Information to Users

INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely afreet reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, begiiming at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & Howell Informaticn Company 300 North Zed) Road, Ann Arbor MI 48106-1346 USA 313/761-4700 800/521-0600

THE IMPACTS OF CHEMICAL MANAGEMENT OF PESTS, DISEASES AND WEEDS ON INVERTEBRATES IN TOMATO AGROECOSYSTEMS

DISSERTATION

Presented in Partial Fulfillment of the Requirements for

the Degree Doctor of Philosophy in the Graduate

School of The Ohio State University

Erdal N. Yardim, B.S., MS.

*****

The Ohio State University 1996

Dissertation Committee; Approved by Clive A. Edwards, Adviser

Celeste Welty V\A>€ . Adviser Richard M. Riedel Environmental Science Graduate Program UMI Number: 9710688

This microform edition is protected against unauthorized copying under Title 17, United States Code.

UMI 300 North Zeeb Road Ann Arbor, MI 48103 Copyright by Erdal Necip Yardim 1996 ABSTRACT

The overall effects of pesticide use in agroecosystems have received a great deal of

attention in recent decades because of the potential undesirable non-target effects of

pesticides. By designing and implementing a field experiment comparing the recommended

chemical management of pests, diseases and weeds in tomato agroecosystems, I was able to evaluate the effects of a comprehensive pesticide treatment program, individual components and lower chemical inputs on the incidence of pests, predators and nematodes over two seasons. The pesticide treatments used included; i) full spectrum pesticide use; based on a comprehensive pesticide application schedule including insecticides (carbaryl, endosulfan, and esfenvalerate), a fungicide (chlorothalonil) and herbicides (trifluralin and paraquat); ii) insecticide only use treatment; based on applications of the same insecticides and doses as in the full spectrum pesticide use treatment; iii) fungicides and herbicides only treatments; based on applications of the same doses fungicide and herbicides as in the full spectrum pesticide use treatment; iv) control plots received no pesticide applications. The pest populations studied included: aphids (potato aphid,

Macrosiphum euphorbiae and green peach aphids, Myzus persicae), flea beetles (Epitrix spp.), and thrips (Thrips spp.). The predatory arthropods investgated included ground beetles (Carabidae), rove beetles (Staphylinidae), tiger beetles (Cicindellidae), lady beetles (Coccinellidae), spiders (Araneae), ants ( Formicidae), and anthocorids (Anhocoridae:

Hemiptera).

Insecticide applications provided efiScient control of aphids and flea beetles, but they led to reductions in numbers of various groups of predators. In general, foliage-dwelling predators were more sensitive to pesticide applications than ground-dwelling predators.

Fungicide and herbicide applications caused decreased populations of anthocorids and foliage-dwelling spiders and consequent increases in the numbers of aphids and flea beetles. Free-living nematode populations were sensitive to pesticide applications in general; in particular insecticide applications consistently led to increase in plant parasitic nematode numbers. Early blight caused by Altemaria solani was a serious disease limiting the yields. An economic comparison of pesticide applications on processing tomatoes showed that fungicide applications were critical to maintain yields and profitability of tomato production in the presence of early blight disease.

m To Hatice, Sehaded and Sehid

IV ACKNOWLEDGMENTS

I wish to express my deepest gratitude to my advisor. Dr. Clive Edwards for his guidance, patience and strong support throughout my graduate training. I am also grateful to the other members of my committee. Dr. Celeste Welty and Dr. R.M. Riedel for their technical advice and comments in my field work and dissertation. I thank Bill Shuster and

Mike Anderson for their assistance during my field work and for their sincere fiiendship. I also wish to thank my wife, Hatice, for her patience and faith, my daughter, Sehaded, and my son, Sehid, for the joy they brought to my life during this endeavor. VITA

March 25, 1966 ...... Bom - Batman, Turkey

1990 ...... M.S. Entomology, Ege University, Izmir, Turkey

1990-199 1...... Research Assistant, Yuzuncu Yil University, Van, Turkey

1991- present...... Graduate Teaching and Research Associate, The Ohio State University

FIELDS OF STUDY

Major Field; Environmental Science (Entomology, Agroecosystem Ecology and Integrated Pest Management)

VI TABLE OF CONTENTS

Page

Dedication...... iv

Acknowledgments...... v

Vita...... vi

List of Tables...... ix

List of Figures...... x

Chapters:

1. Introduction...... 1

2. The effects of chemical management of pests, diseases and weeds on arthropod pests and fungal pathogens in processing tomatoes...... 18

Introduction...... 18 Materials and method ...... 26 Results ...... 29 Discussion...... 33

3. The effects of chemical management of pests, diseases and weeds on surface- dwelling predators in tomatoes...... 59

Introduction...... 59 Materials and method ...... 64 Results ...... 65 Discussion...... 68

4. The effects of chemical management of pests, diseases and weeds on foliage- dwelling predators, commonly found on tomatoes...... 86

Introduction...... 86 Materials and M ethod ...... 87

vii Results ...... 88 Discussion...... 90

5. The effects of chemical management of pests, diseases and weeds on the trophic structure of nematode populations in tomato agroecosystems...... 101

Introduction...... 101 Materials and method ...... 104 Results ...... 105 Discussion...... 106

6. An economic comparison of the use of pesticide applications on processing tomatoes in Ohio...... 119

Introduction...... 119 Materials and method ...... 123 Results...... 124 Discussion...... 125

7. General discussion ...... 132

References...... 149

VUl LIST OF TABLES

Table Page

2.1 Arthropod pests and diseases in Ohio...... 44

2.2 Pesticide application schedule in tomato fields in 1994 and 1995 ...... 46

2.3 Sampling dates of aphids, flea beetles and thrips in experimental plots, 1994 and 1995...... 47

2.4 Seasonal abundance of arthropod pests in experimental plots, 1994 and 1995 (mean ± SEM)...... 48

3.1 Seasonal abundance of predators per pitfall trap in tomato fields, 1994 and 1995 (mean ± SEM)...... 75

4.1 Seasonal abundance of the foliage-dwelling predators in tomato fields, 1994 and 1995 (mean + SEM) ...... 94

5.1 Seasonal abundance of nematode trophic groups in experimental plots in response to pesticide treatments (mean + SEM)...... 114

IX LISTOFnGURES

Figure Page

2.1 Seasonal variations in the aphid populations (mean ± SEM) resulting from pesticide treatments, 1994...... 49

2.2 Seasonal variations in the aphid populations (mean + SEM) resulting from pesticide treatments, 1995...... 50

2.3 Seasonal variations in flea beetle populations (mean + SEM) resulting from pesticide treatments, 1994...... 51

2.4 Seasonal variations in flea beetle populations (mean + SEM) resulting from pesticide treatments, 1995...... 52

2.5 Seasonal variations in thrips populations (mean + SEM) resulting from pesticide treatments, 1994...... 53

2.6 Seasonal variations in thrips populations (mean ± SEM) resulting from pesticide treatments, 1995...... 54

2.7 The effects of pesticides on defoliation of plants resulting from early blight disease of tomatoes, 1994...... 55

2.8 The effects of pesticides on defoliation of plants resulting from early blight disease of tomatoes, 1995...... 56

2.9 The effects of pesticides on the finit blight resulting from early blight disease of tomatoes, 1994...... 57

2.10 The effects of pesticides on the finit blight resulting from early blight disease of tomatoes, 1995...... 58

3.1 Seasonal variations in carabid populations (mean ± SEM) resulting from pesticide treatments, 1994...... 76 3.2 Seasonal variations in carabid populations (mean + SEM) resulting from pesticide treatments, 1995...... 77

3.3 Seasonal variations in staphylinid populations (mean ± SEM) resulting from pesticide treatments, 1994...... 78

3.4 Seasonal variations in staphylinid populations (mean ± SEM) resulting from pesticide treatments, 1995...... 79

3.5 Seasonal variations in cicindelid populations (mean± SEM) resulting from pesticide treatments, 1994...... 80

3.6 Seasonal variations in cicindelid populations (mean ± SEM) resulting from pesticide treatments, 1995...... 81

3.7 Seasonal variations in ground-dwelling spider populations (mean ± SEM) resulting from pesticide treatments, 1994...... 82

3.8 Seasonal variations in ground-dwelling spider populations (mean ± SEM) resulting from pesticide treatments, 1995...... 83

3.9 Seasonal variations in ant populations (mean ± SEM) resulting from pesticide treatments, 1994...... 84

3.10 Seasonal variations in ant populations (mean + SEM) resulting from pesticide treatments, 1995 ...... 85

4.1 Seasonal variations in C.maculata populations (mean ± SEM) resulting from pesticide treatments, 1994 ...... 95

4.2 Seasonal variations in C.maculata populations (mean ± SEM) resulting from pesticide treatments, 1995...... 96

4.3 Seasonal variations in Anhocoridae populations (mean + SEM) resulting from pesticide treatments, 1994...... 97

4.4 Seasonal variations in Anthocoridae populations (mean + SEM) resulting from pesticide treatments, 1995...... 98

4.5 Seasonal variations in foliage-dwelling spider populations (mean ± SEM) resulting from pesticide treatments, 1994...... 99

4.6 Seasonal variations in foliage-dwelling spider populations (mean ± SEM) resulting from pesticide treatments, 1995...... 100

xi 5.1 Seasonal variations in bacterivorous nematode populations (mean + SEM) resulting from pesticide treatments ...... 115

5.2 Seasonal variations in frmgivorous nematode populations (mean + SEM) resulting from pesticide treatments ...... 116

5.3 Seasonal variations in carnivorous nematode populations (mean + SEM) resulting from pesticide treatments ...... 117

5.4 Seasonal variations in plant parasitic nematode populations (mean + SEM) resulting from pesticide treatments ...... 118

6.1 Economic comparisons of the pesticide treatments in 1994...... 130

6.2 Economic comparisons of the pesticide treatments in 1995...... 131

XU CHAPTER 1

INTRODUCTION

The positive and negative impacts of agricultural practices on agroecosystems have

received a great deal of attention by researchers, growers and the general public. Among

the practices used commonly, chemical pest management of arthropod, disease and weed

pests in agroecosystems has remained an important but largely controversial matter

because of the potentially detrimental effects of pesticides on the environment. Although

pesticides are an important component of today’s agricultural productivity, the risk they

pose to the environment raises a question on the benefits derived fi’om their use.

Increasing pesticide prices, decreasing profits per hectare (Metcalf, 1980) and growing

public concern about pesticide use further complicate the pesticide dilemma.

The yearly world-wide pesticide application has been estimated as 2.5 million tons and the amount of pesticides that hit the actual target has been calculated as less than 0.1%

(Pimentel, 1995). The rest moves into the environment and cause adverse effects on aquatic and terrestrial invertebrates and vertebrates, microorganisms as well as microbial processes ( Newsom, 1967; Smith and Stratton, 1986; Edwards, 1987; Baron and Merriam, 1988). In addition, pesticides can have detrimental effects on species diversity,

natural enemy/prey relationships (Edwards and Thompson, 1973), and can cause

development of resistance, as well as secondary resurgence of pest populations (Metcalf

1980; Edwards, 1987).

The effects of chemical pest management, in particular the use of insecticides, upon the

natural enemies of pests in agroecosystems is one of the aspects that has been most

extensively studied. Records go back as early as in the 1940s when agriculture adopted to

synthetic pesticides. Newsom and Smith (1949) believed that organic chemicals such as

pesticides adversely affected the natural balance between pests and their natural enemies

in cotton agroecosytems. DeBach (1951) observed dramatic increases in scale insect, mite

and mealy bug populations in citrus and attributed these increases to use of DDT use that

destroyed natural enemy fauna. Campbell and Hutchins (1952) provided information about

the sensitivity of predators to organochlorine insecticides in cotton. Glick and Lattimore

(1954) showed that proper timing of insecticide applications was usually less detrimental

to natural enemies. Bosch et al. (1956) emphasized the necessity of using the minimum

amounts of detrimental insecticides to minimize their effects on natural enemies and

thereby maximize the use of natural control mechanisms. Ripper (1956) reviewed

research in the1940s and 1950s and categorized the effects of insecticides in to two groups: short-term and long-term. He considered that short-term effects of insecticides would often result in resurgence of pest populations while long-term effects would cause development of resistant strains of pests to insecticides. In addition to these early studies, most of which indicated adverse effects of organochlorine insecticides, other commonly used insecticides have also been shown to be detrimental to natural enemies (Ripper, 1956). Burke (1959) and Fenton (1959) found that organophosphorus insecticides were more toxic to beneficial arthropods than organochlorines. Yun and Ruppel (1964) ranked the toxicity of the three groups of insecticides according to their effect on Coleomegilla maculata (CoIiCoccinellidae).

Carbaryl was most toxic to these beetles followed by organophosphates and organochlorines. Hamilton and Kieckhefer (1969) believed that organophosphorus insecticides were promising for control of pests in cereal crops because predators were less affected than their aphid prey. Elmer et al. (1983) studied the effects of permethrin and azinphosmethyl, that were applied for control of pear psylla,Psylla pyriacola^ on an anthocorid predator, Anthocoris nemoralis, a predatory mite, Balaustrum putmani,

Miridae, and Reduviidae in pear orchards. They recorded that permethrin applied at petal fall reduced A. nemoralis populations significantly while azinphosmethyl did not.

However, both insecticides reduced the predatory mite, mirid, and reduviid populations.

Permethrin was shown to reduce the numbers of predators, staphylinids, coccinellids, lygaeids, anthocorids, lycosids, mirids, nabids, reduviids, elaterids, oxyopids, tetragnathids, salticids, Inyphiids, thomisids, dictynids and theridiids that feed on tetranychid mite in cassava plants (Braun et al. 1987). Kerns and Gaylor (1993) showed the effects of the pyrethroid insecticide cypermethrin and the organophosphate insecticide, sulprofos on natural enemy densities and the efficacy of natural enemies to control the cotton aphid. Aphis gossypii. They found that predator-induced mortality was lower in insecticide treated plots and that cypermethrin suppressed aphid pathogens. Bellows et al.

(1988) demonstrated that the systemic insecticides demeton and aldoxycarb applied to

control citrus thrips, Scirtothrips citri and citrus red mite, Pcmonychus citri did not harm the predaceous mite, Euseius tularensis directly in a citrus agroecosystem. However, they concluded that the predaceous mite would still be afifected by demeton and aldojqfcarb through ingestion of contaminated prey. Another systemic insecticide, aldicarb, did not have any influence on ant, spider, and staphylinid populations while a broad spectrum insecticide, fenvalerate, reduced their numbers in sugarcane (Showier and Reagan, 1991).

In the same study, however, it was shown that predator numbers were reduced on the foliage through aldicarb applications due to the response of the predators to lower prey populations reduced by aldicarb. Granular insecticides, chlorpyrifos, ethoprop, fonophos, and terbufos, were tested by Mack (1992) for their impacts on abundance of predators in peanut fields. All these insecticides caused reduction in carabid, spider, and ant populations. Chlorpyrifos had the greatest impact on the predators. He hypothesized that insecticides would have more lasting effects on ground-dwelling predators because they cannot reinvade agroecosystems in so short a time as flying predators can.

In addition to field observations, many laboratory bioassay studies have reported detrimental effects of the insecticides on natural enemies. Hough-Goldstein and Keil

(1991) showed that endosulfan, oxamyl, and esfenvalerate reduced the numbers of twospotted stink bug, Perillus bioculatus, a predator of Colorado potato beetle,Leptinotarsa decemlineata. They found that endosulfan, esfenvalerate and oxamyl caused 100%, 76% and 68% mortality, respectively, when the predators were exposed to contaminated leaves, and 96%, 60% and 44% mortality, respectively, when the predators ingested contaminated beetles that fed on contaminated foliage. Mizell and Sconyers (1992) demonstrated the effects of foliage residues of the systemic insecticide imidacloprid on common predators: a mirid, Deraecoris nebulosus\ coccinellids, Olla v-nigrum,

Hippodomia convergence-, a chrysopid, Chrysoperla rufiilabrisr, a lygaeid, Geocoris puncipes; and phytoseiid mites. The insecticide, applied at field rate toxic to phytophagous insects, was toxic to all predators except phytoseiid mites. They concluded that although foliar applications were detrimental to the predators, a seed treatment and soil drench of imidacloprid would eliminate the harm to predators. Patel and Vyas (1985) rated the ovicidal toxicity of some insecticides used for Heliothis armigera and

Spodoptera litura control in cotton to the eggs of green lacewings,Chrysopa scelestes, an important predator. Carbaryl had the highest toxicity (32-92%). The toxicity of other insecticides were : 15.73% for malathion, 9.61% for quinalphos, 8.65% for endosulfan,

3.08% for phenthoate, 0.01% for cypermethrin, 0.45% for monochrotophos, 0.52% for phosalone, 0.82% for fenvalerate and 0.86 % for permethrin. They concluded that cypermethrin, monochrotophos, phosalone, fenvalerate, permethrin and endosulfan would be safer than malathion and quinalphos to use in cotton fields. Fabellar and Heinrichs

(1984) tested the toxicity of cypermethrin, deltamethrin, endosulfan, carbophenthion, acephate, azinphosethyl, carbofuran and carbosulfan to predators of rice brown planthopper, Nilaparvata lugens. None of the insecticides were toxic as a stomach poison, through ingestion of the planthoppers, to the spider predator,Lycosa pseudoannulata. Deltamethrin caused 47% mortality in the mirid predator Cyrothinus lividipermis when it was fed on treated N. lugens. By comparison, azinphosmethyl,

carbofuran, carbosulfan, cypermethrin, and deltamethrin residues caused 60% mortality in

L .pseudoannulata and 96% in C. lividipennis numbers. Singh and Verma (1986) studied

the toxicity of some insecticides, that were recommended for the control of cotton

boUworm, to larvae ofChrysoperla camea and adults of Trichogramma basiliensis

Ashmed. At field application rates, endosulfan, quinolphos, monochrotophos,

phenothoate, and fenitrothion caused 74-89% mortality when C. camea larvae consumed the contaminated eggs over a 72 h. period while phosalone, carbaryl and cypermethrin were shown to be moderately toxic and led to 34.1-38.1% larval mortality. All of the insecticides tested were shown to be highly toxic to T. basiliensis adults and caused 84-

100% mortality. Brown et al. (1983) showed that methyl parathion, cypermethrin and primicarb had the highest, intermediate and the least toxicity to common predators,

Coccinella septempunctata (Coleoptera: Coccinellidae); Syrphus spp, (Diptera:

Syrphidae); Pterostichus melanarius, Nebria brevicollis, Agormm dorsale (Coleoptera;

Carabidae); and Erigone spp (Araneae: Lnyphiidae) respectively. They indicated that toxicity studies in laboratory conditions would not reflect the field situation exactly and stated that build up of vapor pressure in test chambers could increase the mortality and laboratory mortality responses would be observed only for a certain time (24h, 48h etc.).

However, in the field, mortality could occur in several ways such as through exposure to residues of pesticides, direct contact during applications, or ingestion of contaminated prey; the toxicity of a pesticide could change with environmental factors such as temperature and other climatic changes; biological availability of pesticides residues could be lower because they would be sprayed on natural surfaces such as soil and plants compared to inert surfaces like glass used in laboratories.

Some insecticides provide a selectivity favoring natural enemies. Free and Hagley

(1985) showed that primicarb was not toxic toChrysopa oculata. Moreover, Lecrone and Smilowitz (1980) reported that primicarb provided a selectivity ratio of 4000:1 in favor of predatorsColeomegilla maculata and Chrysopa oculata over the green peach aphids, Myzus persicae. Coats et al. (1979) found that permethrin and fenvalerate were selective favoring coccinellids over their pests. Hull et al. (1985) noted that some synthetic pyrethroids had the potential to be integrated into biological control in apple orchards.

In many cases, microbial insecticides have been reported not to be detrimental to nontarget arthropods (Hassan et al, 1983; Cheng and Hanlon, 1990). Muckenfuss and

Shepard (1994) showed that Bacillus thuringiensis (Bt) did not harm the natural enemies of diamondback moth, Plutella xylostella on collards. Little mortality of the natural enemies, Aphytis melinus and Rhizobius lophmuhae, was observed by Bellows and Morse

(1993) after the use of B. thuringiensis^ in citrus. However, Flexner et al. (1986) reviewed the effects of microbial pesticides on beneficial arthropods and concluded that indirect mortality of natural enemies caused by their consumption of infected prey by microbial pesticides was usually more significant than direct mortality these caused.

Related to that, Giroux et al. (1994) showed that B. thuringiensis caused a reduction in predation efficiency of Coleomegilla maculata. Webb et al. (1989) evaluated the effects of aerial sprays of nuclear polyhedrosis (NPV), B. thuringiensis, and a growth regulator, difiobenzuron, on natural enemies of gypsy moth, Lymantria dispar in hardwood plots. They observed that a significant reduction in numbers of the parasitoid, Cotesia

melanoscela occurred in NPV and diflobenzuron-treated plots, whereas there was increase

of the parasitoid numbers in B. thuringiensis-\iQax.Qd plots. They concluded that the reason

for the parasitoid reduction through NPV was because the infected gypsy moth larvae

died before the parasitoid in the moth completed development.

Indirect and sublethal effects of insecticides on populations of natural enemies have received less attention. Such indirect effects include starvation through the elimination of prey, consequently emigration of the predators fi'om contaminated fields (Newsom,

1967), feeding on contaminated prey (Croft and Brown, 1975). Hurej and Dutcher (1994) showed that feeding on pecan aphids, Monelliopsis pecanis, treated with esfenvalerate, killed Hippodemia convergens in Ih. Sublethal effects can include changes in reproductive, feeding, dispersal and locomotory behavior (Haynes, 1983). Consequently, fewer prey items may be consumed (Roger et al. 1994). Hardman et al. (1991) reported that insecticide applications to control European red mites caused predatory mites to reduce feeding and oviposition on prey and avoid sprayed surfaces. McMurty et al. (1970) reviewed indirect effects of spray practices on the predatory mite (Phytoseiidae) and stated that sprays would reduce the egg production of phytoseiids, eliminate the food sources by killing phytophagous mites and inhibit development of organisms such as fimgi, lichens that serve as food for alternate prey on which predators feed.

Resurgence of pest populations after pesticide applications is a well-known phenomenon. Elimination of natural enemy activity from agroecosystems has been shown as a major mechanism of this phenomenon (Ripper, 1956; Metcalf, 1980; Godfray and Chan, 1990). Hardman et al. (1991) evaluated the toxicity of azinphosmethyl,

cypermethrin, deltamethrin, dimethoate and primicarb to the European red mite,

Panonychus ulmi and phytoseiid mite, Typhlodromus pyri in apple orchards and made

the assumption that exclusion ofT pyri by pesticide applications would increase red mite

populations. They observed that pesticides reduced T. pyri populations and consequently

increased the red mite populations. They concluded that the effects of pesticides were additive in suppressing the natural enemies and higher numbers of the red mites were present when more pesticides had been applied. In a study by Pike and Allison (1987), applications of carbaryl, cyfluthrin, cypermethrin, fenvalerate and fluvalinate eliminated predatory mites from a com agroecosystem and resulted in significant increases in populations of spotted mite. Similarly, azinphosmethyl applications suppressed phytoseiid mite, Metaseilus occidentalis populations and resulted in resurgences of phytophagous spider mite populations to economically damaging level in almonds (Welter et al. 1987).

Biddinger and Hull (1995) reported similar results in apple orchards in which fenoxycarb had reduced numbers of predatory lady beetles, Stethorus punctvm populations and caused a late-season increase in phytophagous mite populations. Organophosphorus insecticides, azinphosmethyl, malathion and diazinon have been shown to cause outbreaks of two spotted spider mite, Tetrcmychus urticae populations by destroying Stethorus punctum populations in red raspberries (Shanks et al 1992). Bentley et al. (1987) compared the effects of azinphosmethyl, carbaryl, diazinon, permethrin, flucythrinate and cyphlutrin, applied to control of navel orangeworm,Amyleois transitella (Lepidoptera;

Pyralidae), on the abundance of web-spinning spider mite, Tetrcmychus spp. and their predators on almond. They reported that all insecticides increased the abundance of spider

mites significantly. Pyrethroids were the most effective in causing the resurgence. They

concluded that the resurgence was due to the destruction by pesticides of predators

(particularly phytoseiid mites), controlling the spider mites. Penman and Chapman (1988)

reviewed the factors inducing spider mite outbreaks following use of pyrethroid

insecticides and believed that the major fector was the differential toxiciy of pyrethroids to

predatory mites (Phytoseiidae) and phytophagous mites (Tetranychidae); that is, field

application rates of the insecticides were excessively toxic to the predators, although only

slightly toxic to pest mites. Braun et al. (1987) studied the impact of permethrin on the cassava pests, Mononcyhellus progresivus and T. iirticae and natural enemies, Amblyseius limonicus and a staphylinid, Oligota minuta in a laboratory and a cotton field. They showed that predators were an order of magnitude more susceptible to permethrin than their prey in the laboratory. A. limonicus numbers were reduced significantly while M. progresivus numbers were increased by permethrin applications in cotton fields. They noted that M progresivus was 32 to 58 fold more resistant to permethrin thanA. limonicus and O. minuta. Poehling (1988) stated that early timing of insecticide applications prevented the establishment of aphidophagous arthropods and accelerated increases in numbers of aphids in winter wheat. Morrison et al. (1979) found an inverse relationship between numbers of com earworm, Heliothiszea larvae and predators, Orius insidious, Geocoris puncipes and Nabis spp. in aldicarb-treated plots. H. zea populations were seven times higher in treated plots than controls and, this was attributed to reductions in predator numbers. Campbell et al. (1991) stated that endosulfan and

10 methomyl applications led to increases of Heliothis zea egg densities because these

treatments eliminated the egg predators from the system. Etienne (1990) studied the

effects of insecticides on Thrips palmi on eggplant (aubergine). He observed that the density of thrips was significantly higher in profenfos treated fields than in untreated fields and increased exponentially with the same pattern throughout the season. He attributed this increase of the thrips to the reduced activity of the predators, Amblyseius and

Phytoseius mites, dolichropids (Dipt ), Orius insidious, Neohydatothrips porticensis in insecticide treated fields. In another study on tomatoes, Johnson et al. (1980) showed that methomyl affected the parasites, Diglyphus begini and Chrysontomia punctiventris adversely and significantly increased densities of Liriomyzajativae.

Other possible potential mechanisms that could cause resurgence of pest populations by pesticide applications were outlined by Hardin et al. (1995) as alteration of plant quality, induction of insect detoxification enzymes, enhancement of fecundity, changes in insect behavior, reduction in competition, and insecticide resistance. Braun et al. (1987) believed that permethrin enhanced the fecundity of tetranychid pests and this, along with reduction of predators, contributed to outbreaks of the mite populations in cassava plants.

Because the insecticides did not reduce the populations of predatory mites significantly in grain sorghum, Buschman and Depew (1990) discussed the possibility of some physiological changes in plants induced by chlopyrifos and parathion applications that made plants more attractive to pests and caused outbreaks of banks grass mite,

Olygonychus pratensis populations. Similarly, Kerns and Gaylor (1993) reported that outbreaks of cotton aphid. Aphid gossypi populations in sulprofos-treated cotton could be

11 due to increased threanine and essential amino acid levels in plants that enhanced the aphid fecundity, because parasitism of aphids by Lysiphlebus testaceipes was not reduced by insecticide applications. Penman and Chapman (1988) believed that dispersal response of spider mites to the repellent and irritant properties of pyrethroid insecticides resulted in reduction in density-dependent competition, when mites left insecticide-treated foliage and crop that gave the remaining mites opportunity of full reproductive potential. Additionally, they believed that pyrethroids caused a shortened development time by Panonychus ulmi, increased its generation numbers, delayed its diapause, and changed its host physiology.

Fungicides

Although fungicides are designed primarily to kill fungal pathogens, they can also affect beneficial arthropods and pest incidence, however, there is relatively less information on their impacts. Fungicides may not be toxic to arthropods. Bromophos, triform, dichlofluanid and metiram were shown not to be toxic to the aphid predator,

Aphidoletes aphidomyza (Diptera: Cecidomyiidae) (Sell, 1985). Similarly, vinclozolin and iprodione did not affect Stethorus punctum picipes populations that control spider mites,

T. urticae in raspberries. Fungicides may destroy some fungi on which beneficial arthropods feed. For instance, Aebischer and Pots (1990) reported that fungicide applications killed the fungi serving as food sources to rove beetles. They can suppress some entomogenous fungi that control arthropod pests. Redcliffe et al. (1978) reported, fungicide applications enhanced aphid populations in an alfalfa ecosystem because they suppressed fungal enemies of the aphids. Direct toxicity of fungicides to invertebrates is attributed to their insecticidal properties (Sotherton et al. 1987; Sotherton and Moreby,

12 1988). Benomyl is a fungicide commonly used in agroecosystems and there are numerous

reports upon its detrimental effects on natural enemies. Although benomyl was not

reported to be toxic toChrysopa oculata (Free and Hagley, 1985) or had a low toxicity

to Coleomegilla maculata (Roger et al. 1994), Mizell and Schiffhauer (1990) reported

50% mortality resulting from benomyl applications to the aphid predators Chrysoperla

rufilabris, Hippodomia convergens , Cycloneda sanginea and Aphelinus perpallidus, in

pecans. Nakashima and Croft (1974) showed that benomyl was ovicidal to predatory

mites, Amblyseius fallacis. They also showed that ingestion of two spotted spider mites,

Tetroarychus urticae contaminated by benomyl caused a depression in oviposition by the

predatory mites. In another study with mites. Bower et al. (1995) stated that because

fungicides often had suppressive effects on mite populations, applications of benomyl and

mancozeb led to decreases in predatory mite and increase in phytophagous mite numbers

in apple orchards. Boykin et al. (1984) noted that increases in numbers of Tetranychus

urticae to a damaging level in benomyl and mancozeb treated peanut fields were because

fungicides caused a reduction in the efficiency of entomophagous fungi, Neozygites floridana controlling the mites.

Herbicides

Although herbicides are believed to have less direct influence on both pests and natural

enemies of pests, some can be directly toxic to predators (Boiteau, 1984), and can cause

mortality, longer instar duration, reduction in egg production and viability and have

repellent effects on predators (Agnello et al. 1986; Subagie and Snider, 1981). Stam et al.

13 (1978) reported that Coleomegilla maculata, Erotmocerus holdemani, Orius ittsidious,

Geocoris puncipes and Scymnus louisiana were affected by dinoseb in cotton fields.

Adams (1960) stated that 2,4-D amine applications in grain fields caused 50% mortality

and increased of the time to pupation on coccinellid larvae. Ingram et al. (1947) believed

that 2,4-D applications reduced parasitism of sugarcane borer, Diatraea saccharalis by

Trichogramma minitum and led to higher infestations of this pest. Adams and Drew

(1965) observed a resurgence of aphid populations in oats after suppression of coccinellids

by 2,4-D applications. On the other hand. Maxwell and Harwood (1960) believed that

2,4-D dimetyl amine changed the nutritional factors in plants and this in turn increased the

reproductive rate of pea aphid, Macrosiphum pisi.

Indirect effects of herbicides are usually more important in agroecosystems. Herbicides

can influence populations of pests and their natural enemies by removing plant cover that

serves as shelter for natural enemies and alternative hosts for pests (Edwards, 1991).

Powell et al. (1985) believed that a significant decrease in carabid and staphylinid

populations in herbicide-treated fields was a response of predators to destruction of plant

cover. Brust (1990) stated that populations of carabids migrated out of atrazine, simazine,

paraquat and glyphosate-treated fields in a response to destruction of plant cover.

