THE ROLE OF NATURAL ENEMIES IN PEST MANAGEMENT OF CODLING MOTH
(CYDIA POMONELLA [L.]) IN ORGANIC APPLE ORCHARDS IN MICHIGAN
By
Nathaniel James Walton

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Entomology - Doctor of Philosophy
Ecology, Evolutionary Biology and Behavior - Dual Major
2013

ABSTRACT
THE ROLE OF NATURAL ENEMIES IN PEST MANAGEMENT OF CODLING MOTH
(CYDIA POMONELLA [L.]) IN ORGANIC APPLE ORCHARDS IN MICHIGAN
By
Nathaniel James Walton
The codling moth (Cydia Pomonella [L.]) is the number one worldwide pest of
apples, pears and other pome fruits. Conservation of natural enemies is one way to
reduce the costs of managing this pest by reducing the need for insecticide sprays.
Successful conservation of the codling moth’s natural enemies in Michigan requires 1)
identifying them and 2) determining major factors that affect their abundance and
effectiveness in the apple agroecosystem. The first chapter of my dissertation presents
a literature review of the major natural enemies of codling moth by life stage. The
second chapter presents a field study using real time video observation of sentinel prey
to identify codling moth’s key natural enemies in Michigan apple orchards. The third
chapter provides experimental data and analysis aimed at identifying major factors that
may impact the efficacy of natural enemies in Michigan orchards. In chapter 4 I present
the results of two laboratory investigations exploring behavioral factors affecting codling
moth infection by entomopathogenic nematodes. I identified, ants, lacewings, spiders,
and predatory beetles as major natural enemies of codling moth in my experimental
apple orchards. Relative abundances of these organisms were very site-specific.
Organically certified orchards had higher natural enemy abundance and attack rate of
sentinel fifth instar codling moth compared to non-organic orchards. This was not found
for codling moth egg natural enemy abundance or activity. Further investigations
revealed that orchard sanitation and orchard floor vegetation management affect natural

enemy abundance and effectiveness against fifth instar codling moth larvae and codling
moth pupae on the orchard floor. The implications of these findings for organic pest
management and for future avenues of research are also discussed in the concluding
chapter.

ACKNOWLEDGEMENTS
The work contained in this dissertation could not have been completed without
the help of innumerable researchers worldwide who provided the groundwork upon
which this research was based. In addition to those individuals, I would also like to
thank specifically a number of people who were directly involved in helping me to carry
out the research studies contained herein. Firstly, my graduate advisor Matthew
Grieshop without whose support none of this would have been possible. I'd also like to
thank Jim Koan and Steve Tennes for providing their farms as research sites. Michelle
Brosius assisted in carrying out the majority of the field and laboratory experiments
contained in this dissertation and her insight was always invaluable. Other members of
the Organic Pest Management team that provided assistance during the course of this
research include: Emily Pochubay, Ben Phillips, Krista Buehrer, Joe Riddle, Vianney
Willot, Phil Kavouriaris, Shane Celeste, Kristy Morell, Kristin Deroshia, Michael Kelleher,
Matt Quenado, and many others. I would also like to thank the members of my
Graduate Advisory Committee: Doug Landis, Larry Gut, and Chris Klausmeier. This
research was funded primarily by the USDA, Ceres Trust, and Project GREEEN.
Finally, I would like to thank Emily Pochubay, Dave Epstein, Anne Nielsen, Julianna
Wilson, Rufus Isaacs, and Steppin' in it, for providing guidance, advice, moral support,
and dancin' music throughout my entomological career. Thanks also, of course to my
loving family, Bob Walton, Marsha Walton, Amy and Brad McNelly, and Carlos Santana
(the cat).

iv

TABLE OF CONTENTS
LIST OF TABLES .......................................................................................................... viii
LIST OF FIGURES ..........................................................................................................ix
Chapter 1. Codling moth (Cydia pomonella [L.]) natural enemies in apple orchards: over
a century of research....................................................................................................... 1
1.1. Introduction ........................................................................................................... 1
1.2. Major natural mortality factors of the codling moth by life stage: .......................... 3
1.2.1. Eggs and neonates ........................................................................................ 3
1.2.1.1. Environmental conditions......................................................................... 3
1.2.1.2. Predators and Parasitoids ....................................................................... 4
1.2.2. Major natural mortality factors of the codling moth by life stage: Inside the
apple (larval instars 2-4)........................................................................................... 8
1.2.2.1. Environmental conditions......................................................................... 8
1.2.2.2. Predators and Parasitoids ....................................................................... 9
1.2.3. Major natural mortality factors of the codling moth by life stage: Wandering
larvae and overwintering larvae ............................................................................. 10
1.2.3.1. Environmental conditions....................................................................... 11
1.2.3.2. Predators and Parasitoids ..................................................................... 11
1.2.3.3. Pathogens ............................................................................................. 20
1.2.4. Major natural mortality factors of the codling moth by life stage: Adults ...... 21
1.3. Discussion: Biological control and codling moth natural mortality factors ........... 21
Chapter 2. Video observations of the natural enemies of eggs and fifth instar codling
moth Cydia pomonella (L.) in Michigan apple orchards (USA) ...................................... 28
2.1. Introduction ......................................................................................................... 28
2.1.1. Natural enemies in organic agriculture ......................................................... 29
2.1.2. Evaluating the impacts of orchard management on natural enemies of
codling moth ........................................................................................................... 30
2.2. Materials and Methods ....................................................................................... 31
2.2.1 Videography .................................................................................................. 31
2.2.2 Experimental design ...................................................................................... 32
2.2.3. 2010 Experiments ........................................................................................ 32
2.2.3.1. Sentinel Prey ......................................................................................... 32
2.2.3.2. Data collection ....................................................................................... 33
2.2.4. 2011 Experiments ........................................................................................ 34
2.2.4.1. Sentinel Prey ......................................................................................... 34
2.2.4.2. Data collection ....................................................................................... 34
2.2.5. Video observation ........................................................................................ 35
2.2.6. Natural enemy behavior ............................................................................... 36
2.2.7. Data analysis................................................................................................ 36
2.3. Results ................................................................................................................ 37
2.3.1. 2010 First Generation Egg Predation ........................................................... 37

v

2.3.2. 2010 Fifth Instar Predation ........................................................................... 38
2.3.3. 2011 First Generation Egg Predation ........................................................... 39
2.3.4. 2011 Second Generation Egg Predation ...................................................... 40
2.3.5. 2011 Fifth Instar Predation ........................................................................... 41
2.3.6. Yellow Sticky Traps ...................................................................................... 42
2.3.7. Pitfall Traps .................................................................................................. 44
2.3.8 Natural Enemy Behavior ............................................................................... 45
2.4. Discussion .......................................................................................................... 49
2.5. Conclusion .......................................................................................................... 50
Chapter 3. Habitat characteristics influencing foraging success of codling moth
antagonists in apple orchards ....................................................................................... 52
3.1. Introduction ......................................................................................................... 52
3.1.1. Orchard floor complexity and fifth instar codling moth mortality ................... 52
3.1.2. Canopy connectivity and mortality of codling moth eggs .............................. 55
3.2. Methods .............................................................................................................. 56
3.2.1. Study Sites ................................................................................................... 56
3.2.2. Orchard floor complexity experiments .......................................................... 56
3.2.3. Experimental Design .................................................................................... 56
3.2.4. Natural Enemy Treatments .......................................................................... 57
3.2.5. Complexity Treatments ................................................................................ 57
3.2.6. Cages ........................................................................................................... 57
3.2.7. Sentinel Prey ................................................................................................ 57
3.2.8. Experiment 1: Entomopathogenic nematodes.............................................. 58
3.2.9. Experiment 2: Lacewing larvae .................................................................... 59
3.2.10. Experiment 3: Earwigs ............................................................................... 59
3.2.11. Experiment 4: Canopy connectivity ............................................................ 60
3.2.12. Statistical analysis ...................................................................................... 62
3.3. Results ................................................................................................................ 63
3.3.1. Entomopathogenic nematodes..................................................................... 63
3.3.2. Lacewing larvae ........................................................................................... 67
3.3.3. Earwigs ........................................................................................................ 68
3.3.4. Canopy connectivity ..................................................................................... 71
3.4. Discussion .......................................................................................................... 75
3.4.1. Orchard floor complexity .............................................................................. 75
3.4.2. Canopy connectivity ..................................................................................... 78
3.5. Conclusion .......................................................................................................... 78
Chapter 4: An exploration of how moisture and structural complexity may affect the
foraging success of Steinernema feltiae Filipjev against codling moth (Cydia pomonella
[L.]). ............................................................................................................................... 80
4.1. Introduction ......................................................................................................... 80
4.1.2. Experiment 1: Habitat structural complexity and EPN foraging efficiency .... 82
4.1.3. Experiment 2: Soil moisture and codling moth cocoon site selection ........... 83
4.2. Materials and Methods ....................................................................................... 84
4.2.1. Experimental organisms ............................................................................... 84

vi

4.2.2. Habitat structural complexity and EPN foraging efficiency ........................... 84
4.2.3. Soil moisture and codling moth cocoon site selection .................................. 86
4.3. Results ................................................................................................................ 88
4.3.1. Habitat structural complexity and EPN foraging efficiency ........................... 88
4.3.2. Soil moisture and codling moth cocoon site selection .................................. 89
4.4. Discussion .......................................................................................................... 91
4.4.1. Habitat structural complexity and EPN foraging efficiency ........................... 91
4.4.2. Soil moisture and codling moth cocoon site selection .................................. 92
4.5. Conclusions ........................................................................................................ 92
Chapter 5. Towards more sustainable management of codling moth (Cydia pomonella
[L.]) in apple orchards.................................................................................................... 94
5.1. Introduction ......................................................................................................... 94
5.2. Future Research ................................................................................................. 98
5.3. Trade-offs in conservation biological control of codling moth ........................... 100
5.4. Conclusion ........................................................................................................ 101
APPENDIX .................................................................................................................. 103
REFERENCES ............................................................................................................ 106

vii

LIST OF TABLES
Table 1.1. Mortality estimates for codling moth by life stage and mortality agents. ....... 23
Table 2.1. Summary of codling moth egg natural enemy activity in video observations
compared with rank abundances and mean abundances on yellow sticky traps (YSTs)
pooled across years and management regimes. ........................................................... 47
Table 2.2. Summary of codling moth fifth instar natural enemy activity in video
observations compared with rank abundances and mean abundances in pitfall traps
pooled across management regimes. ........................................................................... 48
Table 5.1. Extant estimates of codling moth mortality in the field: Summary................. 95
Table 5.2. Estimates of codling moth mortality in Michigan apple orchards .................. 98
Table A.1 List of voucher specimens........................................................................... 104

viii

LIST OF FIGURES
Figure 2.1. Mean proportion of sentinel codling moth eggs attacked by natural enemies
on video from June 2010 in organic and conventionally managed orchards. ................ 38
Figure 2.2. Mean proportion of sentinel codling moth fifth instars attacked by natural
enemies on video from August and September 2010 in organic and conventionally
managed orchards. ....................................................................................................... 39
Figure 2.3. Mean proportion of sentinel codling moth eggs attacked by natural enemies
on video from June 2011 in organic and conventionally managed orchards. ................ 40
Figure 2.4. Mean proportion of sentinel oriental fruit moth eggs attacked by natural
enemies on video from August 2011 in organic and conventionally managed orchards.
...................................................................................................................................... 41
Figure 2.5. Mean proportion of sentinel codling moth fifth instars attacked by natural
enemies on video from August and September 2011 in organic and conventionally
managed orchards. ....................................................................................................... 42
Figure 2.6. Mean number of natural enemies on unbaited yellow sticky traps in organic
and conventional orchards in June 2010. ...................................................................... 43
Figure 2.7. Mean number of natural enemies on unbaited yellow sticky traps in organic
and conventional orchards in June 2011. ...................................................................... 44
Figure 2.8. Mean number of natural enemies in pitfall traps in organic and conventional
orchards in September 2010. ........................................................................................ 45
Figure 3.1. Mean proportion mortality of sentinel codling moth larvae in experimental
cages at AlMar orchard in June of 2012. ....................................................................... 64
Figure 3.2. Mean proportion mortality of sentinel codling moth in experimental cages at
Clarksville Horticultural Experiment Station in June of 2012. ........................................ 65
Figure 3.3. Mean proportion mortality of sentinel codling moth larvae in experimental
cages at AlMar orchard in June of 2012. ....................................................................... 67
Figure 3.4. Mean proportion mortality of sentinel codling moth pupae in experimental
cages at AlMar orchard in June of 2012. ....................................................................... 68
Figure 3.5. The mean number of predatory beetle larvae found preying on sentinel
codling moth in the field cages at Clarksville in September 2012.................................. 70

ix

Figure 3.6. Proportion codling moth mortality versus the mean number of predatory
beetle larvae in the cage experiments at Clarksville in September of 2012. ................. 71
Figure 3.7. Proportion mortality of sentinel codling moth eggs in high and low
connectivity orchards at AlMar in June 2012. ................................................................ 72
Figure 3.8. The mean number of Chrysopid larvae found in beat samples of the canopy
at the edges of plots at AlMar orchards in June 2012. .................................................. 74
Figure 4.1. Proportion mortality of sentinel codling moth larvae in experimental
microcosms. .................................................................................................................. 89
Figure 4.2. The proportion of larvae found in the Sand vs. Cardboard strata by moisture
level in experimental microcosms. ................................................................................ 90

x

Chapter 1. Codling moth (Cydia pomonella [L.]) natural enemies in apple
orchards: over a century of research
1.1. Introduction
The codling moth is a severe pest of apples in most pome fruit growing regions
worldwide (Barnes, 1991; Lacey and Shapiro-Ilan, 2008; Simon et al., 2007; Stern et al.,
1959). Apple producers in the United States in 2011 produced 4.2 million metric tons of
apples worth 2.7 billion dollars (USDA-NASS, 2012a). In the same year, US apple
producers applied an estimated 621 metric tons of insecticides to 84% of the total US
apple bearing acreage (USDA-NASS, 2012b). Integrated pest management (IPM) and
conservation biocontrol are approaches that apple growers can use to reduce the
frequency of these costly insecticide applications (Jones et al., 2009; Stern et al., 1959;
Van Driesche and Bellows, 1996; Zhang and Swinton, 2009).
Conservation of natural enemies has always been a key element in the theory
behind IPM (Stern et al., 1959). In practice, growers can conserve natural enemies by
withholding insecticide applications, using selective insecticide chemistries or
application methods, and or providing resources for natural enemies (Morris and Winter,
1999). Unfortunately, codling moth directly damages the crop resulting in a low
tolerance for this pest. Thus, in many apple orchards management of the codling moth
still relies largely on suppression of the pest through the use of synthetic or natural
toxins that have a negative impact on natural enemies (Epstein, 2010; Jones et al.,
2009; Simon et al., 2010). This reliance on insecticides has limited the role of natural
enemies in pest management and likely has caused an underestimation of the

1

economic value of the contribution that natural enemies make to pest management in
apple orchards (van der Geest and Evenhuis, 1991; Zhang and Swinton, 2009).
The purpose of this review was to assemble the scientific literature investigating
the contribution that natural enemies make to codling moth management. Sources were
limited primarily to studies that provide estimates of codling moth mortality and the
causal agents of this mortality. Studies surveying natural enemy populations or diversity
were not included unless they also provide data linking codling moth mortality to
measured natural enemy data. References that did not provide functional information on
codling moth biological control were excluded to provide conciseness and clarity.
Studies that only measured the results of augmentative releases were also not included.
Although there is a focus on arthropod natural enemies, I endeavored to include all
natural mortality factors of the codling moth that have been studied.
The review is organized by life stage of the codling moth from egg to adult. This
structure allows comparison of mortality estimates for each life stage to stage structured
matrix population models, which estimate the stages of codling moth that have the
greatest impact on long term population dynamics of the species (Mills, 2005b).
Mortality agents are listed in the life stage in which they cause mortality not the life
stage in which they infect codling moth. This is particularly relevant for the parasitoids of
the codling moth that attack it before during or after the larva penetrates the apple but
do not cause mortality until after the codling moth larva emerges from the apple (Mills,
2005b). There is a great deal of variation among authors in the terminology used to
describe the phenological stages of the codling moth life cycle. I follow the terminology
used by Geier (1963).

2

1.2. Major natural mortality factors of the codling moth by life stage:
1.2.1. Eggs and neonates
The egg and neonate larva (first larval instar) are two of the least protected life
stages of the codling moth. Codling moth eggs are laid individually on leaves or
developing apples and the survival of neonates is dependent on their ability to find a
fruit, successfully penetrate the outer skin, and construct a silken shelter just beneath
the fruit surface (Geier, 1963; MacLellan, 1962; Simpson, 1903). High percentages of
mortality have been estimated for these life stages in the field. These estimates range
from 75% (Ferro et al., 1975; Summerland and Steiner, 1943) to 25% (Jackson, 1982)
total mortality from egg laying to fruit entry.
1.2.1.1. Environmental conditions
Codling moth eggs are probably not damaged significantly by the UV light
frequencies that they are exposed to in the field (Caldwell et al., 2007; Paul and GwynnJones, 2003). However, low humidity (45-50% RH) in a laboratory setting has been
shown to reduce neonate emergence (Hagley, 1972). Jackson and Harwood (1980)
performed a series of laboratory experiments to evaluate the influence of temperature
and humidity on neonate codling moth survival. They found that survival times
decreased as temperature increased and as humidity decreased. The maximum
survival time was 300 h for larvae held at 4 ºC in 100% RH. Neonate codling moth
larvae survived for less than an hour at 45 ºC. They also found that starved first instar
larvae had a harder time penetrating apples at higher temperatures when RH was held
at 100%. The duration of starvation also negatively affected the first instar larvae’s
abilities to penetrate apples (Jackson and Harwood, 1980). Rainfall has been reported

3

to negatively impact neonate survival either by washing them off of foliage or by flooding
their entrance tunnels (Geier, 1963; Jackson, 1982; MacLellan, 1962).
1.2.1.2. Predators and Parasitoids
Jaynes and Marucci (1947) quantified the biological control potential of the
codling moth’s natural enemies in unsprayed orchards in West Virginia, USA. They
monitored codling moth populations in two nearby orchards, one that received
insecticide sprays as needed, and another orchard, which received only a dormant oil
spray, a pink spray of lime-sulfur and a petal fall spray of lime-sulfur, lime, and lead
arsenate. They used sentinel eggs laid on cellophane paper and pinned in orchard trees
to measure predation and parasitization of codling moth eggs. Parasitization in the
sprayed orchard was considerably more variable than in the unsprayed orchard
presumably because insecticide sprays reduced the populations of parasitoids
temporarily. Parasitization was consistently higher in the unsprayed orchard, but this
effect was much greater in the first year of the study (1938) than in the following two
years. Reduced parasitization in the sprayed orchard in the first two years was
attributed to insecticide sprays at the time of second-generation oviposition (August).
All reported parasitization was by the egg parasitoid Trichogramma minutum (Riley)
(Hymenoptera: Trichogrammatidae), which parasitized up to 28% of first-generation
eggs.
In the same study, predation of eggs was very inconsistent and was only different
between the sprayed and unsprayed orchard in one year (1939). They performed visual
searches for egg predators but found very little direct evidence of egg predation in the
field. The only predator that was observed feeding on codling moth eggs in the field was

4

a predaceous thrips, Leptothrips mali (Fitch) (Thysanoptera: Phlaeothripidae), which
migrated from senescing daisies on the orchard floor to the apple foliage where they
attacked second-generation eggs. The predaceous thrips population as recorded by
timed visual counts was correlated with higher predation rates of sentinel codling moth
eggs in both types of orchards (16.4 – 21.9% in unsprayed, 6.4-16.6% in sprayed). In
laboratory experiments, ladybeetle larvae (Coleoptera: Coccinellidae) and green
lacewing larvae (Neuroptera: Chrysopidae) did not feed on codling moth eggs, but
adults of the ladybeetles Anatis quindecimpunctata (Olivier), Hippodamia convergens
Guerin-Meneville, and Coccinella nevomnotata Herbst did feed on codling moth eggs
(Jaynes and Marucci, 1947).
Summerland and Steiner (1943) followed the development of 80 eggs on
unsprayed trees in a 60-acre orchard May-June in Indiana, USA. Of these eggs, 37%
hatched, 32% were parasitized by T. minutum, 19% were destroyed by predators, 5%
died of unknown causes, and 7% disappeared. They identify Chrysopa spp.
(Neuroptera: Chrysopidae) as the primary codling moth egg predator. During JulySeptember the fate of 160 eggs was monitored on the same trees. In this time period,
an average of 25% of the eggs hatched, 43% were parasitized by T. minutum, 23%
were destroyed by predators, 4% died of unknown causes, and 5% disappeared. In
total, 37% of the first brood, and 25% of the second brood eggs hatched (Summerland
and Steiner, 1943).
MacLellan (1962) evaluated the importance of arthropod predators to the control
of eggs and first instar codling moth in an integrated control orchard in Nova Scotia,
Canada. He used weekly examination of marked fruit clusters to evaluate egg density

5

and predation throughout the season. Predators present in the orchards throughout the
season were counted and identified using a non-lethal sampling method (beating).
Predators accounted for 14.4% mortality and Trichogramma sp. for 2.4%. The majority
of predation was believed to be the work of plant bugs (Hemiptera: Miridae). However,
other predators were found, including two predaceous thrips species Haplothrips faurei
Hood and Leptothrips mali (Fitch). One mite, which was identified as Anystis agilis
Banks (Acari: Anystidae), was also implicated, but believed to largely prey only on
young larvae. The loss of larvae between egg hatch and fruit injury was estimated from
comparisons between egg hatch data and fruit injury and was found to be about 58.5%.
Codling moth populations as measured by trunk banding were low in the year of this
study (0.6 larvae per foot of trunk) (MacLellan, 1962).
Geier (1963) collected branches with fruit clusters and measured the ratio of
hatched eggs to deep entries in Australia’s Capital Territory. He found 114 eggs to 38
st

entries. From these he estimated that approximately one third of 1 instar codling
moths are able to damage fruit after hatching. He provides no data on the causal agents
of this mortality (Geier, 1963).
Glen (1975) glued codling moth eggs to leaf surfaces and apples to measure egg
predation in an orchard where no insecticides were used in Long Ashton, UK. The eggs
were glued singly, five per tree and checked twice weekly for signs of predation or
hatching. They categorized damage as eggs disappeared, sucked out, or hatched. This
was done in two years (1972, 1973). In this study three species of plant bugs
(Hemiptera: Miridae) were identified as the most important agents of mortality for
codling moth eggs in the field, since the majority of eggs that did not hatch had been

6

sucked out. Disappearance of eggs was attributed to earwigs (Dermaptera:
Forficulidae). However, their importance as control agents was questioned by the author
because their habit of chewing holes in the apples facilitated colonization by brown rot
fungus (Monolinia sp.), which made them more of a pest than a beneficial in those cider
orchards (Glen, 1975).
Ferro et al. (1975) collected apple fruit clusters and nearby leaves to evaluate
egg mortality, counted eggs, examined them for signs of predation, and reared
parasitoids from them in the lab. They used data from sleeve cages to estimate first
instar larval mortality by subtracting the number of fruit entries from the number of
hatched eggs. They found that approximately 23% egg mortality resulted simply from
non-viability of eggs, but pathogens may have been responsible for some of this
mortality. They also estimated a 50% mortality of eggs from predators and parasitoids
combined. After successfully hatching, they found an average of only 35% of first instars
survived to damage fruit in sleeve cages (Ferro et al., 1975).
Glen (1977) used codling moths eggs glued to fruit and foliage to estimate egg
predation in unsprayed orchards from 1972 -1974 Long Ashton, UK. An average of 1286% of these eggs were “sucked out” and 3-29% were recorded as disappeared. Three
species of plant bugs (Hemiptera: Miridae), a few species of predatory bug (Hemiptera:
Anthocoridae), and a predatory mite (Anystis sp.) were found on the apple foliage in the
study orchard and were able to feed on codling moth eggs in laboratory trials. The
majority of egg mortality in these orchards was attributed to being attacked (sucked out)
by plant bugs. Earwigs were again believed to have been responsible for the majority of
egg disappearances (Glen, 1977).

