.E v v. 5 : 1. .
;:._.L..,.5 fl... Ramps.
,

 

 

.

. has
.- u. . r 1
.star.» u. 7 .
riftfl. ? _ .4
a: 1%?!
.30’ ‘ ‘1'.
.i I . I
.jr \ V V

 

}
I

.. l ‘ £0
.1... x. . .P.r.«tfi2 41
3:..{5 St: 1..
. \I—Cc t,
. tat.
.. 3......
v. .. (I .
5‘ i

, a.

a 9?...

1‘3...“
5:14;».

A... and.

r...

 

.sfi .4 : ,
_. .
:7 h... r 92%.?
I .11....

 

      

hfinfifl. i

 
  

THESIS

2003

This is to certify that the
thesis entitled

A DATABASE OF ARTHROPODS RESISTANT TO
INSECTICIDES AND MITICIDES

presented by

PATRICK S. BILLS

has been accepted towards fulfillment
of the requirements for the

M. S. degree in ENTOMOLOGY

 

 

II MéIflgél/CCRK/

Major Professor’ s Signature

7 MAY 2003

 

Date

MSU is an Afiirmative Action/Equal Opportunity Institution

‘—.-o-0—o-o-o-n—-—-o-.—-—--o----n-u-c-n-u-n—n-o--—---—-—o-o-—n—.—n—o--—-—.—n--o-o—c-o-c-o-p-o-o-o-0-.--o-o-o—o-o-o-o-

 

 

LIBRARY
Michigan State
University

 

 

 

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

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/ClRC/DateDue.p65-p.15

,fi—.—_ __ _

A DATABASE OF ARTHROPODS RESISTANT TO INSECTICIDES AND
MITICIDES

By

Patrick S. Bills

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
Department of Entomology

2003

ABSTRACT

A DATABASE OF ARTHROPODS RESISTANT TO INSECTICIDES AND
MITICIDES

By
Patrick 8. Bills

The resistance of pest organisms to the toxic effects of pesticide is a world-
wide phenomenon defined as “the microevolutionary process of genetic
adaptation through the selection of biocides.” The severe consequences of
resistance have led governments to recommend the development of a central
and permanent ‘data bank’ of resistance information. We designed and
constructed a computerized database using relational database theory to meet
this need. Our measurement of this dynamic, genetic phenomenon was a ‘case’
of resistance: the first documented and peer-reviewed occurrence of a species
resistant to a specific pesticide formulation in a country or region. The
information was drawn from previous authors and a search of the literature for
cases of resistance since the last published tallies. As of 2002, there were 543
arthropod species resistant to one or more of 315 pesticide formulations in at
least one part of the world.

Taking advantage of the relational data model, the resistance database was
linked with the US EPA Pesticide Product Information System to compare
resistance occurrence in the United States .with Pesticide Registration events.
The results show a strong correlation between the exponential rise of resistance
cases in the 19603 and 1970s with pesticide registration patterns, as well as the

more recent decline in the rate of new resistance cases.

To Terri

ACKNOWLEDGEMENTS

First and foremost, I want to express my gratitude to Dr. Mark Whalon for
taking me on as a under-graduate employee ten years ago and granting to me
more challenges and opportunities than I could have ever imagined. This project
was just one of many entrusted to me, all of which will have a profound effect on
my future. This project was based on his tremendous vision. Dr. Whalon
supported me through many tough times and was always generous. Second, I
wish to thank Dr. David Mota-Sanchez. He provided the bulk of the literature
review that makes up the data for this project. I could not have accomplished so
much alone. Unquestionably without his energy and expertise this would not
have happened.

Many undergraduate assistants helped in nearly all aspects of this project
including data entry, literature searches, and reference management. Primary
among these was Ms. Robbie Wong. Andrea Biasi Coombs, who helped me in
countless projects throughout our tenure in the Whalon lab, along with Michael
Bush and Utami Rahardja DiCosty created an atmosphere of camaraderie that
made projects like this one possible.

I’d like to thank the members of my committee for their support and trust to the
drawn out end. First, Dr. Stuart Gage who never failed to provided thought-
provoking, big-picture ideas and showed me the point where Computer Science
and Entomology come together. Second Dr. Bob Hollingworth who provided a

unequalled example of a rigorous and sharp-witted scientist.

PREFACE

World-wide sales for insecticides exceeded $98 in 1999 (Donaldson, 2002
#10), and in the year 2000 insecticide and acaricide (miticide) applications
amounted to over 79 thousand metric tons (WAICENT, 2003). Estimates for all
pesticides are substantially higher. The FAO estimates 1.5 billion hectares of the
world’s land were used for agriculture 1994-1996 (Fischer, 2002 #7).

Resistance is the genetic based, evolutionary response of insects and mites to
the pesticide suppression strategy. Arthropods are without rival in published
documentation of their evolutionary ability to develop resistance with over 543
species resistant to one or more pesticides reported by 2002.(T able 1, Chapter 2,
page 40)

This thesis documents the construction of, output, and analysis of a
computerized database of historical reports of arthropods resistance to
pesticides. The goal was not only to update the work of previous researchers
who've compiled lists of resistance cases but also to deliver this updated
information in a form more useful to a larger, modern audience. The desire was
to openly and freely share information about resistance with the world via the
lntemet.

This thesis is presented in three chapters and one appendix. The first is an
overview of pesticide resistance definitions, policy related to pesticide resistance,
and definition of a “resistance case,” the unit upon which database of pesticide

resistance is based. Chapter two describes the database: it details the aspects

of pesticide resistance therein, the database construction, pesticide resistance
summaries, and several caveats on its use. The third chapter presents a unique
analysis comparing insecticide resistance, insecticide application statistics and
historical pesticide registration records in the United States. This analysis
combined three distinct databases: the arthropod resistance database, the
USEPA's registration database (Office of Pesticide Programs 2000) and a
pesticide use database from the National Center for Food and Agriculture Policy
(Gianessi and Marcelli 2000). The appendix provides the computer code to re-
create the database. There was not space to print the full contents of the
database: a printed table representing all data had nearly 5,800 rows and 33
columns. As of May 9, 2003 the data was available and searchable on the
lntemet at the website http://whalon|ab.msu.edu

Work on this project was supported in part by the Center for Integrated Plant
Systems of Michigan State University, the USDA CSREES, the Insecticide
Resistance Action Committee (IRAC) of the international agrochemical industry,

and Michigan's Project GREEEN.

vi

TABLE OF CONTENTS

List of Tables ....................................................................................................... viii
List of Figures ....................................................................................................... ix
Chapter 1. defining a measurement of Resistance ............................................... 1
Introduction ................................................................................................... 1
History of Counting Pesticide Resistant Arthropods ..................................... 2
Definitions of Resistance .............................................................................. 3
Behavioristic Resistance .............................................................................. 8
Definition of the unit case of resistance ...................................................... 11
Measuring the Consequences of Resistance ............................................. 13
Pesticide Resistance Policy: a call for a database ...................................... 13
Chapter 2. Design and Summary of a Relational Database of Pesticide Resistant
Arthropods. ......................................................................................................... 16
Introduction ................................................................................................. 16
Materials and Methods ............................................................................... 18
Results ....................................................................................................... 25
Resistance Summary by class .................................................................... 26
Top Twenty Resistant Arthropods .............................................................. 26
Pesticide Resistance Timeline .................................................................... 31
Resistance by Chemical Class ................................................................... 32
Discussion .................................................................................................. 35
Chapter 3. Analysis: Resistance and Registration In The US ............................. 48
Introduction ................................................................................................. 48
Methods ...................................................................................................... 53
Results ....................................................................................................... 55
Discussion .................................................................................................. 55
Appendix 1 .......................................................................................................... 59
Appendix 2. Structured Query Language data definition code (Axmark,
2002)”“Hn”n“Hun“”unuunuuu"uHuHHUHHHHHflHHHHHHHHHHHHHH, .............. 59
Bibliography ........................................................................................................ 68

vii

LIST OF TABLES

Table 1. Summary of Cases of Reported Pesticide Resistance; counts of uniqe
occurrences of species, compounds, and cases along with publications
reviewed ....................................................................................................... 40

Table 2. Summary of Insecticide and Acaracide Resistance .............................. 41
Table 3. Cases of Insecticide and Acaricide Resistance by Chemical Class ...... 45

Table 4. Top Twenty Resistant Arthropods, ranked by number of unique
compounds reportedly resistant ................................................................... 46

viii

LIST OF FIGURES

Figure 1. Entity-relationship diagram of our semantic data model of the listing in
Georghiou and Lagunes-Tajeda, 1991 (PK: Primary Key, Fk: Foreign Key) x

Figure 2 Entity-relationship model of enhanced database. PK = primary key,
F K=foreign key; arrows point in direction of unity ......................................... 39

Figure 3. Accumulation of Arthropod Resistance, 1914-2001: Number of First
Reports of Resistance ifor 3 Species per Year, (fit with Loess curve to
emphasize discontinuities) ........................................................................... 40

Figure 4. Time series of resistance case discover by chemical class for those
classes with > 25 cases of resistance. A case is the first discovery of a
species and pesticidecombination. ............................................................. 42

Figure 5. Frequency of new species with no prior resistance, new pesticides,
reports and new cases by decade ............................................................... 43

Figure 6. Timeline of US Pesticide Registration: total number of unique 'use sites‘
for all pesticides in the resistance database, by Chemical Class. OP =
Organophosphate, CAR = carbamate, PYR = Pyrethroid, OCL =
Organochlorine, BAC = Bacterial, IGR = Insect Growth Regulatro, ABM=
Abemectin. Data to 2000 .............................................................................. 56

Figure 7. Timeline of Arthropod Pesticide Resistance and Pesticide Registrations
in the United States ...................................................................................... 57

CHAPTER 1. DEFINING A MEASUREMENT OF RESISTANCE

Introduction

I n the early part of the 20th century, the first pesticide-resistant arthropod
species, the San Jose scale, Aspidiotus pemiciosus (Comstock) was discovered
to “resist” the toxic effects of lime sulfur in deciduous fruits in the state of
Washington (Melander 1914). Since that time, the phenomenon of arthropods
resisting the effects of pesticides is well documented and intensively studied,
producing hundreds of articles per year (Booth et al 1983). By the year 2002,
there were 543 arthropod species reported to be resistant to one or more of 315
pesticide compounds (Table 1, page 39) One species of phytophagous mite,
Tetranycus urticae (Koch), is resistant or cross-resistant to over 75 pesticide
active ingredients formulations from eight chemical classes (Table 4, page 46).
The magnitude of the resistance problem, having grown in less than a century,
would have us believe that the appearance of pesticide resistant populations
seems to be a common result from the use of pesticides. Few, if any, pest
suppression tactics known have not elicited some form of resistance response in
the target pest. Biological pesticides are no exception-including pesticides
derived from Bacillus thuringiensis ((McGaughy and Whalon 1992), viruses (Fuxa
et al. 1988), and parasitoids (Messenger and Bosch. 1971).

The consequences of pesticide resistance problems are dire enough that
resistance management is an accepted part of any integrated pest management
(IPM) program. The extent of pesticide resistance is too large to be easily
enumerated. There is a need to determine the large-scale status of pesticide

resistance, as “what gets measured, gets managed" (attributed to Peter Drucker).
1

History of Counting Pesticide Resistant Arthropods

To appreciate the magnitude of the resistance problem, it is crucial to
enumerate when and where resistance has occurred. Early reviews of pesticide
resistance included Metcalf (1955) and Busvine (1956, 1957), who published lists
of resistant mosquitoes for the World Health Organization (WHO).
A. W. A. Brown also published tables of resistance cases for the WHO and other
agencies in the 1950s until the early 19705 (Brown 1958)(Brown 1971). These
early reviews focused on human and animal disease vectors, which were the
initial targets of worldwide pesticide application (Brown 1951 ). In the mid-19703
Brown and Croft (1975) introduced the novel concept of using pesticide
resistance to improve IPM by determining systems of compatible natural enemies
and pesticides. Compatible natural enemies are species that would be resistant
enough to survive the application of pesticides to manage pests within an agro-
ecosystem. This culminated in a database (SELECTV) of pesticide resistant or
pesticide tolerant biocontrol agents such entomophagous or parasitoid
arthropods names (Croft 1990), (Theiling and Croft 1988). The SELECTV
database has been subsequently updated and portions are available from
Oregon State University (Jepsen and Hennigan 2000).

The penultimate publication to the current database, initiated at the request of
the United Nations Food and Agriculture Organization (UN FAQ), is a thorough
review of resistant arthropod research (Georghiou and Lagunes—Tejeda 1991).

This text documented 511 species that are resistant to one or more compounds

in one or more regions (states, provinces, and countries), covering over 200
pesticide compounds based on 1,263 cited references (Table 1, page 39).
Definitions of Resistance

To enumerate occurrences of pesticide resistance, resistance must be clearly
defined. Today the debate over resistance definitions rages on stronger than
ever, perhaps exacerbated by our increasing technical ability to detect the alleles
of an organism that confers resistance to, or increased tolerance for, a particular
pesticide or class of pesticides. A working definition must be chosen to set the
criteria for inclusion of a resistance report into the database.

The term “resistance" comes from the title of the first modern article
documenting this phenomenon by A. Melander: “Can insects become resistant to
sprays?” (Melander 1914) One of the first explicit definitions came from a panel
of WHO - experts who defined resistance as “the development of an ability in a
strain of insects to tolerate doses of toxicants which would prove lethal to the
majority of individuals in a normal population of the same species” (World Health
Organization 1957). However, it may be very difficult to find the “normal
population” required by this definition, after more than 60 years of selection
pressure from synthetic insecticide applications around the world ((Otto et al.
1992). In addition, the WHO definition says nothing about individuals resistant to
pesticides — resistance in their definition is treated strictly as a population-level
phenomena. This distinction has significance today because new techniques
can detect the presence of resistant alleles in individuals (ffrench-Constant and

Roush 1990). Screening for a very low frequency of resistance alleles, crucial for

resistance management in genetically engineered plant pesticides such as
Bacillus thuringiensis (Bt) toxin producing crops (Zhao et al. 2002a), would not fit
the WHO definition.

