AN ABSTRACT OF THE THESIS OF

Deirdre A. Prischmann for the degree of Master of Science in Entomology presented on

July 6,2000. Title: Biological Control of Spider Mites (Acari: Tetranychidae) on Grape

Emphasizing Regional Aspects. Redacted for privacy

Abstract approval: ,/l' Brian A. Croft

/

During summer of 1998 and 1999,34 and 10 vineyard sites, respectively, were sampled to assess spider mite pests and associated biological control by phytoseiid mites.

Vineyards studied spanned five major valleys in western Oregon where grape production occurs. Leaf samples were taken from site perimeters and centers. One leaf was taken every ten meters of border length, five meters inward from the border to prevent wind- biased or extreme edge effects, while 20 leaves were taken at regular intervals from centers. Variables recorded at each site were: plant age, grape variety, chemical spray information and local vegetation occurring in proximity to vineyards. Sites were

categorized as either agricultural or riparian based on what surrounding vegetation type

was in the majority. Several parametric and non-parametric tests were used to analyze

data, including multiple linear regressions using a computer-based genetic algorithm in

conjunction with the AIC criterion to pre-select a subset of explanatory variables.

Typhlodromus pyri was the predominant phytoseiid mite and Tetranychus urticae was the most abundant tetranychid mite sampled. High levels of T. urticae were found

when predator densities were very low, and low levels of T. urticae occurred when predator densities were moderate or high. Phytoseiid densities were highest in June and

July, while T. urticae densities were highest from August to September. The latter's densities were significantly higher in vineyards surrounded primarily by agriculture, while phytoseiid densities were not significantly different between the two categories.

Predatory phytoseiids had significantly higher densities on vineyard edges, while T. urticae densities were higher in vineyard centers. Caneberry, cherry and grape habitats appeared to be sources of predator immigration, while no vegetation type consistently served as a short-range or nearby immigration source for spider mites. Due to insufficient data, pesticide information was not included in multiple linear regression models, although certain chemicals used in vineyards can potentially impact mite populations. Impacts of surrounding vegetation type, grape variety, regional location, plant age, and presence of other mites on phytoseiid and T. urticae densities are discussed. ©Copyright by Deirdre A. Prischmann July 6, 2000 All Rights Reserved Biological Control of Spider Mites (Acari: Tetranychidae) on Grape Emphasizing Regional Aspects

by

Deirdre A. Prischmann

A THESIS

submitted to

Oregon State University

in partial fulfillment of the requirements for the Degree of

Master of Science

Presented July 6,2000 Commencement June 2001 Master of Science thesis of Deirdre A. Prischmann presented on July 6,2000

APPROVED:

Redacted for privacy

,/'Major Professor, representing Entomology

Redacted for privacy

Redacted for privacy

I understand that my thesis will become part of the permanent collection of Oregon State University libraries. My signature below authorizes releases of my thesis to any reader upon request.

Redacted for privacy

Deirdre A. Prischmann, Author ACKNOWLEDGMENT

I would like to thank Dr. Brian A. Croft for giving me a wonderful opportunity in entomology and for his help with this project and editing. Thank you to my other committee members Drs. Lynn Royce, C. Candolfi-Vasconcelos, and J. Morrell. I would also like to thank the following people for their invaluable help on this project: Drs.

Hang-Kwang Luh, Jim McMurtry, Jack Lattin, Rob Progar, Greg Brenner, Peter

Schausberger and all the vineyards owners and managers who made this possible. Thank

you to the Oregon State University Entomology Department and the Viticulture

Consortium for funding this research. Finally, thank you to my friends and family for

patience, support, and love. TABLE OF CONTENTS

1. INTRODUCTION...... 1

1.1 Biocontrol of Tetranychids by Phytoseiids on Grape...... 1

1.2 Objectives...... '...... 2

1.3 Hypotheses...... 3

2. LITERATURE REVIEW ...... 4

2.1 Oregon's Viticulture Industry ...... 4

2.2 Grape Mite Fauna ...... 5

2.3 Ecology of Tetranychus urticae on Grape...... 6

2.4 Ecology of Typhlodromus pyri on Grape ...... 8

2.5 Role of Other Arthropods on Grape ...... 10

2.5.1 Mites...... 10 2.5.2 Insects and Spiders ...... 10

2.6 Impacts of Surrounding Vegetation on Mite Dynamics ...... 11

2.6.1 Riparian Vegetation ...... 11 2.6.2 Agricultural Crops ...... 12

2.7 Impacts of Vineyard Management on Mite Dynamics ...... 13

2.7.1 Effects of Pesticides...... 13 2.7.2 Grape Variety...... 14 2.7.3 Ground Cover ...... 16

3. EXPERIMENTALDESIGN...... 18

3.1 Site Location ...... 18

3.2 Site Designation as Agricultural or Riparian ...... 18

3.3 Sampling Methods ...... 19 TABLE OF CONTENTS (Continued)

3.3.1 Methods in 1998 ...... 19 3.3.2 Methods in 1999...... 20 3.3.3 Plant Sampling Techniques ...... 21

3.3.3.1 Grape Vines ...... 21 3.3.3.2 Adjacent Vegetation ...... 21

3.4 Mite Identification ...... 22

3.5 Insect Identification ...... 23

3.6 Statistical Methods ...... 24

4. RESULTS...... 28

4.1 Arthropod Fauna ...... 28

4.1.1 Vineyards...... 28 4.1.2 Adjacent Vegetation ...... 30

4.2 Seasonal Density Trends ...... 33

4.2.1 1998 Trends ...... 33 4.2.2 1999 Trends ...... 33

4.3 Effect of Surrounding Vegetation on Mite Density ...... 35

4.3.1 Regional LeveL ...... 35 4.3.2 Adjacent Vegetation Level Effects ...... 39

4.3.2.1 Vegetation Edge Effects ...... 39 4.3.2.2 Quantitative Sampling of Adjacent Vegetation ...... 42

4.4 Multiple Linear Regression ...... 44

4.4.1 Model for Typhlodromus pyri + Amblyseius andersoni Density...... 46 4.4.2 Model for Tetranychus urticae Density...... 48

4.5 Biological Control ...... 50

5. CONCLUSIONS...... 54 TABLE OF CONTENTS (Continued)

BIBLIOGRAPHy...... 57

APPENDICES...... 71

Appendix A Pesticide toxicity ratings for A. andersoni ...... 72 Appendix B Pesticide toxicity ratings for T. pyri...... 77 LIST OF FIGURES

Figure

1. Location of sites sampled in 1998 and 1999 ...... 19

2. Depiction of sampling methods for grapevines and adjacent vegetation ...... 22

3. Mite seasonal density trends in (a) 1998; and (b) 1999 ...... 34

4. Mean mite densities in vineyards with different surrounding vegetation types in 1998 ...... 36

5. Seasonal mite density in vineyards in 1998 in sites (a) primarily surrounded by agriculture; and (b) primarily surrounded by riparian vegetation ...... 37

6. Seasonal density in agricultural and riparian vineyards in 1998 of (a) T. pyri + A. andersoni; and (b) T. urticae...... 38

7. Proportion of pest and predatory mites in 1999 in (a) agricultural crops; and (b) natural vegetation ...... ,...... 43

8. Comparison of T. pyri + A. andersoni and T. urticae densities in 1998 and 1999...... ,..... ,...... ,...... 51 LIST OF TABLES

1. A selection of chemicals used in western Oregon vineyards and associated

median toxicity ratings for T. pyri based on the SELCTV U and IOBC ~ databases ...... 15

2. Arthropods found on grape leaves in western Oregon vineyards ...... 29

3. Arthropods found on vegetation adjacent to vineyards in 1999 ...... 31

4. Mites encountered during sampling, families and functional feeding groups ..... 32

5. Effects of vegetation on pest and predatory mite distributions within vineyards from June to September in (a) 1998; and (b) 1999 ...... 40

6. Effects of vegetation on pest and predatory mite distributions within vineyards combining data from June to September in 1998 and 1999 ...... 41

7. Linear regression model results developed using genetic algorithms and the Ale criterion to compute a subset of explanatory variables ...... 45

8. Biological control in western Oregon vineyards as determined by predator I prey ratio data, based on conservative estimates of data given for other phytoseiids Y••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 52 LIST OF APPENDIX TABLES

A. Median toxicity ratings for A. andersoni based on the SELCTV a and IOBC 13 databases...... 73

B. Median toxicity ratings for T. pyri based on the SELCTV a and IOBC 13 databases...... 77 Biological Control of Spider Mites (Acari: Tetranychidae) on Grape Emphasizing Regional Aspects

1. INTRODUCTION

1.1 Biocontrol of Tetranychids by Phytoseiids on Grape

Biological control, or "biocontrol", is defined as "use of natural enemies to manage populations of pest organisms," and includes the importation, introduction and conservation of beneficial organisms (Lewis et al. 1997). Integrated pest management

(IPM) coordinates a wide array of pest control methods, economic issues, and environmental, societal and production considerations (Kogan 1998, Kogan et a1.l998).

Spider mites of the Tetranychidae are widespread phytophagous pests that cause economic damage to many crops, including grape (Englert and Kettner 1983, Girolami

1987, Pritchard and Baker 1955, Schruft 1985). Attaining successful biocontrol of spider mites in vineyards is highly desirable, as these pests damage plants by destroying chloroplasts and reduce yields by decreasing photosynthetic activity (Englert and Kettner

1983, Flaherty et al. 1992, Garren et al. 1983, Hanna et al. 1996, Hughes 1959, Welter et al. 1989). Spider mites often become secondary pests in agricultural systems, including grape, when toxic materials kill their natural enemies, thus disrupting natural control processes (Bower 1980, Flaherty et al. 1985, Roush et al. 1980, Seymour 1982, Walters

1976). Predatory phytoseiid mites have been used worldwide to control pest tetranychids in vineyards, with variable effectiveness (Caccia et al. 1985, Camporese and Duso 1996,

Duso 1989, 1992, Gambaro 1972, Girolami 1987, Tixier et al. 1998). 2

Several factors affect success of vineyard IPM programs. Understanding mite, plant and human dynamics in target ecosystems is critical for pest control because organisms interact with their environment and are not constrained by artificial boundaries such as crop margins (Kogan 1998, Speight 1983). Climatic conditions, movement patterns, inter- and intraspecific competition, reproduction rates, food source, surrounding vegetation, plant characteristics, vineyard cultural practices and macropredators all influence mite populations and their interactions (Coli et al. 1994,

Collyer 1964, Duso and Pasqualetto 1993, Huffaker et al. 1970, Kennedy and Smitley

1985, McMurtry and Croft 1997, Tixier et al. 1998, Walter 1996).

Pest assessment is critical for development of economic thresholds, allocating resources and research development (Wearing 1975). One step towards implementing effective and economical grape IPM programs in western Oregon is investigating predator-prey dynamics in vineyards and evaluating how their interactions are affected by surrounding vegetation types.

1.2 Objectives

The primary goal of this thesis was to document the degree of spider mite pests and associated biological control by phytoseiid mites within western Oregon vineyards.

A second objective was to clarify how mite density changes seasonally. A third goal was to elucidate effects of surrounding vegetation on dominant predatory and pest mite species. A final objective was to explore how multiple factors, including pesticide use, grape variety, plant age and vineyard appellation (regional designation) affect mite densities. 3

1.3 Hypotheses

In relation to the objectives listed above, several hypotheses reflect expectations about mite populations and their management under growing conditions in western

Oregon vineyards.

1. In most commercial vineyards in Oregon predaceous mites and insects largely control

spider mites at low population densities.

2. Vineyards surrounded by agricultural crops will have more spider mites and fewer

predatory mites than vineyards surrounded by riparian or native vegetation.

3. Vineyards with heavy pesticide usage will have lower beneficial mite densities and

higher pest mite densities than vineyards with low pesticide usage. 4

2. LITERA TURE REVIEW

2.1 Oregon's Viticulture Industry

Domestication of Vitis vinifera Linnaeus grapevines began around 4000 B. C. in central Asia (De Blij 1981, Jackson and Schuster 1987, Winkler 1962). By the late 400s

(A. D.) viticulture had become securely established in France (De Blij 1981). Oregon's viticulture industry is relatively young compared to those in Europe and California

(Pascal 1997a). A major factor that dampened production in Oregon until the 1960s was prohibition of alcohol (Pascal 1997a). Planted acres in Oregon have increased steadily since the 1900s, with 9,800 planted acres in 1999 (Pascal 1997a, Rowley et al. 2000). In

1999, 7,400 acres were harvested, yielding 2.4 tons of grapes worth 23.5 million dollars

(Rowley et al. 2000).

Climate dictates where quality wines can be produced (Jackson and Schuster

1987, Winkler 1962). Grapes are grown on every continent, excluding Antarctica, although most wine producing regions lie in latitudinal bands north and south of the equator where temperatures are 1O-20oC (Jackson and Schuster 1987, Perold 1927,

Winkler 1962). Oregon is located within these temperature boundaries and growing areas west of the Cascade Mountains are considered cool climate regions (1ackson and

Schuster 1987). These areas are characterized by cool autumns and diurnal temperature contrasts (hot days, cool evenings) and generally produce high quality wines (Jackson and Schuster 1987, Lett 1983, Winkler 1962).

Oregon has five main regions (appellations) where grape production occurs, each with a unique climate. These regions are Columbia Valley, North Willamette Valley, 5

South Willamette Valley, Umpqua Valley,and Rogue River Valley (Ahmedullah 1980,

Jackson and Schuster 1987, Pascal 1997b).

2.2 Grape Mite Fauna

Grapevines support a rich mite fauna, including phytophages, fungivores,

omnivores and predators from several acarine families: Tetranychidae, Tenuipalpidae,

Eriophyidae, Tydeidae, Phytoseiidae and Anystidae (Bames 1970, Bames 1992, Flaherty

1992, Flaherty et al. 1992, James and Whitney 1993, Jeppson et al. 1975, Vacante and

Tropea-Grazia 1987, Walter and Denmark 1991).

Several species of tetranychids occur on grape in humid (Willamette) and arid

(Rogue, Hood River) valleys of western Oregon, including: Willamette mite

(Eotetranychus willametti Ewing), European red spider mite [Panonychus ulmi (Koch)],

Tetranychus mcdanieli McGregor, Pacific spider mite (Tetranychus pacificus McGregor)

and two-spotted spider mite (Tetranychus urticae Koch) (Flaherty et al. 1992, Garren et

al. 1983, Schruft 1985). The latter mite is of primary interest in this study because of its

prevalence in western Oregon.

Predaceous phytoseiid mites are important biocontrol agents that are most often

effective at keeping spider mite pests at low densities, although macropredaceous insects

and spiders contribute to mite pest control (Camporese and Duso 1996, Duso and

Pasqualetto 1993, Hussey and Huffaker 1976, McMurtry and Croft 1997). Resident or

introduced phytoseiid populations can suppress pest mite outbreaks and therefore reduce

use of costly pesticide sprays (Duso 1989, Duso et al. 1991). Common phytoseiid species

on grape are Typhlodromus pyri Scheuten, Metaseiulus occidental is (Nesbitt), 6

Amblyseius andersoni (Chant), and Kampimodromus aberrans (Oudemans) (Corino

1985, Duso 1992, Duso and Camporese 1991, Gambaro 1987, Girolami 1987). T. pyri is the primary predatory phytoseiid in western Oregon vineyards and it is the focus of this study, although A. andersoni (A. potentillae) and M. occidentalis (Typhlodromus occidental is) have previously been sampled in Oregon (Chant 1985, Hadam et al. 1986).

2.3 Ecology of Tetranychus urticae on Grape

Spider mites belong to the class Arachnida, subclass Acari, order Acariformes, suborder Prostigmata, cohort Eleutherogonina and family Tetranychidae (Krantz 1971).

Synonyms of T. urticae include T. telarius and T. bimaculatus (Bellotti 1985). Dorsal and ventral striation patterns and male aedeagus morphology are used to identify tetranychid species (Baker and Tuttle 1994).

Two-spotted spider mites have five life stages: egg, larva, protonymph, deutonymph and adult (Lindquist 1985). Production of male offspring parthenogenically

(arrhenotoky) is not uncommon for T. urticae females (Helle and Pijnackee 1985). These spider mites are multivoltine, and females have developmental times of 16.9 days at 59­

83°p (Croft and Morse 1979, Helle and Pijnackee 1985, Jeppson et al. 1975).

Reproduction rates vary according to food source and environmental conditions; females feeding on cotton produced 10.5 eggs per day as opposed to 3.3 when on grape (Sabelis

1985a, Schruft 1987, Wrensch 1985). Under conditions of high temperature and / or low humidity population growth of spider mites is favored over predatory phytoseiid population growth (Sabelis 1985b). 7

Phytophagous spider mites cause damage by piercing leaf cells and ingesting chlorophyll (Croft et al. 1998, Seymour 1982). In high numbers these mites cause leaf discoloration and drying, thus reducing photosynthetic activity and lowering yield and crop quality (Englert and Kettner 1983, Flaherty et al. 1992, Garren et al. 1983, Hanna et al. 1996, Hughes 1959, Welter et al. 1989). Vine vigor, water stress and dusty conditions affect severity of spider mite infestations by increasing leaf carbohydrates and affecting temperature, humidity and airflow (Flaherty et al. 1985, Hanna et al. 1996, Wrensch

1985). Certain tetranychid species, including T. urticae and T. pacificus on grape, produce dense webbing that interferes with plant growth and provides protection to spider mites from predation, environmental conditions (wind, rain) and some pesticides (Gerson

1985, Hussey and Huffaker 1976, Sabelis 1985b).

Two-spotted spider mites aggregate in colonies on grape foliage, most often on ventral, peripheral leaf surfaces (Engle and Ohnesorge 1994b, Sabelis 1985c, Slone 1999,

Zhang and Sanderson 1992). Young females form colonies near parental colonies, creating spider mite patches (Zhang and Sanderson 1992).

T. urticae adult females tum dark orange-red and overwinter in late autumn underneath loose bark (Veerman 1985). Photoperiods, or short daylengths, low temperatures and poor host plant quality can induce diapause (Veerman 1985).

Dispersal of spider mites can occur actively by crawling and passively via air currents (Flexner et al. 1991, Kennedy and Smitley 1985). The spider mite, T. urticae, uses a characteristic posture with raised forelegs or lowers itself on silken threads to become airborne (Kennedy and Smitley 1985). Vertical vegetation profile and wind speed influence how far pest mites are transported aerially (Croft 1997). 8

2.4 Ecology of Typhlodromus pyn on Grape

Phytoseiid mites belong to the class Arachnida, subclass Acari, order

Acariformes, suborder Prostigmata, cohort Gamasina and family Phytoseiidae (Krantz

1971). A synonym of T. pyri is Typhlodromus tiliae (McMurtry et al. 1970). Phytoseiid mites can be identified primarily by their setal number and pattern (Chant 1985).

