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. Prediction or Resistance Risk Assessment JOHANNES KEIDING Resistance risk, or the potentialfor development of Weld resistance to pesticides, depends on genetic and biologicalfactors characteristic of the pest species and the local population and of operational fac- tors, that is, the way pest control is carried out and the history of pesticide use. Thus, for resistance risk assessment (RRAJ these fac- tors must be considered and investigated. As an example the RRA in house py populations on Danish farms from 1948 to 1983 is discussed. Farmers, pesticide producers, and scientists closely co- operated in this work. As a result many new insecticides and types of applications have been rejected owing to high resistance risk, while others have been recommended. Reference is made to RRA for insecticides and acaricides in selected national and international programs to control important veterinary and agricultural pests. For RRA in insecticides the following general points are discussed: (1) the use of laboratory versus field selection, (2 J geographical differences, and (3) the fitness of resistant genotypes and phenotypes. RRA for fungicides, herbicides, rodenticides, and veterinary nematicides is discussed briefly. The paper concludes with lists of elements of RRA and research needs and discussions of the organization, interpre- tation, and use of RRA. INTRODUCTION Before a new pesticide is introduced for wide-scale field use, it is important to estimate the potential for significant "field resistance" (Davies, 1984) in the pests for which it is intended. Resistance risk assessment (RRA) concerns 279
280 DETECTION, MONITORING, AND RISK ASSESSMENT TABLE 1 Genetic, Biological, and Operational Factors Influencing Resistance Risk General Factor Specific Factors Existence of genetic resistance characters (A-genes, it-alleles) Frequency of occurrence of resistance characters Number of genes needed to cause resistance Interaction of genes Dominance of genes Penetrance of genes Past selection by other chemicals Fitness of the R-geno- and phenotypes in the presence or absence of the insecticide Genetic Biological Operational Reproduction (generations, offspring, etc.) Climatic and other ecological conditions Behavior Isolation, migration, and refugia History of insecticide applications Persistence of insecticide Method of insecticide application (frequency, coverage, life stage(s) exposed, residual effect, etc.) SOURCE: Modified from Georghiou and Taylor (1977). the present occurrence of resistance and its potential development, including the rate and extent of development. An RRA should refer to a specific pest species, geographical area, ecological situation, history of pesticide use, and type of formulation/application. Estimating the potential for developing re- sistance to a pesticide can be very difficult, yet an assessment can make the introduction and use of new pesticides more intelligent and thus avoid big problems. In this paper I will discuss (1) how to estimate resistance risk: methods, factors, conditions, difficulties, and research needs; (2) how to organize and coordinate the investigations; and (3) how to interpret and use the results. As I am most familiar with resistance to insecticides, I shall start by discussing RRA for chemical control of insects, ticks, and mites and then deal with special problems concerning other pesticides, fungicides, herbi- cides, rodenticides, and compounds to control parasitic nematodes. INSECTICIDE AND ACARICIDE RESISTANCE Resistance risk depends on genetic, biological, and operational factors, and these must be included in any resistance risk assessment. As shown in Table 1 a resistance risk cannot be assigned to a given insecticide or a given pest species it must relate to the local pest population, with its character- istics and conditions, and the way the insecticide is applied. (For a more
PREDICTION OR RESISTANCE RISK ASSESSMENT 281 detailed discussion of factors influencing the development of resistance, see Georghiou and Taylor, 1977 and Georghiou, this volume.) HOUSE FLY RESISTANCE The Danish Experience As an example of how to estimate resistance risk in practice, I shall briefly describe the work of the Danish Pest Infestation Laboratory (DAPIL) on house fly resistance to insecticides in Denmark and elsewhere (Keiding, 1977~. In Denmark the house fly, Musca clomestica, is primarily a pest on farms with pigs and calves, and in recent years poultry. Chemical fly control is carried out in animal houses using residual sprays, space sprays, and spot treatments with impregnated strips or bait paints or with larvicides. Devel- opment of resistance has been favored by (1) the organized and extensive use of insecticides and (2) the relatively low migration of flies between the farms. Since 1945 DAPIL~ has received good cooperation from the farmers associations and many farms, the pesticide industry, and the research labo- rator~es overseas doing basic research on insecticide resistance in our and other house fly strains. The cooperation with the farmers gave DAPIL the essential current information on the effect of various insecticides, formula- tions, and applications that enabled us to follow the development of resistance and to detect and study early cases. Such cooperation is also necessary for the organization of field trials. The use of insecticides for fly control and the development of resistance from 1945 to 1983 are shown in Figure 1. The main elements we found to be important in conducting our BRA were as follows: ;' Surveillance of Resistance Occurrence . · Obtain information, complaints, inquiries, and so forth, from farmers pest control operators, and others · Determine resistance by standard methods in the laboratory and the field · Conduct systematic surveys to determine the distribution and level of various types of resistance in the state Research on and Surveillance of Cross-resistance and Type of Resistance · Conduct cross-resistance tests · Determine resistance mechanisms and their diagnoses (e.g., by use of a synergist) 'DAPIL combines an advisory service, evaluation of new insectides, formulations and applications and research and development on pest control, biology, and resistance.