7

Hagley et al. (1982) performed a series of laboratory experiments to determine
the feeding preferences of adult ground beetles (Coleoptera: Carabidae) commonly
found in Canadian apple orchards. They identified two species, Amara aenea De Geer
and Stenolophus comma Fabricius that were often collected from apple foliage and
branches and readily consumed early instars of codling moth in lab trials (Hagley et al.,
1982).
Mills (2003; 2005b) reports that Trichogramma platneri Nagarkatti (Hymenoptera:
Trichogrammatidae) can be an important parasitoid of codling moth eggs in California.
Parasitism rates by this species in unsprayed orchards can range from 30-60% (Mills,
2003; Mills, 2005b).
1.2.2. Major natural mortality factors of the codling moth by life stage: Inside the
apple (larval instars 2-4)
Inside of the apple, the codling moth larva is relatively well protected from most
agents of mortality. However, mortality estimates for this life stage have ranged from
10.4% (MacLellan, 1962) to 52% (Ferro et al., 1975). MacLellan (1962), in Nova
Scotia, used weekly examination of marked fruit clusters to evaluate egg density and
predation throughout the season. He estimated based on these data that 10.4%
mortality of larvae occurred inside the fruit. In a field study in eastern Washington (Ferro
et al., 1975), mortality of larvae inside of apples ranged from 35% in early June to 52%
in August. Neither of these studies provides data on the causes of mortality inside the
apple.
1.2.2.1. Environmental conditions

8

Environmental conditions inside of the apple are not thought likely to cause
mortality of codling moth larvae. For example, the relative humidity in the apple is
always high, so higher external temperatures only allow the larva to grow faster (Kuhrt
et al., 2005).
1.2.2.2. Predators and Parasitoids
MacLellan (1958) describes observations of a male downy woodpecker (Picoides
pubescens [L.]) extracting codling moth larvae from infested fruit, but presents no data
to estimate the impact of this behavior on codling moth populations (MacLellan, 1958).
McAtee (1911) also cites several anecdotal reports of P. pubescens removing late instar
codling moth larvae from infested apples.
Ascogaster quadridentata Wesmael (Hymenoptera: Braconidae), Elodia morio
(Fallén) (Diptera: Tachinidae), Hyssopus pallidus (Askew) (Hymenoptera: Eulophidae),
and Pristomerus vulnerator Panzer (Hymenoptera: Ichneumonidae) are the most
common parasitoids reported to attack codling moth larvae before, during, or after they
enter the fruit. However, they do not kill the codling moth larva until after it leaves the
fruit, so estimates of their impact are reported in studies of fifth instar mortality (below).
The only other reported natural enemies of the codling moth inside the apple are
other codling moth larvae. Cannibalism was suspected and/or reported to occur in some
early studies (Geier, 1963). However, Ferro and Harwood (1973) in a laboratory study
found no evidence for cannibalism inside of the apple. They did find that intraspecific
competition resulted in a developmental delay for the inferior competitor inside of apples
(Ferro and Harwood, 1973).

9

1.2.3. Major natural mortality factors of the codling moth by life stage: Wandering
larvae and overwintering larvae
The wandering larva stage is the stage that begins when the codling moth’s fifth
instar leaves the apple and begins to search for a location to pupate and ends when
they spin the cocoon. Wandering larvae are active primarily at night (Borden, 1931;
Garlick and Boyce, 1940; Geier, 1963) and may take several hours to find a suitable
location for their cocoon (Garlick, 1948; Geier, 1963). Fifth instar codling moth larvae
respond to a pheromone emitted by conspecifics that allows them to form aggregations
of from 2-20 individuals for overwintering and pupation (Duthie et al., 2003; Jumean et
al., 2009).
Cocoons can occur anywhere where conditions are relatively dry (Glen and
Milsom, 1978) and they are protected from predators (Chandler, 1928; MacLellan,
1960). The majority of cocoons can be found under flakes of bark on rough barked
apple trees and anywhere from the canopy to the interface with the soil surface
(Chandler, 1928; Garlick, 1948; Jumean et al., 2009; Marshall, 1940). Fewer cocoons
are reported found in the leaf litter and vegetation surrounding the trees (Chandler,
1928; Jaynes and Marucci, 1947; MacLellan, 1960), unless the trees have smooth bark
in which case more larvae seek cocoon sites on the orchard floor (Simpson, 1903).
Artificial substrates are often very attractive to codling moth wandering larvae, so barns,
fence posts, trash, and apple bins have been found to serve as suitable overwintering
and pupation sites (Chandler, 1928; Lacey and Chauvin, 1999; Steiner, 1929).
Estimates of mortality during this life stage are highly variable, from 1.5% to 100%
depending on the type of orchard, location of larvae, and time of year (Geier, 1963;

10

Jaques and MacLellan, 1965; Jaynes and Marucci, 1947; MacLellan, 1960; MacLellan,
1962).
1.2.3.1. Environmental conditions
Newcomer (1920) reported that during an unusual winter storm in eastern
Washington, USA when temperatures reached lower than -29 ºC winter mortality of
codling moth larvae approached 100%. Where the moths were protected by snow cover
or where the temperature stayed above -26 ºC mortality was reportedly closer to 75%
(Newcomer, 1920).
Steiner (1929) performed an experiment to determine the relative overwintering
survival of larvae in cocoons in different materials on the orchard floor. He found that 2
out of 100 larvae spinning cocoons in the soil emerged the following spring. Sixty
percent of larvae that had spun cocoons in plots where the orchard floor was covered in
leaves emerged, in straw 38% survived, and in sod only 6% survived (Steiner, 1929). In
another study, overwintering larvae were kept in an insectary in the orchard and the
average percent mortality for these larvae over the five years of the study was 24.2%
(Garlick, 1948). Finally, MacLellan (1962) used visual searches of tree trunks for
overwintering codling moth larvae to estimate populations, and found that 3% of these
larvae were killed by low temperatures (MacLellan, 1962).
1.2.3.2. Predators and Parasitoids
In an early study of overwintering codling moth mortality, Chandler (1928)
undertook an investigation of the hibernation locations of codling moth larvae in January
and February in southern Illinois. He states, without showing any data, that 90% of the
overwintering larvae in southern Illinois orchards are removed by birds. He also states

11

that larvae hibernating below the soil surface are much safer from bird predators than
larvae on the trunk (Chandler, 1928).
Jaynes and Marucci (1947) in West Virginia, USA, used cardboard bands to
collect larvae and measure predation and parasitization of this life stage, in insecticide
sprayed and unsprayed orchards. The bands were examined periodically, and the
numbers of larvae missing, diseased, or parasitized were recorded. Uninjured larvae
and pupae in the bands were collected and kept to record emergence of parasitoids.
Parasitization rates of this life stage were low (1 – 5.7%) in both orchards. The dominant
parasitoid of this stage was A. quadridentata, but they list 16 other species that
emerged from codling moth larvae and pupae collected from bands (Jaynes and
Marucci, 1947).
Predation of larvae in the bands in this study was highly variable, and often the
species responsible could not be identified based on codling moth remains alone
(Jaynes and Marucci, 1947). Over the three years of the study seasonal average
predation rates ranged from 3.6 – 14.4% in unsprayed and 4.0 – 8.68% in the sprayed
orchard. One ant species, Solenopsis molesta (Say) (Hymenoptera: Formicidae), was
reportedly responsible for the majority of predation of this life stage, but the beetle,
Tenebroides corticalis (Melsheimer) (Coleoptera: Trogossitidae), also consumed a
significant proportion of codling moth larvae in bands in some years (e.g. 14.3% in the
unsprayed orchard in 1940) (Jaynes and Marucci, 1947).
Jaynes and Marucci (1947) also collected drop apples and immature fruit, and
dissected them to determine whether parasitization of larvae in these locations was
higher than in cardboard band traps. They found that parasitization of larvae in drop

12

fruits was very similar to those collected from bands and did not differ between the
sprayed and unsprayed orchard. The parasitism in immature fruit however was higher in
an unsprayed orchard (15.9%) than an insecticide sprayed orchard (4.0%) and was
considerably higher than that found in larvae taken from bands (max. 5%). The
dominant parasitoid of these larvae was A. quadridentata (Jaynes and Marucci, 1947).
To measure codling moth larval predation on the ground, Jaynes and Marucci
(1947) also placed sentinel larvae in pieces of corrugated paper on the soil and under
litter in both orchards. They combined observations in the field and lab to determine the
types of remains that codling moth predators would leave behind and used this to
classify the types of predation that they observed on these sentinel larvae. They found
that predators destroyed almost 100% of these larvae. The majority of predation in the
spring (65% in unsprayed, 69% in sprayed orchard) and fall (78.3% in unsprayed,
87.1% in sprayed) was due to the activity of carabid beetles (adults and larvae) and
cantharid beetle larvae (Chauliognathus spp.) (Coleoptera: Cantharidae). In the
summer, predation by these groups was reduced and the majority of sentinel larva
mortality was caused by Solenopsis spp. and Monomorium spp., ants (Hymenoptera:
Formicidae), which were responsible for 52% and 53% of predation in the unsprayed
and sprayed orchard, respectively. Another notable spring predator in this study was the
elaterid beetle Conderus lividus (De Geer) (Coleoptera: Elateridae), which consumed
about 10% of the sentinel larvae in the unsprayed orchard and 5% in the sprayed
orchard.
In the same study, during one winter (1940 – 41) corrugated paper strips
containing larvae in October were pinned to trees and protected with hardware cloth

13

from predation by birds. These bands were collected in April of the following year and
predation rates of the cocooned larvae were estimated. This yielded figures of 5% and
1.5% predation in the unsprayed and sprayed orchards, respectively. A similar
experiment in the summer showed that 25% of larvae on the tree trunks could be
consumed by predators in the sprayed orchard and 28% in the unsprayed (Jaynes and
Marucci, 1947).
Jaynes and Marucci (1947) also collected the top inch of soil from around apple
trees in both orchards in all three years of the study and used a soil sifter to separate
insects. Additionally, they used old fertilizer bags placed in the orchard and pitfall traps
to catch ground beetles. These sampling methods yielded a great deal of information
about the insect community in these orchards, but no clear patterns were evident. Since
they found predation by ants of the mature larvae of codling moth to be the most
important, they performed several experiments to determine the factors affecting the
effectiveness of ants as natural control agents in their study orchards. What they found
was that primarily two species of ant, Solenopsis molesta (Say) and Formica subsericea
Say, were responsible for the majority of late instar codling moth mortality and that
these two ant species were significantly less abundant in the sprayed orchard compared
to the unsprayed orchard (Jaynes and Marucci, 1947).
Garlick (1948) performed a series of experiments using burlap bands to trap
larvae as part of an investigation of codling moth larval behavior in a heavily infested,
unsprayed orchard in Ontario. Over the course of this five-year-study the study trees
yielded an average of 590 larvae per tree per year. The average parasitism by A.
quadridentata during four years of the study was 18.9% (Garlick, 1948).

14

MacLellan (1958) used cardboard bands baited with codling moth larvae to
measure the importance of woodpeckers (Picoides spp.) for codling moth population
regulation in Nova Scotia, Canada. He also made records of woodpecker predation over
several years (1950 -1956), which were based on characteristic bark punctures made
by woodpeckers when they prey on codling moth under flakes of bark on apple trees.
These data showed that on average woodpeckers removed 52% of the overwintering
codling moth larvae in study orchards, though in individual orchards in some years the
level reached 90% (MacLellan, 1958).
MacLellan (1960) traced the movements of codling moth larvae by dipping them
in a radioactive solution before releasing them in orchards and then using a particle
counter to find them. This was done with first instar larvae, which were recovered later
in the year as mature larvae, or with mature larvae for short-term behavior experiments.
He found that 94% of the larvae that made cocoons on the ground were eaten by
predators (MacLellan, 1960).
MacLellan (1962) evaluated the importance of arthropod predators to the control
of codling moth in an integrated control orchard. In this study visual searches of tree
trunks for overwintering codling moth larvae were used to estimate populations. Using
this method he found that 56% of overwintering larvae were destroyed by predators and
parasitoids, 47% by woodpeckers, and 9% by A. quadridentata (MacLellan, 1962).
Geier (1963) noted that in Australia, larvae that spent a lot of time searching on
the ground were frequently seen being attacked and killed by predators, such as ants,
spiders, and earwigs, but he provides no numerical data to support these observations.

15

Ferro et al. (1975) found that approximately 22% of the first generation fifth
instars emerging from apples were parasitized by A. quadridentata, whereas only about
5% of the second generation were parasitized in Washington, USA. They collected no
data on predation of fifth instars but they estimated based on observations that total
mortality of fifth instar codling moth larvae from the time they left the apple was
approximately 30% in an unsprayed orchard. They also found that about 2.8 – 5% of
pupating larvae did not successfully emerge as adults (Ferro et al., 1975).
Wearing (1975) used cobalt-58 to tag fifth instar codling moth larvae and track
their mortality from the time that they left the fruit to the time that they emerged as
adults the following spring (Wearing, 1975). This study was conducted for six
consecutive years in an orchard of mature trees in New Zealand. The cobalt-58 tagging
method produced an overall estimate of mortality of larvae from fruit exit to cocoon
construction of 57.1% and a total mortality from cocoon construction to adult emergence
of 63.8%. The average total mortality of larvae from fruit exit to adult emergence was
84.8%. The majority of mortality of wandering larvae was attributed to insect predators,
and the majority (82%) of cocoon predation was attributed to birds. Wearing identified
Zoterops lateralis (Latham), as the most common bird predator of codling moth in this
study (Wearing, 1975).
Solomon et al. (1976) used apple logs placed in the field with fifth instar codling
moth on them to measure predation of this life stage by natural enemies in Long
Ashton, UK (1971-1973). They covered some of these logs with netting to exclude
natural enemies and checked the logs at different intervals to record the numbers
remaining and evidence of predation. They recorded an average of 95% predation of

16

larvae on uncovered logs, which they attributed to predation by birds (Solomon et al.,
1976).
Glen and Milsom (1978) used exclusion experiments in a Latin square design to
determine the relative importance of ground beetles and birds to fifth instar codling moth
predation in Long Ashton, UK. They found that predation was always high (90-95%)
when birds were allowed access to trees regardless of whether or not beetles were
excluded. In plots where ground beetles alone were allowed access to trees 56% of the
codling moth larvae survived. There was no statistically significant difference between
codling moth predation in plots with few beetles (beetles excluded) compared to many
beetles. Blue tits (Cyanistes caeruleus [L.]) were found to be the most important bird
predator of codling moth in these orchards. Bird predation was affected by the
availability of alternative prey as well as the density of codling moth (Glen and Milsom,
1978).
Solomon and Glen (1979) observed the foraging behavior of birds in the same
plots with sentinel codling moth larvae from Solomon (1976). In one of these
experiments, logs with similar densities of sentinel larvae were placed in adjacent trees
to see the effect that prey density had on the foraging behavior of the birds. Similar,
more controlled experiments were run in an aviary with great tits (Parus major L.) and
blue tits (C. caeruleus). They found evidence that codling moth larval predation by birds
is consistently ~95%, but were not able to conclusively determine whether this was
density dependent and or regulatory (Solomon and Glen, 1979).
Subrinprasert (1987) in Southern Sweden scraped trunk bark from trees in
unsprayed orchards and measured parasitism of these larvae in the laboratory. This

17

study found a total parasization rate of 38.3% by insect parasitoids. The dominant
parasitoid was the tachinid fly E. morio, which accounted for 50.6% of parasitization in
the study. Other important parasitoids included P. vulnerator, which emerged from
22.9% of the parasitized larvae, and A. quadridentatus, which was found in 14.7% of
parasitized larvae (Subinprasert, 1987).
Hagley et al. (1982) performed a series of laboratory experiments to determine
the feeding preferences of adult carabids commonly found in Canadian apple orchards.
Their results indicate that, of the species tested, Pterostichus melanarius Illiger was the
only one likely to prey on codling moth fifth instar larvae and pupae (Hagley et al.,
1982). Hagley and Allen (1988) in Ontario used serological techniques to detect
predation on fifth instar larvae by eight species of ground beetles. However, they were
unable to correlate the density of any of these species with the number of codling moth
larvae searching for pupation sites in their study orchard. They concluded that P.
melinarius is the most important carabid predator in these orchards, but that carabids as
a group were not a significant factor in the survival of codling moth larvae seeking
pupation sites (Hagley and Allen, 1988).
Riddick and Mills (1994) combined data from laboratory assays and field sentinel
prey studies to determine the potential importance of several species of carabid beetles
as biocontrol agents of fifth instar codling moth in California apple orchards. They
determined that the most important carabid predators of codling moth larvae in these
orchards were Pterostichus spp. They measured a peak consumption of 60% of
tethered fifth instar codling moth larvae per night during early June (Riddick, 1994).

18

Baumgartner (1999) measured predation of overwintering codling moth larvae for
two years in two managed orchards and one abandoned orchard in the Central Coast of
California. In this study the total average mortality of sentinel larvae exposed to all
predators was 53.9%, but in one year in the unmanaged orchard it was as high as
98.7%. When birds were excluded the total average mortality of sentinel overwintering
larvae in the study orchards was about 33%. Although it was not the intention of the
study to test for density dependence, they measured an increase in predation between
years from 15% to 83% when larval density was increased from one or two per tree to
14 larvae per tree (Baumgartner, 1999).
Mathews et al. (2004) used tethered codling moth fourth and fifth instars to
measure nocturnal predation in a newly planted apple orchard in West Virginia. The
study included four different orchard floor treatments, herbicide treated grass stubble,
bare ground, composted manure, and a polyester mulch. They measured the highest
predation rates (81.3%) in the herbicide plots in July. In August and September samples
with human observers present they found no difference between treatments, but
recorded an average predation rate of 51.93% in August and 15.42% in September.
They also found that measured predation rates were significantly lower when human
observers were present. Carabids and ants (Formicidae) were the only predators
observed feeding on sentinel larvae by the human observers in this study (Mathews et
al., 2004).
Mills (2005b) presents data from a trap band study in Europe and Central Asia
from 1992 – 2003. The highest percent parasitism rates of first generation codling moth
fifth instars in that study were for Mastrus ridibundus Gravenhorst (Hymenoptera:

19

Ichneumonidae) in Kazakstan (43.9%), A. quadridentata in Austria (42.6%), Bassus
rufipes Nees (Hymenoptera: Braconidae) in Kazakstan (33.3%), and P. vulnerator in
Switzerland (32.4%). Mills (2005b) also monitored the introduction of three parasitoid
biological control agents of codling moth from 1992-2000 in California apple, pear, and
walnut orchards. Two of the three species introduced did not establish successfully (B.
rufipes and Liotryphon caudatus Ratzeburg), but one species M. ridibundus persisted at
high levels in orchards. This species, which attacks the codling moth cocoon, was
reported to reach parasitism rates as high as 56% in some unsprayed orchards and is
believed to be contributing significantly to codling moth management in California (Mills,
2005a; Mills, 2005b).
1.2.3.3. Pathogens
During the course of an experiment to test the effectiveness of different
formulations of Beauveria bassiana (Bals.-Criv.) Vuill. for codling moth control, Jaynes
and Marucci (1947) measured a background B. bassiana infection rate of 9% on larvae
collected from untreated bands (Jaynes and Marucci, 1947).
Jaques and MacLellan (1965) sampled 1 ft. wide strips of bark around the base
of the tree trunks and evaluated signs of pathogen infection. They found that, from 1950
to 1958, an average of 1.7% of overwintering codling moth larvae were killed by fungal
pathogens. The most common fungal pathogen isolated from these larvae was B.
bassiana (Jaques and MacLellan, 1965).
Subrinprasert (1987) scraped trunk bark from trees in unsprayed orchards in
Southern Sweden and measured pathogen infection rates of these larvae in the
laboratory. They found that 20.5% of the larvae were killed by pathogens. Although they

20

were not able to identify all of the pathogens, B. bassiana and Paecilomyces farinosus
(Holmsk.) A. H. S. Br. and G. Sm., accounted for 34.4 and 29.5% of pathogen infections
in that study, respectively (Subinprasert, 1987).
1.2.4. Major natural mortality factors of the codling moth by life stage: Adults
Adult mortality agents are particularly underrepresented in the literature, because
adult moths are highly mobile, which makes it difficult to study them and their natural
enemies under realistic conditions. Simpson (1903) lists bats, birds, and spiders as
natural enemies of the adult moth, but provides no quantitative estimates for their
impact on pest populations. Hagley (1972) investigated the effect of environmental
conditions on adult mortality by rearing codling moth in green apples at 24 ºC and 75%
RH and a 16 h photoperiod. He then placed recently emerged adult pairs in oviposition
cages at a range of humidity (45 – 90%) and temperatures (15.5 – 36 ºC). He found
that, under constant conditions of 24 ºC and 60-75% RH, 76% of codling moth females
died within 2 days of completing oviposition. He also found that egg viability declined
significantly when females were reared at temperatures exceeding 33 ºC.
1.3. Discussion: Biological control and codling moth natural mortality factors
Table 1.1 summarizes the mortality estimates provided by the studies included in
this review. Some life stages of the codling moth have received more attention in the
literature than others (Table 1.1). Unfortunately, in many of the cited references, sample
sizes, measures of variability, experimental designs, and spatial scales are not given or
are not comparable among studies and or life stages. However, over a hundred years of
research has given us a great deal of useful information about the natural enemies of
the codling moth. For example, we know that plant bugs are important egg predators in

21

some parts of the world (Glen, 1975; Glen, 1977), but that predatory thrips and green
lacewings can also make a contribution to egg predation where plant bugs are
absent(Jaynes and Marucci, 1947; Summerland and Steiner, 1943). We also know that
birds as predators of 5

th

instar larvae and pupae probably make the single largest

contribution to pest management of the codling moth worldwide (Chandler, 1928; Glen
and Milsom, 1978; MacLellan, 1958; MacLellan, 1962; Solomon et al., 1976).
Parasitoids of all life stages have been particularly well studied and we have relatively
good estimates of their impact on codling moth in the major apple growing regions of
the world (Table 1.1). Codling moth pathogens are a group that were somewhat
underrepresented by this review, because they are primarily researched in the context
of augmentative applications (Lacey and Unruh, 2005) and I was restricted to studies of
naturally occurring populations of natural enemies (Table 1.1).

22

Table 1.1. Mortality estimates for codling moth by life stage and mortality agents.
Mortality Agents:
Mortality (%)
Source

Year Geographic region

st

Egg - 1 Instar
Parasitoids
T. minutum
Trichogramma spp.
T. platneri
Green lacewings
Chrysopa spp.
Thrips
Leptothrips mali
Plant bugs

Earwigs
Unspecified

nd

28
32-43
2
30-60

Jaynes and Marucci
Summerland and Steiner
MacLellan
Mills

1947
1943
1962
2003

West Virginia, USA
Indiana, USA
Nova Scotia, Canada
California, USA

19-23

Summerland and Steiner

1943 Indiana, USA

6-22

Jaynes and Marucci

1947 West Virginia, USA

12-86
14
3-29
33
25-50

Glen
MacLellan
Glen
Geier
Jackson

1977
1962
1977
1963
1982

35-52
10

Ferro et al.
MacLellan

1975 Washington, USA
1962 Nova Scotia, Canada

Long Ashton, UK
Nova Scotia, Canada
Long Ashton, UK
Australian Capital Territory
Washington, USA

th

2 - 4 Instar
Unspecified

23

Table 1.1. (cont’d).
Mortality Agents:
5

th

Mortality (%) Source

Year Geographic region

Instar

Birds
Unspecified
Cicoides spp.
C. caeruleus
Z. lateralis
Ants
S. molesta
Solenopsis spp. And Monomorium
spp.
Beetles
T. corticalis
Carabidae and Cantharidae
C. lividus
Carabidae
Parasitoids
A. quadridentata

E. morio
P. vulnerator

90
95
52-90
47
90-95
24

Chandler
Solomon et al.
MacLellan
MacLellan
Glen and Milsom
Wearing

1928
1976
1958
1962
1978
1975

4 – 14
52-53

Jaynes and Marucci
Jaynes and Marucci

1947 West Virginia, USA
1947 West Virginia, USA

14
65-87
5-10
44
60

Jaynes and Marucci
Jaynes and Marucci
Jaynes and Marucci
Glen and Milsom
Riddick and Mills

1947
1947
1947
1978
1994

West Virginia, USA
West Virginia, USA
West Virginia, USA
Long Ashton, UK
California, USA

1-16
19
9
5-22
15
43
51
23
32

Jaynes and Marucci
Garlick
MacLellan
Ferro et al.
Subrinprasert
Mills
Subrinprasert
Subrinprasert
Mills

1947
1948
1962
1975
1987
2005
1987
1987
2005

West Virginia, USA
Ontario, Canada
Nova Scotia, Canada
Washington, USA
Sweden
Austria
Sweden
Sweden
Switzerland

24

Illinois, USA
Long Ashton, UK
Nova Scotia, Canada
Nova Scotia, Canada
Long Ashton, UK
New Zealand

Table 1.1. (cont’d).
Mortality Agents:

Mortality (%)

Source

Year

Geographic region

th

5 Instar (cont’d)
M. ridibundus
B. rufipes
Unspecified predators

44
56
33
94
30
54 - 99
57
15 - 87

Pathogens
B. bassiana

P. farinosus
Adult Moth
No data

9
2
7
6

None

Mills
Mills
Mills

2005
2005
2005

Kazakstan
California, USA
Kazakstan

MacLellan
Ferro et al.
Baumgartner
Wearing
Mathews et al.