Pesticides can affect not only above-ground species but also soil-inhabiting

invertebrates by their direct toxicity and by affecting their reproduction or indirectly by killing their natural enemies (Edwards, 1989). Fungicides can change food supply of invertebrates by killing microorganisms on which they feed (Edwards and Thompson,

1973). Little is known about the effects of pesticides on the community structure of

14 nematode populations in agroecosystems. However, there is evidence that agrochemicals can cause structural changes in nematode communities (Mahn and Kastner, 1985, Bohlen and Edwards, 1994). Use of inorganic fertilizers can increase plant parasitic nematode populations in com (Bohlen and Edwards, 1994). Herbicides, chlormetoxynil and thiobencarb-simetryne mixture can increase numbers of fungivorous and plant parasitic nematodes and decrease numbers of carnivorous nematodes in rice agroecosystems

(Ishibashi et al. 1983). Parmelee et al. (1994) found that in a microcosm study, copper reduced numbers of carnivorous nematodes and increased populations of plant parasitic nematodes. Smolik (1983) found that carnivorous nematodes were more sensitive to aldicarb, terbufos and carbofuran than microbial feeding nematodes, and decreases in populations of carnivorous nematodes in response to the nematicides led to increase in populations of bacterivorous nematodes.

From an economic perspective, pesticide applications are one of the major energy inputs used to increase productivity in agroecosystems. Pimentel (1977) estimated that 33

Kcal energy is needed to be able to apply 1 lb of pesticide per acre. This involves labor, machinery, as well as any kind of energy spent in manufacturing, transportation and distribution. Farmers tend to be willing to spend this energy to protect crops from possible damage and maximize their output, once the investment of labor is made

(Headley, 1972). Pimentel et al. (1993) estimated a $3-$5 return in direct benefit to farmers for every they $1 invested in the use of pesticides. This figure would change however if indirect environmental costs such as human pesticide poisoning, reduction of natural enemies of pests, development of pesticide resistance, honey bee losses, reduction

15 of pollination, crop and tree losses, and fish and wildlife losses were also taken into account.

In spite of the increasing use of pesticides, crop losses to pests have remained at constant rational level of about 33%, because high cosmetic standards and low tolerance levels increase the rate of losses to pests and result in intensified insecticide use. This in turn causes higher pesticide residues, greater environmental and social costs, increased energy use and eventually increased food cost (Pimentel et al. 1977).

Rising costs of pesticides, increasing crop losses despite increasing pesticide use, adverse efifects of pesticides on the environment, and increasing public concern all necessitate the utilization of alternative pest control strategies so that less pesticides can be used in a more ecologically, economically and socially acceptable manner. As an alternative to pesticide use, integrated pest management (IPM) aims to minimize the disadvantages and maximize the advantages of pesticide use (Metcalf, 1980). Pesticide use can be reduced by 50% in IPM programs that utilize an apropriate combination of cultural, biological, ethological and chemical measures (Salas, 1992).

Tomatoes are a crop that receives large amounts of pesticides. Tomatoes are grown worldwide for fi'eshmarket and processing and are subject to many arthropod pest and disease attacks. Control of these pests is the main determining factor in the successful growing on both commercial and small scales because pests can reduce the yield and quality of tomatoes by seriously damaging the plant and their finit. To prevent losses to pests and to meet cosmetic standards, heavy chemical control measures are implemented on tomatoes (Harding, 1971; Lange and Bronson, 1981; Brun, 1981; Mishra, 1984;

16 Wiesenborn et al. 1990; Walgenbach et al. 1991; Walgenbach and Estes, 1992). Because of the environmental and economic costs of pesticides, considerable effort is being made to reduce their use on tomatoes by disease forecasting (Poysa et al. 1993) and pest scouting (Antle and Park, 1986) and utilizing biological control, mating disruption techniques and using microbial pesticides (Trumble and Alvardo-Rodriques, 1993).

The objectives of my research were to determine the environmental impacts of chemical pest management on important components of tomato agroecosystems in Ohio.

More specific objectives were to determine their impacts on commonly found predators, arthropod pests, diseases and soil fauna using nematodes as indices; to evaluate their impact on yield and profitabilty; to compare what component of the management had more influence.

17 CHAPTER 2

THE EFFECTS OF CHEMICAL MANAGEMENT OF PESTS, DISEASES AND

WEEDS ON ARTHROPOD PESTS AND FUNGAL PATHOGENS IN PROCESSING

TOMATOES

Introduction

Tomatoes are the hosts of a wide range of insect pests and diseases in Ohio (Table

2.1). Intensive pesticide applications may be required to keep these pests in acceptable

limits and below the economic threshold in tomato agroecosystems (Brun, 1981;

Hamilton and Toflfolon, 1987; Kennedy et al. 1983; Walgenbach et al. 1989). Tomato fruit

quality or cosmetic standards, primarily related to insect damage, is considered one of the

major factors that results in the emphasis on chemical control in tomatoes to keep the

product free of damage (Lange and Bronson, 1981).

Many attempts have been made to achieve eSicient levels of control of tomato fruit pests. Mishra (1984 ) showed that decamethrin (0.02 kg ai/ha), fenvalerate (0.05 kg ai/ha), permethrin (0.1 kg ai/ha), cypermethrin (0.05 kg ai/ha), endosulfan (0.70 kg ai/ha), and folithion (0.50 kg ai/ha) provided significant control of tomato fruit borer,Heliothis armigera and fruit damage ranged 1.23-9.60% in insecticide treatments

18 compared with 25.33% fruit damage in untreated control treatments. Kennedy et al.

(1983) reported that weekly fenvalerate (0.11 kg/ha) applications between finit set and

harvest significantly reduced fruit losses to fruit pests, Heliothis spp. and shorter

application intervals were needed to achieve adequate control, when Heliothis activity was intense. Walgenbach et al. (1991) determined the persistence of esfenvalerate, carbaryl, methomyl and endosulfan on tomato foliage for implication of tomato finitworm,

Helicoverpa zea control. They observed that esfenvalerate had the longest persistence with a half-life of > lOd while carbaryl, methomyl and endosulfen had a half-life ranging from <2 to <8d, and esfenvalerate caused more than 65% larval mortality, 14 days postapplication, while carbaryl and endosulfan were effective for 4-8 days and methomyl for caused mortality for only 48h. In another study, Walgenbach et al. (1989) showed that a 5-day application interval was necessary for carbaryl and endosulfan treatments while esfenvalerate provided efficient control of H.zea at 15 day intervals. Poe and Everett

(1974) compared single and combined insecticide applications for control of tomato pinworm, Keiferia lycopersicella. They found that the foliar insecticides, acephate (1 lb ai/Acre), diazinon (1 lb ai/Acre), endosulfan (1 lb ai/Acre) and methomyl (1 lb ai/Acre) all provided very efficient control levels and kept tomatoes almost pinworm-free. However, they showed that chlordimeform (0.5 lb ai/Acre) used alone caused phytotoxicity to tomato plants, but provided efficient control level combined with Bacillus thuringiensis at rates of 0.25 lb + 0.25 lb ai/Acre. When they compared the foliar insecticides with granular systemic insecticides such as carbofuran and disulfoton, they considered that granular insecticides were less effective in pinworm control because they provided only

19 29-78% larval mortality. Similarly, Shuster (1978) reported that the foliar insecticides

dimethoate, methomyl, phosphamidon, azinphosmethyl, and lindane were effective on tomato pinworm, whereas granular applications of carbofuran, phorate and disulfoton did not provide any protection from the pinworms. Jaques and Loing (1989) compared the effectiveness of biological and chemical insecticides against Colorado potato beetle,

Leptinotarsa decemlimata, and observed that B. thuringiensis did not differ from permethrin in effectiveness against the beetles. The beetles were significantly fewer in plots treated with these insecticides than in untreated plots. Ghidiu and Getting (1987) showed the effectiveness of oxamyl slow release-tablet, foliar and transplant drench applications on Colorado potato beetles. They found that one (0.14 kg ai/ha), two (0.28 kg ai/ha) and three (0.42 kg ai/ha) slow release tablets applied to the roots were equally effective and they all were more effective than foliar applications (1.12 kg ai/ha) and drenches (0.28 kg Al/ha). They noted that slow release-tablets caused phytotoxicity to tomato plants and that this increased as the numbers of tablets applied increased. Hof&nan et al. (1987) evaluated the efBcacy of commonly-used insecticides applied from the air and ground against stink bugs (Hemiptera; Pentatomidae) on tomatoes. They showed that although insecticide applications by ground equipment did not affect the green bugs, aerial applications of Thiodan, methamidophos and methyl parathion provided adequate control of southern green bug, whereas applications of these insecticides had only minimal effect on consperse stink bug populations because they reached the soil not the plant and spray coverage was poor.

20 There are also many reports indicating adverse eflfeas of insecticide applications on the natural enemies of pests and causing a resurgence of pest populations in tomato agroecosystems. For instance, weekly applications of methomyl and carbaryl significantly reduced parasitism of lepidopterous pests, tomato finitworm, H.zea and cabbage looper,

Trichoplusia ni, (Roltsch and Mayse, 1983). Similarly, HofiBnan et al (1994) and HoflBnan et al. (1996) reported that carbaryl was extremely toxic to the parasites, Trichogramma spp. of tomato fitiitworm, and caused a 67% reduction in the parasitism by the parasites.

Oatman et al. (1983) compared weekly applications ofBacillus thrungiensis Berlinger var kurstaki (0.56 kg Al/ha) plus biweekly releases of Trichogramma pretiosum Riley, with weekly applications of methomyl (0.5 kg Al/ha) for control of tomato fruitworm and cabbage looper. They observed that the parasitism rate of tomato fitiitworm was 95.8, 92 and 100% and the parasitism rate of cabbage looper was 81, 70 and 82.9% in B. thuringiensis, methomyl-treated, and untreated plots respectively, and that there was a significant increase in numbers of vegetable leafininers, Liriomyza sativae in methomyl- treated plots, because leafininer parasites were affected adversely by the applications.

Johnson et al. (1980) showed that methomyl applications for controlling vegetable leafininers, Liriomyza sativae affected its parasitic arthropod populations adversely and resulted in significantly higher population densities of the pest in tomatoes. Similarly,

Campbell et al. (1991) observed that the egg densities of tomato fitiitworm,Heliothis zea were much higher in endosulfan and methomyl-treated tomato fields, where predators of the eggs of this pest were eliminated through insecticide applications. Schuster and Price

(1985) compared the impact of methomyl, endosulfan and permethrin on lepidopterous

21 larvae and the emergence of leafininer parasites in tomatoes. Permethrin and methomyl caused a significant reduction in numbers of larvae of southern armyworm, Spodoptera eridonia, cabbage looper, Tricplushia ni and tomato fitiitworm,Heliothis zea one day after applications. However, endosulfan did not reduce numbers of southern armyworm, whereas it was effective on other pests. They showed that methomyl and permethrin caused a significant reduction in the numbers of leafininer parasites and that endosulfan was the least toxic to the parasites. Chemical control of tomato fruitborer,Neoleucinodes elegantolis, and the leafininer species, Ptharimaea operculella, Scrobipalpula absoluta and Liriomyza sp. caused reductions in the populations of natural enemies and in turn resurgence of the pest populations that necessitated several sprays in one season (Salas,

1992). Pohronezny et al. (1986) reported that 34 insecticide applications were needed to controlLiriomyza spp. in a single crop because of insecticide applications that eliminated natural enemies of this pest and leafininers developed resistance to pesticides.

Trumble and Alvardo-Rodriques (1993) showed the effects of conventional and lower input (IPM) pest management programs on populations of major pests and their natural enemies in tomatoes, where former pest management had been based on multiple applications of two broad-spectrum insecticides, methamidophos and permethrin, while the IPM was based on release of parasitic wasps, T. pretiosum, use of B. thuringiensis, pheromones and minimum amounts of endosulfan if needed. They found that leafininer populations were consistently higher in conventional treatment plots than in those with

IPM and untreated plots and this was because of adverse effects of insecticides on the parasites of leafininers. Schuster et al. (1979) stated that two applications of oxamyl per

22 week did not reduce leafininer populations below threshold (1 active mine per 3 terminal

leaflets) and led to a significant reduction in the numbers of an emerging hymenopterous

parasites; however, based on regular scouting and use of economic threshold, a single

application of oxamyl and methamidophos did not affect parasite emergence, while

reducing numbers of leafininers below threshold level.

There is relatively less information about the indirect and direct effects of fungicides on

insect pest populations. Nevertheless, fungicide applications can affect pest populations in

several ways. Livingston et al. (1978) reported that benomyl applications caused a

resurgence of populations ofPseudoplusia includens in soybean. Bower et al. (1995)

stated that, because benomyl and mancozeb had mite-suppressive activity, they reduced

predatory mite populations and led to higher phytophagous mite numbers in apple

orchards. Redcliffe (1976) concluded that fungicide applications suppressed

entomogenous fungi and caused an increase in the numbers of pea aphids, Acyrtosiphon

pisum and spotted alfalfa aphids, Therioaphis macidata in alfalfa agroecosystems.

Most herbicides are known to be relatively nontoxic to arthropods (Farlow and Pitre,

1983; Stam et al. 1978). Some herbicides can increase pest populations by changing the

nutrient content of plants. Application of the dimethylamine salt of 2,4-D in oat crops

increased the nitrogen content of plants and led to higher numbers of aphids. Probably the

most important effect of herbicides is the removal of alternative hosts which that results in

a greater concentration of pests on crops or removal of shelter for polyphagous predators.

Tomatoes are susceptible to a large number of diseases (Jones et al. 1991). With the warm and humid weather conditions in eastern North America, diseases became a major

23 limiting factor and required fungicide applications at 7-14 d intervals. (Gleason et ai.

1995). Many attempts have been made to achieve an efficient level of various disease control using various fungicides in tomatoes. For instance, Fagg and Fletcher (1987) showed that benomyl provided the best control of tomato stem rot caused by Didymella lycopersici compared with the fungicides carbendazim, thiophanate-methyl, captofol and chlorothalonil. However, they found that benomyl was not effective on damping-off disease of tomatoes caused by Pythium aphanidermatum\ whereas incorporation of ethazol plus thiophanate-methyl, and fenaminosulf plus ethazol were effective against that disease. Gitatis et al. (1992) recommended copper for control of bacterial speck caused by

Pseudomonas syringae and copper mixed with maneb or mancozeb for control of bacterial spot caused by Xanthomonas campestris pv. vesicatoria. Copper was also found to be efficient against bacterial canker caused by Clavibacter michiganensis ( Gleason et al.

1993). Chellemi et al. (1994) showed that fumigation of soil with a 67:33 mixture of methyl bromide and chlorpicrin at the rate of 448kg/ha significantly reduced populations of soil the pathogens,Phytopthora nicotianae, Fusarium oxysporium f.sp. radicis- lycopersici and F.o. lycopersici of tomatoes.

Early blight caused by Altemaria solani and anthracnose caused by Colletotrichum coccodes are destructive diseases of tomatoes in Ohio and require season-long protectant fungicide applications (Reed et al. 1993). Walgenbach et al. (1989) reported that early blight disease had great influence on marketable yield of tomato and protectant fungicides were necessary to prevent establishment and development of this disease. They also showed that 5-d fungicide application intervals was the most effective way to control the

24 disease. Brammal (1993) demonstrated that chlorothalonil reduced disease progress and

severity However, it did not have any effect on yield. He considered this probably was

because environmental conditions did not allow the disease to develop enough to cause

yield losses.

Because of environmental and economic effects of fungicide use, many studies have

been done to develop forecast-based fungicide application systems, to minimize fungicide

use, that is traditionally required on weekly basis. Madden et al. (1978) developed a

computerized forecasting system for early blight of tomatoes to determine the suitability of

the environment that would favor development of the disease and obtain a schedule of

efficient fungicide applications. They stated that timely applications, based on forecasting

provided adequate control of the disease, with fewer fungicide applications than weekly

applications. Penypacker et al. (1983) showed that based on a computerized forecasting

system (FAST), 2 and 5 chlorothalonil applications in subsequent years were not different

from weekly application intervals, which required 9 applications to control early blight of tomatoes. Willamson and Hilty (1988) reported that a forecast-based spray regimen that required 12-13 sprays did not provide better control of early blight than 5-day application intervals that required 14 sprays. Moreover, there was no difference between forecast- based spray and 7-day spray regimen that required 14 applications. Gleason et al. (1992), working with disease severity values and weather sensor locations for the TOM-CAST disease-warning model, stated that 6-10 fewer fungicide applications was obtained for control of early blight and septoria leaf spot caused by Septoria lycopersici in Iowa.

25 The objective of this part of the study was to determine the efifects of different sets of pesticide applications on the main arthropod pests and pathogenic diseases in tomatoes in general, and particularly on aphids (potato aphid, Macrosiphum euophorbiae, green peach aphid, Myzus persicae), flower thrips (Thrips spp.), and flea beetles (Epitrix spp.) that were abundant pests, and early blight caused by Altemaria solani that was the destructive disease in my tomato systems.

Materials and Method

Field and Experiment Description

The field experiment was at a Ohio State University/Ohio Department of Agriculture

Demonstration Farm in Reynoldsburg, OH which has' compared various practices for growing vegetables for several years. The field has a medium textured silty loam soil and receives approximately 7.5 inches rain during the vegetable growing season. The field was cultivated in May and fertilized with a recommended dose of NPK based on a soil test in both 1994 and 1995. The fertility amendment rates used were 2.15 kg/plot 10:20:20 NPK before planting, 0.20 kg/plot 46:0:0 as side dressing in 1994 and 1.50 kg/plot 8:32:16

NPK before planting and 0.36 kg/plot side dressing in 1995.

The study was based on a randomized block design with twelve 4.5 m x 6 m plots in

1994 and sixteen plots in 1995, involving four treatments each with three replicates in

1994 and four replicates in 1995. Each plot consisted of the equivalent of four rows with

48 tomato plants. The same plots were used in 1994 and 1995. Each row contained 12 tomato plants spaced 15 inches apart. Heinz 8704 processing type tomatoes were grown

2 6 in both years and tomato seedlings were transplanted in the field on 15 June in 1994 and 1

June in 1995.

Treatments

The treatments were:

1) Full spectrum pesticide use: This treatment was based on a comprehensive pesticide

application schedule including insecticides, fungicides, and herbicides, targeting certain

insect pests, diseases and weeds. The pesticides used and application schedules and rates

for both years are summarized in Table 2.2. A back pack sprayer (model no: C-18)

manufactured by Berthoud was used to apply pesticides. The sprayer had 3 gal/min flow

rate and 40 psi (pounds per square inch) pressure. The nozzle used was manufactured by

Teejet. A stainless sto type tip (no: TJ60-8003VS) was used. The insecticide

recommendations were based on an intermediate-intensity application schedule for

tomatoes used by a canning company in Ohio. The fungicides were applied based on a

traditional weekly schedule. The trifluralin was applied as preemergence herbicide before

planting and paraquat was applied as needed. The application rates were determined based on the recommendations provided by Ohio Vegetable Production Guide (1994) prepared by Ohio State University Extension. This full spectrum pesticide treatment was used to determine the environmental impacts of a conventional chemical pest, disease and weed management, such as those often employed by farmers, on the main components of tomato agroecosystems and show their effects on the yield of tomatoes and economical feasibility.

27 2~> Insecticide only use: This treatment was based on using only insecticide applications.

The insecticides used and the application schedule were the same as those used in full spectrum pesticide use treatment. The aim of this treatment was to determine the role of the insecticides on the components mentioned above and the yield and economics.

31 Fungicide and herbicide only use: These plots received only the fimgicides and herbicides as used in the full spectrum pesticide use treatment. This treatment aimed the role of fimgicides on plant pathogens, and herbicides on soil and surface faima as well as other components and the yield and economics.

4) Control: This treatment served as the control and no pesticides were used on any plots in order to be able to compare the effects of pesticide treatments on the components in treated plots with effects in imtreated plots. Weeds were cultivated as needed.

Sampling

Experimental plots were scouted every week for pest and disease incidence beginning the second week after planting and up to harvest time. Four tomato plants (one plant from per row) were selected randomly in each plot and all leaves and stems of those plants were searched for flea beetles, aphids and as well as other pests. Numbers of flea beetles were recorded on the stem and leaves of tomato plants. The numbers of aphids were recorded on only three leaflets of upper foliage of tomato plants in 1994, however total numbers of aphid on a plant were recorded in 1995 because aphid population densities were low in that particular year.

Thrips populations were assessed using yellow sticky traps. Two 12cmX 7cm traps in

1994 and four 6cmX 3.5cm traps in 1995 were set up in each experimental plot. The

28 sticky traps were collected one day before each insecticide treatment and replaced one day

after the treatment. The traps were collected from the plots placed in storage bags,

labeled and kept in freezer for later identification, counting and evaluation.

Disease assessment was made every week on four plants in each plot. The last disease

assessment was taken one day before harvest in both seasons and used to show the

severity of early blight in treatment plots. The rate of blighted fruits was estimated on 40

tomato fiuits per plot, picked one day before harvest, to show the contribution of early

blight to marketable quality of the tomato finits.

The data were subjected to an analysis of variance on the basis of the seasonal

abundance and on sampling dates. Group means were compared using t-test.

Results

Seasonal abundances of the pests are summarized in Table 2.4. The seasonal variations

in their abundance are illustrated on figures 2.1-2.6. The defoliation rate and rate of fi-uit

blight of tomato plants by Altemaria solani are demonstrated figures 2.7 and 2.8,

respectively.

1994

The seasonal abundance of the aphids and thrips in response to the pesticide treatments showed considerable variation between years (Table 2.4). Pest populations in general were much lower in 1995 than in 1994. Aphid numbers significantly difiered between treatments (p< 0.001) in 1994. They were significantly higher in the fungicide and herbicide-treated plots than in the control plots (p= 0.0006). Fungicide and herbicide applications caused a 33.2% increase in aphid populations compared with the control.

29 Aphid numbers were reduced by the pesticide treatments in the insecticide-treated plots by

85% and in the full spectrum pesticide-treated plots by 15.2%. The first aphids were detected on 26 July 1994. Aphid populations remained significantly higher (p< 0.01) in the fungicide and herbicide-treated plots for most of the season. This can be attributed to both the reduction of the aphid predators and the presence of more suitable plants in these compared with the ones in the control plots that were defoliated byAltemaria solani The endosulfan (p=0.0001) and esfenvalerate (p= 0.0049) applications decreased aphid populations significantly in the insecticide-treated plots but no significant decrease in aphid populations was observed (p=0.S5, p= 0.40 respectively) in the full spectrum pesticide- treated plots despite the same insecticide application. This was probably due to the similar preference of aphids to healthier plants in the full spectrum pesticide-treated plots.

Flea beetles had a similar response to pesticide treatments as aphids. Pesticide applications reduced the flea beetle populations significantly by 61.3% (p= 0.05) in the full spectrum pesticide-treated plots and by 72.3% (p= 0.0012) in the insecticide-treated plots whereas populations increased significantly by 39.8% in the fungicide and herbicide treated-plots compared relative to the control.

There were no significant differences in flea beetle populations between pesticide treatments after the carbaryl and first esfanvalerate applications. However, flea beetle numbers declined significantly in the insecticide-treated plots after the endosulfan application, compared with the ones in the ftmgicide and herbicide-treated plots (p=

0.0000) and in the control plots (p= 0.0000). Although the same endosulfan applications

30 employed in the full spectrum treated plots the flea beetle numbers were significantly

higher in these plots than in the insecticide-treated plots (p=0.0041).

The pesticide treatments caused increases in thrips populations. Thrips were 16.3%,

12.0% and 0.33% more abundant in the fiill spectrum pesticide, the insecticide and the

fungicide and herbicide-treated plots respectively than in the control plots. This could be

due to the decrease of predator numbers because of the insecticide treatments. No

significant pesticide treatment effects of were detected at any sampling date.

Defoliation caused by early blight was significantly more severe in the control and the insecticide-treated plots than in the fungicide and herbicide, and the full spectrum pesticide-treated plots (p<0.01). Despite the weekly chlorothalonil applications, 33.3% defoliation was observed in the fungicide and herbicide-treated plots. There was no any significant difference between the insecticide-treated plots and the control with respect to defoliation caused early blight of tomatoes. The incidence of blighted fiuits was significantly less in the fungicide and herbicide-treated plots than the insecticide treated

(p=0.0012) and the control plots (p=0.015).

1995

Aphids were 125% and 55% more abundant in the insecticide and the fungicide and herbicide-treated plots respectively, than in the control plots while they had similar abundance in the full spectrum pesticide-treated plots as in the control plots. Nevertheless, there were no significant effects of the treatments on either the seasonal aphid abundance or on the abundance of aphids on individual sampling dates. Carbaryl applications in the insecticide-treated plots caused a resurgence of aphid populations (Fig. 2.2) on 23 and 30

31 June 1995, probably as a response to decreases in numbers of predators. However, this

build up of aphid populations stopped on 29 July after esfenvalerate applicatioiL

Heavy and long lasting rains in 1995 were an important factor that suppressed aphid

populations in July, and thereafter caused dramatic decreases in aphid populations in

every experimental plot.

Flea beetle populations showed the same response to the pesticide treatments in 1995

as they did in 1994. Treatments reduced flea beetle populations significantly by 34% in the

full spectrum pesticide-treated plots (p= 0.05) and by 56.2% in the insecticide-treated

plots, respectively compared with the control plots. However, the flea beetle numbers

were 17% greater in the fungicide and herbicide-treated plots than in the control plots.

Again, this could be because of the decrease in the numbers of predators in the fungicide

and herbicide plots or to the preference of flea beetles to healthier plants over the defoliated plants in the control plots.

In the sampling date analysis, flea beetles were significantly fewer in the insecticide- treated plots (p= 0.05) and in the fungicide and herbicide-treated plots (p= 0.004) than in the control plots on 14 July 1995, after the first esfanvalerate application. However, flea beetle numbers were significantly higher in the full spectrum pesticide-treated plots (p=

0.0039) than in the control plots on 23 July 1995, after the endosulfan treatment and less

(p=0.0039) in the insecticide-treated plots than in the fungicide and herbicide-treated plots, after the second esfenvalerate applications.

Thrips populations in 1995, in contrast to their populations in 1994, were 20.9%, 8.7% and 6.7% less abundant in the full spectrum pesticide-treated, the insecticide and the

32 fungicide treated-plots and herbicide-treated plots than in the control plots, respectively.

In terms of sampling dates, the thrips numbers were significantly higher in the full spectrum pesticide ( p=0.03) and the insecticide-treated plots (p=0.0032) than in the fungicide and herbicide-treated plots, on July 4, after carbaryl application. However, the first esfanvalerate application led to significant reductions in thrips population in the full spectrum pesticide-treated(p= 0.0006) and in the insecticide (p= 0.004)-treated plots compared with the control plots. The second esfenvalerate application reduced thrips numbers only in the insecticide-treated plots. The thrips populations were significantly higher in the fungicide and herbicide-treated plots than in any other plots (p<0.05) on 22

August 1995.

As in 1994, the rates of defoliation were significantly less in the fungicide and herbicide-treated and the full spectrum pesticide-treated plots than in the insecticide- treated and the control plots (p<0.01). Despite weekly chlorothalonil applications the defoliation rate was 87% in the fungicide and herbicide-treated plots. This could be attributed to the warmer and more humid weather that enhanced progress of the disease.

There was no significant difference between the treatments with respect to the rate of occurrence of blighted Suit, probably because the tomatoes were harvested before development of the disease in the fiuit.

Discussion

Most of the major pests of tomatoes that were reported rare in my experimental plots in both the1994 and the 1995 growing seasons. Tomato fiuitworm,H. zea (Campbell et al. 1991; Hoflfinan et al. 1990; 1994; Kennedy et al. 1983; Nfishra, 1984; Oatman et al.

33 1983; Roltsch and Mayse; 1983; Schuster and Price, 1985; Walgenbach et al.

1989;1991), tomato pinworm,K. lycopersicella (Alvardo-Rodriguez, 1988; Poe and

Everett, 1974; Schuster, 1978; Walgenbach et al. 1991), southern armyworm, Spodoptera eridonia and tomato fruit borer, N. elegantolis (Schuster and Price, 1985) were absent from all plots. Colorado potato beetle,L. decemlineata ( Ghidiu and Getting, 1987;

Jaques and Loing, 1989), cabbage looper, T. ni (Oatman et al. 1983; Roltsch and Mayse,

1983; Schuster and Price, 1985), stink bugs ( HofiBnan et al, 1987), leafininers (Salas,

1982; Schuster et al. 1979; Schuster and Price, 1985; Trumble and Alvardo-Rodriques,

1993; Pohronezny et al. 1986) occurred sparsely. Potato aphids, M eubhorbiae (Perring et al. 1988; Walgenbach and Estes, 1992; Walker et al. 1984), green peach aphid, M persicae (Jensen, 1992; Flint and Klonsky, 1985; Kring and Schuster, 1992; Lambdin and

Snodderly, 1984; Perring et al. 1988), flea beetles (Drinkwater et al. 1995), and thrips

(Allen et al. 1993; Boutista, 1994; Cho et al. 1995; Heinz, 1992; Salguero et al. 1991) were abundant in all plots.

The thrips, aphids and flea beetles responded in dififerent ways to pesticide applications.

The efifects of the pesticide applications on pest populations were observed more clearly in

1994 when the weather conditions were more stable. However, long-lasting rains and frequent thunderstorms afifected the arthropod populations adversely as well as the tomato plants in 1995. Similar observations in Ohio by Walker et al. (1984) indicated that rainfall, in combination with higher winds, was a major mortality factor for the on potato aphid,

Macrosiphum euphorbiae.

34 Insecticide applications provided adequate levels of aphid control in the insecticide- treated plots in 1994 and they decreased the seasonal abundance of aphids to 85% those in the control. However, the same insecticide applications in the full spectrum pesticide- treated plots did not provide any significant decrease in aphid populations, compared with the control. Moreover, the aphid populations in the fungicide and herbicide-treated plots were significantly higher than in any other plots throughout the season. Flea beetles responded similarly to aphids to the pesticide treatments in both in 1994 and 1995. All of the insecticides used kept the flea populations under control. Vernon and Mckenzie (1991) reported that foliar sprays were important in controllingEpitrix tuberis in potatoes and pyrethroid insecticides caused higher mortalities than carbaryl because of their longer residual effects. However, Weiss et al. (1991) found that carbaryl was more effective than endosulfan and esfenvalerate on flea beetle populations in canola. My results did not agree with those of Shorey et al. (1962) showing that carbaryl did not affect flea beetles.

Flea beetles were not affected by heavy rainfalls as aphids were in 1995 and their response to pesticides did not show any differences between two season.

One might consider why aphid and flea beetle populations were higher in the fungicide and herbicide-treated plots than in the control, and why they were higher in the full spectrum pesticide-treated plots than in the insecticide treated plots despite the fact that the full spectrum pesticide-treated plots received the same insecticide applications as the insecticide-treated plots. However, the full spectrum pesticide-treated plots received the fungicide and herbicide applications additionally. To explain this phenomenon I can put forward five hypotheses. These are; 1) the fungicide and herbicide applications reduced

35 the populations of predatory arthropods which in turn resulted in higher pest populations,

2) the fungicide and herbicide applications suppressed the fungal parasites of the pests, 3)

the applications increased the fecundity of the pests and resulted in more offspring of the

pests and hence higher populations, 4) the applications caused some physiological changes

in the tomato plants that attracted more pests or stimulated their reproduction and 5) the

fungicide applications provided more nutritious and suitable habitats for the pests by

suppressing the disease (early blight) of the tomatoes.

