PHYTOSEIIDS AS BIOLOGICAL CONTROL AGENTS OF PHYTOPHAGOUS MITES IN WASHINGTON APPLE ORCHARDS

Transcription

1 PHYTOSEIIDS AS BIOLOGICAL CONTROL AGENTS OF PHYTOPHAGOUS MITES IN WASHINGTON APPLE ORCHARDS By REBECCA ANN SCHMIDT-JEFFRIS A dissertation submitted in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY WASHINGTON STATE UNIVERSITY Department of Entomology MAY 2015 Copyright by REBECCA ANN SCHMIDT-JEFFRIS, 2015 All Rights Reserved

2 Copyright by REBECCA ANN SCHMIDT-JEFFRIS, 2015 All Rights Reserved

3 To the Faculty of Washington State University: The members of the Committee appointed to examine the dissertation of REBECCA ANN SCHMIDT-JEFFRIS find it satisfactory and recommend that it be accepted. Elizabeth H. Beers, Ph.D., Chair David W. Crowder, Ph.D. Richard S. Zack, Ph.D. Thomas R. Unruh, Ph.D. Nilsa A. Bosque-Pérez, Ph.D. ii

4 ACKNOWLEDGEMENT I would like to thank Dr. Elizabeth Beers for giving me the opportunity to work in her lab and for several years of exceptional mentoring. She has provided me with an excellent experience and is an outstanding role model. I would also like to thank the other members of my committee, Drs. Thomas Unruh, David Crowder, Nilsa Bosque-Pérez, and Richard Zack for comments on these (and other) manuscripts, and invaluable advice throughout my graduate career. Additionally, I thank the entomology faculty of Washington State University and the University of Idaho for coursework that acted as the foundation for this degree, especially Dr. Sanford Eigenbrode and Dr. James Ding Johnson. I also thank Dr. James McMurtry, for input on manuscripts and identification confirmation of mite specimens. I would like to acknowledge the assistance I received in conducting these experiments from our laboratory technicians, Bruce Greenfield and Peter Smytheman, my labmate Alix Whitener, and the many undergraduate technicians that helped collect data: Denise Burnett, Allie Carnline, David Gutiérrez, Kylie Martin, Benjamin Peterson, Mattie Warner, Alyssa White, and Shayla White. I would especially like to acknowledge my assistants, Jordan Takasugi and Kaitlin Parsons, without which most of this work could not have been done. I also thank Drs. Bahman Shafii, William Price, Mark Evans, and David Crowder for statistical advice. I would also like to express my gratitude to the many cooperating fieldmen and growers who allowed me to work in their orchards. I appreciate the support (both emotional and financial) that I have received from my mother, father, and sister while completing this degree and for encouraging my love of learning and science. I would also like to thank my in-laws, the entomology graduate students in Pullman, and Emily Wolfe for providing me with a family while away from home. Finally, I would like to iii

5 thank my husband, Dennis Jeffris, for being an incredible listener and a constant source of encouragement. iv

6 PHYTOSEIIDS AS BIOLOGICAL CONTROL AGENTS OF PHYTOPHAGOUS MITES IN WASHINGTON APPLE ORCHARDS Abstract by Rebecca Ann Schmidt-Jeffris, Ph.D. Washington State University May 2015 Chair: Elizabeth H. Beers The integrated mite management (IMM) program in Washington has depended on the biological control of spider mites provided by Galendromus occidentalis (Nesbitt). We explored methods of improving this program. A series of inundative releases of G. occidentalis was performed in commercial apple orchards. None of the releases increased G. occidentalis numbers or reduced pest spider mite populations. It was concluded that a cost prohibitive number of predators would be necessary to maintain pest mites below economic thresholds, emphasizing the importance of conservation biological control. Phytoseiids in apple orchards across Washington were sampled to determine diversity and elucidate factors affecting abundances. While the community was expected to be completely dominated by G. occidentalis, Amblydromella caudiglans (Schuester) was also highly abundant. G. occidentalis abundance was positively affected by conventional pesticide regimes and bifenazate use, whereas A. caudiglans was negatively affected by bifenazate and positively affected by herbicide strip weediness. This indicated that A. caudiglans was more susceptible to disruptive pesticide inputs than G. v

7 occidentalis. This was confirmed by an assay wherein several orchard pesticides caused higher mortality of A. caudiglans than G. occidentalis. The recent phase out of certain orchard pesticides may have allowed A. caudiglans to become more common. To understand the mite species complex in the absence of pesticides, an insecticide-free research orchard was monitored at regular intervals during two growing seasons. Generalist phytoseiids were more common than G. occidentalis throughout both seasons. These species were nearly dependent on Aculus schlechtendali (Nalepa) as a food source, as spider mites remained scarce. This emphasizes the role of spider mites as induced pests and suggests that in the absence of pesticide applications, a complement of generalist predators is capable of maintaining spider mite populations at very low densities. G. occidentalis was also found to be less affected by leaf surfaces than generalist phytoseiids examined in previous studies. These findings indicate that G. occidentalis is biological different from the generalist phytoseiids that may replace it as pesticide use changes. Therefore, IMM must constantly be re-evaluated in order to meet orchardists needs. vi

8 TABLE OF CONTENTS Page ACKNOWLEDGEMENT... iii-iv ABSTRACT... v-vi LIST OF TABLES...x LIST OF FIGURES... xii CHAPTER ONE: REVIEW OF BIOLOGICAL CONTROL PRINCIPLES USING PHYTOSEIIDS AS MODELS, WITH SPECIAL REFERENCE TO INTEGRATED MITE MANAGEMENT IN WASHINGTON APPLE... 1 Introduction... 1 Phytoseiids as Model Biological Control Agents... 3 Classical Biological Control... 5 Augmentative Biological Control Conservation Biological Control Integrated Mite Management in Washington Apple References Cited CHAPTER TWO: RELEASES OF INSECTARY-REARED GALENDROMUS OCCIDENTALIS (ACARI: PHYTOSEIIDAE) IN COMMERCIAL APPLE ORCHARDS Abstract Introduction Materials and Methods Results vii

9 Discussion References Cited CHAPTER THREE: ASSESSMENT OF MARKING TECHNIQUES FOR THE PURPOSE OF MONITORING PHYTOSEIID RELEASES Abstract Introduction Materials and Methods Results and Discussion References Cited CHAPTER FOUR: PHYTOSEIIDS IN WASHINGTON COMMERCIAL APPLE ORCHARDS: BIODIVERSITY AND FACTORS AFFECTING ABUNDANCE Abstract Introduction Materials and Methods Results Discussion References Cited CHAPTER FIVE: COMPARATIVE BIOLOGY AND PESTICIDE SUSCEPTIBLITY OF AMBLYDROMELLA CAUDIGLANS AND GALENDROMUS OCCIDENTALIS AS SPIDER MITE PREDATORS IN APPLE ORCHARDS Abstract Introduction Materials and Methods viii

10 Results and Discussion References Cited CHAPTER SIX: PHENOLOGY AND STRUCTURE OF A PHYTOSEIID COMMUNITY IN AN INSECTIDE-FREE APPLE ORCHARD Abstract Introduction Materials and Methods Results and Discussion References Cited CHAPTER SEVEN: EFFECTS OF APPLE TRICHOMES ON PHYTOSEIIDS: PREFERENCES, FECUNDITY, PREY CONSUMPTION, AND POPULATION DENSITY 175 Abstract Introduction Materials and Methods Results Discussion References Cited APPENDICES Appendix 4.1 Data collected with each sample Appendix 4.2 Presence of non-phytoseiid mites in each sample Appendix 4.3 Data collected via survey of pest consultants Appendix 4.4 Phytoseiids collected at each sample location ix

11 LIST OF TABLES Table 2.1. Seasonal mite densities (cumulative mite days) in commercial apple orchards in Washington following release of various rates of G. occidentalis Table 2.2. Mortality, prey consumption, fecundity, egg hatch and larval survival of G. occidentalis placed on apple leaves with and without pesticide residues Table 2.3. Mortality, prey consumption, fecundity, egg hatch and larval survival of G. occidentalis from a commercial insectary and a laboratory culture Table 3.1. Results of the marker effects bioassay Table 3.2. Percent detection and percent of false negatives for each treatment at each DAT evaluated Table 4.1. Species composition and dominance of phytoseiids at 102 sampled apple blocks Table 5.1. Pesticides used in nontarget effects bioassays Table 5.2. Mean (± SE) life stage duration of A. caudiglans and G. occidentalis (d) Table 5.3. Stage-specific survival and sex ratio of A. caudiglans and G. occidentalis Table 5.4. Prey consumption and oviposition by A. caudiglans and G. occidentalis fed on two different diets Table 5.5. Toxicity of orchard pesticides to A. caudiglans and G. occidentalis Table 5.6. Repellency of orchard pesticides to A. caudiglans Table 6.1. Percentage of phytoseiid species that were female on the 10 dates with the highest population of each species Table 7.1. Mean proportion of individuals on the Delicious side of arenas without leaf midribs included x

12 Table 7.2. Mean proportion of T. urticae eggs laid, G. occidentalis eggs laid, and T. urticae eggs consumed by G. occidentalis on Delicious side of arenas, without midribs included Table 7.3. Mean proportion of individuals on the Delicious side of arenas with leaf midribs included Table 7.4. Mean proportion of T. urticae eggs laid, G. occidentalis eggs laid, and T. urticae eggs consumed by G. occidentalis on Delicious side of arenas, with midribs included Table 7.5. Mean proportion of G. occidentalis on the side of arenas with cotton added at each evaluation interval Table 7.6. Mean proportion of G. occidentalis eggs laid and T. urticae eggs consumed on the side of arenas with cotton added Table 7.7. Mean counts of trichomes, pollen, and mites for Delicious and Golden Delicious leaves from an insecticide/fungicide-free orchard xi

13 LIST OF FIGURES Fig Plot layout for mite releases; a single replicate of a single treatment shown Fig Mean (± SEM) seasonal numbers of mites per leaf for Exp Fig Regression of in situ vs. brushed mite counts Fig Mean (± SEM) seasonal numbers of mites per leaf for Exp Fig Mean (± SEM) seasonal numbers of mites per leaf for Exp Fig Galendromus occidentalis under UV light after marking with fluorescent powder Fig Duration of fluorescent powder and egg white powder marks represented as percentage marked over time Fig Locations of sampled commercial apple blocks Fig Mean G. occidentalis abundance and a) bifenazate use; df = 1, χ 2 = 3.89, P = b) pesticide intensity; df = 4, χ 2 = 11.57, P = Fig Mean A. caudiglans abundance and a) bifenazate use; df = 1, χ 2 = 13.50, P = b) herbicide strip weediness ranking; df = 3, χ 2 = 18.21, P = c) cultivar; df = 4, χ 2 = 10.96, P = Fig Number of sites with G. occidentalis present/absent and pesticide intensity Fig Number of sites with A. caudiglans present/absent and cultivar Fig G. flumenis abundance in an insecticide-free orchard over time for two years Fig A. caudiglans abundance in an insecticide-free orchard over time for two years Fig K. corylosus abundance in an insecticide-free orchard over time for two years Fig G. occidentalis abundance in an insectide-free orchard over time for two years Fig Z. mali abundance in an insecticide-free orchard over time for Fig A. schlechtendali abundance in an insecticide-free orchard over time for xii

14 Fig Mean trichome counts per 3 mm reference line and 95% asymmetric confidence intervals for each side of leaf half arenas without midribs included Fig Mean trichome counts per 3 mm reference line and 95% asymmetric confidence intervals for each side of leaf half arenas with midribs included Fig a) Mean T. urticae eggs consumed and b) eggs laid per G. occidentalis female on whole leaf disks (no choice) of two apple cultivars xiii

15 CHAPTER ONE: REVIEW OF BIOLOGICAL CONTROL PRINCIPLES USING PHYTOSEIIDS AS MODELS, WITH SPECIAL REFERENCE TO INTEGRATED MITE MANAGEMENT IN WASHINGTON APPLE INTRODUCTION Arthropods have great potential to damage the food crops produced by humans (Geier 1966). Because of their direct competition with us for food, the development of agriculture also required the development of pest control techniques. Agriculture arose as a consequence of hunter-gatherer societies beginning to live together in larger, more permanent groups. Many hypotheses about agriculture center on (1) the need for these societies to produce larger amounts of food to feed all members and (2) that early crops may have begun as ruderals capable of surviving the disturbed habitats around human dwellings or growing from plant leavings in rubbish heaps (Chrispeels and Sadava 1994). As agriculture developed, members of societies specialized into trades, with farmers producing most of the food. Crop plants were domesticated, first unintentionally, then through deliberate breeding (Chrispeels and Sadava 1994). Traits that were selected allowed for easier harvest, higher yield, better taste, and decreased dispersal ability (Chrispeels and Sadava 1994). These changes, and the development of farming technology, allowed for the production of monocultures, where single crops were grown over a wide area, often in consecutive years (Miller and Spoolman 2009). It was this environment that allowed pests to become significant problems. Monocultures concentrate a single plant resource, favoring specialist pests and allowing them to develop large populations (resource concentration hypothesis) (Price and 1

16 Waldbauer 1982, Schoonhoven et al. 2005). This also disadvantages natural enemies that require non-crop resources (discussed later). Some of the traits of domesticated plants also left them more susceptible to pest attack than their wild-type relatives (Purseglove 1968, Hawkes 1969, Rosenthal and Welter 1995). Early pest control (pre-1940) involved minimal technology, and included practices like crop rotation, hand removal of pests, and eventually applications of very few insecticidal materials (sulfur, nicotine, lead arsenate, petroleum products, soaps) (Smith and Secoy 1975,1976). The development of synthetic pesticides in the 1940s, shifted the pest management focus to rely almost solely on these materials for the control of agricultural pests (Debach 1974). In contrast to the relatively small arsenal of products available before this time, there was a veritable explosion of new materials, and initially efficacy was extremely high, with pesticides available for a broad range of arthropod taxa. The new synthetic pesticides were initially regarded in the light of miracle cures that would end hunger and arthropod-vectored diseases. However, the negative effects of sole reliance on pesticides for pest control were soon felt. Nontarget effects and outbreaks of secondary effects were noted for DDT even before it came into widespread use. The application of additional pesticides to control what had formerly been a secondary pest became known as the pesticide treadmill (Stern et al. 1959, van den Bosch et al. 1982a). Environmental effects (especially those occurring off the application site) became increasingly noticed, culminating in the publication of Silent Spring (Carson 1962, Krupke et al. 2007), an indictment of broad-spectrum pesticide use. Lastly, human health effects were implicated in the use of pesticides, both in terms of safety of agricultural workers, and consumers of treated products (Geier 1966, Debach 1974, Barfield and Stimac 1980, Metcalf 1980). 2

17 The confluence of these events lead to an alternative approach to agricultural pest control. This led to the development of integrated pest management (IPM) as the predominate philosophy used for pest control tactics (Stern et al. 1959, Barfield and Stimac 1980, Metcalf 1980, Luckmann and Metcalf 1982). IPM still allows for the use of chemical controls, but emphasizes the need to understand the ecology of the pest species, including its natural enemies (Stern et al. 1959, Luckmann and Metcalf 1982). Biological control can be defined as the action of parasites, predators, and pathogens in maintaining another organism's density at a lower average than would occur in their absence (DeBach 1964, van den Bosch et al. 1982b). It is important to note that most traditional definitions of biological control do not include biologically-based control tactics that do not use natural enemies of the pest (e.g., sterile male releases, host plant resistance, pheromone technology) (Roush et al. 1980). With IPM, biological control became the centerpiece of pest management. The use of economic thresholds (Stern et al. 1959) permitted for some damage to be acceptable on crop plants (Barfield and Stimac 1980). Without this development, preventative or calendar sprays did not allow for natural enemies to exist in a crop; this is both due to harmful, nontarget effects of pesticides on natural enemies, and the lack of available prey (Debach 1974). Additionally, the emphasis of IPM on applied ecology encouraged investigation into the interactions between pest organisms and their natural enemies. PHYTOSEIIDS AS MODEL BIOLOGICAL CONTROL AGENTS Phytoseiids are the best-known and arguably the most important group of predatory mites due to their ability to control populations of spider mites (Gerson et al. 2003a). They have a nearworldwide distribution and are found in a wide range of climates (Hoy 2011a). Although this 3

18 family has long been recognized for its role in biological control of mites (Parrott et al. 1906, Ewing 1914), extensive research on phytoseiid ecology was not performed until after World War II. Spider mites became serious pests of many crops following the synthetic pesticide boom in the 1940s-1950s (Huffaker et al. 1970, McMurtry et al. 1970). These pests pierce individual cells of leaves and remove the contents, decreasing photosynthetic capacity, reducing plant vigor and crop yield (Huffaker and McMurtry 1969). Plant response can vary considerably in severity; leaf bronzing is a common outcome of mite feeding, but in more sensitive crops (almonds, certain pear cultivars), leaf necrosis or abscission results from even moderate levels of damage (Huffaker and McMurtry 1969, Rice and Jones 1978, Hoyt 1991, Beers and Hoyt 1993). Spider mite outbreaks occur worldwide on many different crops; these outbreaks have been attributed to two key factors: (1) development of resistance to pesticides by spider mites and (2) harmful nontarget effects of pesticide applications on spider mite natural enemies (especially phytoseiids) (Huffaker et al. 1970, McMurtry et al. 1970, Debach 1974, Gerson and Cohen 1989). Spider mites are induced pests, which without perturbation, remain at low densities due the action of natural enemies. Upon realizing the important service provided by phytoseiid mites, research began to focus on how to better use these predators for biological control. This included their introduction, conservation, and release (McMurtry 1981, Hoy 2011a). Because of the variety of research conducted on this family, they make excellent models for highlighting important concepts in biological control. Phytoseiids possess several other characteristics that make them ideal model organisms for studying applied ecology and biological control. Acarine systems have all trophic levels represented, except for primary consumers (Gerson et al. 2003c). Within the mites, over half of all natural enemy studies have been conducted on phytoseiids (Gerson 1998). Tetranychids, their 4

19 primary prey, are more likely to be successfully controlled below economic thresholds than other pests because they are typically indirect pests; tolerance of damage by indirect pests is typically much higher than for direct pests (Turnbull and Chant 1961, Stehr 1982). Phytoseiids are a highly diverse group of predators, making it possible to study both specialists and generalists (McMurtry and Croft 1997, McMurtry et al. 2013). Their small size makes both laboratory assays and mesocosm studies feasible with a modest degree of resources. Phytoseiids were the first natural enemies discovered to have developed field-selected resistance to pesticides (Georghiou 1972), and were the first natural enemies to undergo genetic modification (Hoy 1985). The remainder of this review will focus on how phytoseiids have been used to examine key concepts in biological control. CLASSICAL BIOLOGICAL CONTROL Classical biological control involves the introduction of a species outside of its native range for the control of a pest species (Howarth 1991). Usually, this area focuses on the introduction of natural enemies of invasive species that have escaped their enemies when transported to new regions of the world (Simberloff and Stiling 1996). However, there has also been success in introducing exotic natural enemies for the control of native pests (Debach 1974). In mites, the exemplar of classical biological control introductions is Phytoseiulus persimilis Athias-Henriot. This species is native to Libya, Tunisia, France, and Italy, but is now widespread due to introductions worldwide (McMurtry 1981). While it is typically introduced into greenhouses for control of Tetranychus spp. (usually Tetranychus urticae Koch) in cucumbers, peppers, tomatoes, strawberries and ornamentals, it has also been successfully deployed in outdoor strawberry, hops, corn, and ornamental plots (Oatman et al. 1968, Oatman et 5

20 al. 1977, McMurtry 1981, Pickett and Gilstrap 1986, Trumble and Morse 1993, Drukker et al. 1997, McMurtry and Croft 1997, Opit et al. 2004). This mite has many of the attributes of a good natural enemy, including: fast dispersal, close physical association with pest colonies, high levels of prey consumption, prey specificity, and a high reproductive rate (McMurtry 1981, Stehr 1982, Gilstrap and Friese 1985, McMurtry and Croft 1997). However, it also has weaknesses, mainly in that it prefers humid environments and will frequently completely eliminate pest populations, driving its own populations to local extinction (Gerson et al. 2003a). These traits are likely why its primary use is in greenhouses, where climate can be controlled and the pest in first method can be used to ensure that it has a supply of prey (McMurtry 1981). It is unlikely that this highly specialized predator would be able to persist in perennial cropping systems (Gerson et al. 2003a). Measuring control success. There are a variety of ways to measure success of biological control agent releases (or native predators). These can include comparing pest levels of check plots to areas where natural enemies are left undisturbed. The checks can be in the form of (1) enclosing predator-free space (exclusion cage), (2) use of an insecticide that is toxic to the predator, but not the pest (insecticidal check or chemical exclusion), (3) promotion of ant populations that tend the pest and remove predators (biological check), or (4) hand removal of predators (Debach 1974). All of these measures have been used to study mite biological control (Fleschner 1956, Huffaker and Kennett 1969). However, these methods all pose problems. Use of physical barriers to exclude predators can create microclimatic conditions that favor increased pest populations (Fleschner 1956). The use of some insecticides has been demonstrated to have stimulatory effects on spider mites, increasing their reproduction (Luckey 1968, Huffaker and McMurtry 1969). The biological check is known to introduce many confounding variables (number of ants can vary, promotion of fungal disease-causing honeydew) (Fleschner 1956, 6

21 Debach 1974). Hand removal has been used with success (Fleschner 1956), but is timeconsuming and difficult, due to the small size of the organisms. This is likely why mite biological control is almost exclusively studied via comparisons of release vs. no release plots. Plots are compared for both the number of predators and the number of pests; some studies will also compare crop damage between plots. Number of species to release - biodiversity. There is some debate as to the effect of introducing single or multiple species of predators for biological control. Proponents of the more is better hypothesis indicate that if two biocontrol agents are ecological homologues, the more competitive biocontrol agent in a given environment will replace the less competitive agent (and that the more competitive species will be more effective at controlling the pest) (DeBach 1966), or they might occupy subtly different niches and coexist (Straub and Snyder 2008). Like biodiversity in natural systems, multiple predators can increase the stability of ecosystems by decreasing the risk that the loss of one will result in dramatic herbivore population increases (insurance hypothesis) (Yachi and Loreau 1999). Moreover, some studies have found that natural enemy biodiversity in and of itself does not promote biological control. Instead, diverse enemy populations are more likely to contain the most voracious control agent for a given pest (sampling effect) (Straub and Snyder 2006). Introducing multiple species increases the likelihood that at least one will establish and allows for the possibility of additive effects if more than one species establishes (Hassell and Varley 1969, Stehr 1982). Predators that occupy distinct niches may even act synergistically to improve biological control (Losey and Denno 1998, Sih et al. 1998). However, predator biodiversity and biological control may be incompatible with each other (Straub et al. 2008). Specifically, intraguild competition and predation may lead to 7

22 decreased predation on pests (Sih et al. 1998, Finke and Denno 2004, Casula et al. 2006). Generalist predators may disrupt the biological control of more specific natural enemies (Snyder and Ives 2001), making it difficult to generalize a negative or positive role for biodiversity in biological control (Casula et al. 2006, Straub et al. 2008). In terms of releases of multiple natural enemies, functional biodiversity is likely to be the more important factor; releases of natural enemies that fill distinct niches and perform different functions are more likely to have the good kind of biodiversity (Straub et al. 2008). For the phytoseiid mites, all controlled experiments comparing mixed and single species releases have involved a specialist species and a generalist species (Croft and MacRae 1992b,a, Schausberger 1997, Schausberger and Walzer 2001, Barber et al. 2003, Rhodes et al. 2006). These studies allow for the examination of how functional biodiversity affects biological control. No study found that a mix of two species was harmful to biological control, but mixtures often had the same amount of pest control as the best natural enemy alone (Schausberger 1997, Schausberger and Walzer 2001, Barber et al. 2003, Rhodes et al. 2006). However, in apple orchards (a perennial crop) mixtures of Galendromus occidentalis (Nesbitt) and Typhlodromus pyri Schueten controlled spider mites better than either predator alone (Croft and MacRae 1992b,a). Indeed, it is likely that the beneficial effects of mixed predator releases will only be seen in long term studies in perennial crops. In these systems, generalists can build populations earlier in the season on alternative prey, increasing the duration of control (Croft and MacRae 1992b, Schausberger and Walzer 2001, Cakmak et al. 2009); specialists can suppress pest populations if they reach higher levels later in the season. Additional aspects of the system may also make using multiple natural enemies beneficial. For example, one predator may be better at overwintering than another (Barber et al. 2003), or in the case of T. pyri, be less susceptible to 8

23 intraguild predation than its better counterpart (Croft and MacRae 1992b). Finally, released natural enemies may have differential susceptibility to pesticide applications; releasing multiple species increases the likelihood that at least one will survive (especially if a resistant strain is deliberately chosen). Specialists vs. generalists. It is important to note that in the previous section, the best natural enemy was the specialist in all cases except Rhodes et al. (2006), where the generalist, Neoseiulus californicus (McGregor) decreased T. urticae populations to lower levels than the specialist, P. persimilis. Many guidelines for biological control indicate that prey specificity is an important trait for biological control agents, and priority was given to specialists (DeBach 1964, Debach 1974, Rosen and Huffaker 1983, Symondson et al. 2002). Specificity decreases the risk of nontarget effects (Howarth 1991). However, evidence that generalists can also be effective control agents is mounting. Generalists are often able to establish at low prey densities and maintain populations even when prey become scarce due to successful biological control (Symondson et al. 2002). They are less likely to disperse from a location with low prey levels (McMurtry and Croft 1997, Jung and Croft 2001). Therefore, predator-prey systems involving a generalist predator are likely to be more stable (McMurtry 1992, Jung and Croft 2001) and bring about persistent biological control. In annual crops, or in controlled environments, like greenhouses, this may not be as important; however, in more stable cropping systems, like orchards, long term mite control may rely on more generalist phytoseiids (McMurtry 1992). This is especially true in large canopy crops where constantly releasing predators into the system is economically viable (Schmidt et al. 2013). Finally, phytoseiid generalists may be better able to control tetranychids outside of the genus Tetranychus. While many specialist mites are able to feed on other genera of mites, they are not as attracted to other species and do not track their 9

24 spatial distributions (Sabelis and Van de Baan 1983, Nyrop 1988). This could result in a reduced ability to control these species. However, there are benefits to releasing specialist mites. They often provide better shortterm biological control (McMurtry and Croft 1997), as demonstrated in the discussion of releases of multiple species of predators. If the pest to be controlled is a Tetranychus spp., the use of a specialist is indicated. Specialist predators adapted to feed on Tetranychus spp. are attracted to kairomones produced by these mites and have morphological adaptations for navigating in their dense webbing (Sabelis and Van de Baan 1983, Sabelis and Bakker 1992, Amin et al. 2009). Conversely, generalists often have lower reproductive and prey consumption rates on these pests (if they will eat them at all) (McMurtry and Scriven 1964, Gilstrap and Friese 1985, McMurtry and Croft 1997) or will even get trapped and die in the webbing produced by Tetranychus spp. (Putman 1962, McMurtry and Scriven 1964, Sabelis and Bakker 1992). As a general rule, specialists are likely to be most useful when immediate biological control is necessary (augmentation strategies, greenhouses) or where Tetranychus spp. are pests, whereas generalists will likely be more effective in in longer term management plans (conservation biological control). The use of pesticides can alter the dynamic between specialist and generalist phytoseiids that exist in the same agroecosystem. Specialist phytoseiids have historically been documented to have pesticide resistance (eg. P. persimilis, G. occidentalis), whereas generalists are typically found in minimally sprayed conditions (McMurtry 1992, Croft and Luh 2004) and can be more affected by pesticides than specialists (Downing and Moilliet 1972, Argolo et al. 2013). It is likely that the pesticide resistance selection pressure for specialists is higher than that of generalists, as they are more closely associated with their spider mite prey (the target of the 10

25 pesticide applications) (Croft and Brown 1975). This differential effect emphasizes the importance of selective pesticide programs in IPM. Genetic improvement of biological control agents. One of the many reasons phytoseiids have been well studied is their resistance to agricultural pesticides. After fieldselected organophosphate (OP)-resistant phytoseiids were successfully established in new areas (Readshaw 1975), a natural extension of this work was to deliberately select for other types of pesticide resistance in the laboratory (Hoy 1992). Strains resistant to pyrethroids and carbamates have been successfully created (Hoy and Knop 1981, Thistlewood et al. 1995) and used to control spider mites in the field (Hoy 1982c,a). However, establishment proved to be fairly simple, as long as harmful pesticide sprays were avoided and the resistance trait remained stable in field populations (McMurtry 1981). M These strains were also fairly similar to natural phytoseiids in life table parameters and other biological aspects, indicating a lack of fitness costs for the resistance trait (Bruce-Oliver and Hoy 1990, Li and Hoy 1996). In addition to selecting for pesticide resistance, a non-diapausing strain of G. occidentalis has also been developed (Hoy 1982b). These are useful in greenhouses that do not experience typical seasonal cycles and have been successfully used for pest control (Field 1981, Field and Hoy 1984,1985). In addition to traditional genetic improvement, G. occidentalis has also been genetically transformed via microinjection (Presnail and Hoy 1992). The gene transferred was lacz, which does not have applications in pest management. However, risk assessments and limited field releases with the transgenic phytoseiids have been conducted (McDermott and Hoy 1997, Hoy 2000). Although the gene is very stable in the population under laboratory conditions, it was highly unstable in field trials and disappeared after very few generations (Hoy 2000). Further work is needed to both develop this technique to insert genes that would improve predator 11

26 performance and increase the stability of the gene in the field. Like classical biological control, this technique is not without environment risks. Environmental risks. Risks of classical biological control only began to be critically studied in the 1980s (Strong and Pemberton 2001); previous to this time it was considered a virtually risk-free process (Debach 1974, Hoy 1992). Although it was understood that herbivores introduced for the control of weeds must be carefully screened before introduction, nontarget effects of released predators and parasitoids were not given much consideration (Debach 1974, Rosen and Huffaker 1983). Indeed, displacement of less effective, native control agents was seen as a positive outcome, and the introduction of new species was asserted to be the most important line of research in applied biological control (DeBach 1966,1971). While the harmful effects of some released vertebrates became obvious over time (e.g., the cane toad, Rhinella marinus (L.)), effects of arthropods are more difficult to observe due to their small size (Strong and Pemberton 2001). One of the larger, more conspicuous groups of predators, the coccinellids, has been reviewed for their non-target effects (Obrycki et al. 1999). Releases of ladybeetles have been blamed for the decline in native species (Harmon et al. 2007). Parasitoids have been implicated in the extinction of rare island species (Howarth 1991, Simberloff and Stiling 1996). To date, phytoseiids have not been implicated in non-target effects (Hoy 1992, Gerson et al. 2003d). For instance, while P. persimilis has a higher reproductive rate than most native phytoseiids, there was no evidence of it displacing native phytoseiids on wild plants surrounding the agricultural areas where it had be released in southern California (McMurtry 1981). While introduced phytoseiids replaced natives within the cropping system of release in Australia, they were rarely found outside of the crop (Walter et al. 1998). Indeed, if pesticide applications within the crop ceased, the native species would reestablish (James 1998). There then seems to be little 12

27 risk to native species, since they are only replaced in areas in which they cannot persist (pesticide-treated crops) (Hoy 1992). There is also a lack of evidence for escape of transgenic phytoseiids out of the agriculture systems in which they are released (Hoy 1992). In general, transgenic natural enemies pose the same risks as standard biological control agents (Howarth 1991, Li and Hoy 1996), although the perceived risks may be higher for the public (Hoy 2000). Therefore, predatory mites seem to be among some of the lowest risk introduced biological control agents. However, they have also been involved in far fewer introductions than insects (Hoy 2011b). Although potential environmental risks of biological control should be assessed, it is important to remember that the alternative to biological control is not simply the absence of biological control pesticides will be used to control the pest. The pesticide alternative is likely more harmful to the environment or human health. Fortunately, classical biological control is not the only form of biological control available for spider mite management. AUGMENTATIVE BIOLOGICAL CONTROL Augmentative biological control also involves the release of natural enemies for control of a pest. However, unlike classical biological control, it requires mass rearing of the natural enemy and the natural enemy is not expected to establish permanently (DeBach 1964, Wilson 1966). It has been divided into two types: inoculative and inundative. Inoculative releases are of small numbers of natural enemies, and biological control is provided by the successive generations of the released natural enemies, not the released enemies themselves (DeBach 1964, Wilson 1966). Conversely, inundative releases require large numbers of natural enemies, with the generation released providing control (DeBach 1964, Wilson 1966). Inundative releases of 13

28 natural enemies have been referred to as biological pesticides due to the immediacy of the effect desired (Stehr 1982, Parrella et al. 1992, Hoy 2011b). The distinction between these two types may not be clear in specific situations (Parrella et al. 1992). Phytoseiulus persimilis blurs the line between classical and augmentative control. Although it is not native in most of the systems into which it is introduced (viz., classical biological control), it frequently has to be repeatedly released (augmentative biological control). However, this is likely due to its prevalent use in greenhouses, where there is annual turnover in the crop. In non-protected crops, it has the ability to persist (James 1998) and releases are more likely to be categorized as true classical biological control. Indeed, it is in greenhouses where many cases of augmentative control occur. The common use of augmentative releases in greenhouses is likely due to several reasons. First, predators are unlikely to permanently establish on short-season greenhouse crops, making classical or conservation biological control difficult. Additionally, the value of these crops is higher, which can offset the high cost of repeatedly purchasing natural enemies (Stehr 1982, Parrella et al. 1992, Hoy 2011b). Greenhouses also avoid several of the problems commonly associated with augmentative control: an unfavorable environment and enemy dispersal (Collier and Van Steenwyk 2004), by providing a climate-controlled, enclosed space. Phytoseiid releases in greenhouses have been relatively successful (Gerson and Weintraub 2007). Augmentative releases of phytoseiids have also been successful in field-grown crops; releases of Neoseiulus fallacis (Garman) for control of spider mites in tree fruit in Canada and the United States are considered a key example of successful augmentation (van Lenteren 2006). However, augmentation is still not widely used (Collier and Van Steenwyk 2004). This has been attributed to a variety of factors including environmental or pest mismatch with the natural 14

29 enemy, poor quality of insectary-reared predators, mechanics of the release (timing, method, rate), and use of pesticides after release (Collier and Van Steenwyk 2004). Fortunately, many of these problems can be addressed by further research. However, cost effectiveness may limit adoption of augmentative control (Parrella et al. 1992, Collier and Van Steenwyk 2004). Past problems with phytoseiid augmentation. Although many augmentative programs are successful, unsuccessful releases can provide important information. Many studies have emphasized the number of predators released or the time of the year released as the most likely problem. Both of these aspects address the same issue: the ratio of predators to prey (prey numbers tend to be lower earlier in the season). Many experimental releases have concluded that successful control requires a high ratio of predator to prey, and that releases are best done when pest numbers are low (Pickett and Gilstrap 1986, Hoy and Glenister 1991, Smith and Papacek 1991, Strong and Croft 1995, Stavrinides 2010). There have also been failures when pesticide applications have had nontarget effects on released phytoseiids (Mansour et al. 1993, Ahmad et al. 2013, Schmidt et al. 2013). Finally, there have been instances where either the host plant or the pest were deemed unsuitable for the phytoseiid released (Colfer et al. 2004, Koller et al. 2007). As previously mentioned, these problems are identified in experimental releases and the methods are improved in order to make recommendations for agricultural producers. It is, therefore, not these issues that prevent widespread adoption of augmentative control. Most studies of augmentative control (including those using phytoseiids) do not examine the cost effectiveness of release. While cost effectiveness has been established for strawberries (especially when combined with selective pesticide use) (Trumble and Morse 1993), it has been questioned for other crops (Prokopy and Christie 1992, Shrewsbury and Hardin 2003, Schmidt et al. 2013). Indeed, past successful augmentation programs in tree crops required extremely high 15

