead-Thompson 1991, Cisneros and Rosenheim 1998) and there can be species-
specific (Nyrop and Simmons 1986) and family-specific (Taylor 1974, Humphrey et a1.
1999, Boiteau et al. 2000a, b) vertical distribution patterns driven by dispersal, foraging,
mating and oviposition behaviors.

There is an adaptive benefit to behaviors that maximize abundance of male E.
viteana in regions where gape clusters are numerous, because female oviposition is
strictly on this resource (Clark and Dennehy 1988). Vertical distribution of eggs within
the vineyard canopy is closely correlated with fruit density (Clark and Dennehy 1988)

and so mated female E. viteana are assumed to be most abundant near their oviposition

34

substrate, as predicted for specialized herbivores (Miller and Strickler 1984, Hamilton
and Zalucki 1993). It is not known whether virgin female E. viteana release pheromone
only when on gape clusters but the likelihood of males finding females is assumed to be
geatest if they are in proximity to this oviposition site. Male C. pomonella are trapped in
much geater numbers in pheromone traps placed in the host canopy compared to those
placed outside the canopy (Howell et al. 1990). Riedl et a1. (1979) argued that more C.
pomonella males were caught in traps placed in the higher tree canopy because mating
occurred near the canopy top. In another polyphagous insect, the tortricid Archips
podana, morphological and temporal heterogeneity of populations is tightly related to
larval food preference (Safonkin 1988). Evidence for larval habitat directly influencing
mating behavior of adult moths has recently been described by Takacs et al. (2002) with
webbing clothes moths. In this case, males seek larval habitats and produce pheromone
and sonic sigrals to enhance recruitment of females to a patchy andtemporary resource.

X My vertical sampling results complement the study by Hoffman and Dennehy
(1989) by showing that moths remain abundant high in the woods canopy throughout the
season. These findings can help answer questions posed by these and other researchers
(Dennehy et al. 1990b, Trimble 1993, Lewis and Johnson 1999) of why few male E.
viteana are trapped in woods adjacent to vineyards with high levels of cluster infestation
and why pheromone disruption is less effective at vineyard borders (Taschenberg et al.
1974, Trimble et a1. 1991, Karg and Sauer 1995). Explanations of high larval infestations
where few male moths have been caught have centered on mated females flying into the
vineyard to lay eggs, both in E. viteana (Taschenberg et al. 1974, Biever and Hostetter

1989, Trimble et al. 1991) and in L. botrana (Karg and Sauer 1995). My findings show

35

that a large proportion of the adult population of E. viteana is in areas outside those
targeted by management progams, reinforcing the need to consider the whole landscape
when studying the ecology of native insects (Burel et al. 2000) and tortricids in particular
(Barrett 2000). This approach will also be of value when considering enhancement of
biological control (Wratten and Thomas 1990, Marino and Landis 2000), or cultural
practices such as removal of wild hosts to reduce the impact of gape berry moth on gape
production.

Movement of insects between wild and cultivated habitats has been reviewed by
Macdonald and Smith (1990), Woiwod and Stewart (1990), and Ekbom (2000).
Schumacher et a1. (1997) stated that both mated and virgin female C. pomonella are
capable of movement between orchards, with important implications for pest
management strategies such as pheromone disruption and resistance management (Dom
et al. 1999). Trimble (1993) concluded that high levels of larval infestation by E. viteana
at vineyard borders could be due to mated females entering the vineyard from woods to
lay eggs, but direct movement of E. viteana has as yet to be conclusively demonstrated
Discovery of a female attractant, as recently described for C. pomonella (Light et al.
2001), would geatly assist in determining the sigrificance of immigation by mated

female moths from wild gape into adjacent vineyards.

36

CHAPTER 3:
INFLUENCE OF UNCULTIVATED HABITATS AND NATURAL HOST
PLANTS ON CLUSTER IN FESTATION BY GRAPE BERRY MOTH,

ENDOPIZA WTEANA, IN MICHIGAN VINEYARDS

INTRODUCTION

Insects require varied resources, such as food, water, oviposition substrates,
shelter, and overwintering hosts, and these may not be available within a single habitat
(Kareiva 1983, Toepfer et al. 2002). An ecosystem comprised of multiple habitats will be
required for gowth of insect populations that need multiple resources to complete
development. Habitat heterogeneity is not only important for resource availability and
consequently insect conservation (Welch 1990, Woiwod and Stewart 1990), it may also
contribute to keeping insect pest populations below economic thresholds within
agicultural landscapes (Altieri 1983, Macdonald and Smith 1990, Welch 1990, Banks
2000, Ekbom 2000) by increasing the effectiveness of ‘ natural enemies (Wratten and
Thomas 1990, Williams and Martinson 2000). However, uncultivated habitats may also
harbor insect pests and disease due to movement from wild hosts (Barrett 2000, J eanneret
2000). Thus, pest management strategies that consider movement of insects within this
habitat mosaic may be more effective than those desigred solely for the agicultural
setting (Landis 1994).

Studies on the gape berry moth, E. viteana have used pheromone traps as a tool
to understand pest behavior and phenology. Comparisons of relative moth captures at

different positions within vineyards have been reported by several researchers

37

(Taschenberg et al. 1974, Biever and Hostetter 1989) who pointed out that more E.
viteana males are typically caught at borders versus interiors of vineyards. Hoffman and
Dennehy (1989) used 15 traps to sample moths in two vineyards and two types of
adjacent habitats (two woodlots and one alfalfa field). They found more moths were
captured in the woods but only at the beginning of the season, since by mid-June, male
moth abundance in the vineyard had increased and exceeded abundance in woodlots. In
contrast, few moths were caught in alfalfa fields nearby (Hoffman and Dennehy 1989).
This study also noted that infestation of gapes by E. viteana larvae seemed geater at
vineyard edges than interiors, as reported elsewhere by Biever and Hostetter (1989) and
Trimble et a1. (1991). Taschenberg et al. (1974) first suggested this could be due to
immigation of gavid females from wild areas, thus proposing a link between managed
and uncultivated (or wild) habitats. Based on this, Dennehy et al. (1990a) developed a
risk assessment progam for New York state vineyards, in which risk from gape berry
moth was based partly on the proximity to woods or tree rows (Hoffman and Dennehy
1989, Dennehy et al. 1990a, b, Martinson et al. 1991).

Vineyards across Eastern North America are surrounded by a range of different
uncultivated habitats. Most of them are suitable for wild gapevines, yet the effect of
these habitats on vineyard cluster infestation by E. viteana larvae has not been reported
previously. There is also little information on how abundance of wild gape varies with
uncultivated habitat type, and the relationship between these wild gapevines and
vineyard infestation by gape berry moth. The goal of the present study was to determine
the relationship between the type of adjacent uncultivated habitat and vineyard infestation

by E. viteana, focusing on the role of wild gapevines. Specific objectives were to

38

compare the relative abundance of gape berry moth across habitat types, determine the
effect of habitat type on abundance of wild gape, and determine whether vineyard
infestation by this insect is predicted by the presence, density, or fi'uiting of wild

gapevines in adjacent habitats.

MATERIALS AND METHODS

This study was conducted in commercial juice gape (Vitis labrusca, var. Concord
and Niagara) vineyards in Van Buren County, Michigan, during 2001 and 2002. All
vineyards were approximately 1.6 ha, with at least one side bordered by an uncultivated
habitat that was one of the following four types: deciduous woods, coniferous woods, a
single row of mixed trees (a tree row) 8-12 m wide, or gasses. All uncultivated habitats
sampled, except for the gasses, consisted of mostly mature trees established at least 20
years earlier. The number of replicated sites (vineyard with adjacent uncultivated habitat)
for each habitat type ranged from four to eight in 2001 and from seven to nine in 2002.
Conventional vineyard management practices were implemented by the gowers at all

sites following recommended management measures against insect pests (Table 1.1).

Vineyard infestation

Infestation by gape berry moth was assessed non-destructively by visual
inspection of gape clusters once a month during July, August, and September of 2001
and 2002. A cluster was considered infested if one or more gapes was observed with
symptoms (entrance hole, frass, webbing, larvae) of gape berry moth. Percent cluster

infestation was determined by examining 20 clusters per vine, on five vines at the

39

vineyard border (100 clusters) and five vines at the interior (another 100 clusters), for a
total of 200 clusters per site (Martinson et al. 1991). This procedure was carried out in the

26 vineyards studied in 2001 and the 30 vineyards in 2002.

Moth abundance

At each site, relative moth abundance was measured using four large plastic delta
traps (Suterra LLC, Bend, Oregon) baited with lures containing 0.1 pg of synthetic sex
pheromone of E. viteana (90:10 ratio of (Z)-9-12Ac and (Z)-11-14Ac), lined with sticky
inserts. At each site, a trap was placed 1.5 m high at a position 30 m inside the vineyard,
at the edge of the vineyard, at the edge of the uncultivated habitat, and 30 m inside it
(Figure 3.1). Since tree row habitats were long and narrow (8-12 m width), only one trap
was deployed in this habitat and the position was considered a border. Trap inserts were
replaced as needed and pheromone lures were changed monthly using the same batch of
lures for all traps. Traps were set before the first gape berry moth flight and checked
weekly for the number of moths caught until vineyards were harvested in September. The
cumulative total capture of male E. viteana was determined for each trap at each date that

infestation was assessed.

Wild grape survey

During 2001 and 2002, the presence (or absence) of wild gapevines and wild
gape clusters within the uncultivated habitat was determined in an area 38 m wide by 63
m deep in woods and gasses (2,394 m2), and 15 m deep in tree row habitats (570 m2).

This was done by examining the vegetation in the uncultivated habitats and rating each

40

site with 1 for present or 0 for absent. In 2002, a more extensive survey was carried out to
measure the relative abundance of wild gapevines within the uncultivated habitats
studied, adapting methods of Matteucci and Colrna (1982) for vegetation sampling. At
each site (Figure 3.1), five trained observers were spaced 8 m apart along the edge of the
uncultivated habitat, so that the middle observer was standing next to the pheromone trap
placed at the border. Each observer recorded wild gape presence (1) or absence (0) by
looking in three directions (samples): to their left, in front of them, and to their right,
including the tree canopy, within a 3 m radius. Measurements were taken every 6 m (a
station) until observers reached 60 m inside the uncultivated habitat, for a total of 165
samples (11 stations x 3 samples per observer). In the case of tree rows, the maximum
depth sampled was 18 m, to ensure that both sides of each row were inspected.

A wild gape index describing the abundance of wild gapevines in the
uncultivated habitats was calculated by adding numbers recorded at each station
(minimum 0 and maximum 3 per station) to provide a measure of the number of times
wild gape was observed in each uncultivated habitat site. This sampling was only carried
out in 2002 because wild gapes are perennial and their presence was unlikely to change
in consecutive years.

To determine whether the risk of infestation by gape berry moth could be
predicted by sampling wild gapevine abundance only at the borders of uncultivated
habitats, part of the above data was used to calculate a restricted wild gape index for
each site. This was based on the samples collected within a 38 m wide x 15 m deep area
of each adjacent border habitat (45 samples per site), providing a standard sampling area

to adequately compare uncultivated habitats.

41

Uncultivated habitat

      
      
 
   

     
   
      
 

1%"
rill, ‘ it” rill” 'Vl‘v‘w ‘
. rill {mill NM it‘ll‘ii ‘
i“‘l‘ullllllllllllllllllllmlllllll“ “minim” will i “ii

lull“! till it“

  

, l “till

1
l mimi‘ililllllllmmillilllllmm

:‘i‘vl

llllllllllllllllll mill“

 

       

     

 

 

60
Position of wild grape
samples
(every 6 m from edge up to 60m)

Figure 3.1. Schematic representation of an experimental site (not to scale), with
pheromone traps (triangles) at four positions across the vineyard-uncultivated habitat
transect. The vertical double line indicates the width of the tree row habitat; the interface
width ranged from 5-10 111 depending on the vineyard. Wild gape surveys were done by
five observers Gilack ovals) following a 60 m transect (horizontal dotted lines). Counts of
wild gape were recorded every 6 111 along these transects.

Data analysis

One way ANOVA was used to test the effect of uncultivated habitat type on
cluster infestation by E. viteana and the cumulative number of moths trapped at each
position on each sampling date, using PROC GLM (SAS, Version 8.0). Moth counts
were log (n+1)-transformed and values of percent cluster infestation were arcsine-
transformed to homogenize variance between treatments. Wild gape indices were

compared between habitat types with the GLM Procedure (SAS, Version 8.0). For all

42

sigrificant factors, Tukey’s test was used to determine differences between means at the
5% probability level. Habitats were compared for frequency of wild gapevine presence
and for presence of fruit using the Kruskal-Wallis nonparametric one-way procedure
(SAS, Version 8.0) for rank data. This test’s assumptions are that the samples are
independent and that the variable studied has a continuous distribution and requires that it
is an ordinal measurement (Siegel 1956). Correlations between vineyard infestation
levels and presence of wild gape and wild gape clusters were also analyzed using the
Kruskal-Wallis test (SAS, Version 8.0). Regessions between vineyard cluster infestation
and cumulative moth captures in vineyards and habitats adjacent to them, and between
cluster infestation and wild gape indices were obtained with the REG Procedure (Model
1) (SAS, Version 8.0). All analyses were performed with the SAS progam (SAS

Institute, 1996).

RESULTS
Vineyard infestation by E. viteana

In 2001, levels of infestation by gape berry moth at vineyard borders varied
sigrificantly depending on the type of neighboring uncultivated habitat in July (F332 =
3.71, P = 0.03), August (F332 = 5.69, P = 0.005) and September (F332 = 4.46, P = 0.01)
(Table 3.1). Percent larval infestation was sigiificantly geater in vineyards near
deciduous woods than near gasses (P = 0.02, P = 0.003, and P = 0.01, for each month
respectively). Vineyard interior infestation levels showed similar trend (Table 3.1), but
with no sigiificant difference between habitats (F 332 = 1.43, P = 0.26 for July; F332 =

1.43, P = 0.26 for August; F332 = 1.24, P = 0.32 for September).

43

 

 

2.: H Ném 5d H man «Wm H ficv awd H NS Exam H wém .54. H mdm 8380
3.x H Q: new; H «23 «m6 H odn «Wm H man :36 H 0.2... flaw H oém BE 08,—.
no.2 H fine namm H «Mme .66 H Nam «2 H wnow AWN H 0.2 £6 H Dom €83 380.380
«m6 H Qwh «ed H «.8 3.5 H 53 3...: H v.3. a: H mdm «mm H awn €83 382qu
NOON

3.0 H :N Amfi H q; .3.— H vd And H v.3 and H 5v 5N H ”mm 3330
«ad H fiov £3: H NE. wad H m.~_ 56.: H Nmm 3N H Qw sand H Wm 38 and.
.56 H flow cams H ode .56 H o.m_ namé H o.w~ am.~ H m6 :3: H m2 383 380.380
and H «Km and H Nmn and H 9mm and H «.3 SN H ma: no.0 H oém €83 weed—Econ
Sou

cocoa: 520m 85:: Spam 85:: 820mm 585mm “Sega
confioaom unawa< 98 30>

 

ass»? a ea 8:98? asap am i as:

.God H 5 >835 “cocobmu banana»? “on 2d .832 2:3 05 3 326:3

5:28 a 55:5 £605 50% coco com .88 can Sow wet—6 .80» 05 mo 88: ooh: an :83 083 $383 .3853 c8a>£3§ mo
womb .88 8 “abounds 3.855.» mo mcoummoa BEBE cam bacon 3 23.5 288.3 summonam .3 €838 38w Co 553on 4.». 039—.

In July 2002 border infestation levels were significantly affected by the type of
neighboring habitat (F 3 ,26 = 10.24, P < 0.0001) as in the previous year, and in addition,
the interior infestation in July varied with adjacent habitat (F 3 ,26 = 4.05, P = 0.02) (Table
3.1). Vineyards near deciduous woods had more infested grape clusters at borders than
vineyards near grasses (P = 0.0003), tree rows (P = 0.001), or coniferous woods (P =
0.001). The difference between the two types of woods also was significant in interior
levels of infestation (P = 0.02). In August, uncultivated habitats had no significant effect
on either border (F336 = 1.33, P = 0.28) or interior (F336 = 1.83, P = 0.17) vineyard
infestation levels. In September, as in the previous year, vineyard borders had
significantly higher infestation levels near deciduous woods than near grasses (P = 0.02)
(F 3 36 = 3.20, P = 0.04), but this effect was not found at vineyard interiors (F 3 ,26 = 1.11, P

= 0.36).

