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.
Population Biology of Pesticide Resistance: Bridging the Gap Between Theory and Practical Applications WERE THE EVOLUTION OF PESTICIDE RESISTANCE not of grave concern to human health and well-being, it would have still been important as a major example of the power and potential of adaptive evo- lution. Surprisingly, population geneticists and ecologists have paid little attention to it. Similarly, relatively few investigators involved in management of resistance have directly applied the tools and theoretical concepts of ac- ademic population biology. In this chapter we describe current attempts at bridging the gap between academic and applied population biology, discuss aspects of the genetics and population biology of resistance critical to developing resistance man- agement programs, recommend future work needed in this area, and de- scribe major impediments to developing and implementing programs to manage resistance. A HEURISTIC MODEL OF MANAGING RESISTANCE We present here a simplistic, idealized model of the resistance cycle re- sulting from pesticide use, solely for heuristic purposes (as a "thought ex- periment"), not as a realistic model for the long-term management of resistance. The model assumes that resistant genotypes arise in the pest population and, as a result of selection imposed by pesticide use, field control fails because these genotypes attain high frequencies. The model assumes that stopping use of the pesticide will result in a continuous decline in the frequency of resistant genotypes and, in a reasonable amount of time, the frequency of susceptible genotypes will become sufficiently high for the population to be 143
144 POPULATION BIOLOGY OF PESTICIDE RESISTANCE effectively controlled by that pesticide once again (see Figure 11. Neither assumption is a necessary outcome. The time period between initial use of the pesticide and control failure is the resistance onset interval, TR(i). Stopping treatment with pesticide i results in relaxation of selection pressure for resistance to i and a decline in the frequency of resistant genotypes. The time between the end of treatment with i and a decline in the frequency of resistant genotypes low enough to resume effective control with compound i is the susceptibility recovery interval, TS(i). In theory, pest control is possible indefinitely by cycling through an array of compounds, as long as resistance to each of them is independent of resistance to every other. The total number of pesticides required for this cycling depends solely on the lengths of the resistance onset and susceptibility recovery intervals (Figure 11. In this model, the goal of resistance management is to maximize the resistance onset intervals and minimize the susceptibility recovery intervals. The effect of this strategy would be to minimize the number of independent compounds needed for effective long-term control. USE OF POPULATION BIOLOGY THEORY AND CONCEPTS IN RESISTANCE MANAGEMENT To date, population biology theory has contributed to resistance manage- ment primarily in identifying the factors contributing to the rise of resistance, and to some extent in interpreting factors responsible for resistance in specific populations. We are unaware of any pesticide-use programs that have been entirely planned and executed in a manner prescribed from theoretical and empirical considerations of population genetics of resistance and the ecology of the organisms and ecosystem under treatment. Elements of population biology theory have, however, been applied to some aspects of pest management. For example, the theory of the population biology of infectious disease played a role in the development of the suc- cessful multiline cultivar procedure used to reduce fungicide use in barley cultivation (Wolfe and Barrett, this volume). This theory has also been useful in a retrospective manner. Analyses of resistance cycles are generally con- sistent with those anticipated from population biology theory and laboratory experiments (Gutierrez et al., 1976; Comins, 1977b; Taylor et al., 1983; Tabashnik, this volume). Nevertheless, we are unaware of any cases where a high-dose regime or any other tactic has been actually put into practice based solely upon con- siderations of population genetic theory, even though several theoretical investigations are directly relevant. For example, MacDonald (1959) noted that resistance would develop more slowly if it was recessive. Davidson and
