Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Pesticide Resistance: Strategies and Tactics for Management. 1986. National Academy Press, Washington, D.C. Expenmental Population Genetics and Ecological Studies of Pesticide Resistance in Insects and Mites RICHARD T. ROUSH and BRIAN A. CROFT Current data on the population genetics and ecological aspects of pesticide resistance in insects and mites are reviewed. Very little is known about initialfrequencies of resistance alleles. In some cases dominance depends on pesticide dose. In untreated habitats the fit- nesses of resistant genotypes appear to be 50 to 100 percent of those susceptible genotypes. Up to about 20 percent of the individuals in treated populations escape exposure. Important parameters for fur- ther research include initial allele frequencies and immigration rates. INTRODUCTION One objective of population genetics is to describe evolutionary change. Even though pesticide resistance has long been recognized as evolutionary change (Dobzhansky, 1937), most detailed empirical population studies of insecticide and acaricide resistance have been conducted only during the last decade. Although more work is needed, these experiments complement ex- periences of field entomologists and provide new insights into management of resistance. The rate of allelic substitution in a closed population is a function of allele frequency, dominance, and relative fitnesses of genotypes. Arthropod pop- ulations, however, are rarely completely closed. Gene flow ("migration" in the genetic sense) between populations varies tremendously, depending on species and ecological factors affecting insect and mite dispersal. Thus, the evolution of resistance can be described only by considering both genetic and ecological factors. 257
258 POPULATION BIOLOGY OF PESTICIDE RESISTANCE Most population genetics theory assumes that the traits under consideration are controlled by only one or two loci (for contrast see paper by Via, this volume). Many studies have shown that resistance of practical significance in the field is almost always controlled by one or two loci (Plapp, 1976; Brown, 19671. Although polygenic resistance does occur in nature (Liu et al., 1 98 1 ), it is much more common in the laboratory (Whitten and McKenzie, 1982; Roush, 19831. Therefore, it is not unreasonable to assume that the toxicology of resistance is due to a single allelic variant at one locus (for additional discussion see papers by May and Dobson, Uyenoyama, and Via, this volume). INITIAL ALLELE FREQUENCY Little is known about allelic frequencies prior to pesticide selection, al- though they may range from 10-2 (Georghiou and Taylor, 1977) to 10-13 (Whitten and McKenzie, 19821. These frequencies should be measurable, but this has been accurately done only with dieldrin resistance in Anopheles gambiae, where the frequency is unusually high (Wood and Bishop, 19811. Initial allele frequency is a function of selection against the resistant genotypes and mutation rate (Crow and Kimura, 19701. Although some data exist on selective disadvantages, mutation rates are only estimates. Based on mutation rates for other traits in organisms such as Drosophila (Dobzhansky et al., 1977), these rates may vary from 10-4 to 10-8 or may be as low as 10-~6 if resistance requires a change at two or more sites in the gene (Whitten and McKenzie, 1982~. Measuring initial resistance gene frequencies directly is difficult. The phen- otype of a resistance gene and an efficient means to detect it can be known only when resistance develops in the field. By that time most populations have been exposed to the pesticide. One alternative, laboratory selection, often produces artifacts such as polygenic resistance (Whitten and McKenzie, 1982; Roush, 19831. Laboratory-susceptible strains collected before pesticide use commonly suffer population bottlenecks (LaChance, 1979) that distort rare allele frequencies (Ned et al., 19751. Despite these difficulties initial resistance allele frequencies could and should be measured. Some resistance management strategies depend on allele frequency. For example, high pesticide doses may delay resistance, but only if allele frequency is very low and other conditions are met (Tabashnik and Croft, 19821. Such frequencies could be measured in field populations by screening for resistance before using a new insecticide at a dose that kills more than 99 percent of susceptible individuals. Survivors would have to be held for testing for major resistance alleles. A more efficient approach would be to develop a sophisticated detection test (e.g., electrophoretic) for a cos- mopolitan resistant pest (e.g., Musca domestica L., Tetranychus urticae