Phytoseiids have five life stages: egg, larvae, protonymph, deutonymph and adult

(Duso and Camporese 1991). Fecundity, reproduction and developmental time change with species, temperature and food source (Duso and Camporese 1991, Engle and

Ohnesorge 1994a, Hardman and Rogers 1991, Hussey and Huffaker 1976, McMurtry et al. 1970, Zemek 1993). Humidity affects egg survival and is species dependent; M. occidentalis eggs are more tolerant to desiccation than T. pyri eggs, which are more tolerant than A. andersoni eggs (Croft et al. 1993). Eggs hatch into non-feeding larvae with only six legs, while subsequent stages have eight legs (Collyer 1981, Schausberger and Croft 1999a).

T. pyri is polyphagous and can develop and reproduce on either pollen, grape epidermal outgrowths or pearl hairs, thrips larvae, eriophyids or tetranychids (Collyer

1981, Duso and Camporese 1991, Engle and Ohnesorge 1994a, Kapetanakis and

Cranham 1983). Analysis of mite gut contents collected in German vineyards showed that pollen was the primary food source in spring, eriophyids and thrips larvae in summer, and spider mites from August onward (Engle and Ohnesorge 1994a). With the exception of larvae, cannibalism can occur among mobile life stages (Croft and Croft

1993, Croft et al. 1996, MacRae and Croft 1993, Schausberger and Croft 1999b). 9

Phytoseiids lack eyes and detect prey by olfactory chemicals including contact and volatile kairomones (Sabelis and Dicke 1985, Zhang and Sanderson 1992).

Phytoseiids attack spider mites by grasping them with their front pair of legs and mouthpart appendages, and after subduing prey they extract their body fluids

(Flechtmann and McMurtry 1992). T. pyri has a mean predation rate of 10.5 T. urticae larvae per day at 25°C (MacRae and Croft 1993).

Predator-prey proportions, or ratios, are often used to indicate biocontrol success

(Croft and McGroarty 1977). Ratios of phytoseiids to spider mites larger than one to ten are considered favorable in many agricultural systems for several species of phytoseiids

(Croft and Nelson 1972, Strong and Croft 1995, Wilson et al. 1984).

Mated adult females of T. pyri aggregate and overwinter in crevices on grape

plants and surrounding habitat in late autumn (Gambaro 1986, Hussey and Huffaker

1976, Overmeer 1985, Schulten 1985). Temperature and photoperiod largely control diapause in phytoseiids (Hussey and Huffaker 1976, Overmeer 1985).

T. pyri is often associated with cool, humid areas (Dunley and Croft 1990,

MacRae and Croft 1993). This mite inhabits leaf undersurfaces, often in central areas near leaf veins or domatia (Engle and Ohnesorge 1994b, Walter and O'Dowd 1992b).

Phytoseiids disperse passively via air currents and sometimes via phoresy,

although they can employ active ambulatoriallocomotion as well (Sabelis and Dicke

1985, Tixier et al. 1998). T. pyri and K. aberrans (Amblyseius aberrans) disperse slowly, while M. occidentalis and A. andersoni disperse more rapidly (Croft et al. 1996, Dunley

and Croft 1990, Duso 1989, Strong and Croft 1995). Some phytoseiids assume a 10 characteristic posture to facilitate aerial dispersal that involves raising their upper body and legs (Johnson and Croft 1976, Sabelis and Dicke 1985).

2.5 Role of Other Arthropods on Grape

2.5.1 Mites

Not all acari inhabiting vineyards have a direct economic impact. Tydeidae are omnivorous mites that often feed on fungi found on grape leaves (Kinn and Doutt 1972,

Walter and Denmark 1991). Tydeids are alternate food sources for some predatory arthropods such as M. occidentalis when their preferred prey is absent (Calvert and

Huffaker 1974, Flaherty and Hoy 1971).

Anystis agilis (Acari: Anystidae) is a generalist predator that consumes several mite species, thrips, and other soft-bodied arthropods (Collyer 1953, Wilson et al. 1992).

Anystids only have two generations per year, so their use as biocontrol agents is limited

(Jeppson et al. 1975, Wilson et al. 1992). This mite's susceptibility to pesticides is similar or sometimes greater than that of T. pyri, and this factor can often regulate it to a minor occurrence in commercial vineyards (Croft, personal communication).

2.5.2 Insects and Spiders

There are several generalist macropredators that contribute to pest control in vineyards by consuming tetranychids, although predatory mites are also eaten by some of these larger predators (Camporese and Duso 1996, Duso and Pasqualetto 1993, 11

McMurtry et al. 1970). These predators include: anthocoridae (minute pirate bugs), chrysopidae (lace wings), miridae (plant bugs), Stethorus spp. (coccinellidae),

Scolothrips sexmaculatus (six-spotted thrips) and Aranae (spiders) (Agrawal and Karban

1997, Collyer 1952, 1953, 1964, Daane and Yokota 1997, Easterbrook et al. 1985,

English-Loeb et al. 1998, Fulmek 1930, Jeppson et al. 1975, McMurtry et al. 1970,

Walter 1996, Walters 1976). Densities of these macropredaceous arthropods and the extent to which they provide biocontrol of spider mites in western Oregon vineyards are unclear.

2.6 Impacts of Surrounding Vegetation on Mite Dynamics

2.6.1 Riparian Vegetation

Riparian vegetation bordering agricultural crops can impact resident arthropods by serving as chemical-free refuges (Boller et al. 1988, Croft et al. 1990, Dennis and Fry

1992, Lester et al. 1998, Tuovinen and Rokx 1991). Natural enemies immigrate into nearby crops by aerial dispersal or ambulatory locomotion (Sabelis and Dicke 1985,

Tixier et al. 1998). Vegetation type, proximity, and beneficial arthropod density influence migration patterns and can increase beneficial mite populations within adjacent crops (Tixier et al. 1998). Boller et al. (1988), Kreiter et al. (in press) and Tuovinen and

Rokx (1991) surveyed vegetation adjacent to apple orchards and vineyards to determine which plant species harbor high phytoseiid mite populations. Favorable deciduous plants that contributed to natural enemy survival and immigration into crops included: black currant (Ribes nigrum), caneberry (Rubus spp.) and honeysuckle (Lonicera xylosteum) 12

(Boller et al. 1988, Kreiter et al. in press, Tuovinen and Rokx 1991). Favorable deciduous trees included: unsprayed apple (Malus domestica), ash (Fraxinus excelsior), dogwood (Comus sanguinea), American hazelnut (Corylus avellana) and horse chestnut

(Aesculus hippocastani) (Boller et al. 1988, Kreiter et al. in press, Tuovinen and Rokx

1991). Thus surrounding vegetation can have a large impact on mite population dynamics within vineyards, and is an important factor to consider in areawide pest management.

2.6.2 Agricultural Crops

Agricultural crops can serve as sinks or sources of either pest or predatory mites

(Croft 1997). The tendency for a crop to serve as one or the other depends on many factors, such as suitability as a food for the pest, whether it is treated with pesticides that exclude natural enemies and contribute to pest development, dispersal rates of pests and natural enemies, etc. (Croft 1997). Cultural practices on these crops also affect nearby vineyards, either by exporting chemically resistant pest mites or by chemical drift (Croft

1997, Garren et al. 1983, Hellman 1987). Croft (1997) has suggested that extensively sprayed crops such as com, hops and nursery plants all export resistant pest mites into adjacent fields in August and September. Vertical crop profiles also influences the degree of spider mite dispersal, with tall profile plants providing platforms for dispersing mites over greater distances than short profile plants (Croft 1997). 13

2.7 Impacts of Vineyard Management on Mite Dynamics

2.7.1 Effects of Pesticides

One cultural practice that is potentially disrupting to natural biocontrol systems in vineyards is chemical pesticide application (AliNiazee et al. 1974, Croft 1990, Croft and

Brown 1975). Broad-spectrum pesticide use decimates predator mite and beneficial insect populations and leads to resurgence of pest mite populations via chemical resistance and absence of natural enemies (AliNiazee and Cranham 1980, Croft and

Dunley 1993, Hull and Beers 1985). Vineyard chemical spray regimes can have deleterious impacts on beneficial species in several ways. An obvious effect is mortality by direct toxic dose (AliNiazee and Cranham 1980, Croft 1990, Croft and Brown 1975).

A more obscure effect is mortality of natural enemies and reduced prey searching ability due to chemical residues, some of which persist in the environment for months

(Croft 1990, Croft and Brown 1975, Hussey and Huffaker 1976). Sublethal pesticide doses can negatively affect phytoseiid population density and biocontrol by: reducing fecundity, increasing developmental time, causing egg mortality, rendering pest eggs and mobile stages repellent, eliminating pest and alternate food sources, or by secondary poisoning via contaminated prey species (Croft and Brown 1975, Hussey and Huffaker

1976, Walker and Penman 1978). As is evident, known pesticide effects vary between compounds, arthropod species and life stage, and as yet, many compounds have not been tested for effects on natural enemies (Downing and Moilliet 1972, Overmeer and van Zon

1983). 14

Phytoseiids also can develop resistance to chemicals by natural or artificial selection under field or laboratory conditions, respectively (Croft 1982, Croft and Dunley

1993). Organophosphate resistant strains of T. pyri have been found worldwide and are used in IPM and biocontrol programs (Croft 1982, Croft and Dunley 1993, Kapetanakis and Cranham 1983).

A list of chemicals reportedly used in vineyards in Oregon and their effects on phytoseiid mites, specifically T. pyri is presented in Table 1 (Oregon Agricultural

Statistics Service 1998, Prischmann 1998, 1999 unpublished surveys). Appendices A and

B contain comprehensive lists of chemical effects on A. andersoni and T. pyri, including materials not used in vineyards and details regarding the SELCTV and IOBC

(International organization for biological control of noxious animals and plants) databases

(Hassan et al. 1985, Theiling 1987, Theiling and Croft 1988).

2.7.2 Grape Variety

Grape variety and associated leaf characteristics such as hairiness and presence of domatia affect the presence of some predaceous arthropods and their ecology regardless of prey availability (Agrawal and Karban 1997, Duso 1992, English-Loeb et al. 1998,

Karban et al. 1995, Martinson and Dennehy 1995, Pemberton and Turner 1989).

'Zinfandel' and 'Chardonnay' appear particularly susceptible to damage by both E. willametti and T. pacificus (English-Loeb et al. 1998). Kreiter et al. (in press) found extremely low T. urticae densities on 'Riesling' and higher phytoseiids densities on

'Chardonnay' and 'Riesling' than 'Pinot Noir'. Duso (1992) found that T. pyri and K. aberrans preferred varieties with hairy, as opposed to glabrous, leaf undersurfaces. 15

Table 1. A selection of chemicals used in western Oregon vineyards and associated median toxicity ratings for T. pyri based on the SELCTV U and IOBC ~ databases.

ACTIVE TRADE USE SELCTV a IOBC~ EFFECT INGREDIENT NAME MEDIAN MEDIAN ON T.pyri TOXICITY TOXICITY RATING RATING 2,4-0 Phenoxy Herbicide No info. I Safe compounds BENOMYL Benlate Fungicide 2 No info. Safe CALCIUM Lime Insect!Acaricide 3.5 " 4 Moderate! POLYSULFIDE sulphur /Fungicide High toxicity CARBARYL Sevin Insecticide 2 3.5 Moderate toxicity COPPER Recop Fungicide No info. 1 Safe OXYCHLORIDE DIMETHOATE Champ Insect! Acaricide 5 4 High toxicity DIURON Urea Herbicide 2.5 " No info. Low toxicity ENDOSULFAN Thiodan Insect! Acaricide 2 2" Low toxicity FENARIMOL Rubigan Fungicide 3 I Low! Moderate toxicity GLYPHOSATE Roundup Herbicide 4.5· 1.5 Moderate! High toxicity IPRODIONE Rovral Fungicide 1.5· I Low toxicity MALATHION Maltox Insect! Acaricide 5 No info. High toxicity PARAQUAT Gramoxone Herbicide 4" No info. High toxicity PETROLEUM OIL Mineral oil Insect!Acaricide 2· No info. Low toxicity !Herbicide SIMAZINE Princep Herbicide 2· 1 Low toxicity SULPHUR Cosan Fungicide 4 4 Moderate! high toxicity TRIADIMEFON Bayleton Fungicide 2.5 1 • Low toxicity ct = Data calculated from the SELCTV database (see Appendices A and B; Theiling 1987, Theiling and Croft 1988): SELCTV Toxicity Rating: 1 2 3 4 5 SELCTV Range of effect: 0% <10% 10-30% 30-90% >90%

~ = Data calculated from research conducted by the IOBCIWPRS (International Organization for Biological Control of Noxious Animals and Plants - West Palaearctic Regional Section) (see Appendices A and B; Hassan et al. 1985):

IOBC Toxicity Rating: I 2 3 4 IOBC Range of effect: Harmless Slightly Moderately Harmful Harmful Harmful

.= Data for phytoseiidae, not just T.pyri. 16

There is some conflict as to the preference of A. andersoni. Overrneer and van Zon

(1984) found that A. andersoni prefer hairy leaf surfaces to smooth ones, while Duso

(1992) found this mite was hampered by trichomes and preferred glabrous or only slightly hairy leaves. Differences could be attributed to experimental design, plant odors or plants examined (Overrneer and van Zon 1984). 'Chardonnay', 'Pinot Noir' and

'Riesling' all have glabrous lower leaf surfaces (Duso 1992, Martinson and Dennehy

1995).

Presence of leaf domatia and associated trichomes are related to higher densities ofphytoseiid and tydeid mites (Agrawal and Karban 1997, Grostal and O'Dowd 1994,

Karban et al. 1995, Walter 1996, Walter and Denmark 1991, Walter and O'Dowd 1992a,

1992b). Domatia serve as shelters for vulnerable life stages (eggs, irnrnatures, molting individuals) and provide favorable microc1imates with increased humidity (Duso 1992,

Karban et al. 1995, Pemberton and Turner 1989, Walter 1996, Walter and O'Dowd

1992b). Associations between plants with domatia and predatory arthropods can potentially reduce damage by phytophagous pests by serving as favorable natural enemy reservoirs (Grostal and O'Dowd 1994, Pemberton and Turner 1989, Walter 1996, Walter and O'Dowd 1992a).

2.7.3 Ground Cover

Relationships between cover crop presence, composition, and height and associated spider mite density on nearby agricultural plants are complex and contradictory (Coli et al. 1994). T. urticae can develop and reproduce on several different plant species, although some are preferred (Alston 1994, Coli et al. 1994, 17

Flexner et al. 1991). Alston (1994) found certain cover crops promote damage from phytophagous mites by serving as alternate reproductive hosts, while Flaherty et al.

(1972) found that groundcover decreased the need for chemical controls of T. pacificus.

However, bare ground cultivation increases nitrogen levels in surrounding crop plants, which favors spider mite development, longevity and fecundity (Coli et al. 1994, Hussey and Huffaker 1976, Wilson et al. 1988).

Applications of certain herbicides on cover crop hosts can repel T. urticae and drive mites to seek alternate food sources (i.e. crop foliage) (Flexner et al. 1991).

Herbicides also depress predatory mite populations and can shift ground cover composition of plants and mites (Coli et al. 1994). Unsprayed cover crops provide overwintering refuges and food sources for predatory mites (Alston 1994, Coli et al.

1994, Nyrop et a1.l994). 18

3. EXPERIMENTAL DESIGN

3.1 Site Location

Vineyards studied in 1998 spanned five major valleys in Oregon where grape production occurs, with 18 sites located in North Willamette Valley, eight in South

Willamette Valley, four in Rogue River Valley, two in Umpqua Valley and two in Hood

River Valley (Fig. 1). In 1999, nine vineyards studied were in North Willamette Valley, with one in South Willamette Valley (Fig. 1). Potential sites were chosen randomly from a list compiled by extension agents and by searching for fields via automobile. Site owners were contacted by phone and asked for permission to begin sampling.

3.2 Site Designation as Agricultural or Riparian

Vineyards were categorized as either agricultural or riparian sites based on surrounding vegetation composition. Categories were assigned based on surveying vegetation within 50 meters of vineyard borders. Ifgreater than 50% of adjacent habitat was agricultural (com, grape, nursery, etc) sites were designated as agricultural.

Likewise, if greater than 50% of adjacent habitat was riparian (grass, forest, water, etc.) sites were designated as riparian. This categorization was applied to obtain a general overview of how vegetation affects mite densities. In 1998, agricultural crops primarily surrounded 17 sites and riparian plants primarily surrounded 17. 19

• = \"ine~'aids sampled in 1998 • = Vineyards sampled in 1998 lind 1999 • •

Figure 1. Location of sites sampled in 1998 and 1999.

3.3 Sampling Methods

3.3.1 Methods in 1998

During summer 1998, 32 vineyards (34 field sites) were sampled for levels of pest and predaceous mites. Each vineyard was sampled monthly for one to four months.

Nine percent of sites were sampled once~ 6%, twice~ 56%, three times, and 29%, four times. A map of each site was made showing local and general vegetation that occurred in proximity to vineyards. Several variables were recorded at each site to evaluate impacts on pest and predaceous mite densities, including: direction and number of rows, 20 grape variety, plant age, groundcover and groundcover height, presence of irrigation, acreage, site slope and pest problems. Vineyard owners and / or managers were interviewed about spraying programs, including chemicals used, frequency of application, and amount per acre. Representative samples of adult female phytoseiids were mounted for identification.

3.3.2 Methods in 1999

In summer 1999 vineyard sites sampled were reduced from 34 to 10 to focus research on the ability of phytoseiids to control tetranychids. Sites that were selected harbored spider mites in 1998 and were surrounded primarily by agricultural vegetation.

Sites were sampled once during June and July and twice during August and September.

This was in response to information gathered during 1998, which showed that the majority of spider mite activity was in late season. Vegetation adjacent to vineyards was also sampled to determine immigration potential of resident mites. Methods for sampling vineyards were similar to the previous summer. However, all adult female phytoseiids, males and nymphs were mounted and identified. Representative samples of adult female tetranychids were mounted for identification. Non-phytoseiid mites, non-tetranychid mites, insects (thrips, anthocorids, chrysopids, etc.) and arachnids found on vineyard leaves were counted and at regular intervals, specimens were identified. Maps of local and adjacent vegetation, groundcover and pesticide information were updated from 1998 surveys. 21

3.3.3 Plant Sampling Techniques

3.3.3.1 Grape Vines

Leaf samples were taken from site perimeters (edges) and centers. Edge sampling began at a comer and proceeded until the next comer. One leaf was taken every ten meters, five meters in towards vineyard centers to prevent wind-biased or extreme edge effects. Center samples consisted of twenty leaves taken at regular intervals (Fig. 2).

Leaves were randomly selected from within the canopy at chest height. Leaves from a single sample area (one edge or center) were placed on top of each other in consecutive order. Each perimeter or center sample was placed in a plastic bag, top folded down twice and secured with staples, and labeled with edge, date and site name. All plastic bags from a site were then grouped in one plastic bag and a knot tied. Samples were placed in a cooler (ca. 15°C), taken back to the laboratory, and each leaf scanned under a dissecting microscope at 65x. Leaves were stored at gOC in a refrigerator while awaiting examination. All leaves were examined within one week of sampling.