30 release rates ( mites/tree) (Croft and McMurtry 1972, Fadamiro et al. 2013). This is prohibitively expensive in most instances, especially when compared to the cost of a pesticide application. While augmentation may be a suitable strategy for small canopy, protected crops, a more permanent solution, like classic or conservation biological control, is required for large perennial agroecosystems. CONSERVATION BIOLOGICAL CONTROL Conservation biological control (CBC) involves the manipulation of an environment to enhance the survival, fecundity, longevity, or behavior of natural enemies to increase their effectiveness (Landis et al. 2000). This can involve removal of harmful inputs or the addition of positive inputs (needed resources), usually via habitat management. This aspect of biological control has received the least attention (Landis et al. 2000), possibly because it was considered to require more research that classical biological control (Debach 1974). However, biological control may be moving away from classical biological to focus on CBC (Symondson et al. 2002). Because of its emphasis on natural biological control, CBC may be considered as a first line of defense, rather than an alternative for when pesticides fail (Stehr 1982). Pesticide use. Selective use of pesticides was one of the first CBC techniques developed, and has been the main emphasis of this field (Wilson 1966, Hull and Beers 1985, Landis et al. 2000). Recommendations typically involved leaving unsprayed refuges, using minimal rates, timing applications to minimize harm to natural enemies, and using selective materials (Wilson 1966). Predators tend to not develop resistance to pesticides as quickly, or as often, as pests, perhaps because selection pressure is stronger on the pest than the predator; the pest is more 16

31 closely associated with the area where pesticides are applied (Georghiou 1972, Croft and Brown 1975). The predator relies on the presence of the prey (and will die or emigrate when the pest is eliminated), so the prey will develop resistance first (Georghiou 1972, Croft 1990c). After the pest becomes resistant, the pesticide used is switched, giving the predator insufficient time to gain resistance (Georghiou 1972, Stehr 1982). Some phytoseiids have managed to develop resistance despite these factors; these were the first known cases of a natural enemy becoming resistant to pesticides (Croft and Brown 1975). This could be due to their close association with spider mites (increasing selective pressure), while still having the ability to persist without them on other food sources (allowing them to continue to be exposed to pesticides after the pest is eliminated) (Croft and Brown 1975). Like the spider mites, they also have short generation times (McMurtry et al. 1970), allowing them to evolve quickly (Croft and Brown 1975). Their limited ability to disperse also increases the selection pressure for resistance (Georghiou 1972, Croft and Brown 1975, Stehr 1982). Resistance of some phytoseiids to certain pesticides allowed for the development of integrated mite management (IMM) programs, where selected pesticides were used for control of key pests while maintaining populations of a phytoseiid (resistant to those pesticides) responsible for mite control (Madsen 1964, Hoyt 1969, McMurtry 1981, Croft 1990a, Gerson et al. 2003a, Hoy 2011a). A natural extension of this work was to introduce resistant strains of phytoseiids to new parts of the world. Additionally, as new classes of pesticides are developed, they have been tested for nontarget effects on phytoseiids (Villanueva and Walgenbach 2005, van Driesche et al. 2006, Duso et al. 2008, Bostanian et al. 2009, Bostanian et al. 2010, Lefebvre et al. 2011, Lefebvre et al. 2012, Beers and Schmidt 2014). While previous screening has emphasized acute mortality (Croft 1990b), there is also an increased interest in sublethal effects, which may have 17

32 significant population-level impacts (Croft 1990b, Blumel et al. 1993, Desneux et al. 2007, Beers and Schmidt 2014, Beers and Schmidt-Jeffris 2015). This is still a developing field, and as new pesticides are registered, research regarding their effects on phytoseiids will need to continue if biological control efforts are to succeed. Dust control. Dust is another factor which has been implicated as harmful to some beneficial insects and to cause pest outbreaks when dust residues are present on leaves (Bartlett 1951, Fleschner 1956,1958, Flaherty et al. 1969, Fukushima and Stalford 1969, Kinn et al. 1972). This effect has been attributed to the water absorbent properties of dust particles, resulting in dehydration of arthropods coated in dust (Bartlett 1951, Fleschner 1958). However, more recent research has not found that dust plays a role in mite outbreaks (Oi and Barnes 1989, Pringle et al. 2014). Dust is a sign of dry conditions, which can stress both plants and predatory mites (Kinn et al. 1972, Oi 1987, Pringle et al. 2014). This results in a crop that is more susceptible to herbivory and at the same time lacking natural enemies that could reduce herbivory. Conditions that tend to be dust-free (adequate humidity or irrigation) also promote ground cover, which may be also improve mite biological control (see next section). Therefore, current research suggests that dry conditions, not dust in and of itself, should be prevented where possible to decrease mite outbreaks. Shelter. One of the most common forms of habitat management involves the provisioning of shelter. This includes external plantings outside of the crop (most commonly hedgerows), and within-crop plantings (herbaceous plants, beetle banks) (Landis et al. 2000, Griffiths et al. 2008). Non-crop habitat breaks up an otherwise fairly homogenous environment, increasing landscape diversity (Tscharntke et al. 2007). These new landscape features provide resources that are typically lacking in agroecosystems, including overwintering sites, improved 18

33 microclimate (usually increased humidity), and supplementary food (Altieri and Letourneau 1982, Landis et al. 2000, Griffiths et al. 2008, Pisani Gareau et al. 2013). For phytoseiids, shelter can be provided in the form of increased ground cover, either through relaxed weed control, or deliberate plantings (Alston 1994, Mailloux et al. 2010). Dead plant material (Morris et al. 1996) and hedgerows (Boller et al. 1988, Tixier et al. 1998, Tixier et al. 2002, Duso et al. 2004) may also serve as shelter. In addition to providing shelter, these plantings can also serve as sources of alternative food (discussed in the next section). Ground cover provides a microclimate with lower temperatures and higher humidity than the canopy (Croft and McGroarty 1977, Huang et al. 1981, Liang and Huang 1994), conditions preferred by phytoseiids (Croft et al. 1993, Schausberger 1998, Gerson et al. 2003b). Hedgerows and ground cover have also been suggested to provide refuge sites from pesticide applications (Boller et al. 1988, Liang and Huang 1994). Finally, in addition to the top-down controls provided by CBC, there may be bottom-up effects. Some weeds act as reservoirs for T. urticae; they will disperse into orchard trees if these weeds are removed (Flexner et al. 1991), but may remain in the ground cover if weeds are maintained (Aguilar-Fenollosa et al. 2011b). Artificial shelters may also be beneficial. Kawashima and Jung (2010) found that phytoseiids would use shelters made of several different materials on the orchard floor. Previous research has suggested that winter ground cover can increase biological control provided by phytoseiids in the next season (Croft and McGroarty 1977, Morris et al. 1996), but the hypothesis that artificial shelter can reduce winter mortality has not yet been tested. While use of shelters is a well-known method for studying phytoseiid phenology, it has not yet been adequately tested as a CBC method. 19

34 Alternative food. The same structures that provide shelter may also provide sources of alternative food. Hedgerows (Altieri and Letourneau 1982, Duso et al. 2004), and cover crops can increase natural enemy densities by providing pollen or nectar. Usually these cover crops are planted for the benefit of parasitoids (Leius 1967, Begum et al. 2006, Berndt et al. 2006) or syrphids (Laubertie et al. 2012, Gontijo et al. 2013, van Rijn et al. 2013), but generalist phytoseiids also benefit from the addition of pollen (Kennett et al. 1979, McMurtry 1981, Onzo et al. 2005, González-Fernández et al. 2009). The pollen provided by hedgerows and ground cover has been found to increase phytoseiid populations, possibly by providing higher quality or quantity of pollen than the crop, and over a longer period of time (Boller et al. 1988, Grout and Richards 1992, Tixier et al. 1998, Duso et al. 2004, Mailloux et al. 2010). Diversity of pollen increases the likelihood that the right pollen for a given phytoseiid will be present (Gerson et al. 2003b). In addition to pollen, non-crop plants can also provide a source of spider mite or alternative prey (Boller et al. 1988, Flexner et al. 1991). Phytoseiids use these resources in times of prey scarcity in the crop, keeping their numbers high and preventing the occurrence of pest outbreaks (Waite 1988, Takahashi et al. 1998). Chemical ecology. One of the more recent areas of research developed in CBC is the use of herbivore-induced plant volatiles (HIPV) in applied biological control (Jonsson et al. 2008, Khan et al. 2008). HIPVs are volatiles produced by plants when they are damaged by herbivores. They are known to repel pests and attract predators, and can be produced both constitutively or as an induced defense (de Moraes et al. 1998, Kessler and Baldwin 2001, Khan et al. 2008). These volatiles also have the ability to prime neighboring plants for attack; unharmed plants can detect these volatiles and then produce their own induced defenses (Karban et al. 2000, Turlings and Ton 2006, Heil and Bueno 2007). Knowledge of HIPVs has been applied 20

35 practically in two main ways: (1) push-pull strategies and (2) use of artificial HIPVs in dispensers. Push-pull strategies involve the interplantings of repellent plants into the crop (push) and attractive plants on the margins of the crop (pull)(cook et al. 2007, Hassanali et al. 2008). This strategy is used to bring pests out of crop plant into a plant of less value (the trap crop). While push-pull can be easily integrated into other aspects of CBC, it does not inherently involve biological control agents (Cook et al. 2007). Therefore, while this is a useful strategy, it will not be discussed further in this review. HIPVs have also been artificially synthesized and incorporated into dispensers. They have been found to attract a wide variety of predators (James 2003, James and Price 2004, James 2005, Khan et al. 2008, Jones et al. 2011, Kaplan 2012), and some field studies have found that this leads to reduced pest populations and damage (James and Price 2004, Simpson et al. 2011, Orre Gordon et al. 2013, Kelly et al. 2014). Indeed, dispensers have been incorporated into other aspects of CBC. Attract and reward has successfully used dispensers to bring natural enemies into an area and provision them with nectar resources upon arrival (Simpson et al. 2011, Orre Gordon et al. 2013). Even when these effects do not act synergistically, they at least have the potential to attract a greater diversity of natural enemies than each technique alone (Orre Gordon et al. 2013). HIPVs have also been integrated into augmentative control through the use of release and retain strategies, where the volatiles are used to encourage released predators to remain in a crop (Kelly et al. 2014). These strategies could easily be tested in the context of phytoseiid-spider mite systems. Phytoseiids are known to be attracted to volatiles produced by plants that have undergone spider mite herbivory (Dicke and Sabelis 1988, Mizell and Schiffhauer 1991, Ozawa et al. 2000). They 21

36 are capable of distinguishing between volatiles produced by different species of spider mite (Takabayashi et al. 1994, Amin et al. 2009), different crop plants or cultivars (Dicke et al. 1998, Onzo et al. 2012), and have demonstrated the ability to change their response to volatiles through learning (Takabayashi et al. 1994, Krips et al. 1999, Shimoda and Dicke 2000, de Boer et al. 2005). In order to exploit the behavior of these predators, some strategies for biological control improvement have been suggested. Breeding plants to emit HIPVs attractive to phytoseiids has been suggested (Dicke et al. 1990), but not tested. However, the ability to breed for increased attraction of phytoseiids to HIPVs has been demonstrated (Margolies et al. 1997). Genetic transformation was used to cause Arabidopsis thaliana L. to produce two isoprenoids that resulted in increased attraction of P. persimilis to this plant (Kappers et al. 2005). Preliminary work has also shown that synthetic methyl salicylate is attractive to T. pyri, indicating dispenser technology may also be useful for attracting predatory mites (Gadino et al. 2011). Finally, HIPVs may also play an important role in nectary use by phytoseiids. Lima beans were found to increase production of extrafloral nectar after exposure to T. urticae-induced plant volatiles (Choh et al. 2006). Phytoseiulus persimilis dispersed from plants with additional extrafloral nectar production more slowly, and demonstrated the ability to use this as a food source. Unfortunately, little has been done to examine the effects of HIPVs outside of closedsystem (Y-tube) settings. Indeed, Yano and Osakabe (2009) were unable to find evidence that P. persimilis was attracted to plants that were infested by T. urticae in an open system, even at short distances. However, the plants used in the study were not the plants on which the insectary purchased P. persimilis had been reared, so it is possible that the insectary-reared phytoseiids did not recognize these odor sources as potential prey patches (Krips et al. 1999, Shimoda and Dicke 2000). It is possible commercially purchased phytoseiids may not be attracted to the same 22

37 volatiles as wild counterparts, indicating that HIPV attraction tests would be an important aspect of quality control (Dicke et al. 1990). Further study is needed before the utility of HIPVs in phytoseiid CBC can be established. Diversity. One overarching aspect of CBC is increasing diversity, both of plants and natural enemies (Gurr et al. 2003). Increasing plant diversity increases the number of possible resources provisioned to natural enemies, both in terms of shelter and food, which increases the stability of the community (enemies hypothesis) (Pimentel 1961, Russell 1989, Altieri 1999). It also dilutes the resources of herbivores within the agroecosystem, potentially making the crop more difficult to find (resource concentration hypothesis) (Root 1973). Although these two concepts can act synergistically (Russell 1989, Andow 1991), it has also been suggested that decreased herbivore abundance in more diverse systems is more likely due to decreased resource concentration, than to increased predation (Risch et al. 1983, Andow 1991). However, there is a strong and little-contested body of evidence indicating that landscape diversity increases natural enemy diversity (Dennis and Fry 1992, Asteraki et al. 1995, Bianchi et al. 2006). The current debate is focused on whether increased natural enemy diversity harms or benefits biological control (see section on CBC). This same line of argumentation holds here: that diverse populations of natural enemies may act together (additively or synergistically) to increase biological control or may have increased negative intraguild interactions, decreasing biological control. In studies involving phytoseiids, both cases have been found to occur. The previous discussion regarding multi-species releases of classical biological control indicated either positive or neutral effects of releasing multiple species (compared to a single species). In the CBC literature, adding a cover crop of Festuca arundinaceae Schreb. benefitted a Tetranychus spp. specialist phytoseiid in citrus over the typically more common generalist 23

38 present, and was attributed to increased biological control (Aguilar-Fenollosa et al. 2011a). Alternatively, a wild plant cover favored the generalist phytoseiids, decreasing biological control due to intraguild predation (Abad-Moyano et al. 2010, Pina et al. 2012). The generalist predator winning does not always result in decreased biological control, and indeed may be beneficial (Onzo et al. 2005). These results indicate that the identity of the pest (Wilby et al. 2005) (Tetranychus spp. or not) and natural enemy (Straub and Snyder 2006) may be more important to the success of a particular CBC program than diversity itself. However, there has been little research examining how phytoseiid diversity management affects biological control. More studies in this area may reveal better ways to promote the right kind of phytoseiid biodiversity. Conclusions. In the areas where these CBC strategies have been implemented in the field, they have typically benefitted phytoseiid control. Cover crops and ground cover provide both shelter and alternative food, as do other structures, like hedgerows. There is less evidence that dust management will successfully control mites. Newer technologies, like the use of synthetic HIPV dispensers have yet to be thoroughly tested on phytoseiids. The area of pesticide selectivity is where the greatest quantity of phytoseiid literature exists and the strongest case can be made for CBC as a part of IMM. However, this particular aspect of biological control has been implemented in many IMM programs for many years, including one of the most wellknown. INTEGRATED MITE MANAGEMENT IN WASHINGTON APPLE Apple orchards have been prime targets of non-chemical pest control tactics. Because of how these systems are managed, they have greater potential for biological control. Orchards are more permanent ecosystems and have greater structural and biological diversity, which promotes 24

39 biological control (Altieri and Letourneau 1982, Altieri 1999, Simon et al. 2010). Unlike annual crops, they are highly stratified and physically complex, offering a variety of different habitats in a relatively small space (Simon et al. 2010). In addition to the increased diversity of provisioned resources provided in such an environment, there is also an increase in enemy-free space, decreasing incidence of intraguild predation (Finke and Denno 2006). The physical complexity of the crop plants also reduces the ability to achieve perfect pesticide coverage, providing an additional form of refugia for natural enemies. However, it is unlikely that any of these factors would have been exploited in IMM if not for the spider mite population outbreaks that followed synthetic pesticide era. Spider mites became resistant to pesticides meant to control them (or other pests), while predatory mites decreased due to nontarget effects of these pesticides (Huffaker et al. 1970, McMurtry et al. 1970). In the inception of IMM in Washington, Tetranychus mcdanieli McGregor and Panonychus ulmi (Koch) were the main mite pests of concern in apple orchards (Hoyt 1967). The discovery that an effective predator of these pests, G. occidentalis, was resistant to some OPs was a turning point in the management of mites. This observation allowed for the development of an IMM program that focused on the conservation of G. occidentalis through the selective use of pesticides (Madsen 1964, Hoyt 1969). OPs could still be used for codling moth (Cydia pomonella (L.)) control (although at reduced rates), while causing minimal harm to G. occidentalis. This allowed for minimal application of acaricides for spider mite control. It was also recognized that moderate populations of apple rust mites (Aculus schlechtendali (Nalepa)) were tolerable; they provided alternative food for G. occidentalis and allowed it to sustain populations during times of spider mite scarcity (Hoyt 1966). This program remained successful for several decades. 25

40 Unfortunately, since the early 2000s, mite outbreaks have increased in severity and prevalence (Beers et al. 2005). This has been attributed to the reduced use of OPs and the increased use of reduced-risk pesticides (Beers et al. 2005), due to the legislated phase out of OPs (Agnello et al. 2009). While G. occidentalis is resistant to some OPs, many reduced risk pesticides are known to have harmful nontarget effects (Beers et al. 2005, Bostanian et al. 2009, Lefebvre et al. 2011, Beers and Schmidt 2014). These problems with a critical area of apple pest management indicate that the ongoing changes in the selection of pesticides used for key pest control must be addressed dynamically with changes in IMM. Aspects requiring continued vigilance include nontarget effects (both lethal and sublethal), phytoseiid species diversity and ecology, and horticultural management practices that influence the conservation of natural enemies. 26

41 REFERENCES CITED Abad-Moyano, R., A. Urbaneja, D. Hoffmann, and P. Schausberger Effect of Euseius stipulatus on establishment and efficacy in spider mite suppression of Neoseiulus californicus and Phytoseiulus persimilis in clementine. Exp. Appl. Acarol. 50: Agnello, A. M., A. Atanassov, J. C. Bergh, D. J. Biddinger, L. J. Gut, M. J. Haas, J. K. Harper, H. W. Hogmire, L. A. Hull, L. F. Kime, G. Krawczyk, P. S. McGhee, J. P. Nyrop, W. H. Reissig, P. W. Shearer, R. W. Straub, R. T. Villanueva, and J. F. Walgenbach Reduced-risk pest management programs for eastern U.S. apple and peach orchards: a 4-year regional project. Am. Entomol. 55: Aguilar-Fenollosa, E., M. V. Ibanez-Gual, S. Pascual-Ruiz, M. Hurtado, and J. A. Jacas. 2011a. Effect of ground-cover management on spider mites and their phytoseiid natural enemies in clementine mandarin orchards (II): Top-down regulation mechanisms. Biol. Control 59: Aguilar-Fenollosa, E., M. V. Ibáñez-Gual, S. Pascual-Ruiz, M. Hurtado-Ruiz, and J. A. Jacas. 2011b. Effect of ground-cover management on spider mites and their phytoseiid natural enemies in clementine mandarin orchards (I): tetranychid mite dynamics. Biol. Control 59: Ahmad, S., A. Pozzebon, and C. Duso Augmentative releases of the predatory mite Kampimodromus aberrans in organic and conventional apple orchards. Crop Prot. 52: Alston, D. G Effect of apple orchard floor vegetation on density and dispersal of phytophagous and predaceous mites in Utah. Agric. Ecosyst. Environ. 50:

42 Altieri, M. A The ecological role of biodiversity in agroecosystems. Agric. Ecosyst. Environ. 74: Altieri, M. A., and D. K. Letourneau Vegetation management and biological control in agroecosystems. Crop Prot. 1: Amin, M. M., R. F. Mizell, and R. W. Flowers Response of the predatory mite Phytoseiulus macropilis (Acari: Phytoseiidae) to pesticides and kairomones of three spider mite species (Acari: Tetranychidae), and non-prey food. Fla. Entomol. 92: Andow, D. A Vegetational diversity and arthropod population response. Annu. Rev. Entomol. 36: Argolo, P. S., N. Banyuls, S. Santiago, O. Molla, J. Jacas, and A. Urbaneja Compatibility of Phytoseiulus persimilis and Neoseiulus californicus (Acari: Phytoseiidae) with imidacloprid to manage clementine nursery pests. Crop Prot. 43: Asteraki, E. J., C. B. Hanks, and R. O. Clements The influence of different types of grassland field margin on carabid beetle (Coleoptera, Carabidae) communities. Agric. Ecosyst. Environ. 54: Barber, A., C. A. M. Campbell, H. Crane, R. Lilley, and E. Tregidga Biocontrol of two-spotted spider mite Tetranychus urticae on dwarf hops by the phytoseiid mites Phytoseiulus persimilis and Neoseiulus californicus. Biocontrol Sci. Technol. 13: Barfield, C. S., and J. L. Stimac Pest management: an entomological perspective. BioScience 30:

43 Bartlett, B. R The action of certain "inert" dust materials on parasitic hymenoptera. J. Econ. Entomol. 44: Beers, E. H., and S. C. Hoyt European red mite In E. H. Beers, J. F. Brunner, M. J. Willett and G. M. Warner (eds.), Orchard pest management: a resource book for the Pacific Northwest. Good Fruit Grower, Yakima, WA. Beers, E. H., and R. A. Schmidt Impacts of orchard pesticides on Galendromus occidentalis: lethal and sublethal effects. Crop Prot. 56: Beers, E. H., and R. A. Schmidt-Jeffris Effects of orchard pesticides on Galendromus occidentalis (Acari: Phytoseiidae): repellency and irritancy. J. Econ. Entomol. 108: Beers, E. H., J. F. Brunner, J. E. Dunley, M. Doerr, and K. Granger Role of neonicotinyl insecticides in Washington apple integrated pest management. Part II. Nontarget effects on integrated mite control. J. Insect Sci. 5 (16): available online. Begum, M., G. M. Gurr, S. D. Wratten, P. R. Hedberg, and H. I. Nicol Using selective food plants to maximize biological control of vineyard pests. J. Appl. Ecol. 43: Berndt, L. A., S. D. Wratten, and S. L. Scarratt The influence of floral resource subsidies on parasitism rates of leafrollers (Lepidoptera: Tortricidae) in New Zealand vineyards. Biol. Control 37: Bianchi, F. J. J. A., C. J. H. Booij, and T. Tscharntke Sustainable pest regulation in agricultural landscapes: a review on landscape composition, biodiversity and natural pest control. Proc. R. Soc. B. 273: Blumel, S., F. Bakker, and A. Grove Evaluation of different methods to assess the sideeffects of pesticides on Phytoseiulus persimilis A.-H. Exp. Appl. Acarol. 17:

44 Boller, E. F., U. Remund, and M. P. Candofi Hedges as potential sources of Typhlodromus pyri, the most important predatory mite in vineyards of northern Switzerland. Entomophaga 33: Bostanian, N. J., J. M. Hardman, H. A. Thistlewood, and G. Racette Effects of six selected orchard insecticides on Neoseiulus fallacis. Pest Manag. Sci. 66: Bostanian, N. J., H. A. Thistlewood, J. M. Hardman, M.-C. Laurin, and G. Racette Effect of seven new orchard pesticides on Galendromus occidentalis in laboratory studies. Pest Manag. Sci. 65: Bruce-Oliver, S. J., and M. A. Hoy Effect of prey stage on life-table attributes of a genetically manipulated strain of Metaseiulus occidentalis (Acari: Phytoseiidae). Exp. Appl. Acarol. 9: Cakmak, I., A. Janssen, M. W. Sabelis, and H. Baspinar Biological control of an acarine pest by single and multiple natural enemies. Biol. Control 50: Carson, R Silent spring. Houghton Mifflin, Boston, MA. Casula, P., A. Wilby, and M. B. Thomas Understanding biodiversity effects on prey in multi-enemy systems. Ecol. Lett. 9: Choh, Y., S. Kugimiya, and J. Takabayashi Induced production of extrafloral nectar in intact lima bean plants in response to volatiles from spider mite-infested conspecific plants as a possible indirect defense against spider mites. Oecologia 147: Chrispeels, M. J., and D. E. Sadava Ten thousand years of crop selection, pp Plants, genes, and agriculture. Jones and Bartlett Publishers, Boston, MA. 30

45 Colfer, R. G., J. A. Rosenheim, L. D. Godfrey, and C. L. Hsu Evalution of large-scale releases of western predatory mite for spider mite control in cotton. Biol. Control 30: Collier, T., and R. Van Steenwyk A critical evaluation of augmentative biological control. Biol. Control 31: Cook, S. M., Z. R. Khan, and J. A. Pickett The use of push-pull strategies in integrated pest management. Annu. Rev. Entomol. 52: Croft, B. A. 1990a. Endemic species, pp Arthropod biological control agents and pesticides. John Wiley & Sons, New York, NY. Croft, B. A. 1990b. Sublethal influences, pp Arthropod biological control agents and pesticides. John Wiley & Sons, New York, NY. Croft, B. A. 1990c. Factors affecting resistance, pp Arthropod biological control agents and pesticides. John Wiley & Sons, New York, NY. Croft, B. A., and J. A. McMurtry Minimum release of Typhlodromus occidentalis to control Tetranychus mcdanieli on apple. J. Econ. Entomol. 65: Croft, B. A., and A. W. A. Brown Responses of arthropod natural enemies to insecticides. Annu. Rev. Entomol. 20: Croft, B. A., and D. L. McGroarty The role of Amblyseius fallacis (Acarina: Phytoseiidae) in Michigan apple orchards. Farm Sci. 333: Croft, B. A., and I. V. MacRae. 1992a. Biological control of apple mites by mixed populations of Metaseiulus occidentalis (Nesbitt) and Typhlodromus pyri Scheuten (Acari: Phytoseiidae). Environ. Entomol. 21:

46 Croft, B. A., and I. V. MacRae. 1992b. Persistence of Typhlodromus pyri and Metaseiulus occidentalis (Acari: Phytoseiidae) on apple after inoculative release and competition with Zetzellia mali (Acari: Stigmaeidae). Environ. Entomol. 21: Croft, B. A., and H.-K. Luh Phytoseiid mites on unsprayed apple trees in Oregon, and other western states (USA): distributions, life-style types and relevance to commercial orchards. Exp. Appl. Acarol. 33: Croft, B. A., R. H. Messing, J. E. Dunley, and W. B. Strong Effects of humidity on eggs and immatures of Neoseiulus fallacis, Amblysieus andersoni, Metaseiulus occidentalis and Typhlodromus pyri (Phytoseiidae): implications for biological control on apple, caneberry, strawberry and hop. Exp. Appl. Acarol. 17: de Boer, J. G., T. A. L. Snoeren, and M. dicke Predatory mites learn to discriminate between plant volatiles induced by prey and noprey herbivores. Anim. Behav. 69: de Moraes, C. M., W. J. Lewis, P. W. Paré, H. T. Alborn, and J. H. Tumlinson Herbivore-infested plants selectively attract parasitoids. Nature 393: DeBach, P. (ed.) Biological control of insect pests and weeds. Chapman and Hall, London, UK. DeBach, P The competitive displacement and coexistence principles. Annu. Rev. Entomol. 11: Debach, P The use of imported natural enemies in insect pest management ecology, pp In Proceedings, 3rd Tall Timbers Conference on Ecological Animal Control by Habitat Management, Tallahassee, FL. Tall Timbers Research Station, Tallahasse, FL. 32

47 Debach, P Biological control by natural enemies. Cambridge University Press, London, UK. Dennis, P., and G. L. A. Fry Field margins can they enhance natural enemy population densities and general arthropod diversity on farmland. Agric. Ecosyst. Environ. 40: Desneux, N., A. Decourtye, and J.-M. Delpuech The sublethal effects of pesticides on beneficial arthropods. Annu. Rev. Entomol. 52: Dicke, M., and M. W. Sabelis How plants obtain predatory mites as body guards. Neth. J. Zool. 38: Dicke, M., M. W. Sabelis, J. Takabayashi, J. Bruin, and M. A. Posthumus Plant strategies of manipulating predator-prey interactions through allelochemicals. J. Chem. Ecol. 16: Dicke, M., J. Takabayashi, M. A. Posthumus, C. Schütte, and O. E. Krips Plantphytoseiid interactions mediated by herbivore-induced plant volatiles: variation in production of cues and in responses of predatory mites. Exp. Appl. Acarol. 22: Downing, R. S., and T. K. Moilliet Replacement of Typhlodromus occidentalis by T. caudiglans and T. pyri (Acarina: Phytoseiidae) after cessation of sprays on apple trees. Can. Entomol. 104: Drukker, B., A. Janssen, W. Ravensberg, and M. W. Sabelis Improved control capacity of the mite predator Phytoseiulus persimilis (Acari: Phytoseiidae) on tomato. Exp. Appl. Acarol. 21:

48 Duso, C., V. Malagnini, A. Paganelli, L. Aldegheri, M. Bottini, and S. Otto Pollen availability and abundance of predatory phytoseiid mites on natural and secondary hedgerows. BioControl 49: Duso, C., V. Malagnini, A. Pozzebon, M. Castagnolia, M. Liguori, and S. Simoni Comparative toxicity of botanical and reduced-risk insecticides to Mediterranean populations of Tetranychus urticae and Phytoseiulus persimilis (Acari Tetranychidae, Phytoseiidae). Biol. Control 47: Ewing The common red spider or spider mite. Oregon Agric. Exp. St. Bull. No Fadamiro, H. Y., C. Akotsen-Mensah, Y. Xiao, and J. Anikwe Field evaluation of predacious mites (Acari: Phytoseiidae) for biological control of citrus red mite, Panonychus citri (Trombidiformes: Tetranychidae). Fla. Entomol. 96: Field, R. P Evaluation of genetically-improved strains of Metaseiulus occidentalis (Nesbitt) (Acarina: Phytoseiidae) for integrated control of spider mites on roses in glasshouses. Ph.D. dissertation, University of California, Berkeley, CA. Field, R. P., and M. A. Hoy Biological control of spider mites on greenhouse rose. Calif. Agric. 38: Field, R. P., and M. A. Hoy Diapause behavior of genetically-improved strains of the spider mite predator Metaseiulus occidentalis (Acarina: Phytoseiidae). Entomol. Exp. Appl. 38: Finke, D. L., and R. F. Denno Predator diversity dampens trophic cascades. Nature 429: Finke, D. L., and R. F. Denno Spatial refuge from intraguild predation: implications for prey suppression and trophic cascades. Oecologia 149:

49 Flaherty, D., C. D. Lynn, F. L. Jensen, and D. A. Luvisi Ecology and integrated control of spider mites in San Joaquin vineyards. Ibid 23: 11. Fleschner, C. A Field approach to population studies of tetranychid mites on citrus and avocado in California, pp In E. C. Becker (ed.) Proceedings, 10th International Congress of Entomology, August 1956, Montreal, Canada. Mortimer. Fleschner, C. A The effect of orchard dust on the biological control of avocado pests. California Avocado Society Yearbook : Flexner, J. L., P. H. Westigard, P. Gonzalves, and R. Hilton The effect of groundcover and herbicide treatment on twospotted spider mite density and dispersal in southern Oregon pear orchards. Entomol. Exp. Appl. 60: Fukushima, C., and E. M. Stalford Effects of road dust on spider mites. Ibid 23: 10. Gadino, A. N., V. M. Walton, and J. C. Lee Olfactory response of Typhlodromus pyri (Acari: Phytoseiidae) to synthetic methyl salicylate in laboratory bioassays. J. Appl. Entomol. 136: Geier, P. W Management of insect pests. Annu. Rev. Entomol. 11: Georghiou, G. P The evolution of resistance to pesticides. Annu. Rev. Entomol. 3: Gerson, U Trends in research on acarine biological control agents, pp In R. B. Halliday, D. E. Walter, H. C. Proctor, R. A. Norton and M. J. Colloff (eds.), Proceedings, 10th International Congress of Acarology, 5-10 July 1998, Collingwood, Australia. CSIRO, Melbourne, Australia. Gerson, U., and E. Cohen Resurgences of spider mites (Acari: Tetranychidae) induced by synthetic pyrethroids. Exp. Appl. Acarol. 6:

50 Gerson, U., and P. G. Weintraub Mites for the control of pests in protected cultivation. Pest Manag. Sci. 63: Gerson, U., R. L. Smiley, and R. Ochoa. 2003a. The Phytoseiidae, pp Mites (Acari) for pest control. Blackwell Science, Oxford, UK. Gerson, U., R. L. Smiley, and R. Ochoa. 2003b. The effect of host plants and the ground cover on acarine biocontrol agents, pp Mites (Acari) for pest control. Blackwell Science, Oxford, UK. Gerson, U., R. L. Smiley, and R. Ochoa. 2003c. History, definitions and current work, pp Mites (Acari) for pest control. Blackwell Science, Oxford, UK. Gerson, U., R. L. Smiley, and R. Ochoa. 2003d. Environmental constraints in the use of acarine biocontrol agents, pp Mites (Acari) for pest control. Blackwell Science, Oxford, UK. Gilstrap, F. E., and D. D. Friese The predatory potential of Phytoseiulus persimilis, Amblyseius californicus, and Metaseiulus occidentalis (Acarina: Phytoseiidae). Int. J. Acarol. 11: Gontijo, L. M., E. H. Beers, and W. E. Snyder Flowers promote aphid suppression in apple orchards. Biol. Control 66: González-Fernández, J. J., F. de la Peña, J. I. Hormaza, J. R. Boyero, J. M. Vela, E. Wong, M. M. Trigo, and M. Montserrat Alternative food improves the combined effect of an omnivore and a predator on biological pest control. A case study in avocado orchards. Bull. Entomol. Res. 99:

51 Griffiths, G. J. K., J. M. Holland, A. Bailey, and M. B. Thomas Efficacy and economics of shelter habitats for conservation biological control. Biol. Control 45: Grout, T. G., and R. I. Richards The dietary effect of windbreak pollens on longevity and fecundity of a predacious mite Euseius addoensis addoensis (Acari: Phytoseiidae) found in citrus orchards in South Africa. Bull. Entomol. Res. 82: Gurr, G. M., S. D. Wratten, and J. M. Luna Multi-function agricultural biodiversity: pest management and other benefits. Basic Appl. Ecol. 4: Harmon, J. P., E. Stephens, and J. Losey The decline of native coccinellids (Coleoptera: Coccinellidae) in the United States and Canada. J. Insect Conserv. 11: Hassanali, A., H. Herren, Z. R. Khan, J. A. Pickett, and C. M. Woodcock Integrated pest management: the push-pull approach for controlling insect pests and weeds of cereals, and its potential for other agricultural systems including animal husbandry. Phil. Trans. R. Soc. 363: Hassell, M. P., and G. C. Varley New inductive population model for insect parasites and its bearing on biological control. Nature 223: Hawkes, J. G The ecological background of plant domestication, pp In P. J. Ucko and G. W. Oimbleby (eds.), The domestication and exploitation of plants and animals. Aldine Publishing Co., Chicago, IL. Heil, M., and J. C. S. Bueno Within-plant signaling by volatiles leads to induction and priming of an indirect plant defense in nature. Proc. Nat. Acad. Sci. USA 104:

52 Howarth, F. G Environmental impacts of classical biological control. Annu. Rev. Entomol. 36: Hoy, C. W., and C. S. Glenister Releasing Amblyseius spp. [Acarina: Phytoseiidae] to control Thrips tabaci [Thysanoptera: Thripidae] on cabbage. Entomophaga 36: Hoy, M. A Genetic improvement of Metaseiulus occidentalis implementing pesticide resistant strains - progress and problems, pp In P. J. Cameron, C. H. Wearing and W. M. Kain (eds.), Proceedings, Australasian Workshop on Development and Implementation of IPM, July 1982, Aukland, New Zealand. Entomology Division, Department of Scientific and Industrial Research. Hoy, M. A Genetic improvement of a biological control agent: multiple pesticide resistances and nondiapause in Metaseiulus occidentalis (Nesbitt), pp In Proceedings, 6th International Congress of Acarology, 5-11 September 1982, Edinburgh, Scotland. Ellis Horwood, Sussex, UK. Hoy, M. A. 1982c. Aerial dispersal and field efficacy of a genetically improved strain of the spider mite predator Metaseiulus occidentalis. Exp. Appl. Acarol. 32: Hoy, M. A Recent advances in genetics and genetic improvement of the Phytoseiidae. Annu. Rev. Entomol. 30: Hoy, M. A Criteria for release of genetically-improved phytoseiids: an examination of the risks associated with release of biological control agents. Exp. Appl. Acarol. 14: Hoy, M. A Transgenic arthropods for pest management programs: risks and realities. Exp. Appl. Acarol. 24:

53 Hoy, M. A. 2011a. The Phytoseiidae: effective natural enemies, pp Agricultural acarology: introduction to integrated mite management. Taylor and Francis Group, LLC, Boca Raton, FL. Hoy, M. A. 2011b. The strategy of integrated mite management, pp Agricultural acarology: introduction to integrated mite management. Taylor and Francis Group, LLC, Boca Raton, FL. Hoy, M. A., and N. F. Knop Selection for and genetic analysis of permethrin resistance in Metaseiulus occidentalis: genetic improvement of biological control agent. Entomol. Exp. Appl. 30: Hoyt, S. C The development of an integrated mite control program, pp In Proceedings, 62nd Annual Meeting of the Washington State Horticultural Association, 5-7 December 1966, Yakima, WA. Washington State Horticultural Association, Wenatchee, WA. Hoyt, S. C Population studies of five mite species on apple in Washington, pp In G. O. Evans (ed.) Proceedings, 2nd International Congress of Acarology, July 1967, Sutton Bonington, UK. Akadémiai Kiadó, Budapest, Hungary. Hoyt, S. C Integrated chemical control of insects and biological control of mites on apple in Washington. J. Econ. Entomol. 62: Hoyt, S. C Biology, ecology and control of mites in Washington orchards. In: Williams K. (ed) New directions in tree fruit pest management. Good Fruit Grower, Yakima, WA, pp Huang, M.-D., S.-W. Mai, S.-X. Li, and J. Situ Biological control of citrus red mite, Panonychus citri (McG.) in Guangdong Province, pp In K. Matsumoto (ed.) 39

54 Proceedings, 4th International Citrus Congress, 9 12 November 1981, Tokyo, Japan. International Society of Citriculture. Huffaker, C. B., and C. E. Kennett Some aspects of assessing efficiency of natural enemies. Can. Entomol. 101: Huffaker, C. B., and J. A. McMurtry The ecology of tetranychid mites and their natural control. Annu. Rev. Entomol. 14: Huffaker, C. B., M. Van de Vrie, and J. A. McMurtry II. Tetranychid populations and their possible control by predators: an evaluation. Hilgardia 40: Hull, L. A., and E. H. Beers Ecological selectivity: modifying chemical control practices to preserve natural enemies In M. A. Hoy and D. C. Herzog (eds.), Biological control in agricultural IPM systems. Academic Press, New York, NY. James, D. G History and perspectives of biological mite control in Australian horticulture using exotic and native phytoseiids, pp In R. B. Halliday, D. E. Walter, H. C. Proctor, R. A. Norton and M. J. Colloff (eds.), Proceedings, 10th International Congress of Acarology, 5-10 July 1998, Canberra, Australia. CSIRO, Collingwood, Australia. James, D. G Field evaluation of herbivore-induced plant volatiles as attractants for beneficial insects: methyl salicylate and the green lacewing. J. Chem. Ecol. 29: James, D. G Further field evaluation of synthetic herbivore-induced plant volatiles as attractants for beneficial insects. J. Chem. Ecol. 31: James, D. G., and T. S. Price Field-testing of methyl salicylate for recruitment and retention of beneficial insects in grapes and hops. J. Chem. Ecol. 30:

55 Jones, V. P., S. A. Steffan, N. G. Wiman, D. R. Horton, E. Miliczky, Q.-H. Zhang, and C. C. Baker Evaluation of herbivore-induced plant volatiles for monitoring green lacewings in Washington apple orchards. Biol. Control 56: Jonsson, M., S. D. Wratten, D. A. Landis, and G. M. Gurr Recent advances in conservation biological control of arthropods by arthropods. Biol. Control 45: Jung, C., and B. A. Croft Ambulatory and aerial dispersal among specialist and generalist predatory mites (Acari: Phytoseiidae). Environ. Entomol. 30: Kaplan, I Attracting carnivorous arthropods with plant volatiles: The future of biocontrol or playing with fire? Biol. Control 60: Kappers, I. R., A. Aharoni, T. W. J. M. van Herpen, L. L. P. Luckerhoff, M. Dicke, and H. J. Bouwmeester Genetic engineering of terpenoid metabolism attracts bodyguards to Arabidopsis. Science 309: Karban, R., I. T. Baldwin, K. J. Baxter, G. Laue, and G. W. Felton Communication between plants: induced resistance in wild tobacco plants following clipping of neighboring sagebrush. Oecologia 125: Kawashima, M., and C. Jung Artificial ground shelters for overwintering phytoseiid mites in orchards. Exp. Appl. Acarol. 52: Kelly, J. L., J. R. Hagler, and I. Kaplan Semiochemical lures reduce emigration and enhance pest control services in open-field predator augmentation. Biol. Control 71: Kennett, C. C., D. L. Flaherty, and R. W. Hoffmann Effect of wind-borne pollens in the population dynamics of Amblyseius hibisci. Entomophaga 24:

56 Kessler, A., and I. T. Baldwin Defensive function of herbivore-induced plant volatile emissions in nature. Science 291: Khan, Z. R., D. G. James, C. A. O. Midega, and J. A. Pickett Chemical ecology and conservation biological control. Biol. Control 45: Kinn, D. N., J. L. Joos, and R. L. Doutt Influence of overhead sprinkler systems on spider mite populations in north coast vineyards of California. Environ. Entomol. 1: Koller, M., M. Knapp, and P. Schausberger Direct and indirect adverse effects of tomato on the predatory mite Neoseiulus californicus feeding on the spider mite Tetranychus evansi. Entomol. Exp. Appl. 125: Krips, O. E., P. E. L. Willems, R. Gols, M. A. Posthumus, and M. Dicke The response of Phytoseiulus persimilis to spider mite-induced volatiles from gerbera: influence of starvation and experience. J. Chem. Ecol. 25: Krupke, C. H., R. P. Prasad, and C. M. Anelli Professional entomology and the 44 noisy years since Silent Spring. Part 2: response to Silent Spring. Am. Entomol. 53: Landis, D. A., S. D. Wratten, and G. M. Gurr Habitat management to conserve natural enemies of arthropod pests in agriculture. Annu. Rev. Entomol. 45: Laubertie, E. A., S. D. Wratten, and J.-L. Hemptinne The contribution of potential beneficial insecatary plant species to adult hoverfly (Diptera: Syrphidae) fitness. Biol. Control 61: 1-6. Lefebvre, M., N. J. Bostanian, H. M. A. Thistlewood, Y. Mauffette, and G. Racette A laboratory assessment of the toxic attributes of six reduced risk insecticides on Galendromus occidentalis (Acari: Phytoseiidae). Chemosphere 84:

57 Lefebvre, M., N. J. Bostanian, Y. Mauffette, G. Racette, H. A. Thistlewood, and J. M. Hardman Laboratory-based toxicological assessments of new insecticides on mortality and fecundity of Neoseiulus fallacis (Acari: Phytoseiidae). J. Econ. Entomol. 105: Leius, K Influence of wild flowers on parasitism of tent caterpillar and codling moth. Can. Entomol. 99: Li, J., and M. A. Hoy Adaptability and efficacy of transgenic and wild-type Metaseiulus occidentalis (Acari: Phytoseiidae) compared as part of a risk assessment. Exp. Appl. Acarol. 20: Liang, W., and M. Huang Influence of citrus orchard ground cover plants on arthropod communities in China: a review. Agric. Ecosyst. Environ. 50: Losey, J. E., and R. F. Denno Positive predator-predator interactions: Enhanced predation rates and synergistic suppression of aphid populations. Ecology 79: Luckey, T. D Insecticide hormoligosis. J. Econ. Entomol. 61: Luckmann, W. H., and R. L. Metcalf The pest-management concept, pp In R. L. Metcalf and W. H. Luckmann (eds.), Introduction to insect pest management. Wiley- Interscience, New York, NY. Madsen, H. F Integrated control of phytophagous mites on apple and pear, pp In Proceedings, 60th Annual Meeting of the Washington State Horticultural Association, Dec Washington State Horticultural Association, Wenatchee, WA. Mailloux, J., F. Le Bellec, S. Kreiter, M.-S. Tixier, and P. Dubois Influence of ground cover management on diversity and density of phytoseiid mites (Acari: Phytoseiidae) in Guadeloupean citrus orchards. Exp. Appl. Acarol. 52:

58 Mansour, F., H. Cohen, and Z. Shain Integrated mite management in apples in Israel: Augmentation of a beneficial mite and sensitivity of tetranychid and phytoseiid mites to pesticides. Phytoparasitica 21: Margolies, D. C., M. W. Sabelis, and J. Boyer, J.E Response of a phytoseiid predator to herbivore-induced plant volatiles. J. Insect Behav. 10: McDermott, G. J., and M. A. Hoy Persistence and containment of Metaseiulus occidentalis (Acari: Phytoseiidae) in Florida: risk assessment for possible releases of transgenic strains. Fla. Entomol. 80: McMurtry, J. A The use of phytoseiids for biological control: progress and future prospects, pp In M. A. Hoy (ed.) Proceedings, Conference of the Acarology Society of America, 29 November-3 December 1981, San Diego, CA. University of California Press, Oakland, CA. McMurtry, J. A Dynamics and potential impact of 'generalist' phytoseiids in agroecosystems and possibilities for establishment of exotic species. Exp. Appl. Acarol. 14: McMurtry, J. A., and G. T. Scriven Studies on the feeding, reproduction, and development of Amblyseius hibisci (Acarina: Phytoseiidae) on various food substances. Ann. Entomol. Soc. Am. 57. McMurtry, J. A., and B. A. Croft Life-styles of phytoseiid mites and their roles in biological control. Annu. Rev. Entomol. 42: McMurtry, J. A., C. B. Huffaker, and M. van de Vrie I. Tetranychid enemies: their biological characters and the impact of spray practices. Hilgardia 40:

59 McMurtry, J. A., G. J. de Moraes, and N. Famah Sourassou Revision of the lifestyles of phytoseiid mites (Acari: Phytoseiidae) and implications for biological control strategies. Syst. Appl. Acarol. 18: Metcalf, R. L Changing role of insecticides in crop protection. Annu. Rev. Entomol. 25: Miller, G. T., and S. Spoolman Food, soil, and pest management, pp Living in the environment: principles, connections, and solutions. Brooks/Cole, Cengage Learning, Belmont, CA. Mizell, R. F., and D. E. Schiffhauer Kairomone resonse, pesticide tolerance and field efficacy of the predatory mite, Neoseiulus collegae. J. Environ. Hortic. 9: Morris, M. A., B. A. Croft, and R. E. Berry Overwintering and effects of autumn habitat manipulation and carbofuran on Neoseiulus fallacis and Tetranychus urticae in peppermint. Exp. Appl. Acarol. 20: Nyrop, J. P Spatial dynamics of an acarine predator-prey system: Typhlodromus pyri (Acari: Phytoseiidae) preying on Panonychus ulmi (Acari: Tetranychidae). Environ. Entomol. 17: Oatman, E. R., J. A. McMurtry, and V. Voth Suppression of the two-spotted spider mite on strawberry with mass releases of Phytoseiulus persimilis. J. Econ. Entomol. 61: Oatman, E. R., J. A. McMurtry, F. E. Gilstrap, and V. Voth Effect of releases of Amblyseius californicus, Phytoseiulus persimilis, and Typhlodromus occidentalis on the twospotted spider mite on strawberry in southern California. J. Econ. Entomol. 70:

60 Obrycki, J. J., N. C. Elliott, and K. L. Giles Coccinellid introductions: potential for and evaluation of nontarget effects, pp In P. A. Follett and J. J. Duan (eds.), Nontarget effects of biological control. Kluwer, Dordrecht, The Netherlands. Oi, D. H Factors related to Pacific spider mite outbreaks in roadside almond trees. Ph.D. dissertation, University of California, Riverside, CA. Oi, D. H., and M. M. Barnes Predation by the western predatory mite (Acari: Phytoseiidae) on the Pacific spider mite (Acari: Tetranychidae) in the presence of road dust. Environ. Entomol. 18: Onzo, A., R. Hanna, and M. W. Sabelis The predatory mite Typhlodromalus aripo prefers green-mite induced plant odours from pubescent cassava varieties. Exp. Appl. Acarol. 58: Onzo, A., R. Hanna, K. Negloh, M. Toko, and M. W. Sabelis Biological control of cassava green mite with exotic and indigenous phytoseiid predators effects of intraguild predation and supplementary food. Biol. Control 33: Opit, G. P., J. R. Nechols, and D. C. Margolies Biological control of twospotted spider mites, Tetranychus urticae Koch (Acari: Tetranychidae), using Phytoseiulus persimilis Athias-Henriot (Acari: Phytoseiidae) on ivy geranium: assessment of predator release ratios. Biol. Control 29: Orre Gordon, G. U. S., S. D. Wratten, M. Jonsson, M. Simpson, and R. Hale 'Attract and reward': combining a herbivore-induced volatile with floral resource supplementation - multi-trophic level effects. Biol. Control 64:

61 Ozawa, M., T. Shimoda, M. Kawaguchi, G.-I. Arimura, J.-I. Horiuchi, T. Nishioka, and J. Takabayashi Lotus japonicus infested with herbivorous mites emits volatile compounds that attract predatory mites. J. Plant Res. 113: Parrella, M. P., K. M. Heinz, and L. Nunney Biological control through augmentative releases of natural enemies: a strategy whose time has come. Am. Entomol. 38: Parrott, P. J., H. E. Hodgkiss, and W. J. Schoene The apple and pear mites. New York Agr. Exp. St. No. 283: Pickett, C. H., and F. E. Gilstrap Inoculative releases of phytoseiids (Acari) for the biological control of spider mites (Acari: Tetranychidae) in corn. Environ. Entomol. 15: Pimentel, D Species diversity and insect population outbreaks. Ann. Entomol. Soc. Am. 54: 76. Pina, T., P. S. Argolo, A. Urbaneja, and J. Jacas Effect of pollen quality on the efficacy of two different life-style predatory mites against Tetranychus urticae in citrus. Biol. Control 61: Pisani Gareau, T. L., D. K. Letourneau, and C. Shennan Relative densities of natural enemies and pest insects within California hedgerows. Environ. Entomol. 42: Presnail, J., and M. A. Hoy Stable genetic transformation of a beneficial arthropod, Metaseiulus occidentalis (Acari: Phytoseiidae), by a microinjection technique. Proc. Nat. Acad. Sci. USA 89: Price, P. W., and G. P. Waldbauer Ecological aspects of pest management, pp In R. L. Metcalf and W. H. Luckmann (eds.), Introduction to insect pest management. Wiley-Interscience, New York. 47

62 Pringle, K. L., J. M. Heunis, and M. de Villiers Does dust result in mite outbreaks in apple orchards? J. Appl. Entomol. 138: Prokopy, R. J., and M. Christie Studies on releases of mass-reared organophosphate resistant Amblyseius fallacis (Garm.) predatory mites in Massachusetts commerical apple orchards. J. Appl. Entomol. 112: Purseglove, J. W The origin and spread of tropical crops, pp Tropical crops: dycotiledons. Wiley, London. Putman, W. L Life-history and behaviour of the predacious mite Typhlodromus (T.) caudiglans Schuster (Acarina: Phytoseiidae) in Ontario, with notes on the prey of related species. Can. Entomol. 94: Readshaw, J. L The ecology of tetranychid mites in Australian orchards. J. Appl. Ecol. 12: Rhodes, E. M., O. E. Liburd, C. Kelts, S. I. Rondon, and R. R. Francis Comparison of single and combination treatments of Phytoseiulus persimilis, Neoseiulus californicus, and Acramite (bifenazate) for control of twospotted spider mites in strawberries. Exp. Appl. Acarol. 39: Rice, R. E., and R. A. Jones Mites in almonds and stone fruits. Calif. Agric. 32: Risch, S. J., D. Andow, and M. A. Altieri Agroecosystem diversity and pest control: data, tentative conclusions, and new research directions. Environ. Entomol. 12: Root, R. B Organization of a plant arthropod association in simple and diverse habitats: the fauna of collards (Brassica oleracea). Ecol. Monogr. 43: Rosen, D., and C. B. Huffaker An overview of desired attributes of effective biological control agents, with particular emphasis on mites, pp In M. A. Hoy, G. L. 48

63 Cunningham and L. Knutson (eds.), Biological control of pests by mites. University of California, Special Publ., No. 3304, Berkeley, CA. Rosenthal, J. P., and S. C. Welter Tolerance to herbivory by a stemboring caterpillar in architecturally distinct maizes and wild relatives. Oecologia 102: Roush, R., J. R. Cate, and J. Kalmakoff Defining biological control. BioScience 30: 646. Russell, E. P Enemies hypothesis: a review of the effect of vegetation diversity on predatory insects and parasitoids. Environ. Entomol. 18: Sabelis, M. W., and H. E. Van de Baan Location of distant spider mite colonies by phytoseiid predators: demonstration of specific kairomones emitted by Tetranychus urticae and Panonychus ulmi. Entomol. Exp. Appl. 33: Sabelis, M. W., and F. M. Bakker How predatory mites cope with the web of their tetranychid prey: a functional view on dorsal chaetotaxy in the Phytoseiidae. Exp. Appl. Acarol. 16: Schausberger, P Inter- and intraspecific predation on immatures by adult females in Euseius finlandicus, Typhlodromus pyri and Kampimodromus aberrans (Acari: Phytoseiidae). Exp. Appl. Acarol. 21: Schausberger, P The influence of relative humidity on egg hatch in Euseius finlandicus, Typhlodromus pyri and Kampimodromus aberrans (Acari, Phytoseiidae). J. Appl. Entomol. 122: Schausberger, P., and A. Walzer Combined versus single species release of predaceous mites: Predator-predator interactions and pest suppression. Biol. Control 20:

64 Schmidt, R. A., E. H. Beers, T. R. Unruh, and D. R. Horton Releases of insectaryreared Galendromus occidentalis (Acari: Phytoseiidae) in commercial apple orchards. J. Econ. Entomol. 106: Schoonhoven, L. M., J. J. A. van Loon, and M. Dicke Insects and plants: how to apply our knowledge, pp Insect-Plant Biology, 2nd ed. Oxford University Press, New York, NY. Shimoda, T., and M. Dicke Attraction of a predator to chemical information related to nonprey: when can it be adaptive? Behav. Ecol. 11: Shrewsbury, P. M., and M. R. Hardin Evaluation of predatory mite (Acari: Phytoseiidae) releases to suppress spruce spider mites, Oligonychus ununguis (Acari: Tetranychidae), on juniper. J. Econ. Entomol. 96: Sih, A., G. Englund, and D. Wooster Emergent impacts of multiple predators on prey. TREE 13: Simberloff, D., and P. Stiling How risky is biological control. Ecology 77: Simon, S., J.-C. Bouvier, J.-F. Debras, and B. Sauphanor Biodiversity and pest managment in orchard systems. A review. Agron. Sustain. Dev. 30: Simpson, M., G. M. Gurr, A. T. Simmons, S. D. Wratten, D. G. James, G. Leeson, H. I. Nicol, and G. U. S. Orre-Gordon Attract and reward: combining chemical ecology and habitat manipulation to enhance biological control in field crops. J. Appl. Ecol. 48: Smith, A. E., and D. M. Secoy Forerunners of pesticides in classical Greece and Rome. J. Agric. Food Chem 23:

65 Smith, A. E., and D. M. Secoy A compendium of inorganic substances used in European pest control before J. Agric. Food Chem. 24: Smith, D., and D. F. Papacek Studies of the predatory mite Amblyseius victoriensis (Acarina: Phytoseiidae) in citrus orchards in south-east Queensland: control of Tegolophus australis and Phyllocoptruta oleivora (Acarina: Eriophyidae), effect of pesticides, alternative host plants and augmentative release. Exp. Appl. Acarol. 12: Snyder, W. E., and A. R. Ives Generalist predators disrupt biological control by a specialist parasitoid. Ecology 82: Stavrinides, M. C The effects of timing and rate of release on population growth of Phytoseiulus persimilis reared on Tetranychus urticae. Phytoparasitica 38: Stehr, F. W Parasitoids and predators in pest management, pp In R. L. Metcalf and W. H. Luckmann (eds.), Introduction to insect pest management. Wiley-Interscience, New York, NY. Stern, V. M., R. F. Smith, R. van don Bosch, and K. S. Hagen The integration of chemical and biological control of the spotted alfalfa aphid. Part 1. The integrated control concept. Hilgardia 29: Straub, C. S., and W. E. Snyder Species identity dominates the relationship between predator biodiversity and herbivore suppression. Ecology 87: Straub, C. S., and W. E. Snyder Increasing enemy biodiversity strengthens herbivore suppression on two plant species. Ecology 89: Straub, C. S., D. L. Finke, and W. E. Snyder Are the conservation of natural enemy biodiversity and biological control compatible goals? Biol. Control 45:

66 Strong, D. R., and R. W. Pemberton Food webs, risks of alien enemies and reform of biological control, pp In E. Wajnberg, J. K. Scott and P. C. Quimby (eds.), Evaluating indirect ecological effects of biological control. CABI Publishing, Wallingford, UK. Strong, W. B., and B. A. Croft Inoculative release of phytoseiid mites (Acarina: Phytoseiidae) into the rapidly expanding canopy of hops for control of Tetranychus urticae (Acarina: Tetranychidae). Environ. Entomol. 24: Symondson, W. O. C., K. D. Sunderland, and M. H. Greenstone Can generalist predators be effective biocontrol agents. Annu. Rev. Entomol. 47: Takabayashi, J., M. Dicke, and M. A. Posthumus Volatile herbivore-induced terpenoids in plant-mite interactions: variation caused by biotic and abiotic factors. J. Chem. Ecol. 20: Takahashi, M., M. Inoue, and A. Takafuji Management of the spider-mite population in a vinylhouse vinery by releasing Phytoseiulus persimilis Athias-Henriot onto the ground cover. Jpn. J. Appl. Entomol. Zool. 42: Thistlewood, H. M. A., D. Pree, and L. Crawford Selection and genetic analysis of permethrin resistance in Amblyseius fallacis (Garman) (Acari: Phytoseiidae) from Ontario apple orchards. Exp. Appl. Acarol. 19: Tixier, M.-S., S. Kreiter, P. Auger, and M. Weber Colonization of Languedoc vineyards by phytoseiid mites (Acari: Phytoseiidae): influence of wind and crop environment. Exp. Appl. Acarol. 22: Tixier, M.-S., S. Kreiter, B. A. Croft, and P. Auger Colonization of vineyards by Kampimodromus aberrans (Oudemans) (Acari: Phytoseiidae): dispersal from 52

67 surrounding plants as indicated by random amplified polymorphism DNA typing. Agric. Forest Entomol. 4: Trumble, J. T., and J. P. Morse Economics of integrating the predaceous mite Phytoseiulus persimilis (Acari: Phytoseiidae) with pesticides in strawberries. J. Econ. Entomol. 86: Turlings, T. C. J., and J. Ton Exploiting scents of distress: the prospect of manipulating herbivore-induced plant odours to enhance the control of agricultural pests. Curr. Opin. Plant Biol. 9: Turnbull, A. L., and D. A. Chant The practice and theory of biological control of insects in Canada. Can. J. Zool. 39: van den Bosch, R., P. S. Messenger, and A. P. Gutierrez. 1982a. Naturally occuring biological control and integrated control, pp An introduction to biological control. Plenum Press, New York, NY. van den Bosch, R., P. S. Messenger, and A. P. Gutierrez. 1982b. The nature and scope of biological control, pp An introduction to biological control. Plenum Press, New York, NY. van Driesche, R. G., S. Lyon, and C. Nunn Compatibility of spinosad with predacious mites (Acari: Phytoseiidae) used to control western flower thrips (Thysanoptera: Thripidae) in greenhouse crops. Fla. Entomol. 89: van Lenteren, J. C How not to evaluate augmentative biological control. Biol. Control 39:

68 van Rijn, P. C. J., J. Kooijman, and F. L. Wäckers The contribution of floral resources and honeydew to the performance of predatory hoverflies (Diptera: Syrphidae). Biol. Control 67: Villanueva, R., and J. F. Walgenbach Development, oviposition, and mortality of Neoseiulus fallacis (Acari: Phytoseiidae) in response to reduced-risk insecticides. J. Econ. Entomol. 98: Waite, G. K Integrated control of Tetranychus urticae in strawberries in south-east Queensland. Exp. Appl. Acarol. 5: Walter, D. E., G. N. Azam, G. Waite, and J. Hargreaves Risk assessent of an exotic biocontrol agent: Phytoseiulus persimilis (Acari: Phytoseiidae) does not establish in rainforest in south-east Queensland. Aust. J. Ecol. 23: Wilby, A., S. C. Villareal, L. P. Lan, K. L. Heong, and M. B. Thomas Functional benefits of predator species diversity depend on prey identity. Ecol. Entomol. 30: Wilson, F The conservation and augmentation of natural enemies. Pest Articles and News Summaries. Section A. Insect Control 12: Yachi, S., and M. Loreau Biodiversity and ecosystem productivity in a fluctuating environment: the insurance hypothesis. Proc. Nat. Acad. Sci. USA 96: Yano, S., and M. Osakabe Do spider mite-infested plants and spider mite trails attract predatory mites? Ecol. Res. 24:

69 CHAPTER TWO: RELEASES OF INSECTARY-REARED GALENDROMUS OCCIDENTALIS (ACARI: PHYTOSEIIDAE) IN COMMERCIAL APPLE ORCHARDS ABSTRACT Galendromus occidentalis (Nesbitt) is one of several phytoseiid species that are available for purchase to supplement endemic predator populations that are not providing sufficient control of spider mites. We performed a series of releases of commercially reared G. occidentalis in commercial apple (Malus domestica Borkhausen) orchards in Washington from 2010 to Releases of up to 50,000 mites per acre did not lead to an increase in populations of predatory mites or to a decrease in populations of pest mites. Assessments of mite numbers in shipments and quality (survival and fecundity) of those mites indicated that the commercial insectary was correctly estimating the number of predatory mites in their shipments, and that predator quality was not different than a laboratory colony. Finally, a predator-prey model that used the intrinsic rates of increase of tetranychid prey and the prey consumption rate of the predator indicated that the density of G. occidentalis required to control the prey at the action threshold was not economically feasible. We conclude that G. occidentalis cannot be used to bring about shortterm control via inundative releases in crops such as apple with large canopy volumes. INTRODUCTION Spider mites (Acari: Tetranychidae) have been a serious pest of apple (Malus domestica Borkhausen) since this crop became widely grown in the western United States, and can cause severe leaf damage under some conditions. In Washington State apple orchards, mite control has 55

70 historically been disrupted when new pesticides are introduced for control of codling moth (Cydia pomonella L.) (Hoyt 1969, Tanigoshi et al. 1983). Outbreaks of any of the three most common spider mite species - Panonychus ulmi (Koch), Tetranychus urticae Koch, and T. mcdanieli McGregor - cause noticeable discoloration (or bronzing) of leaves. At severe levels of damage, these effects may lead to premature leaf abscission and increased levels of fruit sunburn, and ultimately to reduced crop yields (Beers and Hoyt 1993). Damaging levels of mites are usually attributed to escape from the regulatory effects of natural control agents, especially those of predatory mites (Madsen 1964, Hoyt 1969, Croft 1975, Tanigoshi et al. 1983). Miticides are commonly applied when natural enemy activity is insufficient to keep spider mite populations below damage thresholds. Many miticides are harmful to predatory mites (Ibrahim and Yee 2000, Bostanian and Akalach 2006, Amin et al. 2009), decreasing their ability to provide control. This is especially detrimental because predators are often more affected by pesticides than pest mites (Hislop and Prokopy 1981, Park et al. 1996, Metzger and Pfeiffer 2002). Therefore, when pest mite populations recover from miticide applications, predatory mites may often not be present in sufficient numbers to provide effective biological control. Finally, repeated applications of pesticides increase the likelihood that pest mite species will become resistant (Hoyt and Caltagirone 1971, Helle and van de Vrie 1974). The primary biological control agent of spider mites in orchards of eastern Washington is Galendromus occidentalis (Nesbitt) (Acari: Phytoseiidae). This species is well adapted to the arid environment of the major apple-growing regions of Washington, and has been incorporated into integrated control programs since the 1960s (Hoyt 1966,1969, Hoyt and Beers 1993). Although its preferred prey are the Tetranychus spp. (McMurtry and Croft 1997), G. occidentalis can also control P. ulmi. The apple rust mite, Aculus schlechtendali (Nalepa) (Acari: 56

71 Eriophyidae) is used by this predator as a secondary food source (Hoyt 1969). However, recent increases in the incidence of spider mite outbreaks in Washington apple orchards (E.H.B., unpublished data) may indicate a breakdown in integrated control. This apparent breakdown in control is commonly attributed to the substitution of reduced risk pesticides for organophosphate insecticides. A number of the reduced-risk and organophosphate-replacement insecticides have been shown to be harmful to predatory mites (Beers et al. 2005, Martinez- Rocha et al. 2008, Bostanian et al. 2009, Lefebvre et al. 2011, Lefebvre et al. 2012). As a result of increased mite populations, some orchardists are releasing predatory mites, including G. occidentalis, in hopes of bringing about short-term control of rapidly rising pest mite populations. Predatory mites are the third most common natural enemy released in Washington organic orchards, after lady beetles and lacewings (Granatstein et al. 2010). Despite these efforts, a quarter of predator releases were either deemed unsuccessful or growers were uncertain of their success (Granatstein et al. 2010). These observations suggest that it is necessary to reevaluate releases in commercial apple orchards to determine objectively if this method is successful or cost-effective. The most conspicuous cases of successful phytoseiid releases are those that are performed in greenhouses (Hussey et al. 1965, Opit et al. 2005, Gerson and Weintraub 2007). Here, the limited area, small canopy sizes, risk of pesticide phytotoxicity (ornamentals), and controlled environment maximize the likelihood of an effective release (McMurtry 1981). Although experimental releases of predatory mites have been conducted in orchards in the past, these have nearly always been inoculative releases (Stinner 1977, Croft and MacRae 1992a). In addition, releases were typically for the purpose of establishing pesticide-resistant mites in an area where most phytoseiids have been extirpated (Croft and Barnes 1971, Thomas and 57

72 Chapman 1978, Roush and Hoy 1981, Hoy et al. 1983, McMurtry et al. 1984, Headley and Hoy 1987, Prokopy and Christie 1992, Lester et al. 1999, Marshall et al. 2000). Some releases have involved the transfer of predatory mites, using cuttings, from a location where they are abundant to a location where they are scarce (Marshall and Lester 2001, Bostanian et al. 2005, Ahmad et al. 2013). These releases were monitored for long periods of time, sometimes spanning multiple growing seasons. In addition, these experiments did not investigate the efficacy of phytoseiids as a "biological pesticide", that is, the inundative release of commercially-reared phytoseiids into orchards with the expectation of preventing economic damage in the succeeding 2-3 wk. Of the few inundative releases, all were characterized by extremely high economically unfeasible release rates (Croft and MacRae 1992b, Takano-Lee and Hoddle 2001, Fadamiro et al. 2013). The purpose of our experiment was to release G. occidentalis purchased from an insectary to evaluate the release practices and rates that are used by growers. We released known densities of G. occidentalis in commercial apple orchards to determine effects on numbers of predatory and phytophagous mites, and evaluated factors that could directly or indirectly have affected success of those releases. Those factors included possible long-term effects of pesticide residues, quality of the insectary-reared predatory mites, and predator-prey ratios theoretically needed to bring about control. MATERIALS AND METHODS The effects of releases on mite densities were evaluated in three separate field trials conducted in three growing seasons ( ). All released G. occidentalis were purchased from Biobest USA, Inc. (McFarland, CA). The company provides estimates of quantities of predators shipped to the nearest 1,000 motile predatory mites. Galendromus occidentalis were 58

73 sent on soybean leaves (Glycene max L.) infested with all stages of T. urticae. The trial in Experiment 1 used an in situ sampling method, which was abandoned in subsequent years due to low sampling fidelity. Higher release rates were also used after this experiment. In Experiment 3, predators were released at very low spider mite densities to increase the likelihood of successful control. Counts of both predator and pest mites were taken immediately before releases and once per week after releases. Experiment 1 (2010). This release was conducted in a mature block of Delicious apples in an orchard near Pasco, WA. Before releases, applications of carbaryl, emamectin benzoate, thiacloprid, and etoxazole were made. During the study, Bacillus thuringiensis subsp. kurstaki Berliner was applied for leafroller control. The experimental design was a randomized complete block design (RCBD) with six replicate plots per treatment. Trees were planted at a 2.1 by 4.6 m tree and row spacing, and the canopy was ~2 m in width and 3 m height. The tree density was ~415 trees per acre (~1024 trees per ha). Each replicate plot consisted of five trees in five rows, with three rows and at least seven trees between each 25-tree plot. Plots contained a single release tree, the four trees immediately within the row and across the row in both directions from the release tree, and the trees directly adjacent to those four trees (Fig. 2.1). Insectary-reared G. occidentalis were deployed onto the release trees on 14 July The treatments consisted of predatory mites released at 0, 5,000, or 15,000 mites per acre (0, 12, and 36 mites per tree, respectively). Treatments were randomly assigned to plots within a replicate block. Mites were released by transferring the appropriate number of adult female G. occidentalis to a soybean leaf infested with T. urticae, and attaching the leaf to the tree with a binder clip. Two sampling methods were used to assess densities of predators and prey. The first method was an in situ count of individual leaves using a headband magnifier (2.5 ) (OptiVisor, 59