In terms of the first factor, the fungicide and herbicide applications might have reduced

the predator populations and resulted in higher pest populations; because fungicides can

affect arthropod fauna (Sotherton and Moreby, 1988), the fungicides used in my

experiments could have reduced aphid predator populations in the fungicide and herbicide- treated plots. Mizzel and Schiffouer (1990) showed that benomyl caused mortality of natural enemies of the aphids, Chrysoperla rufilabris (Chrysopidae: Neu), Hippodemia convergence, Cycloneda sorginena ( Col: Coccinellidae) and parasitic wasp, Aphelinius perpallidus in pecans. Benomyl and mancozeb reduced predatory mite numbers and resulted in increase of phytophagous mite numbers in apple orchards (Bower et al. 1995).

Herbicides can have toxic effects on predator populations (Stam et al. 1978) and cause them to migrate out of treated fields. Adams (1960) showed that 2,4-D applications caused 50% mortality in coccinellid larvae in grain fields. Brust et al. (1990) reported that herbicides led carabids to migrate out of fields through their repellent effects.

It is clear firom many studies in the literature that reduction or elimination of predators in agroecosystems results in increase of pest populations (see also Chapter 1). For

36 instance, endosulfan and methomyl applications for control of tomato fraitwonn,H.zea

eliminated the egg predators of this pest and caused an increase in huitworm populations

(Campbell et al. 1991). Profenfos applications on eggplants increased Thrips palmi

population densities, because applications decreased predatory activity system by

destroying predatory mites, and other predators, dolichropids, Orius insidious, and

Neohydatothrips particensis populations (Etienne, 1990). Poehling (1988) reported that

early season insecticide applications in winter wheat gave rise to accelerated growth of

aphid populations because insecticide applications suppressed the establishment of

aphidophagous arthropods. Organophosphorus insecticides have been shown to cause

outbreaks of the two spotted mite populations by destroying predatory mites and

Stethorus punctum (CokCoccinellidae) in raspberries (Shanks et al. 1992). Similarly,

permethrin applications suppressed populations of predatory mites, Amblyseius limonicus

and staphylinids, Oligota minuta and led to increase of T. urticae populations (Brown,

1987). To summarize, the increases in pest populations in the fungicide and herbicide-

treated plots could be attributed to the decrease in predator numbers (see also chapter 3

and 4).

Addressing the second factor, the fungicide and herbicide applications suppressed the

entomopathogenic fungi that controlled the aphid and flea beetle populations. Kish et al.

(1994) reported that fungal pathogens Conidiobolus sp., Vuillemin sp. and Beauveria

bassiana exerted natural mortality on green peach aphids. There is evidence of

suppressive effects of fungicides on entomopathogenic fungi in agroecosystems. Redclifife et al. (1978) believed that fungicide applications caused suppression of entomophagous

37 fungi and enhanced build up of aphid populations in alfalfa agroecosystems. Similarly benomyl and mancozeb applications led to reduction in efficiency of entomophagous fimgi, Neozygetes floridana that in turn increased T. urticae population densities to damaging level in peanut fields (Boykin et al. 1984).

As a result of the fungicide and probably with a contribution of the herbicide applications, entomophagous fungi populations might have been suppressed in the fungicide and herbicide-treated plots in my experiment. Elimination of such fungi that had exerted natural mortality on parasitic fungi in those plots could have led to enhanced aphid growth and increased in their population densities. Higher aphid densities in the full spectrum pesticide-treated plots, that received fungicide and herbicide applications additionally, relative to the those in the insecticide treated plots, indicated lack of entomophagous fungi activity in those plots.

To consider the third factor, the fungicide and herbicide applications caused changes in physiology and nutrient content of tomato plants that in turn attracted aphids and flea beetles: Some reports have indicated that pesticide applications can cause changes in plants that can lead to outbreaks of pest populations. For instance, Buscham and Depew

(1990) believed that parathion and chlorpyriphos caused physiological changes in grain sorghum and this changes led to outbreaks in Banks grass mite, Olygonycus pratensis.

White (1984), reviewing studies on impact of pesticides on pests, considered that fungicides an herbicides and as well as insecticides can cause changes in plant nutrient content such as by increases in nitrogen, protein and amino acid levels, and such changes can result in resurgence of pest populations. Sulprofos has increased threanine and amino

38 acid levels in plants and enhanced aphid population densities in cotton (Kerns and Gaylor,

1993). Nonetheless, possibility of such an effect on plants, caused by the fungicide and herbicide applications in my experiment, is unknown because it would need further analysis to explain this phenomenoiL

About the fourth factor, the applications enhanced fecundity of pest populations;

Reports that are available on effects of fungicide on arthropods do not provide any evidence on stimulatory effects on pest populations. In contrast, Akhtar and Enden (1992) reported that benomyl caused reductions in fecundity of cherry aphid, Rhopalosiphum podi and his review indicated that benomyl could cause mortality and act as feeding deterrant to green peach aphids. Zhou and Carter (1991) showed that fungicide inputs to winter wheat did not have any effects on cereal aphid population development. However, many reports indicated that some insecticides can stimulate aphid fecundity. Lowery and

Sears (1986 a,b) showed that direct exposure to azinphosmethyl, stimulated the reproduction of green peach aphids and led them to produce 20-30% more offspring than the ones that were not exposed to the insecticide. Similarly, fenvalerate can stimulate growth of green peach aphids (Jackson and Wilkins, 1985). Gordon and Mceven (1984) found that second generation of green the peach aphids, from mothers exposed to azinphosmethyl, produced significantly more offspring. They stated that although exact mechanism was unknown, reproductive hormones could be involved in this. Based on this evidence it is possible to speculate that fungicide and herbicide applications might have stimulated the fecundity of aphid and flea beetle populations in my experimental plots.

39 In terms of the fifth factor; fungicides can increase pest populations by suppressing

fungal disease (early blight). Fungicides may not aftect pest populations in the absence of a

disease on plants, however, in presence of a disease, they may affect populations indirectly

by interfering in interactions between insect pests and diseases on the same plant. In other words, insect pest populations could be expected to be lower on an infected plant relative to a healthy one. Hatcher et al. (1994) reported that whenGastrophyta viridula (Col;

Chiysomelidae) fed on infected Rumex crispus and R. obtusifolius there was greater larval mortality than on healthy plants. They also showed that infected plants caused slower larval development, lower growth, pupation at a lower weight, considerable reduction on the fecundity and longevity and reduction in egg viability. They considered that reductions in nitrogen content by 50-70% in infected plants played a great role in the beetles . Lower population densities of aphids and flea beetles in the control plots than in the fungicide and herbicide-treated plots in my experiment might have been because of severe early blight incidence that caused reductions in nutrients particularly nitrogen content in tomato plants in those plots. The same would justify the lower aphid and flea beetle populations in the insecticide-treated plots compared to the ones in the full spectrum pesticide-treated plots. This was confirmed partially by the field observations made on plants in my experimental plots, during the growing seasons, which indicated that tomato plants in the fungicide-treated plots were more vigorous and had darker green color than those in the control plots. Another explanation in relation to pathogen-pest relationship could have been the possibility of a systemic resistance provided by the pathogen against herbivores. Ajlan and Potter (1991) believed that immunization of plants

40 by pathogens may generate systemic resistance to arthropod herbivores. Nevertheless, I do

not think this phenomenon occurred in my experiment because all plants were infected by

Altemaria solani fungus and fungicides only reduced the development of the disease in the treated plots. Based on the available evidence, it can only be concluded that there was an inverse relationship between arthropod pest numbers and disease severity.

The resurgence of aphid populations in the insecticide-treated plots in 1995, after endosulfan applications could be as because of reduction of natural enemy populations.

Another reason could be that endosulfan has a short half life ranging from <2 to <8 d

(Walgenbach et al. 1991).

The pesticide treatments did not have any significant effects on thrips populations.

Despite the non-significant treatment effect on thrips populations in 1994, after the first esfenvalerate application, they became and remained more abundant in the full spectrum pesticide and the insecticide-treated plots than in control plots. Similarly, thrips numbers increased in the fungicide and herbicide-treated plots after 18 July 1994 and remained higher than in the control plots. Dintenfass et al. (1987) stated that populations of western flower thrips Frankliniella occidentalis resurged after parathion applications to onions and these applications increased the thrip populations. Immaraju et al. (1990) reported that thrips can develop resistance to insecticides readily. Western flower thrips,

Frankliniella ocdcidentalis has showed high level resistance to permethrin and bifenthrin, moderate to high level to methomyl and low level of resistance to chlorpyriphos

(Immaraju et al. 1992). Andoloro et al. (1983), discussing the inadequacy of insecticides in thrips control, stated that, because thrips are parthenogenic, have rapid reproduction and

41 immigration ability into treated fields, so it is diflScult to interpret the effectiveness of

insecticides on thrips. Nasruddin and Smitley (1991) studying application intervals to

control thrips reported that 5-10 d intervals were needed to reduce F.occidentalis populations to a tolerable level in greenhouses. Morse and Zareh (1991) and Morse and

Browner (1986) showed that insecticides stimulated thrips fecundity. The possibility of developing resistance to insecticide by thrips and of increased fecundity would be very low in my experiment because different insecticides were used in a single season. Higher thrips numbers in all of the pesticide treated plots, than those in the control plots was probably because of decrease in predator numbers in all treated plots (see also chapter 3 and 4).

Early blight appeared to be the most destructive factor for tomato plants and affected yields in both seasons. Weekly fungicide applications provided eflScient control levels by reducing the disease progress significantly in 1994. Results of the 1994 season agree with those of Madden et al. (1978), Penypecker (1978) and Williamson and Hilty (1988) indicating that weekly fungicide applications reduced early blight of tomato significantly.

However, when environmental conditions were more favorable for sporulation and disease progress, with warmer and more humid weather in the 1995 season, weekly fungicide applications did not provide adequate control and approximately 86% defoliation occurred in the fungicide and herbicide and the full spectrum pesticide-treated plots. The rates of the defoliation with respect to treatments, reflected fi-uit infection and yield, in other words, the plots that had highest rate of defoliation had the highest number of infected finit and inversely, had the lowest yield. There is limited information on impacts of early blight disease on tomato fhiits. The incidence of infected finit in my

42 experimental plots were an order of magnitude lower than indicated by Walgenbach et

al.(I989)( that was 46%). This difference could be because I harvested before high levels

of early bbgbt occurred on tomatoes. The yield of tomatoes with regard to pesticide treatments is discussed in chapter 5.

43 Pests Ranking of importance

Arthropod pests

Potato aphid XX

Green peach aphid XX

Flea Beetles XX

Colorado potato beetle X

Thrips X

Cutworms XX

Whiteflies X

Grasshoppers X

Stink bugs XX

Mites X

Crickets X

Leafininers X

X: do not cause economical damage

XX: can cause economical damage

XXX: cause economical damages i f not controlled.

Table 2.1 : Arthropod pests and diseases in Ohio.

44 Table 2.1 (continued)

Fruitwonns xx

Horawonns xx

Cabbage loopers x

Diseases

Damping off xx

Early blight xxx

Late blight xx

Anthracnose xxx

Septoria leaf spot xx

Gray mold xx

Bacterial spot xx

Bacterial speck xx

Bacterial canker xx

Buckeye rot x

45 Pesticides Rate of Application Date Target A.L Rate 1994 1995

Carbaryl 2.371/ha 1 qt in 75 gal/A 7 July 21 June beetles, cutworms (Sevin XLR Plus)

Endosulfan 3.151/ha 1.33 qt in 75 gal/A 3 Aug. 17 July beetles, aphids, thrips, (Thiodan 3EC) cabbage looper, stink bugs

Esfenvalerate 0.711/ha 9.6 oz in 75 gal/A 22 July 5 July beetles, homworms, (AsanaXL) 13 Aug 29 July finitworms, leafininers 24 Aug 11 Aug Chlorothalonil 3.551/ha 3 pts in 90 gal/A every week fimgal diseases (Bravo 720)

Trifluralin 2.371/ha 2 pts in 20 gal/A 7 June 23 May pre-emerged weeds (Treflan)

Paraquat 1.771/ha 1.5 pts in 20 gal/A 13 July 21 June emerged weeds (Gramoxone Extra)

Table 2.2: Pesticide application schedule in tomato fields in 1994 and 1995.

46 Pests Sampling dates

1994 1995

Aphids 22 July, 15, 23, 30 June

2,10, 18,26 Aug. 9, 14, 23, 29 July

3, 10 August

Flea Beetles 27 June, 15, 23, 30 June

5, 12, 22, July 9, 16, 23, 29 July

2,10, 18, 26 Aug. 10 Aug.

1 Sept

Thrips 6, 21 July 21 June

2,12, 23 Aug 4, 17, 29 July

7, 20 Sept. 11,22 Aug

Table 2.3: Sampling dates of aphids, flea beetles and thrips in experimental plots, 1994 and 1995.

47 Taxa Treatments

Full spectrum Insecticides Fungicides and Control

Pesticide Herbicides

1994

Aphids 9.580± 1.69 1.600±0.33* 38.33+ 7.180* 11.30±2.500

FleaB. 0.951± 0.155* 0.681±0.091* 0.681±0.290* 2.465±0.206

Thrips 103.4± 16.55 99.6±17.16 89.2± 17.16 88.90±15.80

1995

Aphids 0.404+0.150 0.903± 0.386 0.625+0.175 0.403+0.142

Flea B. 0.427+ 0.070 0.281± 0.063* 0.750± 0.104 0.645+0.089

Thrips 75.11± 5.900 86.68± 7.150 88.66+ 6.590 94.94+ 8.300

* significantly différent fi'om the control (p< 0.001)

Table 2.4; Seasonal abundance of arthropod pests in experimental plots, 1994 and 1995 (mean+SEM).

48 1994 Aphids

— O- - - Full spectrum pesticides — -O— Insecticides — - 6 - • Fin^erb. Control

140

120 -

100

4 0 -•

20 -•

— — D - — ------O-

22-Jul 2-Aug 10-Aug 18-A u g 2 6 -A u g Sampling Dates

Esfenvalerate Endosulfan Esfenvalerate Esfenvalerate 22 July 3 August 13 August 24 August

Figure 2.1: Seasonal variations in the aphid populations (mean ± SEM) resulting from pesticide treatments, 1994.

49 1995 Aphids

— O- - ■ Full spectrum pesticides — -O Insecticides — - 6 - - Fung/herb. C o n tro l

•i.4

S3 -■

0 15-Jun 30-Jun 9-Jul 14-Jul 23-JuI 29-Jul 3-Aug 10-Aug Sam pling Dates

Carbaryl Esfenvalerate Endosulfan Esfenvalerate 21 June 5 July 17 July 29 July

Figure 2.2: Seasonal variations in the aphid populations (mean ± SEM) resulting from pesticide treatments, 1995.

50 1994 Epitrix s pp. -.-o--- Full spectrum pesticide — o— Insecticides - a- Fungherb. — m— Control

5-JuI27-Jun 12-Jul 22-JuI I-Aug 18-Aug 26-Aug 1-Sep t t t t t Carbaryl Esfenvalerate Endosulfan Esfenvalerat Esfenvalerate 7 July 22 July 3 August 13 August 24 August

Figure 2.3; Seasonal variations in flea beetle populations (mean ± SEM) resulting fi’om pesticide treatments, 1994.

51 1995 Eimtnisppi

— O' - - Full spectrum pesticide — o— Insecticides -- Fungterb — M— Control

1.6 -■

1.4 •

1.2 "

■2 0.8 : - y

0.6 M

0 .4

Q2 -■

15-Jun 23-Jun 30-Jun 9-Jul 16-Jui 23-Jul 29-Jul 10-Aug Sam pling Dates t t t t Carbaryl Esfenvalerate Endosulfan Esfenvalerate 21 June 5 July 17 July 29 July

Figure 2.4: Seasonal variations in flea beetle populations (mean ± SEM) resulting from pesticide treatments, 1995.

52 1994 Thrips

— O- - - Full %iectrum — D— Insecticides - 6 -- Fung/Herb. — K— Control 3 5 0

2 5 0 S

ai a . I 150 -

100 -•

5 0 •

6-Ju l 21-Jul 2-Aug 12-A ug 7-Sep

Sampling Date ^ ^ ^ ^

Carbaryl Esfeavalerate Endosulfan Esfenvalerate Esfenvalerate 7 July 22 July 3 August 13 August 24 August

Figure 2.5: Seasonal variations in thrips populations (mean ± SEM)resulting from pesticide treatments, 1994.

53 1995 Thrips

— o-- - Full spectrum pesticide — O ------Insecticides — Fing/herb. — m ------C o n tro l

2 5 0

200 •

B. 150

> -

100

2l-Jun 4-Jul 17-Jul 29-M 1 1 -A ug

Sam pling Date

^ ^

Carbaryl Esfenvalerate Endosulfan Esfenvalerate Esfenvalerate 21 June 5 July 17 July 29 July 11 August

Figure 2.6; Seasonal variations in thrips populations (mean ± SEM)resulting from pesticide treatments, 1995.

54 1994 Eariy blight defoliation 80

^50 o

^ 3 0

Full hsecticides Fung/herb Control Spectrum Pesticide Treatments

Figure 2.7: The effects of pesticides on defoliation of plants resulting from early blight disease of tomatoes, 1994.

55 1995 Early blight defoliation

100

90 -■

§ 60 + 1=0 + %■o vS 40 -

30 -

20 -

0 Full Insecticides Funglherb. Control Spectrum Fteticide Treatments

Figure 2.8; The effects of pesticides on defoliation of the plants resulting from early blight disease of tomatoes, 1995.

56 1994 Eariy blight fruit infection

25 •

u •S 1.5

1 ■

0.5 ■

Full hsecticides Fung/herb. Control Spectrum F eticides Treatments

Figure 2.9: The effect of pesticides on the fixiit blight resulting from early blight disease of tomatoes, 1994.

57 1995 Early blight fruit infection

1.8

1.6 -■

1.4 -•

12 *o I' 3 0.8 35 0.6 -■

0.4 ■

0.2

0 Full Insecticides Fung/herb. Control Spectrum Ftsticide Treatments

Figure 2.10; The effects of pesticides on the fruit blight resulting from early blight disease of tomatoes, 1995.

58 CHAPTERS

THE EFFECTS OF CHEMICAL MANAGEMENT OF PESTS, DISEASES AND

WEEDS ON SURFACE-DWELLING PREDATORS IN TOMATOES

Inroduction

Generalist pest predators can have considerable effects on pest populations (Clark et

al. 1994; Bryan and Wratten,1984) and often maintain pest populations in some form of

equilibrium. Spiders have different hunting species (Agnew and Smith, 1989) and degrees

of feeding specialization (Nyfeller et al. 1992). They are one of the more important

arthropod groups often exerting density dependent mortality on pests (Provencher and

Coderre, 1987) and limiting insect pest populations below the economic threshold both

in natural ecosystems and agroecosystems (Riechert and Lockey,1984). Similarly, most

carabids are carnivorous and can feed on more prey as their number increase (Kharboutli

and Mack, 1991). Edwards et al. (1979) showed that carabid beetles are important

predators of aphids. Brust et al. (1985) found significantly more com plants cut by the black cutworm, Agrotis ipsilon (Lepidoptera; Noctuidae) in the absence of carabid beetles than when they were present. Dennis and Wratten (1991) reported that some staphylinid

59 (Coleoptera) species efifectively decreased numbers of grain aphids, Stobion avenae.

Because of their relatively stable populations and feeding habits, ants have been recognized as important in pest management (Way and Khoo, 1992). Perfecto and Sediles

(1992), showing the effects of ants on pests of maize, reported that reduction of ant foraging activity caused significant increases in fall armyworm, Spodoptera frugiperda and com leafhopper, Dalbulus mcàdis populations in maize.

Most arthropod predators are sensitive to pesticide practices and are nontarget organisms for pesticides in many agroecosystems (Cockfield and Potter, 1983; George et al. 1992; Mack, 1992; Terry et al. 1993). Many of the studies have reported adverse effects of insecticides to predators. Carabids are one of the predator groups that are often exposed to insecticides. Edwards and George (1981) reported that range of organophosphoms insecticides decreased carabid beetle populations. Floate et al. (1989) showed that deltamethrin, dimethoate and carboftiran caused 30%, 73% and 83 to 100% mortality in carabid beetle populations respectively. Los and Alen (1983) noted that

Harpalus pennsylvanicus was much more abundant in untreated alfalfa fields than in insecticide-treated ones. Critchley (1972), investigating the effects of soil applied organophosphoms insecticide thionazin on carabid populations, at 11.2 and 44.8 kg/ha rates in potatoes, detected considerable reductions in carabid numbers eight weeks after applications, and the effect was still observable after the potatoes were removed fi-om field because of the residual activity of the insecticide. Riddick and Mills (1995), studying the impact of microbial and oil insecticides on the seasonal activity of carabids in an apple orchard, found that Harpalus pensylvanicus and Chlaenius sp. became significantly more

60 active because of food (prey) shortage when granulosis virus and oil insecticides were

sprayed to control codling moth,Cydia pomonella.

Andersen and Sharman (1983), investigating the impacts of chlorfenvinphos and isofenphos on Carabidae and Staphylinidae and their predation on eggs of turnip root fly,

Delia floralis, reported that chlofenvinphos did not have any adverse effect on the predators. However, isofenphos reduced the numbers of carabids significantly whüe it did not affect the numbers of staphylinids. However, Shires (1985) showed that parathion- methyl and cypermethrin applied to control cereal aphids in spring wheat caused 10% and

50% reductions, respectively, in carabid and staphylinid numbers one week after insecticide applications, but predator populations recovered 8-12 weeks after application.

Dimethoate applications for control of cereal aphids in winter wheat caused 85% and

60% reductions in numbers of total arthropods one and two weeks after treatment, respectively; among the arthropods, spiders were affected severely with a 90% reduction one week after treatment and the effect was still observable with a 76% reduction six weeks after treatment (Vickerman and Sunderland, 1977). Matcham and Wawkes (1985) reported that deltamethrin at 7.5 g ai/ha for control of cereal aphids, Sitobion avenae,

Metapolophium dirhodum and Rftopalosidim padi in winter wheat resulted in 30% reduction of total polyphagous predators consisting of carabids, staphylinids and spiders.

Similarly, dimethoate, for control of cereal aphids, was reported to cause significant decreases in numbers of carabids, staphylinids and spiders. Basedow (1985) working on the impact of deltamethrin (7.5 g ai/ha) on carabids, staphylinids and Inyphiid spiders in arable fields, showed that carabids were not affected by the insecticide applications,

61 whereas numbers of staphylinids were reduced. However, they noted that Inyphiids were the most affected arthropods, with 92% mortality, and populations did not recover for 6 weeks. Stark (1992) compared impacts on non-target invertebrates of a neem-based insecticide formulation and chlorpyiiphos in turfgrass. He showed that untreated control plots had 3.5 times more carabids and staphylinids than the chlorpyriphos-treated plots while neem insecticides did not have any effects on the beetle populations. He also showed that chlorpyriphos caused 80% and 70% reduction in the numbers of spiders one and three weeks after treatment respectively, but spider populations recovered in six weeks after treatment. Whitford et al. (1987) reported that significant reductions of spider populations occurred due to carbaryl and fenvalerate applications in com. Culin and Yeargan (1983) stated that a single application of carboftiran caused significant reductions in spider numbers, however they concluded that the application had short-term effect for only fourteen days. Gerard and Akkerhuis (1993) detected a 72% decrease in numbers of spiders in 24 hours after deltamethrin application in wheat and no recovery of spider populations occurred within six days of the application. He considered that either deltamethrin could have a repellent effect that kept spiders from migrating into sprayed plots, or deltamethrin remained toxic for a long time and so reduced the migration.

Braman and Pendley (1993) observed that chlorpyriphos applications to centipedegrass had short-term adverse effects on ant populations.

A decrease in predator numbers because of insecticide applications can result in a decrease in predation rates on pests and consequently an increase in pest numbers. For instance, isofenphos reduced predation of turnip root fly, Delia floralis eggs by 40%

62 because of decreases in carabid numbers (Andersen and Sharman, 1983). Shires (1985)

showed that reductions of carabid and staphylinid populations because of DDT

applications resulted in a resurgence of cereal aphid populations in spring wheat. Coaker

(1966) reported that removal of carabids and staphylinds by aldrin and dieldtin caused

rapid upsurge of cabbage root fly, Erioischia brassicae numbers. Cockfield and Potter

(1984) stated that carabids, staphylinids, spiders, and ants accounted for 75%

consumption of sod webworm, Crambus and Pedosia spp., in Kentucky bluegrass turf.

They found that one chlorpyriphos application resulted in no consumption of these pests by predators for one week and reduced predator egg mortality for three weeks. Dunning et al. (1982) evaluating the impacts of contact insecticides onAphis fabae and M yzus persicae in the sugar-beet crop, showed that acephate, demethion, and demeton-s-methyl reduced Pterostichus melonarius (Carabidae) numbers and caused subsequent increases in the aphid populations.

Some herbicides can also be toxic (Boiteau,1984), cause longer instar durations, reduction in egg production and viability (Subagja and Snider, 1981) and have repellent effects on soil arthropods (Agnello et al. 1986). Powell et al. (1985) showed that herbicides reduced carabid and staphylinid populations significantly, however, they considered that reduction of predator population was a response to destruction of weed cover that may have provided a better habitat with favorable humidity for larvae that have a thinner cuticle and more susceptible to desiccation. Similarly, Brust (1990) reported that the four herbicides, atrazine, simazine, paraquat and glyphosate did not have any acute or chronic toxic effects on carabids, but carabids migrated out of fields as a response to

63 destruction of plant cover. However, Boiteau (1984) stated that chlorbromuron and linuron were toxic to carabids and believed that, because of frequent cultivations in untreated plots, the abundance of carabids were not attributable to weed cover. They concluded that sensitivity to herbicides limited biological control ability of carabids.

A major purpose of present study was to determine the effects of comprehensive chemical pest management including insecticide, fungicide and herbicide applications as a single entity and also its individual components separately on surface-dwelling arthropod predators.

Materials and Method

The field, experimental design and treatments were described in Chapter 2.

Sampling

Four pitfall traps placed randomly within the rows in each plot were used to determine the relative abundance of surface-dwelling predators. Each pitfall trap was composed of one large plastic cup 10 cm deep, with 9 cm diameter opening, and one smaller plastic cup, 3.5 cm deep with 6 cm diameter. The large cups were installed into soil so that the opening was level with the soil surface and stayed in the soil throughout the season to avoid disturbing soil from the removal and reburial of the cups. The smaller cups were filled with 5% glycerol + 95% water solution and placed in to the large cups for arthropod sampling. The pitfall traps were covered by plastic lids between sampling periods to protect the traps from rain and debris. The lids were removed from all the traps in all plots approximately 24 h after pesticide treatments and left open for approximately 48 h for sampling. The smaller cups containing trapped arthropods were then removed from

64 the traps, covered with plastics lids and labeled. Arthropods collected in each cup were then transferred to vials containing 70% alcohol and identified to the taxa Aranae

(spiders), Formicidae (ants), Staphylinidae (rove beetles), Carabidae (ground beetles) and

Cicindelidae (tiger beetles). Data on numbers of predators were analyzed by analysis of variance. Group means were compared as whole and at each sampling date using t-tests.

Results

The seasonal abundances of arthropod predators in 1994 and 1995 are summarized in

Table 3.1. Population fluctuations of predators are illustrated on figures 3.1- 3.10.

Carabidae Tground beetles)

In 1994, ground beetles were more abundant in the fungicide and herbicide-treated plots by 14.7% and in the insecticide-treated plots by 8.6% over the control plots.

However carabid numbers were 26.5% fewer in the full spectrum pesticide-treated plots compared with the those in the control plots. The only significant differences observed were between the fungicide and herbicide and the insecticide-treated plots (p<0.01) on

14 August after first esfenvalerate applications (Fig.3.1). The increase in the carabid numbers in the fungicide and herbicide-treated plots on this sampling date was attributed to high numbers of aphids in these plots relative to other experimental plots.

In 1995, carabids were affected by pesticide applications more severely than in the previous year and populations were lower by 26.9%, 36.5% and 41% in the insecticide, the fungicide and herbicide-treated, and the full spectrum pesticide treated plots than in the control plots. Carabid populations did not differ significantly between treatments at any sampling dates (Fig 3.2).

65 Staphylinidae ( rove beetles)

In 1994, rove beetles were more abundant in the insecticide-treated plots by only 4%

and in full spectrum pesticide-treated plots by only 7.7% than in the control plots, whereas they were decreased by 40% in the fungicide and herbicide-treated plots relative to those in the control plots.. The populations of rove beetles declined steadily during the season in the control plots and fluctuated irregularly in the other plots.

In 1995, rove beetle populations in the full spectrum pesticide-treated plots were significantly smaller than in those in the control plots by 60.1% (p< 0.01) and significantly smaller in both the insecticide and the fungicide-treated plots than in those control plots by

53.3% (p<0.01). The statistical analysis at the sampling date level did not reveal any significant treatment effects on the population of rove beetles.

Cicindelidae ( tiger beetles')

In 1994, pesticide applications reduced populations of tiger beetles by 67% in the full spectrum pesticide-treated plots and by 55.1% in the fungicide and herbicide-treated plots compared with the control plots. Nevertheless, the beetle numbers were 22.4% higher in the insecticide-treated plots than in the control plots. The increase of the numbers of the tiger beetle populations in the insecticide-treated plots could be a response of the beetles to prey shortage resulting in increase in beetle activity.

In 1995, tiger beetle populations were significantly more abundant (p<0.01) in the control plots than in any treated plots. The beetle numbers were reduced by 63.3%,

68.63% and 84% in the fungicide and herbicide-treated plots, the insecticide-treated plots and the full spectrum pesticide-treated plots, respectively relative to in the control plots.

66 Araneae ( spiders)

In 1994, spider populations were significantly higher in the fimgicide and herbicide

treated-plots than in the insecticide-treated plots ( p<0.01). Insecticide applications caused

a 20.7% decrease in spider numbers in the insecticide-treated plots compared with those

in the control plots whereas the full spectrum pesticide, and the fimgicide and herbicide

treatments led to a 14.% and 61.12% increases in numbers of spiders. The increases in the

populations of spiders in the full spectrum pesticide and the fungicide-treated plots was

probably a response to abundance of aphids in these plots.

In 1995, spider populations were significantly higher in the control plots than in any of

the treated plots ( p<0.01). There were significantly fewer spiders in the full spectrum

pesticide-treated plots than in the insecticide-treated plots ( p<0.05) and in the fungicide

and herbicide-treated plots ( p<0.01). Pesticide treatments led to reductions in spider

populations by 27.6 % in the fungicide and herbicide-treated, 36.6% in the insecticide-

treated, and 54.7% in the full spectrum pesticide-treated plots. Spider populations were

significantly higher ( p<0.01) in the control plots compared with any pesticide treated

plots on 18 July, 12 and 24 August.

Formicidae (antsJ

In 1994, the pesticide treatments reduced ant populations in the insecticide-treated plots by 7% compared with the control plots, whereas they caused increases in the full spectrum pesticide-treated and the fimgicide and herbicide-treated plots of 24.9% and

46.9%, respectively. Significantly fewer ants were observed in the full spectrum

67 pesticide-treated plots than in the control plots (p<0.01) on 25 August after the second

esfenvalerate treatment.