282 [I ~ ~ k ~ ·o 21~- ~ C E C [___ ~ C ~ ~ L 1[ ~: ~E ~ ~ ~ 1~- r -T ~ ~ E ~ t i T L a ~ E 4 ~ ~ if i Ed ~ a., ~ 0 <, a ~ 0 0 C ~ c a ~ :rn ~ 1: t T: D: Fir ~ ~:E Cat 4_ O ~S O O O Cat Cat Cat a, ~a' . ~C) ~ . ~a ·4_ at ~ ~ ca cat ~ca 04 ~ o a, Cat Cat At cn SO 0, Cat C) C) in · - o Hi Hi 3 a 3 ~ V] o ._' . ~
PREDICTION OR RESISTANCE RISK ASSESSMENT 283 · Determine the genetics of the resistance (genes involved, dominance, fitness of genotypes) · Survey resistance types and frequency of phenotypes · Establish a collection of strains representing the important resistant types and their combinations Studies on the Dynamics of Resistance Development: Operational and Eco- logical Factors · Conduct studies under field conditions, rather than laboratory experi ments · Follow development of resistance through small-scale field trials · Monitor, over several years, the development of resistance and cross- resistance to widespread use of insecticides, formulations, application, effect of alternating treatments, and the like · Collect information on the time of development and the stability of resistance · Study the basic population dynamics and behavior of the pest under field conditions and under different ecological conditions The cooperation between the pesticide industry and DAPIL on the house fly problem has played an important role in the possibility of assessing the resistance risk of new compounds and of using this assessment to: (1) conduct cross-resistance tests using a suitable range of our collection of resistant strains; (2) monitor field populations for resistance to the new compound; (3) conduct small-scale field trials with the new compound, possibly in two or more formulations/applications, to see if resistance may develop rapidly; (4) use the information from (1), (2), and (3) to decide whether and how to introduce the new compound for fly control and how to use it; (5) follow the development of resistance to the new compound when it is widely used and adjust the yearly recommendations for fly control accordingly; and (6) make available to industry our general and specific knowledge of the resis- tance situation and the factors involved, for example, by annual reports. The cooperation with scientists in other countries resulted in much useful, timely information on mechanisms and genetics of resistance (Keiding, 1977; Saw- icki and Keiding, 1981), which could be used for our RRA. First, DAPIL used the RRA to explain to, convince, or persuade companies that certain insecticides or applications with high resistance and cross-resis- tance risks should not be introduced, or that it might be advantageous to make available an insecticide application with a low resistance risk. For example, in 1948 DAPIL found that high DDT resistance extended to avail- able DDT analogues these were not introduced. In the mid-19SOs DAPIL persuaded industry not to sell any organochlorine insecticides for fly control. Owing to rapid development of resistance in small-scale field trials, the following insecticides were not introduced for fly control on Danish farms:
284 DETECTION, MONITORING, AND RISK ASSESSMENT the organophosphorus compounds coumaphos (1955), coumithoate (1957), formothion (1965), phosmet (1968), tetrachlorvinphos (1969), and azame- thiphos residual spray (1981); the carbamate mobam (1967). In other cases DAPIL found that high resistance to new compounds was already present, due to cross-resistance. This happened for most OF com- pounds and carbamates in the 1970s, when dimethoate had been commonly used for five to seven years and high resistance had become widespread. Researchers in England (Sawicki, 1974, 1975; Sawicki and Keiding, 1981) studied the resistance (R) mechanisms, their genetics, and interaction of this resistance and showed that insensitive target (cholinesterase) and several detoxification processes were involved. This research explained the cross- resistance and demonstrated the importance of the sequence of use of insec- ticides (Sawicki, 1975; Keiding, 1977; Sawicki and Keiding, 1981~. The most striking example of RRA was that of pyrethroid resistance. Investigations from 1970 to 1973 showed that house flies on Danish farms had a common potential for developing high resistance to pyrethroids when the selection pressure with pyrethroids was strong, for example, by frequent use of pyrethroid aerosols (Keiding, 1976~. If aerosols were used less fre- quently, however, once a week or less, allowing some unexposed flies to reproduce, the resistance might remain low and the aerosols would remain effective. Knowing that treatments with residual sprays give a strong selection pressure, DAPIL advised the companies and the authorities not to introduce residual sprays with pyrethroids for fly control on farms. The advice was followed, even though there was no proper legal basis for banning residual pyrethroids for fly control until 1980.2 In the meantime DAPIL received further support for this decision. In 1977 and 1978 DAPIL found that heterogeneous resistance to candidate residual pyrethroids was widespread on Danish farms, and the resistance factor kdr, which causes resistance to DDT and pyrethroids (in connection with other factors), occurred in practically all fly populations investigated (Keiding, 1978, 1979, 1980; Keiding and Skovmand, 19X41. The predicted rapid development of general pyrethroid resistance when residual pyrethroids were used was confirmed in Switzerland (Keiding, 1980), in Germany (Skovmand and Keiding, 1980; Kunast, 1979), and England (Chapman and Lloyd, 1981~. In Denmark we continue to avoid the residual pyrethroids for fly control. The aerosols with pyrethrum, and the like, are still effective, and pyrethroid resistance is low or moderate. 2The Danish "Act on Chemical Compounds and Products," passed in 1980, empowers the Danish Ministry of Environment to require, before registration, experimental data on cross-resistance and the potential for developing resistance. If the data indicate that resistance will quickly make the product ineffective and/or its use will result in resistance to useful products, the registration may be refused (Sawicki, 1981). Registrations also may be withdrawn if general development of resistance is found after a period of use.