74 Donegan Optical Company, Lenexa, KS). Fifteen leaves per tree in the nine sample trees were examined in situ without detaching the leaves. Only motile stages of T. urticae, P. ulmi, and G. occidentalis were counted. The second type of sample was a standard brushed leaf sample (Hoyt and Beers 1993). Five leaves from each of four trees adjacent to the release tree (Fig. 2.1) were collected from the field as a 20-leaf composite sample. Leaves were kept cool during transportation and storage. Within 24 h of collection, the sample was brushed onto glass plates coated with a thin layer of dishwashing soap using a mite brushing machine (Leedom, Mi-Wuk Village, CA). All stages of phytophagous and predatory mites were counted and recorded, but only the most common species - G. occidentalis, P. ulmi, T. urticae, and A. schlechtendali - are reported. Sampling was done each week for 3 wk after release, in addition to a prerelease sample immediately before releases were made. On the final sample date (2 August 2010), both types of counts were performed. The in situ counts were made on 15 leaves from each of the nine sample trees. After counting the mites on the attached leaf, it was detached and placed in a bag, and the composite sample of 15 leaves was brushed and counted in the manner described above. This approach allowed us to determine whether the two methods provided similar estimates of mite densities. Experiment 2 (2011). This release was done in a mature block of Golden Delicious apples near Pasco, WA. Applications of spinosad, chlorantraniliprole, imidacloprid were made before releases. No insecticides were applied during the study. The experimental design was an RCBD with six replicate plots per treatment. Trees were planted at a 2.1 by 4.6 m tree and row spacing with ~415 trees per acre (~1024 trees per ha). Replicate plots consisted of 100 trees (5 rows 20 trees). 60

75 A prerelease sample was taken in each plot to determine numbers of endemic predatory mites and numbers of herbivorous mites in plots preceding release of predatory mites. The sampling area consisted of a three row by 13 tree area within each replicate plot. Two to three leaves were randomly picked from each tree, for a composite sample of 100 leaves. Mite densities were estimated from standard brushed leaf samples, as previously described. Treatments consisted of release rates of ~50,000 and 15,000 predatory mites per acre, as well as a no-release check. Treatments were randomly assigned to each plot within a replicate block. The commercial insectary estimated that each bag of bean leaf material contained 50,000 motile predatory mites. This estimate, the weight of the bag contents, and the tree density in the plot were used to determine the amount of bean leaf material to distribute to each tree. The high release rate (50,000 G. occidentalis per acre) required 2.5 stems per tree and the low release rate required 1 stem per tree. To achieve the high release rate, half of the trees received two stems and half of the trees received three stems. The stems were placed within small (14.52 cm 2, 7.62 cm in depth) flower pots and threaded through the drainage holes. The flower pots were stapled to the tree trunk at the junction of the main scaffolds. Releases were made on 8 July Postrelease leaf-brush mite samples were performed once per week for 4 wk after release using the same methods as the prerelease brush sample. Experiment 3 (2012). A third trial was done in a mature block of Golden Delicious and Delicious apples near Mattawa, WA. Before releases, applications of chlorpyrifos, mineral oil, carbaryl, spinetoram, and thiacloprid were made. During the study, B. thuringiensis kurstaki, imidacloprid, oil, and chlorantriniliprole were applied. The experiment design was an RCBD with five replicate plots per treatment. Trees were planted at a 3.2 by 5.1 m tree and row spacing 61

76 with ~245 trees per acre (~603 trees per ha). Replicate plots consisted of 100 trees in a 5 row by 20 tree section. A prerelease sample of each replicate plot was performed as described for Experiment 2. The treatments consisted of a release at a threshold of ~0.5 tetranychids per leaf and a no-release check. Treatments were randomly assigned to each plot within a replicate block. The predators were released at rate of 15,000 per acre, or 1.5 stems per tree. The stems were stapled to the bark of a scaffold limb. Releases were made on 6 July 2012, and postrelease mite samples were performed weekly for 3 wk. Sampling was discontinued when the orchard was sprayed with abamectin and spirodiclofen to control the rapidly increasing numbers of tetranychids. Release site leaf bioassays. Survival and fecundity of the commercially reared G. occidentalis were assessed on leaves from the release (commercial) orchard and a nonsprayed research orchard in a bioassay. Reduced survivorship or fecundity of predators on the commercial orchard leaves relative to nontreated leaves could be indirect evidence that toxic pesticide residues were present. The experimental design of all bioassays was a completely randomized design. Leaves from the release site were compared with those from a nonsprayed experimental orchard at the Washington State University (WSU) Tree Fruit Research & Extension Center in Wenatchee, WA. Each treatment (sprayed or nonsprayed leaves) was replicated five (Experiments 1 and 3) or six (Experiment 2) times with five (Experiments 2 and 3), or 10 (Experiment 1) adult females per replicate. All arthropods were removed from test leaves with a fine brush before the bioassays. A disk (2 cm in diameter) was cut from each leaf and placed on water-saturated cotton in a 30-ml plastic portion cup with the lower surface of the leaf facing up. In the 2010 bioassay, 20 female T. urticae were added to each leaf disk and 62

77 allowed to oviposit for 24 h to provide food for the G. occidentalis females. In all other bioassays, mixed stages of T. urticae were added to the arenas. This was used as a precautionary measure because any miticidal compounds present on the release site leaves could have prevented adult T. urticae from laying sufficient numbers of eggs. One female G. occidentalis was transferred onto each leaf disk. The bioassay was evaluated non-destructively at 24 and 48 h. Galendromus occidentalis females were recorded as live, dead, or runoff (of the leaf), and the number of eggs they produced was counted. Locations of G. occidentalis eggs on each leaf disk were noted by marking the disks with a felt-tip pen. After the 48 h evaluation, G. occidentalis females were removed from the disks. On the fourth day after the 48 h evaluation, the number and status of G. occidentalis life stages were counted (hatched and unhatched eggs, live and dead larvae). Evaluation of predator quality. To determine the accuracy of the insectary s estimate of the number of motile stages of G. occidentalis per bag, 15 stems from each of the two bags from the insectary were weighed. The average weight of the contents of the two bags was calculated. The number of motile mites on each stem and its attached leaves was then determined using a binocular microscope. This quantity was used to estimate the total number of mites per bag. A bioassay was conducted to determine if the commercially purchased predators had unusually high adult or larval mortality, or decreased egg viability or fecundity. The insectaryreared G. occidentalis were compared to those from a laboratory colony started from a commercial apple orchard near Othello, WA in The colony was reared on T. urticae feeding on Henderson Bush lima beans (Phaseolus vulgaris L.). The previously described bioassay methods were used. All leaf disks were from a nonsprayed orchard. Each treatment (insectary-reared or laboratory colony) was replicated six times with five subjects per replicate. 63

78 Calculation of required predator-prey ratios. Because there was clear evidence that the releases were not effective, we determined the theoretical release rate needed to control spider mites of a given density. Latham and Mills (2010) developed a model for estimating predator-prey ratios required to control a target pest arthropod: Nt+1 = Nte rt + gp/r(1 e rt ), where Nt and Nt+1 represent prey population sizes at consecutive sampling dates, r is the intrinsic growth rate of the prey, g is the daily per capita consumption capacity of the predator, and P is the predator density. If Nt is assumed to be zero, model terms can be rearranged to allow predator density to be estimated (P=rNt+1/g). For the purposes of this estimate, Nt was given the value of three mites per leaf, a typical action threshold for Washington apples. Values of r for P. ulmi and T. urticae were calculated from life table analyses (Herbert 1981b, Herbert 1981a) using a matrix projection add-in (Hood 2010) for a spreadsheet (Microsoft Excel 2007, Redmond, WA). The daily per capita consumption of G. occidentalis (g=1.97 mites per day per predator) feeding on T. urticae was obtained from Lee and Davis (1968); similar data for P. ulmi were not found. A range of values for P was calculated using r for both P. ulmi and T. urticae, currently the two most common spider mite pests of apple. Leaf numbers per acre for a spur Delicious apple orchard were taken from Beers and Hull (1987) and E.H.B. (unpublished data). Per-acre tree density (899) and leaf number per tree (2,226 on a third leaf tree) were used to calculate leaves per acre (2,002,111). This estimate was then used to determine the number of predators per acre (on a per-leaf basis) needed to control a hypothetical spider mite population of three per leaf. Data summary and analysis. Cumulative mite days (CMDs) were calculated for each mite species sampled, starting with the first sampled date after the release. CMDs integrate population densities through each sample date: CMD= 0.5(Pa+Pb)Da-b 64

79 where Pa and Pb are the population densities (mean mites per leaf) at times a and b and Da-b is the number of days between time a and time b (Beers and Brunner 1999). Data were analyzed using the Statistical Analysis System (SAS Institute 1988). Brushed mite samples and the results of the bioassays were analyzed by analysis of variance (ANOVA) using PROC MIXED. Treatment means were separated using the pdiff option in the lsmeans statement of PROC MIXED. Linear regression (PROC GLM) was used to examine the relationship between the 2010 in situ versus the brushed-leaf mite counts. Data were tested for homogeneity of variances using Levene s test; nonhomogeneous data were transformed as appropriate for fecundity, egg hatch, and surviving larvae. Percentage data (mortality) were transformed arcsine(sqrt(x)). RESULTS Experiment 1 (2010). Tetranychid counts were at or below the action threshold ( mites per leaf) at the time of predator release in mid-july (Fig. 2.2). Pest mite densities increased over the duration of the test despite a prestudy application of etoxazole (acaricide, IRAC Group 10, mite growth inhibitors) on 24 May Panonychus ulmi was the dominant spider mite species in this test (94.7% of individuals). Aculus schlechtendali populations were moderate initially, between per leaf, but declined during July. Predator populations fluctuated between mites per leaf. There were no statistical differences between treatment means for any mite species or group (Table 2.1; Fig. 2.2). These findings do not provide any evidence that the released predators became established and reproduced in the treated plots. Counting method comparison. The regression of the two mite counting methods (Fig. 2.3) clearly indicates that the in situ counts had consistently lower mite numbers than the brushed 65

80 counts. In situ counts of P. ulmi had the best correlation with the brushed counts. There was a negative relationship between the two types of counts for G. occidentalis, and no obvious numerical relationship for T. urticae. Experiment 2 (2011). Tetranychid counts were at approximately the same levels at the time of release as they were at the site used in Experiment 1. Panonychus ulmi was also the dominant tetranychid mite in this block (99.8%). Aculus schlechtendali were scarce or absent throughout the sampling period. One week after release, the bean leaf material was examined and found to be desiccated and no G. occidentalis were present. Predatory mite densities increased from 0.02 to 0.25 per leaf during the sampling period. There were no differences between the treatment means of predator or prey mites per leaf in the CMDs (Table 2.1; Fig. 2.4). The absence of effects occurred despite release rates of predators that were 3 and 10 higher than the release rate recommended by insectaries (e.g., Rincon-Vitova Insectaries 2013). Experiment 3 (2012). At the time of release, phytophagous mite populations were slightly lower than they were at the sites used in Experiments 1 and 2 (about one mite per leaf). Tetranychus urticae was the dominant spider mite (65.2%), but P. ulmi was also present in substantial numbers (32.7%). Predator populations increased from 0.01 to 0.47 per leaf during the course of the experiment, but A. schlechtendali densities remained low. There were no treatment differences in pest or predator mite densities during the sampling period (Table 2.1; Fig. 2.5). In this test, the release rate was three times higher than the maximum release rate recommended by the insectary, and the predators were released before phytophagous mite numbers had reached a typical action threshold. Release site leaf bioassays. The bioassay comparing the leaves from the Experiment 1 release block to those from a nontreated research orchard indicated no differences in mortality or 66

81 fecundity of the insectary-reared G. occidentalis (Table 2.2). However, there was significantly poorer egg hatch and numbers of live larvae on the release site leaves, indicating the possible presence of sublethally toxic residue(s) on the leaves. Neither Experiment 2 nor Experiment 3 bioassays showed significant differences between release site vs. nonsprayed leaves in terms of mortality, fecundity, egg hatch, or larval survival (Table 2.2). Evaluation of predator quality. The estimate of the number of motile stages within the bags provided by the company (50,000) was similar to our estimate based on the prerelease subsample of leaf material. The 30-stem sample contained 654 motile G. occidentalis and amounted to 1.24% (by weight) of material contained in both bags; we estimated a bag contained 52,624 motile stages of G. occidentalis. Therefore, the release rates used were slightly higher than the nominal rates. The commercial insectary predatory mites tested after shipment did not differ from our laboratory-reared predatory mites in terms of adult mortality, fecundity, egg hatch, or live larvae produced per female (Table 2.3). Predator-Prey Ratio Calculations. The G. occidentalis density (P) needed to maintain T. urticae or P. ulmi populations at three mites per leaf was calculated to be predators per leaf, or a ratio of one predator to prey per leaf. Given a density of ~2 million leaves per acre, an estimated 332,557 to 457,479 G. occidentalis per acre would be required to maintain pest populations at or below the action threshold. This level of predators would cost US$4,323-$5,947 per acre (at US$13 per 1,000 predators). It should be noted that the relatively small trees used in this estimate had not achieved full canopy density, thus the per-acre predator numbers and associated cost represent the best-case scenario. Mature orchards will likely require a much higher number of predators. 67

82 DISCUSSION The experimental releases failed to result either in increased densities of predatory mites or decreased densities of pest mites. Experiment 1 releases may have failed due to toxic pesticide residues present on the leaves, which was suggested by the reduced larval survival and egg hatch associated with rearing on leaves that had been collected from the experimental plots (relative to fitness on leaves collected from a nonsprayed orchard). The orchardist indicated that carbaryl, thiacloprid, and etoxazole had previously been applied in the block. These compounds are known to have some level of toxicity to predatory mites (James and Rayner 1995, Childers et al. 2001, Metzger and Pfeiffer 2002, Kim et al. 2005). It is also possible that the sampling technique used in Experiment 1 did not accurately reflect mite populations. Although in situ counts originally were considered preferable because they are nondestructive, they underestimated the G. occidentalis populations. Their transparent body color, along with the habit of resting in the shadow of the leaf veins likely contributes to this. The in situ method was more effective at detecting P. ulmi than the other species likely because the color differentiation (red) from the leaf surface makes P. ulmi relatively visible. Tetranychus urticae were apparently either more difficult to detect, or were misidentified as P. ulmi in the field. These results indicate that pest management consultants, who typically use hand lenses in situ to estimate mite populations, are likely underestimating predatory mite numbers while more closely approximating P. ulmi numbers. This incorrect perception of predator-prey ratio may cause growers to apply pesticides more often than necessary. Because of these results, only brush counts were used in Experiments 2 and 3. Releases were not successful in Experiments 2 and 3, despite the absence of evidence of toxic residues on the release site leaves and the use of a more accurate sampling method. In 68

83 addition, the subsampling trial determined that the insectary provided more predatory mites than indicated, and our fitness trials showed that insectary mites were similar to a laboratory culture in rates of mortality and fecundity. Therefore, the lack of success was not due to predator quality and release rate estimates. Our studies have made it clear that approach of using per-acre release rates of predators has two major flaws. Release rates cannot be independent of pest density levels, because ratios (at least in the short term) are critical (Thongtab et al. 2001, Opit et al. 2004, Fadamiro et al. 2013). We attempted to address this issue by releasing when pest mite populations were at lower levels (Experiment 3). However, even under these more ideal conditions, the experimental releases strongly indicate that using predator releases to treat an emerging pest mite population does not provide adequate control. The second flaw in this method is that the canopy surface area in an acre of strawberries (for example) is very different than that of a tree crop. The "per-acre" concept, likely borrowed from the way pesticides are recommended, is inappropriate for predator releases. Using the leaf numbers and predator prey ratio calculations, we estimated that the insectary recommended release rate of 5,000 predators per acre would be effective at preventing population increases of pest mites when they do not exceed per leaf for T. urticae and P. ulmi. This target level is detectable with brushed leaf samples, but barely detectable by the more typical hand-lens examination used by Washington pest consultants. Thus, releases would have to be done essentially pro-actively, simulating an inoculative strategy more than an inundative strategy. Prior inoculative studies released more predators (1,000-2,000 per tree) than were used in these studies and a year or more was needed for effects to be observed (Hoy et al. 1983, Prokopy and 69

84 Christie 1992). Because of high release rates, these types of studies should be considered experimental or non-routine, especially because expense was not a consideration. Although our studies highlight several fundamental flaws in the use of G. occidentalis for short term mite control in orchards, these methods are used in Washington. We conjecture that this concept is derived from both greenhouse releases of phytoseiids as well as releases of lady beetles and lacewings. Effects of releases of these predators are seen within the season of release and due to their highly dispersive nature, are not expected to continue in future cropping seasons. Because lady beetles and lacewings are released more commonly than phytoseiids (Granatstein et al. 2010), it is likely that growers use of predatory mites is based from prior experience with these predators. Lastly, our studies reemphasize the importance of conservation biological control as a primary strategy, especially as regards minimizing nontarget effects of pesticides. This was apparent in Experiment 1 when potentially toxic residues may have played a role in creating the mite outbreak and prevented the establishment of released predators. At the least, releases must be accompanied by conservation biological control to achieve the desired result (Hoyt 1966, Headley and Hoy 1987, Mansour et al. 1993). Absence of toxic residues is key to integrated mite management, with or without the addition of releases. 70

85 REFERENCES CITED Ahmad, S., A. Pozzebon, and C. Duso Augmentative releases of the predatory mite Kampimodromus aberrans in organic and conventional apple orchards. Crop Prot. 52: Amin, M. M., R. F. Mizell, and R. W. Flowers Response of the predatory mite Phytoseiulus macropilis (Acari: Phytoseiidae) to pesticides and kairomones of three spider mite species (Acari: Tetranychidae), and non-prey food. Fla. Entomol. 92: Beers, E. H., and L. A. Hull Effect of European red mite (Acari: Tetranychidae) injury on vegetative growth and flowering of four cultivars of apples. Environ. Entomol. 16: Beers, E. H., and S. C. Hoyt European red mite In E. H. Beers, J. F. Brunner, M. J. Willett and G. M. Warner (eds.), Orchard pest management: a resource book for the Pacific Northwest. Good Fruit Grower, Yakima, WA. Beers, E. H., and J. F. Brunner Effects of low rates of esfenvalerate on pest and beneficial species of apple in comparison with a standard program. J. Tree Fruit Prod. 2: Beers, E. H., J. F. Brunner, J. E. Dunley, M. Doerr, and K. Granger Role of neonicotinyl insecticides in Washington apple integrated pest management. Part II. Nontarget effects on integrated mite control. J. Insect Sci. 5 (16): available online. Bostanian, N. J., and M. Akalach The effect of indoxacarb and five other insecticides on Phytoseiulus persimilis (Acari: Phytoseiidae), Amblyseius fallacis (Acari: Phytoseiidae) 71

86 and nymphs of Orius insidiosus (Hemiptera: Anthocoridae). Pest Manag. Sci. 62: Bostanian, N. J., J. Lasnier, and G. Racette A grower-friendly method to transfer predacious mites to commercial orchards. Phytoparasitica 33: Bostanian, N. J., H. A. Thistlewood, J. M. Hardman, M.-C. Laurin, and G. Racette Effect of seven new orchard pesticides on Galendromus occidentalis in laboratory studies. Pest Manag. Sci. 65: Childers, C. C., H. Aguilar, R. Villanueva, and M. M. Abou-Setta Comparative residual toxicities of pesticides to the predator Euseius mesembrinus (Acari: Phytoseiidae) on citrus in Florida. Fla. Entomol. 84: Croft, B. A Tree fruit pest management, pp In R. L. Metcalf and W. H. Luckmann (eds.), Introduction to insect pest management. John Wiley & Sons, New York, NY. Croft, B. A., and M. M. Barnes Comparative studies on four strains of Typhlodromus occidentalis. III. Evaluations of releases of insecticide resistant strains into an apple orchard ecosystem. J. Econ. Entomol. 64: Croft, B. A., and I. V. MacRae. 1992a. Persistence of Typhlodromus pyri and Metaseiulus occidentalis (Acari: Phytoseiidae) on apple after inoculative release and competition with Zetzellia mali (Acari: Stigmaeidae). Environ. Entomol. 21: Croft, B. A., and I. V. MacRae. 1992b. Biological control of apple mites by mixed populations of Metaseiulus occidentalis (Nesbitt) and Typhlodromus pyri Scheuten (Acari: Phytoseiidae). Environ. Entomol. 21:

87 Fadamiro, H. Y., C. Akotsen-Mensah, Y. Xiao, and J. Anikwe Field evaluation of predacious mites (Acari: Phytoseiidae) for biological control of citrus red mite, Panonychus citri (Trombidiformes: Tetranychidae). Fla. Entomol. 96: Gerson, U., and P. G. Weintraub Mites for the control of pests in protected cultivation. Pest Manag. Sci. 63: Granatstein, D., A. Avila, N. Lehrer, and E. Beers Organic orchards: needs and priorities. GS Long Organic Producers Meeting, January, Headley, J. C., and M. A. Hoy Benefit/cost analysis of an integrated mite management program for almonds. J. Econ. Entomol. 80: Helle, W., and M. van de Vrie Problems with spider mites. Outlook Agr. 8: Herbert, H. J. 1981a. Biology, life tables, and intrinsic rate of increase of the European red mite, Panonychus ulmi (Acarina: Tetranychidae). Can. Entomol. 113: Herbert, H. J. 1981b. Biology, life tables, and innate capacity for increase of the twospotted spider mite, Tetranychus urticae (Acarina: Tetranychidae). Can. Entomol. 113: Hislop, R. G., and R. J. Prokopy Integrated management of phytophagous mites in Massachusetts (U.S.A.) apple orchards 2. Influence of pesticides on the predator Amblyseius fallacis (Acarina: Phytoseiidae) under laboratory and field conditions. Prot. Ecol. 3: Hood, G. M PopTools, Version ( Hoy, M. A., P. H. Westigard, and S. C. Hoyt Release and evaluation of a laboratoryselected pyrethroid-resistant strain of the predaceous mite Metaseiulus occidentalis (Acari: Phytoseiidae) in southern Oregon pear orchards and a Washington apple orchard. J. Econ. Entomol. 76:

88 Hoyt, S. C The development of an integrated mite control program, pp In Proceedings, 62nd Annual Meeting of the Washington State Horticultural Association, 5-7 December 1966, Yakima, WA. Washington State Horticultural Association, Wenatchee, WA. Hoyt, S. C Integrated chemical control of insects and biological control of mites on apple in Washington. J. Econ. Entomol. 62: Hoyt, S. C., and L. E. Caltagirone The developing programs of integrated control of pests of apples in Washington and peaches in California, pp In C. B. Huffaker (ed.), Biological Control. Plenum Publishing Company, New York, NY. Hoyt, S. C., and E. H. Beers Western predatory mite In E. H. Beers, J. F. Brunner, M. J. Willett and G. M. Warner (eds.), Orchard pest Management: a resource book for the Pacific Northwest. The Good Fruit Grower, Yakima, WA. Hussey, N. W., W. J. Parr, and H. J. Gould Observations on the control of Tetranychus urticae Koch on cucumbers by the predatory mite Phytoseiulus riegli Dosse. Entomol. Exp. Appl. 8: Ibrahim, Y. B., and T. S. Yee Influence of sublethal exposure to abamectin on the biological performance of Neoseiulus longispinosus (Acari: Phytoseiidae). J. Econ. Entomol. 93: James, D. G., and M. Rayner Toxicity of viticultural pesticides to the predatory mites Amblyseius victoriensis and Typhlodromus doreenae. Plant Prot. Q. 10: Kim, S. S., S. G. Seo, J. D. Park, S. G. Kim, and D. I. Kim Effects of selected pesticides on the predatory mite Amblyseius cucumeris (Acari: Phytoseiidae). J. Entomol. Sci. 40:

89 Lee, M. S., and D. W. Davis Life history and behavior of the predatory mite Typhlodromus occidentalis in Utah. Ann. Entomol. Soc. Am. 61: Lefebvre, M., N. J. Bostanian, H. M. A. Thistlewood, Y. Mauffette, and G. Racette A laboratory assessment of the toxic attributes of six reduced risk insecticides on Galendromus occidentalis (Acari: Phytoseiidae). Chemosphere 84: Lefebvre, M., N. J. Bostanian, Y. Mauffette, G. Racette, H. A. Thistlewood, and J. M. Hardman Laboratory-based toxicological assessments of new insecticides on mortality and fecundity of Neoseiulus fallacis (Acari: Phytoseiidae). J. Econ. Entomol. 105: Lester, P. J., H. M. A. Thistlewood, D. B. Marshall, and R. Harmsen Assessment of Amblyseius fallacis (Acari: Phytoseiidae) for biological control of tetranychid mites in an Ontario peach orchard. Exp. Appl. Acarol. 23: Madsen, H. F Integrated control of phytophagous mites on apple and pear, pp In Proceedings, 60th Annual Meeting of the Washington State Horticultural Association, Dec Washington State Horticultural Association, Wenatchee, WA. Mansour, F., H. Cohen, and Z. Shain Integrated mite management in apples in Israel: Augmentation of a beneficial mite and sensitivity of tetranychid and phytoseiid mites to pesticides. Phytoparasitica 21: Marshall, D., H. Thistlewood, and R. Harmsen Release, establishment and movement of the predator Typhlodromus pyri (Acari: Phytoseiidae) on apple. Can. Entomol. 133:

90 Marshall, D. B., and P. J. Lester The transfer of Typhlodromus pyri on grape leaves for biological control of Panonychus ulmi (Acari: Phytoseiidae, Tetranychidae) in vineyards in Ontario, Canada. Biol. Control 20: Martinez-Rocha, L., E. H. Beers, and J. E. Dunley Effect of pesticides on integrated mite management in Washington State. J. Entomol. Soc. B.C. 105: McMurtry, J. A The use of phytoseiids for biological control: progress and future prospects, pp In M. A. Hoy (ed.) Proceedings, Conference of the Acarology Society of America, 29 November-3 December 1981, San Diego, CA. University of California Press, Oakland, CA. McMurtry, J. A., and B. A. Croft Life-styles of phytoseiid mites and their roles in biological control. Annu. Rev. Entomol. 42: McMurtry, J. A., H. G. Johnson, and M. H. Badii Experiments to determine effects of predator releases on populations of Oligonychus punicae [Acarina: Tetranychidae] on avocado in California. Entomophaga 29: Metzger, J. A., and D. G. Pfeiffer Topical toxicity of pesticides used in Virginia vineyards to the predatory mite, Neoseiulus fallacis (Garman). J. Entomol. Sci. 37: Opit, G. P., J. R. Nechols, and D. C. Margolies Biological control of twospotted spider mites, Tetranychus urticae Koch (Acari: Tetranychidae), using Phytoseiulus persimilis Athias-Henriot (Acari: Phytoseiidae) on ivy geranium: assessment of predator release ratios. Biol. Control 29:

91 Opit, G. P., J. R. Nechols, D. C. Margolies, and K. A. Williams Survival, horizontal distribution, and economics of releasing predatory mites (Acari: Phytoseiidae) using mechanical blowers. Biol. Control 33: Park, C. G., J. K. Yoo, and J. O. Lee Toxicity of some pesticides to twospotted spider mite (Acari: Tetranychidae) and its predator Amblyseius womersleyi (Acari: Phytoseiidae). Korean J. Appl. Entomol. 35: Prokopy, R. J., and M. Christie Studies on releases of mass-reared organophosphate resistant Amblyseius fallacis (Garm.) predatory mites in Massachusetts commerical apple orchards. J. Appl. Entomol. 112: Rincon-Vitova Insectaries Catalog of Beneficials. Roush, R. T., and M. A. Hoy Laboratory, glasshouse, and field studies of artificially selected carbaryl resistance in Metaseiulus occidentalis. J. Econ. Entomol. 74: SAS Institute SAS/Stat User s Guide, Version Cary, NC. Stinner, R. E Efficacy of inundative releases. Annu. Rev. Entomol. 22: Takano-Lee, M., and M. S. Hoddle Biological control of Oligonychus perseae (Acari: Tetranychidae) on avocado: IV. Evaluationg the efficacy of a modified mistblower to mechanically dispense Neoseiulus californicus (Acari: Phytoseiidae). Int. J. Acarol. 27: Tanigoshi, L. K., S. C. Hoyt, and B. A. Croft Basic biology and management components for mite pests and their natural enemies, pp In B. A. Croft and S. C. Hoyt (eds.), Integrated management of insect pests of pome and stone fruits. Wiley- Interscience, New York, NY. 77

92 Thomas, W. P., and L. M. Chapman Integrated control of apple pests in New Zealand 12. Introduction of two predacious phytoseiid mites, pp In M. J. Hartley (ed.) Proceedings, 31st New Zealand Weed and Pest Control Conference, 8-10 August The New Zealand Weed and Pest Control Society Inc., New Plymouth, New Zealand. Thongtab, T., A. Chandrapatya, and G. T. Baker Biology and efficacy of the predatory mite, Amblyseius longispinosus (Evans) (Acari, Phytoseiidae) as a biological control agent of Eotetranychus cendanai Rimando (Acari, Tetranychidae). J. Appl. Entomol. 125:

93 Table 2.1. Seasonal mite densities (cumulative mite days) in commercial apple orchards in Washington following release of various rates of G. occidentalis Cumulative mite days a Predators/acre n Tetranychids Predators (± SEM) (± SEM) Experiment 1 (2010) Apple rust mite (± SEM) 15, ± 0.32a ± 27.75a 6,053 ± 259a 5, ± 1.46a ± 15.11a 6,261 ± 1082a Check (0) ± 0.54a ± 23.07a 6,040 ± 777a F, P 0.76, , , 0.97 Experiment 2 (2011) 50, ± 0.63a ± 14.22a 0.00 ± 0.00a 15, ± 0.86a ± 18.56a 0.23 ± 0.23a Check (0) ± 0.54a ± 33.95a 2.10 ± 1.83a F, P 0.25, , , 0.35 Experiment 3 (2012) 15, ± 0.87a ± 44.34a ± 42.16a Check (0) ± 1.05a ± 26.10a ± 30.92a F, P 0.01, , , 0.07 a Means within columns not followed by the same letter are significantly different. For Exp. 1 and 2, df=2, 15; for Exp. 3, df=1, 8.

