i LIBRARY 3 1293 10140 Coat/3m “3:53:33” This is to certify that the thesis entitled COMPARATIVE RESISTANCE To AZINPHOSMETHYL IN THE PREDATORY MITE AMBLYSEIUS FALLACIS CARMEN (ACARINA: RIF—Mo EIID )W—D ITS PREY TETRANYCHUS URTICAE KOCH (ACARINA: TETW WUSE EXPERIMENTS Joseph Grant Morse has been accepted towards fulfillment of the requirements for M.S. Entomology degree in ,5; 36/24/ Major pr fessor Date j.,//él/7/Y 0-7639 COMPARATIVE RESISTANCE TO AZINPHOSMETHYL IN THE PREDATORY MITE AMBLYSEIUS FALLACIS CARMEN (ACARINA: PHYTOSEIIDAE) SAND ITS PREY TETRANYCHUS URTICAE KOCH (ACARINA: TETRANYCHIUAET_IN—CREENHUUSE EXPERIMENTS BY Joseph Grant MOrse A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Entomology 1978 ABSTRACT COMPARATIVE RESISTANCE TO AZINPHOSMETHYL IN THE PREDATORY MITE AMBLYSEIUS FALLACIS CARMEN (ACARINA: PHYTOSEIIDAE) ANO ITS—PREY_TETRANYCHUS URTICAE KOCH (ACARINA: TETRANYCHIOAES IN GREENHOUSE EXPERIMENTS BY Joseph Grant Morse Susceptible populations of a phytoseiid predator, Amblyseius fallacis Carmen, and one of its tetranychid prey, Tetranychus urticae Koch, were separately selected for resistance to azinphosmethyl under similar greenhouse condi- tions. When a single homogeneous strain was used, resist- ance failed to develop in the predator after 8 generations. Selection of a predator population of heterogeneity similar to the prey population (consisting of the initial strain hybridized with two additional susceptible strains), how- ever, resulted in appreciable resistance development. A comparison of resistance development in the predator and prey showed a 23.87-fold resistance in A, fallacis follow- ing 14 selections vs. a 20.41-fold resistance in T. urticae after 22 selections. Similar rearing programs, population sizes and selection procedures were maintained in both experiments. Selections were initiated with populations of similar Joseph Grant Morse dosage-mortality line slopes although toxicant ranges used in the two experiments differed. Additional experiments were undertaken to determine the importance of the initial gene frequency (of the major gene responsible for resist- ance) in the two comparisons. J... To my Mother and Father -- for their understanding, support and love ii ACKNOWLEDGEMENTS I would like to express gratitude to my major professor, Dr. Brian A. Croft, for suggesting this research project as well as his continuing interest and support. Grateful acknowledgement is also made to the members of my guidance committee, Dr. Erik Goodman, Dr. Roger Hoppingarner, Dr. Frederick Stehr and Dr. Gary Simmons for their suggestions and criticisms. The author also wishes to thank Dr. R.A. Harrison, Department of Entomology, Lincoln College, Canterbury, New Zealand for his assistance in obtaining the susceptible Tetranychus urticae population used in this study. iii TABLE OF CONTENTS LIST OF TABLES . LIST OF FIGURES INTRODUCTION . LITERATURE REVIEW Biological, Physical and Behavioral Comparisons of Tetran chus urticae Koch and Amblyseius faIIaciS Carmen . . . . . Basic Bionomics Reproduction Genetics and mechanism of resistance to organophosphates Mechanism of resistance . Genetics of resistance Previous selection experiments with T. urticae and A. fallacis . MATERIALS AND METHODS . RESULTS AND DISCUSSION . CONCLUSIONS . LIST OF REFERENCES . APPENDICES . Appendix A - Further Figures Appendix B - Experimental Data . iv Page vi 11 ll 13 l4 16 25 33 38 44 44 76 TABLE A-l A-2 LIST OF TABLES A Biological Comparison of T. urticae and A. fallacis . Origin of Mite Strains used in the Experiments Events in the 16-day Generation Cycle . A Summary of Experiments 1-5 A Comparison