3.3.3.2 Adjacent Vegetation

Areas near edges of vegetation external to vineyards and directly bordering each vineyard plot were sampled in 1999 (Fig. 2). One plant sample was taken every 10 meters, five meters into adjacent vegetation. For a majority of vegetation types

(deciduous trees and shrubs, perennials etc.) the plant sample unit consisted of one leaf, 22

CORN GRAPE HOP lOIn ~ 5In 0 0 0 0 0 0 0 0 0 0 ,6.

0 • • 0 0 • • • • North • • • • 0

0 I VINEYARD SllE I • • ® 0 • •...... • • I-vn West • • • • East 0 • CenEr=20 -. leaves 0 • • AN' ® • • lOIn @ South • ~ 5In • • • • II II II II ... • • @ = &ljacl5rl v~iation (l) @ (} (i) Q Q Q (» Q 0 . ' 0 ~le

= Vineyard • leafsaJI1)le RIPARIAN AREA

Figure 2. Depiction of sampling methods for grapevines and adjacent vegetation.

whereas for conifers two-inch apical branch sections were examined. For tall grasses

(wheat, fescue) an entire stem was sampled, while corn samples were composed of twelve-inch apical leaf sections. Short grasses (lawn, pasture), urban areas and extremely young plants were not sampled.

3.4 Mite Identification

Mite numbers, life stages and species were recorded for each leaf. Adult females not sight identified were mounted on microscope slides in Hoyer' s solution and cleared 23 for 24 hours in a hot plate oven at 40°C. A phase-contrast microscope was used to examine slides and identify mites at lOO-450x. Chant's (1959) and Schuster and

Pritchard's (1963) keys were used to identify adult female phytoseiids using spermatheca morphology and dorsal and ventral setal patterns. Keys by Chant (1958) were used to identify phytoseiid immatures. Keys by Baker and Tuttle (1994) and Pritchard and Baker

(1955) were used to identify dorsal striation and setal patterns of adult female tetranychids. Drs. B. A. Croft and 1. A. McMurtry confirmed phytoseiid and tetranychid identity. Non-phytoseiid and non-tetranychid mites were identified by consultation with experts, including Drs. B. A. Croft, G. W. Krantz, 1. A. McMurtry, and P. Schausberger.

A majority of mites were identified to species or genera, although tydeids were only identified to family. Eriophyids were not noted or identified as they were not of key interest in this study.

Phytoseiid eggs not associated with sight-identified mites were assigned as T. pyri eggs. This was justified because only four sites contained significant adult female phytoseiid populations (n>10 in all sample dates) other than T. pyri, so it was reasonable to assume most eggs that were sampled were oviposited by T. pyri females.

3.5 Insect Identification

Presence of macropredatory insects on leaves was assessed qualitatively during

1998, and quantitatively in 1999. Representative insect and spider specimens were mounted as previously described. Insect species were identified to family and spiders to order using keys by Daly et al. (1998). Dr. J. D. Lattin aided in identification of various

Hemiptera. 24

3.6 Statistical Methods

Due to possible misidentification in 1998 from a lack of adequate mite mountings,

T. pyri and A. andersoni densities were grouped for analysis in both 1998 and 1999.

However, these identification problems only were present in <3% of samples because of the strong preponderance of T. pyri and only occasional occurrence of A. andersoni.

Statistical methods used included non-parametric binomial, Kruskal-Wallis,

Mann-Whitney and Dunn's tests, parametric repeated measures anal ysis of variance

(ANOVA) and multiple linear regressions with a log (x + 1) transfonnation. The

Kruskal-Wallis test is a non-parametric equivalent to one-way ANOVA, while the Mann­

Whitney test is comparable to at-test (Sokal and Rohlf 1981). Dunn's test is a multiple comparison method that ranks values, can be adjusted for ties and is used when there are unequal sample sizes (Glantz 1997). Non-parametric binomial tests were used to investigate possible vegetation edge effects. Kruskal-Wallis and Dunn's tests were used for 1998 data when investigating seasonal differences in mite densities, for all sites and between agricultural and riparian vineyards. The Mann-Whitney test was used for 1998 data when making planned comparisons of mite density between agricultural and riparian vineyards. Repeated-measures ANOVA was used with 1999 data to investigate differences in seasonal mite densities (vonEnde 1993).

Levene's test for equality of variance, which is robust against non-nonnality, and graphical exploration, was used to determine if assumptions for parametric statistics were met using non-transfonned and transfonned data (Ramsey and Schafer 1997). Serial correlation coefficients were used to calculate serial correlation test statistics (Z) and investigate possible temporal trends that prohibit use of parametric statistics without prior 25 correction (Ramsey and Schafer 1997). In preliminary assessments there was no evidence for serial correlation of T. pyri + A. andersoni densities (Z statistic =0.026, corresponding P =0.984, n =158). There was suggestive, but inconclusive evidence for serial correlation of T. urticae densities (Z statistic = 1.799, corresponding P = 0.074, n =

158), and therefore no adjustments for serial correlation were implemented. Spatial components of mite populations were not addressed.

A computer-based genetic algorithm and information-based model selection criteria AIC (Akaike's information criterion) was used to identify and pre-select factors important to regression models of T. pyri + A. andersoni and T. urticae densities. This model selection approach was implemented due to a large set of explanatory variables and higher efficiency and accuracy of algorithms compared to statistical modeling when dealing with large number of parameters (Anderson et al. 1994, Luh and Croft 1999).

Genetic algorithms treat data as codes on strings that are evaluated for problem solving fitness (Forrest 1993, Holland 1992, Luh and Croft 1999). Strings are either eliminated or kept in a model according to their fitness level, until after multiple runs the best solution is identified (Luh and Croft 1999). This model incorporates natural selection concepts, including crossover (mating) and mutation (Forrest 1993, Holland

1992, Luh and Croft 1999, Mitchell 1996). AIC values provide a solution to evaluate model performance, where the lowest AIC value represents the best model (Luh and

Croft 1999). This criterion penalizes models with excess parameters to obtain the most accurate, parsimonious solution (Anderson et al. 1994, Bozdogan 1987, Luh and Croft

1999). 26

A subset of explanatory variables was selected using the genetic algorithm and

AIC criterion from an initial complement of thirty variables, including continuous factors

(log + 1 of A. agilis density, log + 1 of Tydeidae density, and either log + 1 of T. urticae density or log + 1 of T. pyri + A. andersoni density), a discrete factor, plant age, and binary or presence / absence factors (site surrounded primarily by agricultural or riparian vegetation, 'Chardonnay', 'Gewtirztraminer', 'Pinot Gris', 'Pinot Noir', 'Riesling', Hood

River Valley, North Willamette Valley, South Willamette Valley, Umpqua Valley, Rogue

River Valley, caneberry, cherry, conifer, corn, dirt, filbert, grape, grass, hop, miscellaneous berry (blackberry and gooseberry), nursery, pasture, riparian, urban and vegetable). Interactions were not considered due to the high number of explanatory variables. Information from both years was combined in order to provide adequate data for analysis. Because data such as macropredaceous insect and thrips density were not recorded both years, these variables were eliminated from analyses. Six data points were eliminated from analysis because of missing information. Insufficient data prevented inclusion of pesticide information in model selection, although certain chemicals used in vineyards the have potential to impact mite populations (Table 1).

Single trials of the genetic algorithm used 62,500 model evaluations (250 generations, 250 models per population). The overall model was programmed by Dr. H.

-K. Luh (2000) and run on MATLAB® (MathWorks 1999). Resulting models were entered in SAS® (SAS Institute 1992) to obtain beta coefficients, or estimates of variable contribution to regression line slope, p-values and 95% confidence intervals. Two models were developed, one using density of T. pyri plus A. andersoni as a dependent variable, and one using T. urticae density as a dependent variable. Because of the 27 biologically important relationship between predatory and prey mites, T. urticae density was always included, or "locked", as an explanatory variable in the regression model when T. pyri + A. andersoni density was the dependent variable regardless of its significance in model selection. Likewise, T. pyri + A. andersoni density was always included as an explanatory variable in the regression model when T. urticae density was the dependent variable regardless of its significance in model selection. Correlations between variables were investigated using correlation coefficients of covariance matrices produced using MATLAB® (Math Works 1999). 28

4. RESULTS

4.1 Arthropod Fauna

4.1.1 Vineyards

Arthropod species from several families of mites and insects were sampled on grape leaves; their densities in 1998 and 1999 are listed in Table 2. When comparing data between years it is important to remember that there were differences in sample sizes and types between 1998 and 1999: 34 versus 10 sites, and a mix of agricultural and riparian sites versus all agricultural sites, respectively. Insects and spiders were only quantitatively sampled in 1999.

Typhlodromus pyri was the most abundant phytoseiid mite collected in both 1998 and 1999, followed by Amblyseius andersoni. Tetranychus urticae was the most abundant tetranychid mite both years, although Eotetranychus willametti and Panonychus ulmi were also observed. Anystis agilis and tydeid mites were frequently encountered during sampling, but population densities of each were not considered high or of major consequence. Families / species of arthropods that were represented by low densities were likely accidental immigrants and not permanent vineyard residents.

Thysanoptera (thrips) were the most abundant insect family sampled, although no predatory six spotted thrips (Scolothrips sexmaculatus) were observed. Beneficial insects commonly sampled included: minute pirate bugs (Anthocoridae; Orius tristicolor), green lacewings (Chrysopidae), ladybird beetles (Coccinellidae, including Stethorus spp.) and plant bugs (Miridae). 29

Table 2. Arthropods found on grape leaves in western Oregon vineyards.

TOTAL ARTHROPODS I TOTAL ARTHROPODS I NUMBER LEAF 1998 & SE a NUMBER LEAF 1999 & SE a 1998 1999 MITES Amblyseius andersoni 72 .0196 ± .Ql05 76 .0235 ± .0147 Anystis agilis 46 .0049 ± .0013 86 .0181 ± .0050 Bdellidae 0 - 1 .0002 ± .0002 Czenspinskia Lordi 9 .0009 ± .0007 0 - Eotetranychus willametti 0 - 541 .1444 ± .1434 Kampimodromus aberrans 2 .0001 ± .0001 1 .0002 ± .0002 'Metaseiulus citri 0 - 1 .0002 ± .0002 W"etaseiulus occidentalis 6 .0006 ± .0005 2 .0005 ± .0005 l,Neoseiulus aurescens 1 .0001 ± .0001 0 - l,Neoseiulus cucumeris 0 - 1 .0001 ± .0001 l,Neoseiulus fallacis 0 - 1 .0001 ± .0001 Panonychus ulmi 0 - 5 .0006 ± .0004 lPionidae 0 - 1 .0001 ± .0001 Platytetranychus spp. 0 - 1 .0001 ± .0001 Stigmaeidae 1 .0001 ± .0001 0 - lTenuipalpidae 1 .0001 ± .0001 5 .0001 ± .0001 Tetranychus urticae 1437 .1690 ± .1041 2126 .5795 ± .4179 Trombidioidea 0 - 1 .0001 ± .0001 lTydeidae 410 .0626 ± .0406 76 .0169 ± .0065 Typhlodromus pyri 1396 .4473 ± .0857 1411 .7697 ± .3953 INSECTS AND SPIDERS Anthocoridae Present - 12 .0021 ± .0009 Aranae Present - 21 .0055 ± .0018 Aranae Egg Sacks Present - 6 .0011 ± .0007 Cercopidae Present - 1 .0001 ± .0001 Chrysopidae Present - 74 .0164 ± .0045 Cicadellidae Present - 5 .0014 ± .0014 Coccidae Present - 5 .OOll ± .0005 Coccinellidae Present - 2 .0004 ± .0003 .....epidoptera - - 6 .0012 ± .0007 Miridae Eggs - - 75 .0222 ± .0124 Pentatomidae - - 14 .0030 ± .0030 Pseudococcidae Present - 13 .0027 ± .0012 Stethoris spp. Present - 80 .0178 ± .0117 Thysanoptera Present - 1026 .3239 ± .0623 Tingidae - - 1 .0002 ±.0002 LEAVES SAMPLED 10007 10007 6020 6020 SITES SAMPLED 34 34 10 10 SURROUNDING AREA II 17 AG, 17 RIP 17 AG, 17 RIP 10AG 10AG fx =Standard Error ~ AG =Site surrounded by a!ITicuItural crops. RIP =Site surrounded by native vegetation. 30

4.1.2 Adjacent Vegetation

Arthropod species that were quantitatively sampled in 1999 from vegetation immediately surrounding vineyards varied greatly (Table 3).

T. pyri occurred on a wide variety of vegetation and was the most abundant phytoseiid, followed by Kampimodromus aberrans, which was frequently collected from filbert. T. urticae had the highest density and greatest host range of any tetranychid mite.

Eotetranychus, Oligonychus and Platytetranychus species were often found in high densities on conifers, nursery plants and filbert, but were not usually observed in adjacent vineyards.

Frequently sampled insects included those from the Anthocoridae, Chrysopidae,

Coccinellidae, Miridae and Thysanoptera, although most of these insects were encountered on grape plants adjacent to studied vineyard blocks. An exception was high densities of mirid eggs on filbert leaves.

All mite species found in vineyards or adjacent vegetation and their family and primary feeding classification are listed in Table 4. A majority of mites were either phytoseiids or tetranychids, although several mite families that are not common on grapevines were encountered, including a parasitic water mite (Pionidae) and a litter or soil dweller (Nanorchestidae) (Krantz 1971). 31

Table 3. Arthropods found on vegetation adjacent to vineyards in 1999.

TOTAL ARTHROPODS I VEGETA TIONAL NUMBER LEAF 1999 OCCURRENCE 1999 WITHSE u MITES Amblyseius andersoni 12 .0048 ± .0021 apple, caneberry, cherry, grape Anystis af?ilis 11 .0040 ± .0018 ~ape, maple, willow Bdellidae 1 .0003 ± .0003 bean Cunaxidae 1 .0004 ± .0004 apple Eotetranychus spp. 2342 1.2247 ± 1.0592 conifer, filbert, juniper Eotetranychus willametti 77 .0264 ± .0211 caneberry, grape, juniper, pigweed Euseius finlandicus 2 .0006 ± .0004 cherry, maple Kampimodromus 80 .0316 ± .0223 filbert, grape aberrans Metaseiulus occidentalis 3 .0011 ± .0011 caneberry, cherry Metaseiuluspini 2 .0009 ± .0006 caneberry, conifer N anorchestidae 1 .0004 ± .0004 com Oligonychus sI>Q. 543 .1894 ± .1015 caneberry, conifer, juniper Oribatei 1 .0004 ± .0004 com Panonychus ulmi 24 .0086 ± .0081 apple, cherry, grape, juniper, pear Paraseiulus soleif?er 1 .0004 ± .0004 pigweed Platytetranychus spp. 238 .0847 ± .0847 nursery Stigmaeidae 5 .0011 ± .0011 apple, juniper Tarsonemidae 4 .0014 ± .0008 com, fern, filbert Tenuipalpidae 55 .0200 ± .0156 cherry, juniper, willow Tetranychidae spp. 3 .0006 ± .0006 caneberry Tetranychus urticae 5564 2.1917 ± .9478 apple, bean, caneberry, cherry, com, gooseberry, grape, hop, maple, nursery, pigweed, strawberry Tydeidae 356 .1255 ± .0773 apple, caneberry, cherry, grape, hop, maple, oak, pear, pigweed, scotchbroom, strawberry, willow Typhlodromus pyri 413 .1665 ± .0769 apple, caneberry, cherry, gooseberry, grape, hop, maple, oak, pigweed INSECTS AND SPIDERS Anthocoridae 8 .0030 ± .0010 com, filbert, grape, strawberry Aranae 1 .0004 ± .0004 caneberry Aranae Egg Sack 1 .0004 ± .0004 grape Chrysopidae 5 .0017 ± .0008 caneberry, Qfape Coccidae 1 .0006 ± .0006 grape Lepidoptera 1 .0006 ± .0006 grape Miridae 1 .0004 ± .0004 filbert MiridaeEgg 19 .0077 ± .0052 filbert, gr~e

Stethoris spp. 31 .0079 ± .0056 ~ape Thysanoptera 36 .0166 ± .0113 grape U =Standard Error 32

Table 4. Mites encountered during sampling, families and functional feeding groups.

MITE SAMPLED FAMILY COMMON ON FUNCTIONAL GRAPE? FEEDING GROUP

Amblyseius andersoni Phytoseiidae Y Omnivore U Anystis aRilis Anystidae Y Predator Bdellidae Bdellidae N Predator Cunaxidae Cunaxidae N Predator Czenspinskia lordi SaprogJyphidae N Fungivore / Herbivore Eotetranychus spp. Tetranychidae Y Herbivore Eotetranychus willametti Tetranychidae Y Herbivore

Euseius finlandicus Phytoseiidae Y Omnivore U

Kampimodromus aberrans Phytoseiidae Y Omnivore U Metaseiulus citri Phytoseiidae Y Predator Metaseiulus occidentalis Phytoseiidae Y Predator Metaseiulus pini Phytoseiidae N Predator Nanorchestidae Nanorchestidae N Fungivore Neoseiulus aurescens Phytoseiidae Y Predator

Neoseiulus cucumeris Phytoseiidae N Omnivore U Neoseiulus fallacis Phytoseiidae Y Predator Oli/?onychus spp. Tetranychidae N Herbivore Oribatei Unknown N Fungivore / Herbivore Panonychus ulmi Tetranychidae Y Herbivore Paraseiulus soleiger Phytoseiidae Y Predator Pionidae Pionidae N Parasite Platytetranychus spp. Tetranychidae N Herbivore Stigmaeidae Stigmaeidae Y Predator Tarsonemidae Tarsonemidae N Fungivore / Herbivore / Predator TenuipaJpidae Tenuipalpidae Y Herbivore Tetranychidae spp. Tetranychidae Y Herbivore Tetranychus urticae Tetranychidae Y Herbivore Trombidioidea Unknown N Parasite / Predator Tydeidae Tydeidae Y Fungivore I Herbivore I Predator Typhlodromus pyri Phytoseiidae Y Omnivore u U = Feed on both animal and plant matter. References: Croft et al. 1998, Flaherty 1992, Jeppson et al. 1975, Krantz 1970, Kreiter et al. in press, Schuster and Pritchard 1963. 33

4.2 Seasonal Density Trends

4.2.1 1998 Trends

In 1998 predatory mite densities reached a peak in July, while spider mites

exponentially increased in late September (Fig. 3); different letter designations show

significant differences in means at the P ::;; 0.05 level; bars represent standard errors.

