Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 194
Pesticide Resistance: Strategies and Tactics for Management.
1986. National Academy Press, Washington, D.C.
F
Computer Simulation as a Too}
tor Pesticide Resistance Management
BRUCE E. TABASHNIK
Computer simulation may be usefulfor devising strategies to retard
pesticide resistance in pests and to promote it in beneficials. This
paper demonstrates the use of simulation to study interactions among
factors influencing resistance development, describes efforts to test
models of resistance development, and illustrates management ap-
plications of computer models. Suggested guidelines for future tests
of resistance models are to (IJ establish baseline data on suscepti-
bility before populations are selectedfor resistance, (2) conduct tests
under field conditions, (3J use experimental estimates of biological
parameters in models, and (4J replicate treatments. Modelers of
pesticide resistance must test models, explore the implications of
polygenic resistance, and incorporate alternative controls such as
biological control in models.
INTRODUCTION
Pest species have developed resistance to pesticides faster than beneficial
organisms, limiting the integration of biological and chemical controls. Re-
sistant strains of more than 400 insect and mite species have been recorded,
but fewer than 10 percent are beneficial (Georghiou and Mellon, 1983; Croft
and Strickler, 19831. The goals of resistance management are to retard re-
sistance in pests and to promote it in beneficials. Models of pesticide resis-
tance can be useful tools for working toward these goals. Various types of
models have played an essential role in building a conceptual framework for
resistance management (Table 1; Taylor, 19831. This paper emphasizes sim-
ulation modeling as a component of management and identifies future di
194
OCR for page 195
COMPUTER SIMULATION
TABLE 1 Modeling Studies of Pesticide Resistance
195
Factors Emphasized
Biological Operational Economic
Studies
Analytical
MacDonald, 1959 X
Comins, 1977a X
Curtis et al., 1978 X X
Gressel and Segel, 1978 X X
Taylor and Georghiou, 1979 X
Cook, 1981 X
Skylakakis, 1981 X X
Wood and Mani, 1981 X X
Muggleton, 1982 X
Simulation
Georghiou and Taylor, 1977a,b X X
Greever and Georghiou, 1979 X X
Plapp et al., 1979 X
Kable and Jeffrey, 1980 X
Curtis, 1981 X X
Taylor and Georghiou, 1982 X X
Tabashnik and Croft, 1982, 1985 X X
Levy et al., 1983 X
Taylor et al., 1983 X X
Knipling and Klassen, 1984 X
Dowd et al., 1984 X X
Optimization
Hueth and Regev, 1974 X X
Taylor and Headley, 1975 X X
Guttierrez et al., 1976, 1979 X X
Comins, 1977b, 1979 X X
Shoemaker, 1982 X X
Statistical/Empir~cal
Georghiou, 1980 X
Tabashnik and Croft, 1985 X
SOURCE: The model classifications are based on Logan (1982) and Taylor (1983). The list of
studies is expanded from Taylor (1983) but is not intended to be exhaustive.
rections for modeling that can increase its usefulness as a resistance manage-
ment tool.
MODELING ASSUMPTIONS
The key assumptions of the models discussed in this paper (Tabashnik and
Croft, 1982, 1985; Taylor and Georghiou, 1982; Taylor et al., 1983) are as
follows:
1. Resistance is controlled primarily by a single-gene locus with two
OCR for page 196
196
POPULATION BIOLOGY OF PESTICIDE RESISTANCE
alleles, R (resistant) and S (susceptible), with a fixed dose-mortality line for
each genotype.
2. The dose-mortality line for RS heterozygotes is intermediate between
the SS (susceptible) and RR (resistant) lines. At low pesticide doses RS
heterozygotes are not killed, and the R gene is effectively dominant; at
high doses RS heterozygotes are killed, and the R gene is effectively
recessive.
3. The insect life cycle is divided into substages, with transition proba-
bilities between substages determined by natural and pesticide mortalities.
4. Immigrants are primarily susceptible and have at least one day to mate
and reproduce before being killed by a pesticide.
INTERACTIONS
There are four main classes of conditions for resistance development: (1)
no immigration, low pesticide dose (R gene functionally dominant); (2) no
immigration, high pesticide dose (R gene functionally codominant or reces-
sive); (3) high immigration, low dose; and (4) high immigration, high dose.