POP UD4TION GENETICS AND ECOLOGICAL STUDIES 259 Koch, Myzus persicae [Sulz.], and Heliothis armigera [Hubner]) in one country and to take the test to another country where the pesticide has never been used. With international cooperation it would then be possible to take advantage of differing pesticide-use patterns to estimate initial allele fre- quencies. DOMINANCE Dominance refers to the resemblance of heterozygotes (usually Fit off- spring) to one of their parents. If heterozygotes (RS) more closely resemble the toxicological phenotypes of the resistant homozygous (RR) parents, re- sistance is dominant. If the heterozygotes show little or no resemblance, resistance is recessive. For many genetic traits, particularly visible mutants, dominance and fitness can be defined independently. For example, "stubby wing" of the house fly can be defined as recessive to the wild type by morphology, even though there may be recessive effects on reproductive fitness. In the field, however, dominance for survival of pesticides may also mean higher relative fitness compared to the susceptible genotypes. In the field, dominance of the toxicological phenotype may depend on dose (Curtis et al., 19781. A dose that would kill RS heterozygotes but not resistant (RR) homozygotes means that the heterozygotes resemble the sus- ceptible homozygotes (SS), and resistance is effectively recessive. Con- versely, a dose that would kill susceptible homozygotes but not the heterozygotes makes resistance functionally dominant, since heterozygotes and RR ho- mozygotes are phenotypically similar. This concept of adjusting the dose is often called alteration of dominance, but could be called alteration of relative fitness. The ultimate reduction in relative fitness results from doses so high that even RR genotypes are killed, which is generally not feasible. At least two research groups have reported on toxicological dominance in the field. Interestingly, the results have not always been consistent with laboratory data. Resistance to lindane and cyclodienes, including dieldrin, ordinarily shows clear discrimination between all three genotypes in laboratory assays (Brown, 19671. Therefore, some pesticide doses in the field should kill all susceptibles and heterozygotes but not all resistant homozygotes. This occurs in anophe- line mosquitoes (Rawlings et al., 19811: SS, RS, and RR adults marked with fluorescent dusts were released into lindane-sprayed village huts in Pakistan. The higher treatments killed all three genotypes at first, but eventually al- lowed some RR individuals to survive as residues decayed. Thus, resistance was rendered effectively recessive. Similarly, McKenzie and Whitten (1982) implanted eggs of RR, RS, and SS sheep blow flies (Lucilia cuprina [Wiedemann]) into artificial wounds in dieldrin-treated sheep. Larvae were later collected from the wounds, reared
260 POPULATION BIOLOGY OF PESTICIDE RESISTANCE to adulthood, and tested with a discriminatory dose to determine genotype. The RS individuals had a relative viability intermediate between the RR and SS larvae, even as the dose decayed. Although higher doses might have made resistance recessive, these results differed from those of Rawlings et al. (1981) despite a similar form of resistance (Whitten et al., 1980~. McKenzie and Whitten (1982) also studied relative Liabilities of diazinon- resistant genotypes. Diazinon resistance was incompletely dominant in lab- oratory assays of sheep blow fly larvae (Arnold and Whitten, 1976~. There- fore, the RR and RS genotypes should have similar relative viabilities in the field, that is, resistance should be dominant. To the contrary the RS genotypes actually showed relative Liabilities very similar to the SS genotypes (i.e., resistance was effectively recessive under field conditions), even as the dia- zinon residues decayed to allow considerable survival of the SS homozygotes. The reason for the contrasting results for dieldrin and diazinon is unclear, but dominance in the field and in the laboratory should not be assumed to be similar. Dominance is important not only in relation to pesticide pressure but also in the absence of pesticide pressure. The important phenotype in this case is relative fitness, which is even more difficult to measure than toxicological dominance in the field. The phenotypic dominance of fitness is most easily discussed in the context of relative fitnesses in untreated habitats. RELATIVE FITNESSES Untreated Habitats Resistant genotypes must be at a reproductive disadvantage in the absence of pesticides. If not, resistance alleles would be more common prior to selection (Crow, 