In 1995, ant populations were reduced by 18% in the insecticide-treated and by 30% in

the full the spectrum pesticide-treated plots, while they were increased by 27% in the

fungicide and herbicide-treated plots. Significantly more ants occurred in the fungicide and

herbicide-treated plots than those in the insecticide-treated plots ( p<0.01) on July 6 and

in the control plots ( p< 0.05) on 18 July.

Discussion

It can be concluded from the extensive literature that most pesticide, and particularly

insecticide, applications lead to a significant reduction in carabid populations (Critchley,

1972, Gholson et al. 1978, Los and Alen, 1983). Andersen and Sharman (1983) detected

significant reductions in carabid populations because of isofenphos applications. Phorate

(Brust et al. 1986), chlorpyriphos (Stark, 1992), aldrin, dieldrin and DDT (Coaker,

1966), thionazin (Critchley, 1972), dimethoate (Vickerman and Sunderland, 1977; Powell

et al. 1984), acephate, demethion and demeton s-methyl (Dunning et al. 1982) were reported to cause decreases in carabid populations. These support my 1995 data in which the carabids were more abundant in the control plots than in any of the treated plots indicating that insecticide applications decreased the overall abundance of carabids.

Herbicides have been shown to cause decreases in carabid populations (Brust, 1990;

Boiteau 1984; Powell et al. 1985). However, in my experiment, no reductions in numbers were observed in the fungicide and herbicide-treated plots in 1994. By contrast, the carabids were more abundant in the fungicide and herbicide-treated plots than in the

68 control. This was probably a response of carabids to higher prey populations (Bryan and

Wratten, 1984, Honek, 1988, Matcham and Hawkes, 1984). This was clearly observed in

the populations sampled on 14 August 1994 when carabid populations were at a peak and

were significantly higher in the fimgicide and herbicide-treated plots than in the

insecticide-treated plots (p<0.001). A similar peak was observed in the full spectrum

pesticide-treated plots when aphid populations increased. It took fi"om 4 to 6 weeks for

carabids to recover fi-om insecticide stress following applications (Terry et al. 1993).

However, since insecticides were applied frequently in the insecticide and the full spectrum

pesticide-treated plots, there was not enough time for carabids to recover fi-om the application stress within the growing seasons. Nevertheless, the increases in carabid numbers on 12 September 1994, 17 days following the last insecticide application could indicate that the beetle populations were recovering.

ffigher carabid populations in the insecticide-treated plots than those in the control in

1994, despite fi-equent insecticide applications throughout the season may have been because insecticides led to hungrier beetle populations with a resultant higher chance of being trapped (Dixon and McKinlay, 1992). Alternatively, the insecticides enhanced the activity of the carabids, by leading them to aggregate in response to prey patches that had fallen to the ground. This was probably the main reason why the carabids were more abundant in my insecticide-treated plots than in the controls because there was high numbers of aphids in the same pitfall traps in which the carabids were caught. Moreover, it can be concluded that since pitfall traps measure a combination of populations and

69 activity, they may not be the most effective tools (Chiverton, 1984) to measure pesticide

effects on carabid populations.

Fungicides may not have any direct toxicity to carabids (Powell et al. 1984). However,

they may affect them indirectly by destroying organisms that carabids feed on (Aerisches

and Potts, 1990). This could have contributed to the decreases in carabid populations in

the full spectrum pesticide-treated plots compared with the insecticide treated plots in

which the beetle populations were higher despite the same insecticide applications.

Another indirect effect of fungicides could be that the applications protected plants from defoliation by Altemaria solani and provided better canopy cover, and hence a more favorable humidity that would attract more carabids compared to defoliated plants in the insecticide-treated and the control plots. This could be another reason for increase of carabids in the fungicide and herbicide-treated plots. Nevertheless, my suggestion would be that prey populations were the main factor influencing carabid populations, and their abundance in the fungicide and herbicide-treated plots and in the insecticide-treated plots was a response to prey densities but in different ways. Lower numbers of carabids in the full spectrum pesticide-treated plots than those in the controls, however could indicate that pesticide applications had cumulative effects and reduced the beetle numbers.

Staphylinids are often sensitive to insecticide applications (Andersen and

Sharman, 1983; Coaker, 1966; Cockfield and Potter, 1984; Matcham and Hawkes, 1985;

Stark, 1992; Terry et al. 1993). Cockfield and Potter (1984) reported that the insecticides, bendiocarb and trichlorfon and chlorpyriphos caused only short-term effects on staphylinid populations in Kentucky bluegrass and they returned to normal levels in two weeks in

70 bendiocarb and trichlorfon and in six weeks in chlorpyriphos-treated plots. By contrast,

Vavrek and Niemzyk (1990) showed that summer treatments with isofenfos reduced staphylinid population densities for 43 weeeks after treatment. In my experiment there was no sufBcient evidence on the effects of the treatments on staphylinid population densities because their populations fluctuated irregularly in pesticide-treated plots and declined steadily in control plots that made it difiScult to compare in 1994. However, the increases in staphylinid population densities, after the last insecticide applications in the insecticide-treated plots, in both years may have indicated that they were recovering from insecticide stress.

The peak staphylinid populations reached 4 August 1994 in the insecticide-treated plots supports the hypothesis that insecticides result in hungrier and consequently more active predators, because aphid numbers were lower in those plots compared with control and staphylinids were more likely to be trapped.

Staphylinid population densities were smaller in all treated plots than in control plots.

The fewest staphylinids were collected from the traps in the full spectrum pesticide-treated plots. This could indicate that pesticides were cumulative in their effects. The decline in staphylinid poulation densities in the fliU spectrum pesticide-treated plots on 23 July 1994 might have been a response to paraquat application.

Spiders are another predatory group that can be affected by insecticide applications

(Gerard and Akkerhius, 1993, Basedow et al. 1985, Brown et al. 1983, Culin and

Yeargan, 1983, Dinter and Poehling, 1995, Mansour, 1987). Braman and Pendley (1993) found significant reductions in spider populations after chloropyriphos applications to

71 centipedegrass. Carboftiran was reported to cause short term effects on spider populations

(Culin and Yeargan, 1983). Gerard and Akkerhius (1993) showed that deltamethrin could

reduce spider catches by 72% in pitfall traps within 24 hours, and remain toxic to them for

long time. Similarly, Basedow et al. (1985) fbimd that deltamethrin caused 92% mortality

in spider populations and its effect lasted for six weeks. Single surface application of

isazofos reduced spider populations significantly for six weeks (Terry et al. 1993). There

is evidence fi’om my experiments that insecticide applications suppressed spider

populations significantly in 1995 because spider population densities were significantly

lower in the full spectrum pesticide-treated and the insecticide-treated plots than those in the control.

Spider populations were higher in the fungicide and herbicide-treated plots than in any

other plots. This was probably an aggregational response of the spiders to high aphid numbers in those plots (Reichert and Lockey, 1984; Sunderland et al. 1986). There was no sign of spider population recovery after 16 days following last insecticide application in

1994. However, there was an increase in spider numbers in the insecticide-treated plots after 13 days, following the last insecticide application on 24 Angust 1995. Despite that, I would not consider that this was a recovery sign fi-om insecticide stress, because the same increase of populations at a higher rate was observed in the control plots.

Sudden decreases in spider numbers in the fungicide and herbicide-treated and the full spectrum pesticide-treated plots on 23 July 1994 was probably because of the paraquat application. The aggregational response of spider populations to higher patches of aphids in the fungicide and herbicide-treated plots compared with the control plots and

72 relatively higher aphid numbers in the full spectrum pesticide-treated plots compared with the insecticide treated plots could have masked any likely fungicide and/or herbicide effects on spiders in 1994. However, in 1995, when aphid numbers were substantially lower in all plots than the previous year, the effects of herbicides were observed because the spider population densities were significantly lower in the fimgicide and insecticide- treated plots than in the control. Spiders were also significantly lower in the full spectrum pesticide-treated plots than in the insecticide-treated plots, indicating the fungicide and herbicide applications contributed to further reductions in numbers of spiders in the full spectrum pesticide-treated plots.

Cicindelid numbers were very low in all of the experimental plots. Terry et al. (1993) believed that because cicindelid numbers were usually low, they would not provide meaningful data to observe the effects of a pesticide such as chlorpyriphos in their experiment in peanut fields. This might be a reason for very limited information on the effects of pesticides on cicindelid populations. However, in my experiments there was a clear treatment effect on cicindelids, showing that all treatments caused significant reductions in cicindelid numbers compared with the control.

Formicids were the dominant arthropod group captured in the pitfall traps. Mack

(1992) reported that chlorpyripfos adversely affected ant populations. Terry et al. (1993) showed that isozofos application caused a significant reduction in ant populations for ten weeks in peanut fields. The insecticide applications were detrimental to ant populations in my experimental plots. Ants require considerably more time than other predators to

73 recover from insecticide stress (Terry et al. 1993). There was no sign of any ant population recovery in my experiment in 1994 and 1995.

Like many of the other predators, ants were more abundant in the fungicide and herbicide-treated plots than in any other plots. This could be because of either a response of ant population to high prey population densities or to the greater honeydew production by aphids that ants utilize as a nutrient source (Way and Khoo, 1992) in those plots.

In conclusion, all of the predator populations were affected by pesticide applications in some ways. Prey abundance seemed to be the main factor influencing predator populations in 1994. The fimgicide applications caused an increase of predator populations indirectly by providing more prey for the predators (see also chapter 2). When there were less prey in the plots in 1995, the effect of pesticides on predators were more obvious.

74 Taxa Treatments Full Spectrum Insecticides Fungicides and Control Pesicide Herbicides

1994 Carabidae 0.845±0.131 1.250±P.163 1.321±0.203 1.155+0.130 Staphylinidae 0.321±0.075 0.309+0.075 0.178±0.053 0.297+0.080 Cicindelidae 0.035±0.020 0.131±0.404 0.059± 0.025 0.107±0.047 Formicidae 4.000±0.437 2.988+0.350 4.702±0.367 3.202±P.287 Araneae 0.845±0.144 0.585±0.095 1.190±0.191 0.738+0.111

1995 Carabidae 0.946±p.208 1.089+0.202 0.875±0.162 1.491+0.255 Staphylinidae 0.053±0.0024 0.062+0.026 0.062±0.022 0.133±0.034 Cicindelidae 0.026+0.015 0.053±0.021 0.062+0.022 0.169±0.060 Formicidae 0.830+0.109 0.964+0.144 1.509±0.160 1.187+0.151 Araneae 1.518+0.166 2.125±0.251 2.437+0.023 3.357±0.263

Table 3.1: Seasonal abundance of predators per pitfall trap in tomato fields in 1994 and 1995 (mean ± SEM).

75 CaraUdK 1994

— O- ■ - Full spectrum pesticide — -O — Insecticides — A— • Fung/herb. C o n tro l

4 J

4 -■

3 .5 -•

2 2.5 -■

IJ -

0 .5 Ï .

21-Jun g.Jul 23-Jul 4-Aug 14 -A u g 25-Aug 12-Sep

Sam pling Date

Carbaiyl Esfenvalerate Endosulfan Esfenvalerate Esfenvalerate 7 July 22 July 3 August 13 August 24 August

Figure 3.1: Seasonal variations in carabid populations (mean + SEM) resulting from pesticide treatments, 1994.

76 CaraUdie 1995

— ^ . Full spectrum pesticide ■ Insecticides - 6 - Fimg/herb. —H------C o n tro l

B. 4

0 IS-Jun 22-Jim 6-Jul 18-Jul 30 -Ju l 12-Aug 24-Aug Sam plig Date

Carbaryl Esfenvalerate Endosulfan Esfenvalerate Esfenvalerate 21 June 5 July 17 July 29 July 11 August

Figure 3.2; Seasonal variations in carabid populations (mean ± SEM) resulting from pesticide treatinents, 1995.

77 1994 Stafihyliiiidae

— o-- - Full spectrum pesticide —-Q— Insecticides — &— Fmg/herb. — M— Control

1.4 r

1.2 •

B i 2 0.8 .

0.6 •

0 .4 ■

4 -A u g ug 2 5 -A u g14-A 12-Sep Sam pling Date

Esfenvalerate EndosulfanCarbaiyl Esfenvalerate Esfenvalerate 7 July 22 July 3 August 13 August 24 August

Figure 3.3: Seasonal variations in staphylinid populations (mean + SEM) resulting from pesticide treatments, 1994.

78 1995 Staphylinidae

- - - o- - - Full spectrum pesticide — Q— Insecticides - A - Fmg/herb. H C o n tro l

0.6

0 .4 -

Ci Si OJ

0.2

^ ^ «I

a— - . y

22-Jun 6-Jul 18-Jul 30-Jul 12-Aug 2 4 -A u g

Sam pling Date

Carbaryl Esfenvalerate Endosulfan Esfenvalerate Esfenvalerate 21 June 5 July 17 July 29 July 11 August

Figure 3.4: Seasonal variations in staphylinid populations (mean + SEM) resulting from pesticide treatments, 1995.

79 1994 Cicindelidae

— o- - - Full spectnim pesticide — -O— Insecticides — A - - Fung/herb. — M ------C o n tro l

0 .7

0.6

S- 0.4

OJ ••

■s •'/

0 U— 2 I-J u n 8 -M 4-A u g 14 -A u g 25-Aug 12-Sep

Sam pling Dates

Carbaryl Esfenvalerate Endosulfan Esfenvalerate Esfenvalerate 7 July 22 July 3 August 13 August 24 August

Figure 3.5: Seasonal variations in cicindelid populations (mean + SEM) resulting from pesticide treatments, 1994.

80 1995 Gcindslidae

— ^ . Full spectrum pesticide — -O— Insecticides — Fung/herb. C o n tro l

0.8

0 .7 •

0.6

0.5

S .0 .4 •

6-JuI 18-Jul 12-Jul 24 -A u g

Sampling Dates

Carbaiyl Esfenvalerate Endosulfan Esfenvalerate Esfenvalerate 21 June 5 July 17 July 29 July 11 August

Figure 3.6; Seasonal variations in cicindelid populations (mean ± SEM) resulting from pesticide treatments, 1995.

81 Araneae 1994

— o- - - Full spectrum pesticide — -o-— Insecticides - 6 - Fing/berb. — * ------C o n tro l

3 .5

0 .5 ■■ D---

21 -Ju n g .Ju l 4 -A u g 14-Aug 25-Aug 12-S ep Sam pling Dates

Carbaiyl Esfenvalerate Endosulfan Esfenvalerate Esfenvalerate 7 July 22 July 3 August 13 August 24 August

Figure 3.7; Seasonal variations in ground-dwelling spider populations (mean + SEM) resulting from pesticide treatments, 1994.

82 Araneae 1995 — o- - • Full spectrum pesticide ' ■ — Insecticides - A - Ftatg/herb. — M— Control

2 4 Ï I / A Z 3 -5 0- /•/ 2

0 — IS-Jun 22-Jun 6-Jul 18-Jul 30-Jul 12-A ug 2 4 -A u g Sam pling Dates

Carbaiyl Esfenvalerate Endosulfan Esfenvalerate Esfenvalerate 21 June 5 July 17 July 29 July 11 August

Figure 3.8; Seasonal variations in ground-dwelling spider populations (mean ± SEM) resulting from pesticide treatments, 1995.

83 1994 Formicidae

— o - - - Full sp ec tru m p e sticid e — - ■ — In se cticid e — Fung/herb. —H------C o n tro l

c .

0 21-Jun 8-Jul 23-Jul 4-Jul 14-Aug 25-Aug 12-Sep Sam pling Dates

Carbaiyl Esfenvalerate Endosulfan Esfenvalerate Esfenvalerate 7 July 22 July 3 August 13 August 24 August

Figure 3.9: Seasonal variations in ant populations (mean ± SEM) resulting from pesticide treatments, 1994.

84 1995 Fonmcidae

— O- - - Full spectrum pesticide — -O— Insecticides — - Fung/herb. Control

2.5

2 0--

?.. a —

0 .5

IS -Ju l 12-Aug 24-Aug Sampling Dates

Carbaiyl Esfenvalerate Endosulfan Esfenvalerate Esfenvalerate 21 June 5 July 17 July 29 July 11 August

Figure 3.10; Seasonal variations in ant populations (mean ± SEM) resulting from pesticide treatments, 1995.

85 CHAPTER 4

THE EFFECTS OF CHEMICAL MANAGEMENT OF PESTS, DISEASES AND

WEEDS ON FOLIAGE-DWELLING PREDATORS, COMMONLY FOUND ON

TOMATOES

Introduction

Pesticide applications to control arthropod pests can often cause both direct and

indirect adverse eftects on predatory arthropod populations in agroecosystems. Direct

efiects are involved with mortality of predator populations and the indirect eftects to

shortage of food because of killing their hosts and also because they may move out of

pesticide treated fields (Newsom, 1967). Roger et al. (1991) reported that at

recommended field rates, carbaryl and malathion caused 100% and 75% mortality of

Coleomegilla maculata, respectively. Pesticide applications can decrease the establishment

rate of predators in fields (Lovei et al. 1991) and often lead to outbreak of pests ( Croft and Brown, 1975) by eliminating one of the most important natural control factors fi’om fields.

Not all foliage-dwelling predators show the same sensitive response to the pesticides.

Dinkins et al. (1971), according to their different responses to various insecticides, ranked

86 Orius insidious as highly sensitive; C. maculata, highly to moderately sensitve; while the

spider complex had only low sensitivity. Whalon and Eisner (1982) showed that malathion

and diazinon applications to different fields against Illnoia pepperi decreased both O rius

sp. and C. m aculata populations significantly, seven days postapplication. Click and

Lattimore (1954) reported that spider populations were not affected by DDT in cotton

fields. On the other hand, Thomas et al. (1990) showed that numbers of spiders, Erigone atra and Oedothorax apicatus were reduced by 89% and 82% respectively, after

deltamethrin applications to winter wheat fields.

Less is known about direct and indirect effects of fungicide and herbicide applications on populations of natural enemies. Roger et al. (1994) reported that benomyl caused mortality in C m aculata populations at recommended field rates and reduced their predation efficiency by inactivating the predators and consequently lowering consumption rates of prey.

This experiment was aimed at determining all the effects of pesticides commonly used in tomato agroecosystems on commonly found foliage-dwelling predators.

Materials and Methods

Field, experimental design and treatments have been described in Chapter 2.

Sampling

In 1994 and 1995 respectively two 12 cm x7 cm and four 6 cm x 3.5 cm yellow sticky traps were used to catch coccinellids and anthocorids. The insecticide application dates were used to construct a timetable on when to replace and collect the sticky traps. Sticky traps were left in the fields for 7-10 days, collected one day before each insecticide

87 treatment and replaced one day after the treatment. The first sampling was made before

the insecticide application. The traps collected fi'om the experimental plots were placed in

storage bags, labeled and kept in a fi-eezer until counting and identification.

Foliage-dwelling spiders were counted on the tomato plants. Four randomly-selected

tomato plants in each plot were searched for spiders. Spider counts were made once every week, beginning 27 June 1994 and 15 June 1995, until harvest.

Data were subjected to an analysis of variance and analyzed based on seasonal abundance using sampling dates. Means were compared using t-test.

Results

All of the treatments, except the fungicide and herbicide applications in 1995, reduced the seasonal abundance of foliage-dwelling predators (Table 4.1). The seasonal variations in the populations of foliage-dwelling predators resulting firom pesticide treatments were summarized in Figures 4.1-4.6.

C. maculata

In 1994, The insecticide treatments caused a 35.2% reduction in C. m aculata populations in the insecticide-treated plots compared with those in the control plots.

Despite the same insecticide treatments in the full spectrum pesticide-treated plots, the numbers of the beetles were reduced by only 22% in those plots. This was attributed to the response of the beetles to higher aphid populations in the full spectrum pesticide- treated plots than in the insecticide-treated plots. The numbers of coccinellids were 6.6% lower in the fungicide and herbicide-treated plots than those in the control plots. Despite the highest numbers of aphids in the fungicide and herbicide-treated plots, a 6.6%

88 reduction in the beetle populations indicated some detrimental effects of the fungicide and

herbicide applications on the coccinellid. The numbers of beetles were lower in all of the

pesticide-treated plots than in the control on 21 June, 4 July, and 17 July. Coccinellid

numbers in the full spectrum pesticide-treated and in the fungicide and herbicide-treated

plots exceeded those in the control plots on 29 July and thereafter while those in the

insecticide-treated plots remained lower.

In 1995, C .m aculata was 10% more abundant in the fungicide and herbicide treated

plots than in the control plots, the beetle populations were lower by 4.1% in the full

spectrum plots and 1% in the insecticide treated plots compared with the control.

Anthocorids

In 1994, anthocorid populations differed significantly in all of the treated plots

(p<0.01). The numbers of anthocorids were 55.8% (p= 0.0002) less in the full spectrum

pesticide-treated and 31.6% (p= 0.036) less in the insecticide-treated plots than in the control plots. Although the anthocorid numbers were lower in the fungicide and herbicide- treated plots by 26.2% they did not differ significantly fi’om those in the control plots

(p=0.074). In terms of sampling dates, the only significant difference in populations was observed on 20 September after the fifth insecticide (esfenvalerate) application; when the predator populations in the full spectrum pesticide-treated plots were significantly less than in the insecticide treated plots (p=0.039), and the fungicide and herbicide-treated plots (p=0.006) than in the control plots (p=0.012).

In 1995, anthocorid populations responded similarly to the pesticide treatments as in

1994. The pesticide applications reduced the numbers of these predators by 38.8, 32.2 and

89 13.5% respectively in the full spectrum pesticide, the insecticide, and the fungicide and

herbicide treated plots compared with control.

The numbers of the anthocorids in the fungicide and herbicide-plots remained lower

than in the control plots for most of the season in 1994 and 1995. However, more

predators occurring in the fungicide and herbicide-treated plots, on 29 July 1994 and 10

Aug. 1995; this may be a response to increasing aphid numbers.

Spiders

In 1994, foliage-dwelling spiders were affected severely by all pesticide treatments.

The full spectrum pesticide-treatments and the insecticide treatments caused 70.9% and

67.5% reduction respectively in spider populations from the controls (p=0.00l8 and p=0.003 respectively). The spider populations were only 44.6% lower in the fungicide and herbicide-treated plots than in the control plots.

In 1995, the effects of the pesticides, particularly the insecticides were more severe on the spider populations. Significant reductions of 91.4% in the full spectrum pesticide- treated (p=0.0000) and of 89.2% in the insecticide-treated plots (p=0.000l) compared with the control plots were observed in spider populations. The numbers of the spiders were 37% lower in the fungicide and herbicide-treated plots than in the control plots.

Discussion

The response of three groups of foliage-dwelling predators to the pesticides demonstrated that not all predator groups were affected to the same degree of severity by the pesticide applications. The foliage-dwelling predators showed various responses to type and application intensity with different pesticides. Anthocorids and spider populations seemed

90 very sensitive to all of the pesticide applications. By contrast, C.m aculata showed only a very low sensitivity. These results do not agree with those of Dinkins et al. (1971) where

C maculata were shown to be highly to moderately sensitive and spiders had a low sensitivity to malathion. However, Dinkins et al. (1971) ranked Orius insidious

(Hemiptera: Anthocoridae) as highly sensitive to pesticides.

Many studies have indicated severe toxicity of insecticides to C. m aculata. Lecome and Smilowitz (1980) reported that carbaryl was very toxic to C. m aculata. Roger et al.

(1991) reported that malathion led to 100% mortality inC. maculata populations in a laboratory study. Yun and Ruppel (1964) showed that the minimal dosage of carbaryl to control cereal leaf beetle caused a rapid elimination of C. m aculata after 48 hours, whereas Guthion, malathion and dieldrin caused mortalities of 97, 87 and 3%, respectively. Coats et al. (1979) noted that C.maculata showed moderate susceptibility to pyrethroid insecticides because of their selectivity in favor of coccinellids. On the other hand, Whitford and Showers (1988) reported that carbaryl and fenvalerate applications used to control European com borer in com fields had only minimal impacts on C. maculata populations This partly agrees with my results, where C. m aculata populations were not affected severely by the pesticide applications, except in the insecticide-treated plots in 1994 where a 35% reduction in the beetle populations was detected. However, the same insecticide applications in the full spectrum pesticide-treated plots, which had more aphids than did the insecticide-treated plots, did not cause a significant reduction in the coccinellid populations. Therefore, the reductions in the coccinellid populations in the

91 insecticide treated plots could have been because of the insuflScient aphid numbers to sustain higher coccinellid populations, rather than reductions by pesticides.

Whalon and Eisner (1982) showed that Orius sp. populations were eliminated seven days after malathion applications in blueberry fields. These predators seemed very susceptible to any kind of pesticides. Like anthocorids, the spiders consistently showed a very highly sensitive response to the pesticide applications. Brown et al. reported that methyl parathion was very toxic to the spider, Erigom spp. Culin and Yeargan (1983) showed that azinphos methyl and dimethoate led to significant reductions in spider populations in an alfalfa ecosystem. Thomas et al. (1990) reported that deltamethrin caused a 89% reduction in spider populations. This is consistent with my results in large reductions in spider populations in the insecticide-treated plots, ranging fi’om 67.5 to

91.5%. Pesticides can also afifect spiders by accumulating on their webs and thus poison spiders directly when they are on the webs or repel them thereby exposing them to predation and contaminated leaf surfaces (Samu et al. (1992).

Fungicides and herbicides are known to be less detrimental to foliar predator populations. For instance, Livingston et al. (1978) reported that spider populations were not influenced by fungicide applications including benomyl and chlorothalonil. Redcliffe

(1976) showed that fungicides increased Orius sp. populations by 1% and spiders and coccinellid larvae by 3% in alfalfa., these increases were attributed to the response of the predators to increasing aphid populations. Stam et al. (1978) stated that the herbicides, dinoseb and MSMA had no effect on C. maculata and spider populations. However,

Mansour (1987) showed that several fungicides and herbicides caused 10-40% mortality in

92 spiders in a laboratory and considered this effect as harmless. In my experiments, the application of fungicides and herbicides caused reductions in the predator populations, except in C maculata populations in 1995. The insecticides combined with fungicides and herbicides intensified the adverse effects on the foliar predator populations.

93 Taxa Treatments

Full spectrum p. Insecticides Fungicide and Control

Pesticide Herbicides

1994

C.maculata 1.639±0.267 1.083+0.223 1.556±0.324 1.667±0.278

Anthocorids 3.000±0.396* 4.639±0.515** 4.972±0.517 6.806+1.290

Spiders 0.093±0.029* 0.104±0.031* 0.177±0.039 0.322+0.065

1995

C.maculata 1.208+0.130 1.250+0.156 1.396+0.161 1.260±0.151

Anthocorids 4.458±0.336* 4.969+0.531* 6.875+0.579 7.292±0.700

Spiders 0.035±0.017* 0.440+0.023* 0.258±0.042 0.410+0.085

* significantly different fi’om control at p<0,01 level.

** significantly different fi’om control at p<0.05 level.

Table 4.1: Seasonal abundance of the foliage-dwelling predators in tomato fields, 1994 and 1995 (mean+SEM).

94 1994 Cmaculata

— O- - - Full spectnan pesticide — D— Insecticides —- 6 —- Fung/herb. — M— Control

3 .5

B. 2.5

0 .5 -■

21-M 2-Aug12-.Aug 23-Aug 7-Scp Sam pling Date

Carbaryl Esfenvalerate Endosulfan Esfenvalerate Esfenvalerate 7 July 22 July 3 August 13 August 24 August

Figure 4.1: Seasonal variations in C. maculata populations (mean + SEM) resulting from pesticide treatments, 1994.

95 1995 Cmaculau

— ^ . . Full spectnan pesticide — -O— Insecticides — A- - Fun^erb. Control

3.5 ■■

0.5 -•

4-Jul 17-Jul 29-Jul 11-A ug 2 2 -A u g Sam pling Date

Carbarv’l Esfenvalerate Endosulfan Esfenvalerate Esfenvalerate 21 June 5 July 17 July 29 July 11 August

Figure 4.2: Seasonal variations in C maculata populations (mean ± SEM) resulting from pesticide treatments, 1995.

96 1994 Anthocoridae

— ^ Full spectrum pesticide — -O Insecticides - A- Fung/herb. — K ------C o n tro l

14 ■

12 ••

6-Jul 21 -Ju l 2 -A u g 12-A ug 23 -A u g

Sam pling Date

Carbaryl Esfenvalerate Endosulfan Esfenvalerate Esfenvalerate 7 July 22 July 3 August 13 August 24 August

Figure 4.3; Seasonal variations in Anthocoridae populations (mean ± SEM) resulting from pesticide treatments, 1994.

97 1995 Anthocoridae

— O- - - Full ^ectrum pesticide — O— Insecticides — A - - Fmgfherb. — * — Control

15 Ë. e a I 10

0 21-Jun 4-Jul 17-Jul 29-Jul 11-Aug 22-Aug Sampling Date

Carbaryl Esfenvalerate Endosulfan Esfenvalerate Esfenvalerate 21 June 5 July 17 July 29 July 11 August

Figure 4.4: Seasonal variations in Anthocoridae populations (mean + SEM) resulting from pesticide treatments, 1995.

98 1994 Foliage Spiders

— O- - - Full spectrum pesticide — O— Insecticides - A - Fing/herb. X— Control 0.8

0 .7 ■

0.6 ■

0.5 ■■

=-0.4

------

27-Jun 5-Jul 12-M 22-Jul 2-Aug 10-Aug 18-Aug 2 6 -A u g Sam pling Date

Carbaryl Esfenvalerate Endosulfan Esfenvalerate Esfenvalerate 7 July 22 July 3 August 13 August 24 August

Figure 4.5: Seasonal variations in foliage-dwelling spider populations (mean + SEM) resulting from pesticide treatments, 1994.

99 199S Foliage Spiders

— O- - - Full spectrum pesticide — D- — Insecticides - A - Fun^erb. — x — Control

1.6

1.2 ■

0.8

0.6 ••

0 .4 -■

0.2

0 15-Jun 30-Jun 9-Jul 23-Jul16-Jul 29 -Ju l lO -A ug Sampling Date t t t t Carbaryl Esfenvalerate Endosulfan Esfenvalerate 21 June 5 July 17 July 29 July

Figure 4.6; Seasonal variations in foliage-dwelling spider populations (mean ± SEM) resulting from pesticide treatments, 1995.

100 CHAPTERS

THE EFFECTS OF CHEMICAL PEST, DISEASE AND WEED MANAGEMENT

PRACTICES ON THE TROPHIC STRUCTURE OF NEMATODE POPULATIONS IN

TOMATO AGROECOSYSTEMS

Introduction

Nematodes have important but different roles in agroecosystems. Some species can have direct and indirect effects on plant growth and yield, through their activities in the rhizosphere where many interactions occur between organisms (Freckman and Caswell,

1985). The first effect occurs by nematodes feeding on plants (Yeates et al. 1993) whereas the latter occur by mineralizing organic matter through their interactions with soil microorganisms (Hendrix et al. 1986; Ingham et al, 1985), regulating its decomposition rate (Parmele and Alton, 1986), contributing to nitrogen dynamics (Didden et al, 1993), and they can also can attack insects and help to control them (Hara and Kaya,1983;

Rovesti and Deseo , 1990; Zimmerman and Cranshaw, 1990).

There have been many studies indicating changes in nematode populations fi’om agronomic practices. Tillage is one of the practices that can cause changes in nematode abundance and trophic structure in agroecosystems (Minton, 1986). For instance.