94 Table 2.2. Mortality, prey consumption, fecundity, egg hatch and larval survival of G. occidentalis placed on apple leaves with and without pesticide residues Leaf Source n Percent mortality a (± SEM) Eggs laid/female a (± SEM) Experiment 1 (2010) Percent egg hatch b (± SEM) Live larvae/female b (± SEM) Release site ± 5.18a 0.21 ± 0.03a ± 3.95b 0.11 ± 0.00b Non-treated ± 4.13a 0.14 ± 0.06a ± 3.25a 0.55 ± 0.15a F, P 0.07, , , < , 0.02 Experiment 2 (2011) Release site ± 10.99a 1.57 ± 0.39a ± 4.09a 0.96 ± 0.30a Non-treated ± 6.67a 2.07 ± 0.52a ± 0.73a 1.57 ±0.41a F, P 0.54, , , , 0.26 Experiment 3 (2012) Release site ± 4.90a 2.56 ± 0.26a ± 1.78a 2.08 ± 0.17a Non-treated ± 4.90a 2.08 ± 0.51a ± 0.00a 2.00 ± 0.45a F, P 0.33, , , , 0.87 a Evaluated two days after placement of females on leaves b Evaluated six days after placement of females on leaves 80

95 Table 2.3. Mortality, prey consumption, fecundity, egg hatch and larval survival of G. occidentalis from a commercial insectary and a laboratory culture Percent mortality a (± SEM) Eggs laid/female a (± SEM) Percent egg hatch b (± SEM) Live larvae/female b (± SEM) Predator Source n Commercial insectary ± 16.06a 1.93 ± 0.41a ± 2.26a 1.63 ± 0.35a Laboratory culture ± 6.67a 2.07 ± 0.52a ± 0.72a 1.57 ± 0.41a F, P 1.27, , , , 0.92 a Evaluated two days after placement of females on leaves b Evaluated six days after placement of females on leaves 81

96 Fig Plot layout for mite releases; a single replicate of a single treatment shown. 82

97 Fig Mean (± SEM) seasonal numbers of mites per leaf for Exp. 1; leaf-brush samples. Releases were made after first date of sampling (14 Jul 2010). 83

98 Fig Regression of in situ vs. brushed mite counts. Theoretical 1:1 relationship indicated by dotted line. Panonychus ulmi (F=148.09, P<0.0001; intercept 3.33 ± 0.43, slope 0.31 ± 0.03); Tetranychus urticae (F=0.18, P=0.674; intercept 0.10 ± 0.02, slope 0.01 ± 0.013), Galendromus occidentalis (F=3.87, P=0.051; intercept 0.04 ± 0.01, slope ± 0.03). 84

99 Fig Mean (± SEM) seasonal numbers of mites per leaf for Exp. 2; leaf-brush samples. Releases were made after first date of sampling (8 Jul 2011). 85

100 Fig Mean (± SEM) seasonal numbers of mites per leaf for Exp. 3; leaf-brush samples. Releases were made after first date of sampling (6 Jul 2012). 86

101 CHAPTER THREE: ASSESSMENT OF MARKING TECHNIQUES FOR THE PURPOSE OF MONITORING PHYTOSEIID RELEASES ABSTRACT When inundative releases of predatory mites are unsuccessful, it can be difficult to determine the cause. Mark/recapture techniques allow for the monitoring of released organisms in their environment. However, the small physical size of many mite species makes traditional marking techniques unfeasible. This study examined the potential of immunomarking and fluorescent powder for marking predatory mites. The effects of four different marking techniques were compared to an unmarked check for mortality, runoff, and fecundity over a 7-d period. Two of these techniques, immunomarking with egg white powder and visual marking with fluorescent powder, were also assessed for mark duration and false positive rate. While none of the marking techniques caused significant mortality or runoff, the use of fluorescent powder reduced fecundity at three of the evaluations. Both egg white powder and fluorescent powder had a low false positive rate, but the detection rate dropped rapidly (<50% for both techniques by 5 DAT), although the fluorescent powder technique was much more durable. Neither fluorescent powder nor egg white powder marking has sufficient durability to be used in a long term mark/recapture study. Further evaluation of marking techniques must be performed before inundative releases of predatory mites can be monitored. There may, however, be other uses for these markers. INTRODUCTION The ability to differentiate released predators from the native population would be of great utility in determining the success of inundative predator releases, the efficacy of which has 87

102 been questioned (Collier and Van Steenwyk 2004, Ahmad et al. 2013), especially in cropping systems with large canopies (Ch. 2). In apple orchards, for example, predatory mites are occasionally purchased by orchardists to control pest mite problems. While there is evidence that such releases can be successful if done on an inoculative basis, using phytoseiids as a biological pesticide may not achieve the desired (near-term) result because of the large numbers of predators required (Ch. 2). Enzyme-linked immunosorbent bioassay (ELISA) is a commonly used tool for disease diagnosis. It has been successfully adapted as a method to track insect dispersal (Hagler et al. 1992). Recently, a relatively inexpensive marking technique using various food proteins (egg white, cow s milk, soy protein) has been developed (Jones et al. 2006). ELISA is a potentially useful tool to determine if predatory mites survive these mass releases and if they disperse effectively. There has been no published attempt at using immunomarking for mites; the smallest arthropods tested have been thrips (V.P. Jones, personal communication). Unlike other arthropods in mark/recapture studies, these organisms are too small to mark with dots of paint or adhesive stickers, and require the use of powders or liquid sprays as markers. Additionally, in large scale releases of biological control agents, the organisms must be marked en masse for efficiency. Genetic markers and pesticide resistance have been used to track dispersal and survival of phytoseiids in the past (Whalon et al. 1982, Dunley and Croft 1994, Navajas et al. 2001), but these methods can be cost-prohibitive or time-consuming; they require establishing known genetic markers or the use of pesticide bioassays. The purpose of this study was to determine whether either immunomarking or fluorescent powder marking could be used as strategies for the detection of released predatory mites. 88

103 MATERIALS AND METHODS Colony information. The predatory mites (Galendromus occidentalis (Nesbitt)) used in the study were taken from a laboratory colony started from a commercial apple orchard near Othello, WA in Prior to the study, these mites were kept on Tetranychus urticae Koch infested lima bean (Phaseolus vulgaris L. Henderson Bush ) leaves. Arena construction. Bean leaf disks (2.2 cm diam.) were placed inside plastic cups (14.7 ml) filled with water-saturated cotton. T. urticae Koch were provisioned as prey by brushing mites from infested foliage onto the leaf disks. Additional T. urticae adults were added as necessary to provide a continuous supply of eggs. Marker treatments. Four marking treatments were tested using the newer immunomarking method and traditional fluorescent powder. The first two used powdered egg white (King Arthur Flour, Norwich, Vermont) applied using different techniques. The first technique involved placing mite-infested foliage into a paper bag. Egg white powder was added to the bag and the bag was thoroughly shaken. This process was repeated until good coverage of the foliage was obtained. For the second treatment, the egg white powder was applied by shaking it on mite-infested foliage spread out on trays using a flour sifter. The third treatment used a 10% solution of liquid egg whites (All Whites, Minnetonka, MN) applied using a Potter Spray Tower (Burkard Mfg, Rickmansworth, England). Mites were placed on a 3.5 cm bean leaf disk and sprayed with 2 ml of the egg white solution at 44.8 kpa using the intermediate nozzle. The fourth treatment used a fluorescent powder (BioQuip Products, Inc., Rancho Dominguez, CA), using the flour sifter method (Fig. 3.1). 89

104 Effects of the markers on mites. Each treatment (marking technique) was replicated with 25 individual G. occidentalis females. After treatment with the designated marker, each female was placed individually on an untreated bean leaf disk. The mites were evaluated at 0-7 days after treatment (DAT), with the 0 DAT reading taken 2 h after marking. Live, dead, and runoff (off the leaf disk) mites and the number of eggs laid (cumulative) were recorded. Mark duration. To analyze the duration of the different mark types, 25 G. occidentalis females marked with fluorescent powder and 25 unmarked females were placed on individual bean leaf disks provisioned with T. urticae prey. The mites were examined under a dissecting microscope with an ultraviolet (UV) flashlight. A blind evaluation was used to determine duration of the mark (the evaluator did not know if the replicate was from the marked or unmarked treatment). A similar method was used to monitor mites marked with egg white powder. Because these mites must be destructively sampled, 25 marked mites were used for each DAT (1, 3, and 7 d). At each sampling interval, the females were removed the leaf disks by touching a toothpick dipped in honey to their dorsum, then placing the toothpick inside an Eppendorf tube containing a 150 µl solution of 1 phosphate buffered saline (PBS; P3813; Sigma-Aldrich, St. Louis, MO) for ~3 min. This 1 PBS solution was made by diluting 10 PBS (one packet of PBS in 1 liter of water); 100 ml of 10 PBS was added to 900 ml of water to make the 1 PBS solution. After the 3 minute period, the honey had dissolved, leaving the mite in the tube when the toothpick was removed. Preliminary trials showed that the presence of honey in a sample did not cause false positives. All samples were stored at -18 C in order to be assayed simultaneously. The assay was conducted ~7 days after the 7 DAT samples were collected and frozen. Samples were completely thawed to a liquid state before use. 90

105 The immunoassays were performed as indirect ELISAs (Crowther 2001). A 50 µl aliquot of each mite sample was transferred using a pipette into the individual wells of 96-well microplates (PRO-BIND TM ; Becton Dickinson Labware, Franklin Lackes, NJ). In addition to the samples from each DAT, a positive control (5 freshly marked mites) and a negative control (5 unmarked mites) were included in each plate (also prepared in 150 µl 1 PBS). Samples were stored in an incubator overnight at 4 C. The contents of each microplate were then discarded. 300 µl 2% milk (0.6 g powdered milk/30 ml) was added to each well (the block) and plates were incubated for 1 h at 37 C. Plates were then washed three times with a solution of 1 PBS and TWEEN 20 (93773; Sigma-Aldrich) (100 ml 1 PBS, 0.5 g TWEEN in 900 ml water), followed by two washes of 1 PBS solution. A solution of anti-chicken egg albumin antibody (C-6534, Sigma-Aldrich) was made by adding 10 µl antibody to 10 ml of 1% milk (0.1 g powdered milk/10 ml water). A 100 µl aliquot of this solution was added to each well and the plate was incubated for 1 h at 37 C. The plates were then tripled washed with 1 PBS and TWEEN solution and then double washed with 1 PBS. Conjugate was made by mixing 1 µl anti-rabbit IgG (A-3687, Sigma-Aldrich) in 10 ml 1% milk. Plates were then incubated for 1 h at 37 C. This incubation was followed by three washes with 1 PBS and TWEEN solution and two washes with PBS. A substrate solution was made by mixing 2.66 g diethanolamine (D8885, Sigma-Aldrich) in 20 ml water and adjusting the ph of the solution to 9.8 using concentrated hydrochloric acid (BDH3026, VWR, West Chester, PA). The final volume of the solution was 25 ml; to reach this volume, additional water was added to the solution as necessary once the ph was adjusted. Finally, g potassium phosphate dibasic (60352, Sigma-Aldrich) was added to the solution. This solution was pipetted into the plate wells in 100 µl aliquots. 91

106 The optical density (OD) of each well was read with a dual wavelength plate reader (Emax plate reader; Molecular Devices, Sunnyvale, CA) at 450 nm. Readings were corrected using wells with 1 PBS only (no antigen). The average OD of the eight buffer-only wells was multiplied by 3 and subtracted from the OD of each well being tested (Crowther 2001). If this corrected value was positive, then that sample was recorded as positive for egg white protein. Data analysis. Data from the marker effects bioassays were analyzed using a logistic regression model (PROC GENMOD, SAS 9.3 (SAS Institute 2014)) using a logit link. Mortality and runoff were analyzed using the binomial distribution and fecundity (eggs per live female) was analyzed using the negative binomial distribution. Comparisons of treatment means were performed using pairwise single degree-of-freedom likelihood ratio contrasts (P<0.05). The percent detection for each DAT was calculated for each marking method by dividing the number of positive controls that were recorded as positive by the total number of known positives. The false negative rate for ELISA was calculated by dividing the number of negative controls that were scored as positive by the total number of negative controls (n=10, 5 negative controls per plate). Because the ELISA marking technique requires destructive sampling, and because all of the marks were read at the same time, there is not a false negative recording for each sample date, but rather for the assay as a whole. For the fluorescent powder mark, the false negative rate was calculated in the same manner as the percent detection (but using the known negatives). RESULTS AND DISCUSSION None of the marking techniques significantly increased mortality or runoff at any of the evaluations (Table 3.1). However, at 2 DAT mortality in several treatments (including the check) exceeded 20%. Mortality accumulated over time; no treatment had less than 20% mortality at 7 92

107 DAT. Runoff also increased in many of the treatments (including the check) over time, although statistical comparisons between dates were not made. In addition to observed runoff (finding mites in the damp cotton), many mites were missing (not found on the disk or on the cotton) by the 7 DAT period, as evidenced by the decreasing numbers of females over time (Table 3.1). The problem of runoff and undocumented attrition in mite leaf-disk bioassays is common, especially for longer-term bioassays (Knight et al. 1990, Bostanian et al. 2009). Based on these results, it appears that none of the types of marker have acutely toxic effects on G. occidentalis. There were differences in fecundity between treatments at 3-5 DAT (Table 3.1). At 3 DAT, mites treated with the egg white liquid and fluorescent powder had reduced fecundity compared to the other treatments. Only the fluorescent powder reduced mite fecundity compared to the check at 4 DAT. None of the treatments were different from the check at 5 DAT, but the mites treated with the two egg white powder application techniques had higher fecundity than those treated with fluorescent powder. However, the cumulative numbers of eggs were the same for all treatments at 7 DAT, indicating that any effects on fecundity are short lived. The detection rate of both the fluorescent powder and the egg white powder decreased over time (Fig. 3.2). However, the fluorescent powder mark was more durable than the egg white powder mark. The higher percentages of false positives in the fluorescent powder method (Table 3.2) were due to in-lab contamination of the controls, which would not be a problem in field marking trials (where it is improbable that large numbers of native mites in the field would come into contact with the marker). As with the marking effects assay, the number of mites found decreased at each DAT due to attrition. In other studies, marking with liquid egg white has produced longer lasting marks. For instance, over 90% of marked Drosophila suzukii (Matsumura) tested positive for egg albumin at 93

108 14 DAT (Klick et al. 2014). In this study, even the less durable marks (soy and milk) had higher detection rates than the egg white powder used in our study. Our studies seem atypical in the short duration of markers (Kelly et al. 2012, Slosky et al. 2012, Swezey et al. 2014). However, we should emphasize that these studies used liquid immunomarks, while our duration test used powders. It is possible that liquid marks could be retained longer due to better adhesion to the organism. Additionally, these studies involved much larger organisms (5-30 larger). However, even in a study involving a minute parasitoid (only 2 the length of G. occidentalis), mark retention was more successful that what we have reported here (but this study also used liquid marks). The duration of liquid egg white marks on predatory mites has not yet been tested and may be more successful. While our marks were too short term to be considered as a means of monitoring inundative releases, they may have other uses. Fluorescent powder was visible in 100% of individuals immediately after marking and remained relatively durable (~70% marked) at up to 4 DAT. Because it can be viewed non-destructively, this mark could be used in short term behavioral assays where distinguishing between two mixed groups (treatments/similar appearing species) of mites is necessary. The results of this study emphasize the difficulty of tracking very small released predators. This could be especially true when they are inundatively released into an agroecosytem where they are easily lost in crops with large canopies that already have a substantial population of native predators (Ch. 2). However, the fluorescent powder was somewhat durable, it may have marking applications in short-term laboratory studies. 94

109 REFERENCES CITED Ahmad, S., A. Pozzebon, and C. Duso Augmentative releases of the predatory mite Kampimodromus aberrans in organic and conventional apple orchards. Crop Prot. 52: Bostanian, N. J., S. Beudjekian, E. McGregor, and G. Racette A modified excised leaf disc method to estimate the toxicity of slow- and fast-acting reduced-risk acaricides to mites. J. Econ. Entomol. 102: Collier, T., and R. Van Steenwyk A critical evaluation of augmentative biological control. Biol. Control 31: Crowther, J. R The ELISA guidebook. Humana Press, Totowa, NJ. Dunley, J. E., and B. A. Croft Gene flow measured by allozymic analysis in pesticide resistant Typhlodromus pyri occurring within and near apple orchards. Exp. Appl. Acarol. 18: Hagler, J. R., A. C. Cohen, D. Dradley-Dunlop, and F. J. Enriquez New approach to mark insects for feeding and dispersal studies. Environ. Entomol. 21: Jones, V. P., J. R. Hagler, J. F. Brunner, C. C. Baker, and T. D. Wilburn An inexpensive immunomarking technique for studying movement patterns of naturally occurring insect populations. Environ. Entomol. 35: Kelly, J. L., J. R. Hagler, and I. Kaplan Employing immunomarkers to track dispersal and trophic relationships of piercing-sucking predator, Podisus maculiventris (Hemiptera: Pentatomidae). Environ. Entomol. 41:

110 Klick, J., J. C. Lee, J. R. Hagler, D. J. Bruck, and W. Q. Yang Evaluating Drosophila suzukii (Diptera: Drosophilidae) immunomarking for mark-capture research. Environ. Entomol. 152: Knight, A. L., E. H. Beers, S. C. Hoyt, and H. Riedl Acaricide bioassays with spider mites (Acari: Tetranychidae) on pome fruits: evaluation of methods and selection of discriminating concentrations for resistance monitoring. J. Econ. Entomol. 83: Navajas, M., H. Thistlewood, J. Lagnel, D. Marshall, and A. Tsagarakou Field releases of the predatory mite Neoseiulus fallacis (Acari: Phytoseiidae) in Canada, monitored by pyrethroid resistance and allozyme markers. Biol. Control 20: SAS Institute SAS/Stat User s Guide, Version Cary, NC. Slosky, L. M., E. J. Hoffmann, and J. R. Hagler A comparative study of the retention and lethality of the first and second generation arthropod protein markers. Entomol. Exp. Appl. 144: Swezey, S. L., D. J. Nieto, C. H. Pickett, J. R. Hagler, J. A. Bryer, and S. A. Machtley Dispersion and movement of the Lygus spp.parasitoid Peristenus relictus (Hymenoptera: Braconidae) in trapcropped organic strawberries. Environ. Entomol. 42: Whalon, M. E., B. A. Croft, and T. M. Mowry Introduction and survival of suceptible and pyrethroid-resistant strains of Amblyseius fallacis (Acari: Phytoseiidae) in a Michigan apple orchard. Environ. Entomol. 11:

111 Table 3.1. Results of the marker effects bioassay. For fecundity, cumulative means of eggs per live female ± SEM are reported. 97 DAT Treatment n Females % Mortality % Runoff Eggs/live female 0 Egg white powder (bag) a 0.00a 0.14 ± 0.07a Egg white powder (sifter) a 0.00a 0.08 ± 0.06a Egg white liquid a 0.00a 0.04 ± 0.04a Fluorescent powder a 0.00a 0.08 ± 0.06a Check a 4.00a 0.18 ± 0.08a χ 2 = 7.44; P= 0.11 χ 2 = 3.19; P= 0.53 χ 2 = 2.54; P= Egg white powder (bag) a 0.00a 1.05 ± 0.24a Egg white powder (sifter) a 0.00a 0.57 ± 0.16a Egg white liquid a 4.35a 0.33 ± 0.13a Fluorescent powder a 0.00a 0.48 ± 0.15a Check a 4.00a 0.77 ± 0.17a χ 2 = 4.50; P= 0.34 χ 2 = 3.58; P= 0.47 χ 2 =9.48; P= Egg white powder (bag) a 5.00a 2.13 ± 0.46a Egg white powder (sifter) a 0.00a 1.78 ± 0.38a Egg white liquid a 5.26a 0.79 ± 0.21a Fluorescent powder a 4.55a 1.00 ± 0.38a Check a 13.04a 1.88 ± 0.44a χ 2 = 7.35; P= 0.12 χ 2 = 3.91; P= 0.42 χ 2 = 8.60; P= Egg white powder (bag) a 6.25a 4.17 ± 0.53a Egg white powder (sifter) a 0.00a 3.33 ± 0.46a Egg white liquid a 5.56a 1.62 ± 0.38b Fluorescent powder a 5.00a 1.33 ± 0.54b Check a 13.64a 3.87 ± 0.73a χ 2 = 8.49; P= 0.08 χ 2 = 3.96; P= 0.41 χ 2 = 19.45; P< 0.00

112 98 DAT Treatment n Females % Mortality % Runoff Eggs/live female 4 Egg white powder (bag) a 6.25a 4.75 ± 0.70a Egg white powder (sifter) a 0.00a 3.72 ± 0.52a Egg white liquid a 5.56a 2.58 ± 0.48ab Fluorescent powder a 5.00a 1.88 ± 0.68b Check a 15.00a 4.62 ± 0.96a χ 2 = 8.74; P= 0.07 χ 2 = 4.31; P= 0.37 χ 2 = 12.46; P= Egg white powder (bag) a 6.25a 5.25 ± 0.75a Egg white powder (sifter) a 0.00a 4.67 ± 0.57a Egg white liquid a 6.25a 3.20 ± 0.63ab Fluorescent powder a 5.00a 2.47 ± 0.74b Check a 15.00a 4.64 ± 1.09ab χ 2 = 6.15; P= 0.19 χ 2 = 4.00; P= 0.41 χ 2 = 9.56; P= Egg white powder (bag) a 7.69a 5.00 ± 1.02a Egg white powder (sifter) a 0.00a 5.50 ± 1.67a Egg white liquid a 6.25a 3.50 ± 0.65a Fluorescent powder a 11.76a 3.36 ± 1.01a Check a 10.53a 6.22 ± 1.56a χ 2 = 1.68; P= 0.79 χ 2 = 2.18; P= 0.70 χ 2 = 4.58; P= Egg white powder (bag) a 7.69a 5.50 ± 1.48a Egg white powder (sifter) a 0.00a 3.67 ± 2.73a Egg white liquid a 20.00a 4.50 ± 0.29a Fluorescent powder a 11.76a 4.10 ± 1.21a Check a 10.00a 7.57 ± 1.94a df=4 for all tests. χ 2 = 3.06; P= 0.55 χ 2 = 2.49; P= 0.65 χ 2 = 3.34; P= 0.50

113 Table 3.2. Percent detection and percent of false negatives for each treatment at each DAT evaluated. Percent Detection Percent False Positives Treatment DAT (detected/total marked) (detected/total unmarked) Fluorescent Powder 0 24/24 (100%) 3/24 (12.5%) 1 14/17 (82.35%) 2/21 (9.52%) 2 12/17 (70.59%) 2/20 (10.00%) 3 12/17 (70.59%) 0/18 (0.00%) 4 11/16 (68.75%) 0/18 (0.00%) 5 5/11 (45.45%) 0/14 (0.00%) 6 3/10 (30.00%) 0/12 (0.00%) Egg White Powder 0 8/10 (80.00%) 1 8/26 (30.77%) 0/10 (0.00%) 3 1/20 (5.00%) 7 1/7 (14.29%) 99

114 Fig Galendromus occidentalis under UV light after marking with fluorescent powder 100 Fluorescent Powder Egg White Powder 80 Percent marked DAT Fig Duration of fluorescent powder and egg white powder marks represented as percentage marked over time 100

115 CHAPTER FOUR: PHYTOSEIIDS IN WASHINGTON COMMERCIAL APPLE ORCHARDS: BIODIVERSITY AND FACTORS AFFECTING ABUNDANCE ABSTRACT Galendromus occidentalis (Nesbitt) is an important biological control agent of spider mites (Acari: Tetranychidae) in Washington apple orchards. It was thought to be essentially the sole phytoseiid existing in this system, due in part to its resistance to commonly used orchard pesticides, and organophosphates in particular. To test this assumption, we conducted a survey of 102 commercial apple blocks in Washington to characterize the community of phytoseiid species. Seven phytoseiid species were found in our samples; G. occidentalis and Amblydromella caudiglans (Schuster) were found in the greatest abundance. We hypothesized that the gradual shift away from the use of organophosphates in recent decades may have caused the change in phytoseiid community structure. The survey data and information regarding the management, location, and surrounding habitat of each block, were used to determine what factors affect phytoseiid abundances. Galendromus occidentalis abundance was positively associated with conventional (vs. organic) spray programs, and the use of the acaricide bifenazate. Amblydromella caudiglans abundance was lower with bifenazate use and increased with greater herbicide strip weediness; it was also less prevalent in Golden Delicious blocks compared to other cultivars. These results indicate that A. caudiglans reaches higher abundances in orchards that lack certain agricultural disturbances, whereas G. occidentalis can survive in more disturbed environments. Surveys of this nature can provide valuable insight to potential drivers of community structure, allowing for the improvement of integrated pest management programs that incorporate conservation of newly recognized biological control agents like A. caudiglans. 101

116 INTRODUCTION Spider mites (Acari: Tetranychidae) are economically important secondary pests of apple (Malus domestica Borkhausen). Tetranychids feed on the cellular contents of leaves, including chlorophyll. Infestations of spider mites can compromise the canopy s photosynthetic function, and in severe cases, cause premature leaf abscission. Loss of photosynthetic ability can lead to poor fruit size or set, and leaf abscission also leads to sunburn (Croft 1982, Walter and Proctor 2004). Outbreaks of spider mites are often severe in commercial apple orchards with intensive pesticide use; however, high populations of spider mites are rarely detected in unmanaged apple trees. Like most secondary pests, flare ups of mite populations are attributed to a disruption of biological control (Madsen 1964, Tanigoshi et al. 1983) caused by intensive agricultural inputs and pesticides in particular (Beers et al. 2005, Prischmann et al. 2005, Martinez-Rocha et al. 2008, Duso et al. 2014). The most extensively studied biological control agent of mite pests in Washington apple orchards is the phytoseiid Galendromus occidentalis (Nesbitt). Integrated mite management (IMM), which includes the conservation of G. occidentalis, began in Washington in the 1960s and was highly successful (Hoyt 1969, Hoy 2011). This program was based on the discovery that G. occidentalis had developed resistance to broad-spectrum organophosphates. This resistance allowed G. occidentalis to persist in orchards that used organophosphates to control codling moth (Cydia pomonella L.), the most damaging pest of Washington apple, while maintaining G. occidentalis populations to provide biological control of spider mites. Unfortunately, pest mite outbreaks attributed to the breakdown of IMM have been noted in increasing frequency and severity since the early 2000s (Beers et al. 2005). This deterioration is thought to have been caused by reduced organophosphate (OP) use coupled with increased use 102

117 of reduced-risk and OP-replacement insecticides (Environmental Protection Agency 2014) for codling moth control (Beers et al. 2005, Beers and Schmidt 2014). While many phytoseiid species, including G. occidentalis, developed resistance to organophosphates (Hoyt 1969, Motoyama et al. 1970, Croft 1990), many reduced-risk/op alternative pesticides appear to have toxic effects on predatory mites (Villanueva and Walgenbach 2005, Bostanian et al. 2009, Bostanian et al. 2010, Gadino et al. 2011, Lefebvre et al. 2011, Lefebvre et al. 2012, Beers and Schmidt 2014). Apple trees that are unsprayed or minimally sprayed have a diverse phytoseiid fauna (Thistlewood 1991, Horton et al. 2002, Croft and Luh 2004). Pesticide use (especially OPs) in commercial orchards of Washington in the 1960s apparently reduced this fauna to a single resistant species, G. occidentalis (Hoyt 1991). The assumption that this was the sole phytoseiid species of any importance in commercial apple orchards went unchallenged during the next 40 years. However, substantial changes in the pesticide program occurred during that time. The number of OPs used declined due to loss of efficacy or loss of registration, and currently only a few remain with limited usage. The purpose of this study was to determine the species composition of phytoseiid mites in commercial apple orchards in Washington. Our goal was to describe the phytoseiid biodiversity of these orchards and identify factors that affect their abundance. Characterization of the species composition will allow for a more informed approach to conserving populations of important spider mite predators. 103

118 MATERIALS AND METHODS We surveyed phytoseiid mites from 102 different commercial apple orchard blocks in fruit-growing districts east of the Cascade Mountains in Washington, USA (Fig. 4.1). From each orchard, we collected a random sample of leaves (n = 100), collecting 1-2 leaves per tree. Samples were taken from late May to early September in 2011 to The date of collection and GPS coordinates were recorded for each block. All phytoseiid mites were removed from the leaf samples using a paintbrush and preserved in 70% ethanol until slide-mounting; when possible, specimens were slide-mounted immediately. The presence or absence of the following potential prey species from the leaf samples was also recorded: Panonychus ulmi (Koch), Tetranychus urticae Koch, T. mcdanieli McGregor, Aculus schlechtendali (Nalepa), and the stigmaeid predator Zetzellia mali (Ewing). Phytoseiids were mounted on a slide in modified Berlese s solution and all adult females were identified to species as per Denmark and Evans (2011). Identifications were confirmed by Dr. James McMurtry, Professor Emeritus of Entomology, University of California, Riverside. To evaluate factors affecting predatory mite populations, a survey was distributed to pest consultants regarding cultural practices, pesticide-use history, and mite problems for each sample location in the survey. The questions pertained to practices used in the three years prior to sampling. We characterized the landscape surrounding each block using Cropland Data Layer maps (US Department of Agriculture 2015); these maps provide remotely sensed data on landuse in the United States at a 30 m resolution. The maps for each sample year were imported into ArcGIS (ESRI 2010) and the area for each of four landscape types (orchards, pasture/shrubland, developed, and other) was extracted from a 50 m radius buffer around each sampled block. A 50 m radius buffer was chosen to reflect edge habitat surrounding orchards. Temperature data were 104

119 collected from the Washington State University AgWeatherNet station ( nearest to each site from 1 January to the date of sample collection. These data were used to calculate cumulative degree days (DD) at the time of sampling (Jones and Brunner 1993). The horizontal temperature cutoffs for degree day calculations were set at 10 C and 37 C, as per Mills (2008). The full list of variables collected from each block where mites were sampled can be found in Appendices Data Analysis. Abundances of the two most common species (Amblydromella caudiglans and G. occidentalis) were assessed relative to the orchard characteristics collected from each block (Appendices ) using generalized linear models. For all models we used a negative binomial distribution (PROC GENMOD, SAS 2014) based on the distribution of predatory mite abundances. We then used mixed stepwise regression to reduce the number of model parameters, using an information theoretic approach where variables were added or removed to minimize Akaike s Information Criterion (AIC). The best-fit models for each species were then used to assess factors influencing predatory mite abundance. For factors that were significant in these analyses, we used contrasts to determine differences between categories within factors. We also analyzed factors affecting the presence/absence of both A. caudiglans and G. occidentalis using step-wise logistic regression, where the presence of mites (present or absent) served as the binary response data. Similar to our generalized linear models, we conducted stepwise logistic regression to select a subset of model parameters that minimized AIC. We used JMP for these analyses (SAS Institute 2014a). Prior to all analyses, categorical survey responses with values less than ~10 were collapsed to fewer categories where possible, or omitted from the subsequent analyses (Appendix 4.3). 105

120 RESULTS Seven species of phytoseiids were found in the survey: Amblydromella caudiglans, Amblyseius andersoni (Chant), Euseius finlandicus (Oudemans), Galendromus flumenis (Chant), G. occidentalis, Neoseiulus fallacis (Garman), and Typhlodromus pyri Schueten (Table 4.1, Appendix 4.4). Galendromus occidentalis and A. caudiglans were the two most common species, both in terms of the number of sites where they were found and overall abundance (Fig. 4.1, Table 4.1). The other species were rare (<5% of individuals) and were only dominant at four of the 102 sites. Amblydromella caudiglans was the dominant species at ~ 20% of the sites and reached densities of up to 1.29 mites/leaf in one sample (Appendix 4.4). In 9.8% of sites surveyed, no phytoseiids were found in the sample. The abundance of G. occidentalis was significantly greater in orchards that had used bifenazate (Acramite 50WS, Chemtura, Middlebury, CT) compared to orchards that did not (Fig. 4.2a). Moreover, the abundance of G. occidentalis was significantly higher in orchards with greater pesticide intensity (Fig. 4.2b). In contrast, the abundance of A. caudiglans was significantly lower in orchards that used bifenazate (Fig. 4.3a). This species was also significantly more abundant in orchards with high herbicide strip weediness (Fig. 4.3b), but it was significantly less abundant in Golden Delicious blocks compared to other cultivars (Fig. 4.3c). No other factors were included in the best fit models for either species. The only orchard characteristic that was found to influence the presence (present/absent) of G. occidentalis was pesticide intensity (Fig. 4.4); this mite was more likely to be present in conventional orchards than in organic orchards, regardless of the intensity of the conventional spray program. Herbicide strip weediness and cultivar (Fig. 4.5) were the only orchard characteristics included in the best-fit model for the presence of A. caudiglans, but herbicide strip 106

121 weediness was only marginally significant (χ 2 = 5.78, P = 0.056). Amblydromella caudiglans was significantly more likely to be present in Gala blocks and less likely to be present in Golden Delicious blocks compared to other cultivars. DISCUSSION Our study shows that a diverse community of phytoseiid mites exists in commercial apple orchards of Washington State. Contrary to our assumptions, G. occidentalis was not the sole phytoseiid in commercial apple orchards. In addition to this well-characterized species, we found Amblydromella caudiglans in many orchards in high abundance, and collected five other, less common species. Similar to our findings that A. caudiglans was most common in less disturbed orchards, previous studies (mostly from eastern North America) showed that A. caudiglans was only abundant in minimally sprayed orchards or unsprayed feral trees (Oatman 1976, Berkett and Forsythe 1980, Strickler et al. 1987, Thistlewood 1991, Croft and Luh 2004, Bostanian et al. 2006). However, few studies in Washington apple orchards have recorded this mite species, and those that did were conducted in research orchards or unsprayed trees (Hoyt 1991, Horton et al. 2002, Croft and Luh 2004). The most likely reason for the high abundances of A. caudiglans in our samples is the shift away from the use of OPs for codling moth control in Washington orchards since Downing and Moilliet (1972) documented this scenario experimentally in British Columbia; G. occidentalis replaced A. caudiglans following the use of azinphosmethyl, but A. caudiglans re-established when those sprays were discontinued. At the time our samples were taken (2011 to 2013), azinphosmethyl was under regulatory phase out (Environmental Protection Agency 2012). This reduced level of OP use created an ecological opportunity for A. caudiglans to establish in a significant number of orchards. In contrast, G. occidentalis appeared 107

122 to benefit from disturbances in orchards given its higher abundance in conventional orchards. An interesting parallel can be found in studies of the shift to reduced-risk/op alternative pesticides in tart cherries in Michigan (Whalon and Korson 2008). These authors noted that growers using the newer alternative pesticides applied acaricides more frequently than did those using azinphosmethyl. The use of bifenazate was the only factor found to affect both G. occidentalis and A. caudiglans. However, by having positive effects on G. occidentalis and negative effects on A. caudiglans, use of bifenazate led to phytoseiid communities dominated by G. occidentalis. These results are in agreement with laboratory tests which show that bifenazate causes higher mortality in A. caudiglans than G. occidentalis (Ch. 5). This acaricide is labeled as safe for several mite predators including G. occidentalis, however, A. caudiglans (because of its relative obscurity) is not listed, and presumably not tested. Our results, however, suggest this species is highly susceptible to bifenazate or that bifenazate usage shifts competition between predatory mites in favor of G. occidentalis. Amblydromella caudiglans was also affected by other factors beyond bifenazate use. Previous research suggests A. caudiglans prefers more pubescent cultivars of apple (Downing and Moilliet 1967). In our samples, Golden Delicious blocks had the lowest abundances of this mite. This cultivar is known to be less pubescent than other apple cultivars (Duso et al. 2009). Trichomes on pubescent varieties can provide phytoseiids shelter from harsh climatic conditions and predators, and can trap pollen, which can be used as a secondary food source (Schmidt 2014). Unlike G. occidentalis, A. caudiglans is capable of surviving and reproducing on pollen (Putman 1962, McMurtry and Croft 1997). Additionally, some research (McMurtry and Croft 1997, McMurtry et al. 2013) has suggested that generalist predators such as A. caudiglans are 108

123 more affected (either positively or negatively) by host plant characteristics than specialists. These studies potentially explain why cultivar influences the abundance of A. caudiglans, but not G. occidentalis. Amblydromella caudiglans were also more abundant in orchards with weedy herbicide strips. Ground cover has been found to increase phytoseiid abundances (Alston 1994, Kawashima and Jung 2010, Mailloux et al. 2010). It has even been suggested that the provision of pollen from ground cover crops may shift community structure in favor of generalist phytoseiids (Aguilar-Fenollosa et al. 2011). Many of the weeds that are common in orchards are flowering plants (e.g., Taraxacum officinale Wigg, Convolvulus arvensis L., Trifolium repens L.) and provide supplementary pollen as food for phytoseiids (Gerson et al. 2003). Increased ground cover, in the form of weedier herbicide strips (as opposed to bare earth), may thus increase abundance of A. caudiglans. Ground cover can also provide a reservoir of spider mite prey, allowing predator populations to build up before spider mites become a problem in the canopy (Waite 1988, Takahashi et al. 1998). Weeds can also modify the microhabitat of the orchard floor, making it more suitable for phytoseiids (Croft and McGroarty 1977, Huang et al. 1981). However, it is also possible that weedy herbicide strips were simply associated with organic blocks or blocks that were managed with minimal chemical applications, and these factors were responsible for increased A. caudiglans abundance. In addition to pollen, A. schlechtendali is also considered an alternative food for phytoseiids. Unlike pollen, both G. occidentalis and A. caudiglans can maintain development on this food source (Blackwood et al. 2004). A. schlechtendali populations are a critical component of the IMM program of Washington because they provide G. occidentalis with prey in times of tetranychid scarcity (Hoyt 1969). Although the presence of certain prey species did not affect the 109

124 abundance of the two most common phytoseiids in our survey, this still has the potential to shape phytoseiid community structure. G. occidentalis is a specialist on Tetranychus spp. and is adapted to finding and traversing the web-nests produced by these pests (Sabelis and Van de Baan 1983, Sabelis and Bakker 1992, McMurtry and Croft 1997). A. caudiglans is more generalized, preferring eriophyid prey over tetranychids, and it becomes tangled in the webbing produced by Tetranychus spp. (Putman 1962, Clements and Harmsen 1993, Blackwood et al. 2004). These prey preferences are significant because the dominant spider mite pest in Washington apple was previously T. mcdanieli, but is now P. ulmi (Hoyt 1969, Beers and Hoyt 1993). Movement away from a Tetranychus spp. as the most common spider mite pest could have contributed to the establishment of A. caudiglans in orchards, given that it is an efficient predator of P. ulmi (Putman 1962, Putman and Herne 1964). Other authors (Whalon and Korson 2008) have considered the mite predator/prey system, especially as regards diversity, a bioindicator of system health. Our results show that the composition of phytoseiid communities in Washington apples has shifted towards more speciesrich communities that are less frequently dominated by G. occidentalis. These shifts could affect biological control of mites. For example, a large body of literature suggests that increasing predator species richness generally strengthens biological control (Straub et al. 2008, Griffin et al. 2013). However, studies examining predatory mite diversity and biological control have shown mixed results. In experiments conducted in Oregon, biological control of European red mite and twospotted spider mite improved when two species of phytoseiids were present compared to either single species alone (Croft and MacRae 1992a,b). In this system, the phytoseiids T. pyri and G. occidentalis were most abundant at different points of the growing season, and thus proved temporal complementarity in terms of mite control. Other studies 110