J. F. Crow (1960) presented a more inclusive definition of resistance that
considers single individuals as well as populations. He proposed “resistance
marks a genetic change in response to selection" (Crow 1960). This definition is
not restricted to high resistance levels: incipient resistance was included.
However, the most significant consequence of pesticide resistance was missing:
field-failure. While Crow’s definition gets to the genetic cause of resistance, it
does not identify the selecting agent, direction or result of selected genetic
change. While Crow’s statement is true, it falls short of a modern understanding
of resistance management because it’s possible to use selection to both increase
and decrease resistant allele frequencies in a population along a continuum from
resistance to susceptibility (McKenzie, 1996 #75, pp 89-121 ).

In 1987, R. M. Sawicki focused Crow's definition closer to an operational
sense of resistance by re-incorporating the notion of field-failure present in the
WHO definition, “Resistance marks a genetic change in response to selection by
toxicants that may impair control in the field” (Sawicki 1987). That is, a genetic
change that leads to increased frequency of resistant individuals that results in a
control problem. Note, however, that Sawicki was careful to consider the
possibility that resistance may or may not reduce the level of pests controlled.

By this definition, therefore, strains of organisms that are selected for pesticide

resistance in the laboratory are considered resistant.

In 1986, the introduction to a classic publication on resistance from the
National Academy of Sciences states: “resistance is a consequence of basic
evolutionary processes” (Glass 1986b). This definition mirrors Crow's statement
above, but adds that resistance is simply an extension of existing natural
processes. That is, it should come as no surprise that insects and other
herbivorous arthropods have become adapted to synthetic or natural toxins when
many of these same species feed unaffected on plants defended by secondary
defense chemistry deadly to most other organisms (Croft and Brown 1975).

Whalon and McGaughy defined resistance as “the microevolutionary process
of genetic adaptation through the selection of biocides” (Whalon and McGaughy
1988). This clearly indicates the type of change, the agent of change, and the
direction. Toxicants and “biocides” are essentially equivalent, however the term
“biocides” is more closely related to our notion of a deliberately applied pesticide,
whereas toxicants might be thought of as naturally occurring plant secondary
defensive compounds. By mentioning 'selection', they imply reduced action of
the biocide, but as with Crow (1960), no mention of the conditions under which
the selection process is made. Clearly this definition includes both lab-selected
populations and field populations among pesticide resistant species.

The agrochemical industry has not been idle in the effort to understand,
define, monitor, and manage pesticide resistance. The exponential increase in
the world-wide cases of resistance during the first three-quarters of this century,
combined with scientific and public pressure, led the pesticide industry to form

various “Resistance Action Committees” including one for insecticides (I RAC),

fungicides (FRAC), and herbicides (HRAC). These resistance action committees
focus on various aspects of resistance management, especially monitoring
resistance. lRAC’s own definition of pesticide resistance, quoted directly is

(IRAC 1997):

An insect should only be viewed as resistant when:

0 The product for which resistance is being claimed carries a use
recommendation against the particular pest mentioned, and has a history
of successful performance.

. Product failure is not a consequence of incorrect storage, dilution or
application, and is not due to unusual climatic or environmental
conditions.

. The recommended dosages fail to suppress the pest population below
the level of economic threshold.

. Failure to control is due to a heritable change in susceptibility of the pest
population to the product.

Mai; Sensitivity to an insect or mite control agent must decrease significantly
before field failure is experienced. The term “resistance” should only be used

once field failure has occurred and been confirmed.

The Committee’s definition stressed that the term “resistance" applies only
when field failure is confirmed. Although the IRAC criteria were sufficient to
ensure that a pest population had truly developed resistance, the definition is still
problematic for the early detection of resistance, setting the stage for a system of
anecdotal reporting and resistance crisis rather than prevention and
management. The implication here is that detection of low frequencies of
resistant alleles in a population does not warrant a resistance outcry. The IRAC

definition emphasizes the disagreement of whether “field failure” is a necessary
6

condition for the definition and hence discovery of a case of resistance. One
implication of a declaration of resistance is that the pesticide will no longer kill
enough of the pest population below the integrated pest management (IPM)
economic threshold, and hencefonivard that pesticide, and perhaps all others in
with the same mode of action, should be avoided, as it no longer works. On the
other hand, if it continues to be effective, it likely will work poorly, which is very
undesirable in agricultural production or human and animal health protection
systems with very low tolerance for pest damage.

The converse of the requirement for field failure is that lab selected resistance
would fit into the definition. Since the transition from anecdotal reporting to
resistance management would require resistance detection, monitoring efforts
can now include the detection of resistant alleles in a sufficient amount of time to
change management, and avert or ameliorate resistance evolution. Consider a
case in which resistant individuals are present in small numbers and the
recommended dose suppresses the pest population below the economic
threshold. In this instance, there is no detected “field failure" and by the IRAC
definition there is no resistance. But the genetic potential is present for the
frequency of resistant individuals to increase in future generations, leading to
control failure. On the other hand, it would be argued, that even with an increase
in resistant allele frequency, a correct insecticide application would guarantee
reduction of pest populations below an economic threshold. Further more, there
are additional factors aside from pesticide application that may affect reduction of

pest population levels. These factors could include the pest-mediating impacts of

predators and parasites, pest movement and spatial distribution, crop phenology
and susceptibility, weather, pest life-stage (e.g. larval instar), and frequency of
resistant individuals (Roush and McKenzie 1991). That said, it is obvious that
special care has to be taken in resistance definition and it’s interpretation.

However, by the time field applications fail to control a pest population and it is
declared 'resistant,’ it may be too late to implement resistance management
strategies for the pesticide in question. Once the resistant allele frequencies
exceed just 1%, no resistance management tactic will suppress resistance from
overtaking the population (Roush and Miller 1986). This resistance scenario
could also be exacerbated by cross-resistance in the pest to other chemicals with
a similar mode of action. Therefore, early detection of resistance is an important
aspect missing in the IRAC definition.

Finally, a change in allele frequencies may also be a product of deliberate
manipulation, as opposed to a by-product of human inputs into the environment.
Examples include the populations of insects select for resistance to crystals of
Bacillus thuringiensis that have been developed in the laboratory (Bolin et al.
1999); (Zhao et al. 2000). While several of these definitions mention the 'field'
indicating the agroecological environment, none explicitly mention such lab-
selected populations which can be useful in understanding some field-resistance
cases.

Behavioristic Resistance
Most documented resistance studies document physiological adaptation as

the resistance mechanism. However, behavior has played a role in resistance

since the phenomeon was first studied (Busvine 1956). The term behavioral (or
“behavioristic”) resistance describes the development of the ability of individuals
within a population to avoid a dose of pesticide that would otherwise prove lethal
(WHO 1957). There are, however, limited examples of behavioral resistance. In
at least one case, behavioral resistance was confounded with an unidentified and
undifferentiated sibling species. Initially, resistance workers believed that a
species of Anopheles mosquito in Africa avoided residues inside houses by
remaining outdoors (Brown 1958). Later, this “behaviorally resistant“ population
was demonstrated to be a complex of sibling species (Colluzzi et al. 1977). One
example of true behavioral resistance can be seen in the Australian sheep
blowfly, Lucillia cupn'na (Wiedemann), in which the oviposition of the fly was
selected for behavioral resistance to cycloprothrin (Mariath et al. 1990). By
essentially all of the definitions above, for a pest to be declared resistant by a
behavioral mechanism, genetic differences must be shown, rather than present
only observations of insects avoiding pesticides (Roush and Daly 1990). For
physiological resistance, this is assumed, even when the gene conferring the trait
remains unidentified, while for behavior it must be explicitly shown. In this case,
genetic studies have shown that this behavioristic resistance is partially dominant
and that the origin is polygenic (McKenzie et al. 1992).

A more recent case of putative behavioral resistance to a pest management
strategy was observed in the corn root worm, Diabrotica vigifera vigifera
(LeConte) (O'Neal 1999), which over-winters as larvae, and lays eggs which

hatch and feed on corn roots. In Illinois, by laying eggs in soybean fields that

require two years to hatch, this insect appears to have overcome alternative
corn/soybean rotation, the dominant strategy of managing D. virgifera population
levels. In the season following soybean, fields with D. vigifera eggs hatch and
larvae attack the corn. If this oviposition behavior is a result of a genetic change
in the population, selected for by the pest management strategy, then perhaps
this case meets Whalon and McGaughey’s (1998) definition. However, there is
some debate about the cause of this newly observed behavior, and the possibility
exists that it is not a change in the organism itself, but that the agro-ecological
landscape has changed. Perhaps the ovenivhelming majority of acreage devoted
to corn-soybean rotation has given D. vigifera no other choices for ovipositional
sites within its dispersal range and females for lack of readily-available com-
acreage are ovenivhelmed by their oviposition drive and simply oviposit
indiscriminately in soybean (O'Neal et al. 1999).

Due to the limited cases of behavioral resistance, the myriad factors affecting
insect behavior, the lack of acceptable documentation, and other issues making
genetic proof extremely difficult, the current database does not include these
cases of resistance unless there is scientific, peer-reviewed acceptance of a
genetic basis for behavioral resistance. However, the development of bioassays
to discriminate between what might be termed behavioral susceptibility and
behavioral resistance, together with genetic studies, certainly could be important
for future resistance development. This will be increasingly important for
pesticides that require the pest to ingest toxins to be effective, such as foliar Bt,

Bt plant-pesticides, or insect growth regulators (IGR).

10

Definition of the unit case of resistance

Resistance workers may disagree as to whether or not pesticide resistance is
a property of a population or an individual, or even if failure to control a field
population is a requirement for resistance. Yet there is general agreement that
resistance occurs in finite units of time and space. To track the global
phenomena of pesticide resistance it is necessary to identify a unit of pesticide
resistance, and the origin and need to declare “case” of resistance for a catalog
of resistance. For our purposes we define a case of resistance as a population
(species per location) with significant alleles conferring resistance to a particular
pesticide formulation. A pesticide, as applied, is often a formulation of multiple
active pesticide compounds and several other ingredients such as synergists,
adjuvants, so-called inert ingredients, and other additional pesticide compounds.
Due the complexity of resistance and the importance of the presence of
synergists for an applied formulation, a case must be distinct for each formulation
that a pest population is tested against. For example, testing with a compound
alone or together with a synergist such as Piperonyl Butoxide.

Identifying the edges of these units may very well be impossible but the
function here again to elucidate the scope of the resistance cases, not
necessarily to delineate them. While evidence points to some cases where an
allele conferring resistance is spread globally (ffrench-Constant et al, 2000),
others may be more isolated, such as Hawaiian Diamondback Moth (Plutella
xylostella) populations resistant to Spinosad (Zhao et al. 2002b). In the latter

example, it was clear when the resistance was selected for: within specific

11

valleys on the island of Oahu between 1999 and 2001 when careful monitoring
was done. Where global or continent-wide resistant populations exist, it may be
difficult to identify a temporally and spatially explicit case. If the resistant
population was selected from rare alleles present in wild-type populations, one
might expect the genotype to be similar enough to consider all locations equal.
However, in many cases resistance workers are only report the phenotype. This
is because of the complex and costly research necessary to investigate the
genetic basis for resistance. Given only the phenotype, one cannot say with
certainty that a resistant population from the same species is distinct, and/or .
'isolated' in a region. Nor can one infer that two different resistant populations
share the same genotype because there is no delineation between a resistant
and susceptible phenotype. Therefore one can only consider populations from
distinct regions as distinct cases of resistance. However, the spatial scale as
which we measure will greatly affect the number of cases that are counted. Geo-
political boundaries are often used to identify regions containing pesticide
resistant populations, in spite of the range of sizes of countries. In many cases
this is all the information that is provided in a report of resistance (see chapter 2).
Merging all cases from the continental US would result in a much-reduced count
versus using individual states.

Therefore resistance may be delineated as the first occurrence of a species
discovered to be resistant to a pesticide formulation (of one or more ingredients)

collect from a defined region. These five elements (time, species, pesticide,

12

place, and document) define a unit, or case, of resistance for inclusion in a
database of resistance.
Measuring the Consequences of Resistance

David Pimentel of Cornell University published one widely cited estimate of the
impact of pesticide resistance at production loss of US$1.4 billion for all major
crops in the United States (Pimentel et al. 1992a). To arrive at this figure, he
started with an estimate of yield loss in California cotton production in 1984 and
applied this proportion the the entire cropping system of the US. However, the
logic of this estimate is flawed. First, cotton is not a good example crop as cotton
has heavier pest pressure, more insecticide use and much higher incidence of
resistance. In 1997, cotton fields around the country received applications from
one or more of 40 distinct insecticides active ingredients (Gianessi and Marcelli
2000). Second, levels of resistance vary across agro-ecosystems, and vary
across regions. Third, the impact of resistance changes over time within an
agro-ecosystem. Costs from resistance since 1984 must certainly have changed
as pest populations change and as new pesticides are registered to address
resistance issues. Certainly the introduction of transgenic cotton has made
dramatic changes to pest management practices for cotton growers.
Pesticide Resistance Policy: a call for a database

Resistance drives not only IPM decisions, but also pesticide policy in the US
and Europe and other parts of the world. In fact, the UN declared pesticide
resistance as one of the top priorities (UNEP 1979). In the US, a 1984 study

initiated by the US Board on Agriculture of the National Research Council made

13

16 recommendations, one of which stated, “Federal agencies should support and
participate in the establishment and maintenance of a permanent repository of
clearly documented cases of resistance” (Dover and Croft 1986). This
recommendation was made law by the Food, Agriculture, and Trade Act in 1990,
which called for a “national pesticide resistance monitoring program.”