PREDICTION OR RESISTANCE RISK ASSESSMENT 285 The experience with fly control illustrates another principle, already men- tioned in this volume (see section on Genetics, Biochemical, and Physio- logical Mechanisms of Resistance to Pesticides), that development of resistance depends on the type of formulation and application used, owing to difference in selection pressure. Thus, for the flies, bait applications promote less resistance than residual sprays, and knock-down sprays less than residual pyrethroids. Several results from DAPIL's lengthy studies have been positive, resulting in recommendations of formulations. Three OF compounds are registered as bait formulations, but not as residual sprays, because of the resistance risk. OF compounds were effective on flies resistant to organochIorines in the early 1950s, and various compounds and applications were recommended (Figure 1~. Fenthion, and especially dimethoate, were effective and were recommended when other OPs failed. Tests with a variety of resistant fly strains, including high multiresistance, showed susceptibility to the devel- opment inhibitors diflubenzuron and cyromazine, used as larvicides, without significant resistance development after selection pressure (Keiding and E1- Khodary, 19831. In addition the extensive data collected on the development of resistance in house fly populations on farms since 194X are being put into a data base, which should provide greater possibilities for analyzing resistance risks under various conditions (Keiding et al., 19831. Resistance in Other Regions Sequential development of resistance in field populations of house flies also has been studied in Czechoslovakia (Rupee et al., 1983), California (Georghiou and Hawley, 1971), and Japan (Yasutomi and Shudo, 1978) and has been used as a guide for choosing new insecticides. In addition house fly samples from many parts of the world have been tested for resistance by Keiding, Hayashi, Kano, and others (Keiding, 1977; Taylor 1982~. These surveys have provided information on the global occurrence of various types of resistance and the resistance risks. An important finding was that DOT resistance occurs everywhere, but only in some areas in northern Europe is the kdr factor for DOT resistance common (Keiding, 1977; Keiding and Skovmand, 1984~. Since kdr is also an important factor for pyrethroid re- sistance, the risk for development of high pyrethroid resistance is still lower in all the areas where kdr is rare or absent. China recently surveyed for resistance in more than 400 field samples of house flies. As no sign of pyrethroid-R or the kdr factor was found, China recommended the use of residual pyrethroids for fly control (Gao lin-ya, Institute of Zoology, Acad. Sinika, Beijing, personal communication, 1983~. In Japan where kdr is rare, high pyrethroid resistance was not found until after six years of fly control with a residual pyrethroid (Motoyama, 1984~;
4 286 DETECTION, MONITORING, AND RISK ASSESSMENT in areas of Europe where kdr is common, high pyrethroid resistance developed in a few months. Examples from Other Insect and Mite Species National Programs The following are examples of some of the many systematic, long-term programs on development, types, and risk of resis- tance. The development of multiple resistance in the cattle tick Boophilus microplus in Australia has been investigated for about 30 years by the Com- monwealth Scientific and Industrial Research Organization (CSIRO) tick laboratory in Queensland (Nolan and Roulston, 1979; Roulston et al., 1981~. This work includes all the factors of importance to RRA: (1) close cooperation with farmers to obtain early detection of resistance, (2) investigation of resistance mechanisms and their genetics to define resistant types and cross- resistance spectra, (3) surveys of the distribution of resistant types, (4) studies and modeling of the dynamics of resistance development, (5) cooperation with industry to test new acaricides against tick strains representing the resistant strains, (6) field trials with promising acaricides and types of ap- plication, and (7) advice to farmers on control methods. Investigations of resistance in the sheep blow fly, Lucilia cuprina, in Australia, begun about 25 years ago, contain the same elements as mentioned for the cattle tick, including surveys of resistance gene frequency and fitness in field populations (Hughes, 1981, 1982, 1983; Hughes and Devonshire, 1982; McKenzie et al., 1980; Whitten and McKenzie, 19821. Among agricultural pests are the following examples. Comprehensive investigations were begun more than 20 years ago on leaf- and planthoppers attacking rice in Japan. These include extensive resistance surveys, studies of resistance mechanisms and genetics, and trials of many new insecti- cides, especially the effect of using mixtures or alternating treatments (Saito and Miyata, 1982; Hama, 1975, 19801. Surveys, resistance mech- anisms, and genetic research have been conducted on the aphids Myzus persicae in Britain (Sawicki et al., 1978) and Phorodon humuli in Czecho- slovakia (Hrdy, 1975, 1979; Sula et al., 1981~. National RRA programs have been conducted in Egypt on cotton pests, especially the leafworm, Spodoptera; in Australia on spider mites (Dittrich, 1979) and Heliothis ssp. (Davies, 19841; and in the United States on spider mites and Heliothis (Sparks, 1981; Bull, 1981~. Spider mites in several countries also have been investigated (Dittrich, 19754. International Programs The World Health Organization (WHO) has or- ganized global programs for detecting and monitoring resistance in vectors and pests of medical importance, especially vector mosquitoes; WHO also
PREDICTION OR RESISTANCE RISK ASSESSMENT 287 has supported many studies on resistance genetics and resistance types oc- curring in vectors (WHO, 1980), as well as trials on the dynamics of resis- tance development (Curtis, 19811. Moreover, WHO has organized a Pesticide Evaluation Scheme including tests of new insecticides from industry with some resistant strains of mosquitoes, flies, and the like. The United Nations' Food and Agriculture Organization (FAG) has organized a global survey of pesticide susceptibility of stored-grain pests (Champ and Dyte, 1976), in- cluding some typing of OP resistance. Laboratory Versus Field Selection Experience and theoretical considerations have shown that the predictive value of investigating resistance risk through laboratory selection is limited. If resistance develops when an insect population is exposed to selection pressure with an insecticide through a number of generations, the ability of resistance exists, but the level, type, and rate at which it develops may be quite different from what happens under field conditions. If a laboratory selection is negative and no resistance develops, one may conclude very little. There is no