Abundance of male E. viteana

Moth abundance accumulated through the season was similar in all vineyards
regardless of the type of uncultivated habitat adjacent to it, both in borders (F 3 .22 = 0.30,
P = 0.83 in 2001; F 3,26 = 0.96, P = 0.43 in 2002) and interiors (F322 = 0.84, P = 0.54 in
2001; F 3 ,26 = 0.24, P = 0.87 in 2002) (data not shown). Data analyzed individually by
month showed no significance either (P > 0.63 for borders and P > 0.27 for interiors in

2001, P > 0.33 for borders and P > 0.41 for interiors in 2002).

45

 

 

 

n56 H m.: .56 H E: Ham H _.: flan H wd aw.m H 2: AWN H Nm 8895
- awéw H 9va - «W: H gum - 55.3 H 9;— 26.— 02H
3.3” H fiavm 3.: H 5.0.3 3.2 H QEN «use H Ramon . 3.3 H m4 3 3.3 H wdo— €003 msohquU
3.8 H @va «Qmm H Dog «Won H ZEN 3.2 H Evo— amén H Ymmm 3.3 H 0.2: €003 95:28.9
Noom
«NV H .2. .36 H fin a: H fie £6 H Q4» 5.6 H fie 36 H We 8380
- .392 H 5.; - assum H 03 - 53mm H N2 38 BC.
.56 H finm «5.3 H finv «Nd H fwm 3.0 H fiwm nod H 5.2.. and H oéw 8003 388:50
«Ngmm H _.om 22.3 H flow namdm H adm nam.m~ H N; na~.mm H flew nam.m~ H 0.2. 383 35:2qu
Sow
Stun: 390m Stun: 890m 3185 320m
“8. a 308..
Honfioaom “mama/x 33. .n : b<
98 30>
@393 338 .«c .3585: gum—=83 Amm Hv 582
h I

 

 

 

3; u 5 >23: Boga Haggai
3a 2a .532 088 05 3 Bko=£ Ens—co a £55 £808 .39» :80 Ba do @2953 33 53882 $53 025 2: 3m cocoa»:
0.8 32? 033383 28 £938 :08 mega >383 :83 825 835mm Eggnog“; 388888 9 «538a 3852 3:32:23
mo 890 .58 MO 30363 SEEM ES .820: 8 moom ES SON macaw 9.33 382.3 QNESSQ 29: .«o con—«ugh? 0333M .N.m 03:.

46

However, moth abundance varied significantly among the uncultivated habitats
(Table 3.2). In July 2001, moth captures at the border (F 3,22 = 3.63, P = 0.03) and interior
(F 2 22 = 5.06, P = 0.02) of uncultivated habitats were significantly different according to
habitat type, similar to August (F 3 22 = 3.83, P = 0.02; F 2 22 = 5.14, P = 0.02, respectively)
and September (F 3 22 = 3.28, P = 0.04; F2 22 = 5.02, P = 0.02). Fewer moths were caught
in borders of grasses near vineyards than in borders of coniferous woods (P = 0.04 in
July, P = 0.03 in August) (Table 3.2). Similarly, in 2002 moth captures at borders and '
interiors of uncultivated habitats varied significantly (borders: F = 27.59 for July, F =
26.03 for August, F = 26.86, for September with df = 3,26 and P < 0.0001 for all;
interiors: F = 43.09, F = 41.88, F = 40.31, respectively, df = 2,26 and P < 0.0001 for all).
The lowest moth abundance was in grasses compared to the other uncultivated habitats (P
< 0.0001 for each comparison in July, August and September) (Table 3.2).

A regression analysis between the total moth captures in uncultivated habitats and
mean vineyard infestation (average of border and interior levels) showed a positive but
weak relationship in September 2001 and August and September 2002 (Table 3.3a).
There also was a weak negative relationship between mean vineyard infestation and moth

abundance in vineyards (Table 3.3b) during both years, except in July 2001.

Survey of wild grape

The probability of wild grapevines being present was similar among the different
uncultivated habitats (Kruskal-Wallis 2% = 1.78, df= 3, P = 0.65 for 2001; 2’ = 4.11, df=
3, P = 0.25 for 2002) (Table 3.4). However, the frequency with which wild grapevines

bore clusters in these uncultivated habitats varied with habitat type (Table 3.4) (12 = 14.0,

47

df = 3, P = 0.003 for 2001; 2? = 10.07, df = 3, P = 0.02 for 2002). As measured by

Wilcoxon rankings, wild grapevines with fruit clusters were more frequently found in

deciduous woods, followed by tree rows then coniferous woods (where no fruit clusters

were found in 2001). Wild grapevines in grasses were never observed with fruit clusters.

Table 3.3. Values obtained for regression analysis of cluster infestation in vineyards and
cumulative number of male Endopiza viteana adults captured in a) uncultivated habitats
and b) in vineyards. Significant regressions are displayed in bold.

 

 

 

 

 

8)

Year df Slope
2001

July 1,46 0.73 0.39 0.02 -0.03
August 1,46 1.42 0.24 0.03 0.05
September 1,46 14.91 0.0004 0.24 0.20
2002

July 1,51 2.91 0.09 0.05 0.07
August 1,51 14.14 0.0004 0.22 0.15
September 1,51 10.93 0.002 0.18 0.14

b)

Year df Slope
2001

July 1,50 3.27 0.08 0.06 -0.06
August 1,50 5.53 0.02 0.10 -0.11
September 1,50 8.85 0.004 0.15 -0.22
2002

July 1,58 23.14 0.0001 0.28 -0.32
August 1,58 4.84 0.03 0.08 -0.23
September 1,58 8.36 0.005 0.13 —0.30

 

48

Wild grape index (W GI) values varied significantly among habitats (F 3 22: 7.81,
P = 0.001 in 2001, and F 3 26 = 7.21, P = 0.001 in 2002, Table 3.4). There were more wild
grapevines in deciduous woods compared to tree rows (P = 0.007 in 2001, P = 0.001 in
2002) and grasses (P = 0.001 in 2001, P = 0.006 in 2002). Although the difference in
WGI values was not significant between the two types of woods (P = 0.10 in 2001, and P
= 0.34 in 2002), wild grapevine abundance was greater in deciduous woods (Table 3.4).
The restricted wild grape index (RWGI) values showed a similar trend to the WGI
values, but with a lower magnitude (Table 3.4). Values varied significantly between
habitats (F 3,22= 3.28, P = 0.04 in 2001, and F 3,26 = 4.31, P = 0.01 in 2002) and there were
more wild grapevines in deciduous woods than any other habitat, though it was only

significant when compared to grasses (Table 3.4).

Table 3.4. Parameters measured in the 2002 survey of wild grape, Vitis spp., within four
types of uncultivated habitats adjacent to vineyards. For wild grape indices, means within
a column followed by the same letter are not significantly different (Tukey a = 0.05).

 

Wild grape Wild grape Restricted wild

 

Type of Wild grape . .
clusters index grape mdex

uncultivated presence

. n presence (WGI) (RWGI)
habitat
Deciduous woods 8 100% 75% 45.011 12.411
Coniferous woods 6 100% 33% 29.0ab 8.5ab
Tree row 7 71% 43% 7.1b 6.7ab
Grasses 9 89% 0% 14.0b 2.1b

 

49

Relationship between wild grape and grape berry math infestation

The presence or absence of wild grapevines in uncultivated habitats was not
predictive of vineyard infestation during any of the sampling times in this study, neither
at vineyard borders (Kruskal-Wallis df = l, P > 0. 21 in 2001; df = 1, P > 0.36 in 2002)
nor vineyard interiors (df = 1, P > 0.33 in 2001; df = 1, P > 0.08 in 2002). On the other
hand, the presence of fruit clusters on wild grapevines in uncultivated habitats was found
to be related to vineyard infestation by grape berry moth larvae during some of the
sampling times of both years (Figure 3.2, Table 3.5). In 2001, infestation levels in
vineyards were not correlated with WGI of the uncultivated habitat adjacent to them,
neither at borders (df = 1,24: F = 4.14, P = 0.05 for July; F = 3.50, P = 0.07 for August;
F = 1.30, P = 0.27 for September) nor at interiors (df = 1,24: F = 0.93, P = 0.34 for July;
F = 1.60, P = 0.22 for August; F = 0, P = 0.95 for September). In contrast, in July 2002,
there were significantly greater infestation levels at vineyard borders when the adjacent
uncultivated habitat had greater WGI (F 1,23 = 7.93, P = 0.009) (Figure 3.3) but not for
the following months (df = 1,28: F = 2.29, P = 0.14 for August; F = 3.10, P = 0.09 for
September). That same year, no significant relationship was observed between WGI of
uncultivated habitat and vineyard interior levels of infestation (Figure 3.3) (df = 1,28: F =
1.38, P = 0.25 for July; F = 0.02, P = 0.89 for August; F = 0.38, P = 0.54, for
September). Regression analysis using RWGI did not show significant correlations in
2002. The only positive significant correlation was obtained in July 2001 between RWGI
and infestation levels at the border of vineyards (F = 4.86, df = 1,24, P = 0.04) (Figure

3.3).

50

a)
100.

 

 

0 border
tn- 0 interior
m‘ o
O
40 *
O
C
.9. 20 ‘ . ' o g
H
s l 2 l a
a 0 . I Y T
'5 absent present absent present
'5
t; b)
2 1oo~ *
0
.\° 8. - g
I
6% : o 0
. O
O
40 . 8 0
° 8
204 i o g o
o a o
0 3 + 8 .

 

 

absent present absent present

Wild grape clusters

Figure 3.2. Infestation levels by Endopiza viteana larvae in vineyard borders (dark) and
interiors (white) adjacent to habitats where wild grape clusters were present or not. a)
July 2001, b) July 2002. The asterisk indicates a significant difference between sites with
wild grape clusters absent or present.

51

Table 3.5. Correlation between vineyard percent cluster infestation by Endopiza viteana
larvae at vineyard borders and interiors, and the incidence of wild grapevines bearing
fi'uit clusters in the uncultivated habitat adjacent to the vineyard. Significant Kruskal-
Wallis-test CHI-square and P-values are in bold.

 

 

 

Vineyard border Vineyard interior

Year df f P j P
2001

July 1 3.02 0.08 4.03 0.04
August 1 3.32 0.07 0.30 0.58
September 1 1 .08 0.30 0.66 0.42
2002

July 1 4.18 0.04 0.47 0.49
August 1 2.49 0.11 0.15 0.70
September 1 0.70 0.40 0.05 0.83

 

52

2001

0 border
1m _ 0 interior 1m.
sol .b0 d? a).
........ Intenor
m a) .
o o .

 

 

 

 

 

% Cluster infestation

   

.. O
. .. 0.9. CO.
8. 9. 'DO . o b

0
0°80
5 10 15 Z) 5 I!)

WGI RWGI

 

0

v

 

figure 3.3. Relationship between percent cluster infestation in vineyards by Endopiza
viteana larvae in July 2001 (top) and July 2002 (bottom) and wild grape index (W GI) or
restricted wild grape index (RWGI) values of the adjacent habitats. Cluster infestation at
the borders is shown by dark circles and solid lines, and infestation at the interiors is
shown by open circles and dashed lines. The asterisk denotes significance.

53

DISCUSSION

This study showed that uncultivated habitats adjacent to vineyards in Michigan
influenced cluster infestation by grape berry moth larvae, and that the effect is most
evident at vineyard borders. Greater larval infestation by E. viteana at vineyard borders
than interiors has been reported previously in other Eastern US grape production regions
in studies of vineyards along wooded borders (Biever and Hostetter 1989, Hoffinan and
Dennehy 1989, Trimble et a1. 1991). The greater infestation at borders could be due to
improved overwintering survival of pupae at borders as opposed to interiors (Trimble et
al. 1991), immigration by gravid females from woods (T rimble et al. 1991, Trimble
1993), E. viteana possibly favoring woods in which it has evolved over cultivated
vineyards (N agarkatti et al. 2002a), or because borders provide a restricted area for
movement by females to lay eggs in comparison to vineyard interiors, making egg
deposition more likely at borders.

1 Comparisons among vineyards with different neighboring habitats revealed that
the greatest infestation was observed near deciduous woods, whereas the lowest
infestation was observed in vineyards near grasses (Table 3.1). The presence of nearby
woods was previously reported to pose a risk of grape berry moth larval infestation in
vineyards (Hoffman and Dennehy 1989). Potential explanations included the observation
that wild grapes were usually (though not always) present in wooded habitats, and that E.
viteana could be responding to structural characteristics of wooded edges (Hoffinan and
Dennehy 1989, Martinson et al. 1991). The study reported here separated woods into two
types according to their primary vegetation, and separated uncultivated habitats into three

different structure types; woods, a single row of trees, and grasses. Deciduous vegetation

54

in woods presents a greater risk than coniferous vegetation (Table 3.1), and that trees,
whether as part of a forest or in a single row, are sufficient to provide greater risk of
infestation than habitats without trees. Herbivores have been shown to respond in a
significant way to landscape complexity within a 1.5 km radius, particularly because of
mortality effects caused by parasitoids that are especially sensitive to this landscape
context (Thies et al. 2003).

Although vineyard infestation levels varied among vineyards with different adjacent
habitats, the abundance of moths in vineyards did not vary in a similar pattern, and no
positive relationship between the level of cluster infestation by larvae and moth
abundance was detected. Rather, the relationship was often negative (Table 3.3b)
indicating that moth captures should not be taken as a predictor of the risk of cluster
infestation in the vicinity. This corroborates studies conducted on the European grape
berry moth Lobesia botrana (Karg and Sauer 1995) and earlier, studies on E. viteana
(Dennehy et al. 1990b). Although abundance of male moths in vineyards did not vary
significantly in relation to the type of adjacent uncultivated habitat, there was significant
variation when abundance of these moths was sampled within these uncultivated habitats
(Table 3.2). Male moth abundance in uncultivated habitats was positively correlated with
vineyard infestation in August 2002 and September of both years (Table 3.3a). By
comparing data in Tables 3.1 and 3.2, it is clear that the greatest infestations were found
in vineyards near those deciduous woods that contained the greatest numbers of male
moths, and that lower levels of vineyard infestation occurred near grasses in which the
fewest moths were captured. Sciarretta et a1. (2001) similarly showed that in the plum

fruit moth, Cydia funebrana, catches were insignificant in landscapes other than orchards

55

where their food substrate was present, and in their studies of E. viteana Hoffman and
Dennehy (1989) demonstrated that this species was more abundant in woods and
vineyards containing grapes than in neighboring alfalfa fields. In a recent study (Chapter
2), male moth abundance in deciduous woods increased with height of trap, with more
than 76% of the moths being caught in the tree canopy at and above 9 m, further
emphasizing that the distribution of E. viteana is related to that of its host.

The variation in abundance of adult E. viteana among different habitats suggests
that the suitability of these habitats for this pest is not identical. This study focused on
variation in the host plant within these habitats as a potential explanation for the
difference in captures of E. viteana males. When wild grape vines were sampled, the
plants were equally likely to be present in each adjacent habitat (Table 3.4). Those wild
grapevines were identified to species following a taxonomic key by Voss (1985) and
corresponded to V. labmsca, V. riparia and V. aestivalis. This, is in agreement with
observations that appropriate habitats for wild grapevines, particularly V. riparia, are
relatively continuous across Eastern North America (Morano and Walker 1995, Downie
and Granett 2000). The habitats did, however, differ in the likelihood of the wild vines
having clusters. These were most commonly observed in the deciduous woods sites,
although never observed in grasses, which could be because wild grapes are poor
competitors of weeds and shrubs when they lack structural support, but also because
grapevines are likely to be mowed with the grass. In addition, Mullins et al. (1992) report
that horizontally-trained shoots of some varieties of cultivated grapes are less fi'uitful than

vertically-trained ones.

56

As measured by WGI values, deciduous woods had six times more wild
grapevines on average than tree rows, three times more than grass fields, and almost
twice as many as coniferous woods (Table 3.4). Although the lower WGI values of tree
rows were a direct result of the smaller area that a single row of trees occupies, they
reflect the total amount of wild grapevines in an uncultivated habitat. On the other hand,
the RWGI quantified wild grape presence within a standard area neighboring vineyards.
Both indices revealed significant differences in wild grapevine abundance among
habitats, with highest values of both indices for deciduous woods.