125 conducted in the greenhouse, laboratory, and field (Schausberger and Walzer 2001, Barber et al. 2003, Rhodes et al. 2006) have similarly found a relationship between increasing phytoseiid richness and mite suppression. However, in these cases the diverse community did not provide more effective control than the most impactful single species; this suggests that the positive effects of biodiversity observed were simply due to the fact that diverse communities contained the most voracious predator species. In contrast, some studies have shown that more diverse communities of phytoseiids are less effective at providing biological control due to intraguild predation (Abad-Moyano et al. 2010, Pina et al. 2012). In apple orchards, Z. mali is known to have the potential to disrupt or enhance biological control by phytoseiids because it consumes phytoseiid eggs, but is also an effective predator of pest mites (Clements and Harmsen 1990, Croft and MacRae 1992a,1993, Villanueva and Harmsen 1998). More work is needed, however, to understand the impacts of shifts in phytoseiid abundance and community structure on biological control in our system. Even though the lone well-adapted predator model was very successful for many decades in Washington apple, it was very reliant on a single class of insecticides and a specific spider mite genus (Tetranychus). However, the success of this model was not necessarily an indicator of ecosystem stability; its fragility became apparent with regulatory changes of the insecticides used. Large scale sampling provides an opportunity to investigate how natural variation across farms in management practices, landscapes, and abiotic conditions affect community structure and abundances of particular species. Analysis of these surveys can highlight pesticides that should be screened for non-target effects in laboratory assays and controlled field experiments. They might also be used to identify conservation biological control strategies for particular species. Finally, extensive surveys of agroecosystems may reveal previously overlooked natural 111

126 enemy biodiversity available for biological control. Our surveys in Washington apple will allow for the re-tooling of IMM through conservation of A. caudiglans via the use of selective spray programs in the post-azinphosmethyl era. 112

127 REFERENCES CITED Abad-Moyano, R., A. Urbaneja, D. Hoffmann, and P. Schausberger Effect of Euseius stipulatus on establishment and efficacy in spider mite suppression of Neoseiulus californicus and Phytoseiulus persimilis in clementine. Exp. Appl. Acarol. 50: Aguilar-Fenollosa, E., M. V. Ibanez-Gual, S. Pascual-Ruiz, M. Hurtado, and J. A. Jacas Effect of ground-cover management on spider mites and their phytoseiid natural enemies in clementine mandarin orchards (II): Top-down regulation mechanisms. Biol. Control 59: Alston, D. G Effect of apple orchard floor vegetation on density and dispersal of phytophagous and predaceous mites in Utah. Agric. Ecosyst. Environ. 50: Barber, A., C. A. M. Campbell, H. Crane, R. Lilley, and E. Tregidga Biocontrol of two-spotted spider mite Tetranychus urticae on dwarf hops by the phytoseiid mites Phytoseiulus persimilis and Neoseiulus californicus. Biocontrol Sci. Technol. 13: Beers, E. H., and S. C. Hoyt European red mite In E. H. Beers, J. F. Brunner, M. J. Willett and G. M. Warner (eds.), Orchard pest management: a resource book for the Pacific Northwest. Good Fruit Grower, Yakima, WA. Beers, E. H., and R. A. Schmidt Impacts of orchard pesticides on Galendromus occidentalis: lethal and sublethal effects. Crop Prot. 56: Beers, E. H., J. F. Brunner, J. E. Dunley, M. Doerr, and K. Granger Role of neonicotinyl insecticides in Washington apple integrated pest management. Part II. Nontarget effects on integrated mite control. J. Insect Sci. 5 (16): available online. 113

128 Berkett, L. P., and H. Y. Forsythe Predaceous mites (Acari) associated with apple foliage in Maine. Can. Entomol. 112: Blackwood, J. S., H.-K. Luh, and B. A. Croft Evaluation of prey-stage preference as an indicator of life-style type in phytoseiid mites. Exp. Appl. Acarol. 33: Bostanian, N. J., J. M. Hardman, H. A. Thistlewood, and G. Racette Effects of six selected orchard insecticides on Neoseiulus fallacis. Pest Manag. Sci. 66: Bostanian, N. J., J. M. Hardman, G. Racette, J. Franklin, and J. Lasnier Inventory of predacious mites in Quebec commercial apple orchards where integrated pest management programs are implemented. Ann. Entomol. Soc. Am. 99: Bostanian, N. J., H. A. Thistlewood, J. M. Hardman, M.-C. Laurin, and G. Racette Effect of seven new orchard pesticides on Galendromus occidentalis in laboratory studies. Pest Manag. Sci. 65: Clements, D. R., and R. Harmsen Predatory behavior and prey-stage preferences of stigmaeid and phytoseiid mites and their potential compatibility in biological control. Can. Entomol. 122: Clements, D. R., and R. Harmsen Prey preferences of adult and immature Zetzellia mali Ewing (Acari: Stigmaeidae) and Typhlodromus caudiglans Schuster (Acari: Phytoseiidae). Can. Entomol. 125: Croft, B. A Apple pest management, pp In R. L. Metcalf and W. H. Luckmann (eds.), Introduction to insect pest management. Wiley-Interscience, New York. Croft, B. A Endemic species, pp Arthropod biological control agents and pesticides. John Wiley & Sons, New York, NY. 114

129 Croft, B. A., and D. L. McGroarty The role of Amblyseius fallacis (Acarina: Phytoseiidae) in Michigan apple orchards. Farm Sci. 333: Croft, B. A., and I. V. MacRae. 1992a. Persistence of Typhlodromus pyri and Metaseiulus occidentalis (Acari: Phytoseiidae) on apple after inoculative release and competition with Zetzellia mali (Acari: Stigmaeidae). Environ. Entomol. 21: Croft, B. A., and I. V. MacRae. 1992b. Biological control of apple mites by mixed populations of Metaseiulus occidentalis (Nesbitt) and Typhlodromus pyri Scheuten (Acari: Phytoseiidae). Environ. Entomol. 21: Croft, B. A., and I. V. MacRae Biological control of apple mites: impact of Zetzellia mali (Acari: Stigmaeidae) on Typhlodromus pyri and Metaseiulus occidentalis (Acari: Phytoseiidae). Environ. Entomol. 22: Croft, B. A., and H.-K. Luh Phytoseiid mites on unsprayed apple trees in Oregon, and other western states (USA): distributions, life-style types and relevance to commercial orchards. Exp. Appl. Acarol. 33: Denmark, H. A., and G. A. Evans Phytoseiidae of North America and Hawaii (Acari: Mesostigmata). Indira Publishing House, West Bloomfield, MI. Downing, R. S., and T. K. Moilliet Relative densities of predaceous and phytophagous mites on three varieties of apple trees. Can. Entomol. 99: Downing, R. S., and T. K. Moilliet Replacement of Typhlodromus occidentalis by T. caudiglans and T. pyri (Acarina: Phytoseiidae) after cessation of sprays on apple trees. Can. Entomol. 104:

130 Duso, C., M. Fanti, A. Pozzebon, and G. Angeli Is the predatory mite Kampimodromus aberrans a candidate for the control of phytophagous mites in European apple orchards? BioControl 54: Duso, C., S. Ahmad, P. Tirello, A. Pozzebon, V. Klaric, M. Baldessari, V. Malagnini, and G. Angeli The impact of insecticides applied in apple orchards on the predatory mite Kampimodromus aberrans (Acari: Phytoseiidae). Exp. Appl. Acarol. 62: Environmental Protection Agency Azinphos-methyl uses cancellation September 30, 2012; use of existing stocks allowed through September Environmental Protection Agency Conventional reduced risk pesticide program. ESRI ArcGIS Desktop: Release 10. Redlands, CA. Gadino, A. N., V. M. Walton, and A. J. Dreves Impact of vineyard pesticides on a beneficial arthropod, Tyhlodromus pyri (Acari: Phytoseiidae), in laboratory bioassays. J. Econ. Entomol. 104: Gerson, U., R. L. Smiley, and R. Ochoa The effect of host plants and the ground cover on acarine biocontrol agents, pp Mites (Acari) for pest control. Blackwell Science, Oxford, UK. Griffin, J. N., J. E. Byrnes, and B. J. Cardinale Predator richness and prey suppression: meta-analysis reveals importance of scale and phylogenetic diversity. Ecology 94: Horton, D. R., D. A. Broers, T. Hinojosa, T. M. Lewis, E. R. Miliczky, and R. R. Lewis Diversity and phenology of predatory arthropods overwintering in cardboard bands 116

131 place in pear and apple orchards of Central Washington State. Ann. Entomol. Soc. Am. 95: Hoy, M. A Integrated mite management in Washington apple orchards, pp Agricultural acarology: introduction to integrated mite management. Taylor and Francis Group, LLC, Boca Raton, FL. Hoyt, S. C Integrated chemical control of insects and biological control of mites on apple in Washington. J. Econ. Entomol. 62: Hoyt, S. C Biology, ecology and control of mites in Washington orchards. In: Williams K. (ed) New directions in tree fruit pest management. Good Fruit Grower, Yakima, WA, pp Huang, M.-D., S.-W. Mai, S.-X. Li, and J. Situ Biological control of citrus red mite, Panonychus citri (McG.) in Guangdong Province, pp In K. Matsumoto (ed.) Proceedings, 4th International Citrus Congress, 9 12 November 1981, Tokyo, Japan. International Society of Citriculture. Jones, V. P., and J. F. Brunner Degree-day models In E. H. Beers, J. F. Brunner, M. J. Willett and G. M. Warner (eds.), Orchard pest management: a resource book for the Pacific Northwest. Good Fruit Grower, Yakima, WA. Kawashima, M., and C. Jung Artificial ground shelters for overwintering phytoseiid mites in orchards. Exp. Appl. Acarol. 52: Lefebvre, M., N. J. Bostanian, H. M. A. Thistlewood, Y. Mauffette, and G. Racette A laboratory assessment of the toxic attributes of six reduced risk insecticides on Galendromus occidentalis (Acari: Phytoseiidae). Chemosphere 84:

132 Lefebvre, M., N. J. Bostanian, Y. Mauffette, G. Racette, H. A. Thistlewood, and J. M. Hardman Laboratory-based toxicological assessments of new insecticides on mortality and fecundity of Neoseiulus fallacis (Acari: Phytoseiidae). J. Econ. Entomol. 105: Madsen, H. F Integrated control of phytophagous mites on apple and pear, pp In Proceedings, 60th Annual Meeting of the Washington State Horticultural Association, Dec Washington State Horticultural Association, Wenatchee, WA. Mailloux, J., F. Le Bellec, S. Kreiter, M.-S. Tixier, and P. Dubois Influence of ground cover management on diversity and density of phytoseiid mites (Acari: Phytoseiidae) in Guadeloupean citrus orchards. Exp. Appl. Acarol. 52: Martinez-Rocha, L., E. H. Beers, and J. E. Dunley Effect of pesticides on integrated mite management in Washington State. J. Entomol. Soc. B.C. 105: McMurtry, J. A., and B. A. Croft Life-styles of phytoseiid mites and their roles in biological control. Annu. Rev. Entomol. 42: McMurtry, J. A., G. J. de Moraes, and N. Famah Sourassou Revision of the lifestyles of phytoseiid mites (Acari: Phytoseiidae) and implications for biological control strategies. Syst. Appl. Acarol. 18: Mills, N. J Spider mites in California vineyards: temperature tolerance, effects of plant water status through leaf temperature, impact of novel pesticides and resistance management. American Vineyard Foundation Research Results

133 Motoyama, N., G. C. Rock, and W. C. Dauterman Organophosphorus resistance in an apple orchard population of Typhlodromus (Amblyseius) fallacis. J. Econ. Entomol. 63: Oatman, E. R An ecological study of arthropod populations on apple in northeastern Wisconsin: Phytoseiid mite species on the foliage. Environ. Entomol. 5: Pina, T., P. S. Argolo, A. Urbaneja, and J. Jacas Effect of pollen quality on the efficacy of two different life-style predatory mites against Tetranychus urticae in citrus. Biol. Control 61: Prischmann, D. A., D. G. James, L. C. Wright, R. D. Teneyck, and W. E. Snyder Effects of chlorpyrifos and sulfur on spider mites (Acari: Tetranychidae) and their natural enemies. Biol. Control 33: Putman, W. L Life-history and behaviour of the predacious mite Typhlodromus (T.) caudiglans Schuster (Acarina: Phytoseiidae) in Ontario, with notes on the prey of related species. Can. Entomol. 94: Putman, W. L., and D. H. C. Herne Relations between Typhlodromus caudiglans Schuster (Acarina: Phytoseiidae) and phytophagous mites in Ontario peach orchards. Can. Entomol. 96: Rhodes, E. M., O. E. Liburd, C. Kelts, S. I. Rondon, and R. R. Francis Comparison of single and combination treatments of Phytoseiulus persimilis, Neoseiulus californicus, and Acramite (bifenazate) for control of twospotted spider mites in strawberries. Exp. Appl. Acarol. 39:

134 Sabelis, M. W., and H. E. Van de Baan Location of distant spider mite colonies by phytoseiid predators: demonstration of specific kairomones emitted by Tetranychus urticae and Panonychus ulmi. Entomol. Exp. Appl. 33: Sabelis, M. W., and F. M. Bakker How predatory mites cope with the web of their tetranychid prey: a functional view on dorsal chaetotaxy in the Phytoseiidae. Exp. Appl. Acarol. 16: SAS Institute. 2014a. JMP Cary, NC, USA. SAS Institute. 2014b. SAS/Stat User s Guide, Version Cary, NC. Schausberger, P., and A. Walzer Combined versus single species release of predaceous mites: Predator-predator interactions and pest suppression. Biol. Control 20: Schmidt, R. A Leaf structures affect predatory mites (Acari: Phytoseiidae) and biological control: a review. Exp. Appl. Acarol. 62: Straub, C. S., D. L. Finke, and W. E. Snyder Are the conservation of natural enemy biodiversity and biological control compatible goals? Biol. Control 45: Strickler, K., N. Cushing, M. Whalon, and B. A. Croft Mite (Acari) species composition in Michigan apple orchards. Environ. Entomol. 16: Takahashi, M., M. Inoue, and A. Takafuji Management of the spider-mite population in a vinylhouse vinery by releasing Phytoseiulus persimilis Athias-Henriot onto the ground cover. Jpn. J. Appl. Entomol. Zool. 42: Tanigoshi, L. K., S. C. Hoyt, and B. A. Croft Basic biology and management components for mite pests and their natural enemies, pp In B. A. Croft and S. C. Hoyt (eds.), Integrated management of insect pests of pome and stone fruits. Wiley, New York. 120

135 Thistlewood, H A survey of predatory mites in Ontario apple orchards with diverse pesticide programmes. Can. Entomol. 123: US Department of Agriculture National Agriculture Statistics Service Spatial Analysis Research Section. Cropland Data Layer. Available: Villanueva, R., and J. F. Walgenbach Development, oviposition, and mortality of Neoseiulus fallacis (Acari: Phytoseiidae) in response to reduced-risk insecticides. J. Econ. Entomol. 98: Villanueva, R. T., and R. Harmsen Studies on the role of the stigmaeid predator Zetzellia mali in the acarine system of apple foliage, pp In R. Harmsen (ed.) Proceedings, Entomological Society of Ontario, Sudbury, Ontario. Entomological Society of Ontario. Waite, G. K Integrated control of Tetranychus urticae in strawberries in south-east Queensland. Exp. Appl. Acarol. 5: Walter, D. E., and H. C. Proctor Mites on plants, pp Mites: ecology, evolution, and behaviour. CABI Publishing, Wallingford, UK. Whalon, M., and P. Korson Tart cherry azinphos-methyl transition strategy. A report to the U. S. Environmental Protection Agency

136 Table 4.1. Species composition and dominance of phytoseiids at 102 sampled apple blocks. Percent individuals found Percent sites dominant Percent sites present Amblydromella caudiglans Amblyseius andersoni Euseius finlandicus Galendromus flumenis Galendromus occidentalis Neoseiulus fallacis Typhlodromus pyri Total (n)

137 Fig Locations of sampled commercial apple blocks. Shape and color of the bubbles indicate which species was dominant; if none were found an is used. Diameter of bubbles indicates total number of phytoseiids found (on a relative scale). 123

138 Fig Mean G. occidentalis abundance and a) bifenazate use; df = 1, χ 2 = 3.89, P = b) pesticide intensity; df = 4, χ 2 = 11.57, P = Con indicates conventional orchard, the number following indicates the pesticide intensity (higher intensities correspond to larger numbers), Org indicates organic orchard. Numbers under each category name indicate n responses. 124

139 Fig Mean A. caudiglans abundance and a) bifenazate use; df = 1, χ 2 = 13.50, P = b) herbicide strip weediness ranking; df = 3, χ 2 = 18.21, P = c) cultivar; df = 4, χ 2 = 10.96, P = indicates a virtually weed free herbicide strip, 3+ were all orchards with herbicide strips ranked moderately weedy or higher, with 2 intermediate between weed-free and moderately weedy. Numbers under each category name indicate n responses. 125

140 Fig Number of sites with G. occidentalis present/absent and pesticide intensity; χ2 = 11.56, df = 1; P = Con indicates conventional orchard, the number following indicates the pesticide intensity (higher intensities correspond to larger numbers), Org indicates organic orchard. 126

141 Fig Number of sites with A. caudiglans present/absent and cultivar; χ 2 = 11.62, df=4; P =

142 CHAPTER FIVE: COMPARATIVE BIOLOGY AND PESTICIDE SUSCEPTIBLITY OF AMBLYDROMELLA CAUDIGLANS AND GALENDROMUS OCCIDENTALIS AS SPIDER MITE PREDATORS IN APPLE ORCHARDS ABSTRACT The successful integrated mite management program for Washington apples was based on conservation of the mite predator Galendromus occidentalis (Nalepa). In the 1960s, this mite was assumed to be the only phytoseiid in Washington commercial apple orchards due to its preference for the most common mite pest of that time, Tetranychus mcdanieli McGregor, and its resistance to organophosphate pesticides. A recent survey of phytoseiids in Washington apple found that another phytoseiid, Amblydromella caudiglans (Schuster) has become common. It is a more generalized predator than G. occidentalis (it is not a Tetranychus spp. specialist) and is not known to be organophosphate-resistant. Here we conducted a series of experiments was conducted to compare the life history, prey consumption, and pesticide tolerance of these two species. Galendromus occidentalis developed more quickly than A. caudiglans, but had slightly lower egg survival. Although A. caudiglans attacked more Tetranychus urticae Koch eggs than G. occidentalis, it could not reproduce on this diet. Both predators performed equally well on a diet of T. urticae protonymphs. Unlike G. occidentalis, A. caudiglans experienced significant mortality when exposed to carbaryl, azinphosmethyl, and bifenazate. Both predators experienced significant mortality due to imidacloprid and spinetoram. These results highlight the key differences between these two predators; the shift away from organophosphate use and the change in dominant mite pest to Panonychus ulmi (Koch) may be driving factors for the observed increased abundance of A. caudiglans in Washington apple. 128

143 INTRODUCTION Integrated mite management (IMM) in Washington apple orchards was developed in the 1960s with conservation of Galendromus occidentalis (Nesbitt) as the cornerstone (Hoyt 1969, Hoy 2011). This predatory mite provided effective control of the most abundant spider mite pest, Tetranychus mcdanieli McGregor (Hoyt 1969). It had also developed resistance to some of the insecticides commonly used to control codling moth (Cydia pomonella L.), including the organophosphate (OP) azinphosmethyl (Hoyt 1969, Croft and Jeppson 1970, Ahlstrom and Rock 1973, McMurtry 1981). This allowed for the selective use of pesticides for control of key pests without the disruption of the biological control provided by G. occidentalis (Hoy 2011). However, pest management practices in Washington apple have changed substantially in the intervening decades. Mating disruption of codling moth was registered in 1990, and use in Washington apple increased steadily over the next decade. OP use has been gradually phased out due to loss of efficacy against codling moth and federal regulations; as of 2013, azinphosmethyl (an OP widely used for codling moth control) can no longer be used on apples (Environmental Protection Agency 2012). This has been part of a shift away from older, broad-spectrum pesticides towards reduced risk/op replacement pesticides from newer chemistry classes. Associated with these changes in IPM practices, a shift in the most common pest species has occurred, with Panonychus ulmi (Koch) outbreaks more commonly reported and T. mcdanieli outbreaks becoming increasingly rare (Beers and Hoyt 1993). Because of these changes, a survey of phytoseiid fauna in commercial apple orchards was conducted in Washington from (Ch. 4). The survey results found several other phytoseiid species in addition to G. occidentalis. Most notably, Amblydromella caudiglans (Schuster) was present at 50% of orchards sampled and was the dominant species in nearly 20% 129

144 of orchards. Higher abundances of this species were correlated with weedy herbicide strips and lack of bifenazate use, whereas G. occidentalis populations were higher in conventional (vs. organic) orchards and where bifenazate had been used (Ch. 4). These results indicate that tolerance of disturbance by certain pesticides could be the key difference between these predators. The two phytoseiids are also known to have different dietary preferences. Galendromus occidentalis is a specialist on spider mites in the genus Tetranychus; it is attracted to volatiles produced by mites while feeding and is capable of navigating the copious amounts of webbing that Tetranychus spp. produce (Sabelis and Van de Baan 1983, McMurtry and Croft 1997). However, it can also feed on other tetranychid species (including P. ulmi) and eriophyids, such as Aculus schlechtendali (Nalepa) (Hoyt and Beers 1993). This places G. occidentalis in the Type II specialist category, indicating that it is loosely specialized on tetranychids, but is commonly associated with web-producing Tetranychus spp. (McMurtry and Croft 1997, McMurtry et al. 2013). Alternatively, A. caudiglans prefers to feed on eriophyids or spider mites outside of the genus Tetranychus, and has difficulty moving in the web-nests produced by Tetranychus spp. (Putman 1962, Clements and Harmsen 1993, Blackwood et al. 2004). Unlike G. occidentalis, it can also reproduce feeding exclusively on pollen (Putman 1962, Blackwood et al. 2004). This species is placed in the Type III category (McMurtry and Croft 1997), or Type III-a (McMurtry et al. 2013). This category indicates that it is more generalized than Type II species (it can reproduce on pollen, is not associated with Tetranychus spp.) and further indicates its preferred habitat type a pubescent leaves. Therefore, a movement away from a Tetranychus 130

145 spp. (T. mcdanieli) as the most common pest species could also favor higher densities of A. caudiglans. The series of studies described here were conducted with the purpose of better describing the differences between these two common orchard phytoseiids. A series of assays was conducted to determine how these species differ in terms of life stage duration, survival, prey consumption, and pesticide resistance. These experiments serve to provide additional clarification as to why a certain phytoseiid may be abundant in one orchard, but not another. MATERIALS AND METHODS Life history. Amblydromella caudiglans were obtained from an unsprayed research orchard in Wenatchee, WA. Adult females were placed individually on lima bean (Phaseolus vulgaris L. Henderson Bush ) leaf disk arenas and monitored for oviposition every 12 h. When an egg was laid, the female was slide-mounted and identified to species following the key of Denmark and Evans (2011); if the female was A. caudiglans, the egg was then used for the life table study. Galendromus occidentalis individuals were obtained from a colony established from a field population collected from a research orchard in near Orondo, WA in June This colony was maintained on Tetranychus urticae Koch kept on lima bean plants. Individual phytoseiids from the colony were slide-mounted to confirm species identity, and the colony was checked before use to ensure contamination by other species had not occurred. Females from the G. occidentalis colony were placed individually on leaf disk arenas until oviposition. When an egg was laid, it was used in the life table study. Leaf disk arenas were constructed by cutting a 2.2 cm diameter disk from a lima bean leaf. A plastic cup (14.7 ml) was filled half-way with agar (Bacto TM Agar, BD, Sparks, MD). 131

146 When the agar had cooled, but not completely solidified, the bean leaf disks were placed into the agar with the abaxial surface facing up. The outer edges of the disks were gently pressed into the agar to hold the leaf flat and to maintain leaf turgor. A band of adhesive material (Tangle-Trap Insect Trap Coating, The Tanglefoot Company, Grand Rapids, MI) was applied in a ca. 1 cm band near the inside upper edge to prevent escape. A small tuft of cotton was adhered to the leaf disk with agar to provide the female mite with shelter. The bioassay cups were sealed with friction-fit lids in which a 1 cm diameter hole was cut and covered with surgical tape (Micropore, 3M, St. Paul, MN) for ventilation while not allowing escape. Individual eggs were monitored in these arenas, with one egg per arena. A sufficient number of individuals (replicates) were monitored so that at least n=50 individuals of each species reached adulthood. Upon hatch, juveniles were provided with T. urticae in various stages, ad libidum. Apple pollen was also brushed onto all leaf disks (regardless of species) to provide a secondary food source for A. caudiglans. After hatching, the life stage and condition (alive or dead) of the mite was recorded every 12 h. Stages were distinguished by the number of legs (larvae v protonymph) and size. Arenas were held at C with a 16:8 h day length, conditions which simulated mid-summer temperatures in central Washington. Each individual (from an egg) constituted a single replication. Life stage durations for each species were compared using PROC GLIMMIX, specifying the negative binomial distribution in the model statement; the model F-test was used to compare the two species (P<0.05). Survival and sex ratio were compared using a logistic regression model (PROC GENMOD) using a logit link. The variables (dead/live, male/female) were treated as binomial, with the binomial distribution specified in the model statement; Wald χ 2 tests were used to compare the two species (P<0.05). 132

147 Prey consumption (eggs) and fecundity. The arena for the bioassays consisted of a 2.2 cm diameter bean leaf disk cut from an untreated, uninfested bean leaf, and placed with the abaxial surface facing up in a plastic cup (14.7 ml) filled with cotton and water. Thirty T. urticae females were added to each arena and allowed to oviposit for 24 h. Adult females were removed from the disks, and egg numbers were adjusted to 40 per disk and the position of each egg was marked with a felt-tip pen. One adult female phytoseiid was added to each of 30 disks (replicates). Galendromus occidentalis were obtained from a research orchard in Orondo, WA, and A. caudiglans were obtained from apple trees in a research orchard in Wapato, WA the day prior to loading; previous experiments indicated that relative treatment differences were maintained with field-collected predators (Beers et al. 2009). Females were identified to species as described above. Remaining T. urticae eggs were counted after 48 h. Arenas were held an additional 48 h to allow the phytoseiids to oviposit. At the end of this period, all phytoseiid eggs and larvae (some eggs had hatched) were counted. Prey consumption (protonymphs) and fecundity. A second experiment, using T. urticae protonymphs instead of eggs, was conducted in a similar manner. However, 20 protonymphs from a T. urticae colony were transferred to the arena, and phytoseiid females were added immediately after. Live and dead T. urticae protonymphs were recorded at the end of the 48 h period; oviposition of the phytoseiids was counted 48 h after that evaluation. Data from both prey consumption/fecundity experiments were analyzed using PROC GLIMMIX (SAS 2014), specifying the negative binomial distribution for count data; the model F-test was used to compare the two species (P<0.05). Replicates where the phytoseiid could not be found at the end of the study were excluded from the analysis. 133

148 Pesticide toxicity. The arena for the bioassays consisted of a 2.2 cm diameter bean leaf disk cut from an untreated, uninfested bean leaf, and placed with the abaxial surface facing up in a plastic cup (14.7 ml) filled with cotton and water. A sufficient number of mixed stages of T. urticae to feed predators were added to each arena by brushing prey from infested bean leaves using a 3 cm paint brush. A small quantity of apple pollen and a single phytoseiid female was also added to each leaf disk. These mites were obtained directly from apple leaves collected the previous day from research orchards in Rock Island, WA (G. occidentalis) and Wapato, WA (A. caudiglans). A total of 25 arenas (replicates) were used for each treatment. The treatments were applied as a topical spray to the phytoseiid females on the disks, and they remained on the same disk throughout the bioassay period. Thus exposure to the pesticide combined contact, residues, and contaminated prey. The pesticide concentration used was based on the maximum label rate of the pesticide per unit surface area applied in 935 liters/ha of water. The solutions were made by mixing the appropriate amount of the formulated pesticide in 1 liter of water. Pesticides were applied with a laboratory sprayer (Potter Spray Tower, Burkard Mfg, Rickmansworth, England) set at 44.8 kpa using the intermediate nozzle. Each arena was sprayed with 2 ml of pesticide mixture (deposition at these settings was 1.94 mg solution/cm2); the checks were sprayed with distilled water. The pesticides tested were those commonly used in eastern Washington apple production, and represented a wide range of modes of action (Table 5.1). Some were included because they were indicated as potentially important factors affecting phytoseiid abundance (Ch. 4). Numbers of live, dead, and runoff phytoseiid females were recorded 48 h after treatment. After the data were recorded, all females were slide-mounted and identified to species. Data from 134

149 females of the incorrect species were not included in the analysis. Arenas were held at 20±2 C and 16:8 L:D photoperiod. Data from the female bioassays were analyzed using a logistic regression model (PROC GENMOD, SAS 9.3 (SAS Institute, 2013)) using a logit link. Mortality (dead+runoff) and runoff (alone) were treated as binomial (live/dead, runoff/live), with the binomial distribution specified in the model statement. Pesticides within a species were compared when the overall model was significant (P<0.05) using pairwise single degree-of-freedom likelihood ratio contrasts (P < 0.05). Pesticide repellency. Pesticides with high levels of runoff in the mortality bioassays were further tested for repellency. Only A. caudiglans was tested; G. occidentalis had already been screened in previous work (Beers and Schmidt-Jeffris 2015). The bioassay arena consisted of a bean leaf disk 3.5 cm in diameter. The disk was cut so that it was bisected by the midvein, which served as the division between the treated and untreated halves. The disk halves were treated by dipping them for 3 s in the appropriate pesticide; the other half was left untreated. After treatment with a pesticide, the disk was allowed to dry for ca. 1 h, and then was placed with the abaxial surface facing up in a plastic cup (30 ml) filled with cotton and water. The pesticide concentrations used were the same as in the non-target effects experiment. Each pesticide treatment was replicated five times. Ten A. caudiglans females were transferred to each disk, placing them on the midvein to avoid bias. The females were allowed to settle for ca. 2 h, then the numbers of live and dead mites on the treated versus the untreated leaf half were recorded. This evaluation was repeated two more times at ca. 2 h intervals, for a total of three evaluations. Runoff (leaving the disk 135

150 arena) was determined by subtraction, and mortality was a composite of all mites found dead, regardless of which side they were found on. Between evaluations, arenas were held at 20±2 C. Repellency was assessed with logistic regression using PROC GENMOD. Each pesticide was tested and analyzed independently. Within each pesticide the proportion at each evaluation was assessed separately using the Wald test. The research hypothesis for each of these tests was that the true underlying proportion deviated from 0.5 (50%) or equivalently, the log-odds ratio deviated from 0. Significance was declared at P<0.0167, using the Bonferroni adjustment for 3 comparisons (3 evaluations) to the 5% level of significance. RESULTS AND DISCUSSION Life history. Galendromus occidentalis had a shorter egg to adult development time than A. caudiglans (Table 4.2), but the difference was only 0.59 days. Although A. caudiglans had shorter egg and larval stages, G. occidentalis had shorter nymphal stages. Unlike G. occidentalis, A. caudiglans larvae do not feed (Putman 1962); non-feeding larvae are known to develop more quickly (Schausberger and Croft 1999). Amblydromella caudiglans had higher survival in the egg stage than G. occidentalis, but survival in all other stages was similar (Table 4.3). Although the small differences in life stage duration and survival were statistically significant, it is unlikely that they will be biologically significant in terms of predator performance. Both A. caudiglans and G. occidentalis had similar sex ratios, 74 and 62% female, respectively (Table 4.3). Although the field-collected A. caudiglans switched hosts (apple to bean) and the G. occidentalis from colony did not, it is unlikely that the host switching had significant effects on results of these experiments; all monitored individuals began life on the bean leaf arenas (as eggs). Additionally, it is unlikely that G. occidentalis had experienced fitness consequences from long- 136

151 term rearing; this study took place 1-2 months after establishing the colony from field-collected mites. Prey consumption (eggs) and fecundity. Amblydromella caudiglans consumed more T. urticae eggs than G. occidentalis (Table 4.4). However, this diet resulted in reduced fecundity for A. caudiglans (Table 4.4). Previous research has suggested that T. urticae eggs are a poor source of nutrition for A. caudiglans, which could explain the reduced oviposition (Putman 1962). Putman (1962) also suggested that A. caudiglans might be puncturing and draining eggs, but not necessarily consuming their contents. Generalist phytoseiids feeding on T. urticae are known to prefer motile life stages over eggs (Blackwood et al. 2001). Prey consumption (protonymphs) and fecundity. Both phytoseiids consumed similar numbers of prey when fed a diet of T. urticae protonymphs (Table 4.4). Both species also laid a similar number of eggs on this diet. The over 4 increase in fecundity of A. caudiglans on this diet, as opposed to the eggs-only diet, supports the hypothesis that the egg diet was nutritionally inadequate. However, because these two studies were not conducted simultaneously, they cannot be compared statistically. Because A. caudiglans is less capable of reproducing on one stage (eggs) of T. urticae, this may have limited its ability to reach high abundances when Tetranychus spp. were more common pests of orchards than they are at present (Beers and Hoyt 1993). Reduced fecundity on this prey may be a contributing factor to the previous dominance of G. occidentalis in commercial apple orchards. Additionally, A caudiglans, unlike G. occidentalis, has difficulty navigating the webbing produced by Tetranychus spp. females and can even become entangled and die in spider mite colonies (Putman 1962, Sabelis and Bakker 1992, McMurtry and Croft 1997). The shift to P. ulmi (which produces little webbing) as the most 137

152 common pest species may have allowed for an increase in A. caudiglans populations. However, the availability of other prey or alternative food (pollen) could mitigate these issues. Pesticide toxicity. Mortality was not significantly different from the check for novaluron, spinetoram, and chlorantraniliprole for A. caudiglans (Table 4.5). For G. occidentalis, all pesticides except imidacloprid and spinetoram did not cause mortality different from the check (Table 4.5). Comparing mortality for these species emphasizes that although some pesticides were acutely toxic to both species (imidacloprid, spinetoram), many pesticides were toxic to A. caudiglans and not G. occidentalis (carbaryl, azinphosmethyl, bifenazate) (Table 4.5). OP resistance is well-documented in G. occidentalis (Croft and Jeppson 1970, Ahlstrom and Rock 1973, McMurtry 1981); carbaryl resistance has also been reported, but to a lesser extent (Babcock and Tanigoshi 1988, Beers and Schmidt 2014). Amblydromella caudiglans is known to be susceptible OPs; OP use was documented to cause a replacement of A. caudiglans by G. occidentalis in an apple orchard in British Columbia (Downing and Moilliet 1967). These studies provide further evidence that the resistance of G. occidentalis to older classes of insecticides may have allowed for its previous uniform dominance in Washington commercial apple orchards. Few OP insecticides remain in use in Washington apple orchards, providing opportunities for susceptible predators, like A. caudiglans, to establish. Unlike azinphosmethyl, carbaryl is still used in some conventional orchards, but not in organic production. Indeed, its lack of acute effects on G. occidentalis (as well the resistance of G. occidentalis to other pesticides) likely explains the higher abundances of G. occidentalis found in conventional orchards when compared to organic (Ch. 4). Bifenazate is an interesting case in that it is considered a selective acaricide; the label lists five species of predatory mites (including G. occidentalis) which are not adversely affected 138