1960
1975
1999
1975
2004

Nova Scotia, Canada
Washington, USA
California, USA
New Zealand
West Virginia, USA

Jaynes and Marucci
Jaques and MacLellan
Subrinprasert
Subrinprasert

1947
1965
1987
1987

West Virginia, USA
Nova Scotia, Canada
Sweden
Sweden

None

25

Unfortunately, quantitative estimates of the role of codling moth natural enemies
are either lacking completely or from research performed 25+ years ago (Table 1.1) in
orchards that are not representative of modern production practices. Modern orchards
are trending towards higher tree densities (>1000 trees per acre) (Jackson et al., 1981;
Wagenmakers, 1991) and more selective pest management practices (Blommers, 1992;
Jones et al., 2009). High-density orchards have a continuous within row tree canopy,
which may affect natural enemy foraging efficiency (Kareiva and Perry, 1989; Mills,
2003). Selective pest management tactics, such as mating disruption for codling moth
(Witzgall et al., 2008) greatly improve the activity of natural enemies in the orchard
environment (Epstein et al., 2000; Epstein et al., 2001), which is likely to have a
significant impact on their contribution to pest management.
Conservation biological control aims to suppress crop pests by increasing the
density or effectiveness of their natural enemies (Barbosa, 1998). In practice, natural
enemy groups respond differently to conservation tactics, and conservation strategies
must be tailored to fit the target pest and cropping system (Landis et al., 2000; Thomas,
1999; Thomas et al., 1991; Unruh et al., 2012; Walton, 2009). The first step in this
process is to identify the natural enemies that have the greatest impact on the pest
population (Mills, 2005b). This can be accomplished by building a stage structured
matrix population model (Leslie, 1945) to identify the pest life stage that most
contributes to population growth of the species (Caswell, 2001).
This approach can be very useful for planning biological control (Mills, 2005b)
and can be used to inform pest management decision-making (Jones et al., 2009).
However, analyses of this kind are dependent on the accuracy of the mortality estimates

26

used to parameterize the model (Caswell, 2001). Extant estimates of the stage specific
mortality of codling moth are highly variable or lacking entirely (Table 1) and depend on
the geographic region, orchard management, and climate. Determining the optimal
targets of conservation biological control of codling moth is likely to be a very regional
and farming system specific process.
Future studies can contribute to present knowledge substantially by using digital
video (Grieshop et al., 2012) or molecular techniques to identify mortality agents (de
Roince et al., 2012) and the factors affecting their impact on codling moth populations
more directly. Studies are also needed that identify the local factors (e.g. farming
practices) as well as landscape-level factors that drive natural enemy impacts on
codling moth.
An abundance of research has been performed concerning the non-target effects
of pest management practices and the impact that naturally occurring insects have on
pest control (Epstein et al., 2001; Hoyt, 1969; Kogan and Lattin, 1993; Ohnesorg et al.,
2009; Van Driesche et al., 1998). However, there is still a lack of data directly linking
natural enemy conservation practices to improvements in pest management in many
agricultural systems (Letourneau and Bothwell, 2008; Wilby and Thomas, 2002).
Understanding the major drivers of pest population dynamics in agroecosystems is
st

essential if we are to develop a sustainable food supply for the 21 century (Thomas,
1999).

27

Chapter 2. Video observations of the natural enemies of eggs and fifth instar
codling moth Cydia pomonella (L.) in Michigan apple orchards (USA)
2.1. Introduction
The conservation of natural enemies has long been fundamental to the theory
and practice of Integrated Pest Management (IPM) (Jones et al., 2009; Stern et al.,
1959; Vandenbosch and Stern, 1962). Numerous studies have shown that broadspectrum insecticides can be harmful to natural enemies and thereby cause secondary
pest outbreaks. (Epstein et al., 2001; Hoyt, 1969; Kogan and Lattin, 1993;
Michelbacher, 1952; Ohnesorg et al., 2009; Van Driesche et al., 1998). Apple orchard
agroecosystems have been a proving ground for IPM and the successful application of
its guiding principles since its inception (Holdsworth Jr, 1970; Hoyt, 1969; Jones et al.,
2009; Maclellan, 1972; Oatman, 1966). Yet, even in these systems very little is known
about the actual contribution of natural enemies to the management of pest species
(van der Geest and Evenhuis, 1991). For example, management of the codling moth in
apples still relies largely on suppression of the pest through the use of synthetic or
natural toxins that have a negative impact on natural enemies (Brunner et al., 2005;
Epstein, 2010; Jones et al., 2009). Fortunately, recent successes in the use of selective
insecticides and biopesticides (including mating disruption) have greatly increased the
suitability of some apple orchards as a habitat for natural enemies (Epstein, 2010).
Specifically, increased use of mating disruption for codling moth management has
reduced the use of broad-spectrum insecticides in apple orchards (Jones et al., 2009;
Witzgall et al., 2008). This has provided an environment where it is possible to
investigate the impact of natural enemies on this pest in a setting where their potential is

28

maximized (Epstein, 2010; Epstein et al., 2001; Gut and Brunner, 1998; Jones et al.,
2009; Witzgall et al., 2008).
2.1.1. Natural enemies in organic agriculture
Organic pest management is IPM characterized by an even heavier reliance on
cultural and biological pest management than conventional IPM. The pest management
products available to certified organic growers in the United States include:
pheromones, horticultural oils, kaolin clay, pyrethrums, spinosads, and bioinsecticides.
Characteristics shared by most of these products are that they degrade quickly in the
field, and or are very target specific, which reduces their impact on natural enemies
(Epstein et al., 2001; Jenkins and Isaacs, 2007). Recent research suggests that organic
orchards have increased biological control from natural enemies compared to
conventional orchards (Crowder et al., 2010; Minarro et al., 2009), but this cannot be
attributed to a reduction in synthetic insecticide use alone. Organically managed
orchards generally have a low disturbance rate, high plant diversity, and are more
structurally diverse than conventionally managed counterparts (Drinkwater et al., 1995).
Insect communities in conventionally managed agroecosystems are dominated by the
relatively few species that are adapted to broad-spectrum pest management strategies
and low-diversity habitats (Crowder et al., 2010; Epstein et al., 2000; Epstein et al.,
2001; Minarro et al., 2009).
In agricultural fields where broad spectrum, contact active insecticides are used
frequently, arthropod communities should be limited to those species that are either
good dispersers (e.g. large, strong fliers), are not active at the time insecticides are
being applied (e.g. sheltered and/or in a dormant state), or that can reproduce quickly

29

(Brown, 1993). It remains to be seen whether this is the type of natural enemy
community is less effective at suppressing pest populations than a more diverse
community (Murdoch, 1975; Straub et al., 2008; van Emden and Williams, 1974).
2.1.2. Evaluating the impacts of orchard management on natural enemies of
codling moth
There are a variety of methods available to researchers interested in measuring
natural enemy populations and their impacts on pests in agricultural settings (Luck et
al., 1988). I chose a combination of video observation of sentinel prey and relative
measures of natural enemy activity (yellow sticky traps and pitfall traps). I focused on
two sessile stages of the codling moth, the egg and cocoon stage. These stages are
important targets for pest management from both a theoretical and practical standpoint
(Geier, 1963; Jones et al., 2009; Kaya et al., 1984; Mills, 2005b).
Codling moth eggs are typically laid singly on foliage or fruit throughout the
canopy of the tree and usually within 20 cm of a developing fruit (Geier, 1963; Stoeckli
et al., 2008). Codling moth is an internal fruit feeder, which makes the eggs an
important target for pest management aimed at preventing damage to the crop by the
first instars entering the fruit (Brunner et al., 2005; Mota-Sanchez et al., 2008). Also,
codling moth larvae are relatively well protected from natural enemies and insecticides
once they gain entry to the fruit (Ferro et al., 1975; MacLellan, 1962), so insecticide
applications targeting first-generation eggs can prevent damage by the second
generation of larvae attacking the crop late in the season (Jones et al., 2009).
Fifth instar codling moth larvae leave the protection of the fruit to spin cocoons in
protected locations either on the trunk of the tree or on the orchard floor (Chandler,

30

1928; Garlick, 1948; Garlick and Boyce, 1940; Geier, 1963). Once they have spun their
cocoons they are relatively stationary as they pupate mid-summer (first generation) or
overwinter as mature larvae (second generation) (Geier, 1963). During this time they
are subject to attack by a variety of actively searching natural enemies (Walton, 2013a).
This stage is also a good target for cultural pest management practices (Judd et al.,
1997; Steiner and Ackerman, 1936), or augmentative (Lacey et al., 2006a; Mills, 2003)
and classical (Mills, 2005a) biological control.
I set out to compare organic and conventionally managed apple orchards in
Michigan (USA) in terms of natural enemy identity, but also in terms of their effect on
codling moth sentinel prey. My principal goal was to test the hypothesis that natural
enemies attack more codling moth prey in organic than conventional orchards. I tested
this hypothesis by comparing attack rates of sentinel eggs and fifth instar larvae on
video in three pairs of organic and conventionally managed orchards in southern
Michigan.
2.2. Materials and Methods
2.2.1 Videography
Video systems used for sentinel prey observations consisted of 4-channel digital
video recorders (DVRs) powered by deep-cycle12 v lead-acid batteries. I used digital
cameras equipped with infrared spectrum LEDs to capture day and night insect activity.
Cameras recorded continuously for 6 d (144 h), and batteries were replaced every 48 h.
Video data was recorded onto removable 132 GB hard drives and transferred to 2 TB
external hard drives in the laboratory. Cameras were fitted with 52 mm +10 or +4 macro

31

lenses for egg and fifth instar predation video, respectively. A detailed description of the
video system and equipment used can be found in Grieshop et al. (2012).
2.2.2 Experimental design
This study was carried out at three farms, two commercial apple orchards (42°
38.089'N 84° 47.337'W and 43° 1.012'N 83° 54.890'W ) and one research station (42°
52.563'N 85° 14.900'W). At each farm, a pair of or chards was selected that were
similar ages and tree densities, but differed in management regime (organic vs.
conventional). All paired orchards were separated by at least 500 m and were at least 1
ha in size. In each orchard, a camera system with four cameras was placed in a tree
row. Due to limitations imposed by the need to power all cameras from one centrally
located battery and the need for farmers to be able to drive tractors between rows, all
four cameras were located in the same tree row. Cameras were placed in separate
trees with at least one tree separating them from the next camera. In general, trees with
cameras in them were separated by 10 m. Each camera was focused on one sentinel
prey patch. Prey patches consisted of either codling moth eggs on synthetic leaves or
fifth instars in cardboard bands.
2.2.3. 2010 Experiments
2.2.3.1. Sentinel Prey
Egg patches contained 10 UV sterilized eggs that were left in place for the entire
six days that video was recorded. Sentinel eggs were directly oviposited onto synthetic
leaves in rearing cages in the lab. Individual gravid females were caged with synthetic
leaves for one night (~12 h). Leaves with eggs were placed under a fluorescent UV-C
bulb for 5 minutes to prevent hatching in the field (Calderon and Navarro, 1971). Trials

32

in the lab found that this procedure reduced egg hatch rate to nearly zero (N.J. Walton,
unpublished data).
Fifth instar patches consisted of three codling moth fifth instars in cardboard
flutes. Sentinel fifth instar larvae were placed individually in plastic containers with flutes
of corrugated cardboard in the lab for one night (~12 h). Flutes in which larvae had spun
cocoons the following day were taken to the field and glued with cyanoacrylate glue to
strips of paper that were stapled to apple tree trunks.
2.2.3.2. Data collection
Video of egg predation was recorded in each orchard for six days during June to
coincide with peak egg laying for first generation codling moth. Video of fifth instar
patches was recorded for six days at each orchard in August or September to coincide
with the overwintering of second-generation codling moth larvae.
Unbaited yellow sticky traps were used to measure natural enemy activity in the
study orchard canopies immediately following the egg video recordings. Ten of these
traps were hung in the middle of the tree canopy, evenly spaced along two tree rows
adjacent to the row that was used for video observations. Yellow sticky traps were in the
orchards for seven days. Insect collected on yellow sticky traps were counted and
categorized by functional and taxonomic group using Borror et al. (1992) and Marshall
(2006).
Pitfall traps were used to measure natural enemy activity in the study orchards
immediately following the fifth instar video observations in 2010. Ten pitfall traps were
placed in each orchard evenly spaced along two rows adjacent to the row that was
being used for the video observations. Traps were deployed for 48 hours. Pitfall traps

33

were constructed from 1000 mL clear plastic deli containers and filled with a dilute
unscented dish soap and water solution. Insects were removed from the traps and
stored in 80% ethyl alcohol for storage and identification. Insects and other arthropods
collected from the pitfall traps were counted and categorized by functional and
taxonomic group using Borror et al. (Borror et al., 1992) and Marshall (Marshall, 2006).
2.2.4. 2011 Experiments
2.2.4.1. Sentinel Prey
In 2011, two egg predation experiments were conducted. The first was carried
out to coincide with first-generation codling moth in June, and the second was
conducted in August to coincide with codling moth second-generation oviposition. Firstgeneration egg patches consisted of 3 codling moth eggs on a synthetic leaf. For the
second-generation egg predation observations, codling moth eggs were not available,
so I used the eggs of a closely related species, the oriental fruit moth Grapholita
molesta Busck (Lepidoptera: Tortricidae).
Sentinel eggs were directly oviposited by female moths on green floral paper in
rearing cages, which was cut and glued to synthetic leaves with cyanoacrylate glue.
Eggs in 2011 were not treated with UV light to prevent hatching. Instead, these eggs
were removed from the field before hatching (36 h) and replaced with new eggs.
Sentinel fifth instar larvae were placed individually in plastic containers with flutes
of corrugated cardboard in the lab for one night (~12 h). Flutes in which larvae had spun
cocoons the following day were taken to the field and glued with cyanoacrylate glue to
strips of paper that were stapled to apple tree trunks.
2.2.4.2. Data collection

34

Video of egg predation was recorded for six days in June and for six days in
August to coincide with peak egg laying for first and second-generation codling moth,
respectively. Eggs were replaced every 36 h to prevent hatching in the field, so each
camera recorded two batches of eggs for 36 h each. Video of fifth instar patches was
recorded for six days at each orchard at the end of August and the beginning of
September to coincide with the overwintering of second-generation codling moth larvae.
Unbaited yellow sticky traps were used to measure natural enemy activity in the
study orchard canopies immediately following the egg video recordings. Twelve of these
traps were hung in the middle of the tree canopy, evenly spaced along two tree rows
adjacent to the row that was used for video observations. Yellow sticky traps were in the
orchards for seven days. Insect collected on yellow sticky traps were counted and
categorized by functional and taxonomic group using Borror et al. (1992) and Marshall
(2006).
2.2.5. Video observation
Video was watched in the laboratory by paid observers. Observers viewed four
video files at a time at up to 4x speed and recorded the start and end time for all video
events as well as a brief description of the event. A video event was defined as
beginning when an organism entered the sentinel prey patch and ending when it left the
prey patch. All interactions with sentinel prey were watched by the authors for
identification of the organisms and determination of whether or not predation had
occurred. Sentinel prey were also collected from the field and survivorship and evidence
of predation were recorded. If a determination could not be made as to whether an
interaction with a sentinel prey item on video was fatal, the prey was assumed to have

35

survived. Attacks on a prey item that had already been killed by another predator were
recorded but not counted as predation events.
2.2.6. Natural enemy behavior
Timing of predation events was quantified using the video files’ built in
timestamps. For egg video observations, handling time was defined as the time from
discovery of the prey item until feeding ceased and foraging behavior was resumed. In
larval video observations, feeding could not be observed directly so handing time was
defined as the time that the natural enemy spent inside of the cardboard flute. For
parasitoid wasps that did not enter the cardboard flute, handing time was defined as the
time spent probing the flute with the ovipositor. For all video observations, event
duration was defined as the total time that an individual natural enemy spent in the
sentinel prey patch. Ants from the same colony were counted as an individual natural
enemy.
For characterization of diel activity patterns, dusk was defined as 2 hours before
sunset. Dawn was defined as 2 hours after sunrise. Morning was the period from the
end of dawn until 12 noon, and afternoon was defined as the period from 12 noon until
dusk. Night began at sunset and ended at Dawn.
2.2.7. Data analysis
The response variable for video predation analysis of all experiments was the
proportion of sentinel prey attacked on video per tree per 6-day observation period.
Proportion values were normalized by arcsine square root transformation for statistical
analysis. Video data were analyzed by mixed effects ANOVA with farm in the model as
a random effect (block) and management regime as a fixed effect (nlme, R v2.9.2.). The

36

2011 fifth instar data did not meet the assumption of normality for ANOVA, so a
nonparametric Kruskal-Wallis rank sum test was used to compare attack rates between
organic and conventionally managed orchards (stats, R v.2.9.2).
Arthropod counts on yellow sticky cards and in pitfall traps were log10(x+1)
transformed in order to meet the normality assumption of ANOVA. These data were
also analyzed with a mixed effects model in R with farm as a random effect,
management as the fixed treatment effect, and transformed natural enemy counts as
the response variable (nlme, R v2.9.2.).
2.3. Results
2.3.1. 2010 First Generation Egg Predation
There was no effect of management on egg predation in 2010 (F=0.53, df=1,20,
P=0.47). The untransformed mean proportion eggs attacked on video (+/- SEM) was
0.24 (+/- 0.10) and 0.13 (+/- 0.06) in organic and conventional orchards, respectively
(Fig. 2.1).

37

Proportion Attacked (+/-SEM)

0.4
0.35

a

Harvestmen

0.3
Mites

0.25
0.2

a

Bugs

0.15
Spiders
0.1
Lacewings (larvae)

0.05
0
Conventional
Organic
Management

Figure 2.1. Mean proportion of sentinel codling moth eggs attacked by natural enemies
1.
on video from June 2010 in organic and conventionally managed orchards. Data shown
are untransformed. Statistical analysis was performed by mixed model ANOVA (R
v.2.9.2) using arcsine square root transformed proportions. Columns with the same
)
letter are not significantly different (
(α=0.05).
2.3.2. 2010 Fifth Instar Predation
3.2.
There were significantly more fifth instars attacked in video in organic orchards
than conventional orchards in 2010 (F=7.30, df = 1,20, P=0.014). The untransformed
mean proportion larvae attacked on video (+/ SEM) was 0.42 (+/- 0.12) and 0.08 (+/(+/(+/
0.06) in organic and conventional orchards, respectively (Fig. 2.2).

38

0.6

Proportion Attacked (+/-SEM)

a
0.5
0.4
Lacewings (larvae)

0.3

Spiders
Parasitoid Wasps

0.2

b

Ants

0.1
0
Conventional
Organic
Management

Figure 2.2. Mean proportion of sentinel codling moth fifth instars attacked by natural
2.
enemies on video from August and September 2010 in organic and conventionally
managed orchards. Data shown are untransformed. Statistical analysis was performed
shown
by mixed model ANOVA (R v.2.9.2) using arcsine square root transformed proportions.
.2.9.2)
Columns with the same letter are not significantly different (
(α=0.05).
2.3.3. 2011 First Generation Egg Predation
3.3.
There was no effect of management on egg predation in June 2011 (F=0.68,
ere
df=1,20, P=0.42). The untransformed mean proportion eggs attacked on video (+/
(+/SEM) was 0.08 (+/- 0.08) and 0.19 (+/ 0.11) in organic and conventional orchards,
(+/respectively (Fig. 2.3).

39

0.4

Proportion Attacked (+/-SEM)

0.35

a
0.3
0.25

Unidentified

0.2

a

Spiders

0.15
0.1
0.05
0
Conventional
Organic
Management

Figure 2.3. Mean proportion of sentinel codling moth eggs attacked by natural enemies
3.
on video from June 2011 in organic and conventionally managed orchards. Data shown
are untransformed. Statistical analysis was performed by mixed model ANOVA (R
v.2.9.2) using arcsine square root transformed proportions. Columns with the same
.2)
letter are not significantly different (
(α=0.05).
2.3.4. 2011 Second Generation Egg Predation
3.4.
There was no effect of management on egg predation in August 2011 (F=0.32,
df=1,20, P=0.58). The untransformed mean proportion eggs attacked on video (+/
0.58).
(+/SEM) was 0.03 (+/- 0.03) and 0.08 (+/ 0.08) in organic and conventional orchards,
(+/respectively (Fig. 2.4).

40

0.4

Proportion Attacked (+/-SEM)

0.35
0.3
Tree Cricket

0.25

Mouse
0.2

a

0.15
0.1

a
0.05
0
Conventional

Organic
Management

Figure 2.4. Mean proportion of sentinel oriental fruit moth eggs attacked by natural
4.
enemies on video from August 2011 in organic and conventionally managed orchards.
Data shown are untransformed. Statistical analysis was performed by mixed model
ANOVA (R v.2.9.2) using arcsine square root transformed proportions. Columns with
the same letter are not significantly different (
(α=0.05).
2.3.5. 2011 Fifth Instar Predation
3.5.
The effect of management on fifth instar predation in 2011 was not statistically
2

significant (Kruskal-Wallis, X =3.27, df=1, P=0.079). The untransformed mean
proportion larvae attacked on video (+/ SEM) was 0.11 (+/- 0.06) and 0.0 (+/- 0.0) in
(+/(+/
organic and conventional orchards, respectively (Fig. 2.5).

41

Proportion Attacked (+/-SEM)

0.6
0.5
Predatory Beetle Larvae

0.4

Ants
0.3

Predatory Beetles

a

0.2
0.1

a
0
Conventional
Organic
Management
Figure 2.5. Mean proportion of sentinel codling moth fifth instars attacked by natural
5.
moth
enemies on video from August and September 2011 in organic and conventionally
managed orchards. Data shown are untransformed. A Kruskal Wallis rank sum test was
Kruskal-Wallis
used to compare attack rates between organic and conventionally managed orchards
conventionally
(stats, R v.2.9.2). Columns with the same letter are not significantly different (
(α=0.05).
2.3.6. Yellow Sticky Traps
There were not significantly more natural enemies on yellow sticky traps in
organic than conventional orchards in 2010 (t=0.61, df=50, P=0.55) (Fig. 2.6).

42

Mean Natural Enemies (+/-SEM)

40

a

a

Lacewings (adult)
Predatory Wasps
Lady Beetles

30

Pirate Bugs
Syrphid Flies

20

Spiders
Predatory Beetles

10

Tachinid Flies
Parasitoid Wasps

0
Conventional
Organic
Management

Figure 2.6. Mean number of natural enemies on unbaited yellow sticky traps in organic
6.
and conventional orchards in June 2010. Data shown are untransformed. Statistical
analysis was performed by mixed model ANOVA (R v.2.9.2) using log10(x+1)
2.9.2)
transformed proportions. Columns with the same letter are not significantly different
(α=0.05).
In 2010, the untransformed mean number of natural enemies on yellow sticky
traps (+/- SEM) was 35.4 (+/- 5.3) and 34.9 (+/- 5.3) in organic and conventional
orchards, respectively. In 2011, there were significantly more natural enemies
(untransformed mean +/- SEM) in organic (98.4 +/ 20.2) than conventional orchards
+/(43.2 +/- 18.6) (t=3.16, df=68, P=0.002) (Fig. 2.7).

43

Mean Natural Enemies (+/-SEM)

140

a

120
100
80
60

b

40
20

Ants
Predatory Wasps
Lady Beetles
Lacewings
Fireflies
Spiders
Rove Beetles
Syrphid Flies
Pirate Bugs
Banded Thrips
Parasitoid Wasps

0
Conventional
Organic
Management
Figure 2.7. Mean number of natural enemies on unbaited yellow sticky traps in organic
ean
and conventional orchards in June 2011. Data shown are untransformed. Statistical
analysis was performed by mixed model ANOVA (R v 2.9.2) using log10(x+1)
v.2.9.2)
transformed proportions. Columns with the same letter are not significantly different
umns
(α=0.05).
2.3.7. Pitfall Traps
When all natural enemy groups were included in the analysis, there were
significantly more natural enemies in pitfall traps in organic than conventional orchards
in 2010 (t=3.17, df=67, P=0.002) (Fig. 2.8). The untransformed mean number of natural
10
enemies (+/-SEM) in pitfall traps in 2010 was 32.8 (4.9) and 17.2 (4.9) in organic and
SEM)
conventional orchards, respectively. No pitfall trap samples were taken in 2011.