guarantee that resistance will not develop in the field (Pal and Brown, 19711. For example, in the 1950s, laboratory strains of house flies were selected at the Riverside Laboratory in California for 19 to 149 generations with various OF compounds; only a slow and moderate increase of tolerance was obtained, compared to what later developed in the field (Pal and Brown, 19711. There may be several reasons for the differences between laboratory and field selection: for example, ( 1) because of the smaller gene pool in laboratory selection, rare resistance genes and ancillary genes may be missing; (2) a difference in insecticide pressure often results in lower mortality in the lab- oratory than in the field; laboratory selection may exploit polygenic variation while field selection tends to act on alleles of single resistance genes (Whitten and McKenzie, 1982~; (3) a difference in the fitness of resistance genotypes; and (4) a difference in natural selection. Therefore, if laboratory selection is used for RRA: (1) the gene pool should be as big as possible and should be taken from natural populations initially; and (2) insecticide pressure, ecological conditions, and natural selection should simulate that occurring in the field. Small-scale control trials on isolated or semi-isolated field pop- ulations with monitoring of resistance often may be better than laboratory trials, provided such field selection is feasible, for example, on farms with house flies in Denmark, pests in greenhouses (Helle and van de Vrie, 1974), isolated fields, and so forth. If small-scale field trials on resistance are not feasible, the first practical applications must be monitored for resistance development. This activity should be organized in collaboration with farmers, state institutions, research laboratories, and industry, and the results should
288 DETECTION, MONITORING, AND RISK ASSESSMENT be made available to all interested parties so that the first experiences can be used for RRA in other areas. Geographical Differences The biological and operational factors influencing the development of resistance in a pest species may vary greatly depending on climate, farming practice, use of insecticides, and the like, and the resistance risk will vary accordingly. The genetic factors also may differ, not only in frequency of resistance genes but also which genes and mechanisms cause resistance lo- cally, as has been found for DOT and pyrethroid resistance in house flies. These possible regional and local differences must be considered for any RRA in a given area and for the use of resistant strains to test for cross- resistance of new compounds. Fitness of Resistant Geno- and Phenotypes The relative fitness of the resistant gene- and phenotypes under field conditions may be difficult to estimate, but the stability or reversion of resistance in the field when the insecticide pressure is relaxed is important. Estimating fitness under laboratory conditions has a limited value (Keiding, 1967), not only because conditions differ from the field but because strains with different periods of adaptation to laboratory conditions may be com- pared. Relatively little is known about the importance of the fitness factor for insecticide resistance. Fitness of resistant types, however, is not constant. With time the resistance genome may be integrated with fitness factors by natural selection, a process called coadaptation (Keiding, 1967~. Mathematical Models A number of simulation models (Taylor, 1983; Section III in this pro- ceedings) have contributed significantly to our general understanding of re- sistance dynamics and are being used for developing strategies to reduce the development of resistance. Their usefulness, however, depends on whether the assumptions and the parameters are realistic. For example, in RRA we need information about factors such as local frequency and number of re- sistance genes, fitness factors, selection pressure, population dynamics, and migration. Such information must be gathered in the field, and assumptions must be tested in the field where possible (Davies, 1984; Denholm, 19811. RESISTANCE OF PLANT PATHOGENS Resistance risk assessment in fungicides has been well discussed in several recent reviews(Dekker, 1981, 1982a,b;Wade, 1982;Staub and Sossi, 19831.
PREDICTION OR RESISTANCE RISK ASSESSMENT TABLE 2 Specific and Systemic Action of Some Fungicides to Emergence of Fungicide Resistance 289 Occurrence of Fungicide or Mode of Action Resistant Strains Fungicide Group Specific Systemic In Vitro On Plants Risksa for Failure of Disease Control Copper compounds - - - - very low Dithiocarbamates - - - - very low Chlorothalonil - - - - very low Phthalimides - - - - very low Organic Hg compounds - - + + low Aromatic hydrocarbons + _b + + high see Butylamine + - + + high Dicarboximides + _ b + + moderate to high Dodine + - + + moderate Organic tin compounds + - + + moderate Acylalanines + + + + high Benzimidazoles + + + + high Dimethir~mol + + 0C + high Ethir~mol + + 0c ~moderate Organic P compounds + + + + moderate Carboxanilides + + + + moderate to low Fenar~mol, nuar~mol + + + - low Imidazoles + + + - low Morpholines + + + - ~low Tr~azoles + + + - ~low Tr~for~ne + + + - very low + with the property; - without the property. aThe risk for failure of disease control is a rough estimation, since it also depends on other factors (type of disease, strategy of fungicide application, etc.). bChloroneb and procymidone have systemic properties. CConcerns obligate parasites. Occurrence of strains with decreased sensitivity to some of these compounds has been reported. SOURCE: Dekker (1981). As with insecticide resistance the RRA is influenced by inherent genetic and biological factors in the pest fungus, including reproduction rate, spore mo- bility, and host range. Moreover, climate and weather play a role, and the operational factors determining the selection pressure (i.e., area treated, coverage and frequency of treatments, duration of exposure, and persistence of the fungicide) are highly important for the development of field resistance. More specifically than in insecticides, resistance risk in fungicides is con- nected with the biochemical mode of action of the fungicide. The resistance risk, therefore, can be classified according to type of fungicide (Table 2~. Within a certain mode of action, for example, the benzimidazoles, a high