The presence of woods per se did not create equal risk of pest infestation, because
the two types of woods differed significantly in their impact on vineyard infestation, male
moth abundance, and likelihood of containing grape clusters. Indeed, wild grapevines
were in 75% of deciduous woods sites and only at 33% of coniferous woods sites studied.
The lower fruit production of grapevines in coniferous woods has not been documented
previously, but increased soil acidity (pH < 6.5) can sometimes limit commercial grape
production. Pine needles on the ground affect soil acidity, inhibit germination of new
grapevines, and decrease the number of flowers available for cross pollination which is
necessary for the primarily dioecious wild grapes (Mullins et al. 1992). The most
important factor affecting the suitability of this habitat for vines is that coniferous woods
have dense canopies throughout the year, reducing light penetration to the forest floor and
creating less favorable mesoclimatic conditions than those found in deciduous woods
(Mullins et al. 1992). I am not aware of studies that describe wild grape fruiting
distribution or abundance in any geographical range, but personal observations of wild

grapevines show that inside deciduous woods, the majority of fruiting typically occurs

57

higher than 12 m within the woods canopy, while at the edges of woods and along tree
rows, fruit clusters are found from low near the ground to high into the tree canopies (N.
Botero-Garcés, unpublished data).

Fruiting frequency and wild grape indices were both related to infestation by
grape berry moth in vineyards. Fruiting was significantly correlated with cluster
infestation in July 2001 (interiors) and July 2002 (borders) (Figure 3.2, Table 3.5),
probably because the presence of wild fi'uit clusters in the uncultivated habitat improved
the quality of the overall landscape for E. viteana. This result agrees with the “ideal free
distribution” prediction in which herbivores distribute themselves so that they utilize
resources optimally (Williams et a1. 2001). Indeed, Nagarkatti et al. (2002a) postulate that
females of E. viteana prefer wild grapes over cultivated ones and are better adapted to
densely wooded habitats with varied vegetation. Research in vineyards and neighboring
deciduous woods (Chapter 2) suggests that E. viteana distribution throughout this
agrolandscape is tightly correlated to the vertical and horizontal distribution of Vitis host
plants. It may be possible that moths are attracted to wild grape clusters in greater
numbers than can be supported, and in such cases females unable to lay eggs in wild
grapes may disperse to locate new hosts in the nearby vineyards.

In July 2002, unprecedented levels of infestation by grape berry moth were
observed, particularly at vineyard borders (Table 3.1, Figure 3.2b). This cluster
infestation tended to be greater when the uncultivated habitat contained fruit clusters or
wherever WGI values were highest, as suggested by the positive correlation between
vineyard infestation and WGI and border infestation and RWGI (Figure 3.3). This result

may have been because wild grape clusters were abundant in the uncultivated habitats,

58

though this can only be inferred from this data and should be tested in another study. The
low predictive power of these three wild grape parameters might be because E. viteana
do not depend on the presence or abundance of wild grapevines, but on the presence and
abundance of wild grape berries for development. It is possible that by factoring presence
of fruit clusters in adjacent sites into any wild grape index we may achieve a better
predictor of vineyard infestation.

This study highlights the importance of landscape management for manipulation
of crop pest populations. Typically, descriptions of the influence of adjacent habitats on
crops have addressed whether the effect is negative due to pest immigration, or positive
because of movement of natural enemies (Seaman et al. 1990, Dom et al. 1999) and the
availability of alternate insect hosts for pest parasitoids (Dennehy et al. 1990a). The grape
berry moth, a specialist pest species that dwells in native habitats next to cultivated
grapes, may be able to cross the area between habitats to colonize vineyards, much like
codling moth, C. pomonella (Dorn et al. 1999). Intercrop movement of insect pests is not
uncommon, as reviewed by Sciarretta et al. (2001) for both large and small spatial scales.
In some cases, the pest’s biology depends on inter-habitat movement, as with Ostrinia
nubilalis which need to fly to grassy surrounding areas in order to mate and rest (Derrick
et al. 1992). In other systems, the incidence of uncultivated habitats neighboring
cultivated land affects pest pressure due to immigration, as occurs with a complex of
thrips in British Columbia, Canada, moving in and out of nectarine orchards (Pearsall and
Myers 2001). Nagarkatti et al. (2002a) argue that inter-habitat movement by E. viteana

may occur only within the immediate vicinity, due to the insect’s lack of flight vigor.

59

Future studies should address potential dispersal capacity of this species and factors
affecting potential inter-habitat movement (Hughes and Dom 2002).

Integrated pest management (IPM) programs may be more effective against grape
berry moth by accounting for the wild grape abundance and fruiting in neighboring
habitats. Future trapping strategies, that may include female baits such as the one
developed for C. pomonella (Light et al. 2001), should consider sampling in uncultivated
habitats, since these appear to be influential to grape berry moth populations. The
removal of wild grapevines in habitats neighboring vineyards may reduce the impact of
this pest by decreasing the amount of larval food substrate available for developing
generations of E. viteana. However, as our analyses of the relationship between wild
grape indices and vineyard infestation by E. viteana show, the area of influence of wild
grapevines is not limited to a border area so vineyard managers may need to consider the
whole uncultivated habitat when considering cultural control practices. A study of the
landscape context over different (or larger than the one described here) spatial scales,
similar to the research by Thies et al. (2003) could help identify critical landscape factors
for grape berry moth infestation in vineyards.

The immediate effects of removing wild hosts could also include increased pest
pressure, since it is not unusual for a portion of a resident population to migrate in
response to environmental cues and selection pressure (Hughes and Dorn 2002). Further
studies should address the effect of wild host removal on associated populations of
parasitoids of E. viteana, as Dennehy et al. (1990a) have indicated that wild hosts are a
“source and refuge for natural enemy populations” in a study in which egg parasitoids

accounted for the greatest E. viteana mortality. Wild hosts in combination with the

60

diverse vegetation that characterizes natural or undisturbed land make up a habitat that is
favored by parasitoids, as discussed by Nagarkatti ct al. (2002a). Since Williams and
Martinson (2000) have shown that leafhopper parasitoids are. better able to successfully
colonize New York vineyards when alternate hosts are present within uncultivated
(wooded) habitats, parasitoids of E. viteana may also require these resources adjacent to
vineyards. The benefits of maintaining a complex landscape (preserving woods and
riparian vegetation) or simplifying it (removal of woods and wild grapevines) should be
further studied in this system before any management plans are implemented, since
differential impacts on parasitoid communities have been demonstrated (Mellaned et al.

1999, Thies and Tscamtke 1999).

61

CHAPTER 4:

MOVEMENT OF GRAPE BERRY MOTH IN VINEYARD AGROECOSYSTEMS

INTRODUCTION

Insect populations are capable of movement from one habitat to another both in
space and time (Wratten and Thomas 1990, Landis 1994, Drake and Gatehouse 1995).
These movements may be related to the need for food or oviposition substrates, mating
sites or refugia, all of which are usually distributed in a patchy pattern (Miller and
Strickler 1984, Demo and Roderick 1991). Herbivorous insects can potentially survive
with these patchy resources available to them in two basic ways: by accepting different
foods (polyphagy) or dispersing in the environment in search of their host.

Within a complex environment, mono- and oligophagous insects are better suited
for finding and accepting the right host than polyphagous ones, due to finely-tuned
mechanisms in their nervous system that govern host selection (Bernays 2001). Some
specialist Lepidoptera, in particular, have coevolved with plant taxa that produce
compounds useful to them in host identification or that convey protection to their larvae
(Dethier 1941, Ehrlich and Raven 1964, Rosenthal and J anzen 1979, Bemays 2001).

The grape berry moth is a monophagous herbivore whose hosts, Vitis spp. vines,
have no known chemical defense against infestation by this insect. Nevertheless, this
association of pest and host may have led to the evolution of fine-tuned host-finding
behaviors that have not yet been studied in detail. This gap in understanding includes
moth movement within and between habitats containing grapes. In Michigan, the grape

agroecosystem is comprised of relatively small (< 2 ha) vineyards interspersed with

62

remnants of woods, windbreaks (single rows of trees) and other crops, where wild
grapevines grow both with and without support (Chapter 3). There are three very
common species of wild vines in Michigan: V. riparia, V. labrusca, and V. aestivalis
(Chapters 2 and 3). Woods are one of the main uncultivated habitats present in Michigan
and across Eastern North America, and are a significant reservoir for grape berry moth
(Chapter 2).

One of the most important but often overlooked characteristics of those woods is
the space occupied by the canopy (branches, leaves) as opposed to the forest floor within
human reach. This three-dimensional structure both determines the distribution of
epiphytes (and vines) and their availability to herbivores (Richards 1983) and creates a
habitat in which some flying insects are more abundant in the tree canopy (Rees 1983,
Sutton 1983, Su and Woods 2001). In the case of E. viteana, which is most abundant
above 9 m in the tree canopy, the distribution is tied to its host’s vertical distribution,
probably in response to fruit cluster distribution (Chapter 2). This effect of food resource
distribution has also been observed in predators that at different life stages forage in
separate host strata for their particular prey (Cisneros and Rosenheim 1998) and also in
insect parasitoids whose abundance is tied to structurally complex habitats (Roland and
Taylor 1997, Thies and Tschamtke 1999). This complies with the ideal free distribution
theory, which predicts that herbivores will be distributed so as to optimally exploit
resources (Williams et a1. 2001).

According to the distribution of its host plant, the three life stages of E. viteana
are likely to follow different dispersion patterns in this agrolandscape (Schowalter 1996).

Adult male moth distribution has been assessed, as was mentioned above, but gaps in

63

knowledge exist regarding distribution of female moths. Larval distribution within
diverse plant communities such as deciduous woods is likely aggregated on grapevines,
whereas within homogenous habitats (such as vineyards) distribution might be
throughout the habitat. It is expected that pupae follow these distribution patterns, as
fourth instars seek grape leaves to cut crescent-shaped sections in order to spin a cocoon
(Chapter 1). However, in the overwintering generation, diapausing pupae are inside
cocoons spun on vine leaves and these leaves fall to the ground during leaf senescence
and remain nearby the grapevine during the winter. Another way that movement may
occur is when (or it) leaves are carried by the wind or rain, and therefore pupae are
transported within them. This type of passive dispersion has never been identified or
studied in the grape berry moth system. Elucidating whether this passive transport takes
place could perhaps shed light on certain peculiarities of patterns of distribution of adults
in the spring, particularly the greater abundance of moths in the woods in spring.

Another possibility is that there are two different populations of E. viteana, one in
uncultivated habitats, spatially separated from the other vineyard population. However,
trapping data (Chapter 2) seem to indicate that these populations mix, either actively by
interchange of adults, or passively by pupae on leaves being taken by winds to the edge
of woods. Indeed, some have proposed that E. viteana abundance is high at the woods
edge in the spring due to improved survival of pupae at vineyard borders (Martinson et al.
1990). Adult trapping studies by Trimble (1993) and Hoffman and Dennehy (1989)
suggest that the uncultivated and cultivated habitats are linked by dispersal of this pest
between these habitats. There is no direct supporting data for this dispersal between

habitats, however.

64

Uncultivated habitats within agricultural landscapes may have positive or
negative impacts on crop pests and beneficial insects (van Emden 1965, Solomon 1981,
Ekbom 2000) (Chapter 3). Others have addressed patterns of insect distribution caused by
winds interacting with these habitats, and have related it to pest impact on crops (Pasek
1988). However, direct insect movement has been more difficult to study, although both

laboratory and field methods are available.

Laboratory methods for measuring flight capacity

Flight of insects has been studied using flight mills, in which an arm of known
length rotates around an axis carrying an insect glued to it by the thorax. During flight,
the arm revolves around the center and by doing so, cyclically interrupts the light on a
sensor that allows automated recording of flight fi'equency and duration, allowing
calculation of flight velocity, duration of flight bouts, distance flown per bout, and
assessment of the tendency to undertake long (migratory) or short (appetitive) flights
(Beerwinkle et a1. 1995).

Flight mills have been used to show that low temperatures limit flight onset and
flight ability, while increasing temperatures in general cause increased locomotory
activity (Sanders et a1. 1978, Fasoranti et a1. 1982, Taylor and Shields 1990). Differences
between sexes in how temperature, relative humidity, and diet affect flight ability have
also been documented (Sharp et al. 1976, Taylor and Shields 1990, Sappington and

Showers 1993) using this method.

65

However, flight mills may present misleading information. Sharp et al. (1976)
stated that the soybean looper, a Noctuid, was a poor flyer in flight mills whilst they were
supposed to be strong fliers in the wild. Sappington and Showers (1992) working with
black cutworm, Agrotis ipsilon, also stated that data obtained fiom flight mills ought to
be interpreted with caution since the experimental conditions were “inherently intrusive”
to the moths and therefore very stringent criteria should be applied to experiments.
Cooter and Arrnes (1993) recommend that flight mill data be used only to compare
estimates of flight performance and not be extrapolated to the moth’s behavior in the
field. Because of the limitations of the flight mill approach, studies of insect movement
have also been canied out directly under field conditions. These methods have centered
on direct observation (night goggles, binoculars) and different variations of mark-release-

recapture methods.

Methods for marking insects

Mark-release-recapture studies use a mark of some kind that has been put on the
insect beforehand, which is then used to identify marked insects in subsequent samples of
the population. This mark can be a tag, a body mutilation of some sort, a paint mark, a
genetic marker, a radioactive-isotope mark, an element mark, or more recently, a protein
mark or some genetically engineered mark (Hagler and Jackson 2001). Recognition of
the mark can be achieved through several techniques such as harmonic radars,
radiotelemetry, radio-activity and metal detectors (Piper and Compton 2002). Of the
simpler available methods, incorporating dyes into meridic diets has been shown to work

well (Showers et al. 1989) but some mortality may occur and a colony must be at hand.

66

Marking with fluorescent pigment dusts depends on insect morphology and has been
shown to be successful for marking minute insects (Garcia-Salazar and Landis 1997,
Cronin et al. 2001) including moths, without affecting survival or ability to find
pheromone traps (Mo et al. 2003).

Mark-release-recapture methods have been used to show that the codling moth, C.
pomonella, has the capacity to engage in long-range flights up to 11 km, preferably in the
2"‘1 to 7th day of life (Schumacher et a1. 1997). Females first oviposited before engaging in
long (between-habitat) flights (Schumacher et al. 1997), in contradiction with the
oogenesis-flight syndrome (Johnson 1969). The oriental fruit moth, C. molesta, tended to
be sedentary and had lower flight capacity than C. pomonella (approximately 1 km),
although a small proportion of the population, most likely gravid females, ventured into

long flights (Hughes and Dom 2002).

Methods for insect recapture under field conditions

Insect recapture depends on the use of some sort of trap suited to the insect’s
environment and behavior (J uillet 1963, Muirhead-Thomson 1991). Food baits and
pheromone traps are usually employed because of their efficacy (higher recapture rates)
but they bias insect movement (Weissling and Knight 1994). Traps without attractants
can give an indication of natural dispersal and flight patterns of foraging insects and they
are generally called passive or interception traps or both (Muirhead-Thomson 1991).
They include the Malaise trap, which can be in many forms and modifications, but
generally consists of a flame on which a vertical fabric has been stretched so as to

interfere with flying insects. When the insect makes contact with the trap it climbs up

67

towards a container that captures it. A killing agent inside the container will ensure that
specimens will not be destroyed by other trapped more robust species. An important
quality of this trap is that direction of flight can be inferred since the trap is mechanically
intercepting insects flying in a given direction. By modifying the trap (a screen in the
middle) or joining two of them, with two different containers at the top, insects flying in
opposite directions can be sampled. In a comparative study, Juillet (1963) found that the
Malaise trap was the second best device for capturing flying insects, including
Lepidoptera species. Passive interception traps can be constructed of transparent plastic
panes coated with sticky material. These are usually hung on tree canopies, and have
been successful in trapping both sexes of codling moth, C. pomonella (Weissling and
Knight 1994, Knight 2000).

Muirhead-Thomson (1991) recommends that at least two types of traps with
different principles of capture and attraction be used when studying insect flight, since
interpretation of only one capturing technique can be “difficult or speculative”. The
limitations of mark-release-recapture methods lie in the possibility of using inadequate
marks (not in accordance with the life stage or morphology of the insect), losing the
marked population (no recapture of any individual), or not recognizing it as marked (tag
lost, color faded). This is why the species considered for these studies needs to be studied
beforehand, so that the marking method is appropriate for the behavior and morphology

of the organism (Hagler and Jackson 2001).