101 Juma and Mishra (1987) reported that conventional tillage increased the numbers of

bacterivorous nematodes. Parmele and Alston (1986) showed that minimum tillage

caused increases in populations of fungivorous nematodes. However, populations of plant

parasitic nematodes were found to be higher in no-tillage systems (Foitnum and

Karlen, 1985). Application of organic amendments to soil is another practice that can

cause reductions in plant parasitic nematode populations (Kabana et al. 1986; Mojtahedi

et al. 1993).

Publications on the effects of agricultural practices such as pesticides have dealt

mostly with their direct impact on plant parasitic nematode populations by killing them

(Castro et al. 1991; Cowles and Villani, 1994; Townshend, 1990; Yeates and Prestidge,

1986) and indirectly suppressing their antagonists ( Jaflfe and Mclnnis, 1990; Pullen et al.

1990). However, there is abundant evidence of the non-target effects of agrochemicals on

plant parasitic nematode populations. Bohlen and Edwards (1994) showed that inorganic

fertilizers increased numbers of plant parasitic nematodes in com. On the other hand, Chen

et al. (1994) reported that inorganic nitrogen caused decreases in plant parasitic nematode numbers in tobacco because it increased numbers of fungal endospores suppressive to the nematodes. The herbicides, veraolate, metribuzin and trifluralin increased Heterodera glycines egg numbers by 37 to 134% in soybeans (Kraus et al. 1982). A combination of the herbicide alachlor and the nematicide fenamiphos led to late season resurgence o f H. glycines populations (Sipes and Schmitt, 1989). Schmitt et al. (1983) considered that herbicides stimulated the hatch of H. glycines eggs, changed plant physiology that made the plants more attractive to the nematodes, and enhanced their fecundity that in turn

102 resulted in increase in their numbers. A mixture of herbicides, chlormetoxyml and

thiobencarb-simetryne increased the numbers of plant parasitic nematodes in rice

agroecosystems (Ishibashi et al. 1983). Thiocarbamate herbicides enhanced the infection

rate of Meloidogyne hapla to alfalfa because they altered the root epidermis composition

to favor the nematodes (Griffin and Anderson, 1979). H. schachtii numbers was found to

be higher in cycloate-treated sugar beet fields, because the herbicide stimulated the

hatching activity of the nematode (Abivardi and Altman, 1978). On the other hand,

Browdie et al. (1994) reported that the herbicide mixture acifluorfen plus bentazon

decreased H.glycines populations. They believed that the herbicides caused root injury that limited root growth, thereby providing fewer infection sites for the nematodes or the herbicides altered the plant’s physiology and led to release of root exudates that affected the nematode’s host-finding behavior or were toxic to nematodes.

Little is known about the direct and indirect impacts of pesticides on nematode community structure and the relative abundance of fi'ee-living and other nematodes in agroecosystems. The herbicides, tetrapron had stimulative effects in fi'ee-living nematode populations on jute (Biswas and Mishra, 1987). Ishibasi et al. (1983) reported that a chlormetoxynil and thiobencarb-simetryne mixture increased numbers of fungivorous and plant parasitic nematode numbers and decreased predatory nematode numbers. Schmitt and Corbin (1981) found that plant bacterivorous nematode numbers increased in alachlor- treated soybeans compared to untreated ones. Mahn and Kastner (1984) reported that herbicide applications favored populations of bacterivorous nematodes because they led to increase in certain microbial populations in response to the immature

103 death of plants. Smolik (1983) found that microbial feeding nematodes were less sensitive to the nematicides, aldicarb, terbufos and carbofuran than predatory nematodes, and reduction in numbers of predatory nematodes caused an increase in numbers of bacterivorous nematode.

There is a need for a much clearer understanding of the impacts of pesticides on the structures of nematode communities in agroecosystems. In the study reported here, I assessed the response of nematode communities to various mixture of insecticides, fungicides and herbicides applications in tomato agroecosystems.

Materials and Method

Field, experimental design and treatments have been described in Chapter 2.

Sampling

Nematode populations were sampled in the first week of June, July, August and

September in 1995 using a core which took soil 5 cm diameter by 15 cm deep. Eight randomly collected soil cores were taken fi'om each plot, four between the rows and four in the rows. 20 g soil sub-samples fi'om eight cores were extracted in Baermann funnels for 48 hours (McSorley and Walter, 1991). Nematodes were counted to bacterivorous, fungivorous, plant parasitic and carnivorous groups. Nematode numbers were analyzed using analysis of variance to assess the main treatment effects. Group means were compared on basis of seasonal abundance and the abundance on individual sampling dates using t-tests.

104 Results

The application of the different pesticides to soil growing tomatoes led to quite different responses in the different nematode trophic groups (Table 5.1 and Figures 5.1-

.4). In general, the pesticide treatments suppressed bacterivorous nematode populations over the whole season. The full spectrum pesticide treatments (p=0.05) and the fungicide and herbicide (p=0.02) applications caused significant reductions of 47% in bacterivorous nematode populations compared with control, whereas the insecticide applications reduced numbers only by 31.8%. Relatively fewer bacterivorus nematodes occurred in the full spectrum pesticide-treated and the fungicide and herbicide-treated plots in the June sampling. This was attributed to the effects of the preemergent herbicide, trifluralin applications because no any other pesticides had been applied at that sampling date. The populations of bacterivorous nematodes did not show any significant differences in pesticide-treated plots at any sampling date despite increasing populations in the control plots. However, the total bacterivorous nematode numbers over the four samplings were significantly fewer in the full spectrum pesticide-treated plots (p= 0.008), the insecticide- treated plots (p= 0.009) and the fungicide and herbicide-treated (p=0.026) plots than in the control plots.

In general, the fungivorous and carnivorous nematode populations were an order of magnitude lower than those of bacterivorous and plant parasitic nematodes in all of the treated plots. The pesticide treatments reduced the numbers of fungivorous nematodes in the full spectrum-treated plots significantly (p= 0.0049) by 73.2%, in the insecticide treated plots (p= 0.0054) by 73.9%, and in the fungicide and herbicide treated plots ( p=

105 0.022) by 59% of the control. The populations of fungivorous nematodes were significantly lower in the full spectrum pesticide-treated plots in July (p=0.037), in the insecticide treated plots in September (p= 0.046), and in the fungicide and herbicide treated plots in July (p= 0.027) over those in the control plots.

Carnivorous nematode populations were 33% and 55.6% lower in the full pesticide- treated plots and the insecticide-treated plots, respectively, whereas they were 15.1% higher in the fungicide and herbicide-treated plots than in the control plots. No significant treatment effects were detected on carnivorous nematode populations at any sampling date.

Plant parasitic nematode populations reacted differently to the pesticide treatments.

Their populations were significantly higher in the full spectrum (p= 0.010) and the insecticide (p= 0.0013)-treated plots by 82.8% and 117.8%, respectively, whereas they were only 21.8% higher in the fungicide and herbicide-treated plots than those in the control plots. At the sampling date level, plant parasitic nematode numbers were significantly higher in the full spectrum pesticide-treated plots, than in the fungicide and herbicide treated (p= 0.019) and the control (p= 0.049) treated plots in July.

Discussion

The various applications of pesticides in my experiment had significant effects on the structure of the nematode communities. The nematodes in each trophic group reacted differentially to the various treatments. The bacterivorous nematodes were the most dominant group in all of the experimental plots. Their populations showed sensitive responses to all pesticides. Fewer bacterivorous nematodes that occurred in the fungicide

106 and herbicide-treated plots than in the control plots in June sampling was probably a response of these nematodes to trifluralin applications. Similarly the decrease of their number in July sampling could be because of the paraquat applications. Fewer bacterivorous nematodes in the fungicide and herbicide-treated plots than in the insecticide treated plots may have indicated that the fungicides and herbicides had more adverse efiects on the nematode populations than did the insecticides. The mechanisms of the decreases in the bacterivorous nematode population densities in response to pesticides that have been reported here and elsewhere are unknown. It could be either because of direct toxicity of the pesticides to the nematodes. Ingham et al. (1986) reported that carbofuran reduced bacterivorous nematode populations significantly. Or it could be because of the indirect efiects of the pesticides by reducing the numbers of bacteria that those nematodes feed on. Mahn and Kastner (1985) and Schmitt and Corbin (1981) showed that herbicides increased numbers of bacterivorous nematodes because they increased numbers of microorganisms. However, my result conflicts their findings because the bacterivorous nematode numbers were significantly fewer in the fungicide and herbicide-treated plots than in the control and the insecticide-treated plots. Nevertheless, it was difficult to infer fi’om my data whether the fungicides or the herbicides were responsible for reducing the bacterivorous nematode densities in the plots treated with these chemicals. It was clear that pesticides used in my experiment were cumulative in their efiects because the lowest bacterivorous nematode populations were recovered fi’om the full spectrum pesticide-treated plots.

107 Like the bacterivorous nematodes, the fungivorous nematodes were more abundant in the control plots than in any of the pesticide-treated plots. Again, this could be either because of direct toxicity of the pesticides to those nematodes (Smolik, 1983) or indirect by reducing amounts of hyphae of fungi serving as food sources for fungivorous nematodes (Ingham et al. 1985). My evidence conflicts with of Biswas and Mishra

(1978) who claimed that herbicides did not have any effect on populations of saprophytic nematodes. A steady decline in the fungivorous nematode populations was observed in the fungicide and the herbicide-treated plots. A plausible explanation could be that the fungicides and herbicides affected the hyphae of fungi that served as food source for the fungivorous nematodes and the soil in the fungicide and herbicide-treated plots could not sustain larger populations of fungivorous nematodes. In this sense, the same explanation would be true for the decreases in the fungivorous nematode populations in the insecticide-treated plots. The decrease of the fungivorous nematode populations in the fungicide and herbicide-treated plots in the August sampling was probably not attributable to paraquat treatment, because the same decrease in the nematode populations was observed in the other plots and it was probably a response of the nematodes to environmental changes.

Carnivorous nematodes were the least abundant nematode groups in my experimental plots. They have been shown to be sensitive to the nematicides, aldicarb, terbufos and carbofuran (Smolik, 1983), inorganic fertilizers (Bohlen and Edwards, 1994), copper

(Parmelee et al. 1993) and herbicides, chlormetoxynil and thiobencarb-simetryne mixture

(Ishibashi et al. 1983). Neither trifluralin nor paraquat had any effects on the carnivorous

108 nematode populations in my experimental plots. The significantly lower carnivorous nematode populations in the full spectrum pesticide-treated plots than in the control in the

July sampling was not attributable to the paraquat applications because the carnivorous nematode populations were higher in the fungicide and herbicide treated plots than in the control plots at that date. Likewise, the fungicide applications probably did not have any efiects on the carnivorous nematode populations. High carnivorous nematode populations in the fungicide and herbicide-treated plots could not be a response to high prey densities in those plots, because the bacterivorous and fungivorous nematode numbers were lower than in the control. The decrease of their numbers in the firll spectrum and the insecticide- treated plots may not have been a response to prey numbers because plant parasitic nematode populations were high and could have sustained carnivorous nematode populations. Probably because they are sensitive to pollutants, carnivorous nematodes may have been affected adversely by the insecticide applications to the insecticide and the full spectrum pesticide treated plots.

The plant parasitic nematode populations were higher in the full spectrum pesticide- treated and in the insecticide-treated plots than in the control. Various reports have indicated that agrochemical applications can cause an increase in plant parasitic nematode population densities in some ways (Bohlen and Edwards, 1994, Kraus et al, 1982,

Parmelee et al. 1993, Schmitt et 1983). Several factors can lead to that increase. For instance, pesticides can induce a stimulatory effect on the hatch or larval growth of nematodes. Bishwas and Mishra (1978) found that direct and residual effects of tetrtapion applications onto jute, had stimulatory efiects on the populations of the plant parasitic

109 nematodes, Hoplolaimus indicus, Helicotylenchus digonychus and Tylenchorynchus sp.

Similarly, Smith and Corbin (1981) and Schmitt et al. (1983), studying the population

dynamics of Heterodera glycines in response to herbicides in soybeans, showed that

alachlor increased population densities of that species. They considered that alachlor

stimulated the egg-hatch and thereby increased H. glycines populations. Alachlor alone

and in combination with a nematicide , fenomiphos, enhanced the survival of H. glycines

(Bastian et al. 1984). Abivardi and Altman (1984) showed similar hatch-stimulating

activities caused by cycloate on H. schachtii in sugar beet. Trifluralin increased the

numbers of H. glycines as many as five times five days postapplication (Riggs and Oliver,

1982). Vemdate, trifluralin and metribuzin stimulated the development oîH. glycines and

increased larvae numbers by 43 to 82% in 14 days after planting, and by 37 to 134% from

planting to harvest.

Although it is unknown whether or not the pesticides in my experiment had such

stimulating effects on the plant parasitic nematodes, it could have been one of the reasons

that caused significantly higher plant parasitic nematode populations in all of the

pesticide-treated plots compared with the control. Besides, such effects, if any, would be

attributed to the insecticide (carbaryl) applications because a significant increase in plant

parasitic nematode populations was observed in the July sampling one week after the

application. Clarke and Shepherd (1966) found that a breakdown product of nabam

(Sodium ethylene bisdithiocarbamate), thiuram monosulphite had hatch-stimulating

activity. Since carbaryl is in this pesticide class, it might also exert such effect. However, this needs further investigation.

1 1 0 Schmitt et al. (1983) and Bostian et al. (1984) reported that nematicide applications

caused resurgences in H. glycines populations. They believed that nematicides led to

nematode-free root development, which allowed plants develop larger roots and in turn

sustain larger nematode populations later in the season. Assuming that the herbicide

applications had suppressed the plant parasitic nematode populations in the fungicide and

herbicide-treated plots, the increases in the numbers of the nematodes after July could fit

the hypothesis by Schmitt et al. (1983) and Bostian et al. (1984). However, despite the

same herbicide treatment, no such effect was observed in the full spectrum-pesticide treated plots.

Alteration of plant physiology by herbicides may be another factor that can increase plant parasitic nematode populations by making the host plant more suitable for the nematodes (Schmitt et al. 1983). Abivardi and Altman (1978) believed that cycloate had toxic effects on sugar beet roots that thereby increased chemical and nutrient gradient exudates from roots consequently increased H. schachtii numbers in response to these changes. It is difficult to confirm such effects of herbicides on the plant parasitic nematode populations in my experiment, because the plant parasitic nematodes in the full spectrum pesticide-treated and in the fungicide and herbicide-treated plots showed different responses to the same herbicide applications. On the other hand, such effect would be attributed to the insecticide applications because they increased the plant parasitic nematode numbers in both the full spectrum pesticide-treated and the insecticide-treated plots.

Ill Biolo^cal changes in the soil environment might cause increases in plant parasitic nematode populations (Schmit et al. 1983). Schmitt and Corbin (1981), without further speculation, stated that increases in populations ofH. glycines and some other small plant parasitic nematodes could have been because of effects of herbicide on bacteria because they found that bacterivorous nematode populations increased in alachlor-treated soybeans. On the other hand, Bostian et al. (1984) stated that alachlor provided protection for nematodes by affecting microorganisms adversely and enhancing of the survival of H. glycines juveniles. Miller and Taylor (1967) demonstrated that the fungicide maneb prevented microbial destruction ofH. tabaccum eggs, so that treated soils had more viable eggs six months after treatment than in untreated soils. The fungicide nabam, was associated with soil microflora and inhibited cellular decomposition that resulted in increase ofH. tabaccum because cellular decomposition was affecting larval emergence

(Miller, 1967).

Fungicides can inhibit numbers of entomophagous fungi in soil. Pullen et al. (1990), working with benomyl, and Jaflfe and Mclnnis (1990), working with carbendazim showed that fungicides reduced the growth, germination and tube elongation of the fungus,

Hirsutella rhosiliensis, a parasite of ring nematode, Criconemella xenoplax. Suppression of fungal activity on nematodes could have caused increases in plant parasitic nematode populations in my experimental plots and might have been one of the main reasons contributing to the plant parasitic nematode population increases because fungi are important in regulating nematode populations (Mankou, 1980) by trapping (Stirling et al.

1979) and parasitizing and killing them (Kerry, 1977; Mankou, 1979)

112 Another reason for the increase in numbers of plant parasitic nematodes in response to pesticides could be that predaceous mites can feed on nematodes (Imbriani and Mankau,

1983) and the insecticide applications might have reduced microarthropod populations that fed on the nematodes. Consequently, reduction of predation pressure by mites would increase plant parasitic nematode numbers.

Carnivorous nematodes can play an important in regulating plant parasitic nematode populations (Small, 1978). A decrease in carnivorous nematode numbers could lead to significant increase in plant parasitic nematode numbers. Parmelee et al. (1993) found that decrease in carnivorous nematode populations in response to copper application resulted in higher plant parasitic nematode numbers. The highest plant parasitic number was recovered fi-om the soils that had least carnivorous nematode populations in the experimental plots. In my experiment, the plant parasitic nematode populations were higher in the full spectrum pesticide-treated and the insecticide-treated plots whereas the carnivorous nematodes were low in these plots relative to the control. This might have indicated that the carnivorous nematodes had a great influence on the plant parasitic nematode populations and the insecticides might have been responsible for the increase of the plant parasitic nematode numbers.

113 Trophic groups Treatments

Full spectrum Insecticides Fungicides and Control

Pesticide Herbicides

Bacterivores 19.66±2.47* 25.51±2.9 19.88±2.15* 37.44+5.50

Fungivores 0.390±P.10* 0.395±0.13* 0.385±0.15** 1.460±0.30

Plant parasitic 9.123±1.26* 10.87±1.41* 6.08±1.36 4.99±0.75

Carnivores 0.355±0.13 0.235+0.07 0.610±0.18 0.53±0.19

* significantly firom control at p<0.01 level.

** significantly fi"om control at p<0.05 level.

Table 5.1: Seasonal abundance of nematode trophic groups in experimental plots in response to pesticide treatments (mean+SEM).

114 Bacterivorous Nematodes

c— Full spectrampesticide — o— Insecticides -- a --Fung/herb. — gg— Control

70 -•

6 0 ' ■

5 0 -•

~ 0-

10 ■

June July August September t t t t t Carbaryl Esfenvalerate Endosulfan Esfenvalerate Esfenvalerate 21 June 5 July 17 July 29 July 11 August

Figure 5.1: Seasonal variations in bacterivorous nematode populations (mean + SEM) resulting from pesticide treatments.

115 Fungivorous Nematodes

3.5 •

M 2.5 ■■

0.5

June Jufy August September

Carbaryl Esfenvalerate Endosulfan Esfenvalerate Esfenvalerate 21 June 5 July 17 July 29 July 11 August

Figure 5.2; Seasonal variations in fungivorous nematode populations (mean ± SEM) resulting from pesticide treatments.

116 Carnivorous Nematodes

1.4

1.2 ■

0.8 i

0 .4 ■

June Jufy August September

Carbaryl Esfenvalerate Endosulfan Esfenvalerate Esfenvalerate 21 June 5 July 17 July 29 July 11 August

Figure 5.3: Seasonal variations in carnivorous nematode populations (mean + SEM) resulting from pesticide treatments.

117 Fiant Parasitic Nematodes

16 ■

14 -■

S 12 -■

M 10 -■

6 /

--Î

June July A ugust September

Carbaryl Esfenvalerate Endosulfan Esfenvalerate Esfenvalerate 21 June 5 July 17 July 29 July 11 August

Figure 5.4: Seasonal variations in plant parasitic nematode populations (mean + SEM) resulting from pesticide treatments.

118 CHAPTER 6

AN ECONOMIC COMPARISON OF THE USE OF PESTICIDE APPLICATIONS ON

PROCESSING TOMATOES IN OHIO

Introduction

There is increasing public concern about the use of pesticides use to increase agricultural productivity in agroecosystems. Pesticides are not only costly inputs for farmers but also they can cause potential damage to the environment. One of the reasons that tend to increase pesticide use is the farmer’s attitude toward alternative or biological methods to keep crops pest free. However, pests that are not present or are below certain population limits, that are high enough to cause damage threshold should not affect the quality and quantity of productivity (Fox and Weersink, 1995). In this case, the cost benefit ratio would be higher and pesticide applications could actually cause monetary losses.

In many cases, insecticides have been shown to prevent yield losses to arthropod pests and led to higher profits. Singh and Singh (1991) evaluated the economics of the insecticides, phosphamidon, monochrotophos, dimethoate, formothion, chloropyriphos and quinalphos against jassid, Apheliona maciilosa infestations on soybean. Glycine max.

119 They showed that all of the insecticides provided higher yields and profits than the controls. The cost : benefit ratio ranged firom 1 : 3.39 for monochrotophos and 1: 10.88 for phosphamidon treated soybeans. Karel and Ashimogo (1991) reported that at least two dimethoate applications was necessary to reduce yield losses significantly to insect pests in soybeans and common beans, Phaseolus vulgaris The yield losses were shown to be

32% in soybeans and 47% in the common bean compared with when they were not treated by the insecticide. The cost ; benefit ratio was 1 : 3 : 2 in soybeans and 1 : 6 : 5 in the common beans. Edelson et al. (1989) showed profitability of cypermethrin applications to control onion thrips,Thrips tabaci on onions.Allium cepa in two growing seasons. Eight cypermethrin applications provided $ 3,180 whereas only four early and only four late treatments and no treatment provided $2,799, $2,116 and $1,155, respectively in 1986.

The same treatment schedules provided less profit in 1987 when weather conditions were unfavorable for onion production and monetary return was $659 for eight, $380 for only four early, $9 for only four late and $-95 for no cypermethrin treatments. Mishra (1984) evaluated the economics of the insecticides, decamethrin, fenvalerate and permethrin against tomato finit borer, Heliothis armigera, whitefly, Bemicia tabaci and jassid,

Amrasca bigutiula Ishida in tomatoes. Decamethrin and permethrin increased yields by

95.12% (20,000 kg/ha) and 82.44% (18,700 kg/ha) and provided higher net profits than fi’om untreated control tomatoes. Wightman et al. (1995) compared the economics of host plant resistance and insecticides application against Helicoverpa armigera in chickpea plants. They noted that insecticide applications were critical to protect the chickpea enterprise for susceptible varieties. Although resistant varieties provided profits when no

120 insecticide was applied, susceptible varieties provided higher yield and profit than

resistant varieties. Trumble and Morse (1993) compared the economics of chemical and

biological, and combination of both control measures %ainst the two spotted spider mite,

Tetranychus urticae in strawberries. The control of the two spotted spider mite by the

predaceous mite Phytoseiulus persimilis at 4,050/ha provided a $2,170-$4,310 higher net benefit compared with the untreated control while applications of abamectin and fenbutatin-oxide provided $5,062-515.802 and $4,401-59.146, respectively. However, abamectin in combination with the releases of the predaceous mites resulted in higher net benefit (56,890-$19,705) than abamectin application alone. Herdt et al. (1984) showed an economic comparison of three management strategies against insects in rice. They noted that one insecticide application based on a threshold level provided higher benefit than two applications and more than two applications because production costs increased as the numbers of sprays increased.

Fungicides are another integral input into modem agricultural productivity that helps to reduce yield losses to diseases and protect the product quality (Babcock et al. 1992).

Takele et al. (1993) reported that fungicide applications provided higher net returns in navel oranges compared with acaricide and nematicide applications and different N fertilizer levels, because they increased the yield, grade and fiuit size.

Wilcut et al. (1987) compared the economics of herbicide use and cultivations in peanuts, Arachis hypogaea infested with Texas panicum, Panicum texamm, sicklepod.

Cassia obstufolia, and pitted morning glory, Ipomoea lacunosa. They reported that herbicide applications with benefin, alachlor, dinoseb and naptalam as preplant and at

121 ground cracking provided a better net return and suflBcient weed control level than four cultivations. However, they achieved the highest yields with herbicide applications in combination with two timely cultivations of weeds in peanut fields. Snipes et al. (1984) showed that herbicide applications to in cotton provided better net financial return than four cultivations and two cultivations plus two hand-hoeing. Cultivations alone resulted in negative net return for four years (-$200 - -S450/ha). Fluometuron applications used alone provided the highest yield ($280 - $420/ha). However, diuron and trifluralin applications required three and two supplementary cultivations to provide positive net returns and efficient weed control levels.

In many cases, reduced pesticide use can provide better returns. Mann et al. (1991) reported that reduced quantities of insecticides to control aphids in winter wheat, gave better financial returns compared to the full recommended rates. Marshall et al. (1991) showed that minimum herbicide applications are more feasible economically than heavier applications in com. Schweizer et al. (1988) studied the profitability of four weed management systems consisting of conventional tillage alone, conventional tillage with minimum, moderate and intensive levels of herbicides in barley, com, pinto bean and sugar beet sequences grown in rotation for four consecutive years. They reported that the conventional tillage, plus intensive herbicide use, system provided the lowest gross retum over the four years period for the barley-com-pinto bean-sugar beet and com-pinto bean- sugar beet-barley and stated that despite the fact that there was two times more weeds in reduced herbicide use systems than the intensive herbicide use systems, the weeds did not reduce the yields and the gross retums. Lybecker et al. (1988) reported that the minimum

122 herbicide system had $440/ha/4 years higher gross retum than the intensive herbicide weed management system. Chase and Duffy, (1991) stated that despite lower yields, com grown in reduced-chemical use 6rms provided competitive financial retums compared to conventional pesticide-based systems. Davies and Whiting (1989) showed that a 50% reduction in the application rates of early-post emergence herbicides, diflufenican + isoproturon in wheat and a spring post-emergence application of metsulfuron-methyl + mecoprop in wheat and spring barley was profitable, because of reduced cost, while providing sufiBcient level of weed control.

The purpose of this part of my study was to provide information on the effects of different chemical pest management strategies on the yield of tomatoes and generate data to which is relevant to the economics of pesticide use in Ohio.

Materials and Method

The field, experimental design and the treatments have been described in Chapter 2.

Sampling

Tomato plants in a five feet row in each plot were taken fi'om the soil. All tomato fiuits on those plants were picked, graded to marketable and cull categories, and weighed to estimate the yield per hectare.

Economical Analvsis

All of the fixed and variable costs were calculated. Fixed costs were estimated, based on a five years farm record and production budget and included spraying, bed preparation, planting and general maintenance costs. The costs consisted of the costs of fertilizer, transplants, pesticides, interest on operating capital and hoeing labor. The costs were

123 recorded in both in 1994 and 1995. The operating cost of equipment was calculated on the basis of the new cost of machinery, salvage value, insurance and useful life expectancy.

Labour was calculated based on a wage of $7.8 per hour. Harvest and marketing costs were calculated based on transportation, commission, packaging and labour. Market value of tomato was estimated based on average prices of processing tomatoes in September.

Data were analyzed ANOVA to separate the treatment effects on net financial retum per hectare.

Results

There were variations between yields and consequently on profits, in all plots in 1994 and 1995. Average yields in 1994 were 80.2 T/ha in the full spectrum pesticide-treated,

52.1 T/ha in the insecticide-treated, 81.4 T/ha in the fungicide and herbicide-treated, and

50.5 T/ha in the control plots. The yields were lower in 1995 than in 1994 and were 13.2

T/ha in the full spectrum pesticide-treated, 13.4 T/ha in the insecticide-treated, 14.4 T/ha in the fungicide and herbicide-treated and 10.6 T/ha in the control plots. The weather conditions, as well as the insect and disease incidences were the influencing factors in the yields. The 1994 growing season was the most profitable for tomatoes and the yields in all pesticide treatment plots were profitable (Fig. 6.1). The fungicide-treated plots (the full spectrum pesticide-treated and the fungicide and herbicide-treated plots) provided higher yields and profits than the control and the insecticide-treated plots. The profit fi-om the plants grown in the fungicide and herbicide-treated plots was higher than those fi-om the full spectrum pesticide-treated (p=0.78), in the insecticide treated (p= 0.07) and in the control (p= 0.081) plots. Despite similar fungicide applications and yield, higher profits

124 were obtained from the tomatoes grown in the fungicide and herbicide-treated plots than from those in the full spectrum pesticide-treated plots based on the costs of insecticide applications. Similarly, despite the higher yields from the plants in the insecticide-treated plots than from those in control plots, slightly lower (p= 0.93) profits were obtained from the insecticide-treated plots than from the control plots.

Heavy rains after a drought period affected the yields adversely in 1995. The profits from every treatment were substantially less than in 1994. This year despite the highest yield in the full pesticide spectrum treated plots, monetary losses of $585 per hectare occurred because production costs exceeded retum (Fig. 6.1). Again, the fungicide and herbicide treatments higher yields and profits. The insecticide treatment provided more profit compared to the controls but less profit than the fungicide and herbicide treatments.

Discussion

Tomatoes are hosts for a wide range of arthropod pests and susceptible to a variety of diseases that reduce quality and quantity of tomato production. Pesticide use is one of the major pest control measures to prevent losses to insect pest, disease and weeds. There are some reports showing the effects of pesticides on tomato yield; there are a few studies showing the economics of insecticide use against major insect pests; however there is little information available indicating a comprehensive economic analysis of pesticide use in tomatoes.

This study revealed that knowing the increase in yield in response to pesticide applications was not enough to conclude that pesticide applications were profitable. In other words, despite slightly higher tomato yields in the insecticide-treated plots than in

125 the control in 1994, I obtained lower profits fi'om the insecticide treatments because

production cost was higher in the insecticide-treated plots than in the control plots

because of the insecticide applications and there was little yield loss attributable to

insects. I found that only a slight différence between the fimgicide and herbicide-treated plots and the full spectrum pesticide-treated-plots with respect to yield. However, the differences with respect to profit were much greater between those treatment plots with respect to profit because the production costs in the full spectrum plots exceeded those in the fungicide and herbicide-treated plots. The loss of profit in the insecticide-treated plots compared to the controls could not be attributed to failure of the insecticides to control the pests, because pests of tomatoes significantly fewer in the insecticide-treated plots than in the control. This was probably because the pests were mainly foliar and seedling feeding pests, did not cause direct damage to the finit and the plants could sustain some pest populations without decreasing the yield and hence did not need extra protection.

Quaglia et al (1993) had similar results with deltamethrin, that had reduced green peach aphid populations on tomatoes but had no effect on finit production. Lambdin and

Snodderly (1984) reported similar results with dimethoate and permethrin that were effective againist potato aphids but did not affect the yield. In such cases, probably a reduction of the dose used to half (Poehling, 1988) or on the evidence of my experiment not to spray may provide better profits.

On the other hand, many reports have indicated that insecticide applications can lead to increases in tomato yield and provide higher profits. Walgenbach et al. (1992) demonstrated the importance of insecticides to optimize profits in tomato production.

126 They found that the tomato fruitwonn, Helicoverpa zea caused $3,385 and $941/ha losses in Western North Carolina when insecticides were not sprayed. They concluded that that even small insect damage could result in large losses in net income, and that insecticides, endosulfan, methomyl or esfenvalerate were required at, 10 d and 5 d intervals in dry and wet seasons respectively, for profitable tomato production.

Wiesenbom et al (1990) showed that up to eight insecticide applications to control tomato pinworm,Keiferia lycopersicella were profitable in California; however after the ninth application, production costs exceeded profit. Kennedy et al. (1983) demonstrated, that under severe finit feeding insects, tomato finitworm, H. zea and tobacco budworm,

H. virescence , pressure from 27 to 84% reduction in tomato yields. Nevertheless , none of these pests were present in my experimental plots. In this regard, if any fiuit-damaging tomato pests had been present in my experiment, probably even a low fiuit insect pressure

(Walgenbach and Estes, 1989) might have caused more economic losses and consequently the profit figure quoted would have been difterent. The insecticide application schedule in this study targeted most possible pests that could cause damage to tomatoes. In this case, it can be concluded, that although insecticide applications reduced aphid and flea beetle population densities significantly, in the absence of fiuit-feeding pests, the applications were not justified and caused losses in monetary retum.