44

40

a

Mean Natural Enemies (+/-SEM)

35
Lacewings (larvae)

30

Bugs
25

Centipedes

b

Carabid Beetles

20

Parasitoid Wasps
15

Harvestmen

10

Rove Beetles
Spiders

5

Ants

0
Conventional
Organic
Management
Figure 2.8. Mean number of natural enemies in pitfall traps in organic and conventional
8.
orchards in September 2010. Data shown are untransformed. Statistical analysis was
performed by mixed model ANOVA (R v2.9.2) using log10(x+1) transformed
proportions. Columns with the same letter are not significantly different (α=0.05).
ns
(α
2.3.8 Natural Enemy Behavior
3.8
The mean event duration (+/ SEM) for egg natural enemies was 16.7 m (+/-4.6).
(+/-SEM)
(+/
The mean handling time (+/-SEM) for egg natural enemies was 5.4 m (+/-0.9). The
SEM)
(+/
mean event duration for fifth instar natural enemies was 252.5 m (+/ 82.4). The mean
(+/-82.4).
handling time (+/-SEM) for fifth instar natural enemies was 644.4 m (+/-219.9
SEM)
219.9).
The mean event duration (+/ SEM) for egg natural enemies in organic and
(+/-SEM)
conventional orchards was 10.8 m (+/
l
(+/-3.7) and 25.3 m (+/-8.6), respectively. The mean
8.6),
handling time (+/-SEM) for egg natural enemies was 3.4 m (+/-0.5) and 7.8 m (+/-1.9) in
SEM)
organic and conventional orchards, respectively. The mean event duration for fifth instar

45

natural enemies was 286.0 m (+/-94.7) and 40.4 m (+/-33.9) in organic and conventional
orchards, respectively. The mean handling time (+/-SEM) for fifth instar natural enemies
was 739.7 m (+/-248.2) and 40.4 m (+/-33.9) in organic and conventional orchards,
respectively. Behavior data for individual natural enemy groups pooled by treatment and
year are presented in Tables 2.1 and 2.2.

46

Table 2.1. Summary of codling moth egg natural enemy activity in video observations compared with rank abundances
and mean abundances on yellow sticky traps (YSTs) pooled across years and management regimes.
Natural Enemy Group
Lacewings
Mites
Spiders
Bugs
Unidentified
Parasitoids
Harvestmen
Mice
Tree Crickets

Predation YST Rank YST Abundance Event duration Handling Time
1
Events Abundance
(±SEM)
(minutes±SEM) (seconds±SEM)
34
6
6
4
4
2
1
1
1

7
NA
5
NA
NA
1
NA
NA
NA

0.2±0.1
NA
1.3±0.2
2.1±0.3
NA
41.7±6.3
NA
NA
NA

12.7±4.8
14.7±3.7
60.2±32.7
3.8±0.3
17.4±10.5
7.5±0.0
4.5±0.0
0.1±0.0
0.7±0.0

1

250.3±143.1
492.7±332.5
702.5±553.8
105.00±35.7
657.0±1183.1
224.0±135.8
1.0±0.0
4.0±0.0
44.0±0.0

Active Period
Dusk/Dawn
Afternoon/Dusk
Afternoon/Night
Night
Afternoon/Night
Dusk
Night
Night
Night

Groups were ranked based on mean abundances in yellow sticky trap counts. Natural enemies that were not counted on

yellow sticky traps were not assigned a rank.

47

Table 2.2. Summary of codling moth fifth instar natural enemy activity in video observations compared with rank
abundances and mean abundances in pitfall traps pooled across management regimes.
Predation Pitfall Rank Pitfall Abundance Event duration Handling Time
1
Events Abundance
(±SEM)
(minutes±SEM) (minutes±SEM)

Natural Enemy Group
Ants
Parasitoids
Lacewings
Beetles
Beetle larvae
Spiders

8
7
3
2
1
1

1
4
9
5
NA
2

11.2±3.3
1.4±0.3
0.03±0.02
1.1±0.2
NA
7.5±1.4

574.3±178.4
4.2±1.4
40.4±33.9
30.0±0.0
360.0±0.0
390.0±0.0

1

1654.3±407.2
1.5±0.7
40.4±33.9
30.0±0.0
360.0±0.0
390.0±0.0

Active Period
Afternoon/Night
Afternoon
Afternoon/Night
Night
Night
Night

Groups were ranked based on mean abundances in pitfall trap counts. Natural enemies that were not counted in pitfall

trap samples were not assigned a rank.

48

2.4. Discussion
Natural enemies in general, and in particular natural enemies of fifth instar
codling moth tended to be numerically more abundant in the organic than the
conventional apple orchards in this study (Figs. 2.1, 2.2, 2.5, 2.7, and 2.8). In the case
of fifth instar codling moth, this trend was also reflected in the attack rates of sentinel
prey on video (Figs. 2.2, 2.5, and 2.8). However, sentinel egg mortality patterns in 2011
did not follow the same trend (Figs. 2.3 and 2.4).
There is relatively little correspondence between the abundance of natural
enemies on the yellow sticky card traps in both years and the abundance of codling
moth egg enemies (Table 2.1; Figs. 2.1, 2.3, 2.4, 2.6, and 2.7). This illustrates the
importance of direct measures of insect mortality in the field to verify conclusions based
on relative measures of activity density of natural enemies in the field. Spiders and
lacewings were the only groups that were detected by both sampling methods and both
groups were poorly represented in the sticky trap samples (Table 2.1). Yellow sticky
traps baited with herbivore-induced plant volatiles (Jones et al., 2011) or beat samples
of foliage (Southwood, 1978) may be a more reliable tool for monitoring codling moth
egg natural enemies in orchards.
Pitfall trap catches exhibited a closer correspondence to mortality of codling moth
fifth instars on tree trunks. While the relative abundances of different natural enemy
groups in pitfall traps did not reflect their importance as codling moth fifth instar
predators (Table 2.2; Figs. 2.2, 2.5, and 2.8), the groups that were observed attacking
codling moth were found in the pitfall traps in appreciable numbers (Table 2.2; Fig. 2.8).
This provides some evidence that pitfall trap catches may continue to be a useful tool

49

for estimating codling moth natural enemy pressure in orchard habitats (Epstein et al.,
2000; Epstein et al., 2001)
2.5. Conclusion
As broad spectrum insecticides are phased out of apple production (Brunner et
al., 2005; Mota-Sanchez et al., 2008) arthropod natural enemies are likely to play an
increasing role in apple pest management (Epstein et al., 2000; Epstein et al., 2001;
Jones et al., 2009). Quantification of this role has also become increasingly important
both for pest management (Zhang and Swinton, 2009) and environmental policy
decision-making (Boyd and Banzhaf, 2007; Isaacs et al., 2009; Losey and Vaughan,
2006; Luck et al., 2009). Unfortunately, previous quantitative estimates of this role in
apple production systems were outdated or lacking entirely (Walton, 2013a). I have
provided quantitative estimates of the mortality of two important codling moth life stages
and identified the natural enemies responsible for that mortality in a small sample of
southern Michigan apple orchards. The natural enemy community and their impact on
codling moth varies substantially with geographic region (Walton, 2013a) and farm
management practices (Walton, 2013a; Walton, 2013c), so more studies will be needed
to adequately characterize the codling moth natural enemy complex in modern apple
production systems.
I advocate the use of direct observation with digital video equipment to measure
the important ecosystem services provide by beneficial arthropods in agricultural
systems. Direct sampling of natural enemy activity with digital video systems has many
advantages over other sampling methods (Frank et al., 2007; Grieshop et al., 2012;
Mathews et al., 2004; Merfield et al., 2004). Of course, there are experimental

50

limitations imposed by cost of equipment, labor needed to watch the video, and
logistical considerations (see Methods above), which make it unlikely that video will
completely replace other sampling methods on a large-scale in the near future. Still,
video systems can be used intensively (as in this study) to calibrate or verify relative
measures of arthropod activity, or they can be used more extensively to sample natural
enemy activity over larger geographic areas (Grieshop et al., 2012). Reliable
quantitative estimates of the impact that natural enemies have on agricultural pests can
provide cost-savings to food producers and increase the sustainability of food
production systems worldwide.

51

Chapter 3. Habitat characteristics influencing foraging success of codling moth
antagonists in apple orchards
3.1. Introduction
The codling moth (Cydia pomonella [L].) is a severe pest anywhere that apples or
pears are grown worldwide (Jones et al., 2009; van der Geest and Evenhuis, 1991).
The two stages of the codling moth life-cycle that are most vulnerable to management
are the egg stage and the mature larva (fifth instar) stage (Jaynes and Marucci, 1947;
Mills, 2005b). Conventional pest management of codling moth in the U.S. generally
relies on the use of broad-spectrum contact insecticides that are harmful to the insect
natural enemies of pests (Epstein et al., 2000; Epstein et al., 2001; Giolo et al., 2009;
Jones et al., 2009). Codling moth natural enemies tend to be more abundant in
orchards utilizing natural enemy conservation practices in an integrated approach
usually due to the use of more selective pesticide chemistries or application methods
(Epstein et al., 2000; Epstein et al., 2001). As U.S. apple growers are forced to phase
out their use of broad-spectrum Organophosphate insecticides under US Environmental
Protection Agency (EPA) regulations, they may expect a substantial increase in the
contribution of these generalist natural enemies to pest management in their orchards
(Jones et al., 2009). However, the impact that these insects have on pest populations
are likely influenced by factors other than the selectivity of a grower’s pest management
program (Norris and Kogan, 2005). I investigated the influence of two of these factors,
orchard floor structural complexity and tree canopy connectivity on the mortality of
codling moth in the field.
3.1.1. Orchard floor complexity and fifth instar codling moth mortality

52

Growers often manage the orchard floor beneath the tree canopy to optimize
nutrient availability to the tree and minimize pest and disease pressure (Atucha et al.,
2011; Mathews et al., 2004). Fifth-instar codling moth larvae leave the fruit and crawl
from the tree canopy to the main trunk or orchard floor to seek a location spin a cocoon
and pupate (mid-summer) or overwinter (fall) (Geier, 1963). In the past, the importance
of the orchard floor as a location for codling moth to seek shelter for pupation and
overwintering was minimal, because they are known to prefer locations under the rough
bark on the trunks of traditional varieties of apple (Geier, 1963). However, with the
growing acreage of dwarf apple varieties with smooth barked trunks, mature codling
moth larvae are more likely to be forced to seek shelter on the orchard floor (Lacey et
al., 2006a; Simpson, 1903).
Evidence from several studies of larval codling moth mortality under field
conditions indicates that their survival is relatively low when they construct their cocoons
on the orchard floor (Chandler, 1928; Geier, 1963; Jaynes and Marucci, 1947;
MacLellan, 1960; Steiner, 1929; Steiner and Ackerman, 1936). However, these studies
were largely observational and or for other reasons do not provide unequivocal tests of
the hypothesis in question (Walton, 2013a). Furthermore, they were conducted in
traditional low-density orchards planted with rough-barked tree cultivars, so the results
may not be relevant to modern high-density orchard systems.
Known predators of fifth-instar larvae, pupae, and/or overwintering cocoons
include: ground beetles, ants, birds, and lacewing larvae (MacLellan, 1958; Mills and
Carl, 1991). Another important natural enemy of overwintering codling moth are
entomopathogenic nematodes. Entomopathogenic nematodes are naturally occurring

53

insect pathogens that are being investigated for use in augmentative biological control
of codling moth (Kaya et al., 1984; Lacey et al., 2006a; Lacey and Unruh, 2005). For
small crawling organisms such as these, with random search behavior, an orchard floor
with high complexity (i.e. vegetation or litter) has a larger search area compared to an
orchard floor with less structural complexity (Kostylev et al., 2005; Randlkofer et al.,
2010). Therefore, the probability of an immobile prey item like a codling moth cocoon
being found by a natural enemy should vary inversely with structural complexity
(Holling, 1959). Orchard floor sanitation practices that reduce complexity of the orchard
floor habitat thus should have a negative impact on codling moth in the presence of
these natural enemies. Alternatively, structural complex orchard floors could provide
alternative prey and a favorable microclimate for natural enemies (Mathews et al.,
2004).
I investigated the effects of reducing complexity of the orchard floor habitat on
the foraging success of codling moth natural enemies in a series of three field cage
experiments. Each experiment was designed to test the hypothesis that reduced
orchard floor complexity will increase the mortality of sentinel codling moth pupae or fifth
instars in the presence of natural enemies. A different natural enemy was used for each
experiment. The natural enemies used in the orchard floor experiments were: lacewing
larvae (Chrysoperla rufilabris, Neuroptera: Chrysopidae), adult earwigs (Forficula
auricularia, Dermaptera: Forficulidae), and the entomopathogenic nematode
Steinernema feltiae (Rhabditida: Steinernematidae). I chose these organisms because
they are all active searchers that have been implicated as important natural enemies of

54

codling moth fifth instars (Epstein et al., 2000; Kaya et al., 1984) (Walton Chapter 2, this
volume).
3.1.2. Canopy connectivity and mortality of codling moth eggs
The structure of the tree canopy is a characteristic of apple orchards that is
changing with modern production practices (Jackson et al., 1981; Wagenmakers, 1991).
In the increasingly dominant high-density orchard, dwarf apple varieties are planted at
close proximity (1 – 1.5 m spacing in some orchards) and pruned to grow in a linear,
“fruiting wall” along supporting trellis wire (Wagenmakers, 1991). This creates, from the
perspective of small crawling insects, a continuous habitat much like a hedgerow. In
contrast, under low-density apple systems, tree canopies are often distinct patches
surrounded by the less favorable habitat of the orchard floor. This difference in spatial
arrangement of trees is most likely to affect the movement of small, flightless arthropods
with primarily random search behavior such as Chrysoperla spp. and other generalist
natural enemies (Grieshop et al., 2008; Heinz, 1998; Kareiva and Perry, 1989;
Suverkropp et al., 2010). These small crawling predators should have reduced dispersal
in low-density apple plantings, which would prevent them from preying on as many prey
as they would in a high-density planting (Huffaker, 1958; Kareiva and Perry, 1989).
Codling moth eggs can be found throughout the canopy of apple trees, but they
are laid primarily on the upper side of the leaf in the vicinity (<20 cm) of developing fruit
(Geier, 1963; Stoeckli et al., 2008). Known predators of codling moth eggs include
thrips, mites, predatory bugs, and lacewing larvae (Mills and Carl, 1991).
I designed an experiment to test the hypothesis that increased canopy
connectivity would increase the mortality of sentinel codling moth eggs compared to

55

control plots in the presence of augmented numbers of flightless natural enemies.
Specifically, I tested this hypothesis by releasing lacewing larvae (Chrysoperla rufilabris,
Neuroptera: Chrysopidae) in experimental plots in an orchard where tree canopies
overlap within tree rows (high connectivity) and an orchard where tree canopies do not
overlap (low connectivity) and measuring predation of sentinel codling moth eggs.
Green lacewing larvae were chosen for this experiment because they are available
commercially in large quantities, and they are flightless active searching predators that
are known to prey on codling moth eggs (Mills and Carl, 1991) (Walton Chapter 2 this
volume).
3.2. Methods
3.2.1. Study Sites
Each experiment was conducted at one of two sites: Michigan State University’s
Clarksville Horticultural Experiment Station (42° 5 2.563'N, 85° 14.900'W) and AlMar
orchards (43° 1.012'N, 83° 54.890'W), a privately o wned certified organic farm. No
insecticides were used while the experiments were running in either location.
3.2.2. Orchard floor complexity experiments
3.2.3. Experimental Design
For each complexity experiment, twelve rows of trees in a high-density orchard
were selected to serve as blocks in a Randomized Complete Block Design. In each row,
6, 1 m by 0.3 m, field cages were placed at equal distances along the row. Each
experiment had 3 levels of natural enemy treatment (described below) and 2 levels of
complexity (described below) for a total of 6 treatment combinations. Within a block, one
of these experimental treatment combinations was randomly assigned to each cage

56

using a random number generator created in Microsoft Excel 14.1.0 (Microsoft
Corporation 2011).
3.2.4. Natural Enemy Treatments
The three levels of natural enemy manipulation in each experiment were: Control
(endemic natural enemies not removed, no natural enemies added), Natural Enemies
Removed, and Natural Enemies Added (endemic natural enemies removed). Endemic
natural enemies were removed from the cages using a modified leaf vacuum as
described in Epstein et al. (Epstein et al., 2011).
3.2.5. Complexity Treatments
Orchard floor complexity was manipulated by either, using a string trimmer and
leaf rake to remove vegetation and litter from beneath the trees in the orchard plots, or
leaving the vegetation and litter in place. This produced two complexity levels, Low
(vegetation removed) and High (vegetation undisturbed).
3.2.6. Cages
Cages were constructed in the field around 1 m by 0.3 m plots by driving a 1 m
wooden stake into the corner of each plot and stapling a 1 m by 3 m sheet of 3 mil
transparent plastic to each stake to make side walls. The bottom 3 cm of each plastic
side wall was then buried and a 1 m by 0.3 m sheet of white nylon tulle attached with
binder clips was used to cover the top of the cage.
3.2.7. Sentinel Prey
Codling moth sentinel prey were presented within each cage in small wooden
shelters to simulate the protected locations where codling moth cocoons are typically
found (Chandler, 1928; Steiner, 1929). Codling moth pupae and diapausing fifth instar

57

larvae were obtained from the USDA-ARS Yakima Agricultural Research Laboratory
(Wapato, WA, USA). Wooden shelters were created by cutting aged (>1 year) 4 cm
diameter apple cuttings into 5 cm sections. A 0.5 cm diameter hole was drilled 2.5 cm
down into the center of one of the cut ends of each section. Then, each section was
split down the center with a chisel to create two shelters. The interior (split) surface of
each shelter was covered with cotton gauze (affixed with hot melt glue), leaving a cavity
formed by the drill-hole where the codling moth sentinel prey were inserted (pupae) or
allowed to build a cocoon (fifth instars).
One day prior to each experiment, fifth instar larvae were removed from cold
storage and placed individually in 500 mL polypropylene deli containers with one
wooden shelter in each. On the day of the experiment shelters were inspected and
those in which larvae had entered the cavity and constructed a cocoon were taken to
the field to be used as sentinel prey. Four such shelters with sentinel prey in them were
placed in each cage on the first day of the experiment. After 72 h in the field, shelters
were collected and returned to the laboratory for mortality assessment. Any insects that
were found in the shelters other than the sentinel prey were collected and preserved for
identification. Beetle larvae that were found during the earwig experiment at Clarksville
were treated with KAAD, preserved in 80% ethyl alcohol and identified using the
taxonomic keys provided in Stehr (Stehr, 1991).
3.2.8. Experiment 1: Entomopathogenic nematodes
This experiment was conducted at AlMar orchards in August 2012 and repeated
at Clarksville Horticultural Experiment Station in September 2012. At both locations, four
codling moth fifth instars in wooden shelters were used as sentinel prey in the natural

58

enemy treatment cages. In the experiment at Clarksville (September 2012), codling
moth pupae were used as sentinel prey in the Control and Natural Enemies Removed
cages only.
The natural enemy used in this experiment was the entomopathogenic nematode
Steinernema feltiae (Rhabditida: Steinernematidae). After sentinel prey were introcuced
to cages, nematodes were applied in water using a hand-held pressurized sprayer at a
rate of 180,000 infective juveniles (IJs) in 250 mL of water per cage. Following initial
application, an additional 125 mL of water was sprayed into the cages to improve
survival of the IJs (Lacey et al., 2006a). No water was applied to Control or Natural
Enemies Removed cages, but previous experiments have showed no detectable effect
of water alone on mortality of sentinel codling moth larvae (N.J. Walton unpublished
data).
3.2.9. Experiment 2: Lacewing larvae
This experiment was conducted at AlMar orchards in August 2012. The sentinel
prey in this experiment were diapausing fifth instar codling moth. Wooden shelters with
sentinel larvae in them were prepared as described above. The natural enemies used in
this experiment were larvae of the green lacewing C. rufilabris. Second and 3rd instar C.
rufilabris were supplied by Rincon Vitova Insectaries Inc., removed individually from the
shipping substrate, and placed in microcentrifuge tubes for transport to the field. In the
field, the microcentrifuge tubes were opened and placed in the center of the
experimental cages to release the lacewing larvae. In cages that had the natural enemy
treatment assigned to them, four lacewing larvae were released.
3.2.10. Experiment 3: Earwigs

59

This experiment was conducted in a 1 ha organic apple block at Clarksville
Horticultural Experiment Station in September 2012. The natural enemy chosen for this
experiment was the European earwig, F. auricularia. Earwigs were obtained from the
Pesticide Alternatives Laboratory at Michigan State University and were wild collected in
corrugated cardboard traps (Epstein et al., 2000) from an organic apple orchard. Prior to
being used in this experiment they were held in 40 x 25 x 14 cm plastic containers and
maintained on a diet of apples, dog food, bee pollen and water ad libitum with a 16:8
L:D photoperiod at 45% RH and 25 °C. Earwigs were starved for 24 h prior to the start
of the experiment, sexed, and placed individually in individual microcentrifuge tubes. In
cages that had the natural enemy treatment assigned to them, four earwigs (2 male, 2
female) were released into the center of each cage.
3.2.11. Experiment 4: Canopy connectivity
This experiment was conducted at AlMar orchards in June 2012. Tree
architecture could not be manipulated directly without removing economically significant
amounts of crop plant material, so experimental plots were set up within orchards with
different pre-existing canopy structures. After consultation with the grower, I selected
two orchards of similar size but different tree spacings. Trees in the high-connectivity
orchard were planted with a 2.2 m in-row tree spacing (~400 trees per acre) whereas
trees in the low-connectivity were planted at a 5 m in-row tree spacing (~150 trees per
acre). The two tree densities have radically different tree architecture resulting in higher
canopy connectivity for the higher density orchard. I estimated the volume of the
orchard canopies in the two orchards using the tree-row-volume method developed by
Manktelow and Pratt (Manktelow and Pratt, 1997). I used these canopy volume

60

estimates to standardize the volume of canopy that was contained with each of my
3

experimental plots, so that each plot contained approximately 200 m canopy. This
translated to 3 trees per plot and 18 trees per plot in the low-density and high-density
orchards, respectively. Each orchard contained 10 plots per treatment arranged in a
factorial treatment design with plots separated by at least 9 m between tree rows or 30
m within tree rows.
Codling moth mortality in the experimental plots was measured with sentinel
codling moth eggs. Codling moth adults were reared from pupae obtained from the
USDA-ARS Yakima Agricultural Research Laboratory (Wapato, WA, USA). Eggs were
3

collected by holding male and female codling moth adults in a 27 m plastic screen
cage with a dilute (~5%) sugar solution provided on cotton rolls. Wax paper sheets were
hung from the ceiling of the cage to serve as an oviposition substrate for the female
moths. Wax paper sheets were collected daily for use in experiments. Pieces of this
paper containing individual ≤24 h old eggs were cut and glued to the heads of steel pins
with cyanoacrylate glue and placed in Styrofoam coolers for transport to the field. Egg
bearing pins were inserted through apple leaves on the morning of the first day of the
experiment. Fifteen sentinel eggs were placed in each plot. Sentinel eggs were spaced
uniformly throughout the canopy of the trees in each plot. Eggs were collected from the
field and returned to the lab for mortality assessment 72 h after they were placed in the
field.
The natural enemy chosen for this experiment was the green lacewing C.
rufilabris. First instar C. rufilabris were obtained from a commercial supplier (Rincon
Vitova Insectaries, Inc.) and released at a density of 1,000 larvae per plot. Lacewing

61

larvae were released in the canopy of the trees in the center of each plot by distributing
their shipping substrate into three Bio-Boxes (Biobest Canada Ltd.), which were then
hung in the tree in the center of the plot. The remaining experimental plots did not have
lacewing larvae released in them (control plots). Egg mortality was assessed in the
laboratory daily for one week following exposure to field conditions.
One day after all sentinel prey were removed from the orchards, the density of
lacewing larvae was evaluated in the plots by beating (Southwood, 1978). Three
samples were taken per plot by striking the apple foliage with a 1 m section of 3 cm
PVC pipe. One such sample was taken from either end of each plot and one sample
was taken from the center. Lacewing larvae were identified visually, counted, and
released.
3.2.12. Statistical analysis
Orchard floor experiments were analyzed by ANOVA in R (v.2.9.2) (R-coredevelopment-Team, 2009). Proportion sentinel prey mortality in each cage was arcsin
square root transformed prior to analysis. In the final model, row was included as a
blocking effect, natural enemy treatment, complexity treatment and their interaction
were included as fixed effects. Preplanned comparisons of simple effects were
performed using Tukey’s Honest Significant Difference (HSD) (TukeyHSD, R v.2.9.2).
The mean numbers of predatory beetle larvae in Low and High complexity cages
in the earwig experiment at Clarksville were compared using Pearson’s Chi-squared
test in R (chisq.test, R v.2.9.2). The relationship between abundance of predatory
beetles and sentinel codling moth mortality in the earwig experiment was analyzed with

62

simple linear regression (lm, R v.2.9.2) using the mean number of predatory beetle
larvae and mean proportion mortality per tree row (block).
Mortality of eggs in the canopy connectivity experiment was analyzed using the
Kaplan-Meier method (Kleinbaum and Klein, 2012; R-core-development-Team, 2009).
Eggs on pins that could not be recovered from the field or that had been parasitized by
Trichogramma spp. were analyzed as censored observations. Survival curves were
compared using Wilcoxon (Gehan) tests (R v.2.9.2, survival package).
The mean number of lacewings per plot in beat samples was log10(x+1)
transformed prior to analysis. These data were analyzed by ANOVA, with orchard
connectivity, lacewing treatment, and their interaction included in the model as fixed
effects (aov, R v.2.9.2).
3.3. Results
3.3.1. Entomopathogenic nematodes
The natural enemy treatment effect was statistically significant at both sites for
this experiment (AlMar: F = 13.0, df = 2,55, p<0.001; Clarksville: F = 7.59, df = 2, 55,
p=0.001). The complexity treatment, however was not statistically significant at either
site (AlMar: F = 1.60, df = 1,55, p=0.21; Clarksville: F = 0.02, df = 1, 55, p=0.87) and
neither was the interaction of natural enemy treatment and complexity (AlMar: F = 1.86,
df = 2,55, p=0.17; Clarksville: F = 0.49, df = 1, 55, p=0.61).
Preplanned pairwise comparisons using Tukey’s HSD revealed some important
differences between and among natural enemy treatments (Figs. 3.1 and 3.2).