The Food Quality Protection Act of 1996 (FQPA) made dramatic changes to
US pesticide policy by amending the Federal Insecticide, Fungicide and
Rodenticide Act (F IFRA). Among the FQPA amendments is an invocation of
resistance as one of the four conditions designating “minor use” status for a
pesticide registration. 'Minor uses' are defined as application sites in the US
where the crop has less than 300,000 acres nationally. “Minor use” registrations
have waived some registration fees (section (4)(i), paragraph (4) ) and fewer data
requirements (FlF RA section (3)(c) paragraph (2) subparagraph (A)). Minor use
status therefore is an incentive to pesticide registrants to pursue EPA approval of
their products in markets with othewvise little or no profit to the registrant. Minor
uses of pesticides are given special status in many parts of section 3
(registration) and section 4 (re-registration) of FIFRA. Specifically, a pesticide
registration may be declared a “minor use” when the US EPA, USDA, and the
pesticide registrant determine that profits from sales of the pesticide for a specific
use “does not provide significant economic incentive to support the initial
registration or continuing registration” and that the use “plays or will play a
significant part in managing pest resistance.” A "minor use" pesticide is given

special provisions that reduces the pesticide registration burden for othenivise the

14

registrant had little to gain economically despite the fact that the pesticide may be
important for the continued production of specific crops. (F IFRA section (3)(c)
paragraph (2) subparagraph (A)). FQPA also added two new sections to the

F IFRA establishing ‘minor use programs’ in both the USEPA (section 32) and the
USDA (section 33).

In the European Union (EU), data eliminating the possibility of cross-
resistance in the target pests is a requirement for every pesticide registration
(EPPO 2002). To help determine which pesticide are useful for resistance
management programs, both US and EU rules require that such pesticides be
identified and cataloged. The FIFRA mentions resistance in that the mandated
USDA minor use program (F IFRA96 section 23) will “carry out...the national
pesticide resistance monitoring program established under section 1651 of the
Food, Agriculture, Conservation and Trade Act (FACTA) of 1990 (7 U.S.C. 5882).

In fact a call for such a monitoring program came earlier when a 1984 study
initiated by the Board on Agriculture of the US National Research Council (NRC)
made 16 recommendations, one of which stated that “Federal agencies should
support and participate in the establishment and maintenance of a permanent

repository of clearly documented cases of resistance” (Glass 1986a).

15

CHAPTER 2. DESIGN AND SUMMARY OF A RELATIONAL DATABASE OF
PESTICIDE RESISTANT ARTHROPODS.

Introduction

Pest management is challenging enough, and pest resistance to pesticides
significantly increases the complexity, expense and perhaps the environmental
impact of pesticides in the agroecosystem (Metcalf 1983). For those pesticide
uses with even low frequency for resistance developing in the target pest
population, resistance management must play a part in any Integrated Pest
Management (IPM) plan (Glass 1986a). If pesticide resistance alleles are
present in a pest population, or soon likely, alternative pesticides must be
selected because the pesticide will not be effective (field failure) if resistance is
not delayed. However, in almost every instance it is difficult to detect resistance
in a local population before field failure occurs. In many of these cases, the
genetic basis for resistance is already fixed in the population (>1% allelic
frequency) and development many not be slowed with any strategy (ffrench-
Constant and Roush 1990). Successful integrated pest management, which
incorporates resistance management, requires diverse and accurate knowledge
about the nature of the pesticide resistant population.

One way to address this problem is to determine what the status of resistance
may be for each pest and pesticide combination in the environment for each
pesticide that is available for use. While it may be overly simplistic or even

specious to reduce the complexity of resistance to a binary state (resistance/not

16

resistant), some indication whether pesticide resistance (or the potential) is
present is one of the first insights necessary for wise pest management.

lnforrnation related to pesticide resistance is scattered throughout the
agricultural research literature. Publications detailing insecticide resistance were
frequent and diverse, with 1003 annually, covering every imaginable aspect of
resistance. Which of these aspects of resistance could or should contribute to a
“permanent repository” as called for by the US Board on Agriculture (Glass
1986a)? How would these publications be used to define the status of resistance
to address both the needs of pest management and of policy makers (see
Chapter 1)?

Some of these resistance information needs in insects and mites (arthropods)
were met in part by the work of George Georghiou at University of California,
Riverside (Georghiou 1972, 1983, 1986, Georghiou and Lagunes-Tejeda 1991).
The approach is straightforward: the authors measure resistance by listing
pesticide resistance “cases” (see chapter 1, pg. 11) by the year the cases were
published. A “case” is the first report of resistance on record, described by five
elements:

1. cited reference,

2. arthropod species,

3. pesticide formulation it is reported to be resistant to,

4. region where the resistant population is found, and

5. the year of the report.

Georghiou (1983) noted that this information was computerized (Georghiou

17

1983), but to the public it was available in book form (Georghiou and Lagunes-
Tejeda 1991) and unfortunately in a very limited printing by the Food and
Agriculture Organization of the United Nations.

Clearly the need existed for a computerized database of arthropod resistance
information, publicly available and amenable for summary and analysis. To
address this need, a computerized database of resistance cases was designed
and constructed using relational database theory.

Materials and Methods '-

Relational database theory was employed as the methodology for designing
the data model to capture the resistance case concept (Codd 1970) (Date 1995).
Nearly every modern commercial DBM software system is based upon this well-
understood theory (Elmarsi and Navathe 1994). The relational database model
is a formal theory of information modeling, defining highly structured data
elements (relations or tables) that consist of sets of tuples (rows), well-defined
mathematical operations on sets of tuples, the results of which also defined as
sets of tuples. Each relation (table) must have one or more attributes (columns)
that unique identify each tuple, designated the primary key (PK). Related
information from two separate tables may be joined by inclusion of the primary
key from one table in another table as a foreign key (FK). This technique was
used extensively for the resistance database.

It was clear that data gathering would be the most costly process for
resistance information given the sheer numbers of resistance research papers.

Just as Georghiou 1991 incorporated A. W. A. Brown's work, we began the

18

system with resistance records contained in Georghiou and Lagunes—Tejeda
(1991). To maintain comparability with reported resistance figures prior to 1991,
we modeled a portion of the database to completely cover the information
included in this text. Timelines previously published in Georghiou and Melon
(1983), Georghiou (1986), and Georghiou and Lagunes-Tejeda (1991) could
therefore be extended.

There were several considerations necessary to arrive at a database structure
given (1) design goals, (2) data and (3) a description of the semantic
relationships within the data itself (Date 1995). While algorithmic procedures
exist for modeling information for construction of a database (Halpin 1996), in this
instance a functional design was clearIy evident based upon the inherent
structure of Georghiou and Lagunes-Tejeda (1991). The columnar form of this
text presented the data directly to a relational database model. The format of this
compilation of resistance cases was much more amenable to mapping into a
relational database design than data from at least one previous author {e.g.
Brown, 1971 #6}, who aggregated key information by reporting chemical class
rather than detail resistance cases by individual compounds.

Entity-relationship techniques (Chen 1981) (Date 1995) were used to design
the semantic model of Georghiou and Lagunes-Tejeda (1991) dataset with the
dual goals of preserving the inherent structure for compatibility and to allow the
expansion the information for more detailed analysis (Figure 1). The core of the
model is the entity 'resistance' (cross-hatch shading), corresponding to a list of

the cases of reported resistance, made up of five attributes discussed previously

19

(page 17). These five attributes uniquely identify a case of resistance; the four
related entities were joined with common attributes indicated by arrows. All of
these relationships have one-to-many cardinality. In Figure 1, arrows point from
the primary key table to the table containing the foreign key. All but the year
attribute was determined to be complex enough to warrant expansion into
separate entities, during the normalization process of the semantic model.

This model, as is standard practice, includes entity attributes with no semantic
meaning that were used to form relationships between entities (Date 1992).
These unique identification codes, typically numeric, were used in practice by join
operations during data analysis. Such codes provide the advantage that text
identifiers may be corrected or changed without a cascade of corresponding
changes in the database. For example, updating genus or species names in the
arthropods table as they change in the literature (Heliothis zea to Helicoverpa
zea.) would not break the relationship between the arthropods table and related
resistance records since the numeric key relates them. Similar examples can be
made for location (e.g.country names) as they change in time. While several
other variations on this semantic model were possible that would capture the
information in the text, all would essentially contain a central entity for a
resistance case.

While this parsimonious model captures the data in Georghiou and Lagunes-
Tajeda 1991, to achieve the design goals for further manipulation required
extending the model through additional attributes and entities. First, a linkage to

external databases, which requires mapping records in the local database with

20

external standardized codes, was built. This innovation was used to make
standard codes such as pesticide active ingredient list compatible with existing
USEPA codes (Office of Pesticide Programs 2000). These codes provided
crucial linkage to the standardized identifiers for pesticides. The relationship
between a chemical structure, a compound common name, the legal term “active
ingredient,” and a commercially available pesticide formulation is much more
complex than a single chemical name. To perform analysis that summarized
counts of core entities for groups of arthropods species, pesticides based upon
chemical classes, and several other operational factors.

Based on this model, instructions were coded to build a relational database
using a variant of standard Structured Query Language (SQL)(ISO 1992). Few
relational database management systems (RDBMS) implement this standard
completely, and many add proprietary extensions that negate the advantage of
wide compatibility. All, however, implement some form of this relational
databases lingua franca. This project used the MySQL RDMBS , which is freely
available, closely adheres to the SQL standard, and is compatible with many
existing computer systems (see program code listed in Appendix 1, page Errorl
Bookmark not defined.)

Data and references from Georghiou and Legunes-Tajeda (1991) were
entered into the database system using forms developed in Microsoft Access
version 7. Resistance cases based upon the review of that covered 50 journals
from 15 countries in four languages, over refereed journal articles were entered

into the database. Like previous efforts, the source data is the result of a review

21

of published resistance accounts, obtained through a series of library and
publication database searches. Only cases published in the peer-reviewed
literature were selected for input. As has been stated previously, a report from
the field of insecticide failure is not a sufficient indication of the presence of
resistant individuals without scientific peer reviewed bioassay data. Obviously,
scientists, resistance workers, and members of the agrochemical industry
requiring empirical proof may view an undocumented claim of resistance with
skepticism, even when such claims may in fact be true.

Significant variation exists in the resistance documenting methods in published
reports. Standardized methods for resistance detection do exist. The UN FAO
has been publishing standardized resistance bioassay methods for species
affecting human health since 1969 (FAO reference). Nevertheless, lab
techniques are constantly improving, and authors often interpret and report
results of standardized tests differently (Croft 1986). Even within these
established standards there are many factors that might cause
misunderstanding, and it is difficult for any reviewer to determine the veracity of
such diverse data. Therefore, part of the data input strategy was to rely upon the
expertise of the reviewers of manuscripts and the editorial boards of publications
as well as upon our own review of the values of the median lethal doses (LDso),
median lethal concentration (LCso). median lethal time (LTso), median knockdown
(K050) and discriminating doses as the final criteria in determining a resistance
case. The primary objective was to a statistical difference between resistant and

susceptible populations for previously unreported species, compounds, and/or

22

regions before the citation qualified for a new resistance case.

A very commonly reported measure of resistance was the resistance ratio
(RR), which is the ratio of dose-mortality of the bioassayed strain (defined by the
morbidity statistic used, eg. L050, L050, K050 or TLso) to a known susceptible
strain. We used reports of RR greater than or equal to ten, a traditional point of
departure as a general rule for declaring a resistance case. However, in some
instances we also included reports with RR smaller than 10 fold when the
authors clearly made a case the resistance was sufficiently high to cause
significant field resistance. No specific RR will predict field failure for all cases.
However, this approach allowed consistency between records from Georghiou
and Lagunes-Tejeda 1991. Factors involved in deciphering a resistance report
included the Whalon and McGaughey (1998) definition of resistance, several
intrinsic and extrinsic factors of the test itself and the type of statistic used to
report the resistance level. Published reports of resistance cases developed in
the laboratory were also included as a warning of potential or impending field
failure. This is consistent with the working definition of resistance, which states
that the genetic change conferring resistance may or may not lead to field failure.

Confounding this categorization of resistance literature was the observed
variability among definitions of a pest ‘population.’ The catalog of resistance
would not be complete without a spatial definition of pesticide resistant
populations. Researchers often reported collecting individuals from multiple,
reproductively isolated locations, but unfortunately reported aggregate bioassay

results. Populations were described with vague spatial definition or overlapping

23

boundaries. This was not surprising as the sampling and bioassay requirements
to map the boundaries of a population are expensive. We used a coarse
geographic resolution to circumvent these problems and thus limited distinction
of regional cases to the national, state, or provincial level. Again, this design was
comparable to Georghiou and Lagunes-Tajeda (1991).

It is presumptuous to say that all new resistance cases since 1991 were
uncovered by this effort given the scope of this world-wide phenomena, the
diversity of reporting mechanisms and the language challenge. Journals
reviewed were published in English and a few in Spanish, French and Italian. In
addition, several cases not included in previous resistance compendiums were
contributed by a resistance researcher and former citizen of the Soviet Union
(Il’lchev, 2000). While the coverage of other languages was spare, the
enumeration of pesticide resistance in arthropods is a dynamic process, not only
as new populations develop resistance, but also as past reports from around the
world are illuminated. One major benefit of a computerized repository for this
information is that it may be updated and analyzed continually. To that end our
database is available, as of May 9, 2003, via the lntemet
(http://whalonlab.msu.edu).

The Insecticide Resistance Action Committee (IRAC), an affiliation of
agrochemical industry representatives, conducted an international survey to
determine the status of resistance in many areas of the world. The result was a
spreadsheet indicating which insects (or class of insects) have resistance to

particular chemical classes together with the type of information source (field,

24

confirmed, publication or rumor). Although this survey does not always indicate
which specific pesticide the resistance is to, the data were incorporated it into the
database, using the chemical class as the formulation, and, if no compound of
this class was currently entered into the database for this species, the case was
entered. The justification for including these data was the evaluation rigor and
expert peer review process IRAC employed (IRAC, 2002). Unfortunately,
however, the majority of cases in this document could not be included in the
database because of lack of specificity either for the pesticide as described
above or because only the genus was provided.
Results

Five hundred and forty three species of arthropods were reported to be
resistant to one or more pesticides: 499 pest species (agricultural, urban or
human health pests), and 44 species that were not pests (natural enemies and
others) (Table 1). Two hundred seventy three genera, one hundred three
families and sixteen orders represented species. Three hundred and five
pesticide compounds were reported with many cases, species were resistant to
more than one compound, either multiply-resistance (simultaneously carrying
the resistance factors for multiple compounds), cross-resistant (having never
been previously exposed yet resistant to that pesticide), or simply resistant to
different compounds at different locations in time and space. Since 1989, 38
previously undocumented arthropod species were reported to be pesticide
resistant in some capacity, just over 6% under increase in twelve years.