Initial modeling studies that focused on different subsets of these four main
classes arrived at apparently conflicting results (e.g., contrast Georghiou and
Taylor, 1977a,b, with Comins, 1977a, and Taylor and Georghiou, 19791.
It was not clear whether contradictions arose from differences in modeling
approaches or from differences in conditions among various studies.
Tabashnik and Croft (1982) examined the influence of various factors on
rates of resistance development under all four main classes of conditions.
Results showed that the way certain factors influence the rate of resistance
evolution depends on which of the four classes of conditions are present. In
other words the same factor may have a different influence under different
background conditions.
One of the most striking examples of the interaction effect is the influence
of pesticide dose on the time to develop resistance (Figure 1~. Without
immigration resistance developed faster as dose increased. With immigration
there were two distinct phases. At low doses resistance developed faster as
dose increased, paralleling the case without immigration. At high doses,
however, resistance developed more slowly as dose increased. These results
are consistent with Comins (1977a). Without immigration the rate of resis-
tance development is determined primarily by the rate at which S genes are
removed from the population. As dose increases, S genes are removed more
rapidly; resistance develops faster. The situation with low doses and immi-
gration is similar. With immigration and doses high enough to kill RS het-
erozygotes, however, pesticide mortality also removes R genes from the
population. As dose increases in this range, more RS heterozygotes are killed,
leaving relatively few resistant (RR) individuals. The RR survivors are ef
OCR for page 197
COMPUTER SIMULATION
Stable
>10
`,, 1 0
_ 9
a)
~ 8
ce
a',
._ ,
In
a,
a:
o
-
a)
i- 4
3
2
1
O
197
~ , ~
I\
\\
1
~1
Air
a,
\\: Aim
\~ ,~y
"\ /
" ~
No Immigration
0~ On --O
.0003 .001 .003 .01 .03 .1
Dose (Log Scale)
FIGURE 1 Effects of dose on the rate of evolution of resistance. Conditions: O or 100
grants daily, biweekly treatments of adults. Source: Tabashnik and Croft (1982~.
fectively swamped out by susceptible immigrants, thereby retarding resis-
tance development.
The simulation results suggest that one of the most important factors
influencing the rate of resistance evolution is the number of generations per
year. Under all four classes of conditions, resistance developed faster as the
number of generations per year increased. Field observations of resistance
development in soil and apple arthropods (Georghiou, 1980; Tabashnik and
Croft, 1985) are consistent with this prediction.
A summary of the influence of various factors on resistance development
(Table 2) highlights the interactions among factors. Increases in the opera-
tional [actors (dose, spray frequency, and fraction of the life cycle exposed
to pesticide) made resistance develop faster when there was no immigration
(both low- and high-dose range) and when there was immigration and a low
dose. The opposite occurred with immigration and a high dose. Some bio-
logical factors (fecundity, survival, and initial population size) had little effect
in the absence of immigration, but increases in these factors made resistance
evolve faster when there was immigration. Two biological factors (genera-
tions/year and immigration) had the same influence under all four classes of
conditions.
OCR for page 198
198
POPULATION BIOLOGY OF PESTICIDE RESISTANCE
TABLE 2 The Influence of Operational and Biological Factors on
Resistance Development under Four Main Classes of Conditions
No Immigration
High Immigration
Low High Low High
Factors Dosea Doseb Dosea Doseb
Operational
Dose + + +
Spray Frequency + + +
Life Stages Exposed + + +
Biological
Generations per Year + + + +
Immigration
Fecundity 0 0 + +
Survivorship 0 0 + +
Initial Population Size 0 0 + +
Initial R Gene Frequency + 0 + +
Reproductive Disadvantage - 0 - -
DominanceC + O + +
NOTE: + shows that increasing the listed factor speeds resistance development; - shows that
increasing the listed factor slows resistance development; 0 shows little or no effect.
aKills only SS, R gene functionally dominant.
bKills SS and some RS, R gene functionally codominant or recessive.
CBased on Comins (1977a), Georghiou and Taylor (1977a), Wood and Mani (1981), and Tabashnik
(unpublished).
SOURCE: Tabashnik and Croft (1982).