63

Proportion Mortality (+/- SEM)

0.6

a
ab

0.5
0.4
0.3
0.2

b

b
b

b

0.1
0
Low
High
Low
High
Low
High
Complexity Complexity Complexity Complexity Complexity Complexity
Control

Control

Nematode Nematode

NEs
Removed

NEs
Removed

Figure 3.1. Mean proportion mortality of sentinel codling moth larvae in experimental
1.
cages at AlMar orchard in June of 2012. Control cages had no manipulation of their
natural enemy communities, Nematode plots had endemic natural enemies removed
and received an application of entomopathogenic nematodes. NEs removed plots had
endemic natural enemies removed. Means separations were performed on arcsin
were
square root transformed data. Tukey’s HSD was used to control the experimentwise
error rate at α=0.05 (R v.2.9.2). Columns with the same letter above them are not
2.9.2).
significantly different.

64

Proportion Mortality (+/- SEM)

0.6

a

0.5

ab

0.4
0.3
0.2

ab

ab

*ab

b

0.1
0

Low
High
Low
High
Low
High
Complexity Complexity Complexity Complexity Complexity Complexity
Control

Control

Nematode Nematode

NEs
Removed

NEs
Removed

Figure 3.2. Mean proportion mortality of sentinel codling moth in experimental cages at
2.
sentinel
Clarksville Horticultural Experiment Station in June of 2012. Control cages had no
manipulation of their natural enemy communities, Nematode plots had endemic natural
enemies removed and received an application of entomopathogenic nematodes. NEs
removed plots had endemic natural enemies removed. Means separations were
performed on arcsine square root transformed data. Tukey’s HSD was used to control
the experimentwise error rate at α=0.05 (R v.2.9.2). Columns with the same letter above
them are not significantly different.
*The comparison, Low Complexity Nematode – Low Complexity Control, was marginally
The
significant (Padj.= 0.065).
The untransformed mean mortality (+/ SEM) in all nematode treated cages was
(+/-SEM)
0.39 (+/- 0.06) at Clarksville, which was at least twice as high as all other natural enemy
treatments (Tukey’s HSD P(adj.)<0.01) (data not shown). At AlMar, the untransformed
mean mortality (+/-SEM) in the nematode cages was 0.40 (+/ 0.05), which was also at
SEM)
(+/-0.05),
least twice as high as all other natural enemy treatments at that site (Tukey’s HSD

65

P(adj.)<0.001) (data not shown). The untransformed mean (+/-SEM) mortality in natural
enemies removed cages at Clarksville was 0.14 (+/-0.05) and was not significantly
different from that of the control cages (0.15 +/- 0.04) regardless of complexity level
(Clarksville: Tukey’s HSD P(adj.)=0.95) (data not shown). The untransformed mean (+/SEM) mortality in all natural enemies removed cages at AlMar was 0.11 (+/-0.03) and
was not significantly different from that in the control cages (0.14 +/- 0.04) (Tukey’s HSD
P(adj.)=0.97) (data not shown).
At AlMar, the nematode treatment in the low complexity cages had higher
proportion mortality (0.48 +/- 0.07) than the low (0.10 +/- 0.05) and high (0.17 +/- 0.06)
complexity control cages (nematode/low - control/low: Tukey’s HSD P(adj.)=0.001;
nematode/low – control/high: Tukey’s HSD P(adj.)=0.016) (Fig. 3.1). The nematode low
complexity treatment at AlMar also had a significantly higher proportion mortality than
the low complexity (0.17 +/- 0.05) and high complexity (0.06 +/- 0.03) natural enemies
removed cages (nematode/low – enemies removed/low: Tukey’s HSD P(adj.)=0.026;
nematode/low – enemies removed/high: Tukey’s HSD P(adj.)<0.001) (Fig. 3.1), but not
than the nematode high complexity cages (0.33 +/- 0.07) (nematode/low –
nematode/high: Tukey’s HSD P(adj.)=0.65) (Fig. 3.1). Also at AlMar, the nematode
treatment in the high complexity cages had higher mortality than the predators removed
high complexity cages (nematode/high – predators removed/high: Tukey’s HSD
P(adj.)=0.026) (Fig. 3.1), but not than any other cage type (Fig. 3.1).
At Clarksville, there was significantly higher proportion mortality in the low
complexity nematode cages (0.42 +/- 0.12) than the high complexity predators removed
cages (0.10 +/- 0.03) (nematode/low – predators removed/high: Tukey’s HSD

66

P(adj.)=0.046) (Fig. 3.2). The difference between proportion mortality in the nematode
2).
low complexity cages and the control lo complexity cages (0.10 +/- 0.03) was also
low
marginally significant at Clarksville (nematode/low – control/low: Tukey’s HSD
P(adj.)=0.066) (Fig. 3.2).
3.3.2. Lacewing larvae
The natural enemy treatment effect (F=0.93, df=2,55, P=0.39), complexity
(F=0.04, df=1,55, P=0.85), or their interaction were not statistically significant in this
f=1,55,
experiment (F=1.48, df=2,55, P=0.24). Pairwise comparisons using Tukey’s HSD also
did not reveal any significant differences between any of the experimental cage types in
this experiment (Fig. 3.3). The mean untransformed proportion mortality (+/ SEM) in
3).
(+/lacewing treatment cages was 0.19 (+/
(+/-0.07) and 0.21 (+/- 0.06) in low complexity and

Proportion Mortality (+/- SEM)

high complexity cages, respectively.
0.6
0.5
0.4
0.3
0.2

a

a

a

a

a
a

0.1
0

Low
High
Low
High
Low
High
Complexity Complexity Complexity Complexity Complexity Complexity
Control

Control

Lacewing

Lacewing

NEs
removed

NEs
removed

Figure 3.3. Mean proportion mortality of sentinel codling moth larvae in experimental
3.
cages at AlMar orchard in June of 2012. Control cages had no manipulation of their
natural enemy communities, Nematode plots had endemic natural enemies removed
and received an application of entomopathogenic nematodes. NEs removed plots had
plication
67

endemic natural enemies removed. Means separations were performed on arcsine
square root transformed data. Tukey’s HSD was used to control the experimentwise
error rate at α=0.05 (R v.2.9.2). Columns with the same letter above them are not
2.9.2). Columns
significantly different.
3.3.3. Earwigs
The natural enemy treatment effect (F=0.30, df=2,55, P=0.75), complexity
(F=0.80, df=1,55, P=0.38), or their interaction were not statistically significant in this
experiment (F=1.06, df=2,55, P=35). Pairwise comparisons using Tukey’s HSD also did
not reveal any significant differences between any of the experimental cage types in this
experiment (Fig. 3.4). The mean untransform proportion mortality (+/- SEM) in earwig
4).
untransformed
treatment cages was 0.13 (+/-0.04) and 0.23 (+/- 0.07) in low complexity and high

Proportion Mortality (+/- SEM)

complexity cages, respectively.
0.6
0.5
0.4
0.3
0.2

a

a

a

a

a

a

0.1
0
Low
High
Low
High
Low
High
Complexity Complexity Complexity Complexity Complexity Complexity
Control

Control

Earwig

Earwig

NEs
removed

NEs
removed

Figure 3.4. Mean proportion mortality of sentinel codling moth pupae in experimental
4.
cages at AlMar orchard in June of 2012. Control cages had no manipulation of their
ar
natural enemy communities, Nematode plots had endemic natural enemies removed

68

and received an application of entomopathogenic nematodes. NEs removed plots had
endemic natural enemies removed. Means separations were performed on arcsine
square root transformed data. Tukey’s HSD was used to control the experimentwise
error rate at α=0.05 (R v.2.9.2). Columns with the same letter above them are not
significantly different.
The mean predatory beetle larva density (+/-SEM) was 0.01 (+/- 0.02) and 0.06
(+/- 0.02) in low complexity and high complexity cages, respectively. The Pearson’s Chisquared test comparing predatory beetle larval density between low and high
2

complexity cages indicated a significant difference between these two cage types (X
=4.63, df=1, p=0.031) (Fig. 3.5). There was also a significant correlation between

predatory beetle density and mortality of sentinel codling moth pupae in this experiment
2

(y=2.16x+0.14, P=0.002, adjusted R =0.58) (Fig. 3.6).

69

Mean Predatory Beetle Larvae (+/- SEM)

0.09

a

0.08
0.07
0.06
0.05
0.04

b
0.03
0.02
0.01
0
Low

High
Complexity

Figure 3.5. The mean number of predatory beetle larvae found preying on sentinel
5.
codling moth in the field cages at Clarksville in September 2012. Means separations
were performed using Pearson’s Chi
Chi-squared test (R v.2.9.2). Columns with the same
letter above them are not significantly different (α=0.05).

70

Proportion Codling Moth Mortality

0.4
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
0

0.02

0.04
0.06
0.08
Mean Predatory Beetle Larvae

0.1

Figure 3.6. Proportion codling moth mortality versus the mean number of predatory
beetle larvae in the cage experiments at Clarksville in September of 2012. Least
2

squares regression was performed in R v.2.9.2 (y=2.16x +0.14, P=0.002, adjusted R =
0.58).
3.3.4. Canopy connectivity
The mean proportion mortality (+/- SEM) of sentinel codling moth eggs was 0.28
(+/-0.02) and 0.27 (+/-0.03) in low connectivity and high connectivity plots, respectively
(data not shown). The mean proportion mortality (+/- SEM) of sentinel codling moth
eggs in high connectivity plots where lacewing were released was 0.30 (+/-0.04) and it
was 0.23 (+/-0.05) in high connectivity plots where they had not been released, but this
2

difference was not statistically significant (X =1.5, df=1, P=0.22) (Fig. 3.7).

71

Proportion Mortality* (+/-SEM)

0.4

a

0.35
0.3

a

a

a

0.25
0.2
0.15
0.1
0.05
0
Control

Lacewing
High

Control

Lacewing
Low

Figure 3.7. Proportion mortality of sentinel codling moth eggs in high and low
7.
connectivity orchards at AlMar in June 2012. Lacewing plots had ~1,000 first instar C.
0
rufilabris released in the tree canopy at the center of the plot. Control plots had no
added natural enemies. Mortality between treatments was compared using Wilcoxon
(Gehan) tests (survdiff, R v.2.9.2). Columns with the same letter above them are not
2.9.2).
significantly different (α=0.05).
=0.05).
*Does not include mortality from parasitism by Trichogramma spp.
There was also no statistically significant difference in mortality of eggs in high
2

compared to low connectivity plots where lacewing larvae were released (X =0.5, df=1,
were
P=0.67) (Fig. 3.7). There was no statistically significant difference in mortality of eggs
7).
2

between plots that had lacewings released compared to control plots (X =0.5, df=1,
P=0.46) (data not shown).

72

Analysis of log10(x+1) transformed Chrysopid counts in beat samples of trees at
the edges of the plots revealed significant effects of lacewing treatment (F=6.19,
df=1,36, P=0.017) and the interaction of lacewing treatment with orchard connectivity
(F=8.31, df=1,36, P=0.007), but no significant effect of orchard connectivity (F=0.07,
df=1,36, P=0.79). Pairwise comparisons using Tukey’s HSD revealed that there were
significantly more lacewing larvae (untransformed mean +/- SEM) in the edges of high
connectivity plots where lacewings had been released (0.55 +/- 0.17) compared to the
edges of high connectivity plots where they had not been released (0.00 +/- 0.00)
(Tukey’s HSD P(adj.)=0.003) (Fig. 3.8), but not significantly more in the edges of low
connectivity plots where they had been released (0.20 +/- 0.09) compared to the edges
of low connectivity plots where they had not been released (0.25 +/- 0.12) (Tukey’s HSD
P(adj.)=0.99) (Fig. 3.8). There was no significant difference between the number of
lacewings in the edges of plots in the high connectivity orchard where they had been
released compared to the low connectivity orchard where they had been released
(Tukey’s HSD P(adj.)=0.13) (Fig. 3.8).

73

Mean Chrysopid Larvae per plot (+/- SEM)

0.8

a

High Connectivity

0.7

Low Connectivity

0.6
0.5

ab

0.4

ab

0.3
0.2
0.1

b

0
Lacewing

Control

Figure 3.8. The mean number of Chrysopid larvae found in beat samples of the canopy
8.
at the edges of plots at AlMar orchards in June 2012. Lacewing plots had ~1,000 first
e
instar C. rufilabris released in the tree canopy at the center of the plot. Control plots had
no added natural enemies. Means separations were performed using Tukey’s HSD with
log10(x+1) transformed data. Columns with the same letter above them are not
significantly different (α=0.05).
=0.05).

74

3.4. Discussion
3.4.1. Orchard floor complexity
My results indicate some species of entomopathogenic nematodes may have
decreased foraging efficiency in complex habitats such as an orchard floor with
abundant vegetation (Figs. 3.1 and 3.2). This observation is consistent with foraging
theory and other studies demonstrating an inverse relationship between habitat
complexity and prey mortality (Anderson, 2001; Cuddington and Yodzis, 2002; Huffaker,
1958; Keitt and Johnson, 1995; MacArthur and Pianka, 1966). Entomopathogenic
nematode foraging behavior is relatively well studied in the laboratory (Lewis et al.,
2006), and S. feltiae has been used effectively against codling moth in the field (Kaya et
al., 1984; Lacey et al., 2006a). Steinernema feltiae exhibits a foraging mode that is
considered intermediate between an ambusher (sit-and-wait) and a cruiser (search-anddestroy) (Campbell and Gaugler, 1993; Shapiro-Ilan et al., 2012), which means that
infective juveniles of this species divide their time between displacement and bouts of
nictation or body-waving (Campbell and Gaugler, 1993). I suspect that the proximal
mechanism behind the reduced infection rate of codling moth in complex cages was
reduced foraging success of the S. feltiae IJs caused by decreased time spent on
displacement and more time spent head or body-waving (Campbell and Gaugler, 1993;
Walton, 2013b).
I was not able to demonstrate a similar effect of habitat complexity on the other
two species of codling moth natural enemies (Figs. 3.3 and 3.4). In the earwig and
lacewing experiments, codling moth mortality was consistently low, which indicates that
these natural enemies did not attack the sentinel prey in the experimental cages. This

75

could have been caused by an inability to locate the sentinel prey, altered foraging
behavior induced by the cage itself, or intraguild predation (incl. cannibalism).
Unfortunately, the data do not allow us to eliminate these alternative hypotheses. I also
did not detect an effect of endemic natural enemy removal in any of my experiments
(Figs. 3.1 – 3.4). This provides some evidence that my natural enemy removal method
did not eliminate all of the endemic natural enemies from the cages, which would have
reduced my ability to detect the effect of adding natural enemies to the cages and
increased the likelihood of intraguild predation. There were a variety of predatory
arthropods present in the cages before they were vacuumed (N.J. Walton, personal
observation) and if vacuuming failed to remove these enemies their impact may have
interfered with my ability to detect the effect of added predators.
In the Clarksville earwig experiment, I was provided with strong evidence that my
endemic natural enemy removal had been ineffective. When the wooden shelters were
brought back to the laboratory for mortality assessment, several contained predatory
beetle larvae. These larvae were primarily clerid larvae (Coleoptera: Cleridae), but a few
cantharids (Coleoptera: Cantharidae) were also collected from wooden shelters. The
beetles in many cases were found still feeding on codling moth pupae at the time that
they were collected and signs of their feeding were seen on other pupae and fifth instar
larvae in the wooden shelters at Clarksville.
Post hoc analysis of the mortality data indicated no relationship between
presence of these beetle larvae and the predator removal treatment (data not shown).
There was however, a clear relationship between presence of these beetle larvae and
proportion mortality of codling moth sentinel prey (Fig. 3.5) and habitat complexity (Fig.

76

3.6). This provides evidence that at Clarksville endemic natural enemy presence was
confounded with one of my experimental variables (complexity), and would have made
it much more difficult to detect the predicted relationship between complexity and
foraging success of the introduced natural enemies.
Additionally, I may have discovered an important natural enemy of the codling
moth in Michigan. Two species of clerid beetle adults were reported to feed on codling
moth in lab trials performed by Jaynes and Marucci (1947) using specimens collected
from pitfall traps in apple orchards in West Virginia (USA). However, this is the first
report of clerid larvae feeding on codling moth larvae in the field. These predatory
beetles may be a promising new target for conservation biological control programs in
apple orchards.
This discovery also brings to light an important implication of my work. Even
though reduced orchard floor complexity improved the effectiveness of
entomopathogenic nematode releases, the removal of vegetation and other material
from under the trees can have important consequences for other important natural
enemies. The microclimate provided by dense vegetation (Thomas et al., 1991),
resources provided by flowering plants (Begum et al., 2007), and alternative prey in
understory litter (Mathews et al., 2004) can be important to beneficial insects in
agricultural habitats (Isaacs et al., 2009; Landis et al., 2000). On the other hand, codling
moth larvae may benefit from spatial refuge on the orchard floor as demonstrated by the
effectiveness of orchard sanitation as a management tactic for this pest (Judd et al.,
1997; Sherman III, 1936; Steiner and Ackerman, 1936). It has also recently been
suggested that adult codling moth may use non-crop vegetation as a daytime refuge

77

(Epstein et al., 2011). This is certainly an area that will benefit from future research,
especially as it pertains to modern apple production as beneficial arthropods become
more important.
3.4.2. Canopy connectivity
My results did not provide strong support for the hypothesis that increased
canopy connectivity would lead to higher levels of codling moth egg predation (Fig. 3.7).
However, I was unable to manipulate the endemic natural enemy community in these
orchards and the high levels of predation in control plots indicates that the effect of preexisting codling moth egg predators in the study orchards may have overwhelmed the
effect of released predators (Fig. 3.7). Visual samples of leaves while this experiment
was running indicated healthy populations of other codling moth egg predators on the
foliage in both study orchards (N.J. Walton, unpublished data). Results of the beat
sampling did indicate that the released natural enemies were able to move farther within
the high connectivity plots than the low connectivity plots (Fig. 3.8). However, this
conclusion is largely based on the non-significant difference between the density of
Chrysopid larvae in the edges of plots in the low connectivity orchard (Fig. 3.8).
Unfortunately I did not differentiate endemic Chrysopids in beat samples from the C.
rufilabris larvae that I released. The dispersion pattern in Figure 8 could have also
resulted from a higher endemic lacewing population in that orchard and or increased
intraguild predation of the introduced C. rufilabris larvae in the low connectivity orchard.
3.5. Conclusion
The effect that the natural enemies of a species have on that species is
determined largely by the frequency of encounter and their ability to inflict mortality once

78

an encounter has occurred (Holling, 1959). For sedentary prey and an actively
searching enemy, the frequency of encounter is a function of the separation of prey
individuals in space. Furthermore, when organisms are small relative to the scale of the
structural heterogeneity of the habitat, separation of prey individuals in space increases
disproportionately with structural complexity (Cuddington and Yodzis, 2002). I found
empirical support for this simplified consumer-resource model in my experiments with
entomopathogenic nematodes on the orchard floor and lacewing larvae in the tree
canopy. However, I also found evidence that other important factors such as endemic
natural enemy population densities are affected by the structural complexity of the
orchard habitat. Future studies are needed to disentangle the effects of structural
complexity on the field efficacy of natural enemies of the codling moth. Understanding
the aspects of habitat structural complexity that govern the effect of natural enemies on
crop pests is one of the major challenges facing practitioners of habitat manipulation for
pest management (Gurr et al., 2004b; Murdoch, 1975).

79

Chapter 4: An exploration of how moisture and structural complexity may affect
the foraging success of Steinernema feltiae Filipjev against codling moth (Cydia
pomonella [L.]).
4.1. Introduction
Entomopathogenic nematodes (EPN) are one of the most promising
bioinsecticides used in organic agriculture worldwide (Denno et al., 2008; Lacey and
Shapiro-Ilan, 2008; Stuart et al., 2008). Entomopathogenic nematodes are obligate
parasites of insects that enter into hosts through the mouth, anus, and spiracles, and
release a symbiotic bacterium, Xenorhabdus spp., which is the agent of target pest
mortality (Shapiro-Ilan et al., 2003). The codling moth (Cydia Pomonella [L.]) is the most
serious insect pest of apple production worldwide and is of critical concern to North
American apple producers (Barnes, 1991; Jones et al., 2009; Solomon, 1991). The
application of entomopathogenic nematodes is a promising codling moth management
option available to apple growers. Entomopathogenic nematodes are available
commercially or can be reared by growers in-house and can inflict high rates of mortality
on codling moth larvae (Lacey and Shapiro-Ilan, 2008).
The codling moth is most susceptible to infection by EPN in the field during the
wandering larva, pupa, and overwintering larva stages (Kaya et al., 1984; Unruh and
Lacey, 2001). The wandering larva stage is the stage that begins when the larva leaves
an apple and begins to search for a location to pupate (summer) or overwinter (fall) and
ends when they spin a cocoon in a protected location. Wandering larvae are active
primarily at night (Borden, 1931; Garlick and Boyce, 1940; Geier, 1963) and may take
several hours to find a suitable location for their cocoon (Garlick, 1948; Geier, 1963).