1957~. Small selection intensities, however, can maintain very low allele frequencies over evolutionary time (Crow and Kimura, 1970~. For resistance management the selective differences between resistant and susceptible genotypes must be accurately quantified. Resistant and susceptible strains of arthropods often are reported to differ in developmental time, fecundity, and fertility. Mating competitiveness might also differ, but of the reports found on this, neither detected differences (Gilotra, 1965; Roush and Hoy, 19811. Table 1 compares R and S strains from some commonly cited studies where reproductive factors were well quantified and where the R strains could be classified as field- or laboratory- selected. In a field-selected strain resistance was diagnosed or suspected before the strain was brought into the laboratory. A laboratory strain was produced by selection from an initially susceptible colony. Whenever possible all data relevant to fecundity (i.e., egg and larval survival) or developmental time (egg, larval, pupal, or mean generation time) were combined. (For
261 - · Ct En V) _ ED 3 o a_ Ct ca _ Cal o Cal as o o Cal ca au .e c at w o ~ ~ .O ~ En .O Cal Cal ~ Cal .O CO c o V) on Go ~ Ch ~ Cal ~ ~ C 3 <^ ~ _ - ~ ~ ~ ~ ~ ~ DOW ~ <,, wow 5 can C (IJ LOCI ~ o2 ~ ~ m m ~ ~ m m _4 c o :^ 5 A \0 ~ c ~ ~ ~· r ~ ~ "1 o~ m 3 ~ ~ c C ~c ~ m ~ m ~ ~ st .~ CQ Z ~ ~ ~ ~ ~ ~ 1 ~ ~ 1 ~ ~ ~ ~ ~ ~ ~ ~ ~ 00 0 c, oo ~ oo oo ~ ~···· ·· 4~. ····- ~00 0 ~ ~. ~0 000 0 .cq ~ ~ u'~ ~z z~ 0 oo 0 oo cr. ~c, ~ ~o `,.. .... I~.. . . . ~4 ~0 ~000 000 0 0 0 :^ - w ~c ( ~ ~ c °c~c ~ _ 9 9 5 ~5 9 E 9 ~5 a E ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ . o. V .~ O ~ w D w c - #o . ~ o ~ ;^ au C t4 ~ ~ fL) t4 · - L~ o - ~ o o c ~ ~ ~ c~ c~ - o c) c~ c) v' . . ~: o c~
262 ' POPULATION BIOLOGY OF PESTICIDE RESISTANCE simplicity throughout this paper the SS genotype will have a relative fitness of 1.00 compared to the RS and RR genotypes.) Two conclusions are apparent from Table 1. First, reproductive disad- vantages are not always associated with resistance. Second, laboratory-se- lected strains suffer more reproductive disadvantages than resistant strains colonized from the field. Even two of the field strains that showed disad- vantages had been selected for 5 to 10 generations in the laboratory. The differences between the laboratory and field strains are consistent with the conclusions of Whitten and McKenzie (1982) and Roush (1983) that labo- ratory and field selection often produce different kinds of resistance. Although the genetic basis of resistance in most of these strains is unknown, resistance in the laboratory-selected DDT-resistant Blatella germanica was polygenic (Cochran et al., 1952~. Studies of the type shown in Table 1 are interesting, but they cannot provide accurate data for resistance management. Strains often differ in fitness, independent of resistance (Babers et al., 1953; Bogglid and Keiding, 1958; Perkins and Grayson, 1961; Birch et al., 1963; Varzandeh et al., 1954; Roush and Plapp, 19821. Even when RR and SS genotypes differ, the more important question is whether there are differences between RS and SS genotypes. During the early stages of selection for resistance and the later stages of resistance reversion, most R alleles in large, randomly mating populations will be carried by heterozygotes. Assuming that selection is not intense, the genotypic frequencies are likely to approximate Hardy-Weinberg proportions ~2:2pq:q24. Thus, at resistance allele frequencies of 20 percent, for example, 32 percent of the population will carry RS, and only 4 percent will carry RR. Clearly resistance management will be best served by com- parisons of RR, RS, and SS genotypes in similar genetic backgrounds. Methods There are two basic methods available for making genotype comparisons. One is to analyze fecundity and developmental-time differences for all three genotypes (Ferrari and Georghiou, 1981; Roush and Plapp, 1982~. The other is to monitor changes in genotypic or phenotypic frequencies in untreated populations where the resistance alleles are initially at some intermediate frequency (often 50 percent). These experiments can be conducted and ana- lyzed by iteratively fitting curves for fitness estimates to the observed data (White and White, 19811. Although not always conducted in cages, the studies will be referred to as "population cage" studies because of their clear analogies to the cage studies long conducted by Drosophila geneticists. The resistance population-cage data available only as LDsoS or resistance ratios are not included here, because such data give only a qualitative appraisal of genotypic fitnesses.