POPULATION BIOLOGY OF PESTICIDE RESISTANCE . · . . . . TR3 TS3 j 1 ~TR, it' TSC . . . TR2 TS2 ' .. . TR1 ~TS1 _1 _ C ~_\ C-1 3 · ~ 145 FIGURE 1 The Pesticide Resistance Cycle. Tit(i) is the time period from the first use of a pesticide, i, to the time resistance precludes its use, the resistance onset interval. TS(i) is the time period between termination of use of compound i to the time the frequency of resistant pests is sufficiently low to maintain effective population control with that compound, the susceptibility recovery interval. C is the total number of compounds required for indefinite control. Pollard (1958) found that higher doses of gamma-BHC (lindane) would kill heterozygotes and indicated that this would slow resistance development. More recently, the high-dose approach has been the subject of much theo- retical work (Tabashnik and Croft, 1982~. By and large, management of resistance to pesticides has made little direct use of population and community ecology theory. Earlier recognition by pesticide users of the "Volterra" principle (when predators and their prey are both killed, prey populations will increase) would have highlighted the danger of indiscriminate use of pesticides on populations where some control of prey species (pests) was achieved by natural enemies (predators). GENETIC AND ECOLOGICAL INFORMATION REQUIRED FOR MODELS OF THE POPULATION BIOLOGY OF RESISTANCE Even though specific resistance management programs should be designed on a case-by-case basis, the following general classes of information are required to develop realistic models of the population biology of pesticide resistance, and thus to design resistance management programs:
146 POPULATION BIOLOGY OF PESTICIDE RESISTANCE · Mode of Inheritance. Knowing the mode of inheritance of the resistant phenotype is critical to developing any model of pesticide resistance. Al- though it sounds formidable, a relatively modest amount of genetic infor- mation would actually be needed for models of the population genetics of resistance. Particularly critical for these models is whether resistance is in- herited as a discrete character (involving one or two major genes) or acts as a continuously distributed (quantitative, or polygenic) character, because different classes of theoretical models are applicable to single-gene and po- lygenic resistance (Via, Uyenoyama, this volume). In the former we must know the number of alleles at the resistance-determining loci and the dom- inance relationship among these alleles (as a function of pesticide concen- tration) (Curtis et al., 19781. It would also be valuable to know the nature of the interactions (epistasis) between the genes determining resistance as well as other, modifying loci (Uyenoyama, this volume). Where resistance acts as a quantitative character, it is particularly critical to know the mean levels of resistance, the phenotypic variances, the additive and nonadditive genetic variance in these levels of resistance, and the genetic covariance in the tolerances to different pesticides (Via, this volume). We recognize that these cannot be known until resistance has evolved, but some generalities on inheritance of resistance are emerging (Chapter 2~. · Fitness Relationships. Estimating genotypic fitness is difficult, even in a well-controlled experiment. Nevertheless, at least rough estimates of the relative reproductive and survival rates of resistant and susceptible genotypes are necessary to consider their rates of increase, frequencies after pesticide treatment, and rate of decline when treatment is stopped (i.e., when selection is relaxed). It is not sufficient to assume that fitness is simply a matter of the kill rate or that resistant genotypes will have a selective disadvantage in the absence of pesticides. These fitness estimates have to be obtained for resistant and susceptible genotypes as functions of stage in the life cycle and concentrations of pesticides. Fitness should not be assumed to be a constant. In obtaining these estimates, it is necessary to control for a variety of other environmental and genetic factors such as temperature, season, physiological state, population density, and genetic background. Again, this information is not available until after resistance has evolved. In the case of insects and rodents, behavioral considerations should also be taken into account (Gould, 1984). · Population Structure. Some details of intrinsic genetic structure of the target population and its spatial and temporal distribution are critical to developing a realistic model, especially: (1) whether generations are discrete or overlapping, (2) the nature of the alternation of haploid and diploid phases of the life cycle, (3) the relative lengths of sexual and asexual stages, and (4) the duration of the whole life cycle and its various stages. The lengths of both the resistance onset and susceptibility recovery intervals depend in
POPULATION BIOLOGY OF PESTICIDE RESISTANCE 147 part on how isolated the treated population is. A high rate of migration (gene flow) from susceptible populations would both delay an increase in the pro- portion of resistant genotypes and increase their rate of decline when treatment is stopped. Migrants from resistant populations, rather than independent mutants, could be the primary reason for the spread of resistance. To de- termine this physical component of population structure, the nature and timing of migration, as well as its absolute rate, should be considered. When con- sidering gene flow, the frequency of resistant genotypes in the untreated population may also be important if that reservoir population is relatively small. In studying migration, an attempt should be made