80

Codling moth cocoons can be found anywhere in the orchard where conditions
are relatively dry (Glen and Milsom, 1978) and they are protected from predators
(Chandler, 1928; MacLellan, 1960). The majority of cocoons are found under flakes of
bark (on rough-barked apple trees), and in any available crack or crevice on the trunk
from the canopy to the interface with the soil surface (Chandler, 1928; Garlick, 1948;
Jumean et al., 2009; Marshall, 1940). In orchards with rough-barked trees, fewer
cocoons are found in the leaf litter and vegetation surrounding the trees (Chandler,
1928; Jaynes and Marucci, 1947; MacLellan, 1960). Artificial substrates are often very
attractive to codling moth wandering larvae, so barns, fence posts, trash, and apple bins
often serve as suitable overwintering and pupation sites (Chandler, 1928; Lacey and
Chauvin, 1999; Steiner, 1929).
Entomopathogenic nematode applications for codling moth management target
cocooned larvae on the tree and in surface litter on the orchard floor. A study performed
in the Pacific Northwest achieved between 72%-86% mortality of cocooned codling
moth larvae in apple orchards (Unruh and Lacey, 2001). However, sentinel prey used in
this and many similar studies are placed on the tree trunk in artificial substrates and
may not reflect actual mortality rates of wild codling moth from EPN applications in the
field (Kaya et al., 1984; Lacey et al., 2006a). Additionally, modern high-density (<1000
trees per acre) orchards with smooth-barked cultivars do not provide as much arboreal
codling moth cocoon habitat as traditional low-density orchards with rough-barked trees
(Walton, 2013c; Walton, 2013d). In the future, the availability of sheltered locations
(structural complexity) on the orchard floor is likely to have an a greater impact on

81

codling moth intergenerational survival and population dynamics than mortality on the
tree trunk (Walton, 2013c).
Two of the most important abiotic factors affecting the ability of EPN IJs to kill
codling moth larvae and pupae on the orchard floor are the structural complexity of that
habitat (Campbell and Gaugler, 1993; Walton, 2013c) and biologically available water
(Navaneethan et al., 2010). I designed two laboratory microcosm experiments to
investigate the mechanisms by which these abiotic factors might affect the survival of
codling moth larvae in the presence of entomopathogens. In the first experiment I
manipulated the structural complexity of microcosms by varying the amount of soil
surface litter and measured the infection rate by entomopathogenic nematodes in
microcosms of different complexities. In the second I held structural complexity
constant, varied the amount of soil moisture in the microcosms and measured the effect
this had on cocoon site selection by codling moth.
4.1.2. Experiment 1: Habitat structural complexity and EPN foraging efficiency
Entomopathogenic nematodes can be classified according to foraging behavior
along a spectrum from ambusher (sit-and-wait) to cruiser (search-and-destroy)(Lewis et
al., 2006). There is considerable evidence that entomopathogenic nematodes with a
cruiser-type foraging strategy are more effective at infecting sessile hosts in protected
locations than ambushers (Campbell et al., 2003; Huey and Pianka, 1981; Lacey et al.,
2006b). However, even active foragers should experience reduced foraging success in
habitats with increased structural complexity (Huffaker, 1958; Janssen et al., 2007;
Walton, 2013c). Unfortunately this hypothesis is difficult to test using EPNs under field
conditions due to confounding variables such as temperature, soil type, alternative

82

hosts, and predation, (Duncan et al., 2007; Ekmen et al., 2010; Greenwood et al., 2011;
Stuart et al., 2008; Walton, 2013c). In addition, EPNs are essentially aquatic organisms
and they are often applied in a water suspension using readily adaptable agricultural
spray equipment (Kaya and Stock, 1997; Lacey et al., 2006a; Unruh and Lacey, 2001).
This makes it difficult to differentiate between purely physical effects of increased
habitat complexity on the orchard floor and effects caused by reduced searching
efficiency of the EPNs (Unruh and Lacey, 2001; Wise et al., 2010).
In other words, is it primarily a spray penetration problem, or a structural
complexity problem? I designed this experiment to answer that question using an
actively searching nematode species that is commonly used in field experiments against
overwintering codling moth fifth instar larvae in orchard habitats (Kaya et al., 1984;
Lacey et al., 2006a; Lewis et al., 2006). The EPN species used was Steinernema feltiae
Filipjev (Rhabditida: Steinernematidae), an EPN with a foraging mode that is
intermediate between ambusher and cruiser (Campbell et al., 2003). This experiment
tested the hypothesis that increasing structural complexity would reduce the infection
rate of codling moth by S. feltiae in laboratory microcosms.
4.1.3. Experiment 2: Soil moisture and codling moth cocoon site selection
One drawback of EPNs for pest management is that nematodes are aquatic
organisms and require a film of water within which to move as they search for and infect
their hosts (Kaya and Stock, 1997; Lacey et al., 2006a; Lacey et al., 2006b; Lewis et al.,
2006; Spence et al., 2008; Unruh and Lacey, 2001; Wallace, 1958). Codling moth have
been reported to avoid moist locations as sites for overwintering or pupation (Glen and
Milsom, 1978). However, no formal investigations have been made regarding moisture

83

effects on cocoon site selection by codling moth since that report. If codling moth fifth
instars actively avoid building cocoons in wet locations or leave cocoons when
conditions become unfavorable (i.e. wet), EPN applications may be less effective than
expected against wild populations in the field. I designed an experiment to test the
hypothesis that codling moth prefer low moisture sites for cocoon construction by
manipulating the moisture content of small plastic arenas filled with moist sand and
corrugated cardboard flutes.
4.2. Materials and Methods
4.2.1. Experimental organisms
Sentinel fifth instar codling moth were obtained from the USDA-ARS Yakima
Agricultural Research Laboratory (Wapato, WA, USA). Larvae were surface sanitized
before each experiment by submerging them briefly in a 0.05% sodium hypochlorite
solution. Steinernema feltiae infective juveniles were produced in Galleria mellonella (L.)
(Lepidoptera: Pyralidae) larvae using the in vivo method described by Kaya and Stock
(Kaya and Stock, 1997). All IJs used in these experiments had emerged from hosts less
than two weeks prior to the experiment.
4.2.2. Habitat structural complexity and EPN foraging efficiency
I manipulated habitat complexity and nematode density in a 3 by 3 factorial
design. The three levels of nematode density were control (water only), low EPN (~12
S. feltiae IJs per microcosm), and high EPN (~28 S. feltiae IJs per microcosm). The
three levels of complexity in the microcosms were bare sand, simple, and complex. The
bare sand microcosms contained 500 g of play sand. Sand was washed three times to
remove water soluble contaminants that might affect nematode survival and then dried

84

in a drying oven at 100 C for 24 h to sanitize and standardize water content.
Complexity was added to the other treatments with the addition of synthetic leaves
made of woven polyester. Leaves were sanitized in an autoclave prior to use in the
experiment. Simple microcosms contained 2 synthetic leaves and 500 g of sand.
Complex microcosms contained 24 synthetic leaves and 500 g of sand. Microcosms
were made of 500 mL polypropylene deli containers with lids (Fabri-Kal®, Kalmazoo,
MI). A small hole was punched in the lid of each container with a push-pin to improve
aeration. Containers were sanitized by soaking in a 0.1% sodium hypochlorite solution
for 1 minute and then allowed to air dry. Microcosms were prepared by putting 500 g dry
sanitized sand into each 500 mL container and then pipetting 20 mL boiled tap water
into the center. Containers with moist sand were then left overnight to allow the water to
distribute itself by capillary action. Previous experiments have shown that this procedure
provides sufficient water content for survival of EPN IJs in 500 g dry sand (N.J. Walton,
unpublished data).
To prepare for the experiment, 45 surface sterilized codling moth larvae were
placed in microcosms designated for the bare sand treatment and placed in an
environmental chamber overnight to spin cocoons. The remaining larvae were placed
between two synthetic leaves stacked vertically in 1 oz. polystyrene soufflé cups
(Solo®, Lake Forest, IL) and placed in the environmental chamber overnight to spin
cocoons.
Steinernema feltiae IJs were applied in a boiled tap water suspension using a
1.75 L hand pump sprayer (Root-Lowell Manufacturing Co., Lowell, MI, USA). IJs were

85

applied to each container in 4 mL of water at two rates: 3 IJs per mL (Low) and 7 IJs per
mL (High). Control containers received an application of 4 mL boiled tap water only.
To begin the experiment, 75 fifth instar codling moth larvae that had spun
cocoons between two synthetic leaves in the soufflé cups (above) were placed
individually in the center of the microcosms that had been designated for the simple and
complex treatments. Complex microcosms (n=30) then had 22 synthetic leaves
distributed haphazardly over the surface of the sand. Simple (n=45) and bare sand
(n=45) microcosms had no additional synthetic leaves added to them. Fifteen containers
of each complexity level then received EPN or water applications according to the EPN
treatments described above. There was no Complex/Control microcosm treatment. After
EPN applications were made, all microcosms were randomized and placed in an
environmental chamber at 13ºC and 85% RH for 48 h. To simulate field conditions for
an evening application of EPN, the chamber was set to start in the dark phase of a
12:12 L:D cycle.
Following the 48 h incubation in the environmental chamber, each microcosm
was opened and the codling moth larva was removed and assessed for mortality and
EPN infection status. Then each larva was placed in a 60 mm petri dish containing
moist filter paper and sealed with parafilm. These dishes were held in the dark at 26ºC
and 45% RH for an additional 6 d and larval mortality and EPN infection status was
recorded every 24 h.
4.2.3. Soil moisture and codling moth cocoon site selection
I manipulated moisture content in a completely randomized design with 20
microcosms per moisture level. The five levels of moisture were Dry (no water), 7 mL,

86

20 mL, 55 mL, and 100 mL of water. Microcosms were made of 500 mL polypropylene
deli containers with lids (Fabri-Kal®, Kalmazoo, MI). A small hole was punched in the lid
of each container with a push-pin to improve aeration. Containers were sanitized by
soaking in a 0.1% sodium hypochlorite solution for 1 minute and then allowed to air dry.
Microcosms were prepared for the experiment by putting 500 g dry sanitized sand into
each 500 mL container. Before use, sand was washed three times to remove water
soluble contaminants and then dried in a drying oven at 100 C for 24 h to sanitize and
remove moisture. Water was boiled, cooled, and added to the center of each microcosm
with a hand pipettor. After the addition of water, microcosms were left overnight to allow
the water to distribute itself by capillary action. Previous experiments have shown that
this procedure with 7 mL of water does not provide optimal water content (-15kPa) for
infectivity of EPN IJs in 500 g dry sand but that it does for water levels >12mL (N.J.
Walton, unpublished data)(Kaya and Stock, 1997). The next day, codling moth cocoon
substrate was added to each microcosm by placing 24 1.75 cm by 1.75 cm squares of
corrugated cardboard haphazardly on top of the sand. Then a single fifth instar codling
moth was placed in the center of each container on top of the cardboard. After codling
moth fifth instars were placed in all microcosms, they were randomized and placed in an
environmental chamber at 13 C and 85% RH for 48 h. To simulate field conditions, the
chamber lighting was set to a 12:12 L:D cycle. Following the 48 h incubation period, all
microcosms were opened and searched for larvae, the locations of cocoons were
recorded, and the mortality or the larvae was assessed
Cocoon locations were quantified by dividing each microcosm into 2 strata:
Cardboard and Sand. A cocoon was recorded as residing in the Cardboard stratum if it

87

spun its cocoon in or on a piece of cardboard and was not in contact with the sand. The
Sand stratum was anywhere from the surface of the sand to the bottom of the container.
4.2.4. Statistical analyses
Statistical analysis for the complexity experiment was performed by survival
analysis using the Kaplan-Meier method in R (survival, v.2.9.2)(R-core-developmentTeam, 2009; Therneau and Lumley, 2009). Preplanned pairwise comparisons of
survival curves were performed using Gehan-Wilcoxon tests (α=0.05)(Harrington and
Fleming, 1982; Therneau and Lumley, 2009).
Data for the moisture experiments were analyzed by logistic regression and
Analysis of Deviance in R v2.9.2 (Dalgaard, 2002; R-core-development-Team, 2009).
The response variable in the logistic regression model was a binomial variable that was
equal to 1 if a larva had built its cocoon in the Cardboard stratum and was equal to 0 if
they had not. The predictor variable was moisture treatment.
4.3. Results
4.3.1. Habitat structural complexity and EPN foraging efficiency
There was a significant overall effect of EPN treatment on codling moth survival
2

(X =32.5, df=2, P<0.0001). The mean proportion EPN infected larvae (+/-SEM) in
Control microcosms was 0 (+/-0). In Low and High microcosms the mean proportion of
larvae infected with EPN was 0.40 (+/- 0.07) and 0.64 (+/- 0.07), respectively. There
was no significant overall effect of complexity treatment on codling moth survival
2

(X =1.3, df=2, P=0.52). The mean proportion EPN infected larvae (+/-SEM) in Bare
sand treatments was 0.31 (+/-0.07). In Simple and Complex microcosms the mean
proportion larvae infected with EPN was 0.42 (+/- 0.07) and 0.46 (+/- 0.09), respectively.

88

Preplanned pairwise comparisons of survival curves revealed some relevant
significant differences between codling moth survival in different microcosm types
(α=0.05) (Fig. 4.1). Codling moth survival in the Low EPN microcosms with complexity
odling
2

added was not significantly different from the Control microcosms (X =1.3, df=1,
P=0.257) (Fig.4.1). Survival in the Low EPN microcosms with only two synthetic leaves
2

added was significantly different from Control microcosms (X =12.4, df=1, P=0.0004)
fferent
(Fig.1). Survival in all High EPN microcosms was significantly different from all Control
microcosms (P<0.001), but not from any of the Low EPN microcosms (P>0.05) (Fig.

Proportion Mortality (+/- 95% CI)

4.1).
1
0.9

a a

a

0.8

a

0.7
0.6

Complex

a
ab

0.5

Simple

Bare Sand

0.4
0.3
0.2
0.1
0

b

b

Control

Low

High

Figure 4.1. Proportion mortality of sentinel codling moth larvae in experimental
n
microcosms. Pairwise comparisons were made using Gehan Wilcoxon tests (α=0.05).
Gehan-Wilcoxon
(
Columns with the same letter are not significantly different. There were no
Complex/Control microcosms.
4.3.2. Soil moisture and codling moth cocoon site selection

89

A total of 48 larvae built cocoons in the Sand stratum and 51 were found in
cocoons in the Cardboard stratum. One larva from a 7 mL microcosm was unaccounted
for and this microcosm was excluded from the data analysis. The results of the Analysis
of Deviance indicated a significant relationship between moisture treatment and the
probability that a larva would be found in the Cardboard stratum (df=4,94, P=0.007)
(Fig. 4.2).
100%
1

OR<1

OR=1

OR=1

OR>1

OR>1

Proportion larvae in sand

90%
80%
0.75
70%
Cardboard
Sand

60%
50%
0.50
40%
30%
0.25
20%
10%
0
0%
0

7

20
mL of water

55

100

Figure 4.2. The proportion of larvae found in the Sand vs. Cardboard strata by moisture
level in experimental microcosms. Odds ratio (OR) indicates results of logistic
regression (α=0.05, glm, R v.2.9.2). When OR<1 larvae were more likely to be found in
Sand than Cardboard. When OR=1 larvae were equally likely to be found in either
stratum. When OR>1 larvae were more likely to be found in Cardboard than Sand.
The 95% confidence interval of the odds ratio for larvae in Dry microcosms
constructing cocoons on cardboard was between 0.10 and 0.86. The odds ratio for

90

larvae in 7 mL microcosms constructing cocoons on cardboard was between 0.57 and
9.02 and the odds ratio for larvae in 20 mL microcosms constructing cocoons on
cardboard was between 0.81 and 12.24 (α=0.05). The 95% confidence interval of the
odds ratio for larvae in microcosms containing 55 mL of water constructing a cocoon in
the Cardboard stratum was from 1.22 to 18.76 and it was between 2.94 and 60.5 for
larvae in the microcosms with 100 mL of water added.
4.4. Discussion
4.4.1. Habitat structural complexity and EPN foraging efficiency
In this experiment, nematode infection was higher compared to the controls when
a higher density of EPN infective juveniles was added to Complex microcosms, but not
Bare Sand and Simple microcosms (Fig. 4.1). In other words, there was no effect of
increasing EPN concentration from Low to High in Simple or Bare Sand microcosms,
but there was an effect of increasing EPN concentration in Complex microcosms. This
provides evidence that EPN IJ foraging or infection success was reduced when
complexity was added to microcosms.
Reduced spray penetration in the Complex microcosms was not likely to have
caused the observed patterns. Assuming a random distribution of nematodes following
application, all codling moth sentinels were well within the foraging range of S. feltiae
IJs at the temporal and spatial scale of this experiment (Campbell and Gaugler, 1993;
Hui and Webster, 2000; Neumann and Shields, 2006; Rolston et al., 2006). The most
likely explanation for the observed difference in response to increasing nematode
concentration between Simple and Complex microcosms (Fig. 4.1) is reduced foraging
success of the S. feltiae IJs caused by increased structural complexity imposed by the

91

addition of synthetic leaves (Cuddington and Yodzis, 2002; Holling, 1959; Huffaker,
1958; Janssen et al., 2007; Kostylev et al., 2005).
4.4.2. Soil moisture and codling moth cocoon site selection
In microcosms containing more than 55 mL water, the codling moth larvae in this
experiment preferred building their cocoons in locations where they were not in contact
with the moist sand (Fig. 4.2). Moist sand is considered to be an ideal foraging substrate
for entomopathogenic nematode IJs (Kaya and Stock, 1997; Nielsen and Lewis, 2012).
It is also known that wetting agents and humectants improve the infection rates of
codling moth by S. feltiae by improving their ability to penetrate into the cryptic locations
where codling moth construct their cocoons (Lacey et al., 2005).
Some insect larvae are known to have defensive behaviors that protect them
from infection by EPN IJs (Gaugler et al., 1994; Koppenhofer et al., 2000; Renn, 1998;
Schroeder et al., 1993). I did not design this experiment to determine whether or not
avoidance of high moisture habitats is an adaptation of codling moth fifth instars to
avoid infection by nematodes. Rather, I was simply interested in whether or not they
would avoid high moisture locations when choosing where to construct their cocoons. A
good next step would be to repeat this experiment with an addition of EPN to see if
codling moth in the upper stratum of such microcosms exhibit lower EPN infection rates.
4.5. Conclusions
Orchardists manage the orchard floor beneath the tree canopy for optimal
nutrient availability and minimized pest and disease pressure (Atucha et al., 2011;
Mathews et al., 2004). The removal of vegetation, litter, and other material from under
orchard trees, a practice know as orchard sanitation, has a long history of usefulness as

92

a tactic to reduce codling moth populations (Judd et al., 1997; Sherman III, 1936;
Steiner and Ackerman, 1936). My results indicate that reducing orchard floor structural
complexity may improve the effectiveness of S. feltiae releases for codling moth
management, a phenomenon that has also been demonstrated in field cages (Walton,
2013c). However, reducing the structural and vegetative complexity of the orchard floor
can have non-target impacts on beneficial arthropods (Walton, 2013c).
Post-application irrigation is a recommended and effective practice that greatly
improves the effectiveness of EPN applications targeting codling moth in their
overwintering cocoons (Lacey et al., 2006a; Unruh and Lacey, 2001). However, most of
the studies demonstrating effective EPN applications targeting codling moth have been
performed in orchards in the arid west with relatively low orchard floor structural
complexity and or have used sentinel codling moth larvae placed on the tree trunk
(Kaya et al., 1984; Lacey et al., 2006a; Unruh and Lacey, 2001). In high complexity
orchard floors such as those commonly found in certified organic orchards in the
eastern United States, EPN applications targeting the overwintering stage of codling
moth may be less effective than would be suggested by previous studies. If codling
moth larvae actively avoid high moisture locations and increased structural complexity
provides alternative cocoon sites, then they may be able to escape infection by
entomopathogenic nematodes. EPN applications targeting the wandering larva stage of
the codling moth may be more effective, in these types of orchards.

93

Chapter 5. Towards more sustainable management of codling moth (Cydia
pomonella [L.]) in apple orchards
5.1. Introduction
The Environmental Protection Agency estimates that the United States
agricultural sector spent $11.8 billion on all pesticides and $1.83 billion on insecticides
and acaricides in 2007 (the latest year for which these data are available) (Grube et al.,
2011). In that same year, U.S. apple producers applied an estimated 3,412.5 metric
tons of insecticides (USDA-NASS, 2012b) to 97% of the total US apple bearing acreage
(USDA-NASS, 2008). In 2011, U.S. apple producers applied an estimated 621 metric
tons of insecticides to 84% of apple bearing acreage (USDA-NASS, 2012b). This
decline in insecticide usage is expected to continue as mating disruption and other
reduced risk approaches replace the broad-spectrum contact active insecticides that
apple producers had previously relied on (Brunner et al., 2005; Epstein, 2010; Grube et
al., 2011; Jones et al., 2009).
As dicussed previously (Walton, 2013a), this reduction in insecticide use is
expected to lead to an increase in the contribution that arthropod natural enemies make
to pest management in apple orchards. Table 5.1 (below) summarizes the previous
estimates of codling moth mortality by life stage (Walton, 2013a). Clearly, there is a
great deal of variation in estimates of the contribution that natural enemies make to
codling moth management as well as the identities of the natural enemies involved
(Table 5.1). Accurate quantification and identification of natural enemies of crop pests is
important because it allows incorporation of this information into economic decision
making tools for pest managers (Radcliffe et al., 2009; Zhang and Swinton, 2009) and

94

environmental policy-makers (Boyd and Banzhaf, 2007; Isaacs et al., 2009; Losey and
Vaughan, 2006; Luck et al., 2009).
Table 5.1. Extant estimates of codling moth mortality in the field: Summary

Egg and 1 Instar

Estimated %
Mortality
2-86

Suspected Mortality
Agents
Parasitoids, green
lacewings, thrips,
mites, plant bugs,
earwigs

Instars 2-4

10-52

Unknown

4-100

Birds, ants, beetles,
parasitoids,
pathogens

Stage
st

5

th

instar

References
(Ferro et al., 1975;
Geier, 1963; Glen,
1975; Glen, 1977;
Jaynes and
Marucci, 1947;
MacLellan, 1962;
Summerland and
Steiner, 1943)
(Ferro et al., 1975;
MacLellan, 1962)
(Baumgartner,
1999; Ferro et al.,
1975; Glen and
Milsom, 1978;
Jaques and
MacLellan, 1965;
Jaynes and
Marucci, 1947;
MacLellan, 1960;
MacLellan, 1962;
Mathews et al.,
2004; Mills,
2005b; Riddick,
1994; Solomon et
al., 1976;
Subinprasert,
1987)
None

Adult
Unknown
Birds, bats, spiders
Total % Mortality
16-100
The identity of the natural enemies that make the greatest contribution to codling
moth pest management is also important to producers and researchers interested in
conservation biological control. Due to the diverse life histories of natural enemies it is
impractical to conserve all natural enemies in a commercial orchard. Crop specific
conservation practices that target the essential natural enemies (‘the right diversity’

95

sensu Landis et al. 2000) need to be developed based on rigorous scientific research.
Such an effort has the potentional to increase the functional biodiversity of both organic
and conventional agriculture (Letourneau and Bothwell, 2008).
Determining the optimal target for conservation biological control in apple
orchards in Michigan and elsewhere will be a difficult process. Apart from determining
natural enemies that are active in orchards and readily conserved, this process will also
require the determination of the optimal codling moth life stage to target. Some of the
natural enemies groups were observed attacking both the egg and fifth instar stages in
my video study (Walton, 2013d), but it is unlikely that these were the same species and
they may not have had a sufficient impact on either life stage to justify conservation
efforts.
Determining the optimal life stage of codling moth to target with conservation
biological control requires close examination of this pest’s life history, demography, and
population dynamics (Brown et al., 1978; Mills, 2005b). This information can be
incorporated into a stage structured matrix model which can and has been used to plan
biological control of the species (Mills, 2005b). Elasticity analysis of such a model can
then provide an estimate for the percentage of mortality that is needed to bring the
pest’s population intrinsic growth rate (r) to less than 1 (Caswell, 2001; Mills, 2005b).
More importantly it can help identify the life stage that has the greatest impact on
codling moth’s long-term population dynamics (Caswell, 2001). For example, in Mills’
(2005) stage structured matrix model for codling moth, the adult stage has the highest
elasticity value. In other words, a marginal increase in adult codling moth mortality
would (given those model parameters) have a larger impact on the pest than a marginal

96

increase in the mortality of any other life stage alone. Given this information we might
conclude that the optimal target for conservation biological control of codling moth
would be a natural enemy of the adult stage. In fact, this is the stage that mating
disruption targets this may be why it is such an effective management strategy.
Unfortunately, natural enemies of codling moth adults are poorly studied because of the
difficulty of collecting data on the mortality of such a mobile stage (Walton, 2013a).
Fortunately, second to the adult in impact on long term population dynamics is
the cocoon stage, followed closely by the wandering fifth instar stage and the egg stage.
The model predicts that a 21.0% or a 34.6% daily mortality in the cocoon and egg stage
respectively will be sufficient to bring the intrinsic growth rate of a codling moth
population to less than 1. Table 5.2 (below) presents mortality rates from my field
studies in Michigan for these two life stages. These rates can be crudely converted to
daily mortality rates by dividing each by the number of days that sentinel prey were
exposed to field conditions (Table 5.2). This yields maximum daily mortality rates (+/SEM) of 8.7% (+/- 1.0) for sentinel eggs on pins in the canopy and 5.9% (+/- 1.5) for
sentinel fifth instars in field cages on the orchard floor. These values are 28% (+/-5) and
25% (+/-4) of the critical daily mortality rates estimated by Mills (2005b) for egg and fifth
instar codling moth, respectively. This suggests that natural enemies are making a
contribution to pest management in the orchards that I studied in Michigan, but that on
their own they would not bring codling moth population growth to a halt under current
conditions. However, conservation targetted at the right natural enemies could bring this
mortality rate within a range where additional economic benefits are realized (Jones et
al., 2009; Zhang and Swinton, 2009).