Due to unequal sample sizes, data were analyzed using non-parametric tests. T. pyri + A. andersoni densities peaked in July with a mean of 0.691 ± 0.125. A Kruskal­

Wallis test showed no significant difference between T. pyri + A. andersoni densities

throughout the sampling period (P = 0.270; n = 104, df = 3), while there was a significant

difference in T. urticae densities (P < 0.001; n = 104, df = 3). T. urticae densities peaked

in September with a mean of 1.505 ± 0.921, and were significantly higher than in June

and July (P::;; 0.05; n = 46, df = 1 June; n = 41, df = 1 July; Dunn's multiple comparison

test). There were no other significant differences among groups. Seasonal patterns

differed between vineyards primarily surrounded by agriculture versus riparian

vegetation (see further details of presentation in 4.3.1).

4.2.2 1999 Trends

T. urticae and T. pyri + A. andersoni densities were transformed using a log (x +1)

transformation to equalize variance. T. pyri + A. andersoni densities were highest in June

with a median value of 0.725 ± 0.180 after backtransformation, although there were no

significant differences between dates (repeated-measures ANOVA, P =0.956; n =60, 34

i 3 I • T. pyri + A. andersoni i T. urticae , 2.5 - .- h I co I -Q) I .J i 2 I (/I -Q) ::: •• I :E 1.5 ! Q) I I C) I ...co 1 I I Q) > I I I ~ _ ~T I 0.5 T I .~ Y'" ~ . - ~ a a 'ab 0 ------. June- July August September (a) Sample Period

_ 0.8 co • T. pyri + A. andersoni Q) T. urticae .J 07 - . .- -(/I ~ 0.6 . :E ~ 8,0.5 ~" ...as ~/II ~ ~ 0.4 \0 I , ~ .", '0 0.3 I " I I ~ ...... II I •• ... .. -: 0.2 I . )( .1 - - .... ~ 0.1 0 .J 0 June July Early Late Early Late Aug. Aug. Sept. Sept. (b) Sample Period

Figure 3. Mite seasonal density trends in (a) 1998; and (b) 1999. 35 df = 5) (Fig. 3). T. urticae densities peaked in early August with a median density of

0.564 ± 0.294 after backtransformation, and a repeated-measures ANOV A showed no

significant differences between dates (P = 0.716; n = 60, df= 5). There were no

significant differences between T. urticae and T. pyri + A. andersoni densities throughout

the 1999 season (repeated-measures ANOVA, P = 0.788; n = 120, df = 10).

4.3 Effect of Surrounding Vegetation on Mite Density

4.3.1 Regional Level

Mite densities were compared between vineyards surrounded primarily by

agricultural vegetation and vineyards surrounded primarily by riparian vegetation in

1998. Because of planned comparisons of T. pyri + A. andersoni densities and T. urticae

densities between the two site categories and outlier data values, non-parametric Mann­

Whitney tests were conducted (Fig. 4).

T. pyri + A. andersoni densities did not differ significantly between vineyards

with separate site designations (P = 0.885; n =34, df =1; indicated by "x's" in Fig. 4),

whereas T. urticae densities were significantly greater in vineyards surrounded primarily

by agriculture (P = 0.001; n = 34, df = 1; indicated by "a" and "b" in Fig. 4).

Seasonal differences in mite densities in both vineyard categories were also

investigated. Graphical comparison revealed one site that was primarily surrounded by

riparian vegetation was an influential outlier. This site was omitted from further analysis

because recent use of a highly toxic (to predaceous mites) organophosphate pesticide was

identified as the cause of extremely high spider mite populations. This effect was quite 36

0.7 ...------i! 0 T. pyri + A. andersoni I • T. urticae 0.6 +------==;;;:======rl

0.5 +---~------~----~ -a:s x x ~ 0.4 +---1 -U) ~ 0.3 := 0.2

0.1

0+----"-­ Agricultural Riparian Surrounding Vegetation Designation

Figure 4. Mean mite densities in vineyards with different surrounding vegetation types in 1998.

different than in other sites and thus data from this site was removed from the population of interest. Seasonal trends in vineyards surrounded primarily by agriculture were similar to Fig. 3(a), with T. pyri + A. andersoni density peaking in July (mean =0.528 ± 0.127) and T. urticae densities exponentially increasing in September (mean = 1.404 ± 1.130)

(Fig. 5). A Kruskal-Wallis test showed T. pyri + A. andersoni densities were not significantly different (P = 0.636; n =60, df =3), while T. urticae densities were significantly different (P =0.001; n =60, df =3). T. urticae densities in September were significantly different from those in June and July (P ~ 0.05; n =26, df =1; Dunn's multiple comparison test). In vineyards surrounded primarily by riparian vegetation T. pyri + A. andersoni densities also peaked in July (mean =0.934 ± 0.257), and mean T. 37

3 • T. pyri + A. andersoni I - T. urticae b 2.5 .- I

as I -CD .J 2 I en I -CD ~1.5- II CD I C) as I ~ , CD 1 > I 0< I . T I .~ I 0.5 .L I a ~I ... - ilab 0 -~ - --.­ - ­ June July August September (a) Sample Period

3 • T. pyri + A. andersoni T. urticae 2.5 - .­ as -CD .J 2 en -CD ~1.5- CD C) as ~ ~ CD 1 > 0< 0.5 /~ ~ Y J...... ~ 0 - - - -iiiiiiI June July August- September (b) Sample Period

Figure 5. Seasonal mite density in vineyards in 1998 in sites (a) primarily surrounded by agriculture; and (b) primarily surrounded by riparian vegetation. 38

3 I • Agricultural I Riparian 2.5 - .­

_ 2 as ~ -;; 1.5 CI) == :! 1 .. , ,.. .. , .. , ~~ .. , 0.5 I .. • .. U o June July August September (a) Sample Period

3 • Agricultural - Riparian b 2.5 .­

2 -as ~1.5

U) -CI) 1 J :!== 0.5 !I / a ab ;/ 0 .. .­ June- July- August- September- a at) a ab -0.5 (b) Sample Period

Figure 6. Seasonal density in agricultural and riparian vineyards in 1998 of (a) T. pyri + A. andersoni; and (b) T. urticae. 39 urticae densities equaled 0.006 ± 0.006. A Kruskal-Wallis test showed neither T. pyri +

A. andersoni or T. urticae densities were significantly different (p =0.240~ n =40, df =3 and 0.377; n =40, df = 3, respectively). Seasonal densities in agricultural and riparian vineyards are shown separately for T. pyri + A. andersoni and T. urticae in Fig. 6. A

Kruskal-Wallis test showed no significant differences between T. pyri + A. andersoni densities (P =0.709; n =100, df =7), while there was a significant difference between T. urticae levels (P < 0.00 1 ~ n = 100, df = 7). Dunn's multiple comparison test, adjusted for ties, showed that the T. urticae density in vineyards primarily surrounded by agriCUlture in September was significantly higher than the T. urticae densities in June and July in

agricultural sites, and the densities in June and August in riparian sites (P ~ 0.05 for all~ n

=26, df =1 June agricultural and July, n =25, df =1 June riparian, n =20, df =1

August). There were no other significant differences among groups.

4.3.2 Adjacent Vegetation Level Effects

4.3.2.1 Vegetation Edge Effects

T. pyri + A. andersoni and T. urticae densities from vineyard centers were

compared to edge densities for all sample dates to investigate evidence of edge effects.

Mite densities were grouped according to corresponding adjacent vegetation types to

determine if specific vegetation influenced mite immigration into vineyards. Only

vegetation types representing a majority of each side were considered. Numbers of

times each mite species was present on a vineyard edge versus a center were analyzed 40

Table 5. Effects of vegetation on pest and predatory mite distributions within vineyards from June to September in (a) 1998; (b) 1999.

(a) T.pyri + T.urticae A. andersoni VEGETATION Center Center p-value Center Center p-value density density * density density * higher lower higher lower CANEBERRY 7 21 0.014 5 0 0.063 CHERRY 2 7 0.180 1 0 n/a CONIFER 3 1 0.625 1 1 1.000 CORN 1 3 0.625 0 1 n/a DIRT 4 12 0.077 1 0 n/a FILBERT 3 1 0.625 2 3 1.000 GRAPE 30 64 0.001 23 8 0.012 GRASS 23 32 0.281 13 5 0.096 HOP 0 4 0.125 0 1 n/a MISCELLANEOUS 2 2 1.000 0 2 0.500 BERRY NURSERY 2 6 0.289 0 0 n/a PASTURE 0 7 0.016 2 1 1.000 RIPARIAN 14 20 0.391 6 1 0.125 URBAN 16 14 0.855 4 2 0.688 VEGETABLE 5 0 0.063 0 0 n/a TOTAL 112 194 <.001 58 25 <.001 *j>-values from a non:parametric binomial test

(b) T.pyri + T. urticae A. andersoni VEGETATION Center Center p-value Center Center p-value density density * density density * higher lower higher lower CANEBERRY 0 1 n/a 1 2 1.000 CHERRY 1 11 0.006 1 5 0.219 CONIFER 0 1 n/a 0 4 0.125 CORN 6 6 1.000 2 6 0.289 DIRT 6 7 1.000 5 0 0.063 FILBERT 8 1 0.039 5 5 1.000 GRAPE 12 31 0.006 11 13 0.839 GRASS 1 6 0.125 1 3 0.625 HOP 1 5 0.219 0 1 n/a MISCELLANEOUS 1 5 0.219 1 0 n/a BERRY NURSERY 1 5 0.219 3 1 0.625 PASTURE 0 4 0.125 6 0 0.031 RIPARIAN 3 6 0.508 1 6 0.125 URBAN 17 12 0.458 4 5 1.000 VEGETABLE 0 0 n/a 0 0 n/a TOTAL 57 101 0.001 41 51 0.348 * p-values from a non-parametric binomial test 41

Table 6. Effects of vegetation on pest and predatory mite distributions within vineyards combining data from June to September in 1998 and 1999.

T.pyri + T. urticae A. andersoni

VEGETATION Center Center p-value Center Center p-value density density * density density * higher lower higher lower CANEBERRY 7 22 0.009 6 2 0.289 CHERRY 3 18 0.001 2 5 0.453 CONIFER 3 2 1.000 1 5 0.219 CORN 7 9 0.804 2 7 0.180 DIRT 10 19 0.137 6 0 0.031 FILBERT 11 2 0.022 7 8 1.000 GRAPE 42 95 <.001 34 21 0.106 GRASS 24 38 0.099 14 8 0.286 HOP 1 9 0.021 0 2 0.500 MISCBERRY 3 7 0.344 1 2 1.000 NURSERY 3 11 0.057 3 1 0.625 PASTURE 0 11 0.001 8 1 0.039 RIPARIAN 17 26 0.222 7 7 1.000 URBAN 33 26 0.435 8 7 1.000 VEGETABLE 5 0 0.063 0 0 nla TOTAL 169 295 <.001 99 76 0.096 * p-values from a non-parametric binomial test

by a non-parametric binomial test. Results from 1998 and 1999 are presented in Table 5; results from the combination of 1998 and 1999 are in Table 6.

Overall, T. pyri + A. andersoni had significantly higher densities on vineyard

edges, while T. urticae was more common in vineyard centers. Caneberry, cherry and

grape habitats appeared to be sources of predator immigration into vineyards, as evidenced by significantly higher densities on edges compared to vineyard centers.

Vineyards surrounded by filbert had significantly higher phytoseiid populations in the

center in 1999. This phenomenon may have been caused by chemical drift from adjacent

orchards, or possible competition by Kampimodromus aberrans, another phytoseiid mite.

K. aberrans can outcompete T. pyri on grape in Europe, although the former mite is 42 found primarily on filbert (Corylus avellanae) in the United States (Duso 1989,

McMurtry and Croft 1997). Sampling of adjacent vegetation in 1999 revealed high

population densities of K. aberrans on filbert, whereas no T. pyri were present.

Low profile vegetation, such as dirt and pasture, significantly affected both

phytoseiid and T. unicae density, which may reflect more of a lack of physical barriers

that impede movement rather than a tendency for these types of borders to have higher

predatory mite populations.

No vegetation type appeared to consistently serve as a short-range or nearby

immigration source for spider mites, although low sample sizes in key crops (i.e. hop,

com and nursery) prevented a complete statistical analysis. This lack of effect may

reflect that long-range spider mite dispersal via wind currents results in more of a random

fallout and spatial distribution into plots rather than any localized concentration of spider

mites in proximity to a high-density plant source (Pratt 1999).

Generally, Tables 5 and 6 suggest that phytoseiid mites were emigrating from

certain vegetation types into vineyards and providing biological control of T. urticae.

The significantly higher densities of spider mites in vineyard centers supports Dunley and

Croft's (1990) hypothesis that T. pyri does not have a high dispersal rate, or at least not

high enough to immigrate into and throughout vineyards within a single season and give

rapid control of spider mites.

4.3.2.2 Quantitative Sampling of Adjacent Vegetation

Mites in vegetation surrounding vineyards were quantitatively sampled in 1999.

Adjacent vegetation was classified as either agricultural or riparian in nature, based on 43

o T. pyri + A. andersoni

.T. urticae

(a)

o T. pyri + A. anderson;

• T. urticae

Figure 7. Proportion of pest and predatory mites in 1999 in (a) agricultural crops; and (b) natural vegetation. 44 the standards that were previously established. Proportions of pest and predatory mites in

agricultural crops compared to riparian vegetation are shown in Fig. 7. T. urticae comprised 92% of resident mite populations in agricultural areas, while T. pyri + A.

andersoni were only 8%. In riparian areas the situation was almost reversed, with T.

urticae being 28%, while T. pyri + A. andersoni were 72% of total mite counts.

4.4 Multiple Linear Regression

A computer-based genetic algorithm and information-based model selection

criteria AIC (Akaike's information criterion) were used to identify and pre-select factors

important to the regression models of predatory and pest mite densities. There were 21

pre-selected explanatory variables in the model with T. pyri + A. andersoni density as a

dependent variable, including: caneberry, cherry, conifer, corn, dirt, filbert, grape, grass,

hop, riparian, urban, vegetable, agricultural or riparian site, 'Chardonnay', 'Pinot Oris' ,

'Pinot Noir', 'Riesling', Hood River, Rogue River, A. agilis density and T. urticae

density. There were 12 pre-selected explanatory variables in the model with T. urticae

density as a dependent variable, including: corn, dirt, filbert, hop, miscellaneous berry,

pasture, riparian, 'Pinot Noir', 'Riesling', Hood River, Umpqua Valley and T. pyri + A.

andersoni density. Models entered in SAS® (SAS Institute 1992) gave results from

running multiple linear regression using the pre-selected explanatory variables (Table 7).

Beta coefficients are interpreted as effects on mite densities when all other

explanatory variables are held constant. Explanatory variables with non-significant p­

values, such as dirt, are important for developing regression models, but do not

substantially influence the response variable, mite density. 45

Table 7. Linear regression model results developed using genetic algorithms and the Ale criterion to compute a subset of explanatory variables.

MODEL DEPENDENT VARIABLE T. pyri + A. andersoni T.urticae ModelAIC -72.22 89.45 Adjusted R2 0.8029 0.4534 Number 158 158 Degrees of 21 12 Freedom MODEL Beta 95 % C. I.

Vegetables -0.1686 -.2488, -.0798 0.0004 - - - Agricultural or 0.1495 .0539, .2537 0.0019 - - - Riparian Site 'Chardonnay' 0.2022 .0439, .3845 0.0110 - - - 'Pi not Oris' -0.1855 -.2814, -.0768 0.0015 - - - 'Pi not Noir' -0.2103 -.2629, -.1540 0.0001 0.1537 .0452, .2734 0.0048 'Riesling' 0.4224 .2601, .6055 0.0001 -0.2218 -.3524, -.0650 0.0078 Hood River -0.3224 -.4076, -.2250 0.0001 0.7628 .4488, 1.1450 0.0001 Umpqua Valley - - - 0.2491 .0137, .5393 0.0370 R~ueRiver -0.4874 -.5744, -.3826 0.0001 - - - (Log + 1) Anystis 8.2062 1.0439, 0.0041 - - - agilis Density 40.4672

q> =Beta Coefficients and 95% Confidence Intervals are backtransformed 46

There were three pairs of explanatory variables with significant correlation coefficients for each model. Caneberry and riparian vegetation had a positive correlation coefficient of 0.6400. Com and nursery had a correlation coefficient of 0.6769, and the northern and southern Willamette regions had a correlation coefficient of -0.7359. Since the variance of each pair of explanatory variables is related, often only one factor is shown as influential after model selection, although both factors may have important effects on a dependent variable (Sokal and Rohlf 1981). Results should be interpreted cautiously because of the correlations between variables and the fact that variable interactions were not considered. There would be 230 or 1,073,741,824 explanatory

variables if all possible binary interactions were considered in model selection, which is

prohibitively large.

4.4.1 Model for Typhlodromus pyri + Amblyseius andersoni Density

Several vegetation types had positive impacts on T. pyri + A. andersoni densities,

which may be because that vegetation serves as an alternate host or pollen source

(caneberry, grape, riparian) or a spider mite source (com, hop) (Table 7). Since com and

nursery were positively correlated, nursery might also have a positive impact on

phytoseiid density by providing prey. The positive beta coefficient associated with urban

areas may reflect intensively managed and sprayed alternate host plants such as shrubs,

gardens, and flowers. Negative impacts of cherry, conifer, filbert, grass and vegetables

could result from toxicity of drifting chemicals. Cherry growers apply insecticides to

approximately 96% of acres, and chemicals toxic to phytoseiids are used in both cherry 47 and filbert orchards, including: azinphos-methyl, chlorpyrifos and endosulfan (Oregon

Agricultural Statistics Service 1998, Rinehold and Jenkins 1998).

'Pinot Gris' and 'Pinot Noir' appeared to negatively influence T. pyri + A. andersoni density, while 'Chardonnay' and 'Riesling' had positive effects. These varietal results confirm similar findings by Kreiter et al. (in press) and could result from the presence of leaf characteristics such as plants having many domatia and trichomes that are known to affect predator abundance (Duso 1992, Walter and O'Dowd 1992a,

1992b).

Drier, and usually warmer, site locations were negatively correlated with predatory mite density. This makes intuitive sense, considering T. pyri prefers humid areas and is a cool adapted mite, whereas, T. urticae is more of a pest in hot, arid regions, and the Hood River and Rogue River regions are considered to be hot and arid (Croft et al. 1990, Pascal 1997b).

Positive associations with A. agilis may reflect that this mite species has similar susceptibility to pesticides as T. pyri, the main phytoseiid mite that provides biocontrol of pest mites (Croft personal communication). Given the complement of explanatory variables, T. urticae density did not explain much variation in phytoseiid density (P =

0.8296, n = 158, df = 21). However, we think that this is largely because most data points for this pest and the predaceous mites were at the extremes (i.e., many spider mites and no predator mites or many predator mites and no spider mites; see further presentation of this point below in section 4.5, also see Fig. 8). 48

4.4.2 Model for Tetranychus urticae Density

Presence of pasture, miscellaneous berry crops and filbert positively influence T. urticae density. For aerially dispersing mites, high profile vegetation would conceivably hinder movement by acting as a physical barrier or enhance dispersal if it was a primary source of the pest mite. Low profile vegetation, such as pasture, would then theoretically enhance dispersal distances by eliminating physical obstacles or limit dispersal by being a poor platform for takeoff if it were a primary source of spider mites. Berry crops are of intermediate size and can support high phytoseiid mite populations (Croft 1990, Croft

1997, Roy et al. 1999). However, if chemicals prohibitive to phytoseiid survival are applied, spider mites can become serious pests on berry crops and they can serve as an excellent platform for aerial dispersal of spider mites (Croft 1990, Croft 1997, Roy et al.