The most important conclusion from this simulation approach is that the
influence of certain factors will depend on the presence or absence of im-
migration by susceptibles and on the functional dominance of the R gene
(i.e., dose). Therefore, it is necessary to develop resistance management
strategies that are appropriate for specific ecological and operational contexts.
TESTING MODELS
Experimental tests of pesticide resistance models are sorely needed (Taylor,
1983~. There have been more than 25 papers describing resistance models
during the past 10 years (Table 1), but only two studies explicitly test such
models (Taylor et al., 1983; Tabashnik and Croft, 19851. These two studies
represent opposite types of validation. The following discussion summarizes
results of the studies and suggests how elements of both approaches can be
combined to produce an especially powerful test of resistance models.
Taylor et al. (1983) used laboratory house fly (Musca domestica) popu-
lations to test a model of evolution of resistance to dieldrin, an organochlorine
insecticide. Resistance to dieldrin is due to a single gene, and three fly
OCR for page 199
COMPUTER SIMULATION
199
genotypes are distinguishable by bioassay (Georghiou et al., 19631. Taylor
et al. (1983) simulated five different treatment regimes, then compared the
predicted resistance gene frequencies and population sizes with those ob-
served in five corresponding experimental cages.
All of the biological parameters used in the simulations were measured
directly from laboratory fly populations. The initial conditions were alike
for all cages (90 SS + 10 RS individuals of each sex per cage), and each
cage received a different treatment: (A) control no insecticide and no im-
migration, (B) slow insecticide decay and immigration, (C) fast decay and
immigration, (D) no decay and no immigration, and (E) no decay and im-
migration. Immigration was achieved by adding 25 individuals (24 SS + 1
RS) to the appropriate cages three times weekly. Dieldr~n was incorporated
in the larval medium and acted only on larvae and newly eclosed adults. The
initial dieldr~n concentration (40 ppm) was the same in treatments B to E,
but decay rates corresponding to insecticide half-lives of 1.0 and 0.5 days
were mimicked by using decreasing dieldr~n concentrations in successive
treatments. Each cage was run for 57 days (about four generations).
The results showed a strong correlation between predicted and observed
values for the final R gene frequency in each treatment (Figure 21. Both the
simulations and experiments support earlier predictions that immigration by
susceptibles can retard the evolution of resistance, especially when the ratio
of immigrants to residents in the treated population is high (Coming, 1977a;
Taylor and Georghiou, 1979; Tabashnik and Croft, 19821.
A 'to
llJ
o
~ 0.8
11
111
Z 0.6
L`J
lo
~ 0.2
LL
o.o
C.
_
MA
B '/
0.4 _
/
BE
/
/
/
/
D ,'
o.o 0.2 0.4 0.6 0.8 1.0
OBSERVED R GENE FREQUENCY
FIGURE 2 Predicted versus observed resistance (R) gene frequencies in caged house
flies. Dashed line shows predicted = observed. Letters indicate treatments (see text)
(Taylor et al., 1983~.
OCR for page 200
200
POPULATION BIOLOGY OF PESTICIDE RESISTANCE
This validation study shows that in a highly defined situation, model
predictions may correspond well with reality. Because virtually all of the
biological and operational parameters were either measured or controlled,
the correspondence between predictions and observations is no accident. The
model appears to incorporate the essential processes affecting evolution of
resistance in the system studied. The system studied, however, was highly
artificial, and its relationship to field systems is unclear. Validation in an
artificial system probably cannot adequately address the question of whether
model predictions apply to field situations.
Tabashnik and Croft (1985) tested a resistance model by comparing sim-
ulated times versus historically observed times to evolve resistance to azin-
phosmethyl in the field for 24 species of apple pests and natural enemies.
Azinphosmethyl is an organophosphorous insecticide that has remained a
major apple pest-control tool in North America for almost 30 years. The
long-term patterns of evolution of resistance to azinphosmethyl among the
diverse apple orchard insects and mites constitute a unique data set for testing
predictions about resistance.
To represent 24 different apple arthropod species in the simulation, the
following population ecology parameters were estimated independently for
each species: generations/year, fecundity, immigration, natural (nonpesti-
cide) mortality, initial population size, development rate, sex ratio, pesticide
exposure in orchards, and percent of time spent in orchards by adults. Pa-
rameter values and historically observed times to evolve resistance for each
species were based on a survey of 24 fruit entomologists (Croft, 19821.