68

Movement by E. viteana

Sampling for adult E. viteana typically relies on the use of pheromone traps which
attract only males. These are usually placed at 1.5 m above the ground at vineyard
borders. Male abundance within vineyards has been studied in order to identify patterns
of spatial distribution that could help predict vineyard infestation, but without much
success (Trimble 1993). Several authors have noted that greater grape cluster infestation
at vineyard borders usually corresponds to lower moth captures and have explained this
by suggesting that wild females may be moving fi'om woods to lay their eggs in vineyard
grapes during the growing season (Taschenberg et al. 1974, Hoffman and Dennehy 1989,
Trimble 1991, 1993). Data presented in Chapter 2 show that moth captures continue
throughout the season when the woods canopy is sampled, indicating that the woods
population may be expanding and subsequently colonizing vineyards.

Elucidating E. viteana movement behavior will potentially improve our
knowledge of the insect’s ability to survive in deteriorating environments or it's ability to
colonize new ones. It will also help in improving pest management strategies, since
mating disruption, for example, is less effective if gravid females immigrate from wild
areas into vineyards (Trimble 1993). The goal of this study was to determine the capacity
of E. viteana adult moths for movement within vineyards, and to determine whether this
species can move from woods to adjacent vineyards. Three different methods for
studying movement of this species were used: mark-release-recapture of fluorescent dust-
treated moths, bi-directional Malaise traps for monitoring movement of adult moths, and

tracking winter movement of pupae using recapture of painted leaves.

69

METHODS
Mark-Release-Recapture study

This study was conducted in a four-year old experimental vineyard ( Vitis
labnrsca, var. Niagara), at the Trevor Nichols Research Complex, Allegan Co.,
Michigan, during 2001 and 2002. The vineyard consisted of three parallel blocks, each of
32 rows and seven vines, with a total of 672 vines. This vineyard was bordered by
another vineyard on its eastern side, by woods to the west, an apple orchard to the south,
and a grassy field to the north. Vineyard management was conventional, following

recommendations for treating insect pests (Table 1.1).

Traps used for recapture

Two different kinds of traps were used to capture grape berry moths: pheromone
traps for males and passive-interception pane traps for both sexes. The first were large
plastic delta traps (Suterra LLC, Bend, Oregon), baited with lures containing 0.1 ug of
synthetic sex pheromone of E. viteana (90:10 ratio of (Z)-9-12Ac and (Z)-11-14Ac), and
lined with sticky inserts. New inserts were used every time moths were found in traps
upon being checked, and lures inside pheromone traps were replaced every month from
the same batch of lures. The passive interception traps were made of 380 x 280 x 3 mm
Plexiglas panes, coated on both sides with tangle trap paste (The Tanglefoot Company,
Grand Rapids, MI). All traps were hung from the trellis at approximately 1.5 m high, and
the plastic panes were secured to the ground by strings to maintain their vertical position.
A total of 33 panes and 56 pheromone traps was deployed during 2001, and 31 panes and

56 pheromone traps during 2002, in a pattern of concentric circles, radiating fi'om the

70

middle to the periphery of the vineyard (Figure 4.1). The first circle of pane traps was 2-
3.30 m away from the central point of release, and consisted of four panes, while the
peripheral ring of traps consisted of pheromone traps positioned around the vineyard and
entering the four adjacent habitats. Traps were placed at the edge of each of these four
habitats separated 20 m from each other: two were within the edge of the grassy field to
the north, two on the first row of trees in an orchard to the south, two on the first row of
vines of the adjacent vineyard to the east, and two at the edge of the woods to the west.

Eight pheromone traps and four pane traps were placed inside the woods (Figure 4.1).

Marking moths

Newly emerged adult grape berry moths were taken from the E. viteana colony at
the Small Fruit Entomology Laboratory, Michigan State University, established in 2000
using larvae collected in infested grapes from a commercial vineyard in Van Buren Co.,
Michigan. The colony was kept in two sets of conditions, according to phenological state.
Adult moths were maintained at 26°C and 70-80% RH, and a photoperiod 16:8 (L:D) h.
Some larvae were reared on commercial table grapes in the laboratory under 22-25°C and
16:8 (L:D) h, 30% RH, or in a meridic diet (Nagarkatti et al. 2000) inside an
environmental chamber at the same conditions except that temperature was 25" C. To
help preserve wild traits, moths reared fi‘om grapes collected in the same vineyard were

added to the colony at the end of the first year.

71

 

 

 

 

grasses

 

 

vineyard

 

+ Pheromone traps

 

I] Panes
0 Release point

 

 

 

orchard

72

 

vineyard

51.1 m

Figure 4.1. Schematic representation (not to scale) of the experimental vineyard-woods
set up where marked moths releases took place. Pheromone and pane traps were arranged
around a central vineyard release point in concentric circles; a release point inside the
woods was also used to determine extent of movement between woods and vineyards.

The moths were held in groups of ~400 in 3.8 L white utility pails (Holiday
Housewares Inc., Leominster, MA), covered on the top with white tulle veil held in place
by a rubber band. A 12 cm-diarneter opening was drilled on the side; a white cotton
sleeve was attached to allow the operator’s arm to be introduced without the insects
flying out. Water was provided by placing damp dental cotton wicks either fixed to the
bottom of the cages or on top of the tulle covering it.

To mark the moths, 0.5 g of dry fluorescent dust dye (Dayglo Color Division,
Switzer, Cleveland, OH) was dissolved in 75 ml of acetone (99.9%) in a small cosmetic
spray bottle. Adult moths were sprayed with the solution through the veil. However, it
was noted that some excess drops of solution tended to form at the bottom of the cage
and catch the wings of moths and pin them to the cage surface. To avoid this, after
removing the cotton wick, the cage was inverted, tapped once so that moths landed on the
veil and then the solution was sprayed through it. Moths on the tulle were marked and
dried immediately. After marking, moths were taken to the release point and placed in the

shade until release.

Moth releases

Releases in the vineyard were made at its central point, at row 16, at the base of
vine 11. Releases took place between 1700-1900 h, when winds were low, it was not
raining, and air temperature was approximately 25 ° C. The weather report was checked
prior to release so that no rain or storms were expected for the following day(s). To
release the moths, the bucket was placed on its side and the rubber band holding the veil

was carefully cut off so that moths could fly out. All live moths had escaped the cage by

73

the next day, when the number of moths released was determined by subtracting the
number of moths found dead inside the bucket from the original number in the bucket.
Moths were released four times in each of 2001 and 2002 during July, August and
September, using moths at a 1:1 sex ratio, as maintained in the colony. During 2002, a
total of 850 moths was released in the adjacent woods during the last three vineyard

releases.

Moth recapture

All traps were checked at regular intervals after release fiom 12 h to 300 h after
release. Pheromone trap inserts with moths were taken to the laboratory and examined by
illuminating the moths with a UV light under a microscope for presence of fluorescent
dust. Moths captured on sticky panes were removed fi'om the panes with a spatula and

taken individually to the laboratory where they were observed in the same way.

Weather data

Weather data were gathered from the Michigan Automated Weather Network
(MAWN) (http://www.agvefiather.geo.msu.edu/rnagv_n_) station at the Trevor Nichols
Research Complex (42.59°, -86.16°) for hourly averages of wind speed, wind direction,
relative humidity, precipitation and air temperature. Hourly averages of these weather
factors were selected for every day between 1 July and 21 September for each year.
Average values of weather factors from 1700 and 2200 h were calculated, since this is the

period during which the gape berry moth flies (G. English-Loeb, pers. comm.) Data

74

from this activity period for each day was again averaged over the period of days from

release to recapture (the day before traps were checked) for each recaptured moth.

Data analysis

In order to evaluate the direction of moth flight in relation to wind direction, and
to compare differences in flight directions between sexes and trap types, the data on
individual moths were analyzed using Oriana software (Version 1.06, 1994). Calculations
of Watson’s F -test were conducted to compare pairs of circular means because this test is
particularly powerful for samples of small size (Batschelet 1981). Circular histogarns, in
which 0" corresponds to the actual North of the spatial location, were produced to show
the mean angle direction (MAD) comprising the mean angle (vector) and 95% confidence
intervals for each of the wind and moth direction samples. The data on individual moths
recaptured were analyzed using the REG procedure (Model 1) (SAS, Version 8.0, SAS
Institute 1996) to determine the relationship between weather factors and the distance
flown by moths. To determine differences in flight distance between sexes or between
moths released in the vineyard and moths released in the woods, the NPARlWAY

procedure (SAS, Version 8.0) was used with a Kruskal-Wallis test (SAS Institute 1996).

75

Malaise traps to measure movement in and out of vineyards

This study was conducted at three juice gape (Vitis labrusca, var. Concord and
Niagara) farms in Van Buren Co, Michigan, during 2000. At each farm, two vineyards
bordered by deciduous woods on at least one side were selected for deployment of
Malaise traps.

Each bi-directional Malaise trap consisted of two traps next to each other facing
opposite directions. Each was made of V2 inch and 3/4 inch PVC, held together by bolts
and covered by white tulle, following the design of Isard et al. (2000) for studies of
western corn rootworrn. A container made of two similar clear plastic 2 liter soda bottles
was placed at the top of each trap to collect flying insects (Figure 4.2a). One of the
bottles was cut in two and its upper half was used as a funnel lodged into the upper half
of the other out bottle. The lid of this latter bottle was filled with paradichlorobenzene
(PDB) as killing agent, covered with tulle, and screwed back.

Traps were deployed at a wooded edge of each vineyard, within the end row of
cultivated gapes (Figure 4.2b), with the vertical poles placed 30 cm into the gound for
support, and standing 2.15 m tall. One side faced the inside of the vineyard and the other
faced the woods. A second bi-directional Malaise trap was placed across from the first at
the edge of the woods with one side open to the woods and the other to the vineyard.
Finally, a third bi-directional Malaise trap was placed at a height of 9.0 m directly above
the second, so that one side faced the woods canopy and the other faced out from the
woods (Figure 4.2b). A PVC pole as described in Chapter 1 was used to place a Malaise
trap 9.0 m above the gound. Traps were lifted to the top of the pole“ with the help of

pulleys hung from loop bolts and 30 m long nylon ropes, which were secured to the side

76

of the pole. Each week, when collecting insect samples, traps were pulled down and the
contents of the containers were examined. The dead insects trapped in the upper section
of the trap were poured into one or several 5 oz. plastic cups labeled with trap number,
location, and direction (of insect flight). The cups were then taken to the laboratory in a
cooler. Specimens were frozen, to be sorted and counted during the winter. The number
of gape berry moths captured in traps, their sex, and the direction of their flight was

recorded.

 

a)

Cap filled with PDB

;\

flwm

t

 

 

 

 

4— Bi—directional Malaise
Traps

  

 

 

 

 

t , ,
’tll r ”H
. W... . ‘ 1
t , t .
l u t ,
’ r ‘ .
. . .
‘l‘ " I, “11,, "t. "It... 1;; "1). WI, "u w. "'t ‘
, 1 t "'1” It... “11-", 0,, "I” "It; ‘14.. "r“. ‘lt. Aim
1 . 1!. ill "I: lit a, t m ‘
‘V '1" it." l'u| .ll‘I "m :l‘.‘ «thingy?» ll. -

Figure 4.2. a) Detail of the Malaise trap insect collecting containers, made of two tops of
2 liter soda bottles encased in one another, with a lid full of a killing agent. b) Schematic
representation of the vineyard set up with two bi—directional Malaise traps placed at the
edge of woods and vineyard (gay boxes) at 1.5 m high, and one bi-directional Malaise
trap placed 9.0 m high near the woods canopy.

77

Painted leaves study

This study was conducted after harvest in the fall of 2002 in four juice gape (V.
labrusca, var. Niagara and Concord) vineyards in Van Buren Co., Michigan. All of the
vineyards had deciduous woods to the north and east of the vineyard tested. At each
vineyard, two sets of vines were painted; one set on the north of the vineyard (N-S), and
the second set on the east (E-W) (Figure 4.3, lower). Within each set, I painted the leaves
on three adjacent border vines located either at the end of three rows or in the middle of
one, and three vines in the same position but located 30 m inside the vineyard. Leaves
were painted with one of four bright colors (blue, red, orange and neon yellow) using
Specialty Lacquer spray (Rust-Oleum Corporation, Vernon Hills, IL) before leaf
senescence in October. Paint was applied to the majority of leaves of each vine by spray-
painting the top and underside of the vines. Vines in each of the four positions were

painted with different colors, to differentiate among leaves from each painted area.

Sampling

Sampling was carried out in the spring afier snow melt (April and May) at the
time of gape berry moth emergence from diapause. For each set (N-S and E-W) of
colored leaves, seven transects were delineated using a measuring tape and colored flags
(Figure 4.3, upper). At each transect, leaves were sampled from ten contiguous
rectangular sampling areas 3.0 x 1.5 m (inside area 4.5 m2) that ran parallel to the woods
and vineyard edges, with the mid-point of the sampling in line with the middle painted
vine. Transect 1 was inside the woods, Transect 2 was at the edge of the woods, Transect

3 was in the middle of the interface between woods and vineyard (6.1-14.5 m wide),

78

Transect 4 was at the edge of the vineyard near to the border painted vines, Transects 5, 6
and 7 were 15, 30 and 45 m inside the vineyard, with Transect 6 running over the interior
painted vines (Figure 4.3). The number of colored leaves inside each rectangle was
recorded for each sample, with a separate record made for the different colors applied to

vines in each of the N—S and E-Wsets.

Data analysis

The NPARlWAY procedure (SAS, Version 8.0) was used in a Kruskal-Wallis
test to establish differences between sets, between positions and among transects for each
set, and between the number of leaves of each color found per transect, including pair-

wise comparisons (SAS Institute, 1996).

79

 

 

 

 

 

 

 

 

 

 

vineyard

 

Figure 4.3. Schematic representation (not to scale) of the arrangement of painted vines in
the leaf dispersal experiment. Grapevines (circles) at the border and interior of the
vineyard in two positions per set (N-S and E-W) were spray-painted with four different
colors (lower). The sampling transects are shown by seven parallel transects from the
inside of the woods to 45 m inside the vineyard (upper).

80

RESULTS
Mark-release-recapture study

A total of 3,505 moths was released in the course of 11 releases during the two
years of the study (Table 4.1), and 246 were recaptured overall (6.9%). A comparison of
the weather conditions during the period that releases took place, between 1 July and 21
September of both years, yielded similar results for all factors (Figures 4.4 and 4.5),
indicating that weather conditions did not vary much between years. The average
temperature during 1700-2200 h from 1 July to 21 September 2001 was 223°C and
241°C in 2002; mean relative humidity was 68.2% the first year and 63.8% on the
second (Figure 4.4). Precipitation was low overall (few rainy evenings, 10 in 2001 and 12
in 2002 over 83 evenings) with a mean 0.09 mm per day in 2001, and 0.04 mm in 2002
(Figure 4.5). Mean wind speed was 1.2 m/s in 2001 and 1.1 m/s in 2002. Average wind
direction (1700-2200 h) was 252.88° (3: 799°) for 2001 and 249.62° (:t 7.19°) for 2002
during the period when releases were made (83 (1) (Figure 4.5); there was no sigrificant
difference between years as shown by Watson’s F-test (F = 0.10, df = 164, P = 0.75).
The eight moth releases in the vineyards across both years were therefore treated as
separate replicates in subsequent analyses.

Vineyard releases. Within the eight vineyard releases, about nine times more
male moths were recaptured than females in 2001, and eight times more in 2002 (Table
4.1), a clear effect of pheromone trap efficacy compared to passive interception traps. In
pheromone traps, 173 males were recaptured over the eight releases, while in pane traps

the total number of moths recaptured was 29. Of these, 69.1% were females (20).

81

Average wind speed (mls)

Average air temperature (°C)

—I—ave atmp —I—ave atmp

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2001
. ave relh .. 100 A 40 - 2002 --o--ave relh 100
, .3 e
. , 2
. ‘ If 7! ”60 g s 20 a:
e ‘ 4O .5 i #40
1 10*"
at i ....
< <
0 fia'tttaettlttjo o tt+e+ +t “t#%0
«9 «94” (9413‘. (52 3969696949 a)” ¢°¢NH a 3‘“ “94%“ b«<‘5\<‘9«°\!°°e"° 6 ’69:? 6T 9.. dqd‘hd's
—I—avewspd —I—avewspd
10*— Wave Wdil'"r 360 10*" anon-wave” .. 350
l t v 315 ‘ ‘ 4- 315
8' ‘ 270gD '2’ 3" “ ‘ ..m
at v 425% g 5th i .225
. «.180_ 2 .1”
. 'O
4': L 0135:g E 4i .135
‘ , a. I l 0
2‘ ‘ m 3 2-0- l Wm
0 0 < affix ............... 0
“N «94° \"«\"- E194” “#0694969 9’ 9‘30 ”he '9 3‘ “9'0" \"«\" «9496’ 960666943169 9“ 99;" “6’9

Figure 4.4. Weather conditions during the period between 1 July and 21 September of
2001 and 2002; wind, temperature, and relative humidity. Data were obtained by
averaging values between 1700 and 2200 h (moth active time) for each day. Vertical
arrows represent releases for each year.