1999). Filbert's positive impact on T. urticae density could be related to it having a suppressive effect on T. pyri + A. andersoni populations in vineyards, possibly from nearby toxic chemical drift.

Negative beta coefficients that were associated with com, hop and riparian vegetation may indicate successful biocontrol in vineyards bordered by these vegetation types. Since the vegetation pairs of nursery and com and riparian and caneberry were positively correlated, nursery and caneberry could also negatively influence T. urticae density. Com, nursery and hop can serve as sources of chemically resistant pest mites, which would provide abundant prey for resident predators (Croft 1997). Conversely, riparian vegetation and caneberry often harbor and can potentially export high densities of predatory phytoseiid mites (Boller et al. 1988, Croft 1997, Tixier et al. 1998). 49

'Pinot Noir' had a positive beta coefficient, perhaps indicating that this was exploitation by spider mites of a variety not favored by predatory phytoseiids. 'Riesling', a variety that positively impacted phytoseiid density, appeared to have a negative impact on T. urticae density.

The Hood River and Umpqua valleys appeared to be highly favorable for T. urticae, perhaps because of their drier climate and warmer temperatures.

Given the complement of explanatory variables, T. pyri + A. andersoni density did not significantly account statistically for much of the variation of T. urticae density (P

= 0.2575, n = 158, df = 12). However, this does not mean that T. pyri + A. andersoni density has no impact on T. urticae density. Rather as noted before, the major reason there was little correlation between these two variables was because there was little or no data that fell in the range of intennediate densities for predators and prey mites (see discussion below in section 4.5 and Fig. 8). It also may reflect the inadequacy of the model to describe the relationship between predator and pest densities within the context of 29 additional variables. Even accounting for model inadequacy, when comparing models, beta coefficients always have opposite signs for each pair of explanatory variables. Therefore, an explanatory variable which has a positive impact on T. pyri +A. andersoni density has a negative impact on T. urticae density, and likewise, an explanatory variable which has a negative impact on T. pyri + A. andersoni density has a positive impact on T. urticae density. Non-random occurrence of beta coefficient signs is unlikely if there was no association between predator and pest density. 50

4.5 Biological Control

Two related types of analyses were used to indicate biological control success in

Oregon vineyards in 1998 and 1999: 1) plots of predator versus spider mites densities on individual sample dates and 2) comparisons of predator:prey ratios at individual sites.

There was an inverse relationship between densities per leaf of T. urticae when plotted against the densities per leaf of T. pyri + A. andersoni at individual vineyard sites for 1998 (Fig. 8). High levels of T. urticae were mostly found when predator densities were very low, and low levels of T. urticae occurred when predator densities were moderate or high. This would seemingly indicate that the major biological control agent in Oregon vineyards is T. pyri and that only modest densities of this natural enemy can keep resident or immigrant populations of spider mites from increasing in vineyards.

Predator-prey ratios were also calculated for each site each year and biocontrol success status assigned using the following classification: predator-prey ratios> 10 predatory mites to 1 pest mite equals excellent control; ratios of ~ 0 < 10 predatory mites to < 10 pest mites equals good biocontrol; and ratios of ~ 1 predatory mite to of ~ 10 pest mites equals poor biocontrol (Table 8; shaded boxes indicate vineyards with poor biocontrol).

According to this above classification, a majority of sites had good or excellent biocontrol. Information regarding sites with poor biological control was investigated to find possible explanations. Although owners of site 14 (Table 8) did not divulge complete pesticide spray information, organophosphate chemicals had been applied within the past year. Sites 21 and 22 were in an area with southwesterly winds during the growing season and were northeast of a large commercial nursery. High spider mite 51

5 4.5 1998 - >­ 6. • :!:: 4 - 0c t:. 1999 cCI) 3.5 Cb ca 3 ~ ,u € 2.5 :l ...: 2 ~~ CI) g» 1.5 - ~ 1 « ... 1:1 0.5 ~""A ... A ...... A o , ...... ""U" o 0.5 1 1.5 2 2.5 3 3.5 4 4.5 Average T. pyri + A. andersoni Density

Figure 8. Comparison of T. pyri + A. andersoni and T. urticae densities in 1998 and 1999.

densities in the adjacent vineyard may have resulted from dispersal from the ornamentals, a crop known to harbor T. urticae (Croft 1997, Pratt 1999). No chemicals highly toxic to phytoseiids were reportedly used in these sites. Sites 4 and 28 had poor biocontrol only during one sampling season. There was not an obvious or clear reason why site 4 had poor biocontrol. This site had a peak T. urticae density of 0.4 mites per leaf, and although this site qualified for poor biocontrol by the criteria used in this study, in practice this density is quite low. Site 28 was surrounded by agricultural crops such as com and strawberry, which often harbor and export high numbers of spider mites (Croft

1997). 52

Table 8. Biological control in western Oregon vineyards as determined by predator / prey ratio data, based on conservative estimates of data given for other phytoseiids Y.

Site # Corresponding Peak Prey Approximate Biological Site AGaor I Predatory Mite Density Predator Control Region Ip RlpP Mite Density (T. unicae I Prey Ratio Success site (Phytoseiids I leaf) leaf) 1998 1 0.04 0.00 .0510 G 2 RIP 2 1.62 0.01 160/1 E 2 AG 3 0.73 0.01 7011 E 2 RIP 4 0.13 0.03 411 G 2 AG 5 0.72 0.07 1011 E 2 RIP 6 1.68 0.00 2/0 G 3 RIP 7 0.61 0.00 110 G 5 RIP 8 1.78 0.01 18011 E 2 AG 9 0.02 0.03 2/3 G 2 AG 10 0.30 om 30/1 E 3 AG 11 0.66 0.00 110 G 2 AG 12 0.42 0.02 20/1 E 2 AG 13 0.02 0.00 .02/0 G 2 RIP 14 0.69 0.00 110 G 1 RIP 15 0.10 0.00 .110 G 5 RIP 16 1.41 0.00 1.510 G 4 RIP 17 0.79 0.00 1/0 G 4 RIP 18 0.00 5.43 015.5 P 1 RIP 19 0.03 0.00 .05/0 G 3 RIP 20 0.08 0.00 .110 G 3 RIP 21 0.04 0.47 2125 P 2 AG 22 0.02 10.32 1/500 P 2 AG 23 0.29 0.25 6/5 G 2 AG 24 1.33 0.00 1.510 G 3 AG 25 0.19 0.02 1011 E 2 AG 26 0.20 0.00 .2/0 G 3 AG 27 2.76 0.00 3/0 G 3 RIP 28 0.01 1.68 11170 P 2 AG 29 0.59 0.00 110 G 3 AG 30 0.30 0.02 15/1 E 2 AG 31 0.91 0.00 110 G 2 RIP 32 0.19 0.00 .2/0 G 5 RIP 33 0.04 0.00 .0510 G 5 RIP 34 0.45 0.00 .510 G 2 AG 1999 2 1.45 0.15 1011 E 2 AG 4 0.02 0.37 1120 P 2 AG 8 3.18 0.09 100/3 E 2 AG 10 0.06 0.05 6/5 G 3 AG 21 0.13 0.91 119 G 2 AG 22 0.07 11.99 11150 P 2 AG 23 0.47 2.04 114 G 2 AG 25 0.67 0.02 35/1 E 2 AG 53

Table 8. (Continued)

28 I 0.00 I 0.08 I 0/.1 I G I 2 I AG 30 j 0.87 J 0.23 l 9/2 I G I 2 I AG E =excellent biocontrol, G=good biocontrol, P =poor biocontrol

Cl =Site primarily surrounded by agriculture, ~ =Site primarily surrounded by riparian vegetation. Y =Estimates by: Croft and Nelson 1972, Strong and Croft 1995, Wilson et al. 1984. 54

5. CONCLUSIONS

This study found that Tetranychus urticae was the most abundant spider mite pest of grape in western Oregon vineyards and Typhlodromus pyri the most abundant predatory phytoseiid mite. Few vineyards had high densities of T. urticae, and those that did had few predatory mites, while T. pyri was almost ubiquitous.

T. pyri + A. andersoni populations were not significantly different during sampling periods or between agricultural and riparian site classifications, while T. urticae had significantly higher densities in August or September and in vineyards primarily surrounded by agriculture. T. pyri + A. andersoni's relatively constant densities may result from their ability to overwinter in vineyards and consume a variety of alternate food sources as they become available (Collyer 1981, Duso and Camporese 1991, Engle and Ohnesorge 1994a, Gambaro 1986, Hussey and Huffaker 1976). Although T. urticae can overwinter in vineyards, spider mite density in June and July was extremely low and only significantly increased within vineyards during the late season in agricultural areas when harvesting occurred.

T. pyri + A. andersoni densities were significantly higher on vineyard edges, while T. urticae densities were higher in vineyard centers. Adjacent riparian vegetation harbored high phytoseiid densities, while adjacent agricultural plants harbored high densities of spider mites. Certain surrounding plant types (caneberry, cherry, grape) appeared to be immigration sources for T. pyri + A. andersoni, while no sources were discovered for T. urticae, possibly because low sample sizes prevented complete statistical analysis. An alternative explanation may be that immigrant T. urticae are 55 originating mostly from regional sources that are composed mainly of agricultural crops.

They are taken up into the air and transported over relatively long distances and they later arrive as random dropouts into vineyards. In contrast, the above results suggest that phytoseiid mites are emigrating from certain local or surrounding vegetation types into edges of vineyards. The latter spread and provide excellent biological control of any immigrant T. urticae that arrive in vineyards during the course of the growing season.

Grape variety, growing region and adjacent vegetation appeared to influence both

T. pyri + A. andersoni and T. urticae densities. Plant age did not appear to influence the density of either mite. Due to insufficient data, pesticide effects on mite density were not investigated through statistical analysis, although some chemicals used in western

Oregon vineyards certainly affect spider and predator mite populations to some extent.

However, as indicated by the levels of biological control that are achieved (Table 8) these side-effects apparently are not enough to make biological control unlikely at most sites.

This is in marked contrast to many other areas of grape production in the western United

States where pesticide use is extensive and highly detrimental to biological control agents

(AliNiazee et al. 1974, Schruft 1985). Associations between grape variety and T. pyri +

A. andersoni densities may reflect preference for certain leaf characteristics, while apparent associations between grape variety and T. urticae density may be an indirect effect of phytoseiid predation. Arid, hot growing regions appeared favorable to T. urticae density and less favorable to T. pyri + A. andersoni densities. Several vegetation types positively influenced T. pyri + A. andersoni densities, possibly due to those plants serving as alternate hosts or food sources. Low profile vegetation positively affected T. urticae density, possibly due to lack of physical barriers that hinder aerial dispersal. 56

T. urticae populations greater than 0.25 mites per leaf only occurred when phytoseiid densities were less than 0.5 mites per leaf. Thus it seems T. pyri is capable of regulating this spider mite pest at low densities. However, in both multiple linear regression models predatory phytoseiid density did not explain much variation in T. urticae density and vice-versa. Yet it is common knowledge that biologically, densities of predators and prey are major factors affecting each other's population dynamics (Duso

1989, Duso 1992, Duso et al. 1991, Strong and Croft 1995). From plots of T. pyri + A. andersoni density against T. urticae density (Fig. 8) linear correlations between the two are weak except at density extremes. Few sites had spider mite populations, and these sites had little or no predatory mites. Thus the non-significance of T. pyri + A. andersoni and T. urticae densities in the mUltiple linear regression models, accounting for all other explanatory variables, stems from the model's failure to adequately describe a relationship between these two variables.

Sites with poor biocontrol often had low predatory mite densities, possibly due to toxic pesticides or lack of adequate food and / or habitat throughout the season.

However, the majority of sampled sites had good or excellent biocontrol, thus confirming the hypothesis that predaceous mites and insects largely control spider mites in western

Oregon vineyards. 57

BIBLIOGRAPHY

Ahmedullah, M. 1980. Grape growing and cultivar review of the Pacific Northwest. Fruit Varieties 10urnal. 34(3): 61-63.

Agrawal, A. A. and R. Karban. 1997. Domatia mediate plant-arthropod mutualism. Nature. 387: 562-563.

AliNiazee, M. T. and 1. E. Cranham. 1980. Effect of four synthetic pyrethroids on a predatory mite, Typhlodromus pyri and its prey, Panonychus ulmi on apples in southeast England. Environmental Entomology. 9: 436-439.

AliNiazee, M. T., E. M. Stafford and H. Kido. 1974. Management of grape pests in central California vineyards: toxicity of some commonly used chemicals to Tetranychus pacificus and its predator Metaseiulus occidentalis. Journal of Economic Entomology. 67(4): 543-547.

Alston, D. G. 1994. Effect of apple orchard floor vegetation on density and dispersal of phytophagous and predaceous mites in Utah. Agriculture, Ecosystems and Environment. 50: 73-84.

Anderson, D. R., K. P. Burnham and G. C. White. 1994. AIC model selection in overdispersed capture-recapture data. Ecology. 75(6): 1780-1793.

Baker, E. W. and D. M. Tuttle. 1994. A guide to the spider mites (Tetranychidae) of the United States. Indira Publishing House, West Bloomsfield, Michigan.

Barnes, M. M. 1970. Calepitrimerus vitis (Acarina: Eriophyidae) on grape leaves. Annals of the Entomological Society of America. 63(4): 1193-1194.

Barnes, M. M. 1992. Grape erineum mites, pp. 262-264. In D. L. Flaherty, L. P. Christensen, W. T. Lanini, J. J. Marois, P. A. Phillips and L. T. Wilson [eds.], Grape pest management, 2nd ed. The Regents of the University of California Division of Agriculture, Oakland, California.

Bellotti, A. C. 1985. Cassava, pp. 333-338. In W. Helle and M. W. Sabelis [eds.], Spider mites, their biology, natural enemies and control, Vol. IB. Elsevier, Amsterdam, Netherlands.

Boller, E. F., U. Remund and M. P. Candolfi. 1988. Hedges as potential sources of Typhlodromus pyri, the most important predatory mite in vineyards of northern Switzerland. Entomophaga. 33(2): 249-255.

Bower, C. C. 1980. An evaluation of the potential of diflubenzuron for integrated pest control in apples. General and Applied Entomology. 12: 41-48. 58

Bozdogan, H. 1987. Model selection and Akaike's infonnation criterion (AIC): the general theory and its analytical extensions. Psychometrika. 52(3): 345-370.

Caccia, R., M. Baillod, E. Guignard and S. Kreiter. 1985. [Introduction of an azinphos­ resistant strain of Amblyseius andersoni Chant (Acari, Phytoseiidae) for the control of phytophagous mites in vine culture.] Revue Suisse de Viticulture, d'Arboriculture et d'Horticulture. 17(5): 285-290.

Calvert, D. J. and C. B. Huffaker. 1974. Predator [Metaseiulus occidentalis] - prey [Pronematus spp.] interactions under sulfur and cattail pollen applications in a noncommercial vineyard. Entomophaga. 19(3): 361-369.

Camporese, P. and C. Duso. 1996. Different colonization patterns of phytophagous and predatory mites (Acari: Tetranychidae, Phytoseiidae) on three grape varieties: a case study. Experimental & Applied Acarology. 20: 1-22.

Chant, D. A. 1958. Immature and adult stages of some British Phytoseiidae Berl., 1916 [Acarina]. Journal of the Linnean Society of London, Zoology. 43(294): 599­ 643.

Chant, D. A. 1959. Phytoseiid mites (Acarina: Phytoseiidae). Part I. Bionomics of seven species in southeastern England. Part II. A taxonomic review of the family Phytoseiidae, with descriptions of 38 new species. The Canadian Entomologist. 91(supplement 12): 2-166.

Chant, D. A. 1985. Systematics and morphology, pp. 3-34. In W. Helle and M. W. Sabelis [eds.], Spider mites, their biology, natural enemies and control, Vol. IB. Elsevier, Amsterdam, Netherlands.

Coli, W. M., R. A. Ciurlino and T. Hosmer. 1994. Effect of understory and border vegetation composition on phytophagous and predatory mites in Massachusetts commercial apple orchards. Agriculture, Ecosystems and Environment. 50: 49­ 60.

Collyer, E. 1952. Biology of some predatory insects and mites associated with the fruit tree red spider mite (Metatetranychus ulmi (Koch» in south-eastern England. I. The biology of Blepharidopterus angulatus (Fall.) (Hemiptera-Heteroptera, Miridae). Journal of Horticultural Science. 27: 117-129.

Collyer, E. 1953. Biology of some predatory insects and mites associated with the fruit tree red spider mite (Metatetranychus ulmi (Koch» in south-eastern England. II. Some important predators of the mite. Journal of Horticultural Science. 28: 85­ 97. 59

Collyer, E. 1964. Phytophagous mites and their predators in New Zealand orchards. New Zealand Journal of Agricultural Research. 7: 551-568.

Collyer, E. 1981. Typhlodromid mite, Typhlodromus pyri Scheuten, life cycle. Department of Scientific and Industrial Research, Wellington, New Zealand.

Corino, L. 1985. [The species of phytoseiids (Acarina: Phytoseiidae) in Piedmont vineyards.] Vignevini. 12(6): 53-58.

Croft, B. A. 1982. Arthropod resistance to insecticides: a key to pest control failures and successes in North American apple orchards. Entomologia experimentalis et applicata. 31: 88-110.

Croft, B. A. 1990. Arthropod biological control agents and pesticides. John Wiley & Sons, Inc. New York, New York.

Croft, B. A. 1997. Management of spider mites for biological control and pesticide resistance: a case of landscape level IPM for diverse agriculture in the Pacific Northwest (PNW) United States, pp. 27-66. In T. Saito and B. A. Croft [eds.], Implementing spider mite resistance management programs. Japan Crop Protection Association, Tokyo, Japan.

Croft, B. A. 2000. Personal communication, 1-5-2000. Entomology Department, 2046 Cordley Hall, Oregon State University, Corvallis, Oregon, 97331, U. S. A., 1-541­ 737-5498.

Croft, B. A. and A. W. A. Brown. 1975. Responses of arthropod natural enemies to insecticides. Annual Review of Entomology. 20: 285-335.

Croft, B. A. and M. B. Croft. 1993. Larval survival and feeding by immature Metaseiulus occidentalis, Neoseiulus jallacis, Amblyseius andersoni and Typhlodromus pyri on life stage groups of Tetranychus urticae Koch and phytoseiid larvae. Experimental & Applied Acarology. 17: 685-693.

Croft, B. A. and M. B. Croft. 1996. Intra- and interspecific predation among adult female phytoseiid mites (Acari: Phytoseiidae): effect on survival and reproduction. Environmental Entomology. 25(4): 853-858.