However, there was a much larger proportional increase in the number of

25

pesticide active ingredients to which arthropods have developed resistance.
Resistance Summary by class

The ovenivhelming majority of reported resistance cases related to
organophosphate (44%) and organochlorine (32%) insecticides (Table 2). This
was not surprising as these classes of compounds include the most popular
pesticides to date, and many have been in use for over half a century (Osteen
and Padgitt 2002). Pyrethroids and carbamates together constitute only about
16% of resistance cases. Bacterial pesticides, primarily those produced from
species of Bacillus thuringiensis (Berliner) (Bt), represent a mere 2% of cases,
and all other remaining chemical classes combined have led to the development
of less than 2% of resistance cases, as reported in the literature.
Top Twenty Resistant Arthropods

Arthropods were ranked based upon the number of unique compounds to
which documented resistance was reported at least once. Table 4 (page 46).
reports the twenty most resistant arthropods according to this ranking. While not
a ranking of pest importance or severity, many of the world’s most destructive
pests were included. It should be stressed that exclusion of a particular pest
from this ranking does not reflect the lack of importance of an arthropod’s
economic importance (e.g. pest status). Indeed, every case of resistance is
important and should be observed in the context of the system of production,
human health protection, geographic area, and other factors. While many of
these reports were decades old, new resistance reports surfaced steadily in the

past ten years. Inclusion on this list may have been indicative not just resistance

26

development from intense selection pressure, but also sheer numbers of
pesticide against which these species was bioassayed. Combination of diversity
pesticides used in the field, world distribution, and scientific interest reporting
results of testing against a wide variety of compounds. Nonetheless,
documentation of resistance to multiple pesticides is a strong indicator of heavy
reliance on pesticides for pest management.

Genetic, biological, and operational factors significantly influence the
development of resistance (Georghiou and Taylor 1986). Members of this list
share worldwide distribution, severe crop damage or human health implications,
and a large body of research. While many of the species listed also share similar
biological and ecological characteristics including high generation turnover, great
mobility and migration, and large numbers of offspring per generation as well as
operational factors.

There were four pests of human health importance and sixteen agricultural
pests in this ranking, a different ratio between these classes of pests than is of
counts of species in the entire database (approximately 2:3 human health pest
species to agricultural pest species in the database overall). In fact, no
anopholene mosquito breached the top twenty. Poor representation by serious
human disease vectors may have been indicative of the fact that the developing
countries were most affected the diseases vectored by these pests, and the
economic reality that many still use cheap organochlorines such as DDT for
human health protection.

A spider mite (Tetranychus urticae (Koch),) for the greatest number of

27

reported cases at 76, just 6 more compounds than the insect which placed
second, the diamond back moth, Plutella xylostella. These species were closely
followed by the green peach aphid, Myzus persicae (Sulzer), with 68 reported.
There were four other homopteran species that have developed resistance to
many conventional and novel compounds: M. persicae Aphis gossypii, Phorodon
humuli, and Bemisia tabaci.(Gennadius) Beside common biological and
ecological characteristics distinctive to this order, some homopterans were
responsible for virus transmission driving economic thresholds very low. The
response by growers is frequent application of pesticides, especially in M.
persicae and B. tabaci. (whitefly) It was reported that lsreali tomato growers
resorted to two insecticide applications per day to reduce the spread of Tomato
Yellow Leaf Curl Virus by whitefly (Berlinger et al. 1993). In addition, frequent
treatments in multiple hosts often cause a great deal of selection of individuals
for resistance. Conversely the damson-hop aphid, Phorodon humuli remains
during the summer only in hops and wild hops and stays close to the crop. But in
this instance monophagy, when combined with high fecundy and primary pest
Importance (hops), make this a “the worst case scenario" for the development of
resistance (Denholm et al. 1998).

Consumer demands for perfect cosmetic standards and a stricter restriction of
"insect parts" present in food force producers to lower economic thresholds
(Pimentel et al. 1992b). On cmcifers the diamondback moth causes qualitative
damage in addition to quantitative both of which growers use lots of insecticide.

The use of Bt has reduced the proliferation of conventional insecticides in

28

crucifers. However, intense use of this alternative pesticide has led to the
development of field resistance to Bt (Tabashnik et al. 1990, Shelton et al. 1993).

Four Lepidopteran species were included in our top twenty including Heliothis
virescens, Spodoptera Iittoralis, and Helicoverpa annigera, all of which have
been exposed to heavy insecticide treatments in cotton. These species
treatments in other hosts have increased the selection pressure. In the past,
industrial cotton had been the recipient of more than 40% of the applied
insecticides produced in the world making it a significant source of pesticide
resistant species (National Research Council 2000).

Distinctive aspects of phytophagous mites, including very high fecundity, many
generations per year (16 generations per year), and high selection pressure,
often leads to pesticide resistance (Kennedy and Smitley 1985). It is not
surprising that the spider mite Tetranychus urticae (Koch), a serious pest of fruit
crops, tops our list. Two others, Panonychus uImi and Rhizoglyphus robin are in
the top ten.

Conversely the Colorado potato beetle (CPB), Leptinotarsa decemlineata
typically has only from one to three generations per year, fewer than the majority
in Table 4. However, this insect has a tremendous capacity to colonize a wide
range of hosts including many species of the Solanacea family. Adaptation to
defensive secondary metabolites produced by solanaceaous species family may
have allowed the Colorado potato beetle to increase its range of hosts from the
original wild hosts to those of the cultivated potato. Adaptation through

thousands of years has made this insect a formidable species to breakdown

29

xenobiotics, a trait which may have extended to insecticides. Another important
factor in the development of resistance for CPB is reduced migration where “local
selection” plays an important role (Grafius 1995). In local selection, individuals in
a population do not disperse or immigrate, reducing gene flow and elevating the
frequency of individuals with resistant alleles.

Similarly, the appearance of the high ranking (fourth) cattle tick, Boophilus
microplus, was related to the method of application and host specificity. Total
coverage of cattle by immersion in insecticide solutions increases the resistant
selection and individuals with resistant alleles were rapidly screened. Given the
limited host range of this insect, in terrestrial agriculture this would correspond to
100% acreage treated.

Insecticide resistance is also a problem in urban areas. For example, the
housefly (Musca domestica) is a significant pest in veterinarian circles. In most
farms, high selection pressures for resistance resulting from insecticide
treatments occur in areas where the treatments were concentrated, the
residuality of the insecticides is long, and the populations were relatively isolated
(Keiding 1977). In addition, the common practice of screening windows and
doors to avoid immigration also has led to rapid selection and an increase in
resistant individuals (Georghiou and Taylor 1986).

Protection of human health has led to an intense use of insecticides. As a
result, there were two species of mosquito: Culex pipens (ranked 9th), Culex
quinquefasciatus (ranked 13). Anophelene mosquitoes did not make this list.

However, there are far and away more resistant species from that genus than

30

any other genus (63) . In fact there are more mosquitoes than any other
arthropods on this list. Anopheles albimanus, that have developed resistance to
many insecticides and have become vectors of diseases. In the year 2000,
billions of people in the world’s tropics were at risk of contracting malaria from
such vectors (Marshall, 2000). Malaria infects 300 to 500 million people per year
and every year 1.5 to 2.7 million individuals die from the disease (Danis and
Gentilis 1998). A. albimanus is one vector of this disease that has developed
resistance to insecticides used to curb the spread of Malaria. Other species in
the genus Anopheles have developed resistance to insecticides as well, yet A.
albimanus has maintained the greatest resistance in comparison with these other
Malaria vectors. One reason for this higher resistance of A. albimanus to multiple
compounds is the intense insecticide selection pressure exerted over the
complex of insect pests in cotton (Georghiou, 1984, Georghiou, 1990) that also
indirectly selected immature stages in breeding sites and adult stages in resting
sites.

Tribolium castaneum, red flour beetle, is a principal pest of stored grains
where complete coverage by insecticide treatment is a common practice to
control insects. High selection pressure and low migration were some of the
causes that have led this insect resistant to many insecticides.

Pesticide Resistance Timeline

Since 1983 published timelines of resistance. The first results, published from

data up to 1979, demonstrated an exponential increase in new resistant species

since the first in 1914 {Georghiou, 1983}. By 1986 that trend had begun to taper

31

off (Georghiou 1986). We see a continuation of that trend in the new species,
compounds and total new cases (species and compound combinations) from
1989 onwards that we have entered into the database (, page). The frequency
of resistant species discovery has declined since the 1960s (, page), when it shot
to the highest level in history. In the 703 and 80s many non-pest species, such as
natural enemies, were added from researchers seeking bio-control agents
compatible with spray programs (Theiling and Croft 1988) (Croft and Brown
1975)

Previous authors attributed this trend to the limit of pest species. Introduction
of Integrated Pest Management (IPM) in the 1970s has probably slowed
insecticide selection pressure and reduced the trends in resistance development
that we have seen in our results. In addition many of the insecticides to which
these pests have developed resistance have been canceled due to
environmental and human health effects (see Chapter 3). However, one of the
collateral effects of multiple-resistance is the presence of a diversity of
mechanisms employed by the species with reported resistance that could be
cross-resistant to existent and new compounds.

Resistance by Chemical Class

Most of the cases of resistance that have occurred with so-called conventional
insecticides were classified in either the organochlorine (OC), organophosphate
(OP), carbamate (CB) or pyrethroid (PY) chemical groupings (Table 2).

Conventional pesticides were those that have controlled a broad spectrum of

species, have worked as contact nerve toxins, were easy to use, and have been

32

in use for many decades. Few cases of resistance had been detected outside of
conventional pesticides such as agonists and antagonists of GABA receptors,
insect growth regulators, Bacillus thuringiensis (Bt) protoxins, and neonicotinoid
compounds. Yet the appearance of insecticide resistance has followed a loose
chronological pattern following the deployment of most insecticides. Generally,
the first cases of resistance have been reported within three to five years after
the compound was extensively used. A specific example comes from Danish
Housefly research (Keiding 1977). Before the organochlorines were introduced,
fifteen cases of resistance were to one of seven inorganic pesticides. From the
time of the first case of DDT resistance in 1947 up until the 19803 the majority of
resistance cases have resulted from the use of organochlorines, followed by
organophosphates, carbamates, and pyrethroids — broad spectrum conventional
pesticides with many formulations available from each chemical class. Adverse
human health effects and negative environmental impacts from organochlorine
compounds led to the cancellation of almost all of the insecticides in this class
after the USEPA was created in 1970, paving the way for the organophosphates
to be the most widely pesticide used. Many of the other classes were introduced
solely to control pests resistant to compounds in the other classes (e.g
forrnmidines) (Ware 1989). In the last two decades of the 20th century, four
groups of pesticides saw resistance develop for the first time: compounds from
the classes organotin (miticides), averrnectin, and pheylpyrazoles, a the group of

compounds that act as insect growth regulators.

33

Reduced registration and use reflects the decline in the rate of newly reported
resistance cases in the organochlorine compounds in the last three decades;
only 0.7% of the total known cases were reported between 1990 and the present
(Table 5). In 2003, the only organochlorine compounds with remaining uses
(worldwide) were DDT, endosulfan, lindane, and dicofol, and these uses are
severely curtailed. In the United States, only endosulfan and dicofol have
registrations remaining. Without organochlorines in the environment, and one
would expect to see a much-declined rate of new species or new cases of
resistance to them. However, species with no exposure to OCLs may be
discovered in the lab to exhibit cross-resistance to them, as many pests with
organoclorine resistance were shown to have cross-resistance to pyrethroids
when they were introduced with little or no exposure to the pyrethroids (Miller,
Salgado and Irving, 1983). It was discovered that the same gene (kdr) was
facilitating resistance to both classes of compounds, which had similar target
sites, many organochlorine-resistant pests developed resistance to pyrethroids
quickly. When organochlorines were replaced by organophosphates,
carbamates, and pyrethroids, more cases of resistance to these replacement
compounds ensued. Although some uses of organophosphates and carbamates
have been cancelled because of adverse human health effects, pesticides from
these classes are still very widely used (Gianessi and Anderson 1995, NASS

1997, 1998, 1999, Gianessi and Marcelli 2000, NASS 2000).

34

Discussion

It should be noted that these totals represent reported resistance discoveries.
The true status of R, the allele frequency per population, was rarely known for
any time period after discovery for any but intensively studied agro-ecosystems.
In some species, when the pesticide was no longer applied and hence the
selection pressure relaxed, any non-dominant resistance alleles decline to pre-
resistance frequencies. However, there is clear evidence that these will quickly
return in the population (Georghiou 1986). A cumulative count is a snapshot of a
phenomenon that varies over time.

To date, the appearance of species resistant to compounds with newly
discovered modes of action has not been as dramatic as in conventional
pesticides (chlorinated hydrocarbons, organophosphates, carbamates,
pyrethroids, etc.). We cannot say that this will be true in the future. This trend
may be due to the limited use of these (often costly) novel compounds, especially
in developing countries where profits are low and regulation of cheaper
alternatives is less stringent than in Europe and the United States. An increasing
emphasis on integrated pest management a key component of sustainable agro-
ecosystems will also increase the use of the compounds that fit with this strategy:
those with characteristics of rapid degradation, narrow spectrum of pest species,
and minimal toxic effects (acute or chronic) to humans and other non-target
organisms. Government restrictions of conventional pesticides, such as through
risk assessment mandated by the US FQPA, will very likely increase the use of

these compounds as alternatives. Increased use of, and hence intensified

35

selection pressure exerted by, these novel compounds on pest populations will
almost certainly result in additional cases of resistant arthropods. The latest
deployment of compounds to combat pest resistance looks promising, but we
should be extremely cautious. It has been suggest early as 1961 that insecticide
resistance is a natural outcome of our use of pesticides, given a 350 million year
history of herbivorous arthropods evolving mechanisms to defeat toxins the
defensive secondary plant chemistry of their hosts (Georghiou 1972). It is not
surprising that, in the last one hundred years, 307 phytophagous arthropod pest
species have evolved resistance to pesticides from selection by humans. A
similar co-evolutionary case may be made for entomophagous arthropods (e.g.
parisitoids) whose hosts sequester secondary compounds from plants. Whether
the toxin is synthesized and applied by humans protecting a host plant, or
synthesized and delivered by the host plant itself, the evolutionary endpoint is the
same: resistance.