290 DETECTION, MONITORING, AND RISK ASSESSMENT degree of cross-resistance is found. Fungi, unlike insects, produce so many spores that resistant mutants can be detected even at a very low frequency. Therefore, the genetic ability for resistance can be demonstrated easily in the laboratory for most pathogens that can be grown on artificial media. Thus, a standard method for RRA in fungicides is to grow spores of pathogens on a medium containing an amount of fungicide just above the minimum inhibitory concentration in which only resistant cells survive. Using laboratory tests resistant mutants have been found for all the specific- site fungicides (Table 2), but resistance may not be a problem in the field. Fitness of the resistant mutants in fungal pathogens is generally a decisive factor for development of field resistance. Therefore, assessments of fitness are important for RRA. These assessments may be done (1) by determining the relative growth of resistant and wild-type strains in vitro, (2) by testing the pathogenicity of strains on plants in the greenhouse, and (3) by infecting plants with a sensitive and a resistant strain and observing the result of competition in the absence of the fungicide over a number of pathogen generations. As with insects, however, laboratory and greenhouse tests may not realistically estimate fitness under field conditions. Therefore, field trials may be necessary for the full answer (Dekker, 1982a). Although laboratory and greenhouse tests can provide much information on resistance risk, negative results cannot exclude the possibility of resistance developing in the field if the selection pressure in area and time is large enough. Moreover, the rate and extent of development of resistance depends mainly on biological, environmental, and operational field factors, as pre- viously mentioned. Field experiments and monitoring of resistance in path- ogens in areas subjected to various schemes of fungicide treatments are therefore essential for RRA of fungicides as well as of insecticides. Coop- eration and rapid exchange of information between producers and users of fungicides, advisers, and research and regulatory institutes are necessary to cope with the rapidly developing problems of fungicide resistance. The in- ternational association of agrochemical industry associations (GIFAP- Groupement International des Associations Nationals de Fabricants de Pro- duits Agronomiques) recognized this need in 1981 when it formed the Fun- gicide Resistance Action Committee (FRAC). The equivalent for insecticide resistance, the Insecticide Resistance Action Committee (IRAC), was formed in 1984. HERBICIDE RESISTANCE Assessing resistance risk to herbicides is simplified because resistance is confined mainly to the s-triazine herbicides, usually with general cross- resistance to all s-triazines and related degrees of resistance or tolerance to the asymmetrical triazinones, ureas, and many other nitrogen-containing pho
PREDICTION OR RESISTANCE RISK ASSESSMENT 291 tosynthetic inhibitors, but as a rule no cross-resistance or negative cross- resistance to herbicides with other modes of action (LeBaron and Gressel, 19821. A careful monitoring and verification of resistance to s-triazines in the field is important for RRA. As for most other pests the rate of resistance development is influenced by the selection pressure, which is a product of the persistence of the herbicide effect after treatment, the dose, the number of years the herbicide has been used alone in an area, and the proportion of the weed population that is exposed. S-triazines have a very specific action and a high persistence. Resistance problems may be expected in new her- bicides having these characteristics. Using selection experiments for.RRA, however, is difficult because of the time required for sufficient generations to be exposed and because the experiments have to be done in field areas of a sufficient size. (The relation between weed ecology and resistance risk is discussed by Slife in this volume.) The fitness of resistant strains does not seem to be of great importance for the development of herbicide resistance. RODENTICIDE RESISTANCE Rodenticide resistance is mainly a problem with one group of rodenticides, the anticoagulants. For practical reasons it is difficult to investigate resistance risk by meaningful selection experiments in the laboratory or in other confined colonies of rats, mice, and other rodents. The best method for RRA is a systematic monitoring of control failures and rodenticide resistance in con- nection with rodent-control campaigns using a given rodenticide. Good col- laboration is therefore essential between the people organizing, conducting, and supervising the control campaigns and a laboratory that can carry out the standard resistance tests on trapped rodents and that can interpret the results. Thus, it is very important to have as complete information as possible on the history of rodenticide use in the area. When resistance has been found a central laboratory should, if possible, keep a colony of each type of resistant strain for use in toxicity tests with new rodenticides to gather information on cross-resistance. Studies on resistance mechanisms and genetics are also important for RRA, as discussed under insecticide resistance. (For more detailed discussions on rodenticide resistance see papers by MacNicoll, Greaves, and Jackson in this volume.) NEMATODE RESISTANCE Nematicide resistance of parasitic nematodes in domestic animals has been found and investigated mainly in sheep, but it may also occur in goats and horses (Prichard et al., 1980; Bj0rn' 19831. Resistance has developed mostly to the benzimidazole compounds having a general cross-resistance within this group, but no cross-resistance to other types of nematicides. Surveys of
292 DETECTION, MONITORING, AND RISK ASSESSMENT nematode resistance are hampered because critical tests to determine the effect of the compound require that a large number of treated host animals be killed. Indications of resistance, however, may be obtained by fecal egg counts after treatments or may be confirmed by in vitro tests on egg hatch for the benzimidazoles. Laboratory selection is fairly simple; colonies of nematodes are exposed to treated hosts for a number of generations. The conditions for resistance development, however, are different in the field, for example, as to natural selection and selection pressure by the nematicide. In the field a high pro- portion of the nematode population may be unexposed, since it is outside the host (Le Jambre, 1978~. CONCLUSION Elements of RRA The important elements of RRA as discussed above are listed in Table 3 (the succession of the elements are not necessarily chronological nor in order of importance). Any RRA program must establish good coordination, collaboration, and exchange of information between (1) the producers (the agrochemical in- dustry), (2) the advisers and organizers of pesticide use, (3) the users of pesticides, and (4) the research institutes. An RRA program may be organized by an international body, for example, FAO or WHO, or it may be a national or state institution. International collaboration and rapid exchange of infor- mation are essential, however, by informal reports, correspondence, con- ferences, and visits. The traveling