97

Table 5.2. Estimates of codling moth mortality in Michigan apple orchards

st

% Daily Mortality
(+/-SEM)
2.2±0.6

Egg and 1 Instar

st

8.7±1.0

Instars 2-4
th
5 instar

Unknown
2.6±0.7

Stage
Egg and 1 Instar

5

th

instar

5.9±1.5

Adult
Unknown
5.2. Future Research

Mortality Agents
Lacewings, spiders,
mites, bugs,
parasitoids,
harvestmen, tree
crickets, mice
Lacewings, mites,
predatory thrips,
bugs
Unknown
Ants, parasitoid
wasps, spiders,
beetles
Unknown, beetle
larvae
Birds, bats, spiders

Method
Video

Sentinel prey on
pins in canopy
(control plots)
Video
Sentinel prey in
control cages
None

I have identified some of the most active natural enemies of both of these
important codling moth life stages in Michigan apple orchards. However, there is still a
great deal of work to be done. For example, many of the natural enemies that were
observed need to be identified to species before conservation efforts can be made.
Lacewings were the most active egg predators in at least one year of the study, but
lacewing species differ significantly in terms of foraging habitat, phenology, and the
resource needs of the adult and larval stages (McEwen et al., 2001). The same can be
said of most of the other natural enemies observed in these studies.
Ants and parasitoid wasps may be two groups that will be particularly promising
as targets of conservation efforts in Michigan apple orchards. Ants were one of the first
organisms ever used in conservation biological control (Hajek, 2004) and in locations
where a foraging colony is present they can have a significant impact on codling moth
fifth instar mortality (Jaynes and Marucci, 1947; Walton, 2013d). Parasitoid wasps can

98

also make a significant contribution to codling moth pest management in orchards and
they have generally been shown to benefit from the addition of flowering habitat as a
food source for the adult stage (Blaauw and Isaacs, 2012; Pfiffner et al., 2004; Walton
and Isaacs, 2011a; Walton and Isaacs, 2011b). The parasitoids observed in my video
studies were unidentifiable based on video images alone. One parasitoid wasp that was
observed on video attacking sentinel codling moth in 2010, however, at least
superficially resembled the species (M. ridibundus) that has become an important
classical biological control introduction in California (Mills, 2005a; Walton, 2013a).
Attempts to rear parasitoids from the sentinel prey that were present at that site were
unsuccessful and no adult specimens were collected from the field. Adults of that
species were released at AlMar orchards in 2008 (David Epstein, personal
communication) and they have been collected subsequently in Michigan (Larry Gut,
personal communication).
Predatory beetles should also be investigated as a possible target for
conservation in apple orchards. Many predatory beetles respond positively to
conservation of grassy areas within crop fields (Gurr et al., 2004a; Thomas et al., 1991;
Walton, 2013c). However, these practices usually target carabid beetles, a group that
was not observed preying on any life stage of codling moth in my video study. Carabids
may be more important as natural enemies of codling moth wandering larvae on the
orchard floor (Glen and Milsom, 1978; Jaynes and Marucci, 1947; Riddick, 1994), a life
stage that I did not collect data on. Regardless, the other beetle groups that I observed
attacking codling moth fifth instars in Michigan and have been observed elsewhere

99

(Jaynes and Marucci, 1947) may also respond positively to conservation practices that
benefit carabids such as maintenance of shelter habitats(Griffiths et al., 2008).
These predatory beetles (clerids and cantharids) might also benefit from flower
resources (Graham et al., 2012; Lovell, 1915; Opitz, 2002). The organic research
orchard at Clarksville, MI where those beetles were observed (Walton, 2013c) is
bordered by a native perenial wildflower planting and adult cantharids were observed on
flowers (Solidago sp.) at the time that those experiments were being carried out
(N.J.Walton, personal observation). Cantharid beetles and some clerid beetles are
known to use flowers as a resource during their adult stages (Graham et al., 2012;
Lovell, 1915; Opitz, 2002). This is a phenomenon that will hopefully be investigated
further.
5.3. Trade-offs in conservation biological control of codling moth
Habitat structural complexity on the orchard floor has been shown to affect
codling moth natural enemies in two opposing ways (Walton, 2013c). Increased
complexity improves the effectiveness of naturally ocurring natural enemies but it
reduces the effectiveness of augmentative applications of entompathogenic nematodes.
This suggests the existence of a trade-off in codling moth pest management between
reducing orchard floor complexity and relying on inundative augmentatiion biocontrol or
increasing complexity and relying on conservation or inoculative augmentation
biocontrol.
Trade-offs are not uncommon in conservation biological control (Gurr and
Kvedaras, 2010; Prokopy, 1994), but there are also notable examples where apparent
trade-offs have been overcome. These examples often involve the application of

100

detailed information about the natural history of the organisms involved, especially
regarding the optimal spatial and temporal scales for habitat manipulations. For
example, strip management at an appropriate spatial scale has been effective in a
number of systems as a means to selectively conserve natural enemies (Bugg and
Waddington, 1994; Hossain et al., 2002; Thomas et al., 1991). Integration of non-crop
plants into or around perrenial fruit crops has also been found to improve biological
control (Brown and Mathews, 2007; Corbett and Rosenheim, 1996; Doutt and Nakata,
1973; English-Loeb et al., 2003; Thomson and Hoffmann, 2009; Thomson and
Hoffmann, 2013; Unruh et al., 2012). Also, reduced mowing frequency can increase the
abundance of some important natural enemies in pear orchards (Horton et al., 2003),
and the use of post-emergence herbicides rather than pre-emergence herbicides has
been suggested as a means to reduce weed competition while conserving soil and a
structurally complex habitat (Atucha et al., 2011).
The challenge for future apple growers using habitat management to increase
the impact of natural enemies on codling moth may be to find a balance between the
opposing effects of habitat complexity. There may exist an optimal level of complexity
that maximizes the impact of the codling moth’s natural enemies. Synergistic
interactions may also exist between conservation practices and other pest management
tools and these should be sought out and exploited where possible (Botto and Glaz,
2010; Grewal et al., 2005; Gurr and Kvedaras, 2010; Vreysen and Robinson, 2011).
5.4. Conclusion
Developing and implementing sustainable pest management strategies and
related food production technologies will be one of the great challenges facing

101

st

agriculture in the 21 century and beyond (Godfray et al., 2010; Gurr and Kvedaras,
2010). One approach to meeting this challenge is to design agricultural systems that
mimic natural systems (Gurr et al., 2004a; Jackson and Jackson, 2002). Unfortunately,
there is a great deal that we still do not understand about the structure and function of
natural and agricultural systems. I have endeavored throughout this dissertation to
provide data that will advance our understanding of both.

102

APPENDIX

103

APPENDIX
RECORD OF DEPOSITION OF VOUCHER SPECIMENS
The specimens listed below have been deposited in the named museum as samples of
those species or other taxa, which were used in this research. Voucher recognition
labels bearing the voucher number have been attached or included in fluid preserved
specimens.
Voucher Number: 2013-09
Author and Title of thesis:
Nathaniel J. Walton
The role of natural enemies in pest management of codling moth (Cydia Pomonella [L.])
in organic apple orchards in Michigan
Museum where deposited:
Albert J. Cook Arthropod Research Collection, Michigan State University (MSU)
Table A.1 List of voucher specimens.
Order*
Opiliones
Subclass:
Acari
Hemiptera
Aranae
Neuroptera
Hymenoptera
Hymenoptera
Coleoptera
Coleoptera
Orthoptera
Rodentia
Rhabditida
Rhabditida

Family*

GenusSpecies

Life Stage Quantity Preservation
Adult

1
1

Digital
Digital
Digital
Digital
Digital
Digital

Chrysopidae
Suborder:
Apocrita
Formicidae
Superfamily:
Cleroidea
Superfamily:
Cleroidea
Gryllidae

Larva
Adult

1
1
1
1

Adult
Larva

1
1

Digital
Digital

Adult

1

Digital

Adult

Steinernematidae Steinernema
feltiae
Steinernematidae Steinernema
feltiae

Infective
Juvenile
Adult

1
1
1

Digital
Digital
Digital

1

Digital

1

Ethanol

Class:
Chilopoda
*Unless specified otherwise.

104

Table A.1 (cont'd).
Order*
Opiliones
Aranae
Neuroptera
Thysanoptera
Hemiptera
Diptera
Coleoptera
Coleoptera

Family*

Hymenoptera
Lepidoptera

Chrysopidae
Aeolothripidae
Anthocoridae
Syrphidae
Staphylinidae
Superfamily:
Cleridae
Superfamily:
Cleroidea
Cantharidae
Suborder:
Apocrita
Formicidae
Tortricidae

Neuroptera

Chrysopidae

Dermaptera

Forficulidae

Diptera
Neuroptera

Tachinidae
Chrysopidae

Lepidoptera

Tortricidae

Lepidoptera

Tortricidae

Coleoptera
Coleoptera
Hymenoptera

GenusSpecies

Life Stage Quantity Preservation

Adult
Adult
Adult
Adult
Adult
Larva

Coleoptera
Carabidae
Coleoptera
Coccinellidae
Coleoptera
Elateridae
*Unless specified otherwise.

105

1
1

Ethanol
Ethanol

Adult
Larva

8
1

Ethanol
Ethanol

Larva

11

Ethanol

Adult

12

Pinned

Adult
Adult

1
7

Pinned
Pinned

Adult

5

Pinned

Adult

5

Pinned

Adult
Adult
Adult

Chrysoperla
rufilabris
Grapholita
molesta
Cydia
pomonella

Ethanol
Ethanol
Ethanol
Ethanol
Ethanol
Ethanol
Ethanol
Ethanol

Larva
Adult
Cydia
pomonella
Chrysoperla
rufilabris
Forficula
auricularia

1
1
4
1
1
1
1
8

1
1
1

Pinned
Pinned
Pinned

REFERENCES

106

REFERENCES
Anderson, T.W., 2001. Predator responses, prey refuges, and density-dependent
mortality of a marine fish. Ecology 82, 245-257.
Atucha, A., Merwin, I.A., Brown, M.G., 2011. Long-term effects of four groundcover
management systems in an apple orchard. HortScience 46, 1176-1183.
Barbosa, P., 1998. Conservation biological control. Academic Press, San Diego.
Barnes, M.M., 1991. Tortricids in pome and stone fruits: Codling moth occurrence host
race formation and damage. In: van der Geest, L.P.S., Evenhuis, H.H., Eds.), Tortricid
pests: Their biology natural enemies and control. Elsevier, New York, pp. 313-327.
Baumgartner, J.A., 1999. Birds and arthropod predation of codling moth (Cydia
pomonella) in apple orchards. San Jose State University, pp. 75.
Begum, M., Gurr, G.M., Wratten, S.D., Hedberg, P.R., Nicol, H.I., 2007. Using selective
food plants to maximize biological control of vineyard pests. Journal of Applied Ecology,
547-554.
Blaauw, B.R., Isaacs, R., 2012. Larger wildflower plantings increase natural enemy
density, diversity, and biological control of sentinel prey, without increasing herbivore
density. Ecological Entomology 37, 386-394.
Blommers, L., 1992. Selective package and natural control in orchard IPM. Acta
Phytopathologica Et Entomologica Hungarica 27, 127-134.
Borden, A.D., 1931. Some field observations on codling moth behavior. Journal of
Economic Entomology 24, 1137-1145.
Borror, D.J., Triplehorn, C.A., Johnson, N.F., 1992. An introduction to the study of
insects. Harcourt Brace, New York.
Botto, E., Glaz, P., 2010. Potential for controlling codling moth Cydia pomonella
(Linnaeus) (Lepidoptera: Tortricidae) in Argentina using the sterile insect technique and
egg parasitoids. Journal of Applied Entomology 134, 251-260.

107

Boyd, J., Banzhaf, S., 2007. What are ecosystem services? The need for standardized
environmental accounting units. Ecological Economics 63, 616-626.
Brown, G.C., Berryman, A.A., Bogyo, T.P., 1978. Simulating codling moth population
dynamics: Model development, validation, and sensitivity. Environmental Entomology 7,
219-227.
Brown, M.W., 1993. Resilience of the natural arthropod community on apple to external
disturbance Ecological Entomology 18, 169-183.
Brown, M.W., Mathews, C.R., 2007. Conservation biological control of rosy apple aphid,
Dysaphis plantaginea (Passerini), in Eastern North America. Environmental Entomology
36, 1131-1139.
Brunner, J.F., Beers, E.H., Dunley, J.E., Doerr, M., Granger, K., 2005. Role of
neonicotinyl insecticides in washington apple integrated pest management. Part i.
Control of Lepidopteran pests. Journal of Insect Science 5, 1-10.
Bugg, R.L., Waddington, C., 1994. Using cover crops to manage arthropod pests of
orchards: A review. Agriculture Ecosystems & Environment 50, 11-28.
Calderon, M., Navarro, S., 1971. Effects of ultra-violet irradiation on the eggs of
Ephestia cautella (wlk.) (Lepidoptera: Physitidae). Journal of Stored Products Research
7, 309-311.
Caldwell, M.M., Bornman, J.F., Ballare, C.L., Flint, S.D., Kulandaivelu, G., 2007.
Terrestrial ecosystems, increased solar ultraviolet radiation, and interactions with other
climate change factors. Photochemical & Photobiological Sciences 6, 252-266.
Campbell, J.F., Gaugler, R., 1993. Nictation behaviour and its ecological implications in
the host search strategies of entomopathogenic nematodes (Heterorhabditidae and
Steinernematidae). Behaviour 126, 155-169.
Campbell, J.F., Lewis, E.E., Stock, S.P., Nadler, S., Kaya, H.K., 2003. Evolution of host
search strategies in entomopathogenic nematodes. Journal of Nematology 35, 142-145.
Caswell, H., 2001. Matrix population models: Construction, analysis, and interpretation.
Sinauer Associates Inc., Sunderland.

108

Chandler, S.C., 1928. Codling moth hibernation studies. Journal of Economic
Entomology 21, 315-318.
Corbett, A., Rosenheim, J.A., 1996. Impact of a natural enemy overwintering refuge and
its interaction with the surrounding landscape. Ecological Entomology 21, 155-164.
Crowder, D.W., Northfield, T.D., Strand, M.R., Snyder, W.E., 2010. Organic agriculture
promotes evenness and natural pest control. Nature 466, 109-U123.
Cuddington, K., Yodzis, P., 2002. Predator-prey dynamics and movement in fractal
environments. American Naturalist 160, 119-134.
Dalgaard, P., 2002. Logistic regression. Introductory statistics with R. Springer, New
York.
de Roince, C.B., Lavigne, C., Ricard, J.M., Franck, P., Bouvier, J.C., Garcin, A.,
Symondson, W.O.C., 2012. Predation by generalist predators on the codling moth
versus a closely-related emerging pest the oriental fruit moth: A molecular analysis.
Agricultural and Forest Entomology 14, 260-269.
Denno, R.F., Gruner, D.S., Kaplan, I., 2008. Potential for entomopathogenic nematodes
in biological control: A meta-analytical synthesis and insights from trophic cascade
theory. Journal of Nematology 40, 61-72.
Doutt, R.L., Nakata, J., 1973. The rubus leafhopper and its egg parasitoid: An endemic
biotic system useful in grape-pest management. Environmental Entomology 2, 381-386.
Drinkwater, L.E., Letourneau, D.K., Workneh, F., Bruggen, A.H.C.v., Shennan, C.,
1995. Fundamental differences between conventional and organic tomato
agroecosystems in California. Ecological Applications 5, 1098-1112.
Duncan, L.W., Graham, J.H., Zellers, J., Bright, D., Dunn, D.C., El-Borai, F.E.,
Porazinska, D.L., 2007. Food web responses to augmenting the entomopathogenic
nematodes in bare and animal manure-mulched soil. Journal of Nematology 39, 176189.
Duthie, B., Gries, G., Gries, R., Krupke, C., Derksen, S., 2003. Does pheromone-based
aggregation of codling moth larvae help procure future mates? Journal of Chemical
Ecology 29, 425-436.

109

Ekmen, Z.I., Hazir, S., Cakmak, I., Ozer, N., Karagoz, M., Kaya, H.K., 2010. Potential
negative effects on biological control by Sancassania polyphyllae (Acari: Acaridae) on
an entomopathogenic nematode species. Biological Control 54, 166-171.
English-Loeb, G., Rhainds, M., Martinson, T., Ugine, T., 2003. Influence of flowering
cover crops on anagrus parasitoids (Hymenoptera : Mymaridae) and erythroneura
leafhoppers (Homoptera : Cicadellidae) in new york vineyards. Agricultural and Forest
Entomology 5, 173-181.
Epstein, D.L., 2010. Development and delivery of biopesticide-based pest management
for Michigan apple producers. Entomology. Michigan State University, Michigan.
Epstein, D.L., Miller, J.R., Grieshop, M.J., Stelinski, L.L., Gut, L.J., 2011. Direct
sampling of resting codling moth (Lepidoptera: Tortricidae) adults in apple tree canopies
and surrounding habitats. Environmental Entomology 40, 661-668.
Epstein, D.L., Zack, R.S., Brunner, J.F., Gut, L., Brown, J.J., 2000. Effects of broadspectrum insecticides on epigeal arthropod biodiversity in pacific northwest apple
orchards. Environmental Entomology 29, 340-348.
Epstein, D.L., Zack, R.S., Brunner, J.F., Gut, L., Brown, J.J., 2001. Ground beetle
activity in apple orchards under reduced pesticide management regimes. Biological
Control 21, 97-104.
Ferro, D.N., Harwood, R.F., 1973. Intraspecific larval competition by the codling moth,
Laspeyresia pomonella. Environmental Entomology 2, 783-790.
Ferro, D.N., Sluss, R.R., Bogyo, T.P., 1975. Factors contributing to the biotic potential of
the codling moth, Laspeyresia pomonella (L.), in washington. Environmental
Entomology 4, 385-391.
Frank, S.D., Wratten, S.D., Sandhu, H.S., Shrewsbury, P.M., 2007. Video analysis to
determine how habitat strata affects predator diversity and predation of Epiphyas
postvittana (Lepidoptera : Tortricidae) in a vineyard. Biological Control 41, 230-236.
Garlick, W.G., 1948. A five-year field study of codling moth larval habits and adult
emergence. Scientific Agriculture 28, 273-292.

110

Garlick, W.G., Boyce, H.R., 1940. A note on the habits of mature codling moth larvae.
The Canadian Entomologist 72, 87.
Gaugler, R., Wang, Y., Campbell, J.F., 1994. Aggressive and evasive behaviors in
Popillia-japonica (Coleoptera, Scarabaeidae) larvae - defenses against
entomopathogenic nematode attack. Journal of Invertebrate Pathology 64, 193-199.
Geier, P.W., 1963. The life history of codling moth, Cydia pomonella (L.) (Lepidoptera:
Tortricidae), in the Australian capital territory. Australian Journal of Zoology 11, 323.
Giolo, F.P., Medina, P., Grutzmacher, A.D., Vinuela, E., 2009. Effects of pesticides
commonly used in peach orchards in brazil on predatory lacewing Chrysoperla carnea
under laboratory conditions. BioControl 54, 625-635.
Glen, D.M., 1975. Effects of predators on eggs of codling moth Cydia-pomonella, in a
cider-apple orchard in southwest england. Annals of Applied Biology 80, 115-119.
Glen, D.M., 1977. Predation of codling moth eggs, Cydia pomonella, the predators
responsible and their alternative prey. Journal of Applied Ecology 14, 445-456.
Glen, D.M., Milsom, N.F., 1978. Survival of mature larvae of codling moth (Cydia
pomonella) on apple trees and ground. Annals of Applied Biology 90, 133-146.
Godfray, H.C.J., Beddington, J.R., Crute, I.R., Haddad, L., Lawrence, D., Muir, J.F.,
Pretty, J., Robinson, S., Thomas, S.M., Toulmin, C., 2010. Food security: The challenge
of feeding 9 billion people. Science 327, 812-818.
Graham, E.E., Tooker, J.F., Hanks, L.M., 2012. Floral host plants of adult beetles in
central Illinois: An historical perspective. Annals of the Entomological Society of
America 105, 287-297.
Greenwood, C.M., Barbercheck, M.E., Brownie, C., 2011. Short term response of soil
microinvertebrates to application of entomopathogenic nematode-infected insects in two
tillage systems. Pedobiologia 54, 177-186.
Grewal, P.S., Ehlers, R.U., Shapiro-Ilan, D.I. Eds.), 2005. Nematodes as biocontrol
agents. CABI, Cambridge, MA.

111

Grieshop, M.J., Flinn, P.W., Nechols, J.R., Campbell, J.F., 2008. Effects of fine-grain
habitat complexity on egg parasitism by three species of trichogramma. Biological
Control 45, 328-336.
Grieshop, M.J., Werling, B., Buehrer, K., Perrone, J., Isaacs, R., Landis, D., 2012. Big
brother is watching: Studying insect predation in the age of digital surveillance.
American Entomologist 58, 172-182.
Griffiths, G.J.K., Holland, J.M., Bailey, A., Thomas, M.B., 2008. Efficacy and economics
of shelter habitats for conservation biological control. Biological Control 45, 200-209.
Grube, A., Donaldson, D., Kiely, T., Wu, L., 2011. Pesticide industry sales and usage:
2006 and 2007 market estimates. In: Office of Chemical Safety and Pollution
Prevention, (Ed.). United States Environmental Protection Agency, Washington, DC.
Gurr, G.M., Kvedaras, O.L., 2010. Synergizing biological control: Scope for sterile insect
technique, induced plant defences and cultural techniques to enhance natural enemy
impact. Biological Control 52, 198-207.
Gurr, G.M., Wratten, S.D., Altieri, M.A., 2004a. Ecological engineering for pest
management. Cornell University Press, Ithaca.
Gurr, G.M., Wratten, S.D., Berndt, L., Irvin, N., 2004b. Ecological engineering, habitat
manipulation and pest management. In: Gurr, G.M., Wratten, S.D., Altieri, M.A., Eds.,
Ecological engineering for pest management. Cornell University Press, Ithaca, pp. 112.
Gut, L.J., Brunner, J.F., 1998. Pheromone-based management of codling moth
(Lepidoptera : Tortricidae) in Washington apple orchards. Journal of Agricultural
Entomology 15, 387-406.
Hagley, E.A.C., 1972. Observations on codling moth longevity and egg hatchability.
Environmental Entomology 1, 123-125.
Hagley, E.A.C., Allen, W.R., 1988. Ground beetles (coleoptera, carabidae) as predators
of the codling moth, Cydia-pomonella (L) (Lepidoptera, Tortricidae). Canadian
Entomologist 120, 917-925.

112

Hagley, E.A.C., Holliday, N.J., Barber, D.R., 1982. Laboratory studies of the food
preferences of some orchard carabids (Coleoptera, Carabidae). Canadian Entomologist
114, 431-437.
Hajek, A.E., 2004. Natural enemies: An introduction to biological control. Cambridge
University Press, Cambridge.
Harrington, D.P., Fleming, T.R., 1982. A class of rank test procedures for censored
survival-data. Biometrika 69, 553-566.
Heinz, K.M., 1998. Dispersal and dispersion of aphids (Homoptera: Aphididae) and
selected natural enemies in spatially subdivided greenhouse environments.
Environmental Entomology 27, 1029-1038.
Holdsworth Jr, R.P., 1970. Codling moth control as part of an integrated program in
Ohio. Journal of Economic Entomology 63, 894-897.
Holling, C.S., 1959. The components of predation as revealed by a study of smallmammal predation of the European pine sawfly. Canadian Ent 91, 293-320.
Horton, D.R., Broers, D.A., Lewis, R.R., Granatstein, D., Zack, R.S., Unruh, T.R.,
Moldenke, A.R., Brown, J.J., 2003. Effects of mowing frequency on densities of natural
enemies in three pacific northwest pear orchards. Entomologia Experimentalis Et
Applicata 106, 135-145.
Hossain, Z., Gurr, G.M., Wratten, S.D., Raman, A., 2002. Habitat manipulation in
lucerne medicago sativa: Arthropod population dynamics in harvested and 'refuge' crop
strips. Journal of Applied Ecology 39, 445-454.
Hoyt, S.C., 1969. Integrated chemical control of insects and biological control of mites
on apple in Washington. Journal of Economic Entomology 62, 74-&.
Huey, R.B., Pianka, E.R., 1981. Ecological consequences of foraging mode. Ecology
62, 991-999.
Huffaker, C.B., 1958. Experimental studies on predation: Dispersion factors and
predator-prey oscillations. Hilgardia 27, 343-383.