The development of resistance to pesticides is an incredibly complex process
crossing biological and social disciplines. However, even with our simple
modeling of resistance as a case, and limiting citations to the first appearance of
a significant resistant population, several trends are apparent. However, there
are several important aspects of resistance missing from this model. One
attribute that is present in nearly all discussions of resistance cases is the
cropping system in which the resistance occurs. There are at least two reasons
for this omission in our database. First, selection of resistance alleles may take

place wherever the population encounters pesticide residues. This may from a

36

different crop farmed next door or thousands of miles away from the field where
resistance was detected, such as Empoasca fabae (Carlson et al. 1991). There
are worldwide examples of documented resistance to organochlorines and
organophosphates in mosquitoes that may have been caused or aggravated by
agrochemicals in the same regions (Georghiou 1990b). Secondly, generalist
pest species whose host range include several difference crops may have been
exposed to pesticide quite some distance from the field in which it was collected.
Databases have been used for some time in agriculture to detail and analyze
broad aspects of pest management that would otherwise be intractable (Whalon
et al. 1984). This project is one more example of how utilizing the relational data

model can help will effective management of large sets of data.

37

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

«Exotica!
masons;
2338282
thing
:06!
5.58
cfiuatumooicto.
Eco. 223:, y...
A ._.l
Era-Ha xx. 7 >
I =E£x3 SE
0&8 53335.8 . . $5: .5233me
«EaclcoEEou .
fifibcfiufi? columoa 7 flag v...
33:52 5: $.89 allzeexs xx.
8.395883 36.38 Boa . 3:3. I
4 2832 3: 8.55 33.55
wag-mafia xx. magma c... . BEES. c: 23839 E
.-1. .- , . v 853 8:.
«238.59 . f Euaascofiuufifioa 23:96:: O...— .muonoztm
>
3.33.8.
--.F
933... Burns. 3:
assuage .
3.... v... 82.633539,
55258633385 251.250“
oz:
52»
.653.
Hugo... v...
12323:
E

Rex cmxmcol SE Xox metQ
in: 53 .mbcxmk-moczumq bum 30.5980 5 mesh: 2: go 3608 Son ozcmEom So co Emcumxb Scot0:&9-§EN .» 93¢

38

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

50005038
0E0Z§E2
“:0: 00
Pawns €580.22,
90:58 5:
fig ~8v.00..<:0_m£
0... 0252250802
8.00..
928058 5: 8.558 zzSfifiwosg
005008>c0§0 :050: 000000305000
00bE>coc>0 . .
:0=0_..0000I::0_ 0.00.9.0. 030500002t0 *
555.0520, 0... .52 200.000 :0 , _ 0.0
0:50:00 0... .350 E 2202 030.0 0... _ al. __
I I 0039 0.. :0
Excogm 0:300:80 I 0.00:0 00025.0 000 :
00.». 5:03:09
3.000m025E300000
.00_E0:000E0E300000
a I I x 3003030.. :0_m0y_00.:0E300000 00000305000023 9.0
H. m: :00 00.0:IEE0 0030?: >=E0~x$ CE
02: lea-.050 0399:0685 858 30.22 8.0.5:
00200 .0 I000E 0053:000 I020: 008:
0 LMHHEQMW «v.0 .80.:6: 00:03.00. 000.0 0E0:I:oE:.8 .0983me
0.90:me .3 I .0Jm00
0300:0305 CE 0 . RBI—HMS N!"— ‘I I :00» 00.0000 1' a: _ 0.0
000 00.05000 2:030. 3.... 0.3000. 04.qu v.0
0E0:0:300E00 . 2:25.323. «v.0 m0___E00I0000E:0
00255 SE ..
030323603 30 23.03000 x0
Hung—dam x0 x0 23:28! 9.0
3:300:50 550.2%2 00.05.00 00:50.00: * 00000200
8030: cue-Ms 5: 0.88.50— 3:
00:03 8.
55000000305 . Band-nag“ :0
I @3083 «802883 .
5:053: 0305 a mama“ QMMHM :0 Hangman CE 0.0
039902.:05 I I E3:I0m000:._3_o> 0E0::0EE00I0000._::0 E>cocwa0000Ecm
0530030..me5 :0=0_:E..0. E 000300500 0:540:50—
0.5
gun—dug .
x0 50>
0305I0:300E8 .0530
flag 0.0
0.200.:

 

 

 

FL

A003 00 00:00:00 5 .500 030:0 S000 0908..qu S000 505.00 u Kl 00000000 00000200 00 0000:: 300000009505 .N 050$

39

Table 1. Summary of Cases of Reported Pesticide Resistance; counts of uniqe occurrences of

species, compounds, and cases along with publications reviewed

 

Count from
literature

Aspect of a Resistance Case: Count
description published in

Georghiou & review from
Lagunes- Michigan State Database
(data to 2002) Tejeda, 1991

Tejeda, 1991
(data to 1989)

University

Total Count

Percent

from Change from
Resistance Georghiou 8-
Lagunes-

 

Species:

arthropod species which were
resistance to one or more
pesticides

511 38

543

6.3

 

Compounds:
225 167

a unique pesticide active ingredient
to which one or more arthropod
species is resistant

315

40.0

 

Unique Cases:
2286 502

a case of a unique species resistant
to a unique compound, eg unique
(species, compound) pairs

2655

16.1

 

National Cases:
4292 598

a case of resistance unique to any
one country, e.g, unique (species,
compound, country)

4793

11.7

 

Regional Cases:

Species, Compound, Country,

Region. May include multiple, 5053 732
identical cases from the same

country (e.g. different states or

provinces)

5785

14.5

 

Referenced Documents:
1 ,263 205

Number of citations of reports of
new regional cases

1 .468

16.2%

 

40

Table 2. Summary of Insecticide and Acaracide Resistance

       

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mpoound ._ 7 7 CateWoistat Arthrood Ttoal Cases
Mode of Action Com- . . . by
I Ch omi cal pounds Agricultural, Med'Ical, Predators Other / PollIn- Chemical

Forest and Veterinary, / Miscella ators
Class with . Class

Ornamenal and Urban ParaSItes neous

Resist-
Plant Pests Pests Arthropo
anco ds
Organo- 112 715 358 52 10 1135 44.1
phosphates %
Organochlorines 26 484 329 10 15 2 840 32.6
%
Pyrethroids 33 133 74 1 1 1 219 8.5%
Carbamates 35 132 57 14 1 204 7.9%
Bacterials 38 42 4 46 1 .8%
Miscellaneous 30 37 1 46 1 .8%
Fumigants 6 21 21 0.8%
Insect Growth 10 16 2 3 21 0.8%
Mators
Organotins 3 8 0.3%
Formamidines 2 4 2 6 0.2%
Arsenicals 2 2 1 1 13 0.5%
Avermectins 2 2 3 1 6 0.2%
Chloronicotinoid 1 2 1 0.1%
Roten one 1 2 2 0.1 %
Dinitrofenols 1 1 1 0.0%
Sulfur 2 1 1 2 0.1%
Compounds
Phenylpyrazoles 1 1 1 0.04
%
Total Cases by 1602 850 90 30 2 2574
Arthropod Category
62.2% 33.0% 3.5% 1.2% 0.1%

 

 

41

 

 

 

 

 

 

 

 

O
8 —-1
-0— species
+ pesticides
O
O - o
v
E
:I
8
2 .
3 .
E
3
O O
o _
N
o .
2 — r.
o —
l r l l l
1 920 1 940 1 960 1980 2000
Year

Figure 3. Accumulation of Arthropod Resistance, 1914-2001: Number of First Reports of
Resistance ifor a Species per Year, (fit with Loess curve to emphasize discontinuities)

42

 

 

 

 

 

 

O

O _

SJ

+ Carbamates

o + Fumigants

8 — + Insect Growth Regulators
’6 —e— Miscellaneous
g —a— Organochlorines
8. —e— Organophosphates I»

. __=:: :n
E g _ —A— Pyrethrouds 9 "~-
8 CD :55; J
m’
(D
'6
(D
Q
3” e
8 co
U)
CG
0
(D
._>.
a as
E
o
A

8 _

(\l

O _.

 

 

 

1950 1960 1970 1980 1990 2000

Reporting Year

Figure 4. Time series of resistance case discover by chemical class for those classes with >
25 cases of resistance. A case is the first discovery of a species and pesticidecombination.

43

6.28% ‘3 mommo go: 25 36%: 6653.5:me 2o: 688,380 Sta 0: 5:: 8.60% so: go Autozuot .m meant

.mw>

ommr ova?
_ _ _ r a
mwwmw

 

0

009 00V 003

mommo 3:223. 55 32523.5 252

.mm>

89 9.9
_ _ _ _ H
o N

     

r

 
 
 

m

3205mm: 262

030

09

00L

suoueonqnd

pauoda; sapgonsad mau

.mo>
ommr ovmr ocmr
_ _ _

F _

m— w n F
mw

 

Fm
wow

$20.33 x msoonwv mommo Boz

.wo>

 

 

«20on is:

0017

008

 

peuodaJ saseo mau

pauodaJ sagoads meu

44

Table 3. Cases of Insecticide and Acaricide Resistance by Chemical Class

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2 Rn by# of
cases of prior to 1980 prior to 1990 prior to 2000
resistance
1 Organochlorines 757 Organophosphates 1050 Organophosphates 1136
2 22 2 2Or-gan-c);;);hates—669—C;r2gano;hl6r2ines 838 Organochlorines 844
2 2 32 22 2C2ar2b2am2ate—s — 22 82922—2C2a2r5a2mates 169 Pyrethroids 224
2 2 42 2 2ll2lliscellaneous 2232122 Pyrethroids _ 166 Carbamates 2 20222
2 25 2 Pyr2ethr6id—s2 22 22262 2111;..ng322 39 Bacterials 46
2 2 2e 2 24.21.131.52 2 18 22F..2m2}§;ts22 2 220222Mi252e12a};§2 2152
2 27 2;;J 2 123 21.88.1106... 2 2212322232515}: 2225122
2 a 2 2 filotirlid: 2 322 32:53:83;th 22% 3:33:33“ 2212
9 Bacterials 2 Formamidines 6 Arsenicals 1 3
2 22 2102 222Fo2rrn2amgmes 2— 2 *Organgnsm 52—2 Organotins 428—2
2 211222 6223.95 212 33391:... 24221635411}: 22 e2 2
2 2 1222 222 1222866; 21 Mme 2223 Formamidines 6 22
22 22 123 222 Summpounds 1 Sulfur Compounds 2 2 Nicotinoids 6
22 214 2 222 2 gye—rmectins 2 1 2 Rotenone 2 —2——
2 2 15 2 2 2 Dinitrofenols 1 Sulfur Compounds 2
2 16 2 i— Rotenone 2 1 2 Dinitrofenols 1 222
2 2172 2 — 22 —2 2 Phenylpyrazoles 2221222

 

 

45

 

 

 

 

 

 

 

 

 

 

 

46

0.000 E35900 000300.00 820200000
50: 00.303“. 65.000 .505 00.90 003 ~03 0m vm 0020020009 EEBQE. or
0.02002. 0.06% 00:00:60
3. 00:01 0:085:20 :09: 00.3 tum? mm mm -0020sz 000:2 0
E0263 000002 0050202 0:0000§
000000... £0300 .500 009.250 39 59 mm. um 002300: £590: 0
500.2000 0.0000500 02:02:00
:0800 9:085:20 :09: 39 80.. co mm 002.0503 00:05 N
03E .0030 _._000 :5:
00. 00: :2“. 000.: :0: 52 $9 on 9» 00039800000 00:0?0201 o
0.000 90E9 .2080 000000.00- 0005:0000
2060 0005.00 .0300 400.080 ooom mmmr um ov 002080950 000005304 m
x0: 0500 _ - 030000.05
02005505000 0500 009 $2 mm 3 _._000 002005 méquom v
00.00H0m0>
2:00 . . . 0.000080:
- 000009 0505 000.: 009 mom? on we - 000.53 03;: m
998 68a. .25 .820 .026: 822:8
£9: . 0.0500200. £00000?
x000-0:0E0_D E03550: 90:00.0 ooow $9 an on -000___0S_0 0:321 N
0.955050
02E 000.00 . . :000 002::
. -. 550; 83989 82 $2 2: E - . F
003000 02:. .35: 60:00 0095280000“ 0393053
0502 £00: tome tome 00000000 8:80.301 ~0bco.b..E0l 00.52% «cmm
=0EE00 0:000; “ME 05 E b0t010¢
000:. .50 00> 00020000.. 5.3.

*0 00:52 0039.301
00 00:52

 

E0508 $000600; 002209000 009:: 00 00:5: 3 00x20; .mbocht< E03030 5:05. QB. .v 0300.

 

.000.

 

 

 

 

 

 

 

 

 

 

9.0. 25 0.9008005 003 000: N NN 00000000000 0000:0350. 8
0002. 9.9.0 000 0000000
€020.08. 08.01028 88 009 a 8 0.0.8002 090$... 9
.0002. .000 :00m .9009 .00:00 000.9000 0.9 000 m
.0260 . 0.0.0...

0020 00.0.02 89.0 0.08 00.: 009 3 mm 002.20....8 0:008 0:03 9
0.002060 90.09 9000.00 0.900200. 0.00.0.0
2.02. 20 .0.00 .00:00 30: 000: N mN -0029000 0E0>00:0I 3

30.00.0000: 0.90.0 090.000.003.00
9.00000. .9005 00000 000.00 30: N02. on N 0022.00 x900 or
0.000 . 99000.0...

:98 20 805. 830.000., 8:8 009 009 E 00 0022.00 .9080 0:0... 0.

>00<£mm3200 0. 00.0900? 30: N o a: v N N 0 0.900200. 0:99.:

00:00 00.. 0 >0m. 9900 00:00 9.9.0 000.9000 0.900025 3

. 0.900.000 :0E0:
2.00 no: :50 8.. 009 009 0. mm 08.2.00 88.80 9
$9.02. 0.900.000
9900:0026 00:00 000: 50: FN mm -00200.>0.0 .0000. 0.0.E0m NF
0.90.0 0:90.:
9.00000. 00:0... 000.02.. ~00: 53 0N mm 0022.00 000.90 .900 E
0502 :0:=:00 0.00: 20:0. 20.00. 00000.00 00:90.03. .00.0.>::.0l 00.0000 «:00.
E000. .0... 05 :. 09.0102
.00.: .0 .00: 0005.90.
.0 .3502 0090.000:
.0 .00.:02

 

390.00. 209.000. 00:00:88 000.:0 .0 .0900: A: 00::0. 00000.53. 0:90:03. 5:00... 00.. «0000::00: .0 0:00.»