to estimate the genetically effective component, not just movement (Comins 1977b; Roush and Croft, May and Dobson, this volume). · Population Regulation. Pest population growth is not necessarily ex- ponential and unregulated in the absence of treatment. Interspecific and intraspecific competition, predation, and parasitism may help limit the rate of growth and densities of pest populations. The nature and importance of the population-regulating mechanisms have to be known and considered in the population biology of resistance. The Volterra principle suggests that pesticide use could exacerbate situations where the pest population is nor- mally limited by parasites or predators that are susceptible to the controlling pesticide. The intensity of selection for and against resistant genotypes could be greatly affected by the nature of the trade-off between density-dependent and density-independent mortality and morbidity factors. Where there is substantial intraspecific competition, sublethal doses of a pesticide could have a strong selection effect by weakening the competitive abilities of susceptible individuals, even when it does not control the density of the population (McKenzie et al., 19821. · Refuges. Reservoirs of susceptible genotypes within the treated area could result from pesticide dose variation in space or time. As is the case for weed seeds, these refuges could be quite substantial and play a sig- nificant role in augmenting the resistance onset interval (Gressel and Segel, 1978). GENERAL AND SPECIFIC MODELS It is possible to construct general models of the population biology of resistance with few possibly no data from natural sources. Models of this type have been used to identify the factors contributing to the rise of resistance and evaluate their relative importance (Coming, 1977a; Taylor, 1983; May and Dobson, this volume). These general models may be the only ones that can be constructed when little population biology information is available, and they can have considerable value. Finally, these models can be used to distinguish between the factors that are really important and those that play
148 POPULATION BIOLOGY OF PESTICIDE RESISTANCE minor roles in the rise of resistance, thus playing a critical role in deciding which empirical studies should be conducted. Where extensive information is available, more detailed applied models can be constructed and analyzed with analytical and computer simulation procedures (Tabashnik, this volume). Although more specific models can provide more quantitatively accurate predictions than general models, we see no justification to postpone developing resistance management programs based on models until all the data are available. SOURCE OF DATA FOR MODEL CONSTRUCTION AND EVALUATION · The Roles of Laboratory and Field Studies. Studies with pesticide- resistant mutants generated in the laboratory and fitness experiments per- formed with laboratory-selected strains may provide some information about the nature of the alleles conferring resistance and their anticipated fate in populations. Whenever possible, however, these investigations should use susceptible and resistant genotypes isolated from natural sources and perform fitness studies under natural conditions. The genetics of resistance in natural populations are probably different from those generated in the laboratory, because, for example, selection pressure under natural conditions might be different from that in short-term laboratory studies (McKenzie et al., 1982; Uyenoyama, this volume). Laboratory studies indicate that fitness differences are likely to exist in natural situations but do not provide accurate estimates of fitness differentials in the field. On the other hand, laboratory studies could provide reliable estimates of toxicological dominance, if they were performed under conditions that approximate field exposure to pesticides. · Extrapolating from Existing Genetic Information and Molecular Pro- cedures. To a great extent the high rate of progress in academic genetics can be attributed to the common use of relatively few species (model systems) that are particularly convenient to study. Unfortunately, real pest organisms are seldom ideal experimental organisms, so genetic information often has to be acquired by extrapolation from related organisms. Using DNA and RNA probes to determine the physical location of genes and to ascertain whether homologous genes are responsible for the same phenotype in different species considerably broadens the range of organisms amenable to genetic analysis. Only limited use has been made of in vitro genetic procedures to investigate the genetics of pesticide resistance (see Georgopoulos, Gressel, Hardy, Hammock and Soderlund, MacNicoll, Plapp, Chapter 2, this volume). Obtaining DNA and RNA probes is not easy when the gene product is not known or known and present in low quantities, or when the physical location on the gene of the model organisms is not known, but molecular techniques should be considered for determining modes of inheritance for population studies of pesticide resistance.