pesticide experts from industry may play a special role for rapid inflation dissemination to national institutions that may serve as a link between users, scientists, and industry. In this way resistance problems may be realized early, such that suitable monitoring and research can be organized, for example, supported by industry. Examples of such collaboration are the work on the cattle tick in Australia, the house fly in Denmark, and rice pests in Japan. Other examples and a discussion of the interagency cooperation are given by Davies (1984~. The WHO and the FAO have organized data bases on the occurrence of pesticide resistance (Georghiou and Mellon, 1983~. The results of unpub- lished investigations, including those in industry, would be useful. One means of providing such information about new findings would be a newsletter on pesticide resistance; WHO had one for several years, but it was discontinued in 1976. luterpretation and Use of the Assessments Two types of interpretation can come from these assessments: scientific- technical interpretation and economic interpretation. For example, a scien
PREDICTION OR RESISTANCE RISK ASSESSMENT TABLE 3 Elements of Resistance Risk Assessment 293 A. Consider the pesticide: mode of action, chemistry, and stability. B. Evaluate the pest species: genetic diversity, resistance potential. 1. Conduct laboratory selection experiments.a 2. Conduct field selection experiments.b 3. Survey for the occurrence and development of resistance in field populations of the pest to pesticide use.C . Determine the cross-resistance spectrum.& 5. Determine the resistant type (mechanism, genetics).e 6. Determine the fitness of resistant biotypes f 7. Monitor for local and regional distribution of resistant types." 8. Investigate factors influencing the development of resistance: genetic, biological, and operational.h 9. Develop mathematical models on the dynamics of resistance development.h 10. Conduct computer simulations of resistance development.' 11. Check and improve simulation models by field experiments. 12. Investigate the effect of sequential use of pesticides for resistance development. aThese experiments have a limited predictive value owing to restricted gene pool, difference of conditions, exposure to pesticides, natural selection, and so forth, and in some pests (e.g., weeds and rodents) they are difficult to perform. bThese experiments, especially in isolated or semi-isolated localities, may be more informative, but also more difficult to arrange. The risk of spreading resistant strains is a limitation. CSurveying is very important and should be a regular activity for applications of new pesticides. Information on the history of pesticide use influencing the previous selection of resistance factors is essential (see item 12). If resistance has reverted in a field population, it usually develops quickly when the pesticide is reintroduced. Determine cross-resistance when resistance to a pesticide is detected. Patterns of cross-resistance are often known or should be investigated. eThis activity is important for predicting and understanding cross-resistance, including the com- ponents of resistance and their genetics. It is also important to know whether resistance depends on one or more resistance factors. fFitness of resistant biotypes under field conditions is of general importance for resistance de- velopment, particularly to fungicides. "Such occurrence may vary locally and regionally. hKnowledge of the dynamics of resistance development and of the parameters in the field is essential for constructing realistic models and for predicting the rate and extent of resistance de velopment. 'Computer simulations are important to evaluate the effects of various genetic, biological, and operational factors and to develop strategies for delaying or avoiding resistance. SOURCE: Keiding (unpublished). tific-technical interpretation may estimate the probability of resistance de- veloping in a pest in an area, the rate and extent of resistance, and the factors influencing it, while an economic interpretation would estimate its economic consequences. Use of the assessments are valuable for regulatory authorities and industry in reaching agreement on formulations and recommended ap- plications for the pesticide. If industry is more interested in getting a quick
294 DETECTION, MONITORING, AND RISK ASSESSMENT profit or recouping investments, however, which could lead to applications and recommendations that conflict with the long-term interest of the users and perhaps of the company, the regulatory authorities and their advisers may want to regulate the use of the pesticide to comply with the strategy of pest control recommended and the hazards of rapid development of resistance. Additionally, the assessments may find that resistance found in the lab- oratory may not apply to the field; resistance found in one area may not occur or develop in the same way in another; and certain pesticides may be useful even if some resistance has developed because they are so much cheaper than the substitutes, as is true with DDT for controlling some malaria vectors. Research Needs The research needed to improve the ability to assess resistance risk may be related to the elements of RRA. The following is a brief list of some general research fields for RRA, with reference to the "element numbers" in Table 3. The need and importance of the research may vary between the groups of pests and pesticides. · Develop and improve methods for detecting and monitoring types of resistance, especially at low frequencies (see Brent in this volume) (3,7) · Research resistance mechanisms, cross-resistance (4,5) · Study the genetics of resistance (5, 6, 8) · Determine the fitness of resistant biotypes (6, 8) · Conduct field investigations of the biology, ecology, and population dynamics of the pest (8, 9, 10, 11) · Conduct field investigations on selection pressure by various applica- tions of pesticides and control schemes (8, 9, 10, 11) · Develop and use more realistic models on the dynamics of resistance development (9, 10, 11) · Investigate the effect of sequential use of pesticides for resistance de- velopment (8, 12) REFERENCES Bj0rn, H. 1983. On aspects of anthelmintic resistance of parasitic nematodes in domestic animals. A review. Pp. 1-116 in Report from Institute of Hygiene and Microbiology. Copenhagen: Royal Veterinary and Agriculture University. Bull, D. L. 1981. Factors that influence tobacco budworm resistance to organo-phosphorous insec- ticides. Bull. Entomol. Soc. Am. 27:193-197. Champ, B. R., and C. E. Dyte. 1976. Pp. 1-297 in Pesticide Susceptibility of Stored Grain Pests. Rome: FAO. Chapman, P. A., and C. J. Lloyd. 1981. The spread of resistance among houseflies from farms in the United Kingdom. Pp. 625-631 in Proc. Br. Crop Prot. Conf. Lavenham, Suffolk: Lavenham, 1981.