POPUC4TION GENETICS AND ECOLOGICAL STUDIES 263 Although both approaches have advantages and disadvantages, the population-cage approach is probably better for most purposes. Fecundity and developmental-time estimates can be measured accurately and fairly quickly. Although these are the only fitness components reported to differ between resistant and susceptible strains, other aspects of fitness could also differ. A population-cage experiment increases the prospect that such dif- ferences will be detected. Another problem for component studies is data analysis. Both cage- and fitness-component studies have generally been conducted on discrete rather than overlapping generations, which is somewhat unrealistic for the field but creates a dilemma in the analysis of fitness-component data. In discrete generations a strain that produces half as many offspring may be only half as fit. For continuous generations population growth rates are more important, as represented by the intrinsic rate of increase, r (Ferrari and Georghiou, 1981). Population growth rates can be more affected by small developmental- time differences than by similar differences in fecundity, as seen in the expression for intrinsic growth rate, r = loge Ro/T, where Ro is the net replacement rate (number of daughters per female) and T = mean generation time (Roush and Plapp, 19821. A 50 percent reduction in fecundity (Ro) may affect r by much less than 50 percent if Ro is large and mean generation time remains unchanged. For example, if Ro = 100 for susceptible females and Ro = 50 for resistant females in the laboratory, the difference in r is only 15 percent. On the other hand realistic values of Ro in the field may be only about 5 (Georghiou and Taylor, 1977), where a 50 percent reduction in Ro (5 to 2.5) means a 43 percent reduction in r. Thus, quantifying fitness with r or similar terms (Roush and Plapp, 1982) depends on an implicit assumption about Ro. For logistical reasons population cages must be maintained at a relatively constant density, so Ro is about 1, which is probably closer to field conditions than if Ro is around 50. In addition cages can be maintained in continuous generations, if appropriate to the species. A third advantage of the population-cage approach is that all three gen- otypes can be compared against a homogenized genetic background. Crossing unrelated R and S strains often results in heterotic Fit heterozygotes, giving biased or ambiguous estimates of fitness specific to the RS genotypes (Roush and Plapp, 1982~. The easiest way to establish a population cage in an unbiased way is with Fit heterozygotes. When fitness differences have been implicated by population cages, the fitness-component approach may be useful for identifying the factors that differ. Fitness estimates should be obtained in the field whenever possible. It is rare, however, that one can monitor populations known to be isolated from R or S immigration and where an allele has been raised to moderately high frequency by pesticide pressure that has ceased. It is generally more feasible
264 POPULATION BIOLOGY OF PESTICIDE RESISTANCE to maintain population-cage experiments in the laboratory. Curtis et al. (1978) compared estimates of fitness from both field and laboratory data and obtained similar results. Therefore, laboratory results may be realistic if the laboratory conditions simulate the field as much as possible. The most realistic studies of this kind may be conducted on species whose behavior and ecology are not too disrupted by laboratory or greenhouse settings, including Musca domestica, Tetranychus urticae, Blatella germanica, and Tribolium casta- neum. It may be particularly important to conduct these studies under different temperatures. Data In a seminal study Curtis et al. (1978) estimated relative fitnesses from field data on changes in frequencies of resistant and susceptible phenotypes of two species of Anopheles mosquitoes during several generations after treatment was discontinued. Although there were some uncertainties about the estimates (Curtis et al., 1978; Wood and Bishop, 1981; Roush and Plapp, 1982), the DDT- and dieldrin-resistant phenotypes in An. culicifacies had relative fitnesses of about 0.44 to 0.97. One important assumption was that susceptibles were not immigrating into the sites, thus causing fitnesses to be underestimated. Some immigration is likely for An. culicifacies, but immi- gration is less likely for An. stephensi (Wood and Bishop, 19811. In this species DDT-resistant phenotypes had estimated fitnesses of 0.91 from field data and 0.96 from a field-selected population held in the laboratory. Muggleton (1983) used methods similar to those of Curtis et al. (1978) in a laboratory study of the Finesses of malathion-resistant phenotypes of the stored products pest Oryzaephilus surinamensis. Relative fitnesses were about 0.63 to 0.76 compared with the S phenotypes when the populations were held at 25°C, but the R phenotypes may have had an advantage at temperatures over 30°C. Only a few studies report data on the fitnesses of RS heterozygotes. In all of these, the fitness disadvantages suffered by the heterozygotes were not more than half of those for resistant homozygotes. In two studies the het- erozygotes suffered no reproductive disadvantage (White and White, 1981; Roush and Plapp, 1982), that is, the reproductive effects of resistance were recessive. Three studies used a fitness-copnonent approach. Ferrari and Georghiou (1981) studied intrinsic growth rate, r, in RR, RS, and SS genotypes of Culex quinquefasciatus. The RR strain had an r of 0.79, but F~ heterozygotes had an r of 0.95. Emeka-Ejiofor et al. (1983) compared the developmental times of dieldrin-resistant, DDT-resistant, and susceptible strains, and F~ crosses of An. gambiae. The differences were small in all comparisons. Roush and Plapp (1982) found that diazinon-resistant (RR) house flies had about