POPULATION BIOLOGY OF PESTICIDE RESISTANCE 149 It is both convenient and traditional to focus on phenotypes that are (or seem to be) discrete characters determined by one of two genes, but it is critical to consider that specific cases of resistance may be determined by multiple alleles and that resistance behaves as a quantitative character. There are well-developed procedures to analyze inheritance of quantitative char- acters and model the behavior of these characters under selection (Via, this volume). EVALUATING MODELS AND PROGRAMS FOR PESTICIDE RESISTANCE MANAGEMENT While we may believe that existing studies of the fit between theory and empirical observation justify the use of population biology theory to develop pesticide use and resistance management programs, a final demonstration of their utility remains necessary. In order to demonstrate the utility of math- ematical and numerical modeling, the programs developed using them must: (1) maintain the required level of pest control, (2) be economically com- petitive, (3) yield lower levels of resistance than would be anticipated for alternative programs employing the same compounders), and (4) be safe from both an environmental and health perspective. While not sufficient in a formal sense, the a posterior) fit between obser- vation and prediction should certainly be considered partial demonstration of the validity of models. Properly controlled pilot studies could provide further evidence, if they were run under field conditions using a few "model" systems with properties similar to those of the intended target species and communities. In cases where the pesticide is already in use, field data could serve as control. These studies should make the evaluation in the minimum time possible, and some acceleration could be achieved by using procedures to detect resistant organisms when they are rare and possibly heterozygous (for one- or two-gene resistance), or when resistance levels are low (for polygenic resistance). The models and data will be quantitative, but fit will have to be evaluated somewhat qualitatively. The extraordinary number of interactions between the biotic and physical factors in a field study cannot all be controlled. On the other hand, if a program is effective, one would anticipate the desired level of pest control and significantly lower rates of increase of resistant genotypes in the experimental populations. FOLKLORE, DOGMA, AND AD HOC PRACTICES There are a number of current pesticide use practices and assumptions about their consequences for resistance management that seem to have little or no base in population biology theory.
150 POPULATION BIOLOGY OF PESTICIDE RESISTANCE · Return to Pesticide Susceptibility. While only occasionally stated ex- plicitly, there seems to be a general belief that a decline in the frequency of resistant genotypes will necessarily follow when use of a compound is stopped. While we expect this to be true in the long run, the length of the susceptibility recovery interval may be effectively indefinite in many cases. In the absence of pesticide use, the selective differential between susceptible and resistant genotypes may be quite small. Even if the original resistant genotypes had a marked disadvantage in the absence of the pesticide, there may be selection for modifier genes that improve the fitness of resistant genotypes. The limited empirical results on the fate of alleles conferring resistance following termination of pesticide use support a mixed view of the fate of resistance genotypes in the absence of pesticide selection. In some cases, the frequencies of alleles conferring resistance declined relatively rapidly (Greaves et al., 1977; Partridge, 1979; McKenzie et al., 1982; also see Greaves, Georgopoulos, this volume). In other cases, there was little change in the frequency of these alleles following the relaxation of selection (White- head et al., 1985; Georgopoulos, Roush and Croft, this volume). Even in cases where the resistant genotypes have a clear selective dis- advantage relative to sensitive genotypes, the intervals for susceptibility re- covery will still be substantially longer than for the corresponding resistance onset. The intensity of selection favoring resistance during pesticide use will certainly be much greater than that favoring susceptibility following the termination of treatment. For a pesticide to be biologically effective for a period as long as that during its first use, the frequency of resistant genotypes in the recovered population would have to be similar to that prior to first use (see May and Dobson, this volume). This conclusion has a number of immediate implications. First, the sim- plistic scheme depicted in our heuristic model