In 1995, when the weather conditions did not favor tomato production, the insecticide treatments provided better yields and increased monetary retums than the control because the plants in the control plots were under severe pressure from both pests and disease.

Little difference was observed between the full spectmm pesticide-treated plots and the

127 fimgicide and herbicide-treated plots with respect to yield. However, the differences with

respect to profit were greater between those plots because production costs exceeded the

retums due to the use of insecticides in the full spectrum pesticide-treated plots and

$585/ha monetary losses occurred.

Yield losses due to early blight was much greater than the losses to insect pests in both

seasons. The fimgicide applications were critical to ensure satisfectory yields. There are

conflicting reports on the effects of fungicides on tomato yields. Pemezny et al. (1996) stated that chlorothalonil made substantial contributions to tomato production in Florida.

Kermedy et al. (1983) showed that 92% of tomato yields were lost to diseases when fungicides were not applied in North Carolina. Walgenbach et al. (1989) found 46% of tomato finit was infected withAltemaria solani if not protected with fungicides in North

Carolina. Septoria leaf spot, caused fi'om 40 to 95% yield losses and weekly fungicide applications were necessary to prevent yield losses fi'om the disease. Pemezny (1994) reported that target spot disease of tomatoes caused 30% yield loss and chlorothalonil applications to prevent tomato yield fi'om this disease provided a $3,000/acre net retum in

Florida. On the other hand, Poysa et al. (1993) stated that a 60% foliage infection by

Colletototrichiim coccodes did not affect tomato yields, so fimgicide applications would not have affected yields in Ontario. Brammal et al. (1993) believed that although chlorothalonil applications suppressed early blight incidence, it did not have any effects on yield in Ontario. This was probably because tomatoes are tolerant to certain degree to defoliation by foliar diseases, without decreasing their yields. For instance, Ferrandino and

Elmer (1992) showed that less than 50% of leaf damage to Septoria leaf spot did not

128 reduce tomato yields. However, above 80% defoliation caused significant yield reductions

(Wolketal. 1983).

In 1995, tomato yields were very low relative to the yields in 1994. This was probably

because of the warmer, more humid and wetter weather conditions in 1995, that enhanced

the progress and development of the disease. The foliar defoliation through early blight disease was much more severe in 1995 than in 1994 (see also Chapter 1). Although weekly chlorothalonil applications provided adequate early blight control in 1994, the same application schedule was not effective to control the disease in 1995 and probably more fi-equent applications were needed to maintain yield. The yield losses were 38% and

26% in the control plots in the 1994 and 1995 season respectively relative to the fimgicide and herbicide-treated plots.

My results showed that the fimgicide applications were critical to ensure tomato yields in Ohio in 1994 and 1995. In the drier season (1994), weekly chlorothalonil applications provided adequate level of early blight control, although an approximately 30% defoliation was still observed in the fimgicide and herbicide-treated plots. As was previously discussed, I have assumed that this degree of defoliation did not affect yields. However, in the wetter season (1995) the defoliation was more severe in the fimgicide and herbicide- treated plots than in 1994 and affected the yields in the fimgicide and herbicide-treated plots.

129 I COST □ RETURNS 0PROFIT(LOSS)

4 5 0 0 0 j

4 0 0 0 0 -•

3 5 0 0 0 -■

3 0 0 0 0 -•

I I 2 5 0 0 0 -- i I g 20000 -■

1 5 0 0 0 -■

10000 5 0 0 0 -■ 1 1FuU Insecticides Fung/herb. Control sp ec tru m p e sticid e

T r e a tm e n ts

Figure 6.1; Economic comparisons of the pesticide treatments in 1994.

130 ■ œ s r □ RETURNS B PROnT(LOSS)

8 0 0 0

7 0 0 0

6 0 0 0

5 0 0 0

2 4 0 0 0

2 3 0 0 0

2000

1000

Full spectrum Insecticides Fung/herb. C o n tro l

-1000 I V e a tm e n ts

Figure 6.2: Economic comparisons of the pesticide treatments in 1995.

131 CHAPTER?

GENERAL DISCUSSION

The results of the studies summarized in this dissertation revealed quite variable responses by populations of pests and diseases, predatory arthropods and nematode communities as well as tomato yields and profitability to various pesticide applications.

Tomato production can be threatened by a wide range of insect pests and disease-causing agents. Pests and diseases can reduce the yield and quality of tomatoes, by damaging plants and their finit and high finit quality standards primarily related to insect damage are required. Hence, chemical control measures are usually implemented to prevent yield losses, keep finit fi’ee firom damage and meet high cosmetic standards (Brun, 1981;

Harding, 1971; Lange and Bronson, 1981; Mishra, 1984; Walgenbach et al. 1991;

Walgenbach and Estes, 1992; Wiesenbom et al. 1990).

The insecticide application schedule used in my study targeted all kinds of insect pests

(seedling, foliage and finit pests) according to the stages of the tomato plants at which the specific pest can occur. However, many of the major pests were not present, or they were

132 rare in my experimental plots. Tomato fruitworm,Heliothis zea is one of the major pests

of tomato damaging fruits, that has been often reported in tomato agroecosystems

(Campbell et al. 1991; HoflSnan et al. 1990; Kennedy et al. 1983; Mishra, 1984; Oatman

et al. 1983; Roltsch and Mayse, 1983; Schuster and Price, 1985; Walgenbach et al.

1989;1991). However, this pest was absent from my experimental plots. Likewise, tomato

pinworm, K. lycopersicella (Alvardo-Rodriquez, 1988; Poe and Everett, 1974; Schuster,

1978; Walgenbach et al. 1991), southern armyworm, Spodoptera eridonia, and tomato

fruit borer, N. elegantolis (Schuster and Price, 1985) do not in Ohio. Many other common

tomato pests such as the Colorado potato beetle,L. decemlineata (Ghidiu and Gettting,

1987; Jaques and Loing, 1989), the cabbage looper, T. ni (Oatman et al. 1983; Roltsch

and Mayse, 1983; Schuster and Price, 1985), stink bugs (Heteroptera: Pentatomidae)

(HoflBnan et al. 1987), leafininers (Diptera; Agromyzidae) (Pohronezny et al. 1986; Salas,

1992; Schuster et al. 1979; Schuster and Price, 1985; Trumble and Alvardo-Rodriques,

1993) were very rare in my experimental plots. However, the potato aphid,Macrosiphum euphorbiae (Perring et al. 1988; Walgenbach and Estes, 1992; Walker et al. 1984) and green peach aphid, Myzus persicae (Jensen, 1992; Flint and Klonsky, 1985; Kring and

Schuster, 1992; Lambdin and Snodderly, 1984; Perring et al. 1988), flea beetles

(Drinkwater et al. 1995) and thrips species (Allen et al. 1993; Boutista, 1994; Cho et al.

1995; Heinz, 1992; Navas et al. 1991) were all commonly found.

Populations of aphids, flea beetles and thrips responded differentially to the mixtures of pesticides in my experiment. The insecticide applications reduced aphid and flea beetle populations in the insecticide-treated plots in 1994, however only flea beetle populations

133 were reduced significantly by the insecticide applications in those plots in 1995. My results

agreed with those of Lamdin and Snodderly (1984) and Quaglia et al. (1993) showing

that insecticides reduced aphid populations on tomatoes. Likewise, my results agreed with

those of Vernon and Mckenzie (1991), showing that carbaryl and pyrethroids reduced

populations of flea beetles in potatoes, and wit those of Weiss et al. (1991) showing that

carbaryl, endosulfan and esfenvalerate reduced flea beetle populations on canola.

The aphid and flea beetle numbers were significantly higher in the fungicide and

herbicide-treated plots than in the control plots in 1994. Despite the same insecticide applications as in the insecticide-treated plots, populations of flea beetles were higher in the full spectrum pesticide-treated plots and in the insecticide-treated plots. Several factors were considered to explain these increases in the populations of the aphids and flea beetles in the fungicide and herbicide-treated and the full spectrum pesticide-treated plots.

One explanation was that fungicide and herbicide applications reduced populations of natural enemies of these pests that in turn caused increases in populations of these pests.

This could be one, if not the only, reason contributing to the increase of the populations of these pests in the fungicide and herbicide-treated plots, because populations of anthocorids and spiders were significantly lower in these plots than in the control plots.

Further suppression of the anthocorids and spider populations in the full spectrum pesticide-treated plots, compared with insecticide-treated plots supported this hypothesis.

There is evidence that fungicides can cause resurgence of pest populations by suppressing populations of natural enemies of pests in agroecosystems (Bower et al. 1995, Boykin et al. 1984). Mizzel and Schiflbuer (1994) reported that benomyl caused mortality of aphid

134 natural enemies, Chrysoperla rufilabris (Neuroptera: Chrysopidae), Hipodemia convergence, Cycloneda sorginena (Coleoptera: CoccinelHdae) and parasitic wasp

Aphellidus perpallidus in pecans. Herbicides could have contributed to the reduced populations of natural enemies of pests in my experiment. Stam et al. (1978) reported that the herbicide dinoseb killed the commonly found predators, Coleomegilla maculata,

Erotmocerus holdemani, Orius insidious and Gecoris puncipes in cotton fields. Adams and Drew (1965) showed that aphids resurged in 2,4-D treated oat fields because the herbicide suppressed population of coccinellids.

Many studies have indicated that fungicide applications can suppress fungal parasites of pests (Boykin et al. 1984; Livigston et al. 1978). Redclifife et al. (1976) reported that fungicide applications suppressed entomophagous fungi and in turn enhanced aphid populations in alfalfa agroecosystems. Suppression of fungal parasites of aphids and flea beetles by fungicide applications, might be one of the reasons that enhanced populations of these pests in the fungicide and herbicide-treated and full spectrum pesticide -treated plots.

Pesticide applications can cause changes in physiology (Buscham and Depew, 1990) and nutrient content (Kerns and Gaylor, 1993) of plants. White (1984) reported that fungicides, herbicides and insecticides could change plant nutrient content, increase in nitrogen protein and amino acid levels which can cause resurgence in populations of pests.

Maxwell and Harwood (1960) believed that 2,4-D dimethyl amine changed the nutritional factors in plants and thereby increased the reproductive rates of the pea aphid,

Macrosiphum pisi. Kerns and Gaylor (1993) reported that sulprofos increased threanine

135 and amino acid levels in cotton plants and as a result enhanced population densities of aphids. The occurrence of such effects due to the fungicide and herbicide applications in my experimental plots is only speculative. However, such effects of the fungicides and herbicides could contribute to the increases in populations of aphids and flea beetles.

Insecticides can stimulate fecundity and growth of aphid populations through their chemical activity. Lowery and Sears (1986 a,b) and Gordon and Mceven (1984) reported that azinphosmethyl enhanced green peach aphid reproduction. Jackson and Wilkins

(1985) showed that fenvalerate stimulated the growth of green peach aphid populations.

However there is no clear evidence that fungicides and herbicides could cause such effects on aphid and flea beetle populations in my experimental plots.

Populations of aphids and flea beetles were significantly higher in the fungicide and herbicide-treated plots than in the control plots and relatively higher in the full spectrum pesticide-treated plots than in the insecticide plots. This could be a response of these pests to changes, if any, in the nutrient content. Hatcher et al. (1994) showed that diseased

Rumex species caused greater larval mortality, slower larval development, lower growth, pupation at a lower weight, reduction on the fecundity and longevity and reduction in egg viability in Gastrophyta viridida (Coleoptera; Chrysomelidae) over those on healthy plants. This effect of the diseased plants on the beetle populations were attributed to a reduction in nitrogen content of 50-70% in plants by fungal pathogens. Another approach that could be considered is that since early blight disease caused severe defoliations of tomato plants, it limited the ability of the foliage or plant canopy to sustain larger pest populations. Because early blight starts affecting the lower leaves first (Watterson, 1985),

136 it could affect flea beetle populations most, since they are mostly present on lower canopy.

However, potato aphids prefer to feed on the newer and young leaves. Since aphids were counted only on three new leaflets, defoliation probably did not affect their populations.

However, some nutrient changes such as reduction in nitrogen content could have affected aphid population estimates. Therefore, it can be hypothesized that the fungicides could have had indirect effects on pest populations in the presence of a disease by altering the balance between arthropod pests and diseases in tomato agroecosystems in favor of pest populations.

Resurgence of pest populations after broad spectrum insecticide applications is a well known phenomenon because many insecticides eliminate natural enemies from agroecosystems (Godfray and Chan, 1990; Metcalf 1980; Ripper, 1956). Aphids are one of the pests groups that often resurge after insecticide applications (Kerns and Gaylor,

1993; Poehling, 1988). The resurgence of the aphid populations in my experiment after the endosulfan application in the 1995 growing season could be attributed to the reductions in the aphid predator numbers in the insecticide-treated plots (see also Chapter 3 and 4). It could also be attributed to the short half-life of endosulfan ranging <2 to <8 days

(Walgenbach, 1991) that kept the aphid populations under control for only a short time.

Thrip populations were higher in the full spectrum pesticide-treated and the insecticide- treated plots as well as in the fungicide and herbicide-treated plots, than in the control plots. Thrips can often develop resistance to insecticides readily ( Andoloro et al. 1983,

Immaraju et al. 1990, 1992). Insecticides can stimulate the fecundity of thrips (Morse and

137 Zareh, 1991, Morse and Brawner, 1986). However, I suggest that the increase in thrips numbers in all of my treated plots was due to the decreases in the predator numbers.

Early blight disease of tomatoes caused by Altemaria solani is one of the most destructive factors to tomato plants and may require weekly fungicide applications to protect the plants (Madden et al. 1978; Penypecker et al. 1983; Williamson and Hilty,

1988). The weekly chlorothalonil applications provided suflBcient control of the disease in the 1994 growing seasoiL The same application schedule however did not control progress of the disease in the warmer and wetter 1995 growing season. Despite the fungicide applications, the disease caused 89% defoliation, which was enough to reduce the yield (Wolk et al. 1983) in the fungicide and herbicide-treated plots.

Not all predator groups responded in the same way to the pesticide applications in both years or to the same applications in each year. This is probably due to the complexity of arthropod communities. In spite of this complexity there have been numerous attempts to explain the diverse and often indirect effects of pesticide applications on non-target organisms as well as on yield of crops.

Ground-dwelling predatory arthropods are one of the groups termed non-target organisms in agroecosystems. They can be affected directly by being exposed to pesticide applications or indirectly by eating poisoned pests (Gholson et al. 1978). Decreases in carabid populations are one of the most common consequences of insecticide use in agroecosystems (Andrsen and Sharman, 1983; Coaker, 1966; Crichley, 1972; Dunning et al. 1982; Los and Alen, 1983; Powell et al. 1984; Vickerman and Sunderland, 1977).

Depending on their properties, insecticides can cause 5%-100% reductions in carabid

138 populations (Floate et al. 1989) and it takes 4 to 6 weeks for carabid populations to recover from insecticide stress. Herbicides can also be toxic to carabid beetles (Boiteau,

1984) and have repellent effect on these beetles or affect their activity (Brust, 1990).

Fungicides can affect surface predators by destroying saprophytic organisms they feed on

(Aerisches and Potts, 1990).

Neither in 1994 nor 1995 did total pitfall trap counts show any significant effects of treatments on carabid abundance, despite the higher numbers usually recovered from the control plots. However, on 14 August 1994 a significant increase in carabid populations was observed, in fimgicide and herbicide-treated plots compared with controls. This occurred when aphid populations were high. This was probably the response of carabid populations to higher prey populations in these plots (Bryan and Wratten,

1984;Honek,1988; Matchkom and Hawkes, 1984). Reductions in pest populations, resulting from pesticide applications, leads to hungrier beetle populations that have higher chance of being trapped ( Dixon and McKinlay, 1992). This may be a possible explanation for the lack of significant reductions in carabid populations in the pesticide-treated plots in spite of continuous pesticide applications throughout the season. Moreover, pitfall traps may not be effective tools to measure the effects of pesticides on carabids since they measure a combination of numbers and activity (Chiverton, 1984).

Staphylinid beetles are a predatory group of insects often exposed to pesticide applications (Andersen, Sharman, 1983; Coaker, 1966; Matchkom and Hawkes, 1985;

Stark, 1992; Terry et al. 1993). Cockfield and Potter (1983) reported that bendiocarb and trichlorfon applications affected staphylinid populations severely in the short term but

139 populations returned to normal level in two weeks after the treatments. However, isofenfos reduced staphylinid population densities for 43 weeks after treatment (Vavrek and Niemzyk 1990). Staphylinid beetle populations were aSected adversely by pesticide treatments in my experimental plots. The pesticides could have had additive effects on the staphylinid populations because the lowest rove beetle population occurred in the full spectrum pesticide-treated plots. The rove beetle populations peaked in my insecticide- treated plots on 4 August 1994 which may support the hypothesis that lack of prey can lead to more hungry and active beetle populations because the aphid populations were the lowest in those plots.

Ground-dwelling spiders were one of the predatory arthropod groups that were exposed to the pesticide treatments and showed immediate responses to the pesticide applications (Gerard and Akerhuis, 1993). Braman and Pedley ( 1993) and Gulin and

Yeargan (1983) observed significant reductions in ground-dwelling spider populations resulted from carbofiiran applications. Deltamethrin reduced spider populations by 92% and its effects could last for six weeks (Gerard and Akkerhius, 1993). In my experiments, the spiders responded differentially between the two seasons and their populations were lower in all of the insecticide-treated plots than in the control plots. However, numbers were significantly higher in the fungicide and herbicide-treated plots, where aphids were more abundant. This probably resulted from an aggregational response of spiders to the pest prey patches (Riechert and Lockley, 1984). Nevertheless, in 1995, when the aphid populations were much smaller in all plots, the effects of pesticides were more noticeable on the spider populations. The insecticide treatments were more detrimental to spider

140 populations than were fungicide and herbicide treatments, and the combinations of

pesticides had additive effects.

Mack (1992) reported that chloropyriphos affected ant populations adversely. Isosofos applications reduced ant populations significantly in peanut fields and its effect lasted for ten weeks. Because of periodic insecticide applications, ants can not rebuild their populations. In my experiment, ant populations were significantly higher in the fungicide and herbicide-treated plots. This may be due to greater honeydew production by aphids in those plots that ants utilize as a nutrient source (Way and Khoo, 1992).

Terry et al. (1993) reported that because cicindelids were present in only low numbers, effects of pesticide treatments on their populations could not be assessed. However, the treatments in my experiments caused significant reductions in tiger beetle populations despite their overall low numbers.

Overall, my results revealed that insecticides were more detrimental to surface-dwelling predators than fungicides and herbicides. The insecticide applications probably resulted in hungrier predators that were more likely trapped because their prey was killed. The fungicide and herbicide applications seemed to be detrimental to surface predators because numbers were reduced in the fungicide and herbicide-treated plots However, this reduction in populations of the predators could have been attributed to the herbicide treatments that were toxic to predators and/or repelling them fi"om the herbicide treated plots. Nevertheless, the fungicides caused increases in numbers of the ground predators in response to increasing pest prey populations.

141 Foliage-dwelling predators seem to have been affected more adversely by the pesticide

treatments than surface-dwelling predators. This may be because pesticides are applied

directly onto the foliage which they inhabit. Although the insecticides killed more C maculata in laboratory studies (Lecrone and Smilowitz, 1980, Roger et al. 1991,1994) the detrimental effects of pesticides are usually minimal and do not often cause high mortality

in fields (Whitford and Showers, 1988). A high population of coccinellids may still remain in a field after insecticide applications (Zoebelein, 1988). By contrast, Redcliffe et al.

(1976) found that insecticide applications reduced the overall numbers of coccinellid larvae and adults by 90% and 67% respectively whereas fungicide applications increased them by only 3% in alfalfa agroecosystems. Insecticide treatments seemed to be detrimental to coccinellid populations in my insecticide treated plots only in 1994. Despite the same insecticide application schedule, higher coccinellid populations occurred in the full spectrum pesticide-treated plots indicating that lower coccinellid populations in the insecticide-treated plots were probably not a response of the coccinellids to insecticides but the response of the beetles to the abundance of prey (Wright and Loing, 1980) in the other plots. Additionally, results fi’om my 1995 trial showed that the insecticide applications had little effects on coccinellid populations. The effects of the fungicides and herbicides on the beetles were not clear because the applications caused reductions in C. maculata populations in 1994 but increases in 1995.

Anthocorids have been shown to be highly susceptible to insecticide applications

(Dinkins et al. 1971, Redcliffe et al. 1976). Malathion applications eliminated Orius sp. populations in seven days in blueberry fields (Whalon and Eisner, 1982). My insecticide

142 treatments significantly reduced anthocorid populations in all of the insecticide-treated

plots. In contrast to the Redclifife et al. (1976) report in which fimgicide applications

caused an 11% increase in Orius populations , the results of my experiment indicated that

the fimgicide treatments reduced anthocorid populations. Again, the combinations of

insecticide, fimgicide and herbicide treatments had more severe effects on anthocorid

populations than when the pesticides used separately.

Foliage-dwelling spiders are more susceptible to pesticide applications than are

ground-dwelling spiders (Whitford et al. 1987). Because spider webs collect pesticides,

they may have greater chances to be exposed to the chemical (Matthews and Lake, 1992)

and chemicals accumulated in their webs can repel them and consequently expose them to

other predators (Samu et al. 1992). Insecticides can cause as much as 89% mortality in

spider populations in field situations (Thomas et al. 1990). It takes at least 29 days for

spider populations to recover fi’om insecticide stress in treated areas (Culin and Yeargan,

1983). Stam et al. (1978) showed that the herbicides, dinoseb and MSMA had no effect

on spider populations. Livingston et al. (1978) reported that fungicides did not have any

effects on spider populations. However, Mansour (1987) reported that fungicides can

cause 10-40% mortality in spider populations. The spider populations in my experimental

plots showed almost the same response to pesticide treatments as the anthocorids.

Numbers remained higher in the untreated plots than in any of the treated plots. The

insecticide treatments were more detrimental to the anthocorid populations than the fungicide and herbicide treatments. The combination of all the pesticides slightly increased their adverse effects on populations of anhocorids.

143 The pesticide treatments used in my experiments affected not only the above-ground fauna but also the below-ground fauna. Pesticides can often cause reductions in free- living nematode populations (Ingham et al. 1986; Smolik, 1983; Yeates and Barker,

1986).

The bacterivorous nematodes which were the most abundant nematode group in my experimental plots were affected adversely by all of the pesticide treatments. The fungicide and herbicide treatments probably had more adverse effects on populations of bacterivorous nematodes, than the insecticide treatments, because the numbers of these nematodes were fewer in the fungicide and herbicide-treated plots. The explanation of these reductions populations in the treated plots could be the response of the bacterivorous nematode populations to the toxicity of the pesticides (Ingham et al. 1986;

Smolik, 1983; Yeates and Barker, 1986) or it could be a response to the reductions in bacterial numbers by pesticides that the bacterivorous nematodes had fed on in the treated plots. However, my results did not agree with those of Mahn and Kastner (1985) and

Schmitt and Corbin (19981) s reporting that herbicides increased had increased numbers of bacterivorous nematodes because they enhanced microbial populations.

Populations of the fungivorous nematodes showed similar responses to pesticides as the bacterivorous nematodes. They were more abundant in the control plots than in any of the treated plots. The pesticide treatments might have caused direct mortality in fungivorous nematode populations (Smolik, 1983) or the treatments might have reduced the fungi abundance on which these nematodes feed (Ingham, 1983).

144 Carnivorous nematodes are very sensitive to pollutants (Bohlen and Edwards, 1994;

Ishibashi et al. 1983; Parmelee et al. 1993; Smolik, 1983). They were the least abundant of

the nematode groups in my experimental site. The insecticide applications had adverse

effects on the carnivorous nematode populations. The fungicide and herbicide applications

probably did not affect the these nematodes much and their populations were higher in the

fungicide and herbicide treated plots.

Plant parasitic nematodes showed differential responses to the pesticide treatments.

Their populations were higher in all of the pesticide-treated plots than in the control plots.

Several reasons can cause such increases in plant parasitic nematode populations. Many

reports have indicated that pesticides can stimulate hatch or larval growth of plant

parasitic nematodes (Abivardi and Altman, 1978; Bostian et al. 1984; Biswas and Mishra,

1978; Clarke and Shepherd, 1966; Riggs and Oliver, 1982; Smith and Corbin, 1978;

Schmitt et al, 1983) . Schmitt et al. (1983) and Bostian et al. (1984) reported that

nematicides applied early in the season led to nematode-free root development and

consequently caused a resurgence of plant parasitic nematode populations later in the season, when nematicide residues had disappeared, because the roots could then sustain larger nematode populations. Herbicides can alter plant physiology (Schmitt et al. 1983) and the chemical and nutrient gradient of roots (Abivardi and Altman, 1984) and consequently make them more suitable hosts. Biological changes in the soil environment such as suppression of microorganisms, that have antagonist effects on plant parasitic nematodes, can increase plant parasitic nematode populations (Bastian et al. 1984; Miller and Taylor, 1967; Nfrller, 1967). Suppression of parasitic frmgal activity against plant

145 parasitic nematodes by the pesticides, could also lead to increase in populations of the

plant parasitic nematodes (JafiFe and Mclnms 1990; Pullen, 1990). Decreases in the numbers of the carnivorous nematodes can cause increases in plant parasitic nematode numbers (Parmelee et al. 1993).

From the time they are planted in the fields to the harvest season, tomato plants are exposed to a wide variety of arthropod pests and diseases, and compete with weeds for water, nutrients and sunlight. Chemical pest management is one of the most common practices used to protect plants fi’om damage and increase the productivity of the tomatoes. However, it is clear that chemical pest management does not benefit the environment. The often question asked is; does it benefit the farmers? Is it always profitable in tomatoes ? This depends on the pest or pest complex, the diseases damaging the plants, the number of sprays needed, type of the pesticides used , weather conditions etc. Walgenbach and Estes (1992) evaluated several insecticide applications against tomato finaitworm,Helicoverpa zea which is a primary pest in North Carolina. They reported that endosulfan, methomyl and esfenvalerate were profitable, while application of carbaryl and Bacillus thrungiensis led to losses in net profits. Wiesenbom (1990) showed that weekly methomyl applications against tomato pinworm,Keiferia lycopersicella provided high net financial returns up to nine applications. However, with over nine applications, management costs exceeded the benefit in net profit return.

On the other hand, Lambdin and Snodderly (1984) reported that although insecticide applications reduced potato aphid populations on fresh market tomatoes, the applications

146 did not increase the overall yields. Quagiia et ai. (1993) showed that 400 green peach

aphids per plant did not have any significant effect on yields of greenhouse tomatoes.

As was already discussed earlier, many of the major pests were absent in my

experiments, and the insecticide application schedule used was designed to control a wide

variety of insect pests. The major pests that were present in my experiments were foliar

and seedling pests. The insecticide applications in insecticide-treated plots provided

efihcient control of all pests and increases in the overall yield; however they resulted in

$429 monetary loss per hectar, because the management costs exceeded the financial

returns in the 1994 growing season. This was probably because the pests in my

experiments did not cause any damage to finits and to significant reductions in the overall yield, and that plants could sustain pest populations without any extra protection.

In the presence of a disease in tomatoes, however, fimgicide applications may be essential to maintain yields (Kennedy et al. 1983) and finit quality (Poysa et al. 1993).

Yield losses to early blight disease was much greater than the losses to insect pests.

Fungicide applicatons provided high profits because the early blight disease was the major problem in the experimental plots and the fungicide applications were critical to ensure yields. Kennedy et al. (1983) showed that 92% of tomato yields were lost when fungicides were not applied. Similarly Walgenbach et al. (1989) reported that 46% of tomato fiuits was infected with Altemaria solani when they were not protected with fungicides.

Pemezny (1996) reported that chlorothalonil applications provided $3,000/acre net financial return when tomatoes had Septoria leaf spot.

147 As was already discussed the weekly fungicide applications were suflHcient to ensure

the yield and net financial return in 1994 despite approximately 30% defoliation occurred

in the fungicide treated plots. However, in 1995, when weather conditions favored the disease development weekly applications did not provide sufficient control and disease caused over 80% defoliation in the plants in the fungicide and herbicide treated plots.

148 REFERENCES

Abivardi, C. and J. Altman. 1978. Effects of cycloate on development oîHeterodera schachtii and growth of three 5eto species.J. Nematol. 10(1): 90-94.

Adams, J.B. 1960. Effects of spraying 2,4-D amine on coccinellid larvae. Con. J. Zool. 38: 285-288.

Adams, J.B. and M E. Drew. 1965. Grain aphids. In New Brunswick. HI. Aphid populations in herbicide-treated oat fields. Can. J. Zool. 43: 789-794.

Aebischer, N.J. and G.R. Potts. 1990. Long-term changes in numbers of cereal invertebrates assessed by monitoring. Brighton Crop Protection Confemce-Pest and Disease : 163-172.

Agnello, A.M., J.W. Van Duyn and J R. Bradley. 19086. Influence of postemergence herbicides on populations of bean leaf beetle, Cerotoma trifurcata (coleoptera: Chrysomelidae), and com earworm, Heliothis zea (Lepidoptera: Noctuidae), in soybeans. J. Econ. Entomol. 79(1): 261-255.

Agnew, C.W. and J.W. Jr. Smith. 1989. Ecology of spiders (Araneae) in a peanut agroecosystem.Environ. Entomol. 18(1): 30-42.

Ajlan, A.M. and D A. Potter. 1991. Does immunization of cucumber against anthracnose by Colletotrichum lagenarium affect host suitability for arthropods ? Entomol. Exp. Appl. 58:83-91.

Akhtar, S. and H.V. Van Emden. 1992. The effect of the systemic fungicide benomyl on survival and reproduction of the bird cherry aphid (Rhopalosiphum padi). Ann. Appl. Biol. 120: 245-255.

Allen, W.R., B.Tehrani and R. Luft. 1993. Effect of horticultural oil, insecticidal soap, and film-forming products on the western flower thrips and the tomato spotted wilt vinxs. Plant Disease. 77: 915-918.

149 Alvardo-Rodriguez, B. 1988. Larval parasites of tomato pinworm (Lepidoptera; Gelechiidae) in Sinola, Mexico. Florida Entomologist 71(1): 84-86.

Andersen, A. and J.A. Sharman. 1983. Effect of chlorfenvinphos and isofenphos on Carabidae and Staphylinidae (Col.) and their predation of eggs of Delia floralis Fallen (Diptera, Anthomyiidae) in field experiments. Z Ang. Ent. 95: 206-213.

Andoloro, J.T., C.W. Hoy, K.B. Rose and A.M. Shelton. 1983. Evaluation of insecticide usage in the New York processing-cabbage pest management program. J. Econ. Entomol. 76(5): 1121-1124.

Antle, J.M. and S.K. Park. 1986. The economics of IPM in processing tomatoes. California Agriculture 40(3-4): 31-32.

Asteraki, E.J., C.B. Hanks and R.O. Clements. 1992. The impact of two insecticides on predatory ground beetles (Carabidae) in newly-sown grass. Arm. Appl. Biol. 120: 25-39.

Babcock, B.A., E. Lichtenberg, and D. Zilberman. 1992. Impact of damage control and quality of output: estimating pest control effectiveness. Amer. J. Agr. Econ. 163- 172. Baron, R.L. and T.L. Merriam. 1988. Toxicology of aldicarb.Rev. Environ. Contam. Toxicol. 105: 1-54.