Croft, B. A. and J. E. Dunley. 1993. Habitat patterns and pesticide resistance, pp. 145­ 162. In K. C. Kim and B. A. McPheron [eds.], Evolution of insect pests / patterns of variation. John Wiley & Sons, Inc., New York, New York.

Croft, B. A., S. S. Kim and D. I. Kim. 1996. Intra- and interspecific predation on four life stage groups by the adult females of Metaseiulus occidentalis, Typhlodromus pyri, Neoseiulus jallacis and Amblyseius andersoni. Experimental & Applied Acarology. 20: 435-444. 60

Croft, B. A. and D. L. McGroarty. 1977. The role of Amblyseiusjallacis (Acarina: Phytoseiidae) in Michigan apple orchards, pp. 2-22. In Research Report 333. Michigan State University Agricultural Experiment Station, East Lansing, Michigan.

Croft, B. A., J. A. McMurtry and H. -K. Luh. 1998. Do literature records of predation reflect food specialization and predation types among phytoseiid mites (Acari: Phytoseiidae)? Experimental & Applied Acarology. 22: 467-480.

Croft, B. A., R. H. Messing, J. E. Dunley and W. B. Strong. 1993. Effects of humidity on eggs and immatures of Neoseiulus jallacis, Amblyseius andersoni, Metaseiulus occidentalis and Typhlodromus pyri (Phytoseiidae): implications for biological control on apple, caneberry, strawberry and hop. Experimental & Applied Acarology. 17: 451-459.

Croft, B. A. and J. G. Morse. 1979. Research advances on pesticide resistance in natural enemies. Entomophaga. 24(1): 3-11.

Croft, B. A. and E. E. Nelson. 1972. An index to predict efficient interactions of Typhlodromus occidentalis in control of Tetranychus McDanieli in southern California apple trees. Journal of Economic Entomology. 65(1): 310-312.

Croft, B. A., P. Shearer, G. J. Fields, and H. W. Riedl. 1990. Distribution of Metaseiulus occidentalis (Nesbitt) and Typhlodromus pyri Scheuten (Parasitiformes: Phytoseiidae) in apple orchards of the Hood River Valley, Oregon. Canadian Entomologist. 122: 5-14.

Daane, K. M. and G. Y. Yokota. 1997. Release strategies affect survival and distribution of green lacewings (Neuroptera: Chrysopidae) in augmentation programs. Environmental Entomology. 26(2): 455-464.

Daly, H. V., J. T. Doyen and A.H. Purcell III. 1998. Introduction to insect biology and diversity, 2nd ed. Oxford University Press, New York, New York.

De Blij, H. J. 1981. Geography of viticulture. Miami Geographical Society, Miami, Florida.

Dennis, P. and G. L. A. Fry. 1992. Field margins: can they enhance natural enemy population densities and general arthropod diversity on farmland? Agriculture, Ecosystems and Environment. 40: 95-115.

Downing, R. S. and T. K. Moilliet. 1972. Replacement of Typhlodromus occidentalis by T. caudiglans and T. pyri (Acarina: Phytoseiidae) after cessation of sprays on apple trees. The Canadian Entomologist. 104: 937-940. 61

Dunley, J. E. and B. A. Croft. 1990. Dispersal between and colonization of apple by Metaseiulus occidentalis and Typhlodromus pyri (Acarina: Phytoseiidae). Experimental & Applied Acarology. 10: 137-149.

Duso, C. 1989. Role of the predatory mites Amblyseius aberrans (Dud.), Typhlodromus pyri Scheuten and Amblyseius andersoni (Chant) (Acari, Phytoseiidae) in vineyards. I. The effects of single or mixed phytoseiid population releases on spider mite densities (Acari, Tetranychidae). Journal of Applied Entomology. 107: 474-492.

Duso, C. 1992. Role of the predatory mites Amblyseius aberrans (Dud.), Typhlodromus pyri Scheuten and Amblyseius andersoni (Chant) (Acari, Phytoseiidae) in vineyards. III. Influence of variety characteristics on the success ofA. aberrans and T. pyri releases. Journal of Applied Entomology. 114: 455-462.

Duso, C. and P. Camporese. 1991. Developmental times and oviposition rates of predatory mites Typhlodromus pyri and Amblyseius andersoni (Acari: Phytoseiidae) reared on different foods. Experimental & Applied Acarology. 13: 117-128.

Duso, C. and C. Pasqualetto. 1993. Factors affecting the potential of phytoseiid mites (Acari: Phytoseiidae) as biocontrol agents in North-Italian vineyards. Experimental & Applied Acarology. 17: 241-258.

Duso, C., C. Pasqualetto and P. Camporese. 1991. Role of the predatory mites Amblyseius aberrans (Oud.), Typhlodromus pyri Scheuten and Amblyseius andersoni (Chant) (Acari, Phytoseiidae) in vineyards. II. Minimum releases ofA. aberrans and T. pyri to control spider mite populations (Acari, Tetranychidae). Journal of Applied Entomology. 112: 298-308.

Easterbrook, M. A., M. G. Solomon, J. E. Cranham and E. F. Souter. 1985. Trials of an integrated pest management programme based on selective pesticides in English apple orchards. Crop Protection. 4(2): 215-230.

Engel, V. R. and B. Ohnesorge. 1994a. [The role of alternative food and microclimate in the system Typhlodromus pyri Scheuten (Acari, Phytoseiidae) - Panonychus ulmi Koch (Acari, Tetranychidae) on grape vines. I. Laboratory investigations.] Journal of Applied Entomology. 118: 129-150.

Engel, V. R. and B. Ohnesorge. 1994b. [The role of alternative food and microclimate in the system Typhlodromus pyri Scheuten (Acari, Phytoseiidae) - Panonychus ulmi Koch (Acari, Tetranychidae) on grape vines. II. Field experiments.] Journal of Applied Entomology. 118: 224-238. 62

Englert, W. D. and J. Kettner. 1983. [Side-effects of plant protection materials on spider mites and predacious mites.] Mitteilungen der Deutschen Gesellschaft flir Allgemeine und Angewandte Entomologie. 4(1/3): 89-91.

English-Loeb, G., R. Karban and M. A. Walker. 1998. Genotypic variation in constitutive and induced resistance in grapes against spider mite (Acari: Tetranychidae) herbivores. Environmental Entomology. 27(2): 297-304.

Flaherty, D. L. 1992. False spidermite, pp. 265. In D. L. Flaherty, L. P. Christensen, W. T. Lanini, J. J. Marois, P. A. Phillips and L. T. Wilson [eds.], Grape pest management, 2nd ed. The Regents of the University of California Division of Agriculture, Oakland, California.

Flaherty, D. L. and M. A. Hoy. 1971. Biological control of Pacific mites and Willamette mites in San Joaquin Valley vineyards: part III. Role of tydeid mites. Research and Population Ecology. 13: 80-96.

Flaherty, D., C. Lynn, F. Jensen and M. Hoy. 1972. Correcting imbalances - spider mite populations in southern San Joaquin vineyards. California Agriculture. 26(4): 10-12.

Flaherty, D. L., L. T. Wilson, V. W. Stem and H. Kido. 1985. Biological control in San Joaquin Valley vineyards, pp. 501-520. In M. A. Hoy and D. C. Herzog [eds.], Biological control in agricultural IPM systems. Academic Press, Inc., Orlando, Florida.

Flaherty, D. L., L. T. Wilson, S. C. Welter, C. D. Lynn and R. Hanna. 1992. Spider mites, pp. 180-192. In D. L. Flaherty, L. P. Christensen, W. T. Lanini, J. J. Marois, P. A. Phillips and L. T. Wilson [eds.], Grape pest management, 2nd ed. The Regents of the University of California Division of Agriculture, Oakland, California.

Flechtmann, C. H. W. and J. A. McMurtry. 1992. Studies on how phytoseiid mites feed on spider mites and pollen. International Journal of Acarology. 18(3): 157-162.

Flexner, J. L., P. H. Westigard, P. Gonzalves and R. Hilton. 1991. The effect of groundcover and herbicide treatment on twospotted spider mite density and dispersal in southern Oregon pear orchards. Entomologia experimentalis et applicata. 60: 111-123.

Forrest, S. 1993. Genetic algorithms: Principles of natural selection applied to computation. Science. 261: 872-878.

Fulmek, L. 1930. Zur kenntnis der entwicklungsstadien von Triphleps minuta L. (Anthocoridae, Hemiptera-Heteroptera). Zeitschrift flir wissenschaftliche insektenbiologie. 25: 82-88. 63

Gambaro, P. I. 1972. [The role of Typhlodromus aberrans (Oudm.) in the biological control of phytophagous mites in Verona province vineyards.] Estratto dal Bollettino di Zoologia agraria e di Bachicoltura, Serie II. 11: 151-165.

Gambaro, P. I. 1986. An ecological study of Amblyseius andersoni Chant (Acarina: Phytoseiidae) in the climate of the Po Valley (North Italy). Redia. 69: 555-572.

Gambaro, P. I. 1987. The biological control of phytophagous mites on vines: role of Kampimodromus aberrans (Oud.) (Acarina, Phytoseiidae), pp. 235-243. In R Cavalloro [ed.], Integrated pest control in viticulture. A. A. Balkema, Rotterdam, Netherlands.

Garren, R, I. MacSwan, I. And G. Fisher. 1983. Problems of disease, disorders and insects affecting grape production, pp. 83-94. In M. Webb [ed.], Oregon wine grape grower's guide. Oregon Winegrower's Association, Oregon.

Gerson, U. 1985. Webbing, pp. 223-232. In W. Helle and M. W. Sabelis [eds.], Spider mites, their biology, natural enemies and control, Vol. lA. Elsevier, Amsterdam, Netherlands.

Girolami, V. 1987. Mites of vineyards and control strategies, pp. 185-197. In R Cavalloro [ed.] , Integrated pest control in viticulture. A. A. Balkema, Rotterdam, Netherlands.

Glantz, S. A. 1997. Primer of biostatistics, 4th ed. McGraw-Hill, New York, New York.

Grostal, P. and D. J. O'Dowd. 1994. Plants, mites and mutualism: leaf domatia and the abundance and reproduction of mites on Viburnum tinus (Caprifoliaceae). Oecologia. 97: 308-315.

Hadam, J. J., M. T. Aliniazee and B. A. Croft. 1986. Phytoseiid mites (Parasitiformes: Phytoseiidae) of major crops in Willamette Valley, Oregon, and pesticide resistance in Typhlodromus pyri Scheuten. Environmental Entomology. 15: 1255-1263.

Hanna, R, L. T. Wilson, F. G. Zalom, D. L. Flaherty and G. M. Leavitt. 1996. Spatial and temporal dynamics of spider mites (Acari: Tetranychidae) in 'Thompson Seedless'vineyards. Environmental Entomology. 25(2): 370-382.

Hardman, J. M. and M. L. Rogers. 1991. Effects of temperature and prey density on survival, development, and feeding rates of immature Typhlodromus pyri (Acari: Phytoseiidae). Environmental Entomology. 20(4): 1089-1096. 64

Hassan, S. A, F. Bigler, P. Blaisinger, H. BogenschUtz, J. Brun, P. Chiverton, E. Dickler, M. A. Easterbrook, P. J. Edwards, W. D. Englert, S. I. Firth, P. Huang, C. Inglesfield, F. Klingauf, C. KUhner, M. S. Ledieu, E. Naton, P. A. Oomen, W. P. J. Overmeer, P. Plevoets, J. N. Reboulet, W. Rieckmann, L. Sams0e-Petersen, S. W. Shires, A. Staubli, J. Stevenson, J. J. Tuset, G. Vanwetswinkel, and A. Q. van Zon. 1985. Standard methods to test the side-effects of pesticides on natural enemies of insects and mites. Bulletin OEPPIEPPO Bulletin. 15: 214-255.

Helle, W. and L. P. Pijnacker. 1985. Parthenogenesis, chromosomes and sex, pp. 129­ 140. In W. Helle and M. W. Sabelis [eds.], Spider mites, their biology, natural enemies and control, Vol. lA. Elsevier, Amsterdam, Netherlands.

Hellman, E. W. 1987. Cultivar sensitivity of grapevines to 2,4-D. Fruit Varieties Journal. 41(3): 111-114.

Holland, J. H. 1992. Genetic algorithms. Scientific American. 267(1): 66-72.

Huffaker, C. B., M. Van de Vrie and J. A. McMurtry. 1970. II. Tetranychid populations and their possible control by predators: an evaluation. Hilgardia. 40: 391-458.

Hughes, T. E. 1959. Mites, or the Acari. Essential Books, Fair Lawn, New Jersey.

Hull, L. A. and E. H. Beers. 1985. Ecological selectivity: modifying chemical control practices to preserve natural enemies, pp. 103-122. In M. A. Hoy and D. C. Herzog [eds.], Biological control in agricultural IPM systems. Academic Press, Inc., Orlando, Florida.

Hussey, N. W. and C. B. Huffaker. 1976. Spider mites, pp. 179- 228. In V. L. Delucchi [ed.], Studies in biological control. Cambridge University Press, Cambridge, Massachusetts.

Jackson, D. and D. Schuster. 1987. The production of grapes & wine in cool climates. Butterworths of New Zealand (Ltd.), Wellington, New Zealand.

James, D. G. and J. Whitney. 1993. Mite populations on grapevines in south-eastern Australia: Implications for biological control of grapevine mites (Acarina: Tenuipalpidae, Eriophyidae). Experimental & Applied Acarology. 17: 259-270.

Jeppson, L. R., H. H. Keifer and E. W. Baker. 1975. Mite injurous to economic plants. University of California Press, Berkeley, California.

Jepson, P. C. and P. A. Heneghan. 2000. Interpreting median toxicity ratings and comparing median toxicity ratings. Unpublished. Dept. of Entomology, Oregon State University, Corvallis, Oregon, 97331, U. S. A., 1-541-737-4733. 65

Johnson, D. T. and B. A. Croft. 1976. Laboratory study of the dispersal behavior of Amblyseius fallacis (Acarina: Phytoseiidae) [Predator of plant feeding mites]. Annals of the Entomological Society of America. 69(6): 1019-1023.

Kapetanakis, E. G. and J. E. Cranham. 1983. Laboratory evaluation of resistance to pesticides in the phytoseiid predator Typhlodromus pyri from English apple orchards. Annual Applications of Biology. 103: 389-400.

Karban, R, G. English-Loeb, M. A. Walker and J. Thaler. 1995. Abundance of phytoseiid mites on Vitis species: effects of leaf hairs, domatia, prey abundance and plant phylogeny. Experimental & Applied Acarology. 19: 189-197.

Kennedy, G. G. and D. R. Smitley. 1985. Dispersal, pp. 233-242. In W. Helle and M. W. Sabelis [eds.], Spider mites, their biology, natural enemies and control, Vol. 1A. Elsevier, Amsterdam, Netherlands.

Kinn, D. N. and R. L. Doutt. 1972. Initial survey of arthropods found in North Coast vineyards of California. Environmental Entomology. 1(4): 508-513.

Kogan, M. 1998. Integrated pest management: historical perspectives and contemporary developments. Annual Review of Entomology. 43: 243-270.

Kogan, M., B. A. Croft and R. F. Sutherst. 1998. Applications of ecology for integrated pest management, pp. 681-736. In C. B. Huffaker and A. P. Gutierrez [eds.], Ecological entomology, 2nd ed. John Wiley & Sons, Inc., New York, New York.

Krantz, G. W. 1971. A manual of acarology. Oregon State University Book Stores, Inc., Corvallis, Oregon.

Kreiter, S., M.-S. Tixier, P. Auger, N. Muckensturrn, G. Sentenac, B. Doublet and M. Weber. In Press. Phytoseiid mites of vineyards in France (Acari: Phytoseiidae). Acarologia.

Lester, P. J., H. M. A. Thistlewood and R Harmsen. 1998. The effects of refuge size and number on acarine predator-prey dynamics in a pesticide-disturbed apple orchard. Journal of Applied Ecology. 35: 323-331.

Lett, D. R 1983. Grape variety and site selection with an emphasis on the Willamette Valley, pp. 7-13. In M. Webb [ed.], Oregon winegrape grower's guide. Oregon Winegrower's Association, Oregon.

Lewis, W. J., J. C. van Lenteren, S. C. Phatak and J. H. Tumlinson. 1997. A total system approach to sustainable pest management. Proceedings of the National Academy of Sciences of the United States of America. 94: 12243-12248. 66

Lindquist, E. E. 1985. Anatomy, phylogeny and systematics, pp. 3-28. In W. Helle and M.W. Sabelis [eds.], Spider mites, their biology, natural enemies and control, Vol. 1A. Elsevier, Amsterdam, Netherlands.

Luh, H.-K. 2000. Personal communication, 5-15-2000. Entomology Department, 2046 Cordley Hall, Oregon State University, Corvallis, Oregon, 97331, U. S. A., 1-541­ 737-6491.

Luh, H.-K., and B. A. Croft. 1999. Classification of generalist or specialist life styles of predaceous phytoseiid mites using a computer genetic algorithm, information theory, and life history. Environmental Entomology. 28(6): 915-923.

MacRae, I. V. and B. A. Croft. 1993. Influence of temperature on interspecific predation and cannibalism by Metaseiulus occidentalis and Typhlodromus pyri (Acarina: Phytoseiidae). Environmental Entomology. 22(4): 770-775.

Martinson, T. E. and T. J. Dennehy. 1995. Varietal preferences of Erythroneura leafhoppers (Homoptera: Cicadellidae) feeding on grapes in New York. Environmental Entomology. 24(3): 550-558.

MathWorks. 1999. Using MATLAB. MathWorks, Inc., Natick, Massachusetts.

McMurtry, J. A. and B. A. Croft. 1997. Life-styles of phytoseiid mites and their roles in biological control. Annual Review of Entomology. 42: 291-321.

McMurtry, J. A., C. B. Huffaker and M. Van de Vrie. 1970. I. Tetranychid enemies: their biological characters and the impact of spray practices. Hi19ardia. 40: 331­ 390.

Mitchell, M. 1996. An introduction to genetic algorithms. The MIT Press, Cambridge, Massachusetts.

Nyrop, J. P., J. C. Minns and C. P. Herring. 1994. Influence of ground cover on dynamics of Amblyseiusfallacis Garman (Acarina; Phytoseiidae) in New York apple orchards. Agriculture, Ecosystems and Environment. 50: 61-72.

Oregon Agricultural Statistics Service. 1998. 1997-1998 Oregon agriculture & fisheries statistics. U. S. Department of Agriculture. Portland, Oregon.

Overmeer, W. P. 1. 1985. Diapause, pp. 95-102. In W. Helle and M. W. Sabelis [eds.], Spider mites, their biology, natural enemies and control, Vol. lB. Elsevier, Amsterdam, Netherlands.