The highest recapture rate was achieved in the release of 31 July 2001, when
almost 32% of the estimated males released (210) were recaptured, for a total of 17.1%

adult moths recaptured overall (including females) (Table 4.1).

82

Average relative humidity (96)

Average wind direction (degrees)

Wind 2001 Wind 2002

 

 

 

 

 

 

 

 

 

 

 

 

E 3

E D2001

: 2.5 ~ .2002 .

i3 2 « r

g .

E’- 1.5 -l

U

2 1 4

‘L 0.5 J l
0 =1 . - fl—l . . # L .D .59.—
x «a tr e e a. o o q o
‘0 «9 .\\" «0 «0' e‘ 9" 9" ‘0' '9'" 9‘ 9\

Figure 4.5. Weather conditions during the period between 1 July and 21 September of
2001 and 2002. Data were obtained by averaging values between 1700 and 2200 h (moth
active time) for each day. a) Circular histogams with mean angular vector and 95%
confidence limits for wind direction for each year. The arrows represent the average wind
direction. b) Daily precipitation during moth active time during the period of releases.

83

The maximum displacement recorded from a vineyard release was that of a male

moth captured in a pheromone trap 58.2 m from the central release point (Table 4.2). The

maximum distance a female moth moved after being released in the middle of the

vineyard was 41.2 m (2001) but the mode for the eight releases was 3.2 m from the

central release point. The average displacement of male moths was more than twice that

of females, although one female (from a woods release) flew almost twice as far as the

average maximum distance male moths flew when released in the vineyard (Table 4.2).

Table 4.1. Total number of marked and recaptured Endopiza viteana moths released and
recaptured in two habitats. Percentages were calculated for each sex based on a 1:1 sex

ratio of released moths.

 

Marked E. viteana moths

 

 

# released # recaptured % recaptured

males females males females

Releases in the vineyard
12 July 2001 350 11 3 6.3 1.7
24 July 2001 500 40 4 16.0 1.6
31 July 2001 420 67 5 31.9 2.4
31 August 2001 365 26 1 14.3 0.6
17 July 2002 300 24 3 16.0 2.0
1 August 2002 280 2 0 1.4 0.0
15 August 2002 200 15 3 15.0 3.0
28 August 2002 240 17 1 14.2 0.8
Total 2655 202 20 15.2 1.5

Releases in the woods

1 August 2002 240 4 1 3.3 0.8
15 August 2002 300 19 0 12.7 0.0
28 August 2002 310 0 0 0.0 0.0
Total 850 23 1 5.4 0.2

 

84

 

m6 H ”An

—.e H five

:5 H v.3.

QM H 1mg

 

 

fieEEMEuaV
- - - - - - No8 semi mm
- ms H 0.8 .. m6 H Own - 9:: Nos «mews/w 3
- a6 H man - Wm H ode v.3 mdofi Noam “mewsaes ~
$893 5 8.483%
fiwn H afifi «an H fine— 5. H v.2 a... H «.m— QM Had" he Hbdm «nefikueseaV
od H ode w.m H mi 06 H Nm w.m H m? Nd m.mm Noam «mamas. mm
05 H camfi fiwm H 0N2 Nd H m.m Em H 5.3 Cd van Noon «ma—wee. 2
- od H odmm - od H 2 - 3 «com seems... a
m.wm H m.mm NON H #62 _.o H m.m 5o H 1w m.m WE Noon b3. 3
od H cda NE H Em: od H Nd v.“ H «.2 Wm 9mm SON sewn—a. E
o._m H odv— Nw H méfi ed H fimm m; H 3: «.3 Nwm SON be. "m
wag H odm— Yfl H was? 9m H 02 fin H v.3 mgv Nam SON 33 em
ON H WEN m.wm H 2.3 On H 08 o.m H v.3 «.3 5mm 58 33. Q
$992.5 S 8333‘
838£ $38 8380.“ SEE mus—«Sow 8.2:
@0233 c5 Beacon—mama AEV
gouache mm H 502 mm H :82 Beacon—3% ea:

 

.Eso: gauche mo 2mg

:38 93 £853 95 E venue—oh £88 38“.: euwmonem 33582 can v8.88 an :30: mooefimme 508 can 825x34 .N... 033—.

85

 

 

 

nod vod mud n g .o mood w—J \
mmd ovd vod ood who w: m
ow; Ed nod «Nam Ed 3; k
meadow
wood wood mod 2d vod n; ..L
5d mwd cod bmd good v Nu— &
mod nod Hmd mod odoo n; k
832
vod mod n _ .o m fl .o ood R; «a
had 3d vod vod _ _.o mm; m
o~.— ood mod voé and R J k
womb 28a 5
Sood oood Sood mood vod v3; \
odd Rd wwd de mood «a: m
Nod 34 mod mod mod 3: k
mag 808823 5
:85 :2 so 6 .o $5 E as ”Basques use:
:euflamooi basins: god 052095: 3‘ 30% 9:? 08F
“Sumo 8583

 

.209
E oh @332on Emu—Mai .508 :80 we choose“: one 0828 5253 bréou Ewe do wooing 3o £83980 .8583 owe—PE
38 33302 2 08¢ fifiame mfioE goon; sacrum woo—.38 oowmfloroumzofiur .3 Esau 358% mo Embeds commmoemom 6.9 03.;

86

During the first four releases in the vineyard, variability in mean direction in
which moths flew were small, ranging fi'om 134-185 degrees (Table 4.2). Male moths
flew on average at a direction of 163.6 degrees (SSE) and females 127.8 degrees (ESE).

Data were split according to the type of trap used. Regression analysis indicated
that for moths caught in pheromone traps, the distance flown was positively correlated
with time after release (Table 4.3). Since this relationship was not significant for pane
trap moths, these were separated by sex, and regression was performed on each sex
separately. No significance was found between distance flown and time elapsed fiom
release for females. However, the distances flown by the nine male moths captured in
pane traps were positively related to the time after release, with a highly significant
coefficient of determination of 0.94. A Kruskal-Wallis test was used to determine
differences between sexes for mean distance flown before landing on panes, but no
significant difference was found (X2 = 0.33, df = 1, P = 0.57).

Weather factors affected the distance flown by moths, in particular the wind speed
and air temperature (Table 4.3). For moths caught in panes, distance flown was
significantly and negatively correlated with wind speed (Distance flown = — 15.50
Windspeed + 28.99, Table 4.3). The same was true for moths caught in pheromone traps,
though not significantly (Distance flown = - 2.61 Windspeed + 16.51, Table 4.3). Air
temperature apparently affected female movement more than male movement, which was

also reflected in pane traps (Distance flown = 1.55 Temperature — 27.59, Table 4.3).

87

 

 

 

 

b)

 

 

 

 

 

 

 

 

 

 

Figure 4.6. Circular histograms of the frequency and distribution of vectors of moths and
wind direction for the period that moths were flying between release and recapture in a)
pheromone traps and b) pane traps. The mean angular vector with 95% confidence limits
runs beyond the outermost circle; the 0° corresponds to the vineyard North and the center
of the histogram the vineyard release point, while the arrows represent the average
directions taken by the moths and the wind.

88

Direction of movement by moths captured in pheromone traps was also analyzed
separately from moths caught in pane traps. For each case, the direction flown by the
moth was compared with the average wind direction during the period the moth flew, to
determine whether moth flight was directly influenced by wind direction. For male moths
captured in pheromone traps, there was a significant difference between their mean
direction and wind direction (F = 47.01, df = 390, P < 0.01). The mean direction vector
for male moths in pheromone traps was 170.34° (i 1358") and for wind 234.63° (i
300°), indicating that moths moved generally at a tangent to the wind (Figure 4.6a). For
the 29 moths captured in pane traps, the difference between average wind direction and
moth direction increased and was significant (F = 16.96, df = 56, P < 0.01), with a mean
direction vector for moths of 105.87° (i 50.92°) and for wind of 250.020 (:1: 789°) (Figure
4.5b). When females were analyzed alone against the average wind direction, the
difference was again significant (F = 5.43, df = 38, P < 0.03). Females moved in more
different directions overall, with a greater variability between individuals and a mean
vector of 88.84° (:1: 13152"). Wind direction was much less variable at 243.86° (:1: 10.92°)
(Figure 4.7a). Males caught in panes also had a significantly different flight direction
from the mean wind direction (F = 12.30, df = 16, P < 0.01) with a mean vector of
113.65° (:1: 53.73°) for males and 260.50“ (i 978°) for the wind (Figure 4.7b). There was
no difference in flight direction between females and males caught in panes (F = 0.10, df
= 27, P = 0.76). No moths released in the vineyard were ever recaptured in the woods or

outside the vineyard.

89

 

 

 

 

 

 

 

 

 

 

 

Figure 4.7. Circular histograms of the frequency and distribution of vectors of a) female
moths and wind direction and b) male moths and wind direction for the period that moths
were flying between release and recapture in pane traps. The mean angular vector with
95% confidence limits runs beyond the outermost circle with 0° corresponding to the
vineyard North. The center of the histogram is the vineyard release point, and the arrows
represent the directions taken by the moths and the wind. '

90

Woods releases. Of the 850 moths released in the woods, 24 were recaptured, 18
of them in the vineyard and the other six in one pheromone trap at the edge of the woods,
facing the vineyard. Male moths released in the woods flew flirther overall: 64.1 m on
average, compared to 13.8 m for males released in the vineyard (Table 4.2). This
difference was significant by Kruskal-Wallis (X2 = 51.24, df = 1, P < 0.0001). Too few
females were recaptured fi'om woods releases to analyze their flight parameters.

A regression analysis of the distance flown by moths released in the woods and
the time it took to fly that distance (Table 4.3) yielded a significant relationship (F132 =
4.46, P = 0.046) with a coefficient of determination of 0.17 %. Of the four weather
factors compared, mean wind speed (F1,22 = 4.51, P = 0.05, r2 = 0.17) and mean
precipitation (F 1,22 = 4.48, P = 0.05, r2 = 0.17) were not significantly associated with
distance flown by moths released in the woods, though these regressions were close to the
critical P-value (= 0.0459). Neither relative humidity (F132 = 0.09, P = 0.77, r2 = 0.04)
nor air temperature (F132 = 1.62, P = 0.22, r2 = 0.07) were significantly correlated with
distance flown by moths released in the woods. There was a significant difference
between the mean angle of moth direction and the mean wind direction (F = 348.53, df =
46, P < 0.01) with a mean vector for moths of 81.68° (d: 620°) and 225.34° (d: 3.06) for
the wind (Figure 4.8). This shows that woods-released moths moved toward the vineyard

in preference to any other direction.

91

 

 

 

 

 

 

Figure 4.8. Circular histograms of the frequency and distribution of vectors of moths
released from the woods and wind direction for the period that moths were flying
between release and recapture in vineyard traps. The mean angular vector with 95%
confidence limits runs beyond the outermost circle. The 0° corresponds to the vineyard
North and the center of the histogram is the vineyard release point. The arrows represent
the average direction taken by the moths and the wind.

Malaise trap study
Only five specimens of E. viteana were captured with the bi-directional Malaise
traps. All five moths were found to be leaving the vineyard at 1.5 m high. This

experiment was not repeated after 2000.

92

Painted leaves study

After seven months in the field, and after the winter snow cover melted, a total of
2,377 painted leaves was found on the ground of the vineyards. Painted leaves were
observed as far as 60 m from the point of release (painted vines) in the case of a north
painted border leaf that was found on the north-east woods edge of the vineyard. A
border leaf painted on the east of the vineyard was observed 29 m away to the northeast,
also at the woods edge. A very similar number of leaves was recovered from the N-S set:
1,249 (53%) and the E-W set 1,128 (47%), but more of them came from border vines
than from interior ones; the total for the border colors was 1,548 (65.5%) leaves and 829
(34.5%) leaves for the interior colors. A Kruskal-Wallis test determined both sets (N-S
and E-W) were not significantly different in the number of leaves recovered, whether
these had come from border vines (X2 = 0.33, df = 1, P = 0.56) or interior vines (X2 =
0.71, df = 1, P = 0.39). An individual analysis of each set suggested border leaves were
not distributed evenly across transects (X2 = 19.06, df = 6, P = 0.004 for N-S, and X2 =
17.13, df = 6, P = 0.009 for E-W) and neither were interior leaves (X2 = 12.64, df = 6, P =
0.049 for N-S, X2 = 18.10, df = 6, P = 0.006 for E-W). To test whether transect
orientation had any effect on leaf distribution, another Kruskal-Wallis tested differences
between sets (N-S and E-W) by comparing transect per transect, the number of leaves
recovered. No significant difference between sets (P > 0.08 for border leaves and P >

0.16 for interiors) was found, so they were analyzed together in subsequent analyses.

93

180 - a
160 -
140 ~
120 «

804

Mean :tSE leaves

 

 

 

b)

120 . l

100 ~
80 ~
604

 

Mean tSE leaves

40— . b

4
20 be be be be c [1;]
sin ,

 

 

 

 

Figure 4.9. Mean :1: SE number of leaves counted per transect (45 m2) from inside the
woods habitat to 45 m inside the vineyard, for leaves that were painted on a) border vines
and b) interior vines. The arrows indicate the locations of painted vines.

94

The number of border and interior painted leaves varied significantly among
transects (Kruskal-Wallis X2 = 35.24, df = 6, P < 0.0001 for border leaves, and X2 =
29.83, df = 6, P < 0.0001 for interior leaves). Therefore, pair-wise comparisons between
transects were made (Figure 4.9a). Significant differences among transects indicated that
the trend of leaf movement was skewed towards the exterior of the vineyard. This was

more noticeable for border leaves but also for interior ones (Figure 4%).

DISCUSSION
Adult E. viteana movement

The different sections of this study show the capacity for different methods of
dispersal by grape berry moth. The mark-release-recapture method used was successful
and suitable for studying E. viteana flight behavior in the field, which had not previously
been done. I assume that the flight capacity of the moths used, which had been reared in
the laboratory, was similar and not different from that of wild moths. This is because the
colony was two years old (about 21 generations) but new genes from the original
population had been added twice, at approximately the 9th and 21St generation to help
conserve feral traits. Given the weather factors of southwest Michigan during the summer
months and the proximity of woodlots (24 m in this case) to vineyards, grape berry moth
adults are able to immigrate into cultivated grapes up to 109 m away from their point of
departure. Although only one female released in the woods was recaptured in the
vineyard, it is nonetheless proof that they are able to move beyond a single habitat in their
search for their host. It is possible to extrapolate this field study to other field conditions

and state that moths within uncultivated habitats such as woods are able to fly fiom these

95

wild habitats to cultivated grapes. Gravid or virgin female moths can immigrate to
vineyards since there was no significant difference between sexes in the distance flown.
This coincides with the general knowledge on other species such as the codling moth, C.
pomonella, which are able to move between distant orchards (Dorn et al. 1999, Keil et al.
2001)

Environmental conditions often determine the extent and occurrence of flight. The
average weather conditions between 1700 and 2200 h during the months this study took
place were amenable to moth flight (Figure 4.4). Wind speed was the most important
abiotic factor affecting moth dispersal (Table 4.3) but the average speed was 1.2 m/s in
2001 and 1.1 m/s in 2002, well into the range of 0.5-2 m/s cited as the range for take off
for many insects (Pasek 1988, Colvin 1995). Researchers in New York (G. English-Loeb,
pers. comm.) have studied grape berry moth flight in wind tunnels and noted that this
species engages in flight at wind speeds of 0.5-1.0 m/s, suggesting that E. viteana are not
strong fliers. The negative correlation between wind speed and the distance flown by
moths before landing on a pane trap was weak (r2 = 0.15) but this was expected given
that wind speeds around 3.0 m/s inhibit flight in small insects (Kisirnoto and Sogawa
1995). Wind speed and precipitation had a greater influence on moths released in the
woods than relative humidity or air temperature, probably because greater distances of
dispersal require more effort than short intra-habitat movements. It is noteworthy to find
that there could be differences in flight capacity or flight direction between different
generations of this species, perhaps in response to vine phenology, though there are no

studies on E. viteana to indicate this.