Operational and genetic factors were held constant for all 24 species. All
species were subjected to the same simulated pesticide dose, spray schedule,
and pesticide half-life because all species were present in the same habitat
and were exposed to a similar treatment regime in the field. The genetic
basis of resistance, dose-mortality lines, and initial R gene frequency were
assumed to be the same for all species because these parameters are virtually
impossible to estimate for most species. Further, Tabashnik and Croft (1985)
sought to determine how much of the variation in rates of evolution of
resistance could be explained by differences among species in population
ecology, with all other factors being constant.
The results show a significant rank correlation between predicted and
historically observed times to evolve resistance for the 12 pest species and
the 12 natural-enemy species (Figure 31. Thus, ecological differences among
apple species are sufficient for explaining observed variation in rates of
resistance development among pests and natural enemies.
There was no consistent bias in the predictions for pests, but predicted
times were consistently less than observed times for natural enemies, sug-
gesting that the original assumptions may omit factors that slow resistance
development in natural enemies. The original assumptions about natural
OCR for page 201
COMPUTER SIMULATION
1
~ a
Tp. P.b
~25
25
- %
c'
~,1 0
o
o3_
-
E
F 2
1
cL
0.5
,
~ Put
Ap,
-
,' ,
Rptn
0.25 , . . ... . .
Dp
Tt'
1 2 5
201
Hh Crr
b | y sr ~
,' Aa Cc|lAm
,," Sp
,'
iAf
'To
. I I I l,,,,l
10 25 >25
10 25 > 25 1 2
Observed Time (years-log scale)
5
FIGURE 3 Predicted versus observed times to evolve resistance to azinphosmethyl for
apple arthropods. Predicted time (~) = simulated time to evolve resistance using means
of estimates of population ecology parameters. Observed time = years after 1955 (first
widespread use of azinphosmethyl) to first report of resistance. Vertical bars show range
of predicted times from sensitivity analysis. Dashed lines show predicted = observed.
A. Pests: n = 12. Spearman's rank correlation coefficient, rS = 0.652, p < 0.05. Key:
Aa = Archips argyrospilus, Ap = Aphis pomi, Av = Argyrotaenia velutinana, Cn =
Conotrachelus nenuphar, Dp = Dysaphis plantaginea, Lp = Laspeyresia pomonella,
Pb = Phyllonorcyter blancardella, Pu = Panonychus ulmi, Qp = Qua~lraspidiotus
perniciousus, Rp = Rhagoletis pomonella, Tp = Typhlocyba pomccria, Tu = Tetran-
ychus urticae B. Natural enemies: n - 12. rS = 0.692, p c 0.025. Key: Aa =
Aphidoletes aphidimyza, Ae = Anagrus epos, Af = Amblyseiusfallacis, Am = Aphelinus
mali, At = Aphelopus typhlocyba, Cc = Chrysopa carnea, Cm = Coleomegilla maculata
lingi, Hh = Hyaliodes hart), 0i = 0rius insidiosus, Sp = Stethorus punctum, Sr =
Syrphus ribesii, To = Typhlodromus occidentalis. Source: Tabashnik and Croft, 19851.
enemies were modified to incorporate the preadaptation and food-limitation
hypotheses. Incorporating the preadaptation hypothesis (pests are preadapted
to detoxify pesticides because they detoxify plant poisons, but natural enemies
are less preadapted) (Croft and Morse, 1979; Mullin et al., 1982) did not
substantially improve the correspondence between predicted and observed
times. Adding the food-limitation hypothesis (a natural enemy evolves re-
sistance only after its prey/host is resistant, because pesticides drastically
reduce food for natural enemies by eliminating susceptible prey/hosts) (Huf
OCR for page 202
202
POPULATION BIOLOGY OF PESTICIDE RESISTANCE
>25
25
_%
a)
c'
O 10
-
u'
c.
c, 5
>'
a)
._
~ 2
a)
._ ~
0.5
0.25
!