Basedow, T., H. Rzhak and K. Vop. 1985. Studies on the effect of deltamethrin sprays on the numbers of epigeal predatory arthropods occuring in arable fields. Pestic. Sci. 16: 325-331.

Bautista, R.C. and R.F.L. Mau. 1994. Prefemces and development of western flower thrips (Thysanoptera: Thripidae) on plant hosts of tomato spotted wilt tospowrus in Hawaii. Environ. Entomol. 23(6): 1501-1507.

Bellows, T.S. J.G. Morse, L.K. Gaston and J.B. Bailey. 1988. The fate of two systemic insecticides and their impact on two phytophagous and a beneficial arthropod in a citrus agroecosystem. J. Econ. Entomol. 81(3): 899-904.

Bellows, T.S. and J.G. Morse. 1993. Toxicity of insecticides used in citrus to Aphytis melirms Debach (Hymenoptera: Aphelinidae) and Rhizobius lophanthae (Blaisd.) (Coleoptera: Coccinellidae).Can. Entomol. 125: 987-994.

Benedict, J.H. M.H. Walmsley, J.C. Segers and M.F. Treacy. 1986. Yield enhancement and insect suppression with chlordimeform (fundal) on dryland cotton. J. Econ. Entomol. 79(1): 238-242.

150 Bentley, W.J., F.G. Zalom, W.W. Bamett and J.P. Sanderson. 1987. Population densities of Tetranychus spp. (Acari: Tetranychidae) after treatment with insecticides for Âmyelois transitella (Lepodoptera: Pyralidae). J.Econ. Entomol. 80: 193-199.

Biddinger, D.J. and LA. Hull. 1995. Eflfects of several types of insecticides on the mite predator, Stethorus punctum (Coleoptera: Coccinellidae), including insect growth regulators and abamectin. J. Econ. Entomol. 88(2): 358-366.

Biggs, A. R. and E.A.C. Hagley. 1988. Effects of two stereol-inhibiting fungicides on populations of pest and beneficial arthropods on apple.Agriculture, Ecosystems and environment. 20: 235-244.

Biswas, D.K. and C. Mishra. 1987. Influence of herbicides on the nematoda fauna in soil of muhiple cropping. Environment & Ecology. 5(1): 133-136.

Bluemberg, A.Y. and D A. Crossley. 1983. Comparison of soil surface arthropod populations in conventional tillage, no-tülage and old field systems. Agro- ecosystems 8:247-243.

Bohlen, P.J. and C.A.Edwards.1994. The response of nematode trophic groups to organic and inorganic nutrient inputs in agroecosystems. Soil.Soc.Am. special publ. no. 35:235-244.

Boiteau, G. 1984. Effect of planting date, plant spacing, and weed cover on populations of insects, arachnids, and entomopthoran fungi in potato fields. Environ. Entomol. 13(3): 751-756.

Bosch, R. V. D., H.T. Reynolds and E.J. Dietrick. 1956. Toxicity of widely used insecticides to beneficial insects in California cotton and alfalfa fields. J. Econ. Entomol. 49(3): 359-363.

Bostian, A.L., DP. Schmitt and K.R. Barker. 1984. In vitro and survival oîHeterodera glycens as affected by alachlor and phenamiphos. J. Nematol. 16(1): 22-26.

Bower, K.N., L.P. Berkett and J.F. Costante. 1995. Nontarget effect of a fungicide spray program on phytophagous and predacious mite populations in a scab-resistant apple orchard. Environ. Entomol. 24(2): 423-430.

Boykin, L.S., W.V. Campbell and M.K. Beute. 1984. Effects of pesticides on Neozygites floridana (Entomopthorales: Entomopthoraceae) and arthropod predators attacking the twospotted spider mite (Acari: Tetranychidae) in North Carolina peanut fields. J. Econ. Entomol. 77(4): 969-975.

151 Braman, S.K and A.F. Pendley. 1993. Relative and Seasonal abundance of beneficial arthropods in centipedegrass as influenced by management practices. J.Econ. Entomol. 86 (2): 494-504.

Brammal, RJD. 1993. Effect of foliar fungicide treatment on early blight and yield of firesh market tomato in Ontario.Plant Disease 77(5): 484-488.

Braun, A.R., JM. Guerrero, A.C. Bellotti and L.T. Wilson. 1987. Relative toxicity of permrthrin toMononychellus progresivus Doreste and Tetranychus urticae Koch (Acari: Tetranychidae) and their predators Amblyseius limonicus Greman&McGregor (Acari: Phytoseiidae) and Oligota minuta Cameron (Coleoptera: Staphylinidae): Bioassays and field validation. Environ. Entomol. 16(2): 545-550.

Braun, A.R. J.M. Guerrero, A. C. Bellotti and L.T. Wilson. 1987. Evaluation of possible nonlethal side effects of permethrin used in predator exclusion experiments to evaluate Amblyseius limonicus (Acari: Phytoseiidae) in biological control of cassava mites (Acari: Tetranychidae). Environ. Entomol. 16(4): 1012-1018.

Browdie, LA. G.L. Tylka, L.P. Pedigo and M.D.K. Owen. 1994. Responses of Heterodera glycines populations to a postemergence herbicide mixture and simulated insect defoliation. J. Nematol. 26(4): 498-504.

Brown, K.C.,J.H. Lawton and S.W. Shires. 1983. Effects of insecticides on invertebrate predators and their cereal aphid (Hemiptera: Aphididae) prey : Laboratory Experiments. Environ. Entomol. 12: 1747-1750.

Brun, L.O. 1981. Experimentation de nauf insecticides destines a la protection des cultures de tomates en Nouvelle-Calodinie. O.R.S.T.OM., ser. Biol. 44: 35-42.

Brust, G.E., B.J.Stinner and D A. McCartney. 1985. Tillage and soil insecticides effects on predator- Black cutworm (Lepidoptera: Noctuidae) interactions in com agroecosystems.J. Econ. Entomol. 78: 1389- 1392.

Brust, G.E., B.J. Stinner and D A. McCartney. 1986. Predator activity and predation in com agroecosystems. £wv/ron. £>/to/no/. 15(5): 1017-1021.

Brust, G.E. 1990. Direct and indirect effects of four herbicides on the activity of carabid beetles (Coleoptera: Carabidae). Pestic. Sci.'iQ: 309-320.

Bryan, M.K. and S.D.Wratte n. 1984. The responses of polyphagous predators to prey spatial heterogeneity: aggregation by carabid and staphylinid beetles to their cereal aphid prey. Ecol. Entomol. 9: 251-259.

152 Burke, HR. 1959. Toxicity of several insecticides to two species of beneficial insects on cotton.J.Econ. Entomol. 52(4): 616-618.

Buschman, L.L. and L.J. DePew. 1990. Outbreaks of Banks grass mite (Acari: Tetranychidae) in grain sorghum following insecticide applications. J. Econ. Entomol. 83(4): 1570-1574.

Campbell, C.D., U . Walgenbach and G.G. Kennedy. 1991. Effect of parasitoid on lepidopterous pests in insecticide-treated and untreated tomatoes in western north Carolina, y. Econ. Entomol. 84(6): 1662-1667.

Campbell, W.V. and R.E. Hutchins. 1952. Toxicity of insecticides to some predaceous insects on cotton.J. Econ. Entomol. 45(5): 828-833.

Carcamo, H.A., J.K. Niemala and J R. Spence. 1995. Farming and ground beetles: Eflfects of agronomic practice on populations and community structiu'e. Can. Ent .127: 123-140.

Castro, C.E., H.E.McKinney and S.Lux. 1991. Plant protection with inorganic ions. JNematol. 23(44): 409-413.

Chase, C and M. Duflfy. 1991. An economic comparison of conventional and reduced- chemical farming systems in Iowa. American Journal o f Alternative Agriculture 6(4): 168-173.

Chellemi, D.O., S.M. Olson and D.J. Mitchell. 1994. Eflfects of soil solarization and fumigation on survival of soilbome pathogens of tomato In Northern Florida. Plant Disease 78: 1167-1172.

Chen, S, D.W. Dixon and E.B. Whitty. 1994. Response oîMeloidogyne spp. toPasteuria penetrans, fungi and cultural practices in tobacco. J. Nematol 26(4S): 620-625.

Cheng, H.H. and J.J. Hanlon. 1990. A note on the use of microbial insecticides for the control of tomato homworms on flue-cured tobacco.Phytoprotection 71: 101- 103.

Chiverton, P. A. 1984. Pitfall-trap catches of the carabid beetle Pterostichus melanarius, in relation to gut contents and prey densities, in insecticide treated and untreated plots in spring barley. Entomol. Exp. Appl. 36: 23-30.

Cho, K., C.S. Eckel, J.F.Walgenbach and G.Kennedy. 1995. Spatial distribution and sampling procedures for Frankliniella spp. (Thysanoptera: Thripidae) in staked tomato, y. Econ. Entomol. 88(6): 1658-1665.

153 Clark, MS. J.MLuna, N.D.Stone and R.R.Youngman. 1994. Generalist predator consumption of armyworm (Lepidoptera: Noctuidae) and effect of predator removal on damage in no-till com.Environ. Entomol. 23(3): 617-622.

Clarke, A.J. and A.M Shepherd. 1966. The action of nabam, metham-sodium and other sulphur compounds. Arm. Appl. Biol. 57: 241-255.

Coaker, T.H. 1966. The effects of soil insecticides on the predators and parasites of the cabbage root fly (Erioischia brassicae (Bouche)). And on the subsequent damage caused by the pest. Ann. Appl. Biol. 57: 397-407.

Coats, S. A., J.R. Coats and C.R Ellis. 1979. Selective toxicity of three synthetic pyrethroids to eight coccinellids, a eulophid parasitoid, and two pest chiysomelids. Environ. Entomol. 8: 720-722.

Cockfield, S.D. and DA. Potter. 1983. Short-term effects of insecticidal applications on predaceous arthropods and oribatid mites in Kentucky bluegrass turf. Environ Entomol. 12: 1260- 1264.

Cockfield, S.D. and D A. potter. 1984. Predation on sod webworm (Lepidoptera: Pyralidae) eggs as affected by chlorpyrifos application to Kentucky bluegrass turf. J. Econ. Entomol. 77(4): 1542-1544.

Cowles, R.S and M.G. Villani. 1994. Soil interactions with chemical insecticides and nemaatodes used for control of Japanese beetle (Coleoptera: Scarabaeidae) larvae. J.EconEntomol. 87(4):1014-1021.

Critchley, B.R. 1972. A laboratory study of the effects of some soil-applied organophosphorus pesticides on carabidae (Coleoptera).Bull. Ent. Res. 62: 229- 242.

Critchley, B.R. 1972. Field investigations on the effects of an organophosphorus pesticide, thionazin, on predacious Carabidae (Coleoptera). Bull. Ent. Res. 62:327-342.

Croft, B.A. and W. A. Brown. 1975. Responses of arthropod natural enemies to insecticides. Arm. Rev. Ent. 285-335.

Culin, J.D., and K.V. Yeargan. 1983. The effets of selected insecticides on spiders in alfalfa. Journal o f the Kansas Entomological Society 56(2): 151-158.

Davies, D.H.K. and A.J. Whiting. 1989. Yield responses to herbicide use and weed levels in winter wheat and spring barley in Scottish trials and consequences for economic models. Brighton crop protection conference-weeds-% :9S5-96Q.

154 Be Bach, P. 1951. The necessity for an ecological approach to pest control on citrus in California. J. Econ. Entomol. 44(2): 443-447.

Dennis, P. and ST). Wratten. 1991. Field manipulation of populations of individual staphylinid species in cereals and their impact on aphid populations. Ecol. Entomol. 16:17-24.

Demoeden, PJî, L.RTCrusberg and S.Sardanelli.l990. Fungicide effects on Acremonium endophyte, plant-parasitic nematodes, and thatch in Kentucky bluegrass and perennial rygrass. Plant Dis. 74: 879-881.

Bidden, W.A.M., J.C.Y.Marinissen, M.J. Vreeken-Bruijs, S.L.G.E. Burgers, R. de Fluiter, M. Geurs and L. Brussaard. 1994. Soil meso-and macrofauna in two agricultural systems: factors affecting population dynamics and evaluation of their role in carbon and nitrogen dynamics. Agriculture, Ecosystem and Environment 5: 171- 186.

Dinkins, R.L., J.R. Brazzel and C.A. Wilson. 1971. Effect of early season insecticide applications on major predaceous arthropods in cotton fields under an integrated control program.J.Econ. Entomol.6A,Ç1): 480-484.

Dintenfass, L.P., DP. Bartell and M.A. Scott. 1987. Predicting resurgence of western flower thrips (Thysanoptera: Thripidae) on onions after insecticide application in the Texas Wgh plains. J. Econ. Entomol. 80(2): 502-506.

Dinter, A. and H.M. Poehling. 1995. Side-effects of insecticides on two erigonid spider species. Ent. Exp. Appl. 74: 151-163.

Dixon, P.L. and R.G. McKinlay. 1992. Pitfall trap catches of aphid predation by Pterostichus melanarius and Pterostichus madidus in insecticide treated and untreated potatoes. Entomol. Exp. Appl. 64: 63-72.

Donald, W.W. and T. Prato. 1992. Effectiveness and economics of repeated sequences of herbicides for cananda thistle (Cirsium arvense) control in reduced-till spring wheat [Jriticum aestivum). Can. J. Plant. Sci. 72: 599-618.

Drinkwater, L.E., D.K Letoumeau, F. Workneh, H.C. Van Bruggen and C. Sherman. 1995. Fundamental differences between conventional and organic tomato agroecosystems in California.Ecological Applications 5(4): 1098-1112.

Dunning, R.A., J.M. Cooper, J.M. Wardman and G.H. Winder. 1982. Susceptibility of the carabid, Pterostichus melanarius (Illiger) to aphicide sprays applied to the sugar-beet crop. Ann. Appl. Biol. 100: 32-33.

155 Edelson, J.V., B. Cartwright and T.A. Royer. 1989. Economics of controlling onion thrips (Thysanoptera: Thripidae) on onions with insecticides in south Texas. J. Econ. Entomol. 82(2): 561-564.

Edwards, C.A. and ÆR. Thompson. 1973. Pesticides and soil fauna. Residue reviews 45: 1-75.

Edwards, C.A. and A_R. Thompson. 1975. Some ejects of insecticides on predatory beetles. .4/iw. Appl. Biol. 80(1): 132-135.

Edwards, C.A., K.D. Sunderland and K.S. George. 1979. Studies on polyphagous predators of cereal aphids. J. Appl. Ecol. 16: 811-823.

Edwards, C.A. 1987. The environmental impact of pesticides, 309-329. In: V. Delucchi (ed). Protection integree: quo vadis?

Edwards, C.A. 1989. Impact of herbicides on soil ecosystems. Critical Reviews in Plant Sciences. 8(3): 221-257.

Edwards, C.A. 1991. Long term ecological effects of herbicides: field studies. Brighton Crop Protection Conference-Weeds : 883-890.

Elmer, A., C. Hagley and C. M. Simpson. 1983. Eflfects of insecticides on predators of the pear psylla, Psyllapyriacola (Hemiptera: Psyllidae), in Ontario. Can. Ent.WS: 1409-1414.

Etienne, J ., J. Guyot and X.V. Waetermeulen. 1990. Effect of insecticides, predation and precipitation on populations ofThrips palmi on eubergine (eggplant) in Gndiàelowpe. F lo rid Entomologist. 73(2): 339-342.

Fabellar, L.T. and E.A. Heinrichs. 1984. Toxicity of insecticides to predators of rice brown planthoppers, Nilaparvata lugetis (Stal) (Homoptera: Delphecide). Environ Entomol 13(4): 832-837.

Fabellar, L.T. and E.A. Heinrichs. 1986. Relative toxicity of insecticides to rice planthoppers and leafhoppers and their predators. Crop protection 5(4); 254-258.

Fagg, J. and J.T. Fletcher. 1987. Studies of the epidemiology and control of tomato stem rot caused by Didymella lycopersici. Plant Pathology. 36: 361-366.

Farlow, R.A., and H.N. Pitre. 1983. Bioactivity of the postemergent herbicides acifluorfen and bentazon on Geocoris punctipes (Say) (Hemiptera: Lygaeidae). J. Econ. Entomol. 76(1): 200-203.

156 Fentom, E.A. 1959. The effect of several insecticides on the total arthropod population in alfalfa. J. Econ. Entomol. 52(3): 428-432.

Ferrandino, F and W.H. Elmer. 1991. The effect of defoliation due to Septoria leaf spot on tomato yield.Phytopathology 81(1): 122 (abstr.).

Ferrandino, F and W Ji Elmer. 1992. Reduction in tomato yield due to Septoria leaf spot. Plant Diseae 76: 208-211.

Flexner, J. L., B. Lighthart and B.A. Croft. 1986. The effects of microbial pesticides on non-target, beneficial arthropods. Agriculture, Ecosystems and Environment. 16: 203-254.

Flint, ML. and K. Klonsky. 1985. Pest management practices in processing tomatoes. California Agriculture 19-20.

Floate, K.D., RH. Elliot, J.F. Doanne and C. Gillot. 1989). Field bioassay to evaluate contact and residual toxicities of insecticides to carabid beetles (Coleoptera: Carabidae). y. Econ. Entomol. 82(6): 1543-1547.

Fortnum, B.A. and D.L.KarIen. 1985. Effects of tillage system and irrigation on population densities of plant nematodes in field com. J.Nematol. 17(1): 25-28.

Fox, G and A. Weersink. 1995. Damage control and increasing returns. Amer. J. Agr. Econ. 77: 33-39.

Franc, G.D., M.D. Harrison and L.K. Lahman. 1988. A simple day-degree model for initiating chemical control of potato early blight in Colorado.Plant Disease 72: 851-854.

Freckman, D.W. and E.P.CasweIl. 1985. The ecology of nematodes in agroecosystems. Ann. Rev. Phytopathol. 23: 275- 296.

Fulling, B.A., EC. Tigchelaar and R Latin. 1995. Integration of host resistance and weather-based fungicide scheduling for control of anthracnose of tomato fiuit. Plant Disease 79: 228-233.

George, T.L.,C.M. Lowell and A. Fowler. 1992. Effects of a carbaryl bait treatment on nontarget wildlife. Environ. Entomol. 21(6): 1239-1247.

Gerard, A.M. and J.O. Akkerhius. 1993. Walking behaviour and population density of adult Inyphiid spiders in relation to minimizing the plot size in short term pesticide studies with pyrethroid insecticides. Environmental Pollution 80: 163-171.

157 Ghidiu, G.M. and R.D. Getting. 1987. Oxamyl slow-release tablets, foliar sprays, and transplant drench applications for Colorado potato beetle (Coleoptera; Chrysomelidae) control on tomato.J. Econ. Entomol . 80(4): 876-879.

Gholson,L.E., C.C.Beegle, R.L.Best and J.C.Owens. 1978. Effects of several commonly used insecticides on cornfield carabids in Iowa. J. Econ. Entomol. 71(3): 416-418.

Giroux, S., J. Cote, C. Vicent, PMartel and D. Coderre. 1994. Bacteriological insecticide M-One effects on the predation efficiency and mortality of adult Coleomegilla maculata lengi (Coleoptera: Coccinellidae).J.Econ. Entomol. 87(1): 39-43.

Gitatis, R_, S. MvCarter and J. Jones. 1992. Disease control in tomato transplants produced in Georgia and Florida. Plant Disease 76(7): 651-656.

Gleason, M.L., A.A. MacNab, R.E. Pitblado, M.D. Ricker. D A. East and R.X. Latin. 1995. Disease-warning systems for processing tomatoes in Eastern North America: Are we there yet? Plant Disease 79(2): 113-121.

Gleason, M.L. S.E. Taylor. H.G. Taber and M L Hockmuth. 1992. TOM-CAST on tomatoes in Iowa: Effects of model threshold and sensor location on control of fimgal diseases. Phytopathology 82(10): 1164 (abstr.)

Glaeson, M.L., R_D. Gitatis and M.D. Ricker. 1993. Recent progress in understanding and controlling bacterial canker of tomato in Eastern North America.Plant Disease 77 (11): 1069-1076.

Glick, P. A. and W.B. Lattimore. 1954. The relation of insecticides to insect populations in cotton fields. J. Econ. Entomol. 47(4): 681-684.

Godfiray, H.G.J. and M.S. Chan. 1990. How insecticides trigger single-stage outbreaks in tropical pests. Funct. Eco/.4: 329-337.

Gordon, P.L. and F.L. McEwen. 1984. Insecticide-stimulated reproduction oiM yzus persicae, the green peach aphid (Homoptera: Aphidiade). Can. Ent. 116: 783-784.

Griffin, G.D. and J.L. Anderson. 1979. Effects of DCPA, EPTC, and chloproham on pathogenicity oîMeloidogyne hapla to alfalfa. J. Nematol. 11(1): 32-36.

Halsall, N.B. and S.D. Wratten. 1988. The efficiency of pitfall trapping for polyphagous predatory Carabidae. Ecol. Entomol. 13: 293-299.

Hamilton, E.W. and R.W. Kieckhefer. 1969. Toxicity of malathion and parathion to predators of the English grain aphid. J. Econ. Entomol. 62(5): 1190-1192.

158 Hamilton, J.T. and R.B. Toffolon. 1987. Pests of tomatoes.Agfacts, Department of agriculture. New South Wales. 19 pp.

Hara, A.H. and H.K.Kaya. 1983. Toxicity of selected organophosphate and carbamate pesticides to infective juveniles of the entomogenous nematodesNeoplectana_ carpocapsae ( Rhabitida: Steimematiddae). Environ. Entomol. 12: 496-501.

Hardin, M.R., B. Benrey, M Coll, W.O. Lamp, G.K. Roderick and P. Barbosa. 1995. Arthropod pest resurgence: an overview of potential mechanisms.Crop protection 14(1): 3-18

Harding, J. 1971. Field comparison of insecticidal sprays for control of four tomato insects in South Texas. J. Econ Entomol. 64(3): 1302-1304.

Hardman, J.M, R.E.L. Rogers, J.P. Nyrop, and Tracey Frisch. 1991. Eflfect of pesticide applications on abundance of European red mite (Acari: Phytoseiidae) in Nova Scotian apple orchards. J. Econ Entomol. 84(2): 570-580.

Hardman, J.M. 1992. Apple pest management in North America: Challenge and response. Brighton Crop Protection Conference-Pests and diseases- 2:506-515.

Harris, C.R. 1977. Biological activity of chlorpyrifos, chlopyrifos-methyl, phorate, and counter in soil. Can. Ent. 109: 115-1120.

Hartman, G.L. 1992. Black leaf mold development and its effect on tomato yield. Plant Disease 76(5): 462-465.

Hassan, S.A., F. Bigler., H. Bogenschutz, J.U. Brown, S.I. Firth, P. Huang, M.S. Ledieu, E. Naton, P.A. Oomen, W.P.J. Overmeer, W. Rieckmann, L. Samsque-Pettreson, G. Viggiani and A.Q. V. Zon. 1983. Results of the second joint pesticide testing programme by the lOBC/WPRS-working group “pesticides and beneficial arthropods. Z. Ang. Ent. 95: 151-158.

Hatcher, P.E., N.D. Poul, P.G. Ayres and J.B. Whittaker. 1994. The effect of a foliar disease (rust) on the development of Gastrophyta viridula (Coleoptera: Chrysomelidae). Ecol. Entomol.\9: 349-360.

Haynes, K.F. 1988. Sublethal effects of neurotoxic insecticides on insect behavior. A nn Rev. Entomol. 33:149-168.

Headley, J.C. 1972. Economics of agricultural pest control. Ann. Rev. Ent. 273-286.

Heinz, K.M., M.P. Parella and J.P. Newman. 1992. Time-efficient use of yellow sticky traps in monitoring insect populations.J. Econ Entomol. 85(6): 2263-2269.

159 Hendrix, P.F., R-WJarmelee, D.A.Crossley, D.C.Coleman, E.P.Odum and MGroffinan. 1986. Detritus food webs in conventional and no-tillage agroecosystems-5/ojc/e«ce 36(6): 374-380.

Herdt, R.V., L.L Castillo and S.K. Jayasuriya. 1984. The economics of insect control on rice in the Philippines.p. 41-56. In: Proceedings of the FAO/IRRI workshop on judicious and efGcient use of insecticides on rice. International rice research institute.

Hofifinan, M.P., L.T. Wilson, F. Zalom. 1987. Control of stink bug in tomatoes. California Agriculture 41(5-6): 4-6.

HoflSnann, MP. L.T. Wilson, F.G. Zalom and R.J. Hilton. 1990. Parasitismo f Heliothis zea (Lepidoptera: Noctuidae) eggs : Effect on pest management rules for processing tomatoes in the Sacramento Valley of California.Environ. Entomol. 19(3): 753-763.

Hofifinann, M.P., L.T. Wilson, F.G. Zalom. R.J. Hilton and C.V.Weakley. 1996. Parasitoid helps control fruitworm in Sacramento Valley processing tomatoes.California Agriculture 44(1): 20-23.

Holt, J., D.R. Wareing and G.A Norton. 1992. Strategies of insecticide use to avoid resurgence o f Nilaparvata lugens ( Homoptera: Delphacidae) in tropical rice: a stimulation analysis. J.Econ. E n to m o l. 1979-1989.

Honek, A 1988. The effect of crop density and microclimate on pitfall trap catches of Carabidae, Staphylinidae (Coleoptera), and Lycosidae (Araneae) in cereal fields. Pedobiologia 32: 233-242.

Hough-Goldstein, J and C.B. Keil. 1991. Prospects for integrated control of the Colorado potato beetle (Coleoptera: Chrysomelidae) usingPerillus bioculatus (Hemiptera: Pentatomidae) and various pesticides. J. Econ. Entomol. 84(6): 1645-1651.

Hull, L A , E.H. Beers and R. L Meagher. 1985. Integration of biological and chemical tactics for apple pests through selective timing and choice of synthetic pyrethroid insecticides. J. Econ. Entomol. 78(3): 714-721.

Hurej, M. and J.D. Dutcher. 1994. Indirect effect of insecticides on convergent lady beetle (Coleoptera: Coccinellidae) in pecan orchards.J.Econ. Entomol 87(6): 1632-1635.

Imbriani, J.L. and R. Mankau. 1983. Studies on Lasioseius scapulatus, a mesostigmatid mite predaceous on nematodes. J. Nematol. 15(4): 523-528.

160 Immaraju, A.J., J.G. Morse, and R_F. Hobza. 1990. Field evaluation of insecticide rotation and mixtures as strategies for citrus thrips (Thysanoptera: Thripidae) resistance management in California.J. Econ, Entomol. 83(2): 306-314.

Immaraju, J.A ., T. D. Paine, J.A. Bathke, K.L. Robb and J.P. Newman. 1992. Western flower thrips (Thysanoptera: Thripidae) resistance to insecticides in coastal California greenhouses. J. Econ. Entomol. 85(1): 9-14.

Ingham, R.E., J.A.Trofymow, E.R. Ingham and D.C.Coleman. Interactions of bacteria, fungi, and their nematode grazers: Effects on nutrient cycling and plant growth. Ecol. Monog. 55(1): 119-140.

Ingham, E.R, J.A. Trofymow, R.N.Ames, H.W. Hunt, C_R_Morley, J.C. Moore and D C. Coleman, 1986. Trophic interactions and nitrogen cycling in a semi-arid grassland soil. JAppl. Ecol. 23: 615-630.

Ingram, J.W., E.K. Bynum and L.J. Charpentier. 1947. Effect of 2,4-D on sugarcane borer J. Econ. Entomol. 40(5): 745-746.

Ishibashi, N., E. Kondo and S. Ito. 1983. Effects of application of certain herbicides on soil nematodes and aquatic invertebrates in rice paddy fields in Japan. Crop Protection 2(3): 289-304.

Jackson, A.E.M. and R.M. Wilkins. 1985. The effect of sub-lethal dosages of synthetic pyrethroid fenvalerate on the reproductive rate of the aphid, Myzus persicae. Pestic. Sci. 16:364-368.

Jaffe, B.A and T.M. Mclnnis. 1990. Effect of carbendazim and the nematophogous fungus Hirsutella rhossiliensis and ring nematoe. J. Nematol. 22(3): 418-419.

Jaques, R.P. and D R. Laing. 1989. Effectiveness of microbial and chemical insecticides in control of the Colorado potato beetle (Coleoptera: Chrysomelidae) on potatoes tomatoes.CcauEnt. 121: 1123-1131.

Jensen, E. 1992. Integrated pest mangement at the grower level. Pestic. Sci. 36: 355-357.

Johnson, K.S., J.M. Scriber, J.K. Nitao and D R. Smithley. 1995. Toxicityo îBacillus thuringiensis var. kurstaki to three nontarget Lepidoptera in field studies. Environ. Entomol. 24(2): 288-297.

Johnson, M.W., E.R. Oatman and J.A. Wyman. 1980. Effects of insecticides on populations of the vegetable leafininer and associated parasites on summer pole tomatoes.J. Econ. Entomol. 73(1): 61-66.

161 Jones, J.B., J.P. Jones, R.E. Stall and T.A. Zitter. 1991. Compendium of tomato diseases. APS press, Minnesota, 73 pp.

Juma, N.G. and C.C. Mishra. 1988. Effect of an annual and perennial crop on trophic group dynamics of nematodes. Can. J. Soil Sci. 68; 101-109.

Karel, AJC, and G.C. Ashimogo. 1991. Economics of insect control on common beans soybeans in Tanzania. J. Econ. Entomol. 84(3): 996-1000.

Kennedy, G.G., L. R. Romanov, S J. Jenkins and D C. Sanders. 1983. Insects and diseases damaging tomato Suits in the Coastal Plain of North Carolina.J. Econ. Entomol. 76(1): 168-173.

Kems, DX. and M.J. Gaylor. 1993. Induction of cotton aphid outbreaks by insecticides in cotton.Crop Protection 12(5): 387-393.

Kems, D.L. and M.J. Gaylor. 1993. Biotic control of cotton aphids (Homoptera: Aphididae) in cotton influenced by two insecticides. J. Econ. Entomol. 86(6): 1824-1834.

Kerry, B.R. and D.H. Crump. 1977. Observations on fungal parasites of females and eggs of the cereal cyst-nematode, Heterodera avenae, and other cyst nematodes. Nematologica 23: 193-201.

Kharboutli, M.S. and T.P. Mack. 1991. Relative and seasonal abundance of predaceous arthropods in Alabama peanut fields as indexed by pitfall traps. J. Econ. Entomol. 84(3): 1015-1023.

Kish, L.P., I. Majchrowicz and K.D. Biever. 1994. Prevalence of natural fungal mortality of green peach aphid (Homoptera: Aphididae) on potatoes and nonsolanaceous hosts in Washington and Idaho. Environ. Entomol. 23(5): 1326-1330.

Kring, J.B. and D.J. Schuster. 1992. Management of insects on pepper and tomato with UV-reflective mulches. F/ondh £«ro/n/og7s/. 75(1): 119-129.

Kraus, R, G,R Noel and D.I. Edwards. 1982. Effect of preemergence herbicides and aldicarb on Heteridera glycines population dynamics and yield of soybean. J. Nematol. 14(4): 452.

Lambdin, P.L. and L.J. Snodderly. 1984. Chemical control of the potato aphid, Macrosiphum euphorbiae (Thomas), on tomato in Tennessee.Tennessee Farm and Home Science 129: 6-8.

162 Lambdin, P i . and L.J. Snoddrely. 1984. Green peach aphid control on tomato with foliar applied insecticides. Tennessee Farm and Home Science 132: 6-8.