PREDICTION OR RESISTANCE RISK ASSESSMENT 295 Curtis, C. F. 1981. Possible methods of inhibiting or reversing the evolution of insecticide resistance in mosquitoes. Pestic. Sci. 12:557-564. Davies, R. A. H. 1984. Insecticide resistance: an industry viewpoint. Pp. 593-600 in Proc. Br. Crop Prot. Conf. Lavenham, Suffolk: Lavenham, 1984. Dekker, J. 1981. Impact of fungicide resistance on disease control. Pp. 857-863 in Proc. Br. Crop Prot. Conf. Lavenham, Suffolk: Lavenham, 1981. Dekker, J. 1982a. Can we estimate the fungicide resistance hazard in the field from laboratory and greenhouse tests? Pp. 128-138 in Fungicide Resistance in Crop Protection, J. Dekker and S. G. Georgopoulos, eds. Wageningen, Netherlands: Centre for Agricultural Publishing and Documen- tation. Dekker, J. 1982b. Countermeasures for avoiding fungicide resistance. Pp. 177-186 in Fungicide Resistance in Crop Protection, J. Dekker and S. G. Georgopoulos, eds. Wageningen, Netherlands: Centre for Agricultural Publishing and Documentation. Denholm, I. 1981. Present trends and future needs in modeling for the management of insecticide resistance. Pp. 847-855 in Proc. Br. Crop Prot. Conf., Vol. 3. Lavenham, Suffolk: Lavenham, 1981. Dittrich, V. 1975. Acaracide resistance in mites. Z. Angew. Entomol. 78:28-45. Dittrich, V. 1979. The role of industry in coping with insecticide resistance. Pp. 249-253 in Proc. Symp. 9th Int. Congr. Plant Prot., Vol. 1, T. Kommedahl, ed. Minneapolis, Minn.:-Burgess. Georghiou, G. P., and M. K. Hawley. 1971. Insecticide resistance resulting from sequential selection of house flies in the field by organophosphorus compounds. Bull. W.H.O. 45:43-51. 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. 1976. Pesticide resistance as an evolutionary phenomenon. Pp. 759-785 in Proc. 15th Int. Congr. Entomol., Washington, D.C. College Park, Md.: Ento- mological Society of America. Hama, H. 1975. Resistance to insecticides in the green rice leafhopper. Jpn. Pestic. Inf. 23:9-12. Hama, H. 1980. Mechanism of insecticide resistance in green rice leafhopper and small brown planthopper. Rev. Plant. Prot. Res. (Japan) 13:54-73. Helle, W., and M. van de Vrie. 1974. Problems with spider mites. Outlook Agric. 8:119-125. Hrdy, I. 1975. Insecticide resistance in aphids. Pp. 739-749 in Proc. Br. Insectic. Fungic. Conf. Lavenham, Suffolk: Lavenham, 1975. Hrdy, I. 1979. Insecticide resistance in aphids. Pp. 228-231 in Proc. Symp. 9th Int. Congr. Plant Prot., Vol. 1, Washington, D.C., Minneapolis: Burgess. Hughes, P. B. 1981. Spectrum of cross-resistance to insecticides in field samples of the primary sheep blowfly, Lucilia cuprina. Int. J. Parasitol. II:475-479. Hughes, P. B. 1982. Organophosphorus resistance in the sheep blowfly, Lucilia cuprina (Wiede- mann) (Diptera: Calliphoridae): A genetic study incorporating synergists. Bull. Entomol. Res. 72:573-582. Hughes, P. B. 1983. Biochemical and genetic studies on resistance to organophosphorus insecticides in Lucilia cuprina. Ph.D. dissertation. Macquarie University, Australia. Hughes, P. B., and A. L. Devonshire. 1982. The biochemical basis of resistance to organophosphorus insecticides in the sheep blowfly, Lucilia cuprina. Pestic. Biochem. Physiol. 18:289-297. Keiding, J. 1967. Persistence of resistant populations after the relaxation of the selection pressure. World Rev. Pest Control 6:115-130. Keiding, J. 1976. Development of resistance to pyrethroids in field populations of Danish houseflies. Pestic. Sci. 7:283-291. Keiding, J. 1977. Resistance in the housefly in Denmark and elsewhere. Pp. 261-302 in Pesticide Management and Insecticide Resistance, D. A. Watson and A. W. A. Brown, eds. New York: Academic Press. Keiding, J. 1978. Insecticide resistance in houseflies. Danish Pest Infest. Lab. Annul Rep. 1977:37- 54.