is unlikely to be a realistic long-term solution to the problem of pesticide resistance. The recovery period following the rise of resistance could be extremely long and, for practical purposes, too long for individual pesticides to be used more than once. Thus, long-term control by pesticides alone would require an almost infinite supply of independent compounds. In a short-term view, the factors affecting ev- olutionary rates also illustrate the utility of (1) terminating pesticide use before the frequency of resistance is high; (2) developing procedures that increase the selection pressures favoring susceptible genotypes; and (3) programs that increase rates of gene flow from sensitive populations. · Pesticide Mixing and Cycling. A current controversy is whether pesti- cides should be in rotations or mixtures before their target pesos) become resistant. The answer is equivocal. Models can be constructed in which pesticide cycling or mixing either increases or decreases the resistance onset interval. The outcome depends critically on the way the different pesticides interact in determining the fitness of resistant and sensitive genotypes. Also
POPULATION BIOLOGY OF PESTICIDE RESISTANCE 151 important are: modes of inheritance of resistance; frequency of mutations for resistance; rates of recombination between the loci involved; and population dynamics of pest growth, refuges, migration and pesticide action. These qualifications emphasize the need for considering tactics on a case by case basis with validation prior to implementation. The population biology of each type of pesticide use regime can be readily modeled, and the relative merits and liabilities of these pesticide use regimes can be . . ~ assessed a prlorl. · Directed Evolution of Resistance. A fundamental premise of evolution- ary theory is that mutations occur at random; their incidence and nature are independent of specific selection pressure. Starting with the classical fluc- tuation test experiment of Luria and Delbruck (1943), there have been a number of lines of evidence in support of this interpretation (Crow, 19571. There have been suggestions, nevertheless, that pesticides will promote the generation of resistant organisms (as well as select for increase in their frequency) or that resistance to one compound will increase the rate of mutation to a second compound (Wallace and MacSwiney, 19761. While it may be easy to discount these (or any) neolamarckian interpre- tations, we believe that the hypothesis that the rate and nature of mutation is influenced by selection for that mutation is interesting from both an aca- demic and applied perspective and certainly worth testing. We can speculate on mechanisms that make mutations appear to be directed. In nonlethal doses, pesticides could cause "genomic shocks" that increase frequencies of trans- position of chromosome pieces. If pesticide resistance is the result of inserting movable elements of chromosomes, then conceivably the initial transposition could increase the future rates of transposition. In cases where resistance to specific pesticides requires two mutations, one in a gene that is common to resistances to different compounds and one that is unique to each, mutation could appear to be directed. IMPEDIMENTS Implicit in this discussion is the assumption that the pesticide resistance problem is amenable to a technical solution. There is some justification for this assumption; for specific agricultural or clinical situations, programs using combinations of chemical and biological agents could be developed to prolong the useful life of compounds. On the other hand, we see little justification in maintaining the polite fiction that pesticide resistance is solely a technical problem and therefore solvable with the right tools. The design, execution, monitoring, and evaluation of pesticide-use programs and their ultimate im- plementation are major endeavors, even for single agricultural or clinical situations. Development and testing require cooperation of investigators in a variety of fields: chemistry, genetics, population biology, toxicology, bot