,' ~
L
''
I'
1 2 5 10
,' Aa ~ Cc)
Tol' jAf Asp l
~1
I ~ I
25 >25
Observed Time (years-log scale)
F~GuRE 4 Effects of the food-limitation hypothesis on predicted times for natural enemies
to evolve resistance. n = 12. rS = 0.806, p < 0.005 (see Figure 3 for key to species
names). Open circles indicate predictions with the food-lim~tation hypothesis incorporated;
dark circles indicate predictions under initial assumptions. Arrows show change in pre-
dictions due to food-limitation hypothesis.
faker, 1971), however, substantially improved the correspondence between
predicted and observed times for all six natural enemies that were initially
predicted to evolve resistance too fast (Figure 4~.
These results suggest that food limitation following pesticide applications
may be an important factor in retarding evolution of resistance in natural
enemies. If this is so it may be possible to promote resistance development
in natural enemies by ensuring them an adequate food supply following
sprays either by reducing mortality to their prey/hosts or by providing an
alternate food source when prey/hosts are scarce.
The validation study of Tabashnik and Croft (1985) provides encourage-
ment that model results can be applied to field situations. That study, how-
ever, relies on estimated values for many important parameters. Tabashnik
and Croft (1985) address this problem in part by a sensitivity analysis dem-
onstrating that many of the model's predictions were minimally affected by
substantial variation in some key parameters that are difficult to estimate,
but that are potentially influential (immigration, initial population size, and
fecundity; see sensitivity bars in Figure 31.
OCR for page 203
COMPUTER SIMULATION
TABLE 3 Predicted Time (years) for the European Red Mite (Panonychus
ulmi) to Evolve Pesticide Resistance under Different Pesticide Doses and
Application Frequencies
203
Application Frequency
Pesticide Initial (Sprays/Year)
Dosea Mortality 6 3 1 112b
0.01 93% 1.5 1.7 2.6 5.7
0.002 73% 1.6 1.9 6.5 19.6
0.001 50% 1.5 2.2 13.6 >25
aArbitrary units
bOne spray every 2 years
SOURCE: Tabashnik and Croft (1985).
It seems that a powerful approach to testing resistance models can be
developed by combining elements from both of the studies described above.
Guidelines are as follows:
· Establish baseline data on susceptibility before populations are selected
for pesticide resistance. Rates of resistance development can be measured
only if initial susceptibility is known.
· Conduct tests under field conditions or conditions similar to the field.
It may be especially important to use large initial population sizes if genes
conferring resistance are rare.
· Obtain experimental estimates of basic biological parameters (e.g., fe-
cundity) required for modeling.
· Replicate treatments.
Field experiments that might promote rapid evolution of new resistances
in pests should not be performed. Although experimental selection for re-
sistance is costly and time-consuming (Taylor, 1983), unintentional selection
for resistance is widespread. Extremely valuable data bases on resistance
could be developed by concomitant monitoring of field treatment regimes
and susceptibility levels in field populations. Such data would provide a
sound basis for evaluating management tactics as well as models of pesticide
resistance.
MANAGEMENT APPLICATIONS
Computer simulations can be used to project the consequences of alter-
native control strategies. For example, Tabashnik and Croft (1985) simulated
resistance development by the European red mite (Panonychus ulmi) under
12 management schemes based on three pesticide doses and four application
schedules (Table 31. Resistance was predicted to occur within three years
when intermediate to high acaricide doses (causing 50 to 93 percent initial
OCR for page 204
204
POPULATION BIOLOGY OF PESTICIDE RESISTANCE
mortality) and frequent applications (three to six per season) were simulated.
If both dose and application frequency are reduced, resistance in the European
red mite is predicted to be delayed from 7 to more than 25 years.
The projected times for resistance development in the European red mite
are consistent with observed patterns of resistance to the acaricide cyhexatin
in the United States. Since cyhexatin was introduced in 1970, resistance has
not occurred in apple orchards, where it has been used judiciously in con-
junction with biological control by predators. Cyhexatin resistance has oc-
curred rapidly, however, in pear-apple interplants, where biological control
is difficult and acaricide use is more intensive (Croft and Bode, 19831.