Overmeer, W. P. J. and A. Q. van Zon. 1983. Resistance to parathion in the predacious mite Typhlodromus pyri Scheuten (Acarina: Phytoseiidae). Med. Fac. Landouww. Rijksuniv. Gent. 48(2): 247-251. 67

Overmeer, W. P. J. and A. Q. van Zon. 1984. The preference of Amblyseius potentillae (Garman) (Acarina: Phytoseiidae) for certain plant substrates, pp. 591-596. In D. A. Griffiths and C. E. Bowman [eds.], Acarology VI, Vol. 1. Ellis-Horwood, Chichester, England.

Pascal, C. R. 1997a. Soil, climate, and passion: Blazing the Oregon Trail, part 1. American Vineyard. May: 6-9.

Pascal, C. R. 1997b. Soil, climate, and passion: Blazing the Oregon Trail, part 2. American Vineyard. June: 27-31.

Pemberton, R. W. and C. E. Turner. 1989. Occurrence of predatory and fungivorous mites in leaf domatia. American Journal of Botany. 76(1): 105-112.

Perold, A. I. 1927. A treatise on viticulture. MacMillan and Co., Limited, London, England.

Pratt, P. D. 1999. Biological control of spider mites by the predatory mite Neoseiulus fallacis (Acari: Phytoseiidae) in ornamental systems. Dissertation, Dept. Entomology, Oregon State University. pp. 187.

Pritchard, A. E. and E. W. Baker. 1955. A revision of the spider mite family Tetranychidae. The Pacific Coast Entomological Society, San Francisco, California.

Ramsey, F. L. and D. W. Schafer. 1997. The statistical sleuth: A course in methods of data analysis. Duxbury Press, Belmont, California.

Rinehold, J., and J. J. Jenkins. 1998. Oregon pesticide use estimates for small fruits (1995) and tree fruits (1996). EM 8673. Oregon State University Extension Service. Corvallis, Oregon.

Roush, R. T., W. L. Peacock, D. L. Flaherty and M. A. Hoy. 1980. Dimethoate-resistant spider mite predator survives field tests. California Agriculture. 34(5): 12-13.

Rowley, H. K., D. C. Marousek and R. F. Kriesel. 2000. 1999 Oregon vineyard report. Oregon Agricultural Statistics Service, United States Department of Agriculture. Portland, Oregon.

Roy, M., J. Brodeur and C. Cloutier. 1999. Seasonal abundance of spider mites and their predators on red raspberry in Quebec, Canada. Environmental Entomology. 28(4): 735-747.

Sabelis, M. W. 1985a. Reproductive strategies, pp. 265-278. In W. Helle and M. W. Sabelis [eds.], Spider mites, their biology, natural enemies and control, Vol. 1A. Elsevier, Amsterdam, Netherlands. 68

Sabelis, M. W. 1985b. Capacity for population increase, pp. 35-42. In W. Helle and M. W. Sabelis [eds.], Spider mites, their biology, natural enemies and control, Vol. lB. Elsevier, Amsterdam, Netherlands.

Sabelis, M. W. 1985c. Predation on spider mites, pp. 103-130. In W. Helle and M. W. Sabelis [eds.], Spider mites, their biology, natural enemies and control, Vol. lB. Elsevier, Amsterdam, Netherlands.

Sabelis, M. W. and M. Dicke. 1985. Long-range dispersal and searching behavior, pp. 141-160. In W. Helle and M. W. Sabelis [eds.], Spider mites, their biology, natural enemies and control, Vol. lB. Elsevier, Amsterdam, Netherlands.

SAS Institute. 1992. SAS/STAT user's guide. SAS Institute, Inc., Cary, North Carolina.

Schausberger, P. and B. A. Croft. 1999a. Activity, feeding, and development among larvae of specialist and generalist phytoseiid mites (Acari: Phytoseiidae). Environmental Entomology. 28(2): 322-329.

Schausberger, P. and B. A. Croft. 1999b. Predation on and discrimination between con­ and heterospecific eggs among specialist and generalist phytoseiid mites (Acari: Phytoseiidae). Environmental Entomology. 28(3): 523-528.

Schruft, G. A. 1985. Grape, pp. 359-366. In W. Helle and M. W. Sabelis [eds.], Spider mites, their biology, natural enemies and control, Vol. 1B. Elsevier, Amsterdam, Netherlands.

Schruft, G. A. 1987. Biological control of spider mites in viticulture, pp. 227-233. In R. Cavalloro [ed.], Integrated pest control in viticulture. A. A. Balkema, Rotterdam, Netherlands.

Schulten, G. G. M. 1985. Mating, pp. 55-66. In W. Helle and M. W. Sabelis [eds.], Spider mites, their biology, natural enemies and control, Vol. 1B. Elsevier, Amsterdam, Netherlands.

Schuster, R. O. and A. E. Pritchard. 1963. Phytoseiid mites of California. Hi19ardia. 34(7): 191-285.

Seymour, J. 1982. Spray-resistant mites to the rescue. Ecos. 33: 3-7.

Slone, D. H. 1999. Spatial patterns of predaceous and phytophagous apple mites (Acari: Eriophyidae, Phytoseiidae, Stigmaeidae, Tetranychidae): intra- and interspecific relationships. Dissertation, Dept. Entomology, Oregon State University. pp. 130.

Sokal, R. R. and F. J. Rohlf. 1981. Biometry, 2nd ed. W. H. Freeman and Company, San Francisco, California. 69

Speight, M. R. 1983. The potential of ecosystem management for pest control. Agriculture, Ecosystems and Environment. 10: 183-199.

Strong, W. B. and B. A. Croft. 1995. Inoculative release of phytoseiid mites (Acarina: Phytoseiidae) into the rapidly expanding canopy of hops for control of Tetranychus urticae (Acarina: Tetranychidae). Environmental Entomology. 24(2): 446-453.

Theiling, K. M. 1987. The SELCTV database: the susceptibility of arthropod natural enemies of agricultural pests to pesticides. MS thesis, Dept. Entomology, Oregon State University. pp. 170.

Theiling, K. M., and B. A. Croft. 1988. Pesticide side-effects on arthropod natural enemies: a database summary. Agriculture, Ecosystems and Environment. 21: 191-218.

Tixier, M. S., S. Kreiter, P. Auger and M. Weber. 1998. Colonization of Languedoc vineyards by phytoseiid mites (Acari: Phytoseiidae): influence of wind and crop environment. Experimental & Applied Acarology. 22(9): 523-542.

Tuovinen, T. and 1. A. H. Rokx. 1991. Phytoseiid mites (Acari: Phytoseiidae) on apple trees and in surrounding vegetation in southern Finland. Densities and species composition. Experimental & Applied Acarology. 12: 35-46.

Vacante, V. and G. Tropea-Grazia. 1987. Grape mites in Sicily - contribution I, pp. 207-215. In R. Cavalloro [eq.], Integrated pest control in viticulture. A. A. Balkema, Rotterdam, Netherlands.

Veennan, A. 1985. Diapause, pp. 279-316. In W. Helle and M. W. Sabelis [eds.], Spider mites, their biology, natural enemies and control, Vol. lA. Elsevier, Amsterdam, Netherlands.

vonEnde, C. N. 1993. Repeated-measures analysis: growth and other time-dependent measures, pp. 113-137. In S. M. Scheiner and 1. Gurevitch [eds.], Design and analysis of ecological experiments. Chapman & Hall, New York, New York.

Walker, 1. T. S. and D. R. Penman. 1978. Integrated control of apple pests in New Zealand. 11. The influence of field application of azinphos-methyl on predation of European red mite by Typhlodromus pyri. Proceedings of the 31 51 New Zealand Weed and Pest Control Conference. 208-213.

Walter, D. E. 1996. Living on leaves: Mites, tomenta, and leaf domatia. Annual Review of Entomology. 41: 101-114.

Walter, D. E. and H. A. Denmark. 1991. Use of leaf domatia on wild grape (Vitis munsoniana) by arthropods in central Florida. Florida Entomologist. 74(3): 440­ 446. 70

Walter, D. E. and D. J. Q'Dowd. 1992a. Leaf morphology and predators: Effect of leaf domatia on the abundance of predatory mites (Acari: Phytoseiidae). Environmental Entomology. 21(3): 478-484.

Walter, D. E. and D. J. Q'Dowd. 1992b. Leaves with domatia have more mites. Ecology. 73(4): 1514-1518.

Walters, P. J. 1976. Effect of five acaricides on Tetranychus urticae (Koch) and its predators, Stethorus spp. (Coleoptera: Coccinellidae) in an apple orchard. Journal of the Australian Entomological Society. 15: 53-56.

Wearing, C. H. 1975. Problems of pest assessment in horticulture, particularly orchards. The New Zealand Entomologist. 6(1): 17-21.

Welter, S. C., P. S. McNally and D. S. Farnham. 1989. Effect ofWillamette mite (Acari: Tetranychidae) on grape productivity and quality: a reappraisal. Environmental Entomology. 18(3): 408411.

Wilson, L. T., D. L. Flaherty and W. L. Peacock. 1992. Grape leafhopper, pp. 140-152. In D. L. Flaherty, L. P. Christensen, W. T. Lanini, J. J. Marois, P. A. Phillips and L. T. Wilson [eds.], Grape pest management, 2nd ed. The Regents of the University of California Division of Agriculture, Oakland, California.

Wilson, L. T., J. M. Smilanick, M. P. Hoffmann, D. L. Flaherty and S. M. Ruiz. 1988. Leaf nitrogen and position in relation to population parameters of Pacific spider mite, Tetranychus pacificus (Acari: Tetranychidae) on grapes. Environmental Entomology. 17(6): 964-968.

Winkler, A. J. 1962. General viticulture. University of California Press, Berkeley, California.

Wrensch, D. L. 1985. Reproductive parameters, pp. 165-170. In W. Helle and M. W. Sabelis [eds.], Spider mites, their biology, natural enemies and control, Vol. lA. Elsevier, Amsterdam, Netherlands.

Zemek, R. 1993. Characteristics of development and reproduction in Typhlodromus pyri on Tetranychus urticae and Cecidophyopsis ribis. I. Overwintered females. Experimental & Applied Acarology. 17: 405421.

Zhang, Z. -Q. and J. P. Sanderson. 1992. Short-distance location of spider mite colonies by three predatory mites (Acari: Tetranychidae, Phytoseiidae): predator responses to prey- and predator- associated stimuli. Environmental Entomology. 21(4): 799-807. 71

APPENDICES 72

APPENDIX A

Pesticide toxicity ratings for A. andersoni

The SELCTV database (Theiling 1987, Theiling and Croft 1988) is an extensive compilation of pri mari Iy pre-1986 literature detailing pesticide effects on non-target arthropods. IOBC toxicity ratings refer to data calculated from research conducted by the

IOBCIWPRS (International Organization for Biological Control of Noxious Animals and

Plants - West Palaearctic Regional Section) (Hassan et al. 1985). Toxicity ratings for these two systems have different rating systems and are listed under each chart (Hassan et al. 1985, Theiling 1987). Since toxicity ratings are simply categories, averages of toxicity ratings would not have practical meaning and would be difficult to interpret.

Therefore median toxicity ratings were used here in lieu of mean toxicity ratings.

Several factors can influence interpretation of resulting toxicity ratings, including: test methods, response measured, resistance status of test organism and number of organisms tested (Jepson and Heneghan 2000). Test type (field or laboratory) can impact resulting toxicity ratings, due to droplet physics, chemical adsorption onto substrates and degradation processes (Jepson and Heneghan 2000). Different methods for reporting responses to chemical exposure (lethal dose, percent mortality, etc.) are difficult to compare (Theiling and Croft 1988). Organisms resistant to a particular chemical will have different toxicity ratings for that chemical compared to susceptible organisms (Theiling and Croft 1988). Toxicity ratings for chemicals based on a single reference can be misleading, and stronger conclusions can be drawn for chemicals with several records (Jepson and Heneghan 2000). 73

Table A. Median toxicity ratings for A. andersoni based on the SELCTV (L and IOBC II databases.

Active Chemical Class IOBC a SELCTV i3 Test Response Resist- Ingredient Median Median Type Type ance Toxicity Toxicity y 5 Status Rating Rating £

2-( I-naphthyl) synthetic auxin I - lab PH U acetamide 2-(l-naphthyl) synthetic auxin I - lab PH U acetic acid 2,4-D aryloxyalkanoic acid I - lab PH U acephate organophosphate 4 - lab PH U amitraz amidine 4 - lab PH U atrazine 1,3,5-triazine I - lab PH U azinphos-methyl organophosphate - I field M R

azinphos-methyl organophosphate - 4.5 lab LC S azinphos-methyl organophosphate 4 - lab PH U aziJ!Phos-methyl organophosphate - 5 med LC S azocyc1otin organotin - 5 lab M S azocyc1otin organotin 4 - lab PH U Bacillus bacterium I - lab PH U thurinl? iensis benomyl benzimidazole - 4 lab LC S benomyl benzimidazole - 2 lab LC T benzoximate unknown - 3 lab M S benzoximate unknown I - lab PH U bitertanol azole 2 - field PH U bitertanol azole I - lab PH U bromacil uracil I - lab PH U bromofenoxim hydroxybenzonitrile 4 - lab PH U precursor bromophos unknown 4 - lab PH U bromoxynil hydroxybenzonitrile 4 - lab PH U bupirimate pyrimidine 1 - lab PH U captafol N-trihalomethylthio I - lab PH U captan N-trihalomethylthio - 2 lab M T captan N-trihalomethylthio 1 - lab PH U carbaryl carbamate - 4 med LC S carbaryl carbamate 4 - lab PH U carbendazim benzimidazole - 3 lab M S carbendazim benzimidazole 4 - lab PH U chinomethionat unknown 4 - lab PH U chlorfenvinphos organophosphate 4 - lab PH U chlormequat quaternary I - lab PH U ammonium chlorothalonil unknown 3 - lab PH U chlo_rpyrifos organophosphate 4 - lab PH U c10fentezine unknown 2 - field PH U 74 clofentezine unknown 1 - lab PH U copper inorganic 2 - field PH U oxychloride copper inorganic 1 - lab PH U oxychloride cyfluthrin pyrethroid 4 - field PH U cyhexatin organotin - 3 lab LC S cyhexatin organotin 4 - lab PH U cy~ermethrin pyrethroid 4 - lab PH U deltamethrin pyrethroid 4 - lab PH U demeton-S­ organophosphate 4 - lab PH U methyl desmetryn 1,3,S-triazine - 3 lab M S desmetryn 1,3,S-triazine 1 - lab PH U dialifos unknown 4 - lab PH U diazinon organophosphate 4 - lab PH U diazinon organophosphate - 4 med LC S dichlofluanid N-trihalomethylthio 3 - field PH U dichlofluanid N-trihalomethylthio 2.S - lab PH U dicofol organochlorine 4 - lab PH U difenzoquat unknown - S lab M S difenzoquat unknown 4 - lab PH U diflubenzuron benzoylurea 1 - field PH U diflubenzuron benzoylurea 1 - lab PH U dimethoate organophosphate 4 - lab PH U dinocap dinitrophenol - 3 lab LC S derivative dinoseb unknown 4 - lab PH U ditalimfos unknown - 4 lab M S ditalimfos unknown 2 - lab PH U dithianon unknown 2 - field PH U dithianon unknown 1 - lab PH U endosulfan organochlorine - 4 lab LC S endosulfan organochlorine 2 - lab PH U ethiofencarb carbamate 4 - lab PH U ethirimol pyrimidine 1 - lab PH U etrimfos unknown 4 - lab PH U fenarimol pyrimidinyl carbinol 1 - lab PH U fenbutatin oxide organotin - 2 lab LC T fenbutatin oxide organotin 1 - lab PH U fenitrothion organophosphate 4 - lab PH U fenoxycarb carbamate 1 - field PH U feno~ycarb carbamate 2 - lab PH U fenpropathrin pyrethroid 4 - field PH U fenpropimorph morpholine 1 - lab PH U fen valerate pyrethroid - 5 lab M S fen valerate pyrethroid 4 - lab PH U fluazifop-butyl 2-(4-aryloxy 1 - lab PH U phenoxy) propionic acid flubenzimine unknown 4 - lab PH U flutriafol azole 1 - lab PH U 75 fluvalinate unknown 4 - field PH U folpet N-trihalomethylthio 1 - lab PH U glufosinate­ unknown 4 - lab PH U ammonium glyphosate unknown 1 - lab PH U heptenophos organophosphate 2 - field PH U heptenophos organophosphate 4 - lab PH U hexythiazox unknown 1 - field PH U hexythiazox unknown 1 - lab PH U iprodione dicarboximide 1 - field PH U iprodione dicarboximide 1 - lab PH U lindane organochlorine - 4 lab M S lindane organochlorine 2 - lab PH U mancozeb alkylenebis 3.5 - lab PH U (dithiocarbamate) maneb alkylenebis - 3 lab LC S (dithiocarbamate) maneb alkylenebis 4 - lab PH U (dithiocarbamate) methabenzthiazu urea 4 - lab PH U ron methamidophos organophosphate 2 - field PH U methidathion organophosphate - 5 lab M S methidathion organophosphate 4 - lab PH U methomyl oxime carbamate - 5 lab M S methomyl oxime carbamate 4 - lab PH U metiram alkylenebis 4 - lab PH U (dithiocarbamate) metsulfuron­ sulfonylurea 1 - lab PH U methyl mevinphos organophosphate 4 - lab PH U monolinuron urea - 5 lab M S monolinuron urea 4 - lab PH U nuarimol pyrimidinyl carbinol 1 - lab PH U oxamyl oxime carbamate 4 - lab PH U penconazole azole 1 - field PH U penconazole azole 1 - lab PH U permethrin pyrethroid - 5 lab M S permethrin pyrethroid 4 - lab PH U permethrin pyrethroid - 5 med LC S phenmedipham bis-carbamate 4 - lab PH U phosalone organophosphate 4 - lab PH U phosalone organophosphate - 5 med LC S phosphamidon organophosphate 4 - lab PH U pirimicarb carbamate - 4 lab LC S pirimicarb carbamate 3 - lab PH U pirimiphos­ organophosphate - 5 lab M S meth}'l pirirniphos­ organophosphate 4 - lab PH U methyl prochloraz azole 3 - lab PH U procymidone dicarboximide 3 - field PH U propachlor chloroacetanilide - 4 lab M S 76 j:>fopachlor chloroacetanilide 3 - lab PH U propiconazole azole 2 - lab PH U propineb alkylenebis 4 - lab PH U (dithiocarbamate) propoxur carbamate - 5 lab M S propoxur carbamate 4 - lab PH U pyrazophos organophosphate 4 - lab PH U ester simazine 1,3,5-triazine 1 - lab PH U sulfur unknown 3 - field PH U sulfur unknown - 3 lab LC S sulfur unknown - 2 lab LC T sulfur unknown - 4 lab M S sulfur unknown 3 - lab PH U teflubenzuron benzoylurea 1 - field PH U tetradifon unknown 1 - lab PH U thiocyclam 2-dimethylamino 2 - lab PH U propane-l,3-diol thiophanate­ benzimidazole 4 - lab PH U methyl precursor thiram Dimethyldithio 3 - lab PH U carbamate triadimefon azole - 2 lab M T triadimefon azole 1 - lab PH U triazophos organophosphate 4 - lab PH U trichlorfon or~anophosphate - 5 lab M S trichlorfon organophosphate 4 - lab PH U triforine unknown 1 - field PH U vamidothion organophosphate 4 - lab PH U vinclozolin dicarboximide - 2 lab M T vinclozolin dicarboximide 1 - lab PH U zineb alkylenebis - 3 lab LC S (dithiocarbamate) zineb alkylenebis - 2 lab LC T (dithiocarbamate) a = Data calculated from research conducted by the IOBCIWPRS (International Organization for Biological Control of Noxious Animals and Plants - West Palaearctic Regional Section; Hassan et al. 1985): IOBC Toxicity Rating: 1 2 3 4 IOBC Range of effect: Harmless Slightly Moderately Harmful Harmful Harmful

~ = Data calculated from the SELCTV database (Theiling 1987, Theiling and Croft 1988):

SELCTV Toxicity Rating: 1 2 3 4 5 SELCTV Range of effect: 0% <10% 10-30% 30-90% >90%

Y • field = field tests, lab = laboratory tests, med = median lethal assay & probit I) : LC = lethal concentration, M = mortality, PH = percent harm £ : R = resistant, S = susceptible, T = tolerant, U= unknown 77

APPENDIXB

Pesticide toxicity ratings for T. pyri

Table B. Median toxicity ratings for T. pyri based on the SELCTV a and IOBC f3 databases.