152 POPULATION BIOLOGY OF PESTICIDE RESISTANCE any, microbiology, zoology, epidemiology, and medicine. These activities have to be coordinated with people actually running and monitoring the program in the field or clinic. Pesticide-producing companies, primary users of these compounds (growers, physicians, veterinarians, and public health personnel) and government agencies regulating their use will have to partic- ipate. · The Dilemma of Interdisciplinary Programs. Pesticide-use programs are interdisciplinary, yet universities, research institutes, and funding orga- nizations responsible for their development and support are rigidly structured along traditional, disciplinary lines. In universities in the United States, academic and applied biology departments are almost always separate, both geographically and administratively, and have been maintained that way for 50 years or more. Most evolutionary geneticists, ecologists, and population biologists are in academic departments while biologists directly involved in pesticide use and management are in agricultural, clinical, and other more applied departments. Academic and applied biologists primarily publish in different journals and receive funding from different sources. As a result, there is relatively little intellectual intercourse between investigators in these two types of biology departments and often considerable xenophobia. While there are many situations where these administrative and geographic barriers have been breached (e.g., a number of papers in the bibliographies of the population biology papers in this volume and cited here), these are rare exceptions. More extensive breakdown of the traditional separation between applied and academic biology would be a major step toward the solution to the pesticide resistance problem as well as other biological-technical prob- lems. We see no easy general solution to this problem. While lip service is frequently given to the value of interdisciplinary programs, their active de- velopment has been limited at best, and this situation is likely to persist as long as universities, research institutes, and funding agencies are adminis- tratively partitioned into academic and applied areas. As long as these separate administrative units have primary control over personal rewards (salary, promotion, tenure), and as long as the kudos (invitations, travel, awards, and other recognition) are generated along disciplinary lines, from a purely careerist perspective, there is little positive incentive for individuals to engage in interdisciplinary projects; in some cases, there is pressure to avoid doing so. Funding may well be the greatest impediment to jointly applied and theoretical research. As long as research is funded either explicitly or im- plicitly (via the peer review system) along disciplinary lines, interdisciplinary projects will be at a disadvantage. In the long run academic and applied biology could be somewhat unified, despite existing administrative barricades, with a more ecumenical approach
POPULATION BIOLOGY OF PESTICIDE RESISTANCE 153 to teaching. The genetics and population biology of pesticide resistance are certainly interesting applied problems that merit investigation even from the perspective of the most basic biology. Many other applied examples could replace more traditional model systems or natural populations used as ex- amples in genetics and population biology courses. RECOMMENDATIONS RECOMMENDATION 1. Pesticide use practices based on considerations of the population biology of pesticide resistance should be developed and im- plemented. Although the theory and observations of academic population biology have been used to explain past resistance episodes, at this juncture there have not been significant pesticide use programs developed and implemented from con- siderations of the principles of population biology. RECOMMENDATION 2. General models of the population biology of resis- tance can be used to develop pesticide-use practices, as long as the basic premises of these models can be empirically justified. While it may be a long-term goal to develop precise analogs of specific pesticide-use situations, population biology theory may be applied to develop pesticide-use regimes before specific models are developed. The fact that general population biology theory has been successful in a retrospective manner, by providing mechanistic explanations for past resistance episodes, justifies the use of this theory in a prospective manner. RECOMMENDATION 3. While general models may have broad utility, it remains necessary to gather the genetic and ecological information needed to construct specific models. In cases where general models prove inadequate, it will be necessary to employ specific and precise analogs of the populations and pesticides under considera- tion. RECOMMENDATION 4. The continuous monitoring of resistance frequencies should be an integral part of all programs to manage resistance. If the models are realistic analogs of the effects of the pesticide use regime on the genetic structure of the target population, there should be a good corre- spondence between the observed and predicted resistance frequencies and changes in those frequencies. RECOMMENDATION 5. Population biology theory should be used to examine current pesticide-use practices and controversies.