296 DETECTION, MONITORING, AND RISK ASSESSMENT Keiding, J. 1979. Insecticide resistance in houseflies. Danish Pest Infest. Lab. Annul Rep. 1978:43- 55. Keiding, J. 1980. Insecticide resistance in houseflies. Danish Pest Infest. Lab. Annul Rep. 1979:40- 53. Keiding, J., and A. El-Khodary. 1983. Investigations of cross-resistance involving laboratory strains (of houseflies). Danish Pest Infest. Lab. Annul Rep. 1982:58-59. Keiding, J., and O. Skovmand. 1984. The occurrence of the kdr-factor in houseflies and the potential resistance to pyrethroids in Denmark and elsewhere. P. 737 in 17th Int. Congr. Entomol., Ham- burg, 1984. (Abstr.) Keiding, J., O. Skovmand, and H. O. H. Nielsen. 1983. Establishing a data base for data on insecticide resistance in houseflies at DPIL 1948-82. Danish Pest Infest. Lab. Annul Rep. 1982. Kunast, C. 1979. Die Entwicklung der Permethrinresistenz bei der Stubenfliege (Musca domestica) im suddeutschen. Z. angew. Zool. 66:385-390. LeBaron, H. M., and J. Gressel, eds. 1982. Herbicide Resistance in Plants. New York: John Wiley and Sons. LeJambre, L. F. 1978. Anthelmintic resistance in gastrointestinal nematodes of sheep. Pp. 109- 120 in The Epidemiology and Control of Gastrointestinal Parasites of Sheep in Australia, A. D. Donald, W. H. Southcott, and J. K. Dineen, eds. Australia: CSIRO. McKenzie, J. A., J. T. Dearn, and M. J. Whitten. 1980. Genetic basis of resistance to diazinon in Victorian populations of the Australian sheep blowfly, Lucilia cuprina. Aust. J. Biol. Sci. 33:85- 95. Motoyama, N. 1984. Pyrethroid resistance in a Japanese colony of the housefly. J. Pestic. Sci. (Japan) 9:523-526. Nolan, J., and W. J. Roulston. 1979. Acaricide resistance as a factor in the management of Acari of medical and veterinary importance. Pp. 3-13 in Recent Advances in Acarology, Vol. II, J. D. Rodriquez, ed. New York: Academic Press. Pal, R., and A. W. A. Brown. 1971. Insecticide Resistance in Arthropods. Geneva: World Health Organization. Prichard, R. K., C. A. Hall, J. D. Kelly, I. C. A. Martin, and A. D. Donald. 1980. The problem of anthelmintic resistance in nematodes. Aust. Vet. J. 56:239-251. Roulston, W. J., R. H. Wharton, J. Nolan, J. D. Kerr, J. T. Wilson, P. G. Thompson, and M. Sehotz. 1981. A survey for resistance in cattle ticks to acaricides. Aust. Vet. J. 57:362-371. Rupes, V., J. Pinterova, J. Ledvinka, J. Chmeia, J. Placky, M. Homolac, and V. Pospisil. 1983. Insecticide resistance in houseflies, Musca domestica in Czechoslovakia 1976-80. Int. Pest Control 25:10~108. Saito, T., and T. Miyata. 1982. Studies on insecticide resistance in Nephotettix cincticeps. Pp. 377- 382 in Proc. Int. Conf. Plant Prot. in Tropics, K. L. Heong, B. S. Lee, T. M. Lim, C. H. Teoh, and Yusof Ibrahim, eds. Kuala Lumpur: Malaysian Plant Protection Society. Sawicki, R. M. 1974. Genetics of resistance of a dimethoate-selected strain of houseflies to several insecticides and methylenedioxyphenyl synergists. J. Agric. Food Chem. 22:344-349. Sawicki, R. M. 1975. Effect of sequential resistance on pesticide management. Pp. 799-811 in Proc. 8th Br. Insectic. Fungic. Conf., Lavenham, Suffolk: Lavenham, 1975. Sawicki, R. M. 1981. Problems in countering resistance. Philos. Trans. R. Soc. London, Ser. B 295: 143-151. Sawicki, R. M., and J. Keiding. 1981. Factors affecting the sequential acquisition by Danish houseflies (Musca domestica) of resistance to organophosphorus insecticides. Pestic. Sci. 12:587- 591. Sawicki, R. M., A. L. Devonshire, A. D. Rice, G. D. Moores, S. M. Petzing, and A. Cameron. 1978. The detection and distribution of organophosphorus and carbamate insecticide-resistant Myzus persicae (Sulz.) in Britain in 1976. Pest. Sci. 9:189-201. Skovmand, O., and J. Keiding. 1980. Chemical fly control and resistance on farms in North Germany. Danish Pest Infest. Lab. Annul Rep. 1979:39-40.
PREDICTION OR RESISTANCE RISK ASSESSMENT 297 Sparks, T. C. 1981. Development of insecticide resistance in Heliothis zea and Heliothis virescens in North America. Bull. Entomol. Soc. Am. 27:186-192. Staub, T., and D. Sossi. 1983. Recent practical experiences with fungicide resistance. Pp. 591-598 in 10th Int. Congr. Plant. Prot., Vol. 2. Lavenham, Suffolk: Lavenham, 1983. Sula, J., J. Kuldova, and I. Hrdy. 1981. Insecticide-resistance spectrum in the hop aphid (Phorodon humuli) populations from different regions: Notes on resistance mechanisms. IOBC/WPRS Bull. IV. 3:46-54. 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, R. N. 1982. Insecticide resistance in houseflies from the Middle East and North Africa with notes on the use of various bioassay techniques. Pestic. Sci. 13:415-425. Wade, M. 1982. Resistance to fungicides. Span. 25:8-10. 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: S. A. Government Printer. World Health Organization. 1980. Resistance of vectors of disease to pesticides. 5th Rep. WHO Exp. Comm. Vector Biol. Contr. Tech. Rep. Ser. 655. Yasutomi, K., and C. Shudo. 1978. Insecticide resistance in the houseflies of the third Yumenoshima, a new dumping-island of Tokyo (II). Jpn. J. San. Zool. 29:205-208.