96

The difference in direction of moth movement and wind shown above indicates
that moths have some control over their dispersal. Pheromone trap-recaptured moths
generally flew south when they were released at the vineyard center (Figure 4.6a), at a
tangent from where the wind direction. Male moths exhibited more directed flight than
females (Figure 4.7) who were distributed in many directions. Perhaps this is because
males fly across oncoming wind in order to catch pheromone plumes from females as
they are emitted further downwind, whereas females are moving to locate oviposition
substrate. The case for a directed flight in grape berry moth can be better made with the
example of moths released in the woods (Figure 4.8). In this case, regardless of the
presence of pheromone traps located in directions fi'om 135-360° from the woods release
point, moths flew east towards the vineyard. This would signify that pheromone plumes
were not responsible for driving the movement, but rather that moths were flying towards
hosts. , Because six of these moths were recaptured in the woods (before crossing the
interface towards the vineyard, it is to be expected that they encountered the pheromone
plume and were diverted from their direction. An important finding is that moths released
in the woods flew significantly greater distances than moths released in the middle of the
vineyard, which is perhaps a sign that E. viteana is able to exhibit short and long-range
flights.

Pheromone traps were successful at re-capturing marked male moths and passive-
interception traps proved useful in capturing moths of both sexes. Special attention
should be given to pane trap placement, since they need to be sufficiently within the

grape canopy so as to interfere with moth flight, but yet leaves should not stick to them.

97

The fact that more females than males were captured in pane traps suggests a
difference in flight strategies between sexes. For example, females tend to fly inside the
canopy, where clusters are located, and hit the pane traps located within the leaves and
fruit more easily than males who are flying more externally to catch pheromone drift. A
difference in flight movement between sexes may also be indicated by the fact that time
between release and recapture was positively correlated with distance flown for males
captured in pane traps but not females (Table 4.3). Females released in the vineyard did
not fly long distances before their recapture (Table 4.2). This could mean that females
move in short bouts and displace little in a given amount of time, whereas males engage
in more prolonged flights and therefore reach farther distances sooner. The difference in
female and male maximum displacement was somewhat consistent throughout eight
releases, with males displacing farther (Table 4.2) which may be due to difference in
behaviors of each sex while searching for virgin mates or oviposition substrate.

The data obtained are the first record of displacement for E. viteana of either sex,
and it will be important to follow with studies on maximum flight capacity in this species.
The average maximum displacement of female E. viteana was ten times lower than for
the oriental fruit moth, C. molesta (Hughes and Dom 2002), but more studies are
necessary to assess how this translates into flight ability and capacity. This species may
be poor fliers as has been suggested by Nagarkatti et al. (20023), or short fliers that make

repetitive bouts of movement.

98

Malaise trapping

Malaise trapping was not an effective passive method for capturing E. viteana,
although the traps themselves may be usefiil for other species. Multiple non-target
specimens were collected, mainly Diptera, some Hymenoptera and a few Lepidoptera.
However, E. viteana did not appear to respond to the trap at the rate expected (only five
moths captured), perhaps because the grape berry moth is not a very good flier and
because flight takes place mostly within the canopy of the vines which are not thoroughly
sampled by the trap. In addition to the behavioral aspect, several physical factors
contributed to the failure of the Malaise trapping experiment. Powerful winds exerted
excessive force against the bi-directional Malaise traps at the top of the poles, made more
susceptible by the fact they stood 9.0 m high. Several of the poles bent and/or fell apart,
bringing down the Malaise traps with them. The traps had sturdy and flexible PVC
frames that never broke, but the veil (tulle) ripped and the insect collectors fell apart from
the impact, scattering the insects collected during the week. Samples from different sites
and different weeks were thus lost, in addition to the fact that some insect containers of
the ground traps (at 2.15 m high) disappeared. The experiment was kept going
nonetheless all through the summer, but because of the missing samples, it was deemed

incomplete.

Passive movement of E. viteana
The study on leaf movement yielded interesting results regarding the potential for
passive movement of the overwintering stage of E. viteana. This showed that there is

movement of dead leaves in vineyards during the winter season, and that this movement

99

is away from the vine on which the leaf grew. Furthermore, this movement is directed
towards the exterior of the vineyard toward adjacent habitats. This movement may
potentially affect the distribution and survival of overwintering grape berry moths that
pupate in leaves. The position of painted leaves in the spring indicated that leaves on
border vines disperse during the winter months towards the woods more than into the
vineyard (Figure 4.9), while interior leaves disperse less and tend to stay near the vines
from which they fall (Figure 9). Sampling along the woods transects for the N-S sets
indicated that leaf movement not only occurred towards the north, it also tended towards
the east, to distances up to 29 m away from the vine of origin. Likewise, sampling for the
E-W sets yielded painted leaves 60 m away to the north of the vine they dropped fiom.
This reveals not only a considerable capacity for dispersal of a passively moving pupa,
but also that winds may be a factor to take into consideration when assessing the impact
of grape berry moth on a vineyard. Wind direction will influence dead leaf transport
during the winter months. The impact of snow cover and rain and patterns of water
drainage should be explored further to see if it has any impact on leaf dispersal during the
winter months. However, it is important to state that I do not know whether a painted leaf
moves as would an unpainted leaf. I assume that they do, and that it is likely the paint did
not hamper the passive transportation of the leaves. Likewise, I do not know how long
pupae stay on the leaves before these rot, since painted and unpainted leaves were found
in varying levels of decomposition. It would be necessary to address these questions in
order to confirm movement of pupae within the vineyard.

There are several final points to be made about E. viteana movement within the

grape system of Michigan. The displacement distances observed in this study would

100

suggest that E. viteana does not move too far within the grape agroecosystem. On the
other hand, we do not know whether this species undertakes long flights because the
farthest trap in this study was located less than 150 m frommoth origin. It is important to
study flight capacity in the grape berry moth, to determine expected distances covered by
moths moving between habitats, the propensity for long-range flights in E. viteana, and
whether this dispersal is affected by age, sex or physiological state. Afterwards, it will be
important to remember that some exchange of moths between cultivated and wild grapes
can be favorable to growers because the exchange of genes may help delay development
of resistance (N agarkatti et al. 2002b). The studies described here demonstrate that moths

can move between wild and cultivated habitats by both active and passive transport.

101

CHAPTER 5:

CONCLUSIONS AND FUTURE RESEARCH NEEDS

Research presented in this thesis helps to explain why pheromone traps, helpful in
determining the onset of grape berry moth flight, fail to provide accurate information on
the pest pressure in vineyards later in the season. These traps attract only males, and may
lose attractiveness as the season progresses due to competition with virgin females, as has
been found for codling moth (Howell 1974) and the sunflower moth (Pyralidae) (Aslarn
et al. 1990). Traps placed at 1.5 m height in the woods near vineyards may misrepresent
the degree of pest pressure expected fi'om the surrounding habitat because they would
underestimate the population size in the latter part of the season. The findings in Chapter
2 suggest that male moths preferably fly in the woods canopy, and they may aggregate in
areas where females are more likely to be, near fruit clusters where oviposition must take
place. Wild grapes seldom fruit in the shade of the woods under a dense canopy, and
grapevines are more likely to survive by climbing on trees to reach sunlight. High male
abundance in the woods canopy therefore could indicate that females are present near
fi'uit clusters, similar to males of the webbing clothes moth (Tineidae) that are more likely
to successfully mate if they first find adequate larval substrate for females (Takacs et al.
2002). Future studies should verify the distribution of female moths in this system, and
understand the mechanisms that trigger grape berry moth attraction to grapes. The
development of attract and kill strategies for grape berry moth would be improved if we
knew more about mating behavior (e. g. how many times E. viteana mates), since there is
the possibility that males that have contacted a toxicant may pass it along to the female

during mating.

102

The data in Chapter 3 demonstrate that high levels of infestation in vineyard
borders are largely influenced by the proximity of the vineyard to habitats containing
wild grape. The area of influence of these adjacent habitats would be a suitable object of
future studies, similar to landscape context research of Thies et al. (2003). We need to
find out if the effect of deciduous woods, for example, varies with distance from the
vineyard. We need to explore the effect that woods size and density of woodlots at the
landscape level may have on grape berry moth pest pressure. This can and should all be
linked to studies on grape berry moth flight capacity, because the spatial context should
take into consideration the dispersal capacity of the insect. Ultimately, these studies could
help elucidate why mating disruption practices have failed in some areas (Dennehy et al.
1990, R. Isaacs, pers. comm), or have great potential in others (Trimble et al. 1993).

The ability to predict deleterious pest impact on a vineyard may depend on more
than one parameter of the wild grape or its habitat near to a vineyard. Presence of wild
grape fi'uit clusters in woods were found to be insufficient for reliable prediction of pest
impact in adjacent vineyards (Chapter 3), and the same was found for grapevine
abundance. However, improved predictive power may be possible if these two parameters
could be combined to develop a factor representing the abundance of fruiting vines. To
illustrate this, results fi'om Chapter 3 were combined so that sites where wild fi'uit clusters
were observed would carry greater weight than sites where clusters were not observed,
independent of each site’s wild grape index. The following equation was applied to
calculate a combined predictor (CP):

CP=WGI+(FxWGI)

103

where WGI = sum of vines observed at the site (same as in Chapter 3) and F = fruit
presence at the site (0 or 1). Taking July of 2001 as an example, by using CP values in a
regression analysis (REG PROCEDURE, SAS Institute, 1996) with cluster infestation
levels in vineyard, coefficients of determination increased while significance was
observed for both borders (P = 0.004, :3 = 0.30) and interiors (P = 0.04, I? = 0.17),
compared to data from Chapter 3 (Figures 3.2 and 3.3). This example illustrates the
possibilities of being able to better predict E. viteana pest impact on a given month
through the use of a new wild host parameter.

Having determined at least that the presence of wild gapevines in adjacent
habitats can be injurious to neighboring vineyards; the removal of wild vines may
decrease pressure fiom this pest. However, removing wild hosts may be deleterious to
natural enemies of E. viteana (Seaman et al. 1990, Landis 1994, Nagarkatti et al. 2002a).
Future research should focus on a thorough survey of parasitoids and other natural
enemies of this species in Michigan, to compare with the studies of Slingerland (1904),
Seaman et a1. (1990), and Nagarkatti et al. (2002a) and determine how much they benefit
grape culture by reducing the incidence of pest insects.

Another topic that should be addressed in the future is how different species of
Vitis are used by gape berry moths. We do not know whether E. viteana prefer a
particular species of wild vines, but there are reports of some varieties of cultivated
labrusca (e.g. Catawba) that are less susceptible to this insect (B. Blum, pers. comm.)
These studies should include female moth preference for these different species of wild
gapes and also survival rates of the several stages from 1St instar to adult. The results of

such studies should then be linked to studies on natural enemies, to determine whether E.

104

viteana population dynamics differ not only according at the vineyard-uncultivated
habitat scale, but also within uncultivated habitats.

Movement of gavid females fiom woods into vineyards to lay eggs on cultivated
gapes is a sigrificant possibility (Dennehy et al. 1990, Trimble 1993) as was concluded
in Chapter 4. Future research should address flight capacity of the gape berry moth as
conducted previously with other fi'uit pests (Dom et al. 1999) to determine differences
between sexes, age, and mating status. Flight behavior data can then be joined with
information on landscape context studies to be able to rationally assess gape berry moth
impact on vineyards in a certain region. It is also important to consider the possibility that
different races of E. viteana (proposed by Tobin et a1. 2002) have different strategies and
may behave differently in different gape agoecosystems.

Although not presented in this thesis, I often detected moth presence in the woods
traps when vineyard traps did not catch any. Consequently, one. implication of this
research for gape gowers is that by monitoring moth presence in uncultivated habitats,
the quality of the information (e. g. onset of flight, hotspots of abundance near vineyards)
would geatly improve if gowing degee day models are shown to help predict optimal
spray timings with more accuracy. Pheromone traps should be placed inside and at the
edge of woods, at the highest level the scout can reach without it taking too much time to
check the traps. Secondly, these studies indicate the need to take the entire habitat into
consideration when selecting a site for vineyard establishment. Uncultivated habitats
nearby may cause future pest pressure on the gapes, as is the case with deciduous woods.
Increasing the distance between these areas and the prospective vineyard should serve as

a preventive measure in pest management, though there is still a need to determine this

105

minimum distance. Cultivation of other crops around gapes can serve as a ‘safety belt’
against immigation by the gape berry moth, and casual observations of vineyards in SW
Michigan indicate that these buffers are highly effective in maintaining a low population
of gape berry moth.

Some of the vineyards studied indicated very high pest pressure, pointing to the
need for integated strategies to reduce gape berry moth populations in and around
vineyards. This includes, as stated above, locating uncultivated areas, and within these,
sampling for wild gape and determining the potential risk they pose for gape berry moth
infestation (i.e. presence of clusters). Deciduous woods and fruiting wild vines will be the
most important factors to sample. Wild gape clusters and clusters in the vineyard border
should be examined for presence of natural enemies. Finally, returning to cultural
practices implemented in the 18003 and early 1900s, such as raking and chopping of
leaves in the fall would be advisable, since this would help destroy part of the population
that is overwintering in the vineyard.

The conclusions of this study are a lesson in not underestimating nature.
Grape berry moth populations have been affected by the removal of woodlots and their
replacement with crops of different kinds, but they have been able to colonize vineyards
and flourish in the wood fragnents left within this landscape mosaic. Uncultivated
habitats are typically considered beneficial to agriculture by their provision of refuges for
parasitoids and predators (e. g. Kareiva 1983). However, when they harbor pest insects
and allow them to escape management practices applied to the vineyard crop, they can be

detrimental to the production of gapes. Insects have dynamic populations that thrive to

106

occupy every possible niche present in agoecosystems, and this research shows that this

is true for E. viteana in the gape gowing regions of Michigan.

107

REFERENCES CITED

Alm, S.R., D.L. Reichard, RN. Williams, and ER. Hall. 1989. Mortality of first-instar
gape berry moths (Lepidoptera: Tortricidae) due to dosage and size of spray drops
containing fenpropathrin. Journal of Economic Entomology 82: 1180-1184.

Altieri, M.A. 1983. Vegetational designs for insect habitat management. Environmental
Management 7: 3-7.

Anonymous. 1911. 40th Annual Report of the Secretary of the State Horticultural Society
of Michigan for the year 1910. Lansing, MI.

Aslam, M., (3.13. Wilde, and T.L. Harvey. 1990. Monitoring flight activity of sunflower
moth (Lepidoptera: Pyralidae) in Kansas. Environmental Entomology 19: 1639-1645.

Ball, B. D., and D.F. Lovitt. 1968. Orchard and vineyard inspection and removal. Plant
Pest Control Progams, Michigan Department of Agriculture, MI.

Ball, B. D., and D.F. Lovitt. 1969. Neglected orchard and vineyard removal. Pest
Control Progams, Michigan Department of Agriculture, MI.

Ball, B. D., and D.F. Lovitt. 1971. Neglected orchard and vineyard removal. Plant Pest
Control Progams, Michigan Department of Agriculture, MI.

Banks, J.E. 2000. Natural vegetation in agoecosystems: pattern and scale of
heterogeneity, pp. 215-230. In B. Ekbom, M.E. Irwin and Y. Robert (eds), Interchanges
of insects between agicultural and surrounding landscapes. Kluwer Academic
Publishers, Dordrecht, The Netherlands.

Barrett, G.W. 2000. The impact of corridors on arthropod populations within simulated
agolandscapes, pp. 71—84. In B. Ekbom, M.E. Irwin and Y. Robert (eds), Interchanges of
insects between agicultural and surrounding landscapes. Kluwer Academic Publishers,
Dordrecht, The Netherlands.

Batschelet, E. 1981. Circular statistics in biology, Academic Press, London, UK.
Beerwinkle, K.R., J.D. L6pez, Jr., D. Cheng, P.D. Lingren, and R.W. Meola. 1995.
Flight potential of feral Helicoverpa zea (Lepidoptera: Noctuidae) males measured with a
32-channel, computer-monitored, flight-mill system. Environmental Entomology 24:
1 122-1 130.

Bernays, E.A. 2001. Neural limitations in phytophagous insects: Implications for diet
breadth and evolution of host affiliation. Annual Review of Entomology 46: 703-727.