POPULATION GENETICS AND ECOLOGICAL STUDIES 265 57 to 89 percent of the reproductive potential of an SS strain, but RS flies had 100 percent of that potential. White and White (1981) reported on a population-cage study of diazinon resistance in sheep blow flies. The frequency of resistance phenotypes de- clined quickly from an initial frequency of about 90 percent, then slowed dramatically (White and White, 1981), as is typical for selection against a recessive allele (Crow and Kimura, 19701. Approximately 10 percent of the population was still resistant at generation 38 (White and White, 19811. The fitness estimates for generations 13 to 38 were 0.61 for RR and 1.0 for RS and SS. Coadaptation The above studies were conducted on long-established R strains and may underestimate the fitness disadvantages suffered by RR and RS genotypes during the early stages of a resistance episode if "resistance coadaptation" is common. According to this theory the fitnesses of resistant genotypes are improved by "coadaptive" modifying genes that change the genetic back- ground (Whitehead et al., 19851. Coadaptation of fitness and resistance may, however, be rare (Roush, 19831. The only reliable approach to evaluating whether coadaptation has occurred in a strain is to use repeated backcrossing to a susceptible strain (Crow, 1957) to isolate the major resistance gene in a susceptible genetic background. Perhaps the first researcher to use repeated backcrossing and to report on fitness was Helle (1965~. The Leverkusen-S strain of Tetranychus urticae was selected for more than 30 generations to produce an R strain. This strain was inferior to the S strain in fitness, and resistance reverted after relaxing selection. Contrary to what would be expected if coadaptation was occurring, fitness of the R strain was improved, not worsened, by repeated backcrossing. More recently a backcrossing study on sheep blow fly has demonstrated that resistance coadaptation can occur. McKenzie et al. (1982) found that diazinon resistance was not deleterious in population cages established from Fat and BC3 RS flies, but was significantly deleterious in cages established from BC6 and BC' RS flies. The decline in the frequency of the R allele in the BC' cages can be approximated by fitnesses of 0.5 for RR and 0.75 for RS. The major resistance modifiers were on a different chromosome than the major resistance locus (McKenzie and Purvis, 1984~. In contrast fitness coadaptation was not found in diazinon resistant house flies collected in Mississippi (Whitehead et al., 19851. Even after six gen- erations of backcrossing to a laboratory-susceptible strain, there were no significant differences in developmental time or fecundity. There are major differences, however, between house flies in Mississippi and sheep blow flies in Australia. Fitness modifiers can only be at an advantage when in the
266 POPULATION BIOLOGY OF PESTICIDE RESISTANCE presence of the resistance allele. Thus, selection for fitness modification must be fairly weak until the resistance allele reaches high frequency. Charlesworth (1979) gives a similar argument. The frequency of the diazinon-resistance allele appears to be about 0.27 in Mississippi house flies (Whitehead et al., 19851. The resistance allele in the sheep blow fly was maintained in very high frequency by continuous diazinon use against the insect for more than 10 years, which is rather unusual (McKenzie et al., 19821. Thus, it is rea- sonable that modification occurred in the sheep blow fly but not in the house fly. In sum, fitness modification has been observed in only one of three cases. More such studies are needed. The available data show that fitnesses of RR range from 0.5 to 1.0; fitnesses of RS range from 0.75 to 1.0. At least in laboratory studies, organophosphorous (OP) insecticide-resistant genotypes generally seem to suffer larger reproductive disadvantages than DDT- or cyclodiene-resistant genotypes, consistent with a suggestion by Zilbermints (19751. Treated Habitats How do fitnesses in treated habitats compare with those in untreated hab- itats? Data on increases in frequencies of DDT- and dieldrin-resistant phen- otypes of Anopheles sup. in the field show that resistant cenotv~es mav have -a 1~ r -A err - -I ---I -~~~ ~~~~~ ~~~~~~~~~~ =~~~~~~ rip ~ fitnesses of 1.3 to 6.1 (Curtis et al., 1978; Wood and Cook, 19831. Such fitnesses are a complex function of genotypic mortality (which depends on treatment intensity) and reproductive potential, refugia, and immigration (Georghiou and Taylor, 19771. In some circumstances the overall fitnesses of R phenotypes are probably much higher than 6.1. ECOLOGICAL STUDIES Although selection for resistance can proceed very quickly in closed pop- ulations where each individual is exposed, such intense treatment is uncom- mon in resistance episodes. Usually, some portion of the controllable individuals escapes significant exposure in protected or overlooked spots or "refugia" within the treated area. Also, some individuals, usually adults, will disperse into the treated areas from outside after pesticide residues have decayed. Both concepts are interrelated and emphasize the maintenance of susceptible individuals in the population. Refuges The importance of refugia is clear in models (Georghiou and Taylor, 1977) and can be readily noted in field experience. In spider mites, for example, resistance generally appears first in greenhouses, where all host plants are