113

Hui, E., Webster, J.M., 2000. Influence of insect larvae and seedling roots on the hostfinding ability of Steinernema feltiae (Nematoda: Steinernematidae). Journal of
Invertebrate Pathology 75, 152-162.
Isaacs, R., Tuell, J., Fiedler, A., Gardiner, M., Landis, D., 2009. Maximizing arthropodmediated ecosystem services in agricultural landscapes: The role of native plants.
Frontiers in Ecology and the Environment 7, 196-203.
Jackson, D.L., Jackson, L.L., 2002. The farm as natural habitat. Island Press,
Washington.
Jackson, D.M., 1982. Searching behavior and survival of 1st-instar codling moths.
Annals of the Entomological Society of America 75, 284-289.
Jackson, D.M., Harwood, R.F., 1980. Survival potential of first instars of the codling
moth in laboratory experiments. Annals of the Entomological Society of America 73,
160-163.
Jackson, J.E., Parry, M.S., Stephens, C.P., 1981. Intensification of tree fruit production:
Current constraints, relevant research and an alternative system strategy for the 1980s.
Acta Horticulturae 114, 399-406.
Janssen, A., Sabelis, M.W., Magalhaes, S., Montserrat, M., Van der Hammen, T., 2007.
Habitat structure affects intraguild predation. Ecology 88, 2713-2719.
Jaques, R.P., MacLellan, C.R., 1965. Fungal mortality of overwintering larvae of the
codling moth in apple orchards in Nova Scotia. Journal of Invertebrate Pathology 7,
291-296.
Jaynes, H.A., Marucci, P.E., 1947. Effect of artificial control practices on the parasites
and predators of the codling moth. Journal of Economic Entomology 40, 9-25.
Jenkins, P.E., Isaacs, R., 2007. Reduced-risk insecticides for control of grape berry
moth (Lepidoptera : Tortricidae) and conservation of natural enemies. Journal of
Economic Entomology 100, 855-865.
Jones, V.P., Steffan, S.A., Wiman, N.G., Horton, D.R., Miliczky, E., Zhang, Q.H., Baker,
C.C., 2011. Evaluation of herbivore-induced plant volatiles for monitoring green
lacewings in Washington apple orchards. Biological Control 56, 98-105.

114

Jones, V.P., Unruh, T.R., Horton, D.R., Mills, N.J., Brunner, J.F., Beers, E.H., Shearer,
P.W., 2009. Tree fruit IPM programs in the western united states: The challenge of
enhancing biological control through intensive management. Pest Management Science
65, 1305-1310.
Judd, G.J.R., Gardiner, M.G.T., Thomson, D.R., 1997. Control of codling moth in
organically-managed apple orchards by combining pheromone-mediated mating
disruption, post-harvest fruit removal and tree banding. Entomologia Experimentalis Et
Applicata 83, 137-146.
Jumean, Z., Wood, C., Gries, G., 2009. Frequency distribution of larval codling moth,
Cydia pomonella L., aggregations on trees in unmanaged apple orchards of the Pacific
Northwest. Environmental Entomology 38, 1395-1399.
Kareiva, P., Perry, R., 1989. Leaf overlap and the ability of ladybird beetles to search
among plants. Ecological Entomology 14, 127-129.
Kaya, H.K., Joos, J.L., Falcon, L.A., Berlowitz, A., 1984. Suppression of the codling
moth (Lepidoptera, Olethreutidae) with the entomogenous nematode, Steinernema
feltiae (Rhabditida, Steinernematidae). Journal of Economic Entomology 77, 1240-1244.
Kaya, H.K., Stock, S.P., 1997. Techniques in insect nematology. In: Lacey, L.A., (Ed.),
Manual of techniques in insect pathology. Academic Press, San Diego, pp. 281-324.
Keitt, T.H., Johnson, A.R., 1995. Spatial heterogeneity and anomalous kinetics emergent patterns in diffusion-limited predatory prey interaction. Journal of Theoretical
Biology 172, 127-139.
Kleinbaum, D.G., Klein, M., 2012. Survival analysis. Springer, New York.
Kogan, M., Lattin, J.D., 1993. Insect conservation and pest-management. Biodiversity
and Conservation 2, 242-257.
Koppenhofer, A.M., Grewal, P.S., Kaya, H.K., 2000. Synergism of imidacloprid and
entomopathogenic nematodes against white grubs: The mechanism. Entomologia
Experimentalis Et Applicata 94, 283-293.

115

Kostylev, V.E., Erlandsson, J., Ming, M.Y., Williams, G.A., 2005. The relative
importance of habitat complexity and surface area in assessing biodiversity: Fractal
application on rocky shores. Ecological Complexity 2, 272-286.
Kuhrt, U., Samietz, J., Dorn, S., 2005. Thermoregulation behaviour in codling moth
larvae. Physiological Entomology 30, 54-61.
Lacey, L.A., Arthurs, S.P., Unruh, T.R., Headrick, H., Fritts, R., 2006a.
Entomopathogenic nematodes for control of codling moth (Lepidoptera : Tortricidae) in
apple and pear orchards: Effect of nematode species and seasonal temperatures,
adjuvants, application equipment, and post-application irrigation. Biological Control 37,
214-223.
Lacey, L.A., Chauvin, R.L., 1999. Entomopathogenic nematodes for control of
diapausing codling moth (Lepidoptera : Tortricidae) in fruit bins. Journal of Economic
Entomology 92, 104-109.
Lacey, L.A., Granatstein, D., Arthurs, S.P., Headrick, H., Fritts, R., 2006b. Use of
entomopathogenic nematodes (Steinernematidae) in conjunction with mulches for
control of overwintering codling moth (Lepidoptera : Tortricidae). Journal of
Entomological Science 41, 107-119.
Lacey, L.A., Neven, L.G., Headrick, H.L., Fritts, R., 2005. Factors affecting
entomopathogenic nematodes (Steinerneniatidae) for control of overwintering codling
moth (Lepidoptera : Tortricidae) in fruit bins. Journal of Economic Entomology 98, 18631869.
Lacey, L.A., Shapiro-Ilan, D.I., 2008. Microbial control of insect pests in temperate
orchard systems: Potential for incorporation into IPM. Annual Review of Entomology 53,
121-144.
Lacey, L.A., Unruh, T.R., 2005. Biological control of codling moth (Cydia pomonella,
Lepidoptera: Tortricidae) and its role in integrated pest management, with emphasis on
entomopathogens. Vedalia 12, 33 - 60.
Landis, D.A., Wratten, S.D., Gurr, G.M., 2000. Habitat management to conserve natural
enemies of arthropod pests in agriculture. Annual Review of Entomology 45, 175-201.
Leslie, P.H., 1945. On the use of matrices in certain population mathematics. Biometrika
33, 183-212.
116

Letourneau, D.K., Bothwell, S.G., 2008. Comparison of organic and conventional farms:
Challenging ecologists to make biodiversity functional. Frontiers in Ecology and the
Environment 6, 430-438.
Lewis, E.E., Campbell, J., Griffin, C., Kaya, H., Peters, A., 2006. Behavioral ecology of
entomopathogenic nematodes. Biological Control 38, 66-79.
Losey, J.E., Vaughan, M., 2006. The economic value of ecological services provided by
insects. Bioscience 56, 311-323.
Lovell, J.H., 1915. The origin of anthophily among the Coleoptera. Psyche 22, 67-84.
Luck, G.W., Harrington, R., Harrison, P.A., Kremen, C., Berry, P.M., Bugter, R.,
Dawson, T.P., de Bello, F., Diaz, S., Feld, C.K., Haslett, J.R., Hering, D., Kontogianni,
A., Lavorel, S., Rounsevell, M., Samways, M.J., Sandin, L., Settele, J., Sykes, M.T., van
den Hove, S., Vandewalle, M., Zobel, M., 2009. Quantifying the contribution of
organisms to the provision of ecosystem services. Bioscience 59, 223-235.
Luck, R.F., Shepard, B.M., Kenmore, P.E., 1988. Experimental methods for evaluating
arthropod natural enemies. Annual Review of Entomology 33, 367-391.
MacArthur, R.H., Pianka, E.R., 1966. On optimal use of a patchy environment. The
American Naturalist 100, 603-609.
MacLellan, C.R., 1958. Role of woodpeckers in control of codling moth in Nova Scotia.
The Canadian Entomologist 90, 18-22.
MacLellan, C.R., 1960. Cocooning behaviour of overwintering codling moth larvae. The
Canadian Entomologist 92, 469-479.
MacLellan, C.R., 1962. Mortality of codling moth eggs and young larvae in an integrated
control orchard. The Canadian Entomologist 94, 655-666.
Maclellan, C.R., 1972. Codling moth populations under natural, integrated, and
chemical control on apple in Nova-Scotia (Lepidoptera-Olethreutidae). Canadian
Entomologist 104, 1397-1404.

117

Manktelow, D.W.L., Pratt, J.-P., 1997. The tree-row-volume spraying system and its
potential use in new zealand. 50th New Zealand Plant Protection Conference, pp. 119124.
Marshall, G.E., 1940. Some newly discovered habits of the codling moth. Journal of
Economic Entomology 33, 200.
Marshall, S.A., 2006. Insects: Their natural history and diversity: With a photographic
guide to insects of eastern north america. Firefly Books Inc., Buffalo.
Mathews, C.R., Bottrell, D.G., Brown, M.W., 2004. Habitat manipulation of the apple
orchard floor to increase ground-dwelling predators and predation of Cydia pomonella
(L.) (Lepidoptera : Tortricidae). Biological Control 30, 265-273.
McEwen, P., New, T.R., Whittington, F.B. Eds.), 2001. Lacewings in the crop
environment. Cambridge University Press, Cambridge, UK.
Merfield, C.N., Wratten, S.D., Navntoft, S., 2004. Video analysis of predation by
polyphagous invertebrate predators in the laboratory and field. Biological Control 29, 513.
Michelbacher, A.E., 1952. Walnut insect and spider-mite control in Northern California.
Journal of Economic Entomology 45, 1020-1027.
Mills, N., 2003. Augmentation in orchards: Improving the efficacy of Trichogramma
inundation. In: Van Driesche, R.G., (Ed.), 1st International Symposium on Biological
Control of Arthropods. USDA Forest Service, pp. 130-135.
Mills, N., 2005a. Classical biological control of codling moth: The California experience.
Second International Symposium on Biological Control of Arthropods. USDA Forest
Service, Davos, Switzerland, pp. 126-131.
Mills, N., 2005b. Selecting effective parasitoids for biological control introductions:
Codling moth as a case study. Biological Control 34, 274-282.
Mills, N.J., Carl, K.P., 1991. Parasitoids and predators. In: van der Geest, L.P.S.,
Evenhuis, H.H., Eds.), Tortricid pests. Elsevier, New York, pp. 235-252.

118

Minarro, M., Espadaler, X., Melero, V.X., Suarez-Alvarez, V., 2009. Organic versus
conventional management in an apple orchard: Effects of fertilization and tree-row
management on ground-dwelling predaceous arthropods. Agricultural and Forest
Entomology 11, 133-142.
Morris, C., Winter, M., 1999. Integrated farming systems: The third way for european
agriculture? Land Use Policy 16, 193-205.
Mota-Sanchez, D., Wise, J.C., Poppen, R.V., Gut, L.J., Hollingworth, R.M., 2008.
Resistance of codling moth, Cydia pomonella (L.) (Lepidoptera: Tortricidae), larvae in
Michigan to insecticides with different modes of action and the impact on field residual
activity. Pest Management Science 64, 881-890.
Murdoch, W.W., 1975. Diversity, complexity, stability and pest control. Journal of
Applied Ecology 12, 795-807.
Navaneethan, T., Strauch, O., Besse, S., Bonhomme, A., Ehlers, R.U., 2010. Influence
of humidity and a surfactant-polymer-formulation on the control potential of the
entomopathogenic nematode Steinernema feltiae against diapausing codling moth
larvae (Cydia pomonella L.) (Lepidoptera: Tortricidae). BioControl 55, 777-788.
Neumann, G., Shields, E.J., 2006. Interspecific interactions among three
entomopathogenic nematodes, steinemema carpocapsae weiser, S. Feltiae Filipjev,
and Heterorhabditis bacteriophora Poinar, with different foraging strategies for hosts in
multipiece sand columns. Environmental Entomology 35, 1578-1583.
Newcomer, E.J., 1920. Winterkilling of codling moth larvae. Journal of Economic
Entomology 13, 441-443.
Nielsen, A.L., Lewis, E.E., 2012. Designing the ideal habitat for entomopathogen use in
nursery production. Pest Management Science 68, 1053-1061.
Norris, R.F., Kogan, M., 2005. Ecology of interactions between weeds and arthropods.
Annual Review of Entomology 50, 479-503.
Oatman, E.R., 1966. Studies on integrated control of apple pests. Journal of Economic
Entomology 59, 368-&.

119

Ohnesorg, W.J., Johnson, K.D., O'Neal, M.E., 2009. Impact of reduced-risk insecticides
on soybean aphid and associated natural enemies. Journal of Economic Entomology
102, 1816-1826.
Opitz, W., 2002. Flower foraging behavior of the Australian species Eleale aspera
(Newman) (Coleoptera: Cleridae: Clerinae). The Coleopterists Bulletin 56, 241-245.
Paul, N.D., Gwynn-Jones, D., 2003. Ecological roles of solar UV radiation: Towards an
integrated approach. Trends in Ecology & Evolution 18, 48-55.
Pfiffner, L., Wyss, E., Gurr, G.M., Wratten, S., Altieri, M.A., 2004. Use of sown
wildflower strips to enhance natural enemies of agricultural pests. Ecological
engineering for pest management. Cornell University Press, Ithaca, pp. 165-186.
Prokopy, R.J., 1994. Integration in orchard pest and habitat management - a review.
Agriculture Ecosystems & Environment 50, 1-10.
R-core-development-Team, 2009. R: A language and environment for statistical
computing. R Foundation for Statistical Computing, Vienna, Austria.
Radcliffe, E.B., Hutchinson, W.D., Cancelado, R.E. Eds.), 2009. Integrated Pest
Management: Concepts, tactics, strategies, and case studies. Cambridge Universiy
Press, New York.
Randlkofer, B., Obermaier, E., Casas, J., Meiners, T., 2010. Connectivity counts:
Disentangling effects of vegetation structure elements on the searching movement of a
parasitoid. Ecological Entomology 35, 446-455.
Renn, N., 1998. Routes of penetration of the entomopathogenic nematode Steinernema
feltiae attacking larval and adult houseflies (Musca domestica). Journal of Invertebrate
Pathology 72, 281-287.
Riddick, E.W., Mills, N. J., 1994. Potential of adult carabids (Coleoptera: Carabidae) as
predators of fifth-instar codling moth (Lepidoptera: Tortricidae) in apple orchards in
California. Environmental Entomology 23, 1338-1345.
Rolston, A.N., Griffin, C.T., Downes, M.J., 2006. Emergence and dispersal patterns of
two isolates of the entomopathogenic nematode Steinernema feltiae. Journal of
Nematology 38, 221-228.

120

Schroeder, P.C., Villani, M.G., Ferguson, C.S., Nyrop, J.P., Shields, E.J., 1993.
Behavioral interactions between japanese-beetle (Coleoptera, Scarabaeidae) grubs and
an entomopathogenic nematode (Nematoda, Heterorhabditidae) within turf microcosms.
Environmental Entomology 22, 595-600.
Shapiro-Ilan, D.I., Lewis, E.E., Campbell, J.F., Kim-Shapiro, D.B., 2012. Directional
movement of entomopathogenic nematodes in response to electrical field: Effects of
species, magnitude of voltage, and infective juvenile age. Journal of Invertebrate
Pathology 109, 34-40.
Shapiro-Ilan, D.I., Stuart, R., McCoy, C.W., 2003. Comparison of beneficial traits among
strains of the entomopathogenic nematode, Steinernema carpocapsae, for control of
Curculio caryae (Coleoptera : Curculionidae). Biological Control 28, 129-136.
Sherman III, F., 1936. Eight seasons' tests in codling moth control in Michigan. Journal
of Economic Entomology 29, 653-655.
Simon, S., Bouvier, J.C., Debras, J.F., Sauphanor, B., 2010. Biodiversity and pest
management in orchard systems. A review. Agronomy for Sustainable Development 30,
139-152.
Simon, S., Defrance, H., Sauphanor, B., 2007. Effect of codling moth management on
orchard arthropods. Agriculture Ecosystems & Environment 122, 340-348.
Simpson, C.B., 1903. The codling moth. Government Printing Office, Washingon D.C.
Solomon, M.E., 1991. Ecology and population dynamics. In: van der Geest, L.P.S.,
Evenhuis, H.H., Eds.), Tortricid pests. Elsevier, New York, pp. 209-234.
Solomon, M.E., Glen, D.M., 1979. Prey density and rates of predation by tits (parus spp)
on larvae of codling moth (Cydia-pomonella) under bark. Journal of Applied Ecology 16,
49-59.
Solomon, M.E., Glen, D.M., Kendall, D.A., Milsom, N.F., 1976. Predation of
overwintering larvae of codling moth (Cydia pomonella (L.)) by birds. Journal of Applied
Ecology 13, 341-352.
Southwood, T.R.E., 1978. Ecological methods with particular reference to the study of
insect populations. Chapman and Hall, London.

121

Spence, K.O., Lewis, E.E., Perry, R.N., 2008. Host-finding and invasion by
entomopathogenic and plant-parasitic nematodes: Evaluating the ability of laboratory
bioassays to predict field results. Journal of Nematology 40, 93-98.
Stehr, F.W., 1991. Immature insects. Kendall/Hunt, Dubuque.
Steiner, L.F., 1929. Miscellaneous codling moth studies. Journal of Economic
Entomology 22, 648-654.
Steiner, L.F., Ackerman, A.J., 1936. Large-scale test of orchard sanitation to control
codling moth Journal of Economic Entomology 29, 648-653.
Stern, V.M., Smith, R.F., van den Bosch, R., Hagen, K.S., 1959. The integrated control
concept. Hilgardia 29, 81 - 101.
Stoeckli, S., Mody, K., Dorn, S., 2008. Influence of canopy aspect and height on codling
moth (Lepidoptera : Tortricidae) larval infestation in apple, and relationship between
infestation and fruit size. Journal of Economic Entomology 101, 81-89.
Straub, C.S., Finke, D.L., Snyder, W.E., 2008. Are the conservation of natural enemy
biodiversity and biological control compatible goals? Biological Control 45, 225-237.
Stuart, R.J., El-Bopai, F.E., Duncan, L.W., 2008. From augmentation to conservation of
entomopathogenic nematodes: Trophic cascades, habitat manipulation and enhanced
biological control of diaprepes abbreviatus root weevils in florida citrus groves. Journal
of Nematology 40, 73-84.
Subinprasert, S., 1987. Natural enemies and their impact on overwintering codling moth
populations (Laspeyresia-pomonella L) (Lep, Tortricidae) in south Sweden. Journal of
Applied Entomology-Zeitschrift Fur Angewandte Entomologie 103, 46-55.
Summerland, S.A., Steiner, L.F., 1943. Codling moth oviposition and fate of eggs.
Journal of Economic Entomology 36, 72-75.
Suverkropp, B.P., Bigler, F., van Lenteren, J.C., 2010. Movement and host finding of
Trichogramma brassicae on maize plants. Bulletin of Insectology 63, 115-127.

122

Therneau, T., Lumley, T., 2009. Survival: Survival analysis, including penalised
likelihood. http://CRAN.R-project.org/package=survival.
Thomas, M.B., 1999. Ecological approaches and the development of "truly integrated"
pest management. Proceedings of the National Academy of Sciences of the United
States of America 96, 5944-5951.
Thomas, M.B., Wratten, S.D., Sotherton, N.W., 1991. Creation of 'island' habitats in
farmland to manipulate populations of beneficial arthropods: Predator densities and
emigration. Journal of Applied Ecology 28, 906-917.
Thomson, L.J., Hoffmann, A.A., 2009. Vegetation increases the abundance of natural
enemies in vineyards. Biological Control 49, 259-269.
Thomson, L.J., Hoffmann, A.A., 2013. Spatial scale of benefits from adjacent woody
vegetation on natural enemies within vineyards. Biological Control 64, 57-65.
Unruh, T.R., Lacey, L.A., 2001. Control of codling moth, Cydia pomonella (Lepidoptera :
Tortricidae), with Steinernema carpocapsae: Effects of supplemental wetting and
pupation site on infection rate. Biological Control 20, 48-56.
Unruh, T.R., Pfannenstiel, R.S., Peters, C., Brunner, J.F., Jones, V.P., 2012. Parasitism
of leafrollers in Washington fruit orchards is enhanced by perimeter plantings of rose
and strawberry. Biological Control 62, 162-172.
USDA-NASS, 2008. Agricultural chemical usage 2007 field crops summary. United
States Department of Agriculture, pp. 120.
USDA-NASS, 2012a. Noncitrus fruits and nuts. United States Department of Agriculture
- National Agricultural Statistics Service, pp. 79.
USDA-NASS, 2012b. Quick stats United States Department of Agriculture.
van der Geest, L.P.S., Evenhuis, H.H. Eds.), 1991. Tortricid pests. Elsevier, New York.
Van Driesche, R.G., Bellows, T.S., 1996. Biological control. Chapman & Hall, New York.

123

Van Driesche, R.G., Mason, J.L., Wright, S.E., Prokopy, R.J., 1998. Effect of reduced
insecticide and fungicide use on parasitism of leafminers (Phyllonorycter spp.)
(Lepidoptera : Gracillariidae) in commercial apple orchards. Environmental Entomology
27, 578-582.
van Emden, H.F., Williams, G.F., 1974. Insect stability and diversity in agroecosystems. Annual Review of Entomology 19, 455-475.
Vandenbosch, R., Stern, V.M., 1962. Integration of chemical and biological control of
arthropod pests. Annual Review of Entomology 7, 367-&.
Vreysen, M.J.B., Robinson, A.S., 2011. Ionising radiation and area-wide management
of insect pests to promote sustainable agriculture. A review. Agronomy for Sustainable
Development 31, 233-250.
Wagenmakers, P.S., 1991. Planting systems for fruit-trees in temperate climates.
Critical Reviews in Plant Sciences 10, 369-385.
Wallace, H.R., 1958. Movement of eelworms. Annals of Applied Biology 46, 74-85.
Walton, N.J., 2009. Habitat management using native flowering perennials to increase
beneficial insects in Michigan highbush blueberry. Entomology. Michigan State
University, Lansing, pp. 115.
Walton, N.J., 2013a. Codling moth (Cydia pomonella [L.]) natural enemies in apple
orchards: Over a century of research. The role of natural enemies in the management of
codling moth (Cydia pomonella [L.]) in organic apple orchards in Michigan.
Walton, N.J., 2013b. An exploration of how moisture and structural complexity may
affect the foraging success of Steinernema feltiae Filipjev against codling moth (Cydia
pomonella [L.]). The role of natural enemies in pest management of codling moth (Cydia
pomonella [L.]) in organic apple orchards in Michigan.
Walton, N.J., 2013c. Habitat characteristics influencing foraging success of codling
moth antagonists in apple orchards (Cydia pomonella [L.]) The role of natural enemies
in the management of codling moth (Cydia pomonella [L.]) in organic apple orchards in
Michigan.

124

Walton, N.J., 2013d. Video observations of the natural enemies of eggs and fifth instar
codling moth Cydia pomonella (L.) in Michigan apple orchards (USA). The role of
natural enemies in the management of codling moth (Cydia pomonella [L.]) in organic
apple orchards in Michigan.
Walton, N.J., Isaacs, R., 2011a. Influence of native flowering plant strips on natural
enemies and herbivores in adjacent blueberry fields. Environmental Entomology 40,
697-705.
Walton, N.J., Isaacs, R., 2011b. Survival of three commercially-available natural
enemies exposed to Michigan wildflowers. Environmental Entomology 40, 1177 - 1182.
Wearing, C.H., 1975. Integrated control of apple pests in New Zealand 2. Field
estimation of fifth-instar larval and pupal mortalities of codling moth by tagging with
cobalt-58. New Zealand Journal of Zoology 2, 135-149.
Wilby, A., Thomas, M.B., 2002. Natural enemy diversity and pest control: Patterns of
pest emergence with agricultural intensification. Ecology Letters 5, 353-360.
Wise, J.C., Jenkins, P.E., Schilder, A.M.C., Vandervoort, C., Isaacs, R., 2010. Sprayer
type and water volume influence pesticide deposition and control of insect pests and
diseases in juice grapes. Crop Protection 29, 378-385.
Witzgall, P., Stelinski, L., Gut, L., Thomson, D., 2008. Codling moth management and
chemical ecology. Annual Review of Entomology 53, 503-522.
Zhang, W., Swinton, S.M., 2009. Incorporating natural enemies in an economic
threshold for dynamically optimal pest management. Ecological Modelling 220, 13151324.

125