Active Chemical Class IOBC u SELCTV~ Test Response Resist­ Ingredient Median Median Type Type ance Toxicity Toxicity 'Y I) Status Rating Rating £

2-(l-naphthyl) synthetic auxin - 2 field M S acetamide 2-( I-naphthyl) synthetic auxin 1 - lab PH U acetamide 2-( I-naphthyl) synthetic auxin I - lab PH U acetic acid 2,4-D aryloxyalkanoic acid I - lab PH U acephate organophosphate 4 - lab PH U amitraz amidine - 4 field M S amitraz amidine 4 - field PH U amitraz amidine - 4 lab M R amitraz amidine 4 - lab PH U amitraz amidine - 5 unk. M R anilazine unknown I - lab PH U atrazine 1,3,5-triazine I - lab PH U azamethiphos organophosphate - 5 lab M S azinphos-methyl organophosphate - 3 field M R azinphos-methyl organophosphate - 2 field M S azinphos-methyl organophosphate - 4 field M S azinphos-methyl organophosphate - 5 field M S azinphos-methyl organophosphate - 2 field M T azi~hos-methyl organophosphate 4 - field PH U azinphos-methyl orRano~hosphate - 1 lab M R azinphos-methyl organophosphate - 4 lab LC S azinphos-methyl organophosphate - 2 lab M S azinphos-methyl organophosphate 4 - lab PH U azinphos-methyl organophosp_hate - 3.5 med LC R azinphos-methyl orRanophosp_hate - 4 med LC S azinphos-methyl organophosphate - 2 unk. LC S azinphos-methyl organophosphate - 2 unk. M S azocyclotin organotin - 4 field M T Bacillus bacterium I - field PH U thuringiensis Bacillus bacterium I - lab PH U thuringiensis benomyl benzimidazole - 4 field M S 78 benomyl benzimidazole - 2 lab M R benomyl benzimidazole - 2 lab LC S benom~l benzimidazole - 1 lab M T benomyl benzimidazole - 5 unk. M R bentazone unknown 1 - lab PH U benzoximate unknown - 2 field M S binapacryl unknown - 4 field M R binapacryl unknown - 4 field M S binapacryl unknown - 5 unk. M S bitertanol azole 1 - field PH U bitertanol azole 1 - lab PH U Bordeaux mixture inorganic - 3 field M S Bordeaux mixture inorganic - 2 field M T bromacil uracil 1 - lab PH U bromofenoxim hydroxybenzonitrile 4 - lab PH U precursor bromophos unknown - 4 field M T bromophos unknown 3.5 - lab PH U bromo~hos unknown - 3 med LC R bromopropylate benzilate - 5 field M S bromoxynil hydroxybenzonitrile 4 - lab PH U bupirimate pyrimidine - 2 field M S bupirimate pyrimidine - 3 unk. M S buprofezin unknown 2 - lab PH U calcium unknown 4 - lab PH U polysulfide captan N-trihalomethylthio - 2 field M R captan N-trihalomethylthio - 2 field M S captan N-trihalomethylthio - 2 field M T captan N-trihalomethylthio - 2 lab M T captan N-trihalomethylthio - 3 unk. M S carbaryl carbamate - 2.5 field M S carbaryl carbamate - 4 field M T carbaryl carbamate 3 - field PH U carbaryl carbamate - 2 lab LC R carbaryl carbamate - 2 lab M T carbaryl carbamate 4 - lab PH U carbaryl carbamate - 3 med LC R carba~yl carbamate - 2.5 med LC T carbaryl carbamate - 2 unk. M S carbaryl carbamate - 2 unk. LC T carbendazim benzimidazole - 4 field M T carbendazim benzimidazole - 5 unk. M R carb<:>pheno-thion unknown - 5 field M S chinomethionat unknown - 5 field M S chinomethionat unknown - 3 lab M T chinomethionat unknown 4 - lab PH U chlordecone unknown - 2 field M T chlordimeform unknown - 5 lab M S chlordimeform unknown - 5 lab M T chlordimeform unknown - 5 unk. LC S chlorfenson unknown - 1.5 field M S 79 chlorfenvinphos organophosphate 4 - lab PH U chlormequat quaternary 1 - lab PH U ammonium chloropropylate unknown - 5 field M T chloropropylate unknown - 5 lab M S chloropropylate unknown - 5 unk. LC T chlorothalonil unknown 3 - lab PH U chlorpyrifos organophosphate - 4 field M R chlorpyrifos organophosphate 4 - lab PH U chlorthiophos unknown - 4 lab M S cIofentezine unknown 1 - field PH U cIofentezine unknown 1 - lab PH U cIopyralid pyridinecarboxylic 1 - lab PH U acid copper inorganic 1 - field PH U oxychloride copper inorganic 1 - lab PH U oxychloride cyfluthrin pyrethroid 4 - field PH U cyfluthrin pyrethroid 4 - lab PH U cyhexatin organotin - 3.5 field M R cyhexatin organotin - 5 field M S cyhexatin organotin - 3 lab LC S cyhexatin or~anotin - 2 lab M S cJlhexatin or~anotin - 3 lab M T cyhexatin organotin - 3 unk. M R cypermethrin pyrethroid 4 - field PH U cypermethrin pyrethroid - 5 lab M S cypermethrin pyrethroid 4 - lab PH U cypermethrin pyrethroid - 4 med LC R cypermethrin pyrethroid - 5 unk. M S cyproconazole azole 3 - lab PH U cyromazine unknown 4 - lab PH U DDT or~anochlorine - 4 field M R DDT or~anochlorine - 3 field M S DDT organochlorine - 4 field M S DDT organochlorine - 2.5 field M T deltamethrin pyrethroid 4 - field PH U deltamethrin pyrethroid - 5 lab M R deltamethrin pyrethroid 4 - lab PH U deltamethrin pyrethroid - 5 med LC S deltamethrin pyrethroid - 5 unk. M S demeton unknown - 1.5 lab M S demeton unknown - 3 unk. M R demeton unknown - 2 unk. LC S dialifos unknown 4 - field PH U dialifos unknown - 4 lab M T dialifos unknown 3.5 - lab PH U diazinon organophosphate - 4 field M S diazinon organophosphate 2.5 - lab PH U dichlofluanid N-trihalomethylthio - 4 field M S dichlofluanid N-trihalomethylthio 3 - field PH U 80 dichlotluanid N-trihalomethylthio 2 - lab PH U dichlone unknown - 4 field M S dicofol organochlorine - 2 field M S dicofol organochlorine - 4 field M T dicofol organochlorine - 2 lab M S dicofol organochlorine - 5 unk. M S difenoconazole azole 1 - field PH U difenoconazole azole 3 - lab PH U ditlubenzuron benzoyl urea - 2 field M S ditlubenzuron benzoylurea 1 - field PH U ditlubenzuron benzoylurea 1 - lab PH U ditlubenzuron benzoylurea - 1 unk. M R ditlubenzuron benzoylurea - 3 unk. M S dimethoate organophosphate - 5 field M S dimethoate organophosphate 3 - field PH U dimethoate organophosphate - 4 lab LC R dimethoate organophosphate - 4 lab M R dimethoate organophosphate - 4 lab M S dimethoate organophosphate - 5 lab M T dimethoate organophosphate 4 - lab PH U dimethoate organophosphate - 5 unk. LC T dinocap dinitrophenol - 4 field M R derivative dinocap dinitrophenol - 5 field M T derivative dinocap dinitrophenol - 3 lab LC S derivative dinocap dinitrophenol - 5 unk. M R derivative dithianon unknown 1 - field PH U dithianon unknown 1 - lab PH U dithianon unknown - 3 unk. M S dodine guanidine - 2 field M S dodine guanidine - 2 lab M S dodine guanidine - 3 unk. M S endosulfan organochlorine - 2 field M T endosulfan organochlorine - 3 lab LC S endosulfan organochlorine - 1 lab M T endosulfan organochlorine - 2 unk. LC T epofenonane unknown - 2 field M T ethephon ethylene generator - 3 unk. M S ethiofencarb carbamate - 3 lab M S ethiofencarb carbamate - 3 lab M T ethiofencarb carbamate 4 - lab PH U ethion organophosphate - 4 field M R ethirimol pyrimidine 1 - lab PH U ethofumesate benzofuranyl 2 - lab PH U alkanesulfonate etrimfos unknown 4 - field PH U etrimfos unknown 4 - lab PH U fenarimol pyrimidinyl carbinol 1 - field PH U fenarimol pyrimidinyl carbinol 1 - lab PH U 81 fenarimol pyrimidinyl carbinol - 3 unk. M S fenbutatin oxide organotin - 3 field M R fenbutatin oxide organotin - 2 field M T fenbutatin oxide oroanotin - 2 lab M S fenbutatin oxide organotin - 2 lab LC T fenitrothion organophosphate - 1 lab M S fenitrothion organophosphate - 4 lab M T fenitrothion organophosphate 4 - lab PH U fenoxycarb carbamate 1 - field PH U fenoxycarb carbamate 2 - lab PH U fenpropathrin pyrethroid 4 - lab PH U fenpropimorph morpholine 1 - lab PH U fenvalerate pyrethroid - 5 field M S fenvalerate pyrethroid - 4 field M T fen valerate pyrethroid - 5 lab M S ferbam dimethyldithio - 2 field M S carbamate ferbam dimethyldithio - 2 field M T carbamate fluazifop-butyl 2-(4­ 1.5 - lab PH U aryloxyphenoxy) propionic acid flubenzimine unknown - 3 field M S flubenzimine unknown 3 - field PH U flubenzimine unknown 4 - lab PH U flufenoxuron benzoyl urea 1 - field PH U fluroxypyr aryloxyalkanoic acid 2 - lab PH U flutriafol azole 1 - lab PH U fluvalinate unknown 4 - field PH U fluvalinate unknown 4 - lab PH U folpet N-trihalomethvlthio 1 - field PH U folpet N-trihalomethylthio 1 - lab PH U formetanate carbamate - 5 lab M S formetanate carbamate - 5 unk. LC T glufosinate­ unknown 4 - lab PH U ammonium glyodin unknown - 2 field M T glyodin unknown - 4 lab M S glyphosate unknown 1.5 - lab PH U haloxyfop 2-(4­ 3 - lab PH U aryloxyphenoxy) propionic acid heptenophos organophosphate 2 - field PH U heptenophos organophosphate 4 - lab PH U hexaconazole azole 1 - lab PH U hexadecyl unknown - 4 field M R cyclopropane carboxylate hexadecyl unknown - 2 lab M S cyclopropane carboxylate hexythiazox unknown 1 - field PH U hexythiazox unknown 1 - lab PH U 82 ioxynil hydroxybenzonitrile 3 - lab PH U iprodione dicarboximide 1 - field PH U iprodione dicarboximide 1 - lab PH U isoproturon urea 1 - lab PH U lambda­ pyrethroid 4 - field PH U cyhalothrin lambda­ pyrethroid 4 - lab PH U cy_halothrin lecithin unknown 1 - field PH U lecithin unknown 1 - lab PH U leptophos unknown - 5 field M R malathion organophosphate - 5 field M S malathion organophosphate - 4.5 field M T mancozeb alkylenebis - 5 field M R (dithiocarbamate) mancozeb alkylenebis - 5 field M T (dithiocarbamate) mancozeb alkylenebis 4 - field PH U (dithiocarbamate) mancozeb alkylenebis 4 - lab PH U (dithiocarbamate) mancozeb alkylenebis - 5 unk. M R (dithiocarbamate) maneb alkylenebis 3 - field PH U (dithiocarbamate) maneb alkylenebis - 3 lab LC S (dithiocarbamate) maneb alkylenebis 4 - lab PH U (dithiocarbamate) menazon unknown - 4 field M S metamitron 1.2,4-triazinone 1 - lab PH U methabenzthiazur urea 3.5 - lab PH U on methamidophos organophosphate 4 - lab PH U methidathion organophosphate - 4 field M S methomyl oxime carbamate - 5 lab M S metiram alkylenebis - 5 field M S (dithiocarbamate) metiram alkylenebis 4 - field PH U (dithiocarbamate) metiram alkylenebis - 2 lab M S (dithiocarbamate) metiram alkylenebis 3.5 - lab PH U (dithiocarbamate) metsulfuron­ sulfonylurea 1 - lab PH U methyl mevinphos organophosphate 4 - field PH U mevinphos organo~hosphate 3 - lab PH U nicotine unknown - 2 field M S nuarimol pyrimidinyl carbinol 1 - lab PH U oxamyl oxime carbamate 4 - lab PH U oxydemeton­ organophosphate - 4.5 field M S methyl 83 oxydemeton­ organophosphate - 4 lab M S methyl parathion organophosphate - 4.5 field M S parathion organophosphate - 4 field M T parathion organophosphate - 4 lab M S parathion organophosphate - 2.5 med LC R parathion organophosphate - 5 med LC S parathion organophosphate - 5 unk. M S parathion-methyl organ~hosphate - 3 lab M S penconazole azole 1 - field PH U penconazole azole 1 - lab PH U permethrin pyrethroid - 5 med LC S perrnethrin pyrethroid - 4 field M S permethrin pyrethroid - 5 field M S perrnethrin pyrethroid - 5 field M T perrnethrin pyrethroid - 5 lab M S permethrin pyrethroid - 5 unk. M S phosalone organophosphate - 2 lab LC R phosalone organophosphate - 1 lab M S phosalone organophosphate - 2 med LC T phosalone organophosphate - 2 unk. M R phosalone organophosp_hate - 2 unk. LC T phosmet orEanophosphate - 4 field M R phosmet organophosphate - 5 field M S phosmet organophosphate 4 - field PH U phosmet organophosphate - 4 lab M T phosmet organophosphate 4 - lab PH U phosmet organophosphate - 3 med LC S phosphamidon organophosp_hate - 5 field M S phosphamidon mganophosppate - 4 lab M S pirimicarb carbamate - 2 field M S pirimicarb carbamate - 4 lab LC S pirimicarb carbamate - 2 lab M S pirimicarb carbamate - 3 unk. M R pirimiphos-methyl organophosphate - 4 lab M R pirimiphos-methyl organophosphate - 5 unk. M S prochloraz azole 4 - lab PH U procymidone dicarboximide 1 - field PH U procymidone dicarboximide 1 - lab PH U propargite unknown - 4 field M R pro~aI"gite unknown - 2 lab M T propargite unknown - 3 unk. M S propiconazole azole 3 - lab PH U propineb alkylenebis 4 - field PH U (di thiocarbamate) propineb alkylenebis 4 - lab PH U (dithiocarbamate) propineb alkylenebis - 5 unk. M S (dithiocarbamate) pro~oxur carbamate - 2 med LC R propoxur carbamate - 2 med LD S pro~xur carbamate - 5 med LD U 84 quizalofop-ethyl 2-(4­ 4 - lab PH U aryloxyphenoxy) propionic acid sethoxydim cycIohexanedione 2.5 - lab PH U oxime simazine 1,3,5-triazine 1 - lab PH U sulfur unknown - 4 field M S sulfur unknown - 3.5 field M T sulfur unknown 4 - field PH U sulfur unknown - 2 lab LC S sulfur unknown - 2 lab M T sulfur unknown 3.5 - lab PH U sulfur unknown - 5 unk. M S tebuconazole azole 2 - field PH U tebuconazole azole 3 - lab PH U tefl ubenzuron benzoyl urea 1 - field PH U tefl ubenzuron benzoylurea 1 - lab PH U tetradifon unknown - 2 field M S tetradifon unknown - 2 field M T tetradifon unknown 1 - field PH U tetradifon unknown 1 - lab PH U tetradifon unknown - 3 unk. M S thiocycIam 2-dimethylamino 3 - field PH U propane-l,3-diol thiocycIam 2-dimethylamino 2 - lab PH U propane-l,3-diol thiometon organophosphate - 5 lab M S thiophanate­ benzimidazole - 2 field M S methyl precursor thiophanate­ benzimidazole - 3 unk. M S methyl precursor thiram dimethyldithio - 5 field M T carbamate thiram dimethyldithio 4 - lab PH U carbamate tralkoxydim cycIohexanedione 1.5 - lab PH U oXIme triadimefon azole - 2 field M S triadimefon azole - 3 unk. M S triadimenol azole 1 - field PH U triadimenol azole 1 - lab PH U triazophos organophosphate 4 - lab PH U tridemorph morpholine 1 - lab PH U triforine unknown 1 - lab PH U vamidothion organophosphate 4 - field PH U vamidothion organophosphate 4 - lab PH U vamidothion organophosphate - 5 unk. M S Verticillium fungal 2 - field PH U lecanii Verticillium fungal 1 - lab PH U lecanii zineb alkylenebis - 5 field M T (dithiocarbamate) 85 zineb I alkylenebis 2 LC T I lab I l (dithiocarbamate) I I I a = Data calculated from research conducted by the IOBCfWPRS (International Organization for Biological Control of NoxiousAnimals and Plants - West Palaearctic Regional Section; Hassan et al. 1985): IOBC Toxicity Rating: 1 2 3 4 IOBC Range of effect: Harmless Slightly Moderately Harmful Harmful Harmful

J3 =Data calculated from the SELCTV database (Theiling 1987, Theiling and Croft 1988):

SELCTV Toxicity Rating: 1 2 345 SELCTV Range of effect: 0% <10% 10-30% 30-90% >90% y field =field tests, lab =laboratory tests, med =median lethal assay & probit, unk. =methods or analysis not found II : LC =lethal concentration, LD =lethal dose, M =mortality, PH =percent harm t : R = resistant, S = susceptible, T = tolerant, U= unknown