CONCLUSION
Modelers of pesticide resistance face three major challenges in the im-
mediate future. First, and most important, models of pesticide resistance
must be tested. Second, the implications of polygenically based pesticide
resistance need to be explored. With few exceptions models of pesticide
resistance assume one locus-two allele genetics, but many resistances may
be polygenic (Plapp et al., 1979~. Two of the papers in this volume take
important steps toward addressing this challenge (Uyenoyama, Via). Third,
alternative control methods such as biological control should be incorporated
into models of pesticide resistance. The most promising way to retard resis-
tance is to reduce pesticide use by integrating pesticides with other controls,
yet current models generally assume that pesticides are the sole control
method. If these challenges are addressed, modeling will play an increasingly
important role in managing pesticide resistance.
ACKNOWLEDGMENTS
Special thanks to B. A. Croft for his assistance and encouragement. R.
T. Roush and R. M. May provided valuable comments. Support was provided
by the Research and Training Fund, University of Hawaii and USDA-
HAW00947H. Paper Number 2919 of the Hawaii Institute of Tropical Ag-
nculture and Human Resources journal series.
REFERENCES
Comins, H. N. 1977a. The development of insecticide resistance in the presence of migration. J.
Theor. Biol. 64:177-197.
Comins, H. N. 1977b. The management of pesticide resistance. J. Theor. Biol. 65:399-420.
Comins, H. N. 1979. The control of adaptable pests. Pp. 217-266 in Pest Management, Proceedings
of an International Conference, G. A. Norton and C. S. Holling, eds. Oxford: Pergamon.
Cook, L. M. 1981. The ecological factor in assessment of resistance in pest populations. Pestic.
Sci. 12:582-586.
OCR for page 205
COMPUTER SIMULATION
205
Croft, B. A. 1982. Arthropod resistance to insecticides: A key to pest control failures and successes
in North American apple orchards. Entomol. Exp. Appl. 3:88-110.
Croft, B. A., and W. M. Bode. 1983. Tactics for deciduous fruit IPM. Pp. 270-291 in Integrated
Management of Insect Pests of Pome and Stone Fruits, B. A. Croft and S. C. Hoyt, eds. New
York: Interscience.
Croft, B. A., and J. G. Morse. 1979. Recent advances on pesticide resistance in natural enemies.
Entomophaga 24:3- 11.
Croft, B. A., and K. Strickler. 1983. Natural enemy resistance to pesticides: Documentation,
characterization, theory and application. Pp. 669-702 in Pest Resistance to Pesticides, G. P.
Georghiou and T. Saito, eds. New York: Plenum.
Curtis, C. F. 1981. Possible methods of inhibiting or reversing the evolution of insecticide resistance
in mosquitoes. Pestic. Sci. 12:557-564.
Curtis, C. F., L. M. Cook, and R. J. Wood. 1978. Selection for and against insecticide resistance
and possible methods of inhibiting the evolution of resistance in mosquitoes. Ecol. Entomol.
3:273-287.
Dowd, P. F., T. C. Sparks, and F. L. Mitchell. 1984. A microcomputer simulation program for
demonstrating the development of insecticide resistance. Bull. Entomol. Soc. Am. 30:37-41.
Georghiou, G. P. 1980. Insecticide resistance and prospects for its management. Residue Rev.
76:131-145.
Georghiou, G. P., R. B. March, and G. E. Printy. 1963. A study on genetics of dieldrin resistance
in the house fly (Musca domestica L.). Bull. W. H. O. 29:155-165.
Georghiou, G. P., and R. B. Mellon. 1983. Pesticide resistance in time and space. Pp. 1-46 in
Pest Resistance to Pesticides, G. P. Georghiou and T. Saito, eds. New York: Plenum.
Georghiou, G. P., and C. E. Taylor. 1977a. Genetic and biological influences in the evolution of
insecticide resistance. J. Econ. Entomol. 70:319-323.
Georghiou, G. P., and C. E. Taylor. 1977b. Operational influences in the evolution of insecticide
resistance. J. Econ. Entomol. 70:653-658.
Greever, J., and G. P. Georghiou. 1979. Computer simulations of control strategies for Culex
tarsalis (Diptera: Culicidae). J. Med. Entomol. 16:18~188.
Gressel, J., and L. A. Segel. 1978. The paucity of plants evolving genetic resistance to herbicides:
Possible reasons and implications. J. Theor. Biol. 75:349-371.