POPUL4TION GENETICS AND ECOLOGICAL STUDIES 267 likely to be thoroughly treated, and later in orchard and field crops, where treatment is less intense or complete (Dittrich, 1975~. Few estimates have been made of the portion of populations that ordinarily escape treatment. Such data could be gathered from mark-recapture or population sampling data. For example, population sampling data show that about 20 percent of Heliothis larvae in cotton fields escape lethal exposure (Wolfenbarger et al., 1984~. The portion of 12 apple pests escaping in refugia ranges from 0.2 percent (apple maggot, Rhagoletis pomonella) to 17 percent (San Jose scale, Quad~raspidliotus perniciosus), depending on species (Tabashnik and Croft, 1985~. From practical considerations 20 percent may be an upper limit for the portion in refugia. Failure to obtain at least 80 percent control from insecticide or acaricide applications is probably unsatisfactory for almost any pest and would lead to changes in treatment practices until higher levels of control were achieved. immigration A recent experimental laboratory study on house flies has demonstrated the importance of both susceptible immigration and the influence of pesticide persistence on such immigration (Taylor et al., 1983; Uyenoyama, Via, this volume). Yet immigration is difficult to quantify in terms that relate to resistance development. Rates of immigration for a species depend not only on distances to the source of the untreated population and its size but also on weather and the quality and distribution of host plant species (Stinner, 1979; Follett et al., 1985; Whalon and Croft, 19861. A better understanding of dispersal is a key component of many emerging pest-management tactics, but resistance management has some rather special needs. It is not enough to conduct mark-recapture studies on adults. Knowing where the individuals mate and oviposit is also necessary for understanding the impact they have on the susceptibility of a population. Genetic markers, including pesticide resistance and allozymes, may be particularly useful in such studies. Based on a survey of orchard entomologists, ratios of migrants to the resident population among 12 apple pests range from 0.1 to 10-5, depending on species (Tabashnik and Croft, 19851. As was true for initial R allele frequencies, and in contrast to factors like refugia, current estimates of im- migration rates vary over several orders of magnitude. This emphasizes not only the need to tailor resistance management programs to individual species but also the need to improve estimates of immigration. RESEARCH NEEDS Most resistance models are based on fairly reasonable genetic assumptions (Tabashnik, this volume). Most resistance seems to be associated with single
268 POPULATION BIOLOGY OF PESTICIDE RESISTANCE locus changes. Fitness disadvantages clearly occur, although they may be "slight" to "moderate" rather than "severe," as defined in some modeling studies (Georghiou and Taylor, 1977; Tabashnik and Croft, 19821. None- theless, more studies must be conducted on the fitnesses of resistant geno- types, with emphasis on coadaptation, to determine if the studies reviewed here are representative across a range of species. More important, however, better estimates must be obtained for R allele frequencies in untreated pop- ulations, since current estimates vary over several orders of magnitude. Although migration and refugia are important, they are poorly understood compared with their potential impact. The quantification of immigration, in particular, requires continued improvement in understanding the basic ecol- ogy of pest species. Presumably, such understanding will also allow better control of these species without pesticides and will.further deter resistance development, which is at the heart of modern pest management. ACKNOWLEDGMENTS We thank J. C. Schneider, M. J. Whitten, and B. E. Tabashnik for dis- cussion. Paper approved as No. 5985 by Director, Mississippi Agricultural and Forestry Experiment Station. REFERENCES Arnold, J. T. A., and M. J. Whitten. 1976. The genetic basis for organophosphorous resistance in the Australian sheep blowfly, Lucilia cuprina (Wiedemann) (Diptera: Calliphoridae). Bull. En- tomol. Res. 66:561-568. Babers, F. H., J. J. Pratt, Jr., and M. Williams. 1953. Some biological variations between strains of resistant and susceptible house flies. J. Econ. Entomol. 46:914-915. Bhatia, S. K., and S. Pradhan. 1968. Studies on resistance to insecticides in Tribolium castaneum Herbst 1. Selection of a strain resistant to p, p' DDT and its biological characteristics. Indian J. Entomol. 30:13-32. Bielarski, R. V., J. S. Roussel, and D. F. Clower. 1957. Biological studies of boll weevils differing in susceptibility to chlorinated hydrocarbon insecticides. J. Econ. Entomol. 50:481-482. Birch, L. C., T. Dobzhansky, P. O. Elliott, and R. C. Lewontin. 1963. Relative fitness of geographic races of Drosophila serata. Evolution 17:72-83. Bogglid, O., and J. Keiding. 1958. Competition in house fly larvae. Oikos 9:1-25. grower, J. H. 1974. Radio-sensitivity of an insecticide-resistant strain of Tribolium castaneum (Herbs") . J. Stored Prod. Res . 10: 129- 131. Brown, A. W. A. 1967. Genetics of insecticide resistance in insect vectors. Pp. 505-552 in Genetics of Insect Vectors of Disease, J. W. Wright and R. Pal, eds. New York: Elsevier. Charlesworth, B. 1979. Evidence against Fisher's theory of dominance. Nature (London) 278:848- 849. Cochran, D. G., J. M. Grayson, and M. Levitan. 1952. Chromosomal and cytoplasmic factors in transmission of DDT resistance in the German cockroach. J. Econ. Entomol. 45:997-1001. Crow, J. F. 1957. Genetics of insect resistance to chemicals. Annul Rev. Entomol. 2:227-246. Crow, J. F., and M. Kimura. 1970. An Introduction to Population Genetics Theory. New York: Harper and Row.