Lange, W.H. and L. Bronson. 1981. Insects pests of tomatoes.Ann. Rev. Entomol. 26: 345-71.

Lecrone, S. and Z Smilowitz. 1980. Selective toxicity of primicarb, carbaryl and methamidophos to green peach aphid, {Myzus persicae), Coleomegilla maculata lengi (Timberlake) and Chrysopa oculata Say. Environ. Entomol. 9: 952-755.

Livingston, J.M., W.C. Yearian and S.Y. Young. 1978. Effect of insecticides, fungicides and insecticide-fungicide combinations on developments of lepidopterous larval populations in soybean. Environ. Entomol. 7823-828.

Los, L.M. and W.A. Allen. 1983. Abundance and diversity of adult carabidae in insecticide-treated and untreated alfalfa fields. £wv/row.£htomo/. 12: 1068-1072.

Lowery, D.T. and M.K. Sears. 1986. Simulation of reproduction of the green peach aphid (Homoptera: Aphididae) by azinphosmethyl applied to potatoes. J. Econ. Entomol. 79(6): 1530-1533.

Lowery, D.T. and M. K. Sears. 1986. Effect of exposure to the insecticide azinphosmethyl on reproduction of green peach aphid (Homoptera: Aphididae).J. Econ. Entomol. 79(6): 1534-1538.

Lovei, G.L., M.Sarospataki and Z.A. Radwan. 1991. Structure of ladybird ( Coleoptera: Coccinellidae) assemblages in apple: changes through developmental stages. Environ. Entomol. 20(5): 1301-1308.

Lybecker, D.W. E. E. Schweizer and R.P. King. 1988. Economic analysis of four weed management systems. Weed Science. 36: 846-849.

Madden, L., S.P. Pennypacker and A.A. MacNab. 1978. FAST, a forecast system for Altemaria solani on tomato.Phytopathology 68: 1354-1358.

Mack, T.P. 1992. Effects of five granular insecticides on the abundance of selected arthropod pests and predators in peanut fields. J.Econ. Entomol. 85(6): 2459- 2466.

Mahn, E-G. and A. Kastner. 1984. Effects of herbicide stress on weed communities and soil nematodes in agroecosystems.Oikos. 44: 185-190.

Mankau,R_1980. Biological control of nematode pests by natural enemies.Ann. Rev. Phytopathol 18: 415-440.

163 Mann, B.P., SJD. Wratten, M-Poehling and C. Borgemeister. 1991. The economics of reduced-rate insecticide applications to control aphids in winter wheat. Anth Appl. Biol. 119: 451-464.

Mansour, F. 1987. Spiders in sprayed and unsprayed cotton fields in Israel, their interactions with cotton pests and their importance as predators of the Egyptian cotton leaf worm,Spodoptera littoralis. Phytoparasitica 15(1): 31-41.

Martin, M.A., M.M. Schreiber, J.R.Riepe and J.R.Bahr. 1991. The economics of alternative tillage systems, crop rotations, and herbicide use on three representative east-central com belt farms. Weed Science 39: 299-307.

Martin, A.M., M.M. Schreiber, J.R.Riepe and J.R. Bahr. 1991. The economics of alternative tillage systems, crop rotations, and herbicide use on three representative east-central com belt farms. Weed Sciece. (39)2:299-307.

Matcham, E.J. and C. Hawkes. 1985. Field assessment of the effects of deltamethrin on polyphagous predators in winter wheat. Pestic. Sci. 16: 317-320.

Maxwell, R.C. and R.F. Harwood. 1960. Increased reproduction of pea aphids on broad beans treated with 2,4-D. Ann. Ent. Sac. Am. 53:199-205.

Mayer, D.F., J.D. Lunden and L. Rathbone. 1987. Evaluation of insecticides for Thrips (abaci (Thysanoptera: Thripidae) and effects of thrips on bulb onions. J. Econ. Entomol.ZQ(A): 930-932.

McMurty, J.A ., C.B. Huffaker and M. van de Vrie. 1970. Ecology of Tetranychid mites and their natural enemies: A review. Hilgardia 40(11): 331-390.

McSorley, R. and E.E. Welter. 1991. Comparison of soil extraction methods for nematodes and microarthropods, p. 201-208.In D A. Crossley et al. (ed) Modem techniques in soil ecology. Elsevier, New York.

Metcalf R.L., 1980. Changing role of insecticides in crop protection.Ann. Rev. Entomol. 25: 219-256.

Miller, P.M. 1967. Stimulating emergence of larvae q î H eterodera tabacum with fungicides. Plant Disease Reporter. 51(3): 202-206.

Miller, P.M. and G.S. Taylor. 1967. Effects of fungicides on hatching, survival, and nematodal killing of egges ofHeterodera tabacum in the field. Plant Disease Reporter 51(7): 609-613.

164 NCngochi, D.S. and A. Jensen. 1986. Pests and Disease in tomato cultivars in Zambia, their seasonal occurrence and possible control.Acta Horticulturae 190; 131-138.

Minton, N. A. 1986. Impact of conservation tillage on nematode populations.J.NematoL 18(2): 135-140.

Mishra, P.N. 1984. Studies on bio-efiScacy of some insecticides against the pest complex of tomato,Lycopersicon esculentum Mill., var pusaruby. Madras. Agric. J. 71(10): 673-676.

Mizell m , R.F. and D £ SchiflBiauer. 1990. Effects of pesticides on pecan aphid predator Chrysoperla rufilabris (Neuroptera: Chiysopidae), Hippodemia convergens, Cyclomda sanguinea (L.), Olla v-nigrum (Coleoptera: Coccinellidae), and Aphelirtus perpallidus (Hymenoptera: Encyrtiadae). J. Econ. Entomol. 83(5): 1806-1812.

Mizell m, R.F. and M C. Sconyers. 1992. Toxicity of imidacloprid to selected arthropod predators in the laboratory. Florida Entomologist 75(2): 277-280.

Moctahedi, H, G.S.Santo and R.E.Ingham. 1993. Suppression oïMeloidogyne chitwoodi with sudangrass cultivars as green manure. J. Nematol. 25(2): 303-311.

Morrison, D.E., J.R. Bradley and J.W. Van Dyun. 1979. Populations of com earworm and associated predators after applications of certain soil-applied pesticides to soybeans.,/. Econ. Entomol. 72(1): 97-100.

Morse, J.G. and O.L. Brawner. 1986. Toxicity of pesticides to Scirtothrips citri (Thysanoptera: Thripidae) and implications to resistance management. J. Econ. Entomol. 79(3): 565-570.

Morse, J.G. and N. Zareh. 1991. Pesticide-induced hormoligosis of citrus thrips (Thysanoptera: Thripidae) fecundity./. Econ. Entomol. 84(4): 1169-1174.

Muckenfuss, A.E. and B.M. Shepard. 1994. Seasonal abundance and response of diamodback moth, Flutella xylostella (L.) (Lepidoptea: Plutellidae), and natural enemies to esfenvalerate and Bacillus thuringiensis Berliner in Coastal South Carolina.J.Agric. Entomol. 11(4): 361-373.

Nakashima, M.J. and B.A. Croft. 1974. Toxicity of benomyl to the life statges of Ambyseius fallacis. J. Econ. Entomol. 67(5): 675-677.

Nasmddin, A and D.R. Smitley. 1991. Relationship o ïFranliniella occidentalis (Thysanoptea: Thripidae) population density and feeding injury to the frequency of applications to gloxinia.J. Econ. Entomol. 84(6): 1812-1817.

165 Napit, K.B., G.W. Norton, R.F. Kazmierczak and E.G. Rajotte. 1988. Economic impacts of extension integrated pest management programs in several states. J. Econ. Entomol. 81(1): 251-256.

Newsom, L.D. 1967. Consequences of insecticide use on nontarget organisms.Ann. Rev. Ent.: 257-287.

Newsom, L.D. and C.E. Smith. 1949. Destruction of certain insect predators by applications of insecticides to control cotton pests.J. Econ. Entom. 42(6): 904- 908.

Nyfeiler, M., D. A. Dean and W.L. Sterling. 1989. Prey selection and predatory importance of orb-weaving spiders (Araneae: Araneidae, Uloboridae) in Texas cotton.Environ. Entomol. 18(3): 373-380.

Nyfeiler, M., D A. Dean and W.L. Sterling. 1992. Diets, feeding specialization, and predatory role of two lynx spiders, Oxyopes salticus and Peucetia viridcms Araneae: Oxyopidae), in a Texas cotton agroecosystem.Environ. Entomol. 21(6): 1457-1465.

Oatman, E.R., J.A. Wyman, R.A Van Steenwyk and M.W. Johnson. 1983. Integrated control of the tomato huitworm (Lepidoptera: Noctuidae) and other lepidopterous pests on fresh-market tomatoes in Southern California.J. Econ. Entomol. 76: 1363-1369.

Ghep, J., R.T. McMillian, H.H. Bryan and D.J. Cantliffe. 1984. Control of Damping-oflf of tomatoes by incorporation of fungicides in direct-seeding gel. Plant Disease 68(1): 66-67.

Parmelee, R.W. and D.G.Alston. 1986. Nematode trophic structure in conventional and no-tillage agroecosystems.J.NematoL 18(3): 403-407.

Parmelee, R.W., KS.Wentsel, C.T. Phillips, M. Siminis and R.T. Chickai.1993. Soil microcosm for testing the effects of chemical pollutants on soil fauna communities and trophic structure. Æ/rv/ro/r. Toxicol. Chem. 12:1477-1486.

Patel, K.G. and H.N. Vyas. 1985. Ovicidal evaluation of certain insecticides against the eggs of green lacewing, Chrysopa scelestes Banks-an important predator-under laboratory condition.Indian J. Ent. 47(1): 32-35.

Penman, D.R. and R.B. Chapman. 1988. Pesticide-induced mite outbreaks: Pyrethroids and spider mites. Exp. Appl. Acarol. 4: 265-276.

166 Pennypacker, S.P., L.V. Madden and A. A. Macnab. 1983. Validation of an early blight forecasting system for tomatoes.Plant Disease 67(3); 287-289.

Perfecto, Ivette and A. Sediles. 1992. Vegetational diversity, ants (Hymenoptera: Formicidae), and herbivorous pests in a neotropical agroecosystem.Environ. Entomol. 21(1): 61-67.

Pemezny, K., L.E. Datnoflf and J.CoUins. 1994. Yield losses associated with foliar diseases of fresh-market tomatoes in Florida and benefits of protectant fimgicides. Phytopathology 1141 (abstr.)

Pemezny, K., L.E. Datnofi^ T. Mueller and J. Collins. 1996. Losses in fresh-market tomato production in Florida due to target spot and bacterial spot and the benefits of protectant fungicides. Plant Disease 80(5); 559-563.

Perring T. M., C.A. Farrar, and N.G. Toscano. 1988. Relationship among planting date, potato aphids (Homoptera: Aphididae), and natural enemies. J. Econ. Entomol. 81(4): 1107-1112.

Pike, K.S. and D. Allison. 1987. Effects on twospotted spider mite of insecticides applied to sweet com for control of com earworm.J. Agric. Entomol. 44(4): 327-332.

Pimentel, D., E C. Terhune, W. Dritschillo, D. Callahan, N. Kiimer, D. Nafus, R.Peterson, N. Zareh, J. Misiti and O. Haber-Schaim. 1977. Pesticides, insects in foods, and cosmetic standards. Bioscience 37(3): 178-185.

Pimentel, D. L. McLaughlin, A Zepp, B. Lakitan. T. Kraus. P. Kleinman, F. Vancini. W. J. Roach, E. Graap, W.S. Keeton and G. Selig. 1993. Environmental and economic effects of reducing pesticide use in agriculture. Agriculture, ecosystems and environment. 46: 273-288.

Pimentel, D. 1995. Amounts of pesticide reaching target pests: environmental impacts and tX\ùcs. Journal o f agricultural and environmental ethics. 8(1): 17-29.

Poe, S.L. and P H. Everett. 1974. Comparison of single and combined insecticides for control of tomato pinworm in Florida.J. Econ. Entomol. 67(5): 671-674.

Poehling, H.M. 1988. Influence of cereal aphid control on aphid control on aphid specific predators in winter wheat. Ew/omo/.Gener. 13(3/4): 163-174.

Pohronezny, K., V.H. Wadill, D.J. Schuster and R.M. Sonoda. 1986. Integrated pest management for Florida tomatoes.Plant Disease 70(2): 96-102.

167 Potter, D.A., MC. Buxton, C.T. Redmond, C.G. Patterson and A.J. Powell. 1990. Toxicity of pesticides to earthworms (Oligocheta; Lumbricidae) and effect on thatch degredation in Kentucky bluegrass turf. J. Econ. Entomol. 83(6): 2362- 2369.

Powell, W ., G.J. Dean and R. Bardner. 1985. Effects of primicarb, dimethoate and benomyl on natural enemies of cereal aphids in winter wheat. Ann. Appl. Biol. 106: 235-242.

Powell, W., G.J. Dean and A. Dewar. 1985. The influence of weeds on polyphagous arthropod predators in winter wheat. Crop Protection. 4(3): 298-312.

Poysa, V ., R.A. Brammal and R.E. Pitblado. 1993. Effects of foliar fungicide sprays on disease and yield of processing tomatoes in Ontartio. Ccm. J. Plant. Sci. 73: 1209- 1215.

Pre, D.J. and A.C. Hagley. 1985. Toxicity of pesticides to Chrysopa oculata Say (Neuroptera: Chiysopidae). J. Econ. Entomol. 78(1): 129-132.

Provencher, L. and D. Coderre. 1987. Functional response and switching of Tetragnatha laboriosa Hentz (Araneae: Tetragnathidae) and Clubiona pikei Gertsh (Araneae: Clubionidae) for the aphids Rhopalosiphum maidis (Fitch) and Bhopalosiphum padiÇ^.) (Homoptera: Aphididae). Environ. Entomol. 16(6): 1305-1309.

Pullen, M.P., E.LZehr and GJE.Carter. 1990. Influences of certain fungicides on parasitism of the nematode, Cricinimella xenoplax by the fungus Hirsutella rhossilensis. Phytopathology 80: 1142-1146.

Quaglia, F., E. Rossi, R Petacchi and C.E. Taylor. 1993. Observations on an infestation by green peach aphids (Homoptera: Aphididae) on greenhause tomatoes in Italy. J. Econ. Entomol. 86(4): IOI9-I025.

Redcliflfe, E.B., RW. Weires, RE. S tucker and D.K. Barnes. 1978. Influence of cultivars and pesticides on pea aphid, spotted alfalfa aphid, and associated arthropod taxa in a Minnesota alfalfa ecosystem.Environ. Entomol. 5(6): 1195-1207.

Reed, J.P., F.R Hall and RM. Riedel. 1993. Biological implications of drift from sprayers in tomato fungicide field trials. Plant Disease 77(2): 186-189.

Riddick, E.W. and N.J. Mills. 1995. Seasonal activity of carabids (Coleoptera: Carabidae) affected by microbial and oil insecticides in an apple orchard in California. Environ. Entomol. 24(2): 361-366.

168 Riechert, S.E. and T. Lockl^, 1984. Spiders as biological control agents.Arm. Rev. Entomol. 29: 299-320.

Riggs, R.D. and L.R. Oliver. 1982. Eflfect of trifluralin (treflan) on soybean cyst nematode. J. Nematol. 14(4): 466.

Ripper, W.E. 1956. Effects of pesticides on balance of arthropod populations.Ann. Rev. Ent. 1:403-439.

Rodriguez-Kabana,R. 1986. Organic and inorganic nitrogen amendments to soil as nematode suppressants. J.Nematol. 18(2): 129-135.

Roger, C., D.Coderre and C.Vincent. 1994. Mortality and predation efGciency of Coleomegilla maculata lengi (Coleoptera: Coccinellidae) following pesticide applications. J. Econ. Entomol. 87(3): 583-588.

Roger, C., D. Coderre and C. Vincent. 1991. Apparent mortality ofColeomegilla maculata (Coccinellidae) following pesticide treatment: possibility of overlooking predators. In Behaviour and Impact o f Aphidophaga:‘i29-326, edts: L.Polgar, R.J. Chambers, A.F.G. Dbcon and I. Hodek.

Roltsh, W.J. and M.A. Mayse. 1983. Parasitic insects associated with Lepidotera on fresh- market tomato in Southeast Arkansas. Environ. Entomol. 12(6): 1708-1713.

Rovesti,L and K.V. Deseo. 1990. Comparability of chemical pesticides with the entomopathogenic nematodes,Steimema carpocapsae Weiser and S. feltiae Filipjev. (Nematoda: Steimematidae). Nematologica. 36: 237-245.

Salas, J. 1992. Integrated pest-management program for tomato crops (Lycopersicon esculentum Mill.) Lara State, Venezuella. Acta Horticulturae 301: 199-203.

Navas, V.E.S. J.E. Funderburk, R.J. Beshear, S.M. Olson and T.P. Mack. 1991. Seasonal patterns ofFranliniella spp. Thysanoptera: Thripidae) in tomato flowers.J. Econ. Entomol. 84(6): 1818-1822.

Samu, F. G.A. Matthews, D. Lake and F. VoUrath. 1992. Spider webs are efiBcient collectors of agrochemical spray.Pestic. Sci. 36: 47-51.

Schmitt, D P. and F.T. Corbin. 1981. Interaction of fensulfothion and phorate with preemergence herbicides on soybean parasitic nematodes. J. Nematol. 13(1): 37- 41.

169 Schmitt, D P. F.T. Corbin and L.A. Nelson. 1983. Population dynamics oiHeterodera glycines and soybean response in soils treated with selected nematicides and herbicides. J. Nematol. 15(3): 432-437.

Schuster, D.J. 1978. Tomato pinworm: chemical control on tomato seedlings for transplant. J. Econ. Entomol. 71: 194-196.

Schuster, D.J., C.A. Musgrave and J.P. Jones. 1979. Vegetable leafininer and parasite emergence fi'om tomato foliage sprayed with oxamyl.J. Econ. Entomol. 72(2): 208-210.

Schuster, D.J., J.F. Price and P.H. Everett. 1981. Insecticides for management of insect pests on tomatoes in Florida. IF AS, Bradenton AREC Research Report, 9 pp.

Schuster, D.J. and J.F. Price. 1985. Impact of insecticides on lepidopterous larval control and leafininer parasite emergence on tomato.Proc. Fla. State Hort. Soc. 98: 248- 251.

Schweizer, E.E., D.W. Lybecker and R.L. Zimdahl. 1988. Systems approach to weed management in irrigated crops. Weed Science 36: 840-845.

Sell, P. 1985. Leistungan des blatlausraubers Aphidoletes aphidimyza (Rond.) (Diptera: Cecdomyiidae) nach einwirkung von insectizid-und fiingizid-belagen auf zwei und drei tage alte larven. Journal o f Plant Diseases and Protection. 92(2): 157-163.

Shanks, CH., A.L. Antonelli and B. D. Congdon. 1992. Effect of pesticides on twospotted mite(Acari: tetranychidae) populations on red raspberries in western Washington. .4gr/cM/ft/re, Ecosystems and Environment 3%: 159-165.

Shean B. and W.S. Cranshaw. 1991. Diffemtial susceptibilities of green peach aphid (Homoptera: Aphidida) and two endoparasitoids (Hymenoptera: Encyrtidae and Braconidae) to pesticides. J. Econ. Entomol. 84(3): 844-850.

Shires, S.W. 1985. A comparison of the effects of cypermethrin, parathion-methyl and DDT on cereal aphids, predatory beetles, eartworms and litter decomposition in spring wheat. Crop Protection 4(2): 177-193.

Shorley, H.H., H.T. Reynolds and I. D. Anderson. 1962. Effect of zectran, sevin and other new carbamate insecticides upon insect populations found on vegetable and field crops in Southen California.J. Econ. Entomol. 55(1): 5-11.

Showier, A.T. and T. E. Reagan. 1991. Effects of sugarcane borer, weed, and nematode control strategies in Luisiana sugarcane. Environ. Entomol. 20(1): 358-370.

170 Singh, K.J. and OJ*. Singh. 1991. EfiBcacy and economics of some insecticides against jassid, Aphelioina maculosa Dist. Infestation on soybean.J. Insect. .Sc/.4(2): 187- 189.

Singh, PJ*. and G.C. Varma. 1986. Comparative toxicities of some insecticides to Chrysoperla camea (Chiysopidae: Neuroptera) and Trichogramma brasiliensis Trichogrammatidae: Hymenoptera), two arthropod natural enemies of cotton Agriculture, Ecosystems and Environment 15: 23-30.

Sipes, B.S. and DP. Schmitt. Development ofH. glycines as affected by alachlor and fenamiphos.J. Nematol. 21(1): 24-32.

Small, R.W. 1979. The effects of predatory nematodes on populations of plant parasitic nematodes in pots. Nematologica 25: 94-103.

Smith, R.G., TX. Archer and E.D. Bynum. Low-rate applications of insecticides for greenbug (Homoptera: Aphididae) control and effect on beneficial insects on sorghum. J. Econ. Entomol. 78(4): 988-991.

Smith, T.M. and G.W. Stratton. 1986. Effects of synthetic pyrethroid insecticides on nontarget organisms.Residue Reviews. 97: 93-120.

Smolik, J.D. 1983. Effect of nematicide treatments on nontarget nematode populations associated with com. Plant Dis. 67(1): 28-30.

Snipes, C.E., R.H. Walker, T. Whitwell, G. A. Buchanan, J.A. McGuire and N.R. Martin. 1984. Efficacy and economics of weed control methods in cotton(Gossypium hirsutum). Weed science. 32: 95-100.

Sotherton, N.W ., S.J. Moreby and M.G. Langley. 1987. The effects of the foliar fungicide pyrazophos on beneficial arthropods in barley fields. Ann. Appl. Biol. 111:75-87.

Sotherton, N.W. and S.J. Moreby. 1988. The effects of foliar fungicides on beneficial arthropods in wheat ^LeXésElntomophaga. 33(1): 87-99.

Stam, P.A., D.F. Glower, J.B. Graves and P.E. Schilling. 1978. Effects of certain herbicides on some insects and spiders found in Louisiana cotton fields. J. Econ. Entomol. 71(3): 477-480.

Stark, J.D. 1992. Comparison of the impact of a neem seed-kerael extract formulation, Margosan-0’ and chlorpyrifos on Non-target invertebrates inhabiting turf grass. Pestic. Sci. 36: 293-299.

171 Sterk, G. 1993. Studies on the effects of pesticides on the beneficial arthropods. Acta Horticulturae 347: 223-243.

Stirling,G.R. and R. Mankau. 1978. Parasitism oïMeloidogyne eggs by a new fungal parasite./.Aema/o/. 10(3): 236-240.

Subagja, J and R.J. Snider. 1981. The side effects of the herbicides atrazine and paraquat upon Falsomia C andida and Tulibergia gram lata (Insecta, Collembola). Pedobiologia-. 22: 141-152.

Sunderland, K.D., A_M. Fraser and A.F.G. Dixon. 1986. Field and laboratory studies on money spiders (Linyphiidae) as predators of cereal aphids. J. Appl. Ecol. 23: 433- 447. Takele, E., J.A. Menge, J_E. Pehrson, J.L.Meyer. C.W. Coggins, M.L. Arpaia, J.DUare, D.R. Atkin and C. Adams. 1993. Economic analysis of integrated crop management practices of navel oranges.J. Amer. Soc. Hort. Sci. 118(6): 910-915.

Terry, L.A., D.A.Potter and P.G.Spicer.l993. Insecticides affect predatory arthropods and predation on Japenese beetle (Coleoptera: Scarabaeidae) eggs and Fall armyworm (Lepidoptera: Noctuidae) pupae in turfgrass. J.Econ. Entomol. 86(3): 871-878.

Thomas, C.F.G., E.H.A. Hoi and J.W. Everts. 1990. Modelling the diflfusion component of dispersal during recovery of a population of linyphiid spiders fi*om exposure to an insecticide. Funct. Ecol. 4: 357-368.

Townshend, J.L. 1990. Growth of carrot and tomato fi’om oxamyl-coated seed and controloïMeloidogyne hapla. J.Nematol. 22(2): 170-175.

Trumble, J.T., W.G. Carson and K.K White. 1994. Economic analysis o ï a Bacillus thuringiensis-h?iseà integrated pest-management program in fi’esh market tomatoes.J. Econ. Entomol. 87(6): 1463-1469.

Trumble J. T. and J.P. Morse. 1993. Economics of integrating the predaceous mite Phytoseilus persimilis (Acari: Phytoseiidae) with pesticides in strawberries. J. Econ. Entomol. 86(3): 879-885.

Trumble, J.T., and B. Alvardo-Rodriguez. 1993. Development and economic evaluation of an IPM program for fi’esh market tomato production in Mexico.Agriculture, Ecosystem and Environment. 43: 267-264.

Walgenbach, J.F., P.B. Shoemaker and K. A. Sorenson. 1989. Timing pesticide applications for for control ofHeliothis zea (Boddie) (Lepidoptera: Noctuidae), Altemaria solani (Ell and G.Martin),and Phytopthora infestans (Mont.) De Bary, on tomatoes in Western North Carolina.J. Agric Entomol. 6(3): 159-168.

172 Walgenbach, J.F., R.B. Leidly and TJ. Sheets. 1991. Persistence of insecticides on tomato foliage and implications for control of tomato fruitworm (Lepidoptera: Noctuidae). J. Econ. Entomol. 84(3): 978-986.

Walgenbach, JJ. and E.A. Estes. 1992. Economics of insecticide use on staked tomatoes in Western North Carolina.J. Econ. Entomol. 85(3): 888-894.

Walker, GJ*., L.R. Nault and D.E. Simonet. 1984. Natural mortality factors acting on potato aphid (Macrosiphum euphorbiae) populations in processing-tomato fields in Ohio. Environ. Entomol. 13: 724-732.

Walker, G.P., L.V.Madden and D.E. Simonet. 1984. Spatial dispersion and sequential sampling of the potato aphid,Macrosiphum euphorbiae (Homoptera: Aphididae), on processing tomatoes in Ohio.Can. Ent. 116: 1069-1075.

Watterson, J. C. 1985. Tomato diseases. A practical guide for seedsman, growers and agricultural advisors. Petoseed Co., Inc. 47 pp.

Way, M.J. and K.C.Khoo.1992. Role of ants in pest management.Ann. Rev.Entomol. 37: 479-503.

Webb, R.E., M. Shapiro, J.D. Podgwaite, R.C. Reardon, K.M. Tataman, L. Venables and D.M. Kolodny-Hirsch. 1989. Effect of aerial spraying with dimilin, dipel, or gypcheck on two natural enemies of the gypsy moth (Lepidoptea: Lymantridae). J. Econ. Entomol. S2(6): 1695-1101.

Weiss, M.J., P. Mcleod, B.G. Schatz and B. K. Hanson. 1991. Potential for insecticidal management of flea beetle (Coleoptera: Chrysomelidae) on Canola.J. Econ. Entomol. 84(5): 1597-1603.

Welter, S.C., M.V. Johnson, N.C. Toscano, T.M. Perring and L.Varela. 1989. Herbivore effects on fr-esh and processing tomato productivity before harvest.J. Econ. Entomol. 82(3): 935-941.

Welter, S.C., W.S. Seaman, C.E. Engle and M.M. Barnes. 1987. Reduced insecticide use in almonds through selective application . J. Econ. Entomol. 80(3): 650-654.

Whalon, M.E. and E.A. Eisner. 1982. Impact of insecticides in Illinoia pepperi and its predators, y. Econ. Entomol. 75: 356-358.

White, T.C.R. 1984. The abundance of invertebrate herbivores in relation to the availability of nitrogen in stressed food plants. Oecologia. 63: 90-105.

173 Whitford, F., W.B. Showers and G 3 .Edwards. 1987. Insecticide tolerance of ground- and foliage-dwelling spiders (Araneae) in European com borer (Lepidoptera: Pyralidae) action sites. Environ. Entomol. 16: 779-185.

Whitford, F and W 3. Showers. 1988. Abundance of nontarget foliage-dwelling insects after insecticide treatment for adult European com borer (Lepidoptera: Pyralidae) J. Econ. Entomol. 81(5): 1323-1327.

Wiesenbom, W.D., J.T. Trumble and E.R. Oatman. 1990. Economic comparison of insecticide treatment programs for managing tomato pinworm (Lepidoptera: Gelechiidae) on fell tomatoes.J. Econ. Entomol. 83(1): 212-216.

Wightman, J.A, M.M. Anders, V. R. Rao and L.M. Reddy. 1995. Management of Helicoverpa armigera (Lepidoptera: Noctuidae)on chickpea in southem India: thresholds and the economics of host plant resistance and insecticide application.

Wilcut, J.W., G.R. Wehtje. 1987. Economics of weed control in peanuts(Arachis hypogaea) with harbicides and cultivations. Weed Science. 35: 711-715.

Williamson, J.W. and J.W. Hilty. 1988. Comparison of a forecast-generated system and a weekly spray regimen for early blight control in trellised tomatoes. Tennessee Farm and Home Science 146: 8-12.

Wolk, J.O., D.W. Kretchman and D.G. Ortega. 1983. Response of tomato to defoliation. J. Amer. Soc. Hort. Sci. 108(4): 536-540.

Wright, E.J. and J.E. Laing. Numerical response of coccinellids to aphids in com in southem Ontario. Can. Ent. 112: 977-988.

Vavrek, R.. C. and H.D. Niemczyk. 1990. Effect of isofenphos on nontarget invertebrates. Environ. Entomol. 19(5): 1572-1577.

Vemon, R.S. and J R. Mackenzie. 1991. Evaluation of foliar sprays against the tuber flea beetle, Epitrix tuberis Gentner (Coleoptera: Chrysomelidae), on potato.Can. Ent. 123:321-331.

Vickerman, G.P. and K.D. Sunderland. 1977. Some effects of dimethoate on arthropods in winter wheat. J. Appl, Ecol. 14: 767-777.

Yeates, G.W. and R.A. Prestige. 1986. Nematode populations and their effects on herbage production in a volkanic plateau pasture. New Zealand Journal o fAgricultural Research 29: 517-523.

174 Yeates,G.W. and G.M. Barker. 1986. Influence of repeated oxamyl application on nematode populations beneath 4 rygrass cultivars on a yellow-brown loam. New Zealand Journal o f Agricultural Research 29: 501-505.

Yeates, G.W., D.A. Wardle and R.N.Watson. 1993. Relationships between nematodes, soil microbial biomass and weed management strategies in maize and asparagus cropping systems. Soil. Biol. Biochem. 25(7): 869-876.

Yu, S.J. 1988. Selectivity of insecticides to the spined soldier bug (Heteroptera: Pentatomidae) and its lepidopterous prey. J. Econ. Entomol. 81(1): 119-122.

Yun, Y.M. and R.F. Ruppel. 1964. Toxicity of insecticides to a coccinellid predator of the cereal leaf beetle. J. Econ. Entomol. 57(4): 835-837.

Zhou, X., and N. Carter. 1991. The effects of nitrogen and fungicide on cereal aphid population development and the consequences for the aphid-yield relationship in winter wheat. Ann. Appl. Biol. 119: 433-441.

Zimmerman, R.J. and W.S. Cransshaw. 1990. Compatibility of entomogenous nematodes (Rhabditia) in aqueous solutions of pesticides used in turfgrass meintenance. J. Econ. Entomol. 83(1): 97-100.

Zoebelein, G. 1988. Long-term field studies about pesticide effects on ladybird beetles (Coleoptera: Coccinellidae).Entomol. Gener. 13(3/4): 175-187.

175