Guttierez, A. P., U. Regev, and C. G. Summers. 1976. Computer model aids in weevil control.
Calif. Agric. April:8-9.
Guttierez, A. P., U. Regev, and H. Shalet. 1979. An economic optimization model of pesticide
resistance: Alfalfa and Egyptian alfalfa weevil an example. Environ. Entomol. 8:101-109.
Hueth, D., and U. Regev. 1974. Optimal agricultural pest management with increasing pest resis-
tance. Am. J. Agric. Econ. 56:543-552.
Huffaker, C. B. 1971. The ecology of pesticide interference with insect populations. Pp. 92-107
in Agricultural Chemicals- Harmony or Discord for Food, People, and the Environment, J. E.
Swift, ed. Berkeley: University of California, Division of Agricultural Science.
Kable, P. F., and H. Jeffery. 1980. Selection for tolerance in organisms exposed to sprays of biocide
mixtures: A theoretical model. Phytopathology 70:8-12.
Knipling, E. F., and W. Klassen. 1984. Influence of insecticide use patterns on the development
of resistance to insecticides: A theoretical study. Southwest. Entomol. 9:351-368.
Levy, Y., R. Levi, and Y. Cohen. 1983. Buildup of a pathogen subpopulation resistant to a systemic
fungicide under various control strategies: A flexible simulation model. Phytopathology 73: 1475-
1480.
Logan, J. A. 1982. Recent advances and new directions in phytoseiid population models. Pp. 49-
71 in Recent Advances in Knowledge of the Phytoseiidae, M. A. Hoy, ed. Publ. 3284. Berkeley,
Calif.: Agricultural Sciences Publications (Publ. No. 3284.)
MacDonald, G. 1959. The dynamics of resistance to insecticides by anophelines. Riv. di Parassitol.
20:305-315.
OCR for page 206
206
POPULATION BIOLOGY OF PESTICIDE RESISTANCE
Muggleton, J. 1982. A model for the elimination of insecticide resistance using heterozygous
disadvantage. Heredity 49:247-251.
Mullin, C. A., B. A. Croft, K. Strickler, F. Matsumura, and J. R. Miller. 1982. Detoxification
enzyme differences between an herbivorous and predatory mite. Science 217:1270-1272.
Plapp, F. W., Jr., C. R. Browning, and P. J. H. Sharpe. 1979. Analysis of rate of development of
insecticide resistance based on simulation of a genetic model. Environ. Entomol. 8:494-500.
Shoemaker, C. A. 1982. Optimal integrated control of univoltine pest populations with age structure.
Oper. Res. 30:40-61.
Skylakakis, G. 1981. Effects of alternating and mixing pesticides on the buildup of fungal resistance.
Phytopathology 71:1119-1121.
Tabashnik, B. E., and B. A. Croft. 1982. Managing pesticide resistance in crop-arthropod complexes:
Interactions between biological and operational factors. Environ. Entomol. 11:1137-1144.
Tabashnik, B. E., and B. A. Croft. 1985. Evolution of pesticide resistance in apple pests and their
natural enemies. Entomophaga 30:37-49.
Taylor, C. E. 1983. Evolution of resistance to insecticides: The role of mathematical models and
computer simulations. Pp. 163-173 in Pest Resistance to Pesticides, G. P. Georghiou and T.
Saito, eds. New York: Plenum.
Taylor, C. E., and G. P. Georghiou. 1979. Suppression of insecticide resistance by alteration of
gene dominance and migration. J. Econ. Entomol. 72:105-109.
Taylor, C. E., and G. P. Georghiou. 1982. Influence of pesticide persistence in evolution of
resistance. Environ. Entomol. 11:746-750.
Taylor, C. E., F. Quaglia, and G. P. Georghiou. 1983. Evolution of resistance to insecticides: A
cage study on the influence of migration and insecticide decay rates. J. Econ. Entomol. 76:704-
707.
Taylor, C. R., and J. C. Headley. 1975. Insecticide resistance and the evolution of control strategies
for an insect population. Can. Entomol. 107:237-242.
Wood, R. J., and G. S. Mani. 1981. The effective dominance of resistance genes in relation to the
evolution of resistance. Pestic. Sci. 12:573-581.
Representative terms from entire chapter:
resistance development