154 POPULATION BIOLOGY OF PESTICIDE RESISTANCE There are a variety of ad hoc pesticide-use practices, e.g., alternating and mixing pesticides to extend the useful life of compounds, which may or may not be justifiable. Mathematical models of the population biology of pesticide use represent an efficient way to evaluate these practices in a prospective manner. RECOMMENDATION 6. An extensive effort should be made to encourage both research on pesticide use and resistance by academic biologists and the study of the population biology by applied biologists involved in pesticide use. Pesticide resistance is a long-term problem that will require the coordinated efforts of investigators representing several disciplines that currently suffer from a lack of interdisciplinary communication. While unlikely to be sufficient as a unique solution to the problem of coordinating efforts, some funds specifically earmarked for joint basic and applied research on the population biology of pesticide resistance may help surmount some of the institutional impediments to this type of interdisciplinary activity. RECOMMENDATION 7. A considerable effort should be put into developing pest-control measures that do not rely on the use of chemical pesticides. The continuous control of pest populations by cycling through novel chemical pesticides is unlikely to be a viable long-term strategy. There is no biological or evolutionary justification for the assumptions that (1) pest populations will return to sensitive states relatively quickly following the termination of the use of specific pesticides, or (2) an adequate supply of novel and safe pesticides can be developed and made available continuously to replace compounds that have lost their effectiveness due to resistance. ACKNOWLEDGMENT We would like to thank Ralph V. Evans for his comments on this manu- scrtpt. REFERENCES Comins, H. N. 1977a. The management of pesticide resistance. J. Theor. Biol. 65:399-420. Comins, H. N. 1977b. The development of insecticide resistance in the presence of migration. J. Theor. Biol. 64:177-197. Crow, J. F. 1957. Genetics of insect resistance to chemicals. Annul Rev. Entomol. 2:227-246. Curtis, C. F., L. M. Cook, and R. J. Wood. 1978. Selection for and against insecticide resistance and possible methods for inhibiting the evolution of resistance in mosquitoes. Ecol. Entomol. 3:515-522. Davidson, G. and D. G. Pollard. 1958. Effect of simulated field deposits of gamma-BHC and
POPULATION BIOLOGY OF PESTICIDE RESISTANCE 155 Dieldr~n on susceptible hybrid and resistant strains of Anopheles gambiae Giles. Nature 182:739- 740. Gould, F. 1984. The role of behavior in the evolution of insect adaptation to insecticides and resistant host plants. Bull. Entomol. Soc. Am. 30:34-41. Greaves, J. H., R. Redfern, P. B. Ayers, and J. E. Gill. 1977. Warfare resistance: a balanced polymorphism in the Norway rat. Genet. Res. Camb. 89:295-301. 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. Gutierrez, A. P., U. Regev, and C. G. Summers. 1976. Computer model aids in weevil control. Calif. Agr. Apr~1:8-18. Curia, S. E., and M. Delbruck. 1943. Mutations of bacteria from virus sensitivity to virus resistance. Genetics 28:491-511. MacDonald, G. 1959. The dynamics of resistance to insecticides by anophelines. Revista di Par- assitologia 20:305. McKenzie, J. A., M. J. Whitten, and M. A. Adena. 1982. The effect of genetic background on the. fitness of rlia,.on resistance aenotv~es of the Australian sheen blowfly, Lucilia cuprina. Heredity Lll~ ~110~O Vet Vie a_ J as 19:1-19. Partridge, G. G. 1979. Relative fitness of genotypes in a population of Rattus norvegicus polymorphic for warfare resistance. Heredity 43:239-246. 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. 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., 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. Wallace, M. E., and F. MacSwiney. 1976. A major gene controlling warfare resistance in the house mouse. J. Hyg. Camb. 76:173-181. Whitehead, J. R., R. T. Roush, and B. R. Norment. 1985. Resistance stability and co adaptation in diazinon-resistant house flies (Diptera:Muscidae). J. Econ. Entomol. 78:25-29. WORKSHOP PARTICIPANTS Population Biology of Pesticide Resistance. Bridging the Gap Between Theory and Practical Applications BRUCE R. LEv~N (Leader), University of Massachusetts J. A. BARRETT, Cambridge University E~NoR C. CRUZE, National Research Council ANDREW P. DOBSON, Princeton University FRED Gould, North Carolina State University JOHN H. GREAVES, Ministry of Agriculture, Fisheries and Food, Great Britain DAVID HECKEE, Clemson University ROBERT M. MAY, Princeton University HAROLD T. REYNOLDS, University of California, Riverside RICHARD T. ROUSH, Mississippi State University
156 POPULATION BIOLOGY OF PESTICIDE RESISTANCE BRUCE E. TAsAsHN~K, University of Hawaii MARCY UYENOYAMA, Duke University SARA VIA, University of Iowa MAX J. Whirred, Commonwealth Scientific and Industrial Research Organization M. S. Wolfe, Plant Breeding Institute, Cambridge