108

Biever, K.D., and D.L. Hostetter. 1989. Phenology and pheromone trap monitoring of
the grape berry moth, Endopiza viteana Clemens (Lepidoptera: Tortricidae) in Missouri.
Journal of Entomological Science 24: 472-481.

Boiteau, G., W.P.L. Osborn, X. Xiong, and Y. Bousquet. 2000a. The stability of
vertical distribution profiles of insects in air layers near the gound. Canadian Journal of
Zoology 78: 2167-2173.

Boiteau, (2., Y. Bousquets, and W. Osborn. 2000b. Vertical and temporal distribution
of Carabidae and Elateridae in flight above an agicultural landscape. Environmental
Entomology 29: 1 157-1163.

Burel, F., J. Baudry, Y. Delettre, S. Petit, and N. Morvan. 2000. Relating insect
movements to farming systems in dynamic landscapes, pp. 5-32. In B. Ekbom, M.E.
Irwin and Y. Robert (eds), Interchanges of insects between agicultural and surrounding
landscapes. Kluwer Academic Publishers, Dordrecht, The Netherlands.

Cardé, R.T., and AK. Minks. 1995. Control of moth pests by mating disruption:
Successes and constraints. Annual Review of Entomology 40: 559-5 85.

Cisneros, JJ., and J.A. Rosenheim. 1998. Changes in the foraging behavior, within-
plant vertical distribution, and microhabitat selection of a generalist insect predator: an
age analysis. Environmental Entomology 27: 949-957.

Clark, L.G., and TJ. Dennehy. 1988. Oviposition behavior of gape berry moth.
Entomologia Experimentalis et Applicata 47: 223-230.

Clemens, B. 1860. [Title unknown]. Proceedings of the Academy of Natural Sciences,
Philadelphia. p. 359 [original description].

Colvin, J. 1995. The regulation of migation in Helicoverpa armigera. pp. 265-277. In
V.A. Drake and AG. Gatehouse (eds.) Insect Migration. Tracking resources through
space and time. Cambridge University Press, Cambridge, UK.

Cooter, M., and NJ. Armes. 1993. Tethered flight technique for monitoring the flight
performance of Helicoverpa armigera (Lepidoptera: Noctuidae). Environmental
Entomology 22: 339-345.

Cronin, J.T., K. Hyland, and W.G. Abrahamson. 2001. The pattern, rate, and range of
within-patch movement of a stem-galling fly. Ecological Entomology 26: 16-24.

Davis, M.A. 1980. Why are most insects short fliers? Photocopy, pp. 103- 111.
Dartmouth College, Hanover, NH.

109

Deland, J.P., G.J.R. Judd, and B.D. Roitberg. 1994. Disruption of pheromone
communication in three sympatric leafroller (Lepidoptera: Tortricidae) pests of apple in
British Columbia. Environmental Entomology 23: 1084-1090.

Dennehy, T.J., C.J. Hoffman, and J.S. Kamas. 1990a. Grape berry moth risk
assessment reduces insecticide use 70 percent. Cornell Cooperative Extension
Chautauqua County Agicultural News 72(5).

Dennehy, T.J., C.J. Hoffman, J.P. Nyrop, and M.C. Saunders. 1990b. Development
of low-spray, biological and pheromone approaches for control of gape berry moth,
Endopiza viteana Clemens, in the Eastern United States, pp. 261-282. In N. Bostanian, L.
Wilson and T.J. Dennehy (eds.) Monitoring and integated management of arthropod
pests of small fruit crops. Intercept Ltd, Andover, UK.

Denno, RF. and G.K. Roderick. 1991. Influence of patch size, vegetation texture, and
host plant architecture on the diversity, abundance, and life history styles of sap-feeding
herbivores, pp. 169-196. In S.S. Bell, B.D. McCoy, and HR. Mushinsky (eds.), Habitat
structure: The physical arrangement of objects in space. Chapman and Hall, London, UK.

Derrick, M.E., J.W. van Duyn, C.E. Sorenson, and G.G. Kennedy. 1992. Effect of
pheromone trap placement on capture of male European corn borer (Lepidoptera:
Pyralidae) in three North Carolina crops. Environmental Entomology 21: 240-246.

Dethier, V. G. 1941. Chemical factors determining the choice of foodplants by Papilio
larvae. American Naturalist 75. 61-63.

DeVries, P.J., T.R. Walla, and H.F. Greeney. 1999. Species diversity in spatial and
temporal dimensions of fruit-feeding butterflies from two Ecuadorian rainforests.
Biological Journal of the Linnean Society 68: 333-353.

Dorn, S., P. Schumacher, C. Abivardi, and R. Meyhofer. 1999. Global and regional
pest insects and their antagonists in orchards: spatial dynamics. Agriculture, Ecosystems
and Environment 73: 111-118.

Downie, D.A., and J. Granett. 2000. Distribution, abundance, and short-term
persistence of gape phylloxera (Homoptera: Phylloxeridae) populations in two regions of
the native range. Environmental Entomology 28: 1004-1013.

Drake, V.A., and A.G. Gatehouse. 1995. Insect Migation. Tracking resources through
space and time. Cambridge University Press, Cambridge, UK.

Ehrlich, P.P., and PH. Raven. 1964. Butterflies and plants: A study in coevolution.
Evolution 18: 586-608.

110

Ekbom, B. 2000. Interchanges of insects between agicultural and surrounding
landscapes, pp. 1-3. In B. Ekbom, M.E. Irwin and Y. Robert (eds.) Interchanges of insects
between agicultural and surrounding landscapes. Kluwer Academic Publishers,
Dordrecht, The Netherlands.

Fasoranti, J.O., R.I. Gara, and D.R. Geiszler. 1982. Laboratory studies on the flight
capacity of the Mahogany shootborer, Hypsipyla grandella (Zeller) (Lepidoptera,
Pyralidae). Journal of Applied Entomology (Z. ang. But.) 93: 182-186.

Feldhege, von M., F. Louis, and H. Schmutterer. 1995. Untersuchungen fiber
Falterabundanzen des Bekreuzten Traubenwicklers Lobesia botrana Schiff. Irn Weinbau.
Anz. Schadlingskde., Pflanzenschutz, Umweltschutz 68: 85-91.

Fergusson-Kolmes, L.A., and T.J. Dennehy. 1993. Differences in host utilization by
populations of North American gape phylloxera (Homoptera: Phylloxeridae). J oumal of
Economic Entomology 86:1502-1511.

Fermaud, M., and R. Le Menu. 1989. Association of Botrytis cinerea with gape berry
moth larvae. Phytopathology 79: 651-656.

Fermaud, M., and R. Le Menn. 1992. Transmission of Botrytz’s cinerea to gapes by
gape berry moth larvae. Phytopathology 82: 1393-1398.

Gabel, B., and V. Renczés. 1985. Factors affecting the monitoring of flight activity of
Lobesia botrana and Eupoecilia ambiguella (Lepidoptera, Tortricidae) by pheromone
traps. Acta Entomologia Bohemoslov 82: 269-277.

Galet, P. 1979. A practical arnpelogaphy. Grapevine identification. American Species.
140-153.

Garcia-Salazar, C., and D.A. Landis. 1997. Marking T richogramma brassicae
(Hymenoptera: T richogarnmatidae) with fluorescent marker dust and its effect on
survival and flight behavior. Journal of Economic Entomology 90: 1546-1550.

Hagler, J.R., and CG. Jackson. 2001. Methods for marking insects: current techniques
and future prospects. Annual Review of Entomology 46: 511-543.

Hamilton, J.G., and M.P. Zalucki. 1993. Interactions between a specialist herbivore,
Crocidosema plebejana, and its host plants Malva parviflora and cotton, Gossypium
hirsutum: Oviposition preference. Entomologia Experimentalis et Applicata 66: 207-212.

Hamstead, E.O., R.E. Dolphin, and W. Roelofs. 1972. Field test of virgin female and
synthetic sex-lure traps for male redbanded leafrollers. Environmental Entomology 1:
488-489.

111

. Hendrickson, AH. 1913. 42nd Annual Report of the Secretary of the State Horticultural
Society of Michigan for the year 1912, Lansing, MI.

Hodges, R.W. 1983. Check list of the Lepidoptera of America North of Mexico. E.W.
Classey Ltd and The Wedge Entomological Research Foundation. London, UK.

Hoffman, C.J., and T.J. Dennehy. 1989. Phenology, movement, and within-field
distribution of the gape berry moth, Endopiza viteana Clemens (Lepidoptera:
Tortricidae), in New York vineyards. Canadian Entomologist 121: 325-335.

Hoffman, C.J., T.J. Dennehy, and J.P. Nyrop. 1992. Phenology, monitoring, and
control decision components of the gape berry moth (Lepidoptera: Tortricidae) risk
assessment progam in New York. Journal of Economic Entomology 85: 2218-2227.

Howell, J.F. 1974. The competitive effect of field populations of codling moth on sex
attractant trap efficiency. Environmental Entomology 3: 803-807.

Howell, J.F., R.S. Schmidt, D.R. Horton, S.U.K. Khattak, and L.D. White. 1990.
Codling moth: male moth activity in response to pheromone lures and pheromone-baited
traps at different elevations within and between trees. Environmental Entomology 19:
573-577.

Hughes, J., and S. Dorn. 2002. Sexual differences in the flight performance of the
oriental fruit moth, Cydia molesta. Entomologia Experimentalis et Applicata 103: 171-
182. ,

Humphrey, J.W., C. Hawes, A.J. Peace, R. Ferris-Kaan, and M.R. Jukes. 1999.
Relationships between insect diversity and habitat characteristics in plantation forests.
Forest Ecology and Management 113: 11-21.

Isard, S.A., J.L. Spencer, M.A. Nasser, and E. Levine. 2000. Aerial movement of
western corn rootworm, Diabrotica virgifera virgifera (Coleoptera: Chrysomelidae): Diel
periodicity of flight activity in soybean fields. Environmental Entomology 29: 226-234.

Isely, D. 1917. Control of the gape berry moth in the Erie-Chautauqua gape belt.
(USDA) U. S. Department of Agiculture Bull. 550. 44 pp. Washington DC.

J eanneret, P. 2000. Interchanges of a common pest guild between orchards and the
surrounding ecosystems: a multivariate analysis of landscape influence, pp. 85-108. In B.
Ekbom, M.E. Irwin and Y. Robert (eds.) Interchanges of insects between agicultural and
surrounding landscapes. Kluwer Academic Publishers, Dordrecht, The Netherlands.

Johnson, CG. 1969. Migation and dispersal of insects by flight. Methuen, London, UK.

Johnson, F., and A.G. Hammar. 1912. The gape berry moth. (USDA) U. S.
Department of Agiculture Entomological Bull. 116, Part II .71 pp. Washington DC.

112

Johnson, D.T., R.L. Mayes, B.A. Lewis, and J.M. Domek. 1988. Grape berry moth,
geen June beetle and gape root borer studies. Proceedings of the Arkansas State
Horticultural Society 109: 197-202.

Juillet, J.A. 1963. A comparison of four types of traps used for capturing flying insects.
Canadian Journal of Zoology 41: 219-223.

Kareiva, P. 1983. Influence of vegetation texture on herbivore populations: Resource
concentration and herbivore movement. pp. 259-289. In R.F. Denno and MS. McClure
(eds.) Variable plants and herbivores in natural and managed systems. Academic Press,
New York, NY.

Karg, G., and A.E. Sauer. 1995. Spatial distribution of pheromone in vineyards treated
for mating disruption of the gape vine moth Lobesia botrana measured with
electroantennogams. J. Chem. Ecol. 21: 1299-1314.

Karg, G., and A.E. Sauer. 1997. Seasonal variation of pheromone concentration in
mating disruption trials against European gape vine moth Lobesia botrana (Lepidoptera:
Tortricidae) measured by EAG. Journal of Chemical Ecology 23: 487-501.

Keil, S., H. Gu, and S. Dorn. 2001. Response of Cydia pomonella to selection on
mobility: laboratory evaluation and field verification. Ecological Entomology 26: 495-
501.

Kisimoto, R., and K. Sogawa. 1995. Migation of the brown planthopper Nilaparvata
lugens and the white-backed planthopper Sogatella filrcrfera in East Asia: the role of
weather and climate. pp. 67-91. In V.A Drake and AG. Gatehouse (eds.) Insect
Migation. Tracking resources through space and time. Cambridge University Press,
Cambridge, UK.

Knight, A.L. 2000. Monitoring codling moth (Lepidoptera: T ortricidae) with passive
interception traps in sex pheromone-treated apple orchards. Journal of Economic
Entomology 93: 1744-1751.

Landis, D.A. 1994. Arthropod sampling in agicultural landscapes: ecological
considerations. pp. 15-31. In L.P. Pedigo and GD. Buntin (eds.) Handbook of sampling
methods for arthropods in agriculture. CRC Press, Boca Raton, FL.

Lewis, B., and D. Johnson. 1999. Grape berry moth pest management progam.
Proceedings Annual Hort. Industries Show 18:56-60.

Light, D.M., A. K. Knight, C.A. Henrick, D. Rajapaska, B. Lingren, J.C. Dickens,
K.M. Reynolds, R.G. Buttery, G. Merrill, J. Roitman, and B.C. Campbell. 2001. A
pear-derived kairomone with pheromonal potency that attracts male and female codling

moth, Cydia pomonella (L.). Naturwissenschaften. 88: 333-338.

113

Macdonald, D.W., and H. Smith. 1990. Dispersal, dispersion and conservation in the
agicultural ecosystem, pp. 18-64. In R.G.H. Bunce and DC. Howard (eds.) Species
dispersal in agricultural habitats. Belhaven Press, London, UK.

Marine, P.C., and D.A. Landis. 2000. Parasitoid community structure, pp. 183-193. In
B. Ekbom, M.E. Irwin and Y. Robert (eds), Interchanges of insects between agricultural
and surrounding landscapes. Kluwer Academic Publishers, Dordrecht, The Netherlands.

Martinson, T.E., C.J. Hoffman, T.J. Dennehy, J .S. Kamas, and T. Weigle. 1991. Risk
assessment of gape berry moth and guidelines for management of the Eastern
leafhopper. New York’s Food and Life Sciences Bull. 138.

Matteucci, S.D., and A. Colma. 1982. Metodologia para el estudio de la vegetaeién. The
General Secretariat of the Organization of American States, Washington DC.

Mellaned, F.D., P.C. Marine, S.H. Gage, and D.A. Landis. 1999. Does agricultural
landscape structure affect parasitism and parasitoid diversity? Ecological Applications 9:
634-641.

Miller, J.R., and KL. Strickler. 1984. Finding and accepting host plants, pp.127-157.
In W.J. Bell and RT. Cardé (eds), Chemical Ecology of Insects. Chapman and Hall Ltd.,
New York, NY.

Mo, J., G. Baker, M. Keller, and R. Roush. 2003. Local dispersal of the diamondback
moth (Plutella xylostella (L.)) (Lepidoptera: Plutellidae). Environmental Entomology 32:
71-79.

Morano, L.D., and M.A. Walker. 1995. Soils and plant communities associated with
three Vitis species. American Midland Naturalist 134: 254-263.

Muirhead-Thomson, RC. 1991. Trap responses of flying insects: the influence of trap
desigi on capture efficiency. Academic Press Limited, London, UK.

Mullins, M.C., A. Bouquet, and LE. Williams. 1992. Biology of the gape vine.
Cambridge University Press, Cambridge, UK.

Nagarkatti, S., A.J. Muza, and M.C. Saunders. 2000. Meridic diet for rearing
Endopiza viteana Clemens (Lepidoptera: Tortricidae). Canadian Entomologist 32: 259-
261.

N agarkatti, 5., RC. Tobin, and M.C. Saunders. 2001. Diapause induction in the gape
berry moth (Lepidoptera: Tortricidae). Environmental Entomology 30: 540-544.

114

Nagarkatti, S., M. Muza, M.C. Saunders, and P.C. Tobin. 2002a. Role of the egg
parasitoid Trichogramma minutum in biological control of the gape berry moth,
Endopiza viteana. BioControl 47: 373-385.

Nagarkatti, S., P.C. Tobin, A.J. Muza, and M.C. Saunders. 2002b. Carbaryl resitance
in populations of gape berry moth (Lepidoptera: Tortricidae) in New York and
Pennsylvania. Journal of Economic Entomology 95: 1027-1032.

Nyrop, J.P., and G.A. Simmons. 1986. Temporal and spatial activity patterns of an
adult parasitoid, Glypta filmiferanae (Hymenoptera: Ichneumonidae), and their influence
on parasitism. Enviromnental Entomology 15: 481-487.