47

CHAPTER 3. ANALYSIS: RESISTANCE AND REGISTRATION IN THE US
Introduction

The first case of pesticide resistance was reported in the United States in 1908
and resistance has had dire consequences in the US ever since (see Chapter 1).
From that first case of resistance until 2002, the US had over 1,500 new cases
reported (see Chapter 2), ranking it as the number one country in the world.
Given these sheer numbers of cases, it is crucial to have some notion of what
drives resistance development in the US. This chapter exploits a number of
databases to explore how pesticide resistance, use, and implementation trends
are drivers of resistance cases.

Arthropod pesticide resistance is not a new phenomenon, contrary to the
opinion some authors (Horowitz and Denholm 2001), resistance populations
appeared in the 403 just years after DDT was introduced in the US (Brown 1951).
For example, in eastern Florida, resistance occurred after less than 3 years of
DDT use to control Salt Marsh Flies. Certainly some arthropod pest species did
not develop pesticide resistant populations. For example, no pesticide resistance
report existed by 2003 for potato aphid, Macrosiphum euphorbiae (Thomas), a
Potato-Y virus (PYV) vector, although the US potato crop received intense
pesticide application (National Research Council 2000) for defense of one of the
most resistant defoliating insects known, the Colorado Potato Beetle (CPB),
Leptinotarsa decimlineata (Error! Reference source not found.,Chapter 2).
However, when pesticide resistance does develop in any species, it may do so

within a few seasons. It has been claimed, for those species with reported

48

resistance, that resistance develops within 3-5 yrs after pesticide implementation
(Mote-Sanchez, Bills, and Whalon 2001).

Since the first DDT-resistant Musca domestica population was discovered in
1947, pesticide resistance discoveries have accumulated steadily over time
(Georghiou 1983 and see Chapter 2). The 19603 witnessed the discovery of 19
newly resistant species per year and 15 per year in the 19705. (Figure 4).
However, in the 19903 that rate declined to an average of just over 2 species
annually (see chapter 2). Georghiou speculated on the cause of this decreased
rate (Georghiou, 1986 #62) , however he did not correlate this decline with
historical records of these causes.

The declining trend may well have resulted from reduced pesticide use more
rapid introduction of new modes of action or IPM. A complex confluence of an
arthropod’s agroecosystem, genetic potential, life history traits, and other factors
select pesticide-resistant populations (see chapter 4). However, without the
pesticide application there would be no selection for a resistant population.
Therefore, when seeking a causal agent, the first step is to review a pesticide’s
history in the pest host-range.

Many aspects described pesticide usage, but typical reports included
averages of annual sales (lbs), seasonal applications (lbs, number), in which
crop it was applied, area applied (acres), rate per unit area (lbs/acre) (e.g.
(NASS 1996, 1997, 1998, 1999, 2000, Donaldson et al. 2002), etc).

Assuming all pesticide sold is applied, pesticide sales indicate a measure of

the total pesticide load in the environment. No federal agency collects pesticide

49

sales data or comprehensive pesticideuse quantities. However, the USEPA
produces annual estimates. Donaldson (2002) reported sales as total lbs active
ingredient for fungicides, herbicides, insecticides (including acaricides), and
‘other’ pesticides. Yet, using these statistics as indicators of resistance selection
pressure may be problematic as the weight of a substance has little bearing on
its toxicity (Graham-Bryce 1987).

Insecticides and acaricides were incredibly diverse. By the end of the century,
there were at least 17 classes of insecticides and acaricides in existence, more
depending on one’s classification scheme. In the US in 1997, over 67
insecticides and acaricides were applied (>1% crop acreage per state, 81 crops
on 307 million acres) . Excluding oils, this meant active ingredient application
rates ranged from 0.3to 30 lbs per acre for the season. Over two hundred
insecticides and acaricides were available worldwide in the year 2000 (Tomlin
2000). Toxicity varies among all pesticides more than one-thousand fold
(Copplestone 1977). Indeed, the mix of pesticides available has changed over
time, since recent dramatic pesticide policy changes (see below). Overall
tonnage of active ingredients sold gave little indication of the selection pressure.
More active ingredient details would be needed before conclusions about
resistance are drawn. However, in many specific agro-ecosystems, pesticide
application histories are well known and use and resistance may be correlated on
a case-by-case bases.

Long-term and consistent use measurements for specific active ingredients

were largely not available in the public literature. Proprietary data were available

50

but cost prohibitive. National pesticide usage information was not systematically
collected by any federal agency until 1990 when the National Agricultural
Statistics Service (NASS) began a complex of chemical use surveys. In addition,
while evolution of resistance is a phenomenon of a population’s exposure to
pesticides over time and space, pesticide use reporting has rarely been spatially
or temporally explicit. Data are typically annual aggregates for the entire US or
specific US states. Some exceptions are data collected by the California
Department of Pesticide Regulation (Federighi 2001) and infrequent, detailed
grower surveys from NASS (NASS 1996). In Michigan, a pilot survey
administered by NASS in 1998 (vegetables) and 1999 (fruit) that was temporally
but not spatially definite (Harris 2000). Prior to 1990 when NASS began their
surveys, pesticide use data was sparse or overly summarized. Pesticide use
details are critical for understanding pest management practices, however
existing pesticide use statistics as anything but a general guide would be
challenged as indicators of resistance development (Osteen 1992).

Agrochemical use is often reported per crop. The US Pesticide Impact
assessment program produced several detailed assessment of pesticide use and
benefits for individual crops, but these were not persistent over time. However,
while the crop in which resistance discovery was made may have been
documented, it is not recorded in the MSU database as many population cross
farm-boundaries and population structure influence by exposure in any crop.

To correlate Ps and R, at minimum it should be known when the pesticide was

first used anywhere. Pesticide use is not a good indicator of that fact, as survey

51

does not necessarily coincide with a compound’s introduction. NASS, for
example, conducted surveys for fruit and vegetables on alternating years (NASS
1991-2001). However a good indicator of this is the date of registration, as a
pesticide cannot be used, other than for experimental or emergency purposes,
without being federally registered. The 1947 US Federal Insecticide, Fungicide,
and Rodenticide Act (F IF RA) required that all pesticides distributed or sold in
interstate commerce had to be registered and labeled federally. Responsibility for
registration was first under the direction of the United States Department of
Agriculture (USDA), but the Environmental Protection Agency (EPA) assumed
registration activities after its establishment in 1970. In 1972, FIFRA was
expanded, and for the first time required reassessment of pesticide risk, allowing
reregistering only if the risk was not unreasonable when compared to the benefits
of its use. The Food Quality Protection Act of 1996 (FQPA) not only removed
benefits from the risk assessment equations, but also required a cumulative risk
assessment of compounds per class. These and other dramatic changes served
to further restricted registration. FQPA introduced a strict timetable for re-
assessment and 38% of pesticide uses and ingredients were voluntarily or
otherwise cancelled in the first five years especially for many belonging to older
chemical classes such as organophosphates, carbamates and remaining
organochlorines (Mulkey 2002). For any point in time, those pesticides with
active registrations served as good proxy for which pesticides were generally put

into the market.

52

Using the year of registration as a proxy for the date of first use, and year of
first reported resistance as a proxy for when pesticide resistance first developed,
some broad exploration of time to resistance was be made. It was thought that
such an analysis would help describe the trend in reduced rate of resistance
discovery. To accomplish this, US EPA registration records since 1947 were
compared with pesticide resistance reports. This was a step towards meeting
the need to anticipate the development of resistance in numerous systems, and
or to contribute what measured factors will help such prediction.

Methods

To manage registrations, USEPA stored pesticide data in a computerized
database at the NC Computing Center names the Pesticide Product lnforrnation
System (PPIS) (Office of Pesticide Programs 2000). Periodic snapshots of the
data were made available the public as a collect of 24 data sets \as fixed-width
ASCII files. Data files dated December 2000 were downloaded from USEPA.
The well-documented database included registered product code, percent of
each active ingredients, legal uses of the pesticides, or ‘sites’ (e.g. Apple, foliar),
the target pests, original registration date, and cancellation date if any. The date
appearing is the first date it was registered in FIFRA section 3 (FIFRA 1996) or
240. All historical product registrations since 1947 passage of original FIFRA
except so-called ‘emergency’ registrations under F lFRA section 18 section were
included in this data. Products were coded by the class of organisms against

which there were affected using the standard terminology (insecticide, herbicide,

53

etc). Only products with insecticidal and acaricidal designations were
considered.

USEPA pesticide products were formulations of one or more active ingredients
(AI). Active ingredients and products are stored in distinct tables in the PPIS
database, and share a key table with the product registration number, the active
ingredient Shaugnessy code, and the percent concentration of the Al in the
product, by weight. First registration date is per product but linkable to Al.

This analysis assumes that, once registered, a pesticide is uniformly adopted
wherever pest populations might occur. This is a limitation as sites as added
gradually to a pesticide’s registration, especially minor uses, Use sites are
added over time, however dates are not stored. Therefore this analysis assumes
all use sites were available at time of registration.

The MSU database was basis for cases of resistance in the US (see Chapter
2). Pesticide compounds in the MSU Resistance Database were coded with
Chemical Abstract Service (CAS) numbers (Ware 1989) and then cross-coded
with the USEPA Shaughnessy code using a PPIS data table matching these
codes. The relational database structure also linkage of pesticide tables in both
database to be linked once common codes are entered.

All data tables were stored and manipulated using relational database system
software MySQL version 4.1 (Axmark et al. 2003). Aggregated results were
created using standard SOL code (ISO 1992). Statistical summaries and
graphics were produced with R-language and computing environment for

statistics (lhaka and Gentleman 1996), version 1.7.0.

54

Results

Of 98,000 registered products with insecticidal and acaricidal uses registered
between 1947 and 2002, approximately 29,000 were identified in the resistance
database as compound with a case of resistance in the US. The number of use-
sites was tracked for each class of active ingredient (Figure 6). Use sites are
associate with pesticide products. However, by associating this with active
ingredients it may then be correlated with compounds for which resistance has
occurred. While cases of ogranochlorine (OCL) resistance exceed
organophosphates (OP) until the 19803 ( Error! Reference source not found.,
Chapter 2, page ), registration records show far greater use capacity for the
organophophates since the 19503. This may be due the sheer numbers of CPS

registered.

Discussion

The trend for the reduced rate of new species today correlates well with the
number of distinct pesticide ingredients input into the environment. Over time
pesticides have been developed in new chemical families, increasing the
diversity of pesticide chemical classes. These increased in classes we would
expect to see an increase in cases, that it, either more insects becoming multiply
resistant, adding the novel compounds to the list. The diverse toxins have
diverse binding sites or modes of action; hence we don‘t expect to see cross-

resistance. Certainly this has been true for Leptinotara decimlineata and

55

.88 s 3.3 £8583.
"Ema. .efisuom £38m 885 u mos .3338 u 05 32538590 u do Eocsei u .51 .smsmfimo u 56 3.2828590 u do
.350 Eugene 3 .ommomfib oozfimfiog 05 5 83233 E 52 beam 23. ESE: Co 892:: :32 ..:o.:§m.&mm $3;me m: go 053E...» .0 Eat

30>
ooou camp 08—. 2.2. owmv 002. 03—.

unmuuuefqlrr...j a nnnnnnnn fistula.» o

 

 

O 4

 

 

j I .. . 82

 

 

 

 

 

 

.l 88 t

n
.29 . \\ m.
2?. < \t m
400' I coon 5
”Eli m
«20 IT m

m

 

é \\ 009.
< ooom

 

 

 

 

 

 

 

 

 

ooow

56

# of Active Ingredlents Registered

83% USED ms 5 meoagflumm mutual bum mo:$w..mmm 23.6281 onEct< v6 «SEER N 93mm

 

 

 

50>
mmmw Saw mwmw owmw mnmr osmr mgr 08F 93.. 0mm; mvmw
0 all \lllplllllllilr llIf I I . I O
8“ Im . .
oooeoooooeoooeooo lIoom
CO». I
I i I ooor
coo Ia #
m...
8
o
s
I . . m
cool I .. I I l .83 m
. - o
. . I e
. - . I . m
ooor I I I I IowaaflmyamslwfidflcomolgamiElaoIll m
. . 8.6on Emfiwmm o>zmSanI6|
I llll ‘ II I , 9:28.55 m>=o< Co CenozoohEqu ooow
, $30 262 @253an +
89 I- # 8566mm mefiemc. m>=o<
I I I oomw
00¢? 4:

 

 

o8 w coon

57

neriestoxins. But in general could we say that novel compounds are less
likely to have resistance? Perhaps the slow appearance of resistant species to
classes such as fipronil, tebufenozide, etc is because the areas in which these
are used is dramatically less than past compounds.

One argument for the slow down in new resistant species appearance is that
we have exhausted the available pests (Georghiou 1990a). As pesticides
increase the cover of our landscape in time and space, selection is increased,
and those that could become resistant, have been.

Refinements to this analysis would attempt to link where the resistance
occurred and a combination of the level of pesticide use, pesticide registration
counts, relevant biological factors (weather) may help estimate probability of
exposure to pesticide residues and hence selection. Pesticide registration is not
a perfect indicator of selection pressure as pesticides are not deployed
immediately nationwide but gradually as sites and pests are add to pesticide

labels during an on-going registration and re-registration process.

58

APPENDIX 1

Appendix 1. Structured Query Language data definition code
The following code can be used to recreate the database using MySQL

version 4.0 (Axmark, 2002).