153 by the product. However, it did cause significant mortality in A. caudiglans. The absence of A. caudiglans from the list of species tested is understandable, since its discovery as an important predator in commercial orchards is very recent; however, it emphasizes the importance of species-specific testing for nontarget effects of reduced-risk pesticides. In terms of the population dynamics of these two phytoseiids, bifenazate has been correlated with both higher G. occidentalis populations and lower A. caudiglans populations (Ch. 4). Use of this pesticide clearly has strong potential to drive competition in favor of G. occidentalis. For many of the pesticides tested, the most impactful negative effects on A. caudiglans were acute, rather than sublethal; this is in contrast to previous work with G. occidentalis (Beers and Schmidt 2014). Sublethal effects (with greatly reduced acute toxicity) of pesticides to G. occidentalis are possibly evidence of resistance development. The differences in mortality are likely a mechanism capable of shifting competition in favor of G. occidentalis in orchards where these pesticides are sprayed, even if G. occidentalis experiences some sublethal effects (e.g., azinphosmethyl) (Beers and Schmidt 2014). The exception would be pesticides that cause overwhelming sublethal effects in G. occidentalis (e.g,. spirotetramat) (Beers and Schmidt 2014). Analysis of runoff indicated that azinphosmethyl, spinetoram, and bifenazate were irritant to A. caudiglans, whereas imidacloprid and spinetoram were irritant to G. occidentalis (Table 4.5). The repellency assay confirms that all three pesticides causing significant runoff were also repellent to A. caudiglans (Table 4.6). These results further highlight the sensitivity of A. caudiglans to bifenazate and azinphosmethyl, whereas both species were sensitive to spinetoram. In contrast, previous research with G. occidentalis found little evidence of spinetoram irritancy or repellency (Beers and Schmidt-Jeffris 2015); it appears that the effects of this pesticide may vary with population. Imidacloprid is known to be repellent to G. occidentalis (Bostanian et al. 139

154 2009) and in our study (where mortality was calculated as dead+runoff), runoff made a larger contribution to total mortality. The opposite was true for A. caudiglans, indicating that this is yet another pesticide where the acute effects seem to be more pronounced for this phytoseiid than G. occidentalis. Conclusions. These results provide experimental evidence supporting a recent Washington survey (Ch. 4), and previous work with these species in other regions (Downing and Moilliet 1972). While G. occidentalis can thrive under intensive OP use, several of the newer pesticides have the ability to decrease A. caudiglans or G. occidentalis densities. A. caudiglans also cannot survive on a diet of eggs of a Tetranychus spp., and has difficulty navigating their dense webbing. Thus it is possible that a change in both pesticide regimes (away from OPs) and dominant prey species (to P. ulmi) provided the conditions that allowed this predator to become more common in Washington apple orchards in recent years. 140

155 REFERENCES CITED Ahlstrom, K. R., and G. C. Rock Comparative studies on Neoseiulus fallacis and Metaseiulus occidentalis for azinphosmethyl toxicity and effects of prey and pollen on growth. Ann. Entomol. Soc. Am. 66: Babcock, J. M., and L. K. Tanigoshi Resistance levels of Typhlodromus occidentalis (Acari: Phytoseiidae) from Washington apple orchards to ten pesticides. Exp. Appl. Acarol. 4: Beers, E. H., and S. C. Hoyt European red mite In E. H. Beers, J. F. Brunner, M. J. Willett and G. M. Warner (eds.), Orchard pest management: a resource book for the Pacific Northwest. Good Fruit Grower, Yakima, WA. Beers, E. H., and R. A. Schmidt Impacts of orchard pesticides on Galendromus occidentalis: lethal and sublethal effects. Crop Prot. 56: Beers, E. H., and R. A. Schmidt-Jeffris Effects of orchard pesticides on Galendromus occidentalis (Acari: Phytoseiidae): repellency and irritancy. J. Econ. Entomol. 108: Beers, E. H., L. Martinez-Rocha, R. R. Talley, and J. E. Dunley Lethal, sublethal, and behavioral effects of sulfur-containing products in bioassays of three species of orchard mites. J. Econ. Entomol. 102: Blackwood, J. S., P. Schausberger, and B. A. Croft Prey-stage preference in generalist and specialist phytoseiid mites (Acari: Phytoseiidae) when offered Tetranychus urticae (Acrari: Tetranychidae) eggs and larvae. Environ. Entomol. 30: Blackwood, J. S., H.-K. Luh, and B. A. Croft Evaluation of prey-stage preference as an indicator of life-style type in phytoseiid mites. Exp. Appl. Acarol. 33:

156 Bostanian, N. J., H. A. Thistlewood, J. M. Hardman, M.-C. Laurin, and G. Racette Effect of seven new orchard pesticides on Galendromus occidentalis in laboratory studies. Pest Manag. Sci. 65: Clements, D. R., and R. Harmsen Prey preferences of adult and immature Zetzellia mali Ewing (Acari: Stigmaeidae) and Typhlodromus caudiglans Schuster (Acari: Phytoseiidae). Can. Entomol. 125: Croft, B. A., and L. R. Jeppson Comparative studies on four strains of Typhlodromus occidentalis. II. Laboratory toxicity of ten compounds common to apple pest control. J. Econ. Entomol. 63: Denmark, H. A., and G. A. Evans Phytoseiidae of North America and Hawaii (Acari: Mesostigmata). Indira Publishing House, West Bloomfield, MI. Downing, R. S., and T. K. Moilliet Relative densities of predaceous and phytophagous mites on three varieties of apple trees. Can. Entomol. 99: Downing, R. S., and T. K. Moilliet Replacement of Typhlodromus occidentalis by T. caudiglans and T. pyri (Acarina: Phytoseiidae) after cessation of sprays on apple trees. Can. Entomol. 104: Environmental Protection Agency Azinphos-methyl uses cancellation September 30, 2012; use of existing stocks allowed through September Hoy, M. A Integrated mite management in Washington apple orchards, pp Agricultural acarology: introduction to integrated mite management. Taylor and Francis Group, LLC, Boca Raton, FL. 142

157 Hoyt, S. C Integrated chemical control of insects and biological control of mites on apple in Washington. J. Econ. Entomol. 62: Hoyt, S. C., and E. H. Beers Western predatory mite In E. H. Beers, J. F. Brunner, M. J. Willett and G. M. Warner (eds.), Orchard pest Management: a resource book for the Pacific Northwest. The Good Fruit Grower, Yakima, WA. McMurtry, J. A The use of phytoseiids for biological control: progress and future prospects, pp In M. A. Hoy (ed.) Proceedings, Conference of the Acarology Society of America, 29 November-3 December 1981, San Diego, CA. University of California Press, Oakland, CA. McMurtry, J. A., and B. A. Croft Life-styles of phytoseiid mites and their roles in biological control. Annu. Rev. Entomol. 42: McMurtry, J. A., G. J. de Moraes, and N. Famah Sourassou Revision of the lifestyles of phytoseiid mites (Acari: Phytoseiidae) and implications for biological control strategies. Syst. Appl. Acarol. 18: Putman, W. L Life-history and behaviour of the predacious mite Typhlodromus (T.) caudiglans Schuster (Acarina: Phytoseiidae) in Ontario, with notes on the prey of related species. Can. Entomol. 94: Sabelis, M. W., and H. E. Van de Baan Location of distant spider mite colonies by phytoseiid predators: demonstration of specific kairomones emitted by Tetranychus urticae and Panonychus ulmi. Entomol. Exp. Appl. 33: Sabelis, M. W., and F. M. Bakker How predatory mites cope with the web of their tetranychid prey: a functional view on dorsal chaetotaxy in the Phytoseiidae. Exp. Appl. Acarol. 16:

158 Schausberger, P., and B. A. Croft Activity, feeding, and development among larvae of specialist and generalist phytoseiid mite species (Acari: Phytoseiidae). Environ. Entomol. 28:

159 Table 5.1. Pesticides used in nontarget effects bioassays. Common name Mode of action a Chemical class Brand name/formulation b Formulation Use rate (g ai/ha) c mg AI/liter (bioassay rate) carbaryl 1A carbamate Sevin 4F d 479 g/liter azinphosmethyl 1B organophosphate Guthion 50W d 500 g/kg imidacloprid 4A neonicotinyl Provado 1.6F d 192 g/liter spinetoram 5 spinosyn Delegate 25WG e 250 g/kg novaluron 15 IGR - benzoyl urea Rimon 0.83EC f 99 g/liter spirotetramat 23 tetramic acid Ultor 1.25L d 150 g/liter chlorantraniliprole 28 anthranilic diamide Altacor 35WDG g 350 g/kg bifenazate UN - Acramite 50WS f 500 g/kg a Mode of action classification taken from Insecticide Resistance Action Committee (IRAC) v 7.3 ( or the fungicide Resistance Action Committee (FRAC). b The Registrant listed is from the time the experiments were begun. c The concentrations were based on an application rate of 935 liters/ha, or 100 US gallons/acre. d Bayer CropScience, Research Triangle Park, NC. e Dow Agrosciences LLC, Indianapolis, IN. f Chemtura Corporation, Middlebury, CT. g E.I DuPont de Nemours & Co., Wilmington, DE

160 Table 5.2. Mean (± SE) life stage duration of A. caudiglans and G. occidentalis (d). Duration in days Species n Egg n Larva n Protonymph n Deutonymph n Total A. caudiglans ± ± ± ± ± 0.14 G. occidentalis ± ± ± ± ± 0.12 df1, df2 1, 218 1, 184 1, 149 1, 124 1, 124 F P < < < < Table 5.3. Stage-specific survival and sex ratio of A. caudiglans and G. occidentalis. 146 % Survival % Species n Egg n Larva n Protonymph n Deutonymph n Female A. caudiglans G. occidentalis χ 2 (df=1) P

161 Table 5.4. Prey consumption and oviposition by A. caudiglans and G. occidentalis fed on two different diets. Numbers in parentheses are the asymmetric 95% confidence limits. T. urticae eggs T. urticae protonymphs n Prey consumed/48 h Eggs laid/96 h n Prey consumed/48 h Eggs laid/96 h A. caudiglans (21.72, 36.54) 0.37 (0.18, 0.74) ( 8.35, 11.51) 1.67 (1.22, 2.27) G. occidentalis (6.54, 12.36) 1.57 (1.05, 2.35) (8.73, 11.99) 1.33 (0.94, 1.87) df1, df2 1, 17 1, 24 1, 29 1, 29 F P <

162 Table 5.5. Toxicity of orchard pesticides to A. caudiglans and G. occidentalis. A. caudiglans G. occidentalis Treatment n % Mortality a % Runoff n % Mortality a % Runoff Carbaryl a 12.00bc b 13.04b Azinphosmethyl b 45.83a b 16.67b Imidacloprid b 20.00b a 65.22a Spinetoram ab 50.00a a 50.00a Novaluron cd 20.83b c 0.00c Spirotetramat cd 20.00b bc 9.52bc Chlorantraniliprole e 0.00c bc 8.00bc Bifenazate c 43.48a bc 8.33bc Check de 8.00bc bc 8.33bc χ 2 (df=8) P < < a White cells indicate <25%, light grey 25-75%, dark grey >75%. For pesticide concentrations, see Table

163 Table 5.6. Repellency of orchard pesticides to A. caudiglans. Pesticide Evaluation n % on treated half z P Azinphosmethyl * * * Spinetoram <0.0001* <0.0001* * Bifenazate * <0.0001* <0.0001* *Significant at P<

164 CHAPTER SIX: PHENOLOGY AND STRUCTURE OF A PHYTOSEIID COMMUNITY IN AN INSECTIDE-FREE APPLE ORCHARD ABSTRACT Commercial orchards have acarine communities that are reduced in biological diversity compared to their undisturbed counterparts. Examining the phenology of an unsprayed orchard allows for the examination of non-pesticide factors that drive changes in populations. This study examined the mite community in a largely unsprayed research orchard in The phytoseiids Galendromus flumenis (Chant), Amblydromella caudiglans (Schuster), Kampimodromus corylosus Kolodochka, and G. occidentalis (Nesbitt) were found, in addition to Zetzellia mali (Ewing) and Aculus schlechtendali (Nalepa). Although G. occidentalis is typically the dominant phytoseiid in commercial orchards, G. flumenis was much more abundant. A. schlechtendali appeared to be the main source of prey for all predator species. The availability of this prey item and the lack of pesticides are likely the factors that allowed for G. flumenis to reach high abundances. Tetranychids were scarce, emphasizing the role of these mites as induced pests; without the application of disruptive sprays, the predatory mite community was able to maintain biological control. This study demonstrates that the species complex of generalist phytoseiids that is present in orchard systems undisturbed by pesticides is sufficient to maintain spider mite populations below damaging levels throughout the season. 150

165 INTRODUCTION Phytoseiids (Mesostigmata: Phytoseiidae) are important biological control agents of spider mites in a variety of cropping systems, including apple. Although past research has examined the phenology of mite communities in apple orchards, most of this work is several decades old (Hoyt 1967, Oatman 1973, Readshaw 1975, Oatman 1976, Woolhouse and Harmsen 1984). More recent studies are needed, especially those that examine the whole community structure in unsprayed apple orchards. In examining the entire mite community structure, potential drivers of population change can be more easily identified. Additionally, by studying unsprayed orchards, we can more easily identify population dynamics that are strictly due to seasonal changes as opposed to agricultural inputs (which vary highly though time and between orchards). As pesticide use and other integrated pest management practices become more selective, the typical orchard may have more in common with an unsprayed system than conventional orchards of the past. In addition to changing management practices, anthropogenic climate change also has the potential to affect the phenology of mites. A warming climate has resulted in longer growing seasons, species distribution changes, and asynchrony in predator-prey and plant-insect interactions (Parmesan 2006, Jeffs and Lewis 2013). Therefore, even in cases where the phenology of an organism or community is historically well-established, newer studies may find that significant change has occurred. We examined the phenology of the mite community in a research apple block to compare and contrast this community to that commonly found commercial orchards (Ch. 4). In undisturbed phytoseiid communities, generalist phytoseiids can become highly abundant (Prischmann and James 2003, Croft and Luh 2004). These species, referred to as Type III and IV 151

166 phytoseiids, can either subsist on eriophyids, tetranychids (but usually prefer to feed outside of the genus Tetranychus), other mite species, or even pollen (Type III), or are exclusive pollen feeders (Type IV) (McMurtry and Croft 1997). In commercial orchards, Type II predators frequently dominate, especially those that have developed resistance to pesticides (Strickler et al. 1987, Croft and Luh 2004). More notably, tetranychids are usually only present in low numbers in settings undisturbed by agriculture and virtually never cause damage in these situations (Croft and Luh 2004). Our study was conducted in a research apple block that has been rarely sprayed with insecticides or fungicides to determine the phenology of a mite community in a relatively undisturbed agricultural environment. Monitoring this orchard at an unusually high intensity allowed for the identification of both gradual and sudden changes in species abundance and composition throughout the growing season. MATERIALS AND METHODS All samples were collected from an unsprayed 42 year-old research block of Delicious apples in Wenatchee, WA. This block has not received routine sprays of insecticides or fungicides; single trees received sprays during the life of the orchard, but none were applied during two years of the study. Herbicides (glyphosate, norflurazon, simazine) were applied for weed control. Mating disruption and nutrient sprays were not used, but a regular pruning schedule was followed. In 2013, samples of 100 leaves were collected at approximately seven-day intervals. All adult phytoseiid mites were removed from the leaves, slide-mounted (in modified Berlese medium in 2013, in polyvinyl alcohol in 2014), and identified to species as per Denmark and 152

167 Evans (2011). Sampling intensity and frequency was increased in 2014; Samples of leaves were taken and sampling occurred two to three times in a seven-day period. Additionally, in 2014, immature stages of phytoseiids and Zetzellia mali Ewing, were counted without slide mounting; Z. mali can be identified without slide mounting, and juvenile phytoseiids cannot reliably be determined to species using morphological characteristics. A standard brushed-leaf sample was done after all other counts were completed. This allowed Aculus schlechtendali (Nalepa) to be counted; the small size of this species makes it difficult to count while it is on the surface of a leaf. In both years, collection began in April and ended in November, after the majority of leaves had become senescent. Trend lines for each species of mite for each year were generated by using a cubic spline smoother, λ=0.05 (SAS Institute Inc. 2013). To determine the sex ratio of each species, the ten dates with the largest population of each species were identified. The percentage of individuals of each species that were female for the sum of these dates was calculated. RESULTS AND DISCUSSION Phytoseiids found in the research orchard, in order of abundance, were Galendromus flumenis (Chant), Amblydromella caudiglans (Schuster), Kampimodromus corylosus Kolodochka, and G. occidentalis (Nesbitt). A stigmaeid predator, Zetzella mali (Ewing) was also present in the orchard and usually in higher abundances than the phytoseiids. Tetranychids of any species were rarely found in the samples (<0.06/leaf). However, the phytophagous mite Aculus schlechtendali (Nalepa) was present in substantial numbers until late August. Galendromus flumenis was the dominant species in this orchard throughout most of both seasons. Populations in 2013 were somewhat higher than those in 2014 (Fig. 6.1) had two 153

168 distinct population peaks: mid-july and mid-september. There was a similar mid-september peak in 2013, but the earlier peak in mid-july was virtually unnoticeable; the population dip in mid-july was not as apparent. The final (and highest) peak in mid-september was followed by a rapid population decline. It is important to note that, because all phytoseiids have overlapping generations, these peaks do not correspond with generations; these are points in the phenology of the predator when its population increased. G. flumenis is a moderate generalist (Type III), being less specialized that G. occidentalis but more specialized than A. caudiglans (Blackwood et al. 2004, Croft and Luh 2004). It prefers to feed on eriophyids over tetranychids (and has higher reproduction on this prey), and cannot sustain significant reproduction on most pollens (unlike A. caudiglans) (Croft and Jorgensen 1969,1977, Blackwood et al. 2004). A previous survey of feral apple trees found that it is more abundant at low elevations in river valleys and in older trees (Croft and Luh 2004), similar to the conditions present at the research orchard (elevation: 241 m). However, the Croft and Luh (2004) survey also indicated that this species was highly competitive and found with other phytoseiid species less frequently than expected. In this study, G. flumenis populations were around 2-4 higher than that of the other species, but they did not completely exclude other phytoseiids from the orchard. This species is also known to be highly susceptible to pesticides and is typically only found in unsprayed conditions (Duke et al. 1970, Croft and Jorgensen 1977, Prischmann et al. 2005). Therefore, it is likely the combination of high populations of eriophyids and lack of pesticide disruption that allowed this predator to become abundant in the research orchard. In addition to G. flumenis, A. caudiglans was found in this orchard. Its populations had a slight tendency to be higher in 2013 than 2014, but only in the middle of the season (July- 154

169 October) (Fig. 6.2). There were fewer extremely high counts in Both years have the same maxima in the trend line around mid-july. Like G. flumenis, A. caudiglans prefers to feed on eriophyids over tetranychids, especially lacking preference for Tetranychus spp. (Putman 1962, Clements and Harmsen 1993, Blackwood et al. 2004). However, this predator is known to be more generalized than G. flumenis (classified as Type III or even Type IV) (Blackwood et al. 2004, McMurtry et al. 2013), because it can maintain reproduction on pollen-only diets (Putman 1962). While it has also been characterized as predominately existing in unsprayed habitats (Downing and Moilliet 1972, Prischmann et al. 2005), it is also known to exist in abundance in some commercial orchards (Ch. 4). Although the Croft and Luh (2004) survey found that populations of this species tend to be found at higher elevations, it was found in relative abundance in the research orchard (at low elevation). In addition, no correlation was found between elevation and A. caudiglans abundance in a survey of commercial apple in Washington (Ch. 4). Although its populations were not as high as those of G. flumenis, the absence of insecticides and high abundance of A. schlechtendali also favored A. caudiglans in the research orchard. Kampimodromus corylosus is unusual in that its populations were low when the populations of the other phytoseiid species were high (Fig. 6.3). In 2013, this species was virtually undetected until mid-september. In 2014, it was found in low numbers throughout the season, until September when its population began to increase. In both years, the abundance of this species continued to increase through the last sample date. Kampimodromus corylosus is unlike the other phytoseiids found in our study in that it was not found in the Croft and Luh (2004) survey. 155

170 References to this species in the literature are scarce, as it has been recently described (Kolodochka 2003) and there is still some debate as to whether it is a distinct species from K. aberrans (Oudemans) (Tixier et al. 2008). For the purposes of this discussion, we will refer to literature on K. aberrans s.l. for information regarding the biology of K. corylosus. Currently, only individuals historically identified as K. aberrans s.l. from France and the United States are believed to be K. corylosus (Tixier et al. 2008), unfortunately, the majority of literature examines K. aberrans outside of these countries. However, these species are closely related and can therefore be reasonably assumed to have similar biology. This species was virtually absent in apple in the Willamette Valley of Oregon, but was common in filbert (Hadam et al. 1986). It has also been reported in western Washington, where it was rare and found only on Rubus spp. and Tsuga heterophylla (Raf.) Sarg. (Congdon 2002). An extreme preference for Corylus spp. was also reported for Italian populations of K. aberrans s.l. (Duso et al. 2004).We are unaware of any record of this species in substantial abundance on apple in the United States, although it has previously been found in low numbers on recently planted apple in the Wenatchee area(rathman and Brunner 1990). K. aberrans s.l. in Europe has recently been well-studied, primarily due to the discovery of pesticide resistant strains that have been successful in biological control (Duso 1989, Duso et al. 2009). However, this strain is from Italy (and not France), so it is likely to be K. aberrans s.s. and not K. corylosus. Other populations of K. aberrans s.l. are known to be susceptible to pesticides (Hluchy et al. 1991); K. corylosus is likely to be susceptible. K. aberrans s.l. is a Type III generalist phytoseiid (McMurtry and Croft 1997). Although it can survive and reproduce on spider mites, it performs better on a pollen diet (Kasap 2005). It also appears to be less heat tolerant than other phytoseiids; its maximum intrinsic rate of 156

171 population increase occurs at 25 C (Broufas et al. 2007), whereas for G. occidentalis it occurs at 33 C (Tanigoshi et al. 1975). It is also sensitive to low humidities (Schausberger 1998). This potentially explains its low abundance during summer and its gradual increase starting in September. G. occidentalis, although present, remained at low densities throughout both years of the study (Fig. 6.4). This is in contrast to sprayed orchards, where G. occidentalis has been the dominant species in eastern Washington apple orchards for decades (Hoyt 1969, Hoyt 1991, Ch. 4). Indeed, nearness to agriculture and the use of certain pesticides are known to promote populations of this G. occidentalis, likely by selecting against competing, susceptible species (Croft and Luh 2004, Ch. 4). Additionally, as a Type II species, it has a preference for and has higher rates of population growth on Tetranychus spp., although it can feed on other tetranychids and eriophyids (Hanna and Wilson 1991, McMurtry and Croft 1997, Blackwood et al. 2004, Rioja and Vargas 2009). It is also unable to use pollen as a secondary food source (Blackwood et al. 2004). Therefore, the low population of G. occidentalis in this block can be attributed to lack of pesticide applications and appropriate prey items (tetranychids). Zetzellia mali was typically found in higher abundances than phytoseiids (Fig. 6.5). Populations began increasing gradually in July, then rapidly early in August. The population peak occurred mid-september, around the same time as the G. flumenis peak (Fig. 6.1), but after rust mite populations began to decline (Fig. 6.6). Phytoseiid eggs can make up a large portion of Z. mali prey when apple rust mites are scarce (Clements and Harmsen 1990). As in this study, the populations of Z. mali have been noted to relate to those of A. schlechtendali, but to not decline as rust mite populations drop (Hoyt 1967). This is potentially why Z. mali were consistently more abundant in the research orchard than phytoseiids; Z. mali was able to 157

172 consume both rust mites and phytoseiids. Conversely, phytoseiids are not known to use Z. mali as prey, giving Z. mali a competitive advantage. Densities of Z. mali and phytoseiids have been found to be inversely related (Croft and MacRae 1993, Croft et al. 2004); This makes the role of Z. mali as biological control difficult to discern, as it feeds on phytophagous mites but also has the potential to disrupt biological control by phytoseiids through intraguild predation (Croft and MacRae 1992). While G. occidentalis is common in commercial orchards, and the generalist phytoseiids found in this study are known to be susceptible to pesticides, Z. mali is known to have the potential to survive in both circumstances (Strickler et al. 1987, Thistlewood 1991). It has been found to be less sensitive to field applications of neonicotynl and pyrethroid pesticides than G. occidentalis (Villanueva and Harmsen 1998, Beers et al. 2005). Z. mali can also has a high rate of reproduction on Panonychus ulmi (Koch), a pest that reach outbreak populations in commercial orchards (Santos 1976, Clements and Harmsen 1993). Resistance to pesticide applications and the ability to reproduce on a common pest gives this predator the ability to persist in commercial orchards, but it can also survive in unsprayed conditions where it can feed on eriophyids and phytoseiids. Spider mite populations remained very low throughout the season (<0.06 per leaf). Bryobia rubrioculus (Scheuten) and Tetranychus urticae Koch were found, but not P. ulmi. The only prey species available in substantial numbers was A. schlechtendali. This species was first found in June and increased in abundance rapidly, peaking in mid-july (Fig. 6.6). By mid- September, this species was no longer found in samples. Availability of rust mites was likely the factor that drove phytoseiid and Z. mali population increases. Indeed, the only predator that does not prefer eriophyids as prey (G. occidentalis) was also the most scarce. While rust mites are 158

173 depended upon as an alternative prey for G. occidentalis in times of tetranychid scarcity in commercial orchards (Hoyt 1969), they also played the key role in the acarine community of this unsprayed orchard. In addition to A. schlechtendali, the predatory mites may have also consumed pollen or other species of predators in order to sustain populations. Amblydromella caudiglans and K. corylosus are both known to feed on pollen, but Galendromus flumenis and G. occidentalis are not able to reproduce when feeding on pollen exclusively (Putman 1962, McMurtry and Croft 1997, Blackwood et al. 2004). However, both G. flumenis and A. caudiglans have the highest ovipositional rates on eriophyids (Blackwood et al. 2004). This provides further evidence that the high abundance of A. schlechtendali was critical to the maintenance of the rest of the mite species in our study. As with most phytoseiid species, the population of G. flumenis, A. caudiglans, and K. corylosus were female-biased (Dyer and Swift 1979, Hoy 1985). There was some variability in the proportion of females between 2013 and 2014, especially for A. caudiglans (Table 6.1). Natural variation in sex ratio in the field has been previously reported for many phytoseiid species (Dyer and Swift 1979). Because populations of G. occidentalis were so low, sex ratio calculations were not performed for this species. While juvenile phytoseiids cannot be identified to species, examination of the proportion of the community at a given life stage gives a sense of when a population begins reproducing in spring and starts diapause in fall. The phytoseiid community in this research orchard produced deutonymphs by May, indicating that one generation was nearly complete at this point (Fig. 6.7). The proportion of total individuals that were juveniles significantly decreased by October. This is roughly the initiation of diapause of the three most numerous species, which all overwinter as 159

174 diapausing females (McMurtry et al. 1970). This indicates the potential for predatory mites to suppress pest mite species throughout the entire growing season, as spider mites also begin diapause at this point (Hoyt 1967, Beers and Hoyt 1993b,a). However, biological control requires that phytoseiid populations are not disrupted by pesticides with nontarget effects. In commercial orchards, certain pesticides can reduce populations of G. occidentalis and cause pest mite flare ups (Huffaker et al. 1970, McMurtry et al. 1970). However, in the complete absence of pesticides (the study orchard), a different community of phytoseiids can still provide biological control. Indeed, the low abundance of spider mites in this orchard emphasizes the role of spider mites as induced pests. Without the use of pesticides for key pests (eg. Cydia pomonella (L.)), spider mite populations do not reach damaging levels (Croft and Luh 2004). This study provides insight into the seasonal changes in a phytoseiid community free from pesticide pressure. Phytoseiid populations in this orchard were likely driven by the availability of both prey and alternative food, and competition with each other and Z. mali. Consistent with previous studies in unsprayed (or minimally sprayed) apple, generalist phytoseiids appear to dominate and are successful at suppressing pest mite species below damaging levels (McMurtry 1992, Croft and Luh 2004, Funayama et al. 2015). In an ideal orchard system, selective pesticide use for control of key pests will allow predatory mite communities to maintain biological control of pest mite populations. With the adoption of more selective pest management practices, future commercial orchards may have acarine communities similar to those of an unsprayed orchard and the reduced pest mite pressure associated with this habitat. 160

175 REFERENCES CITED Beers, E. H., and S. C. Hoyt. 1993a. Twospotted spider mite In E. H. Beers, J. F. Brunner, M. J. Willett and G. M. Warner (eds.), Orchard pest management: a resource book for the Pacific Northwest. Good Fruit Grower, Yakima, WA. Beers, E. H., and S. C. Hoyt. 1993b. European red mite In E. H. Beers, J. F. Brunner, M. J. Willett and G. M. Warner (eds.), Orchard pest management: a resource book for the Pacific Northwest. Good Fruit Grower, Yakima, WA. Beers, E. H., J. F. Brunner, J. E. Dunley, M. Doerr, and K. Granger Role of neonicotinyl insecticides in Washington apple integrated pest management. Part II. Nontarget effects on integrated mite control. J. Insect Sci. 5 (16): available online. Blackwood, J. S., H.-K. Luh, and B. A. Croft Evaluation of prey-stage preference as an indicator of life-style type in phytoseiid mites. Exp. Appl. Acarol. 33: Broufas, G. D., M. L. Pappas, and D. S. Koveos Development, survival, and reproduction of the predatory mite Kampimodromus aberrans (Acari: Phytoseiidae) at different constant temperatures. Environ. Entomol. 36: Clements, D. R., and R. Harmsen Predatory behavior and prey-stage preferences of stigmaeid and phytoseiid mites and their potential compatibility in biological control. Can. Entomol. 122: Clements, D. R., and R. Harmsen Prey preferences of adult and immature Zetzellia mali Ewing (Acari: Stigmaeidae) and Typhlodromus caudiglans Schuster (Acari: Phytoseiidae). Can. Entomol. 125: Congdon, B. D The family Phytoseiidae (Acari) in western Washington State with descriptions of three new species. Int. J. Acarol. 28:

176 Croft, B. A., and C. D. Jorgensen Life history of Typhlodromus mcgregori (Acarina: Phytoseiidae). Ann. Entomol. Soc. Am. 63: Croft, B. A., and C. D. Jorgensen Typhlodromus mcgegori (Acarina: Phytoseiidae) and its potential control of phytophagous mites in southern California and central Utah apple orchards. Great Basin Nat. 37: Croft, B. A., and I. V. MacRae Persistence of Typhlodromus pyri and Metaseiulus occidentalis (Acari: Phytoseiidae) on apple after inoculative release and competition with Zetzellia mali (Acari: Stigmaeidae). Environ. Entomol. 21: Croft, B. A., and I. V. MacRae Biological control of apple mites: impact of Zetzellia mali (Acari: Stigmaeidae) on Typhlodromus pyri and Metaseiulus occidentalis (Acari: Phytoseiidae). Environ. Entomol. 22: Croft, B. A., and H.-K. Luh Phytoseiid mites on unsprayed apple trees in Oregon, and other western states (USA): distributions, life-style types and relevance to commercial orchards. Exp. Appl. Acarol. 33: Croft, B. A., J. S. Blackwood, and J. A. McMurtry Classifying life-style types of phytoseiid mites: diagnostic traits. Exp. Appl. Acarol. 33: Downing, R. S., and T. K. Moilliet Replacement of Typhlodromus occidentalis by T. caudiglans and T. pyri (Acarina: Phytoseiidae) after cessation of sprays on apple trees. Can. Entomol. 104: Duke, K. M., B. A. Croft, and C. D. Jorgensen Distribution and searching behavior of Typhlodromus mcgregori. J. Econ. Entomol. 63: Duso, C Role of the predator mites Amblyseius aberrans (Oud.), Typhlodromus pyri Scheuten and Amblyseius andersoni (Chant) (Acari, Phytoseiidae) in vineyards. I. The 162

177 effects of single or mixed phytoseiid population releases on spider mite densities (Acari, Tetranychidae). J. Appl. Entomol. 107: Duso, C., M. Fanti, A. Pozzebon, and G. Angeli Is the predatory mite Kampimodromus aberrans a candidate for the control of phytophagous mites in European apple orchards? BioControl 54: Duso, C., V. Malagnini, A. Paganelli, L. Aldegheri, M. Bottini, and S. Otto Pollen availability and abundance of predatory phytoseiid mites on natural and secondary hedgerows. BioControl 49: Dyer, J. G., and F. C. Swift Sex ratio in field populations of phytoseiid mites (Acarina: Phytoseiidae). Ann. Entomol. Soc. Am. 72: Funayama, K., M. Komatus, S. Sonoda, I. Takahashi, and K. Hara Management of apple orchards to conserve generalist phytoseiid mites suppresses two-spotted spider mite, Tetranychus urticae (Acari: Tetranychidae). Exp. Appl. Acarol. 65: Hadam, J. J., M. T. Aliniazee, and B. A. Croft Phytoseiid mites (Parasitiformes: Phytoseiidae) of major crops in Willamette Valley, Oregon, and pesticide resistance in Typhlodromus pyri Scheuten. Environ. Entomol. 15: Hanna, R., and T. Wilson Prey preference by Metaseiulus occidentalis (Acari: Phytoseiidae) and the role of prey aggregation. Biol. Control 1: Hluchy, M., Z. Pospisil, and M. Zacharda Phytophagous and predatory mites (Acari: Tetranychidae, Eriophyidae, Phytoseiidae, Stigmaeidae) in South Moravian vineyards, Czechoslovakia, treated with various types of chemicals. Exp. Appl. Acarol. 13: Hoy, M. A Recent advances in genetics and genetic improvement of the Phytoseiidae. Annu. Rev. Entomol. 30:

178 Hoyt, S. C Population studies of five mite species on apple in Washington, pp In G. O. Evans (ed.) Proceedings, 2nd International Congress of Acarology, July 1967, Sutton Bonington, UK. Akadémiai Kiadó, Budapest, Hungary. Hoyt, S. C Integrated chemical control of insects and biological control of mites on apple in Washington. J. Econ. Entomol. 62: Hoyt, S. C Biology, ecology and control of mites in Washington orchards. In: Williams K. (ed) New directions in tree fruit pest management. Good Fruit Grower, Yakima, WA, pp Huffaker, C. B., M. Van de Vrie, and J. A. McMurtry II. Tetranychid populations and their possible control by predators: an evaluation. Hilgardia 40: Jeffs, C. T., and O. T. Lewis Effects of climate warming on host-parasitoid interactions. Ecol. Entomol. 38: Kasap, I Life-history traits of the predaceous mite Kampimodromus aberrans (Oudemans) (Acarina: Phytoseiidae) on four different types of food. Biol. Control 35: Kolodochka, K. A A new species of the genus Kampimodromus (Parasitiformes, Phytoseiidae) from Ukraine and Moldova. Acarina 11: McMurtry, J. A Dynamics and potential impact of 'generalist' phytoseiids in agroecosystems and possibilities for establishment of exotic species. Exp. Appl. Acarol. 14: McMurtry, J. A., and B. A. Croft Life-styles of phytoseiid mites and their roles in biological control. Annu. Rev. Entomol. 42:

179 McMurtry, J. A., C. B. Huffaker, and M. van de Vrie I. Tetranychid enemies: their biological characters and the impact of spray practices. Hilgardia 40: McMurtry, J. A., G. J. de Moraes, and N. Famah Sourassou Revision of the lifestyles of phytoseiid mites (Acari: Phytoseiidae) and implications for biological control strategies. Syst. Appl. Acarol. 18: Oatman, E. R An ecological study of arthropod populations on apple in northeastern Wisconsin: population dynamics of mite species on the foliage. Ann. Entomol. Soc. Am. 66: Oatman, E. R An ecological study of arthropod populations on apple in northeastern Wisconsin: Phytoseiid mite species on the foliage. Environ. Entomol. 5: Parmesan, C Ecological and evolutionary responses to recent climate change. Annu. Rev. Ecol. Evol. Syst. 37: Prischmann, D. A., and D. G. James Phytoseiidae (Acari) on unsprayed vegetation in southcentral Washington: implications for biological control of spider mites on wine grapes. Int. J. Acarol. 29: Prischmann, D. A., D. G. James, L. C. Wright, R. D. Teneyck, and W. E. Snyder Effects of chlorpyrifos and sulfur on spider mites (Acari: Tetranychidae) and their natural enemies. Biol. Control 33: Putman, W. L Life-history and behaviour of the predacious mite Typhlodromus (T.) caudiglans Schuster (Acarina: Phytoseiidae) in Ontario, with notes on the prey of related species. Can. Entomol. 94:

180 Rathman, R. J., and J. F. Brunner Abundance and composition of predators on young apple, Malus domestica Borkhausen, within sagebrush and riparian species pools in north central Washington. Melandria 46: Readshaw, J. L The ecology of tetranychid mites in Australian orchards. J. Appl. Ecol. 12: Rioja, T., and R. Vargas Life table parameters and consumption rate of Cydnodromus picanus Ragusa, Amblyseius graminis Chant, and Galendromus occidentalis (Nesbitt) on avocado red mite Oligonychus yothersi (McGregor) (Acari: Phytoseiiday, Tetranychidae). Chil. J. Agric. Res. 69: Santos, M. A Evaluation of Zetzellia mali as a predator of Panonychus ulmi and Aculus schlectendali. Environ. Entomol. 5: SAS Institute Inc Using JMP 11, Cary, NC: SAS Institute Inc. Schausberger, P The influence of relative humidity on egg hatch in Euseius finlandicus, Typhlodromus pyri and Kampimodromus aberrans (Acari, Phytoseiidae). J. Appl. Entomol. 122: Strickler, K., N. Cushing, M. Whalon, and B. A. Croft Mite (Acari) species composition in Michigan apple orchards. Environ. Entomol. 16: Tanigoshi, L. K., S. C. Hoyt, R. W. Browne, and J. A. Logan Influence of temperature on population increase of Metaseiulus occidentalis (Acarina: Phytoseiidae). Ann. Entomol. Soc. Am. 68: Thistlewood, H A survey of predatory mites in Ontario apple orchards with diverse pesticide programmes. Can. Entomol. 123:

181 Tixier, M.-S., S. Kreiter, B. A. Croft, and B. Cheval Kampimodromus aberrans (Acari: Phytoseiidae) from the USA: morphological and molecular assessment of its density. Bull. Entomol. Res. 98: Villanueva, R. T., and R. Harmsen Studies on the role of the stigmaeid predator Zetzellia mali in the acarine system of apple foliage, pp In R. Harmsen (ed.) Proceedings, Entomological Society of Ontario, Sudbury, Ontario. Entomological Society of Ontario. Woolhouse, M. E. J., and R. Harmsen The mite complex on the foliage of a pesticidefree apple orchard: population dynamics and habitat associations, pp In C. R. Ellis, P. G. Kevan, D. J. Pree and R. M. Trimble (eds.), Proceedings, Entomological Society of Ontario. Entomological Society of Ontario. 167

182 Table 6.1. Percentage of phytoseiid species that were female on the 10 dates with the highest population of each species Species n Percent female n Percent female A. caudiglans G. flumenis K. corylosus

183 Fig G. flumenis abundance in an insecticide-free orchard over time for two years 169

184 Fig A. caudiglans abundance in an insecticide-free orchard over time for two years 170

185 Fig K. corylosus abundance in an insecticide-free orchard over time for two years 171

186 Fig G. occidentalis abundance in an insectide-free orchard over time for two years 172

187 Fig Z. mali abundance in an insecticide-free orchard over time for

188 Fig A. schlechtendali abundance in an insecticide-free orchard over time for

189 CHAPTER SEVEN: EFFECTS OF APPLE TRICHOMES ON PHYTOSEIIDS: PREFERENCES, FECUNDITY, PREY CONSUMPTION, AND POPULATION DENSITY ABSTRACT Studies regarding the effects of plant structures, like trichomes, on phytoseiids have become common in recent decades. However, there is a paucity of literature on this topic for Galendromus occidentalis (Nesbitt), which is considered the most important tree fruit mite biological control agent in the western United States. We conducted a series of laboratory experiments to determine the preferences of G. occidentalis for a pubescent ( Delicious ) apple cultivar compared to a more glabrous cultivar ( Golden Delicious ); we also studied oviposition and prey consumption on these varieties when no choice was given. Finally, observations of phytoseiid populations in a mixed unsprayed block the two varieties were performed to determine if preferences observed in the laboratory translated to the field. While its prey, Tetranychus urticae Koch demonstrated some preference for the more glabrous Golden Delicious leaves, G. occidentalis had a marginal preference for ovipositing on the more pubescent Delicious leaves. However, when no choice in cultivar was given, G. occidentalis did not oviposit or consume more prey on either cultivar. Although the laboratory results indicated a slight preference for Delicious leaves, G. occidentalis populations were numerically higher on Golden Delicious trees in the field; however, the same was true for other phytoseiid species. The conflicting results of the laboratory preference tests vs. field observations highlight the importance phytoseiid species identity in examining tritrophic interactions with host plants. 175

190 INTRODUCTION Many phytophagous arthropods are host-specific or have preferences for certain crops. These preferences can be exploited by crop breeders, allowing for the production of varieties that are resistant to pest damage. However, only recently has the relationship between predators and prey host plants been thoroughly explored. Because these interactions do not involve the use of the plant as a food source, there is a strong potential for mutualistic relationships. The host plant may provide the predator with a resource, enabling the predator to better control phytophagous arthropod populations (Agrawal 2000). Conversely, there is also the potential for plant defenses to unintentionally inhibit the biological control abilities of natural enemies. One such plant defense, trichomes, are hair-like outgrowths of the plant epidermis (Johnson 1975). These structures can reduce plant desiccation, decrease ultraviolet light exposure, and provide defenses against herbivorous arthropods (Levin 1973, Johnson 1975). Glandular trichomes produce sticky or toxic exudates that can entangle or kill plant enemies and hooked trichomes can impale potential phytophages (Levin 1973, Southwood 1986). Trichomes also can serve as a barrier, preventing herbivores from reaching the leaf tissues (Levin 1973, Johnson 1975, Southwood 1986). Because of their small size, mites can be greatly affected by leaf surface features, including trichomes. Mites in the family Phytoseiidae, which are important biological control agents of phytophagous mites in a variety of cropping systems, can be strongly influenced by the level of leaf pubescence (Schmidt 2014). Trichomes may increase leaf surface humidity (Grostal and O'Dowd 1994), trap pollen (Kreiter et al. 2003, Roda et al. 2003), or provide shelter from predators (Roda et al. 2000, Seelmann et al. 2007), factors which may provide a beneficial environment for phytoseiids. Trapped pollen may provide an alternative food source for 176

191 generalist phytoseiids in times of prey scarcity (Croft et al. 2004). However, like other natural enemies (Riddick and Simmons 2014), phytoseiid prey search efficacy may suffer if trichomes are very dense (Camporese and Duso 1996, Krips et al. 1999, Cedola et al. 2001). Additionally, glandular or hooked trichomes may also increase phytoseiid mortality (Cedola et al. 2001). In Washington apple orchards, Galendromus occidentalis (Nesbitt) is usually the dominant phytoseiid species (Ch. 4). It is an important biological control agent of the most common spider mite pests, Panonychus ulmi (Koch), Tetranychus urticae Koch, and Tetranychus mcdanieli McGregor (Hoyt 1969, Hoy 2011). Apple trichomes are neither glandular nor hooked, therefore, these structures have the potential to benefit G. occidentalis. However, unlike more generalized phytoseiids, G. occidentalis does not consume pollen (McMurtry 1992). Therefore, the most likely benefits provided by these structures are shelter from either predation or harsh climate conditions. Relatively few predators of phytoseiids exist in commercial apple orchards. Although small life stages of predatory hemipterans will consume phytoseiids (Cloutier and Johnson 1993), these organisms are often scarce in orchards. Another predatory mite, Zetzellia mali (Ewing), is also known as an intraguild predator of phytoseiid eggs (Clements and Harmsen 1990), but it is not found in significant abundance in the majority of commercial apple orchards (Thistlewood 1991). Additionally, the apple growing regions of eastern Washington are arid (Kottek et al. 2006), so it is possible that trichomes may benefit G. occidentalis by providing a more humid microclimate. For the purposes of shelter, the egg stage is most vulnerable to both predation (Clements and Harmsen 1990) and desiccation (Croft et al. 1993). Very few studies have examined the trichome density preferences of G. occidentalis. Downing and Moillet (1967) found that G. occidentalis populations were higher in certain apple varieties and concluded that differences in leaf pubescence could explain these differences. The 177

192 presence of domatia, another type of leaf structure, was found to increase G. occidentalis reproduction at low humidity, indicating that surface structure complexity may benefit this predator in suboptimal climate conditions (Grostal and O'Dowd 1994). The following series of studies were conducted in order to better describe the trichome density preferences of G. occidentalis and elucidate possible causes for these preferences. MATERIALS AND METHODS Choice tests apple leaves. Choice test experiments were performed in 2011, using Delicious leaves as the pubescent cultivar and Golden Delicious leaves as the glabrous cultivar. Ten leaves of each cultivar were picked from unsprayed trees at the Washington State University Research and Extension Center (WSU-TFREC) in Wenatchee, WA. Dual choice arenas were created using a method adapted from Seelman et al. (2007). Plastic cups (96.1 ml) were filled with water-saturated cotton. From each leaf, a mm rectangle was cut, avoiding the midrib. Double-sided tape was used to attach a leaf piece from each cultivar, lower surface facing up, onto a thin piece of plastic film to form a mm square. The edges of each leaf piece were flush with each other and excess plastic film was removed from the edges of the arena. Each arena was replicated ten times. Trichome density of each leaf piece was measured using a method adapted from Roda et al. (2003); a 3 mm reference line was randomly placed in an area between leaf veins. The number of trichomes that crossed the reference line was counted for three different random locations on each leaf piece. Tetranychus urticae from a laboratory colony originally established at Cornell University and maintained on Phaseolus vulgaris L. Henderson Bush were used in these studies. Twenty adult female T. urticae from the laboratory colony were placed at the junction of the two leaf 178

193 pieces. The number of T. urticae females on each side of the arena was recorded at four times at one hour intervals after loading. Twenty-four hours after loading, the T. urticae females and number of eggs laid on each side of the arena was recorded. The adult female T. urticae were removed from the arenas. Egg numbers were adjusted so that there were twenty on each side of the arena. Each T. urticae egg position was marked with a felt-tip pen. A single G. occidentalis female was transferred to the arena at the junction of the two leaf pieces. The G. occidentalis were taken from a laboratory colony originating from a commercial apple orchard near Othello, WA in The predatory mite colony was kept on T. urticae-infested bean leaves. The location of the G. occidentalis female (relative to cultivar), was recorded four times at one-hour intervals after transfer. Forty-eight hours after transferring the G. occidentalis females, the number and location of the G. occidentalis eggs and the number of T. urticae eggs consumed from each side of the arena were recorded. A second trial arena was tested where the leaf midrib was included, given that this is a favored resting place (Hoyt and Beers 1993). The methods were similar to those described above, except that the midrib was included in each leaf piece, centered lengthwise in each rectangle. The trichome density of each leaf piece was measured using the reference line technique described previously. Mite response data (position on arena, oviposition, prey consumption) were analyzed using a generalized linear mixed model (PROC GLIMMIX, SAS 9.3), specifying a binary distribution for position on arena and a binomial distribution for oviposition and prey consumption. The denominator degrees of freedom for F tests were adjusted as per Kenward and Roger (1997), in order to test fixed effects within the mixed model. PROC GLIMMIX was also used to analyze the trichome count data, specifying a log-normal distribution. A small number 179

194 (0.001) was added to each value to prevent non-convergence due to zero values. Significance was declared at P= for the mite position data, using the Bonferroni adjustment for four comparisons to the 5% level of significance. All other analyses used P=0.05 to determine significance. Choice tests addition of cotton. Galendromus occidentalis from a colony established approximately three months prior to testing were used. This colony was collected from a research apple orchard near Orondo, WA. Because the preference for a specific apple cultivar could be due to other varietal differences other than trichome density, a second choice test was performed in Artificial trichomes were created by attaching cotton fibers to trichome-free bean leaves (Roda et al. 2001). Bean leaf disks (2.2 cm diam.) were cut with the midrib of the leaf bisecting the disk. Leaves were placed abaxial side up on water-saturated cotton inside plastic cups (14.7 ml). A very small amount of cotton fiber was removed from a cotton ball with forceps. One end of the fibers was dipped into unsolidified agar and then placed in the center of one side of the leaf disk and allowed to dry. Twenty T. urticae females were added to the arena and allowed to oviposit for 24 h., after which T. urticae were removed from the arena. Egg numbers were adjusted so that there were 20 on each side of the arena. Each T. urticae egg position was marked with a felt-tip pen. A single female G. occidentalis was placed on the midrib of each arena. The arena was replicated 25 times. The side of the arena (with or without cotton fibers) on which the G. occidentalis female was found was recorded at 1-h intervals for 6 h, and also at 24 h intervals for 72 h. The number of T. urticae eggs consumed and the number of eggs laid by the G. occidentalis on each side of the arena was recorded after 72 h. These data were analyzed as described in the apple cultivar 180

195 choice tests. Statistical significance in the side preference test was adjusted for 9 comparisons by using P= No-choice tests. Although the prior tests tested preferences for different leaf substrates, they did not determine if one substrate is more beneficial for G. occidentalis survival than another. To examine the hypotheses that increased leaf pubescence 1) positively affects oviposition or 2) negatively affects prey consumption, no-choice tests were conducted in Disks (2.2 cm diam.) were cut from untreated Delicious and Golden Delicious apple leaves from WSU-TFREC, with the midrib bisecting the disk. Leaf disks were placed on watersaturated cotton in 14.7 ml plastic cups. All arthropods were removed from the disks using a fine brush. Thirty T. urticae females were added to each leaf disk and allowed to oviposit for 24 h. Egg numbers were adjusted to 40 per disk, marked, and the T. urticae females were removed. A single phytoseiid female was added to each leaf disk. These mites were obtained directly from apple leaves collected the previous day from a research orchard in Rock Island, WA. Thirty replicates per treatment of this arena were tested. The number of T. urticae eggs consumed and phytoseiid eggs laid per disk was recorded 48 h after the introduction of the phytoseiid female. The female was slide-mounted for identification, and only those determined to be G. occidentalis were used in the analysis. Data were analyzed using PROC GLIMMIX, specifying the negative binomial distribution for count data. A small number (0.001) was added to each specified response to prevent non-convergence due to zero values. Field observations. Leaves were collected from apple trees in an untreated, experimental orchard block located in Wenatchee, WA on 12 Sep This block consisted of Delicious cultivar apples, with Golden Delicious planted as pollenizer trees. A 100-leaf sample was 181

196 collected from each of 16 Golden Delicious and Delicious tree pairs immediately adjacent to one another (two treatments, 16 replications). The number of trichomes on 10 randomly selected leaves of each 100-leaf sample was counted using the reference line technique previously described. The trichome counts were always taken from an area within 2 cm of the midrib because of the findings from the previous experiments. For each leaf, three reference line counts were recorded. All 100 leaves were then examined under a dissecting microscope for arthropods. All adult phytoseiids found in the samples were removed with a fine brush and slide mounted for identification. Juvenile phytoseiids were identified by life stage (egg, larva, protonymph, deutonymph), but not slide mounted. All tetranychid motile stages and eggs were also counted, as were Aculus schlechtendali (Nalepa) and Zetzellia mali. Finally, pollen counts were taken from five of the ten leaves used for trichome counts. A strip of double-sided tape (1.3 cm wide) was applied to the abaxial side of each leaf, centered over the midrib. The tape was pressed down over the midrib and the leaf lamina. The strip of tape was then removed and placed on a slide, with the side that contacted the leaf facing up. The edges of the tape were cut to fit the length of the slide (7.5 cm). Several drops of glycerin were added to the surface of the tape, causing swelling of pollen grains, making them easier to count (Addison et al. 2000). A second slide was placed on top of the tape with glycerin. The slide covering the tape was marked into three 1.0 cm 2 areas using a felt pen. Pollen grains within each marked square was counted under phase-contrast microscopy at 200 magnification. The counts of the most commonly found mite species were analyzed with PROC GLIMMIX, assuming a negative binomial distribution. Per leaf trichome and pollen counts were 182

197 averaged and then PROC GLIMMIX was used to analyze these data for each variable, specifying the log-normal distribution. RESULTS Choice tests apple leaves. In the first trial (no midribs), the trichome counts of Delicious leaves were not different than those of Golden Delicious (Fig. 7.1). Tetranychus urticae were only found to have a significant preference for positioning themselves on Golden Delicious in final evaluation (Table 7.1). Tetranychus urticae also exhibited a marginal preference for laying eggs on the Golden Delicious side of the arenas. Galendromus occidentalis had no apparent preference for arena side; the proportion of individuals on the Delicious side was highly variable between Evaluation 1 (9%) and the other evaluations (60%) (Table 7.1). There was no G. occidentalis preference for either prey consumption or oviposition (Table 7.2). In arenas where midribs were included, trichome counts were higher for Delicious than Golden Delicious (Fig. 7.2). Similar to the arena trials where the midrib was not included, a marginally significantly higher proportion of T. urticae were found on Golden Delicious at only one evaluation, but were always greater in number on this side than the Delicious side (Table 7.3). While the spider mite preference was non-significant (P=0.07), there was a tendency for ovipositing on the Golden Delicious side of arenas (Table 7.4). Similarly, the predator position preference was non-significant, but showed a trend towards ovipositing on the Delicious side of the arena (Table 7.4). Galendromus occidentalis did not exhibit a preference for consuming prey on a particular cultivar (Table 7.4). 183

198 Choice test addition of cotton. A preference for position on the cotton addition side of the arena was found at two of the six hourly evaluations (Eval. 3 and 6), with one additional observation being marginally significant (Table 7.5). However, no preference was found in any of the three daily evaluations (Table 7.5). More eggs were laid, and fewer prey were consumed on the cotton-added side of arenas, but these differences were not statistically significant (Table 7.6). In the case of the oviposition counts, this is likely due to a lower number of eggs laid overall (several repetitions had no oviposition). No-choice test. Slide mounting and species identification revealed that only half (n=15) of the replicates contained G. occidentalis females; all other species were excluded from the analysis. G. occidentalis consumed similar amounts of prey, and laid similar numbers of eggs on the two cultivars tested (Fig. 7.3). Field observations. In order of abundance, Galendromus flumenis (Chant) (n=556), Amblydromella caudiglans (Schuster) (n=95), Kampimodromus corylosus Kolodochka (n=42), G. occidentalis (n=19), and Euseius finlandicus (Oudemans) (n=3) were found in the block sampled. Only the counts of G. flumenis and A. caudiglans were high enough to warrant statistical analysis. Very few tetranychids of any species were found, however apple rust mite was present in substantial numbers, as was Z. mali. As with the arena studies, trichome counts were higher on Delicious than Golden Delicious leaves (Table 7.7). However, this did not result in higher pollen counts. Both G. flumenis and A. caudiglans counts were higher on Golden Delicious than Delicious. There were no differences in Z. mali or A. schlechtendali populations between the varieties. Although not statistically analyzed, more G. occidentalis (84%) were found on Golden Delicious as well, although K. corylosus was evenly distributed, with 50% of individuals found on each cultivar. 184

199 DISCUSSION Previous studies have indicated that Golden Delicious is a consistently less pubescent cultivar (Roda et al. 2003, Duso et al. 2009), Delicious is known to have trichomes concentrated around the midrib (Roda et al. 2001). This may explain the lack of difference found in trichome counts for the two varieties when the midribs were excluded from arenas. When whole leaves were tested for trichome differences, Delicious leaves had trichome counts over 4 those of Golden Delicious (Table 7), indicating that Delicious was the more pubescent cultivar in our studies. In both choice tests, T. urticae displayed at least a marginally significant preference for Golden Delicious leaves over Delicious leaves. Despite the preference of its prey, the predators preferred to oviposit on the opposite side of the arena, but only when the midrib was included. This could indicate an important benefit of this cultivar to the predators, if they are willing to leave the preferred area of their prey (Faraji et al. 2002). The cotton addition tests indicate that this preference is in fact related to the structural complexity of the leaves of this cultivar, and not some other difference between the two varieties; Galendromus occidentalis consistently preferred to position itself near the added tuft of cotton during the daily evaluations. However, this did not result in statistically higher oviposition on this side of the arena. This could be attributed to low statistical power, because so few eggs were laid during this experiment. Despite the lack of statistical difference, the majority (75%) of G. occidentalis eggs were laid on the side of the arena with the added cotton. When given a choice, G. occidentalis appeared to at least marginally prefer the Delicious side of the arena in some instances. However, with no choice in trichome density, predators laid equal numbers of eggs on both varieties. This indicates that while predators may 185

200 have a preference to oviposit in structurally complex leaves, they are capable of ovipositing in less complex substrates. Unfortunately, an insufficient number of G. occidentalis were found in the field observations for statistical comparisons. However, there were numerically more individuals of this species on the Golden Delicious trees. Additionally, the two species that were sufficiently abundant for statistical analysis (G. flumenis and A. caudiglans) were also more prevalent on Golden Delicious. This runs contrary to previous studies indicating A. caudiglans is more abundant on pubescent apple varieties (Downing and Moilliet 1967), although no similar experiments have been conducted with G. flumenis. However, one survey of unsprayed vegetation in Washington (in which G. flumenis and A. caudiglans were found) did not find a correlation between ventral leaf surface pubescence and the presence of phytoseiid mites (Prischmann and James 2003). Although Delicious leaves do have more trichomes than Golden Delicious, they have equal numbers of domatia (Duso et al. 2009). It is possible that Golden Delicious has more or higher quality domatia and this explains the field preference for this cultivar. At first it appears unusual that there was a clear preference for Golden Delicious in the field, but no preference (or a marginal preference for Delicious ) in our laboratory experiments. However, G. occidentalis is a specialist phytoseiid, whereas the other species found in the field study are generalists (McMurtry and Croft 1997, McMurtry et al. 2013). Research has suggested that generalist phytoseiids are more affected by the environment provided by their host plant; they are more likely to have preferences for glabrous or pubescent leaves (Camporese and Duso 1996, McMurtry and Croft 1997). The newer classification system for phytoseiid life styles even divides Type III generalists into subgroups based on host plant preferences (McMurtry et al. 186

201 2013). Therefore, it is not unexpected that G. occidentalis did not exhibit strong preferences for a variety in our assays. However, the preference of the phytoseiids for the more glabrous Golden Delicious in the field study does not reflect previous findings. Amblydromella caudiglans is known to prefer pubescent host plants, as does K. corylosus (K. aberrans s.l.) (Downing and Moilliet 1967, Seelmann et al. 2007, McMurtry et al. 2013). Similar research has not been conducted with G. flumenis. A possible explanation for these results is the use of McIntosh as the pubescent variety and Delicious as the glabrous variety in the Seelmann et al. (2007) study. It is possible that the difference in trichome density between these two cultivars is greater than the difference between Golden Delicious and Delicious. Also, our field study was conducted during the end of the season, at a single time point, so phytoseiid densities at this point may not reflect how they were distributed during the rest of the year. This study emphasizes the need to further evaluate effects of mite trichome/cultivar preferences in both the laboratory and the field and to not generalize effects across multiple species of phytoseiids, especially before this information is applied as part of a pest management strategy. While many phytoseiid species have demonstrated preferences for physically complex leaf surfaces, these structures may be detrimental to other species or have no effect (Schmidt 2014). Additionally, these effects may vary based on time of year and which cultivars are being compared. Clearly, more research is needed to clarify the role of plant physical structures in tritrophic interactions involving phytoseiids, phytophagous mites, and plants. 187

202 REFERENCES CITED Addison, J. A., J. M. Hardman, and S. J. Walde Pollen availability for predaceous mites on apple: spatial and temporal heterogeneity. Exp. Appl. Acarol. 24: Agrawal, A. A Mechanisms, ecological consequences and agricultural implications of tritrophic interactions. Curr. Opin. Plant Biol. 3: Camporese, P., and C. Duso Different colonization patterns of phytophagous and predatory mites (Acari: Tetranychidae, Phytoseiidae) on three grape varieties: a case study. Exp. Appl. Acarol. 20: Cedola, C. V., N. E. Sanchez, and G. G. Liljesthrom Effect of tomato leaf hairiness on functional and numerical response of Neoseiulus californicus (Acari: Phytoseiidae). Exp. Appl. Acarol. 25: Clements, D. R., and R. Harmsen Predatory behavior and prey-stage preferences of stigmaeid and phytoseiid mites and their potential compatibility in biological control. Can. Entomol. 122: Cloutier, C., and S. G. Johnson Predation by Orius tristicolor (Hemiptera: Anthocoridae) on Phytoseiulus persimilus (Acarina: Phytoseiidae): Testing for compatibility between biocontrol agents. Environ. Entomol. 22: Croft, B. A., J. S. Blackwood, and J. A. McMurtry Classifying life-style types of phytoseiid mites: diagnostic traits. Exp. Appl. Acarol. 33: Croft, B. A., R. H. Messing, J. E. Dunley, and W. B. Strong Effects of humidity on eggs and immatures of Neoseiulus fallacis, Amblysieus andersoni, Metaseiulus occidentalis and Typhlodromus pyri (Phytoseiidae): implications for biological control on apple, caneberry, strawberry and hop. Exp. Appl. Acarol. 17:

203 Downing, R. S., and T. K. Moilliet Relative densities of predaceous and phytophagous mites on three varieties of apple trees. Can. Entomol. 99: Duso, C., M. Fanti, A. Pozzebon, and G. Angeli Is the predatory mite Kampimodromus aberrans a candidate for the control of phytophagous mites in European apple orchards? BioControl 54: Faraji, F., A. Janssen, and M. W. Sabelis Oviposition patterns in a predatory mite reduce the risk of egg predation caused by prey. Ecol. Entomol. 27: Grostal, P., and D. J. O'Dowd Plants, mites and mutualism: leaf domatia and the abundance and reproduction of mites on Viburnum tinus (Caprifoliaceae). Oecologia 97: Hoy, M. A Integrated mite management in Washington apple orchards, pp Agricultural acarology: introduction to integrated mite management. Taylor and Francis Group, LLC, Boca Raton, FL. Hoyt, S. C Integrated chemical control of insects and biological control of mites on apple in Washington. J. Econ. Entomol. 62: Hoyt, S. C., and E. H. Beers Western predatory mite In E. H. Beers, J. F. Brunner, M. J. Willett and G. M. Warner (eds.), Orchard pest Management: a resource book for the Pacific Northwest. The Good Fruit Grower, Yakima, WA. Johnson, H. B Plant pubescence: an ecological perspective. Botanical Review 41: Kenward, M. G., and J. H. Roger Small sample inference for fixed effects from restricted maximum likelihood. Biometrics 53:

204 Kottek, M., J. Grieser, C. Beck, B. Rudolf, and F. Rubel World Map of the Köppen- Geiger climate classification updated. Meteorologische Zeitschrift 15: Kreiter, S., M.-S. Tixier, and T. Bourgeois Do generalist phytoseiid mites (Gamasida: Phytoseiidae) have interactions with their host plants? Int. J. Trop. Insect Sci. 23: Krips, O. E., P. W. Klejin, P. E. L. Willems, G. J. Z. Gols, and M. Dicke Leaf hairs influence searching efficiency and predation rate of the predatory mite Phytoseiulus persimilis (Acari: Phytoseiidae). Exp. Appl. Acarol. 23: Levin, D. A The role of trichomes in plant defense. Q. Rev. Bio. 48: McMurtry, J. A Dynamics and potential impact of 'generalist' phytoseiids in agroecosystems and possibilities for establishment of exotic species. Exp. Appl. Acarol. 14: McMurtry, J. A., and B. A. Croft Life-styles of phytoseiid mites and their roles in biological control. Annu. Rev. Entomol. 42: McMurtry, J. A., G. J. de Moraes, and N. Famah Sourassou Revision of the lifestyles of phytoseiid mites (Acari: Phytoseiidae) and implications for biological control strategies. Syst. Appl. Acarol. 18: Prischmann, D. A., and D. G. James Phytoseiidae (Acari) on unsprayed vegetation in southcentral Washington: implications for biological control of spider mites on wine grapes. Int. J. Acarol. 29: Riddick, E. W., and A. M. Simmons Do plant trichomes cause more harm than good to predatory insects? Pest Manag. Sci. online first. 190

205 Roda, A., J. Nyrop, and G. English-Loeb Leaf pubescence mediates the abundance of non-prey food and the density of the predatory mite Typhlodromus pyri. Exp. Appl. Acarol. 29: Roda, A., J. Nyrop, M. Dicke, and G. English-Loeb Trichomes and spider-mite webbing protect predatory mite eggs from intraguild predation. Oecologia 125: Roda, A., J. Nyrop, G. English-Loeb, and M. Dicke Leaf pubescence and two-spotted spider mite webbing influence phytoseiid behavior and population density. Oecologia 129: Schmidt, R. A Leaf structures affect predatory mites (Acari: Phytoseiidae) and biological control: a review. Exp. Appl. Acarol. 62: Seelmann, L., A. Auer, D. Hoffmann, and P. Schausberger Leaf pubescence mediates intraguild predation between predatory mites. Oikos 116: Southwood, R Plant surfaces and insects- an overview, pp In B. Juniper and R. Southwood (eds.), Insects and the Plant Surface. Edward Arnold, Oxford. Thistlewood, H A survey of predatory mites in Ontario apple orchards with diverse pesticide programmes. Can. Entomol. 123:

206 Table 7.1. Mean proportion of individuals on the Delicious side of arenas without leaf midribs included. Evaluation Species Prop. on Delicious a df F P 1 T. urticae (0.28, 0.58) 1, G. occidentalis (0.01, 0.57) 1, T. urticae (0.34, 0.59) 1, G. occidentalis (0.24, 0.88) 1, T. urticae (0.23, 0.52) 1, G. occidentalis (0.24, 0.88) 1, T. urticae (0.26, 0.46) 1, * G. occidentalis (0.24, 0.88) 1, a Mean, followed by asymmetric 95% confidence interval in parentheses. *Indicates significance at P= Table 7.2. Mean proportion of T. urticae eggs laid, G. occidentalis eggs laid, and T. urticae eggs consumed by G. occidentalis on Delicious side of arenas, without midribs included. n Prop. On Delicious a df F P T. urticae eggs laid (0.22, 0.53) 1, G. occidentalis eggs laid (0.23, 0.95) 1, T. urticae eggs consumed (0.43, 0.89) 1, a Mean, followed by asymmetric 95% confidence interval in parentheses. 192

207 Table 7.3. Mean proportion of individuals on the Delicious side of arenas with leaf midribs included. Evaluation Species n Prop. on Delicious a df F P 1 T. urticae (0.29, 0.54) 1, G. occidentalis (0.12, 0.76) 1, T. urticae (0.34, 0.54) 1, G. occidentalis (0.19, 0.87) 1, T. urticae (0.27, 0.47) 1, G. occidentalis (0.13, 0.81) 1, T. urticae (0.29, 0.62) 1, G. occidentalis (0.14, 0.86) 1, a Mean, followed by asymmetric 95% confidence interval in parentheses. *Indicates significance at P= Table 7.4. Mean proportion of T. urticae eggs laid, G. occidentalis eggs laid, and T. urticae eggs consumed by G. occidentalis on Delicious side of arenas, with midribs included. n Prop. on Delicious a df F P T. urticae eggs laid (0.18, 0.52) 1, G. occidentalis eggs laid (0.44, 0.96) 1, T. urticae eggs consumed (0.32, 0.74) 1, a Mean, followed by asymmetric 95% confidence interval in parentheses. 193

208 Table 7.5. Mean proportion of G. occidentalis on the side of arenas with cotton added at each evaluation interval. Prop. G. occidentalis Eval. n on addition side a df F P Hour (0.46, 0.84) 1, Hour (0.54, 0.89) 1, Hour (0.71, 0.98) 1, * Hour (0.58, 0.92) 1, Hour (0.63, 0.94) 1, Hour (0.74, 1.00) 1, * Day (0.22, 0.64) 1, Day (0.10, 0.35) 1, Day (0.33, 0.74) 1, a Mean, followed by asymmetric 95% confidence interval in parentheses. *Indicates significance at P= Table 7.6. Mean proportion of G. occidentalis eggs laid and T. urticae eggs consumed on the side of arenas with cotton added. n Prop. on Addition Side a df F P G. occidentalis eggs laid (0.00, 1.00) 1, T. urticae eggs consumed (0.25, 0.51) 1,

209 Table 7.7. Mean counts of trichomes, pollen, and mites for Delicious and Golden Delicious leaves from an insecticide/fungicidefree orchard; asymmetric 95% confidence intervals in parentheses. Cultivar Trichomes/ref. line Pollen/cm 2 A. caudiglans G. flumenis Delicious' (14.72, 20.09) (11.27, 17.87) 1.42 (0.80, 2.53) (8.07, 17.44) Golden Delicious' 3.97 (3.39, 4.66) (10.14, 16.08) 4.41 (2.84, 6.86) (13.75, 28.72) F (df = 1,15) P < Cultivar Z. mali A. schlectendali Delicious' (44.11, 84.76) (32.51, ) Golden Delicious' (47.18, 90.58) (31.98, ) F (df = 1,15) P

210 Fig Mean trichome counts per 3 mm reference line and 95% asymmetric confidence intervals for each side of leaf half arenas without midribs included. F(1,18)=2.37, P= Fig Mean trichome counts per 3 mm reference line and 95% asymmetric confidence intervals for each side of leaf half arenas with midribs included. F(1,18)=5.22, P=

211 Fig a) Mean T. urticae eggs consumed and b) eggs laid per G. occidentalis female on whole leaf disks (no choice) of two apple cultivars, with asymmetric 95% confidence intervals. a) F(1,11)=0.47, P= b) F(1,11)=1.55, P=