POPULATION GENETICS AND ECOLOGICAL STUDIES 269 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. Dittrich, V. 1975. Acaricide resistance in mites. Z. Angew. Entomol. 78:28-45. Dobzhansky, T. 1937. Genetics and the Origin of Species. New York: Columbia University Press. Dobzhansky, T., F. J. Ayala, G. L. Stebbins, and J. W. Valentine. 1977. Evolution. San Franciseo, Calif.: Freeman. Emeka-Ejiofor, S. A. I., C. F. Curtis, and G. Davidson. 1983. Tests for effects of insecticide resistance genes in Anopheles gambiae on fitness in the absence of insecticides. Entomol. Exp. Appl. 34:163-168. Ferrari, J. A., and G. P. Georghiou. 1981. Effects of insecticidal selection and treatment on reproductive potential of resistant, susceptible, and heterozygous strains of the southern house mosquito. J. Eeon. Entomol. 74:323-327. Follett, P. A., B. A. Croft, and P. H. Westigard. 1985. Regional resistance to insecticides in Psylla pyricola from pear orchards in Oregon. Can. Entomol. 117:565-573. Georghiou, G. P., and C. E. Taylor. 1977. Genetic and biological influences in the evolution of insecticide resistance. J. Econ. Entomol. 70:319-323. Gilotra, S. K. 1965. Reproductive potentials of dieldrin-resistant and susceptible populations of AnophelesalbimanusWiedemann. Am. J. Trop. Med. Hyg. 14:165-169. Grayson, J. M. 1953. Effects on German cockroaches of twelve generations of selection for survival to treatments with DOT and benzene hexachloride. J. Econ. Entomol. 45:124-127. Grayson, J. M. 1954. Differences between a resistant and a nonresistant strain of the German cockroach. J. Econ. Entomol. 47:253-256. Helle, W. 1965. Resistance in the Acarina: Mites. Pp. 71-93 in Recent Advances in Acarology, Vol. II, J. D. Rodriguez, ed. New York: Academic Press. LaChance, L. E. 1979. Genetic strategies affecting the success and economy of the sterile insect release methods. Pp. 8-18 in Genetics in Relation to Insect Management, M. A. Hoy and J. J. McKelvey, Jr., eds. New York: The Rockefeller Foundation. Liu, M-Y., Y-J. Tzeng, and C-N. Sun. 1981. Diamondback moth resistance to several synthetic pyrethroids. J. Eeon. Entomol. 74:393-396. March, R. B., and L. L. Lewallen. 1950. A comparison of DDT-resistant and nonresistant house flies. J. Econ. Entomol. 43:721-722. McKenzie, J. A., and A. Purvis. 1984. Chromsomal localization of fitness modifiers of diazinon resistance genotypes of Lucilia cuprina. Heredity 53:625-634. McKenzie, J. A., and M. J. Whitten. 1982. Selection for insecticide resistance in the Australian sheep blowfly, Lucilia cuprina. Experientia 38:84-85. McKenzie, J. A., M. J. Whitten, and M. A. Adena. 1982. The effect of genetic background on the fitness of diazinon resistance genotypes of the Australian sheep blowfly, Lucilia cuprina. Heredity 49:1-9. Muggleton, J. 1983. Relative fitness of malathion-resistant phenotypes of Oryzaephilus surinamensis L. (Coleoptera: Silvanidae). J. Appl. Ecol. 20:245-254. Nei, M., T. Maruyama, and R. Chakraborty. 1975. The bottleneck effect and genetic variability in populations. Evolution 29:1-10. Perkins, B. D., Jr., and J. M. Grayson. 1961. Some biological comparisons of resistant and nonresistant strains of the German cockroach, Blattella germanica. J. Econ. Entomol. 54:747- 750. Pimentel, D., J. E. Dewey, and H. H. Sehwardt. 1951. An increase in the duration of the life cycle of DDT-resistant strains of the house fly. J. Econ. Entomol. 44:477-481. Plapp, F. W., Jr. 1976. Biochemical genetics of insecticide resistance. Annul Rev. Entomol. 21: 179- 197. Rawlings, P., G. Davidson, R. K. Sakai, H. R. Rathor, K. M. Aslamkhan, and C. F. Curtis. 1981.