/*
MysSQL Server version 4.0.12-max
# create database if not exists

*/

/*
Table struture for arthropod_families

*/

drop table if exists ‘arthropod_families‘;

CREATE TABLE ‘arthropod_families‘ (
‘taxfamily‘ varchar(20) NOT NULL default ",
‘taxorder‘ varchar(20) NOT NULL default ",
‘taxsuborder‘ varchar(SO) default NULL,
‘commonname‘ varchar(lSO) default ",
PRIMARY KEY (‘taxfamily‘,‘taxorder‘),
KEY ‘arthropod_familiesorder‘ (‘taxorder‘)

) TYPE=MyISAM;

/*
Table struture for arthropod_groups

*/

drop table if exists ‘arthropod_groups‘;

CREATE TABLE ‘arthropod_groups‘ (
‘arthropodgroupcode‘ char(S) NOT NULL default ",
‘arthropodgroup‘ char(50) default NULL,

KEY ‘insectgroupcode‘ (‘arthropodgroupcode‘)

) TYPE=MyISAM;

59

/*
Table struture for arthropod_orders

*/

drop table if exists ‘arthropod_orders‘;

CREATE TABLE ‘arthropod_orders‘ (
‘taxorder‘ varchar(50) NOT NULL default ",
‘taxclass‘ varchar(SO) default NULL,
‘previousorder‘ varchar(SO) default NULL,
‘commonname‘ varchar(lSO) default ",
PRIMARY KEY (‘taxorder‘)
) TYPE=MyISAM COMMENT='from MS Access table arthropod_orders';

/*
Table struture for arthropod_refs

*/

drop table if exists ‘arthropod_refs‘f

CREATE TABLE ‘arthropod_refs‘ (

‘arthropodID‘ int(11) NOT NULL default '0',

‘referenceID‘ int(11) NOT NULL default '0',

‘topics‘ varchar(lOO) default ",

PRIMARY KEY (‘arthropodID‘,‘referenceID‘)
) TYPE=MyISAM COMMENT='citations to references with general information
about this';

/*
Table struture for arthropod_synonym

*/

drop table if exists ‘arthropod_synonym‘;

CREATE TABLE ‘arthropod_synonym‘ (
‘arthropodid‘ int(11) NOT NULL default '0',
‘speciessynonym‘ char(50) NOT NULL default ",
PRIMARY KEY (‘arthropodid‘,‘speciessynonym‘),
KEY ‘arthropodid‘ (‘arthropodid‘)

) TYPE=MyISAM;

60

/*
Table struture for arthropods

*/

drop table if exists ‘arthropods‘;

CREATE TABLE ‘arthropods‘ (
‘arthropodid‘ int(11) NOT NULL default '0',
‘taxfamily‘ varchar(20) NOT NULL default ",
‘genus‘ varchar(20) NOT NULL default ",
‘species‘ varchar(30) NOT NULL default ",
‘pest_of‘ varchar(lOO) default NULL,
‘arthropodgroup‘ varchar(SO) default NULL,
‘common_name‘ varchar(50) default NULL,
‘notes‘ text,
‘arthropodgroupcode‘ varchar(SO) default NULL,
‘author‘ varchar(lOO) default NULL,
‘url‘ varchar(255) default NULL,
‘taxorder‘ varchar(50) NOT NULL default ",
‘citation‘ text,
PRIMARY KEY (‘arthropodid‘l,
UNIQUE KEY ‘indexspecies‘ (‘genus‘,‘species‘)
) TYPE=MyISAM;

/*
Table struture for compound_group

*/

drop table if exists ‘compound_group‘;

CREATE TABLE ‘compound_group‘ (
‘chemgroupid‘ char(10) NOT NULL default ",
‘chemgroupsubid‘ char(50) NOT NULL default ",
‘chemicalgroup‘ char(255) default NULL,
‘group_description‘ char(50) default NULL,
‘groupedasother‘ tinyint(4) default NULL,
‘newgen‘ tinyint(4) default NULL,
KEY ‘chemgroupid‘ (‘chemgroupid‘),
KEY ‘chemgroupsubid‘ (‘chemgroupsubid‘)

I TYPE=MyISAM;

/*

61

Table struture for compound_synonym

*/

drop table if exists ‘compound_synonym‘;

CREATE TABLE ‘compound_synonym‘ (
‘msuchemid‘ int(11) NOT NULL default '0',
‘chemsynonym‘ char(SO) NOT NULL default ",
‘synonymtype‘ char(SO) default NULL,
‘synonymsource‘ char(SO) default NULL,
PRIMARY KEY (‘msuchemid‘,‘chemsynonym‘),
KEY ‘chem_id‘ (‘msuchemid‘)

) TYPE=MyISAM;

/*
Table struture for compounds

*/

drop table if exists ‘compounds‘;

CREATE TABLE ‘compounds‘ (
‘compoundid‘ int(11) NOT NULL default '0',
‘compoundname‘ char(75) default NULL,
‘cas‘ int(11) default NULL,
‘shaugnessy‘ int(11) default NULL,
‘msuchemgroup‘ char(10) default NULL,
‘entry_notes‘ char(100) default NULL,
‘mode_of_action‘ char(255) default NULL,
‘other_properties‘ char(255) default NULL,
PRIMARY KEY (‘compoundid‘),
KEY ‘name‘ (‘compoundname‘),
KEY ‘shaugnessy‘ (‘shaugnessy‘)

) TYPE=MyISAM;

/*
Table struture for compounds_in_formulation

*/

drop table if exists ‘compounds_in_formulation‘;

CREATE TABLE ‘compounds_in_formulation‘ (
‘formulationid‘ int(11) NOT NULL default '0',

62

‘compoundid‘ int(11) NOT NULL default '0',
PRIMARY KEY (‘formulationid‘,‘compoundid‘l
) TYPE=MyISAM;

/*
Table struture for countries

*/

drop table if exists ‘countries‘;

CREATE TABLE ‘countries‘ (
‘countryid‘ int(10) NOT NULL auto_increment,
‘country‘ char(SO) NOT NULL default ",
‘continent‘ char(SO) default NULL,
‘westerncountry‘ tinyint(1) default '0',
PRIMARY KEY (‘countryid‘I

) TYPE=MyISAM;

/*
Table struture for documentation

*/

drop table if exists ‘documentation‘;

CREATE TABLE ‘documentation‘ (
‘id‘ int(11) NOT NULL default '0',
‘author‘ varchar(ZSS) NOT NULL default ",
‘year‘ double default NULL,
‘journal_title‘ varchar(255) default NULL,
‘volumepage_num‘ varchar(ZSS) default NULL,
‘title‘ varchar(ZSS) NOT NULL default ",
‘key_words‘ varchar(255) NOT NULL default ",
‘summary‘ blob,
‘category‘ varchar(lOO) default NULL,
‘select_field‘ tinyint(4) default NULL,
‘checkedoutby‘ varchar(20) default NULL,
‘checkedoutdate‘ datetime default NULL,
‘document_type‘ varchar(lOO) default NULL,
‘lyear‘ year(4) default NULL,
KEY ‘authorindex‘ (‘author‘),
KEY ‘key_words‘ (‘key_words‘),
KEY ‘id‘ (‘id‘).
KEY ‘titleindex‘ (‘title‘)

) TYPE=MyISAM;

63

/*
Table struture for formulation_type

*/

drop table if exists ‘formulation_type‘;

CREATE TABLE ‘formulation_type‘ (
‘form_id‘ int(11) NOT NULL default '0',
‘form‘ char(SO) NOT NULL default ",
‘form_description‘ char(100) default NULL,
PRIMARY KEY (‘form_id‘)

) TYPE=MyISAM;

/*
Table struture for literature

*/

drop table if exists ‘literature‘;

CREATE TABLE ‘literature‘ (
‘id‘ int(11) NOT NULL default '0',
‘author‘ varchar(255) NOT NULL default ",
‘year‘ double default NULL,
‘title‘ varchar(255) NOT NULL default ",
‘journa1_title‘ varchar(255) default NULL,
‘volumepage_num‘ varchar(255) default NULL,
‘summary‘ blob,
‘category‘ varchar(lOO) default NULL,
‘key_words‘ varchar(255) NOT NULL default ",
‘lyear‘ year(4) default NULL,
PRIMARY KEY (‘id‘)

) TYPE=MyISAM;

/*
Table struture for pesticide_formulation

*/

drop table if exists ‘pesticide_formulation‘;

64

CREATE TABLE ‘pesticide_formulation‘ (
‘formulationid‘ int(11) NOT NULL default '0',
‘pesticide‘ char(100) default NULL,
‘form_type‘ int(11) NOT NULL default '0',
‘with_synergist‘ tinyint(4) default NULL,
‘fromlabel‘ tinyint(4) default NULL,
‘epanumber‘ char(SO) default NULL,
‘msuchemgroup‘ char(10) default NULL,
‘entry_notes‘ char(255) default NULL,
‘reviewedby‘ char(3) default NULL,

PRIMARY KEY (‘formulationid‘),
KEY ‘formulation_typepesticide_formulation‘ (‘form_type‘)

) TYPE=MyISAM;

/*
Table struture for ref_epa_chemname

*/

drop table if exists ‘ref_epa_chemname‘;

CREATE TABLE ‘ref_epa_chemname‘ (
‘pc_code‘ int(11) NOT NULL default '0',
‘cname_type‘ char(1) NOT NULL default ",
‘seq_nr‘ char(3) default NULL,
‘pc_prefix‘ char(15) default NULL,
‘pc_name‘ char(253) NOT NULL default ",
KEY ‘chemnamecname_type‘ (‘cname_type‘),
KEY ‘chemnamepc_code‘ (‘pC_code‘),
KEY ‘name‘ (‘pc_prefix‘,‘pc_name‘)

I TYPE=MyISAM;

/*
Table struture for ref_years

*/

drop table if exists ‘ref_years‘;

CREATE TABLE ‘ref_years‘ (
‘year‘ double default NULL,
‘dYear‘ date default NULL

) TYPE=MyISAM;

65

/*
Table struture for res_backup

*/

drop table if exists ‘res_backup‘;

CREATE TABLE ‘res_backup‘ (
‘arthropodid‘ int(11) NOT NULL default '0',
‘formulationid‘ int(11) NOT NULL default '0',
‘literatureid‘ int(11) NOT NULL default '0',
‘chemical‘ char(SO) NOT NULL default ",
‘year‘ int(11) NOT NULL default '0',
‘region‘ char(SO) default ",
‘cross_resistance‘ tinyint(4) default NULL,
‘error_check‘ tinyint(4) default NULL,
‘notes‘ char(SO) default NULL,
‘source‘ char(6) default NULL,
‘subref‘ char(255) default NULL,
‘resistancedev‘ char(3) default NULL,
‘faorefok‘ tinyint(4) default NULL,
‘reviewer‘ char(SO) default NULL,
‘subrefid‘ int(11) default NULL,
‘rYear‘ date default NULL,
‘countryid‘ int(10) default '0'

I TYPE=MyISAM;

/*
Table struture for resistance

*/

drop table if exists ‘resistance‘;

CREATE TABLE ‘resistance‘ (
‘arthropodid‘ int(11) NOT NULL default '0',
‘formulationid‘ int(11) NOT NULL default '0',
‘literatureid‘ int(11) NOT NULL default '0',
‘chemical‘ char(SO) NOT NULL default ",
‘year‘ int(11) NOT NULL default '0',
‘region‘ char(SO) default ",
‘cross_resistance‘ int(11) default '0',
‘error_check‘ tinyint(4) default NULL,
‘notes‘ char(SO) default NULL,
‘source‘ char(6) default NULL,
‘subref‘ char(255) default NULL,
‘resistancedev‘ char(3) default NULL,
‘faorefok‘ tinyint(4) default NULL,
‘reviewer‘ char(SO) default NULL,

66

‘subrefid‘ int(11) default NULL,

‘rYear‘ date default NULL,

‘countryid‘ int(10) default '0'
) TYPE=MyISAM;

drop table if exists ‘subrefs‘;

CREATE TABLE ‘subrefs‘ (
‘subref‘ varchar(255) NOT NULL default ",
‘full_citation‘ text,
‘year‘ int(11) default NULL,
‘msulitid‘ int(11) NOT NULL default '0',
‘lib_call_num‘ varchar(SO) default NULL,
‘subrefid‘ int(11) NOT NULL auto_increment,
PRIMARY KEY (‘subrefid‘),
KEY ‘short_ref‘ (‘subref‘),
KEY ‘msulitid‘ (‘msulitid‘)
) TYPE=MyISAM COMMENT='Full citations for embedded references in
GLT91';

67

BIBLIOGRAPHY

Axmark, D., M. Widenius, J. Cole, A. Lentz, and P. DuBois 2003. MySQL
Database Server computer program, version 4.0.12-nt. By Axmark, D., M.
Widenius, J. Cole, A. Lentz, and P. DuBois, Uppsala, Sweden.

Berlinger, M. J., S. Leibush-Mordechai, N. More, and D. Fridja. 1993.
Insecticide efficiency in controlling the whitefly Bemisia tabaci and the
Tomato Yellow Leaf Curl Virus. Phytoparasitica 21: 176.

Bolin, P. C., W. D. Hutchison, and A. D. A. 1999. Long-term selection for
resistance to Bacillus thuringiensis Cry1Ac endotoxin in a Minnesota
population of European corn borer (Lepidoptera: Crambidae). Journal of
Economic Entomology 92: 1021-1030.

Brown, A. W. A. 1951. Insect Control by Chemicals. Wiley & Sons, Inc., New
York.

Brown, A. W. A. 1958. Insecticide Resistance. World Health Organization,
Geneva.

Carlson, J. D., M. E. Whalon, D. A. Landis, and S. H. Gage. 1991. Evidence
for long-range transport of potato leafhopper into Michigan, pp. 123-128,
10th Conf. On Biometeorology and Aerobiology and Special Session on
Hydrometeorology, Genva.

Chen, P. P.-S. 1981 . A preliminary framework for entity-relationship models,
Entity-Relationship Approach to lnforrnation Modeling. ER Institute,
Saugus, CA.

Codd, E. F. 1970. A relational model of data for large shared data banks.
Communications of the Association Computing Machinery 13: 377-387.

Colluzzi, M., A. Sagatini, V. Petrarca, and M. DiDeco. 1977. Behavioural
divergence between mosquito with different inversion karyotypes in
polymorphic populations of the Anopheles gambiae complex. Nature 226:
832-833.