Oriana, 1994. Oriana for windows. Ver. 1.06. Kovaeh Computing Services, Pentraeth,
UK.

Pasek, J.E. 1988. Influence of wind and windbreaks on local dispersal of insects.
Agriculture, Ecosystems and Environment 22/23: 539-554.

Pearsall, I.A., and J.H. Myers. 2001. Spatial and temporal patterns of dispersal of
Western flower thrips (Thysanoptera: Thripidae) in nectarine orchards in British
Columbia. Journal of Economic Entomology 94: 831-843.

Peterson, A. 1926. Additional information on baits attractive to the oriental peach moth,
Laspeyresia molesta Busck. Journal of Economic Entomology 19: 429-239.

Pettit, RH. 1932. The grape berry moth in 1932. (MAES) Michigan Agricultural
Experimental Station Quarterly Bull. 14, 4 pp. East Lansing, MI.

Pettit, RH. 1933. The principal gape insects of Michigan. Agricultural Experimental
Station Special Bulletin 239. 19 p. Michigan State College of Agriculture and Applied
Science, East Lansing, MI.

Pfeiffer D.G., T. J. Boucher, M.W. Lachance, and J.C. Killian. 1990. Entomological
research in Virginia (USA) vineyards. pp. 45-61. In N. Bostonian, L. Wilson, and J.J.
Dennehy (eds.) Monitoring and integated management of arthropod pests of small fi'uit
crops, Intercept Ltd, Andover, UK.

Piper, R.W., and S.G. Compton. 2002. A novel technique for relocating concealed
insects. Ecological Entomology 27: 251-253.

Poole, R.W., and P. Gentili. 1996. Nomina insecta neartica. A checklist of the insects of

North America. Volume 3: Diptera, Lepidoptera, Siphonaptera. Entomological
Information Service, Rockville.

115

Rees, C.J.C. 1983. Microclirnate and the flying Hemiptera fauna of a primary lowland
rain forest in Sulawesi. pp. 121-135. In S.L. Sutton, T. C. Whitrnore, and AC. Chadwick
(eds.), Tropical rain forest: ecology and management, Blackwell Scientific Publications,
Oxford, UK.

Richards, P.W. 1983. The three—dimensional structure of tropical rain forest. pp. 3-10. In
S.L. Sutton, T. C. Whitrnore, and AC. Chadwick (eds) Tropical rain forest: ecology and
management, Blackwell Scientific Publications, Oxford, UK.

Riedl, H., S.A. Hoying, W.W. Barnett, and J.E. DeTar. 1979. Relationship of within-
tree placement of the pheromone nap to codling moth catches. Environmental
Entomology 8: 765-769.

Roelofs, W.L, J.P. Tette, E.F. Taschenberg, and A. Comeau. 1971. Sex pheromone of
the gape berry moth: Identification by classical and electro-antennogam methods and
field tests. Journal of Insect Physiology 17: 2235-2243.

Roland, J. and RD. Taylor. 1997. Insect parasitoid species respond to forest structure at
different spatial scales. Nature 386:710-713.

Rosenthal, G.A., and D.H. Janzen. 1979. Herbivores: Their interaction with secondary
plant metabolites. Academic Press, Orlando FL.

Rothschild, G.H.L., and AK. Minks. 1977. Some factors influencing the performance
of pheromone traps for oriental fruit moth in Australia. Entomologia Experimentalis et
Applicata 22: 171-182.

Safonkin, A.F. 1988. Infrastructure of the population of the geat brown tortricid Archips
podana Sc. (Lepidoptera: Tortricidae) in relation to coincidence with its various food-
plants. Doklady Akademia Nauk. SSSR. 296: 499-500.

Sanders, C.J., D.R. Wallace, and G.S. Lucuik. 1978. Flight activity of female Eastern
spruce budworrn (Lepidoptera: Tortricidae) at constant temperatures in the laboratory.
Canadian Entomologist 110: 627-632.

Sappington, T.W., and W.B. Showers. 1992. Reproductive maturity, mating status, and
long—duration flight behavior of Agrotis ipsilon (Lepidoptera: Noctuidae) and the
conceptual misuse of the oogenesis-flight syndrome. Environmental Entomology 21:
677-688.

Sappington, T.W., and W.B. Showers. 1993. Influence of larval starvation and adult
diet on long-duration flight behavior of the migatory moth Agrotis ipsilon (Lepidoptera:
Noctuidae). Environmental Entomology 22: 141-148.

SAS Institute. 1996. PROC user’s manual, version 8.0, 5‘h ed. SAS Institute Inc., Cary,
NC.

116

Sauer, A.E., and G. Karg. 1998. Variables affecting pheromone concentration in
vineyards treated for mating disruption of gape vine moth Lobesia botrana. Journal of
Chemical Ecology 24: 289-302.

Savopoulou-Soultani, M., P.G. Milonas, and D.G. Stavridis. 1998. Role of availability
of food to the adult Lobesia botrana (Lepidoptera: T ortricidae) in its reproductive
performance. Journal of Economic Entomology 91: 1341-1348.

Schmitz, V., R. Roehrich, and J. Stockel. 1995. Disruption mechanisms of pheromone
communication in the European gape moth Lobesia botrana Den. & Schiff. (Lep.,
Tortricidae) 11. Influence of the population density and the distance between insects for
males to detect the females in atmosphere impregnated by pheromone. J oumal of Applied
Entomology 119: 303-308.

Schowalter, T.D. 2000. Insect ecology. An ecosystem approach. Academic Press, San
Diego, CA.

Schumacher, P., A. Weyeneth, D.C. Weber, and S. Dorn. 1997. Long flights in Cydia
pomonella L. (Lepidoptera: Tortricidae) measured by a flight mill: influence of sex,
mated status and age. Physiological Entomology 22: 149-160.

Sciarretta, A., P. Trematerra, and J. Baumgartner. 2001. Geostatistical analysis of
Cydia fimebrana (Lepidoptera: Tortricidae) pheromone trap catches at two spatial scales.
American Entomologist 47: 174-184.

Seaman, A.J., J.P. Nyrop, and T.J. Dennehy. 1990. Egg and larval parasitism of the
gape berry moth (Lepidoptera: Tortricidae) in three gape habitats in New York.
Environmental Entomology 19: 764-770.

Sharp, J.L., J.R. McLaughlin, and D.R. Bennett. 1976. Flight mill studies with the
soybean looper (Lepidoptera: Noctuidae). Journal of the Georgia Entomological Society
11: 105-110.

Showers, W.B., R.B. Smelser, A.J. Keaster, F. Whitford, J.F. Robinson, J.D. L6pez,
and S.E. Taylor. 1989. Recapture of marked black cutwonn (Lepidoptera: Noctuidae)
males after long-range transport. Environmental Entomology 18: 447-458.

Siegel, S. 1956. Nonparametric statistics for the behavioral sciences. McGraw-Hill Book
Company, New York, NY.

Slingerland, M.V. 1904. The gape berry moth. Cornell University Agricultural
Experimental Station of the College of Agriculture Bull. 223: 43-60.

Smythe, RA. 1913. 42'“1 Annual Report of the Secretary of the State Horticultural
Society of Michigan for the year 1912, Lansing, MI.

117

Solemn, M.G. 1981. Windbreaks as a source of orchard pests and predators, pp. 273-
283. In J .M. Thresh (ed.), Pests, pathogens and vegetation. Pitrnan Advanced Publishing
Progam, Boston, Massachussetts.

Su, J.C., and S.A. Woods. 2001. Importance of sampling along a vertical gadient to
compare the insect fauna in managed forests. Environmental Entomology 30: 400-408.

Sutton, S.L. 1983. The spatial distribution of flying insects in tropical rain forests. pp.
77-91. In S.L. Sutton, T. C. Whitrnore, and AC. Chadwick (eds.), Tropical rain forest:
ecology and management, Blackwell Scientific Publications, Oxford, UK.

Takacs, S., G. Gries, and R. Gries. 2002. Where to find a mate? Resource-based sexual
communication of webbing clothes moth. Naturwissenschaflen 89: 57-59.

Taschenberg, E.F., R.T. Cardé, and W.L. Roelofs. 1974. Sex pheromone mass
trapping and mating disruption for control of redbanded leaf roller and gape berry moths
in vineyards. Environmental Entomology 3: 239-242.

Taschenberg, E.F., and W.L. Roelofs. 1977. Mating disruption of the gape berry moth,
Paralobesia viteana, with pheromone released from hollow fibers. Environmental
Entomology 6: 761-763.

Taylor, LR. 1974. Insect migation, flight periodicity and the boundary layer. Journal of
Animal Ecology 43: 225-238.

Taylor, P.S., and EJ. Shields. 1990. Flight thresholds of the armyworm (Lepidoptera:
Noctuidae). Environmental Entomology 19: 1410-1417.

Thies, C., and T. Tscharntke. 1999. Landscape structure and biological control in
agoecosystems. Science 285:893-895.

Thies, C., I. Steffan-Dewenter, and T. Tscharntke. 2003. Effects of landscape context
on herbivory and parasitism at different spatial scales. Oikos 101: 18-25.

Tobin, P.C., S. Nagarkatti, and M.C. Saunders. 2001. Modeling development in gape
berry moth, Endopiza viteana Clemens (Lepidoptera: T ortricidae). Environmental
Entomology 30: 692-693.

Tobin, P.C., S. Nagarkatti, and M.C. Saunders. 2002. Diapause maintenance and
termination in the gape berry moth (Lepidoptera: Tortricidae). Environmental
Entomology 31: 708-713.

Toepfer, S., H. Gu, and S. Dorn. 2002. Phenological analysis of spring colonization of

apple trees by Anthonomus pomorum. Entomologia Experimentalis et Applicata 103:
151-159.

118

Torres-Vila, L.M., J. Stockel, and M. C. Rodriguez-Molina. 1997. Physiological
factors regulating polyandry in Lobesia botrana (Lepidoptera: Tortricidae). Physiological
Entomology 22: 387-393.

Trimble, R., D. Pree, P. Vickers, and K.W. Ker. 1991. Potential of mating disruption
using sex pheromone for controlling the gape berry moth, Endopiza viteana Clemens
(Lepidoptera: Tortricidae), in Niagara peninsula, Ontario vineyards. Canadian
Entomology 123: 451-460.

Trimble, RM. 1993. Efficacy of mating disruption for controlling the gape berry moth,
Endopiza viteana Clemens (Lepidoptera: Tortricidae), a case study over three consecutive
gowing seasons. Canadian Entomology 125: 1-9.

van Emden, H.F. 1965. The role of uncultivated land in the biology of crop pests and
beneficial insects. Scientific Horticulture 17: 121-136.

Voss, E.G. 1985. Michigan flora. A guide to the identification and occurrence of the
native and naturalized seed-plants of the state. Part II. DICOTS (Saururaceae-Comaceae).
Cranbrook Institute of Science Bull. 59, and University of Michigan Herbarium.

Weissling, T.J., and A.L. Knight. 1994. Passive trap for monitoring codling moth
(Lepidoptera: Tortricidae) flight activity. Journal of Economic Entomology 87: 103-107.

Welch, RC. 1990. Dispersal of invertebrates in the agricultural environment, pp. 203-
218. In R.G.H. Bunce and D.C. Howard (eds.) Species dispersal in agicultural habitats.
Belhaven Press, London, UK.

Williams III, L., and T.E. Martinson. 2000. Colonization of New York vineyards by
Anagrus spp. (Hymenoptera: Mymaridae): overwintering biology, within-vineyard
distribution of wasps, and parasitism of gape leafhopper, Erythroneura spp. (Homoptera:
Cicadellidae), eggs. Biological Control 18:136-146.

Williams, I.S., T.H. Jones, and S.E. Hartley. 2001. The role of resources and natural
enemies in determining the distribution of an insect herbivore population. Ecological
Entomology 26: 204-21 1

Wise, J.C., L.J. Gut, R. Isaacs, A.M.C. Schilder, B. Zandstra, E. Hanson, and B.
Shane. 2003. Michigan fi'uit Management Guide. Extension Bull. E-154, Michigan State
University Extension, East Lansing, MI.

Witzgall, P., M. Bengtsson, and RM. Trimble. 2000. Sex pheromone of the gape
berry moth (Lepidoptera: T ortricidae). Environmental Entomology 29: 433-436.

Woiwod, I.P., and J.A. Stewart. 1990. Butterflies and moths — migation in the

agicultural habitat, pp. 189-202. In R.G.H. Bunce and D.C. Howard (eds), Species
Dispersal in agicultural habitats. Belhaven Press, London, UK.

119

Wratten, S.D., and C.F.G. Thomas. 1990. Farm-scales spatial dynamics of predators
and parasitoids in agricultural landscapes, pp. 219-237. In R.G.H. Bunce and D.C.
Howard (eds.) Species dispersal in agicultural habitats. Belhaven Press, London, UK.

120

APPENDICES

121

APPENDIX 1
Taxonomic note on Endopiza viteana Clemens.

As described in Chapter 1, there has been some confusion regarding the taxonomy
of E. viteana. In Clemens' (1860) description of viteana, he placed a question mark next
to the genus to which he assigned the species, Endopiza. In point of fact, Endopiza is a
misspelling of Endopisa (J. Brown', pers. comm), which is now considered a synonym
of Grapholita (Poole and Gentili 1996). When Obraztsov revised the goup in 1953, he
recogrized that viteana and relatives did not belong in the same genus as botrana (i.e.,
Lobesia), nor did they belong in Endopisa; hence, he described Paralobesia to
accommodate these species (J. Brown, pers. comm.). Diakonoff in 1973 interpreted
Clemens' Endopiza as a valid genus (with the type species of viteana), rather than a
misspelling, and relegated Paralobesia to the status of a junior synonym of Endopiza
(i.e., Endopiza is the senior synonym by priority) (Hodges 1.983). Powell in 1983
followed Diakonoff, recogrizing Endopiza as a valid genus, and since the mid-19808 the
species has been referred to most frequently in the literature as Endopiza viteana
Clemens. However, according to J. Brown (pers. comm.) Diakonoff was incorrect in
resurrecting Endopiza because a misspelling cannot be interpreted as the proposal of a
new genus. For this reason, the Obraztsov genus Paralobesia is valid and the correct
name should be Paralobesia viteana (Clemens). This is the name that Brown uses in the
new World Catalogue on Tortricids to be published in spring of 2004 (J. Brown, pers.
comm); however, until then and for simplicity, I refer to this species by the name in the
title of this dissertation.

1 Systematic Entomology Laboratory, USDA, c/o National Museum of Natural History, Washington D.C.

122

APPENDIX 2
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.: 2003-07
Title of thesis or dissertation (or other research projects):
BEHAVIORAL ECOLOGY OF THE GRAPE BERRY MOTH,
ENDOPIZA VITEANA CLEMENS, IN MICHIGAN AGROECOSYSTEMS
Museurn(s) where deposited and abbreviations for table on following sheets:

Entomology Musemn, Michigan State University (MSU)

Other Museums:

Investigator’s Name(s)
Natalia Botero-Garcés ..................

 

 

Date Augr_1_st 22, 2003

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

Deposit as follows:
Original: Include as Appendix 1 in ribbon copy of thesis or dissertation.

Copies: Include as Appendix 1 in copies of thesis or dissertation.
Museurn(s) files.
Research project files.

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

123

Appendix 2.1

Voucher Specimen Data

 

New“ six .3.

amaze: o.

 

59:22 2.: E gmoaou

 

moom «mound. NN Ema

 

 

 

.8 mcoEBoam 3E. o>onm 05 32081

 

hormoow .oZ .mco:o>

mooamoéeom m=Emz
A825 @952 $2859...
$8800: 2 98% $5225 33

 

 

Page 1 of 1 Pages

ooowriqrow
$5:on moomN
.00 cmEmm .=2

Sammie.
$0506.28 m__9mz

:233 ..oo :23 cm> ...2

88-82 8.8 cocoa,
mwocmwrouooom £582
25n— 3m& :00 £0.50 55> ._2

.328 £22 Edam 880 Set
Foomomow 8.8 30:?
$0506.23 m__£mz
ocficmn 6mm :00 £29.. ...2

95.290 memo»: muqobcm

 

Museum

where

deposited

 

Other

 

Adults
Adults

 

 

Pupae

 

Nymphs

 

Larvae
Eggs

 

cogmoaou cam tom:
.6 880:8 29:63.... no“. Sun .33

 

:98» 550 do 83on

 

 

a

O

donEnz

 

 

124