270 POPULATION BIOLOGY OF PESTICIDE RESISTANCE Field measurement of the effective dominance of an insecticide resistance in anopheline mos- quitoes. Bull. W.H.O. 49:631-640. Roush, R. T. 1983. A populational perspective on the evolution of resistance. Speech presented at Nat. Meet. Entomol. Soc. Am. Detroit, Mich., November 27-December 1, 1983. Roush, R. T., and M. A. Hoy. 1981. Laboratory, glasshouse and field studies of artificially selected carbaryl resistance in Metaseiulus occidentalis. J. Econ. Entomol. 74:142-147. Roush, R. T., and F. W. Plapp, Jr. 1982. Effects of insecticide resistance on biotic potential of the house fly. (Diptera: Muscidae). J. Econ. Entomol. 75:708-713. Shaw, D. D., and C. J. Lloyd. 1969. Selection for lindane resistance in Dermestes maculates de Geer (Coleoptera: Dermestidae). J. Stored Prod. Res. 5:69-72. Stinner, R. E. 1979. Biological monitoring essentials in studying wide area moth movement. Pp. 199-208 in Movement of Highly Mobile Insects, R. L. Rabb and G. G. Kennedy, eds. Raleigh: North Carolina State University. 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., 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. Thomas, J. G., and J. R. Brazzel. 1961. A comparative study of certain biological phenomena of a resistant and a susceptible strain of the boll weevil, Anthonomus grandis. J. Econ. Entomol. 54:417- 420. Varzandeh, M., W. N. Bruce, and G. C. Decker. 1954. Resistance to insecticides as a factor influencing the biotic potential of the house fly. J. Econ. Entomol. 47:129-134. Whalon, M. E., and B. A. Croft. 1986. Dispersal of apple pests and natural enemies in Michigan. Michigan State University, Agricultural Experiment Station: Research Report No. 467. White, R. J., and R. M. White. 1981. Some numerical methods for the study of genetic changes. Pp. 295-342 in Genetic Consequences of Man Made Change, J. A. Bishop and L. M. Cook, eds. New York: Academic Press. Whitehead, J. R., R. T. Roush, and B. R. Norment. 1985. Resistance stability and coadaptation in diazinon-resistant house flies (Diptera: Muscidae). J. Econ. Entomol. 78:25-29. Whitten, M. J., and J. A. McKenzie. 1982. The genetic basis for pesticide resistance. Pp. 1-16 in Proc. 3rd Australas. Conf. Grassl. Invert. Ecol., K. E. Lee, ed. Adelaide, Australia: S.A. Government Printer. Whitten, M. J., J. M. Dearn, and J. A. McKenzie. 1980. Field studies on insecticide resistance in the Australian sheep blowfly, Lucilia cuprina. Aust. J. Biol. Sci. 33:725-735. Wolfenbarger, D. A., J. A. Harding, and S. H. Robinson. 1984. Tobacco budworm (Lepidoptera: Noctuidae): Variations in response to methyl parathion and permethrin in the subtropics. J. Econ. Entomol. 77:701-705. Wood, R. J., and J. A. Bishop. 1981. Insecticide resistance: Populations and evolution. Pp. 97-127 in Genetic Consequences of Man Made Change, J. A. Bishop and L. M. Cook, eds. New York: Academic Press. Wood, R. J., and L. M. Cook. 1983. A note on estimating selection pressures on insecticide resistance genes. Bull. W.H.O. 61:129-134. Zilbermints, I. V. 1975. Genetic change in the development and loss of resistance to pesticides. Pp. 85-91 in Proc. 8th Int. Congr. Plant Prot., Vol 2.