Copplestone, J. F. 1977. A global view of pesticide safety, pp. 147-155. In D. L.
Watson and A. W. A. Brown [eds], Pesticide Management and Insecticide
Resistance. Academic Press, New York.

Croft, B. A. 1990. Arthropod Biological Control Agents and Pesticides. John
Wiley & Sons, New York.

Croft, B. A., and A. W. A. Brown. 1975. Response of arthropod natural enemies
to insecticides. Annual Review of Entomology 20: 285-355.

68

Crow, J. F. 1960. Genetics of insecticide resistance: general considerations.
Misc Publ Entomol Soc Am 2: 69-74.

Date, C. J. 1992. Dont' Encode Information into Primary Keysl, pp. 99-110. In C.
J. Date and H. Daniven [eds.], Relational Database Writings 1989-1991.
Addison Wesley, Reading, Massachusetts.

Date, C. J. 1995. An Introduction to Database Systems. Addison-Wesley,
Reading, Massachusetts.

Denholm, I., A. R. Horowitz, M. Cahill, and I. Ishaaya. 1998. Management of
Resistance to Novel Insecticides, pp. 260-281. In I. Ishaaya and D.
Degheele [eds.], Insecticides with Novel Modes of Action: Mechanism and
Application. Springer-Vedag, Berlinz.

Donaldson, D., T. Kiely, and A. Grube. 2002. Pesticides Industry Sales and
Usage: 1998 and 1999 Market Estimates. United States Environmental
Protection Agency

Dover, M. J., and B. A. Croft. 1986. Integration of Policy for Resistance
Management, pp. 422-435. In National Research Council [ed.], Pesticide
Resistance: Strategies and Tactics for Management. National Academy
Press, Washington, DC.

Elmarsi, R., and S. Navathe. 1994. Fundamentals of Database Systems.
Benajmine/Cummings, Redwood City, CA.

Federighi, V., ed. 2001. Regulating Pesticides: The California Story, A Guide to
Pesticide Regulation in California, pp. 135. California Department of
Pesticide Regulation, Sacramento, CA.

ffrench-Constant, R. H., and R. T. Roush. 1990. Resistance Detection and
Documentation: The Relative Roles of Pesticidal and Biochemical Assay,
pp. 4-38. In R. T. Roush and B. E. Tabashnik [eds.], Pesticide Resistance
in Arthropods. Chapman and Hall, New York.

Fuxa, J. R., F. L. Mitchell, and A. R. Richter. 1988. Resistance of Spodoptera
frugiperda (LepzNoctuidae) to a nuclear polyhedrosis virus in the field and
laboratory. Entomophaga 35: 55-63.

Georghiou, G. P. 1972. The Evolution of Resistance to Pesticides. Ann. Rev.
Ecol. Systematics. 3: 113.

Georghiou, G. P. 1983. Pesticide resistance in time and space, pp. 1-46. In G.
P. Georghiou and T. Saito [eds.], Pest Resistance to Pesticides. Plenum
Press, New York.

69

Georghiou, G. P. 1986. The Magnitude of the Resistance Problem, pp. 14-44. In
National Research Council [ed.], Pesticide Resistance: Strategies and
Tactics for Management. National Academy Press, Washington, DC.

Georghiou, G. P. 1990a. Overview of insecticide resistance, pp. 18-41. In M. B.
Green, H. M. LeBaron and W. K. Moberg [eds], Managing Resistance to
Agrochemicals: From Fundamental Research to Practical Strategies.
American Chemical Society, Washington, D. C.

Georghiou, G. P. 1990b. The Effect of Agrochemicals on Vector Populations,
pp. 183-187. In R. T. Roush and B. E. Tabashnik [eds], Pesticide
Resistance in Arthropods. Chapman and Hall, New York.

Georghiou, G. P., and C. E. Taylor. 1986. Factors Influencing the Evolution of
Resistance, pp. 157-169. In National Research Council [ed.], Pesticide
Resistance: Strategies and Tactics for Management, Washington, DC.

Georghiou, G. P., and A. Lagunes-Tejeda. 1991. The Occurrence of
Resistance to Pesticides in Arthropods: an index of cases reported
through 1989. Food and Agriculture Organization (FAO) of the United
Nations, Rome.

Gianessi, L. P., and J. E. Anderson. 1995. Pesticide Use in US Crop
Production. National Center for Food and Agrictulture Policy, Washingon,
DC.

Gianessi, L. P., and M. Marcelli. 2000. Pesticide Use in US. Crop Production:
1997. National Center for Food and Agricultural Policy, Washington, DC.

Glass, E. H. 1986a. Executive summary, pp. 1-9. In National Research Council
[ed.], Pesticide Resistance: Strategies and Tactics for Management.
National Academy Press, Washington, DC.

Glass, E. H., Chairman, Committee on Strategies for the Management of
Pesticide Resistant Pest Populations. 1986b. Introduction, pp. 11-13. In
National Research Council [ed.], Pesticide resistance: strategies and
tactics for management. National Academy Press, Washington, DC.

Graham-Bryce. 1987. Chemical Methods, pp. 148-151. In A. J. Burn, T. H.
Coaker and P. C. Jepson [eds], Integrated Pest Management. Academic
Press, London.

Halpin, T. 1996. Conceptual Schema and Relational Database Design. Prentice
Hall, Sydney, Austrailia.

Harris, C. 2000. Summary of Pilot Michigan Pesticide Use Survey for
Vegetables. Unpublished Data.

70

Horowitz, A. R., and I. Denholm. 2001. Impact of Insecticide Resistance
Mechanisms on Management Strategies, pp. 334-335. In I. Ishaaya [ed.],
Biochemical Sites of Insecticide Action and Resistance. Springer, Berlin.

Ihaka, R., and R. Gentleman. 1996. R: A Language for Data Analysis and
Graphics. Journal of Computational and Graphical Statistics 5: 299-314.

IRAC. 1997. Insecticide Resistance Action Committee. In C. Tomlin [ed.], The
Pesticide Manual. British Crop Protection Council, Farnham, U.K.

ISO, I. O. f. S. 1992. Database Language SQL.

Jepsen, P., and P. Hennigan. 2000. SELECTV Database of Natural Enemies
Resistant to Pesticides. On-Iine:
http://www.ent3.orst.edu/Phosure/database/selctv/selctv.htm. Oregon
State University.

Keiding, J. 1977. Resistance in the Housefly in Denmark and Elsewhere, pp.
261-265. In D. L. Watson and A. W. A. Brown [eds], Pesticide
Management and Insecticide Resistance. Academic Press, New York.

Kennedy, G. G., and D. R. Smitley. 1985. Dispersal, pp. 233-242. In W. Helle
and M. W. Sabelis [eds], Spider Mites, Their Biology, Natural Enemies
and Control. Elsevier, Amsterdam.

Mariath, H. A., C. J. Orton, and C. J. Shivas. 1990. Resistance to oviposition
suppressants in Lucilia cupn'na, pp. 52. In J. A. McKenzie, P. J. Martin and
J. H. Arunde [eds], Resistance Management in Parasites of Sheep.
Australian Wool Corporation, Melbourne.

McGaughy, W. H., and M. E. Whalon. 1992. Managing Insect Resistance to
Bacillus thuringiensis toxins. Science 258: 1451 -1455.

McKenzie, J. A., A. G. Parker, and J. L. Yen. 1992. Polygenic and single gene
responses to selection for resistance to diazinon in Lucilia cupn'na.
Genetics 130: 613-620.

Melander, A. L. 1914. Can insects become resistant to sprays? Journal of
Economic Entomology 7: 167-173.

Messenger, P. S., and R. v. d. Bosch. 1971. The adaptability of introduced
biological control agents, pp. 68-92. In C. B. Huffaker [ed.], Biological
Control. Plenum Press, New York.

Metcalf, R. L. 1983. Implications and prognosis of resistance, pp. 718-725. In G.
P. Georghiou and T. Saito [eds], Pest Resistance to Pesticides. Plenum
Press, New York.

71

Mulkey, M. E. 2002. FY 2001 Annual Report, pp. 16. Office of Pesticide
Programs (OPP), Enviromental Protection Agency (EPA), Washington,
DC.

NASS, N. A. S. S. 1996. Agricultural Chemical Usage: 1995 Fruit Summary
Enterprise Level. US Department of Agriculture, Washington, DC.

NASS, N. A. S. S. 1997. Agricultural Chemical Usage: 1996 Vegetable
Summary. US Department of Agriculture, Washington, DC.

NASS, N. A. S. S. 1998. Agricultural Chemical Usage: 1997 Fruit and Nuts
Summary. US Department of Agriculture, Washington, DC.

NASS, N. A. S. S. 1999. Agricultural Chemical Usage: 1998 Vegetable
Summary. US Department of Agriculture, Washington, DC.

NASS, N. A. S. S. 2000. Agricultural Chemical Usage: 1999 Field Crops
Summary. US Department of Agriculture, Washington, DC.

National Research Council. 2000. The Future Role of Pesticides in US
Agriculture. National Academy Press, Washington, DC.

Office of Pesticide Programs, 0. 2000. Pesticide Product Information System
(PPIS). US Environmental Proection Agency.

O'Neal, M. E., M. E. Gray, and C. A. Smyth. 1999. Population Characteristics of
a Western Corn Rootworm (Coleoptera: Chyrsome. Journal of Economic
Entomology 92: 1301-1310.

Osteen, C. 1992. Pesticide Use Trends and Issues in the United States, pp. 307-
338. In D. Pimentel and H. Lehman [eds], The Pesticide Question.
Chapman & Hall, New York.

Osteen, C. D., and M. Padgitt. 2002. Economic Issues of Agricultural Pesticide
Use and Policy in the United States, pp. 59-96. In W. B. Wheeler [ed.],
Pesticides in Agriculture and the Environment. Marcel Dekker, New York.

Otto, D. E., E. Moll, and R. P. 1992. A Critical comment on the Evaluation of the
Resistance Level in Field Populations, pp. 363-376. In D. Otto, B. and B.
Webber [eds], Insecticides: Mechanisms of Action and Resistance.
Intercept Limited, Andover, England.

Pimentel, D., H. Acquay, M. Biltonen, P. Rice, M. Silva, J. Nelson, and V.
Lipn. 1992a. Assessment of environmental and economic impacts of
pesticide use, pp. 47-84. In D. Pimentel and H. Lehman [eds], The
Pesticide Question. Chapman & Hall, New York.

72

Pimentel, D. C., C. Kirby, and A. Shroff. 1992b. The relationship between
'cosmetic standards' for foods and pesticide use, pp. 56-58. In D. C.
Pimentel and H. Lehman [eds], The Pesticide Question. Chapman and
Hall, New York.

Roush, R. T., and G. L. Miller. 1986. Considerations for the design of insecticide
resistance monitoring programs. Journal of Economic Entomology 79:
293-298.

Roush, R. T., and J. C. Daly. 1990. The role of population genetics in resistance
research and management, pp. 97-152. In R. T. Roush and B. E.
Tabashnik [eds], Pesticide Resistance to Pesticides. Chapman and Hall,
New York.

Roush, R. T., and J. A. McKenzie. 1991. Ecological genetics of insecticide and
acaracide resistance. Annual Review of Entomology 32: 361-380.

Sawicki, R. M. 1987. Definition, detection and documentation of insecticide
resistance, pp. 105-117. In M. G. Ford, D. W. Holloman, B. P. S. Khambay
and R. M. Sawicki [eds], Combating Resistance to Xenobiotics: Biological
and Chemical Approaches. Ellis Horwood Ltd, England.

Shelton, A. M., J. L. Robertson, J. D. Tang, C. Perez, S. D. Eigenbrode, H. K.
Preisler, W. T. Wilsey, and R. J. Cooley. 1993. Resistance of
Diamondback Moth (Lepidoptera:Plutellidae) to Bacillus thruingiensis
Subspecies in the the Field. Journal of Economic Entomology 86: 697-
705.

Tabashnik, B. E., N. L. Cushing, N. Finson, and M. W. Johnson. 1990. Field
Development of Resistance to Bacillus thuringiensis in Diamonback Moth
(Lepidoptera: Plutellidae). Journal of Economic Entomology 83: 1671-
1676.

Theiling, K. M., and B. A. Croft. 1988. Pesticide Side Effects on Arthropod
Natural Enemies: a Database Summary. Agric Ecosys Environ 21: 191-
218.

UNEP. 1979. The State of the environment: Selected topics-1979. United
National Environmental Programme, Governing Counci|,Seventh Session.,
Nairobi.

Ware, G. 1989. The Pesticide Book. Thomson Publications, Fresno, CA.

Whalon, M. E., and W. H. McGaughy. 1988. Bacillus thuringiensis: Use and
resistance management, pp. 106-137. In I. Ishaaya and D. Degheele
[eds], Insecticides with Novel Modes of Action: Mechanism and
Application. Springer-Verlag, Delhi, India.

73

Whalon, M. E., S. G. Gage, and M. Dover. 1984. Estimation of Pesticide Use
Through the Cooperative Crop Monitoring System in Michigan Apple
Production. Journal of Economic Entomology 77: 559-564.

World Health Organization. 1957. Expert Comittee on Malaria, Seventh Report,
Geneva.

Zhao, J.-Z., Y.-X. Li, H. L. Collins, and A. M. Shelton. 2002a. Examination of
the F2 Screen for Rare Resistance Alleles to Bacillus thuringiensis Toxins
in the Diamondback Moth (Lepidoptera: Plutellidae). Journal of Economic
Entomology 95: 14-21.

Zhao, J.-Z., Y.-X. Li, H. L. Collins, L. Gusukuma-Minuto, R. F. L. Mau, G. D.
Thompson, and A. M. Shelton. 2002b. Monitoring and Characterization
of Diamondback Moth (Lepidoptera: Plutellidae) Resistance to Spinosad.
Journal of Economic Entomology 95: 430-436.

Zhao, J.-Z., H. L. Collins, J. D. Tang, J. Cao, E. D. Earle, R. T. Roush, S.
Herrero, B. Escariche, J. Ferre, and A. M. Shelton. 2000. Development
and characterization of diamondback moth resistance to transgenic
broccoli expressing high levels of Cry1 C. Applied Environmental
Microbiology 66: 3784-3789.

74

 

llllllllllil!lllllllllllllllllIlllllillllllllllllllIlll
3 1293 02460 4237