of Predicted and Actual Mortalities Following Spray Application for I. urticae and A. fallacis Selection Data for Experiment 1 (A, fallacis) Selection Data for Experiment 2 (A. fallacis) Selection Data for Experiment 3 (T. urticae) Selection Data for Experiment 4 (A, fallacis) Selection Data for Experiment 5 (I. urticae) Relative Changes in LC50 for Experiments 2 and 3 . . . . Relative Changes in LC50 for Experiments 4 and 5 . . . . . . . . . . . . A Comparison of Predicted vs. Actual Mortalities for Pooled A. fallacis and T. urticae Populations Page 20 22 30 32 76 77 78 80 81 82 83 84 FIGURE 10. ll. 12. 13. LIST OF FIGURES Outline of Experiments 1-5 . Diagram of Experiment 4 and 5 Dosage- Mortality Lines for Experiment 1: Selection of an A. fallacis Population Through 7 Generations Change in LC50 with Time for Experiment 1. Dosage-Mortality Lines for the Three Susceptible Strains and the Resulting Composite Strain Used in Experiment 2 . Dosage- Mortality Lines for Experiment 2: Selection of an A. fallacis Population Through l4 Generations Dosage- Mortality Lines for Experiment 3: Selection of a T. urticae Population Through 22 Generations Relative Change in LC50 for Experiments 2 and 3 . . . . . A Comparison of Population Sizes Prior to Selection in Experiments 2 and 3 . A Comparison of Percent Mbrtality After Selection in Experiment 2_and 3 '. Dosage- Mortality Lines for Experiment 4: Selection of an A. fallacis Population Through 6 Generations Dosage- Mortality Lines for Experiment 5: Selection of a T. urticae Population Through 6 Generations X—Fold Change in LC50 for Experiments 4 and 5 . . . . . . . . . . . . . . . vi Page 18 18 45 47 49 51 53 55 57 59 61 63 65 LIST OF FIGURES (Continued . . .) FIGURE 14. 15. l6. 17. 18. Y-Fold Change in LCSO for Experiments 4 and 5 . . . . . . . . . . . . . . . . . A Comparison of Population Sizes Prior to Selection in Experiments 4 and 5 A Comparison of Percent MOrtality After Selection in Experiments 4 and 5 Relationship Between Predicted Probit Mortality, Actual Spray Concentration and the Dosage-Mortality Line . A Comparison of Mortality Probit Ratios (Actual/Predicted Mortality) Following Selection for A. fallacis and T. urticae . vii Page 67 69 71 73 75 INTRODUCTION Public disenchantment with the use of pesticides began in 1962 with Rachel Carlson's "Silent Spring". Increasing public concern over the problems of objection- able pesticide residues, adverse effects on nontarget organisms and direct hazards to the user (Smith 1970) have led to severe limitations on the types and amounts of pesticides available for pest control. Pesticides, however, remain the entomologists most powerful tool (NAS 1969, Metcalf 1975) and their use will most likely increase in the foreseeable future (Mrak Commission 1969). Problems other than those related to environmental contamination have also been associated with the use of chemical pesticides. Pest resistance resulting in escala- tion of dosage levels, often accompanies repeated pesticide applications. Secondary pest outbreaks following elimina- tion of natural control by previously unnoticed natural enemies is a similar symptom of the "pesticide syndrome" (Doutt and Smith 1969). The above problems have led to the formulation of the concept of integrated control (Bartlett 1956) and subsequently the concept of pest management (Geier and Clark 1961, Luckmann and Metcalf 1975). Pest 2 management involves evaluation of all available techniques and their consolidation into a unified program to manage pest populations so that economic damage is avoided and adverse side effects on the environment are minimized (NAS 1969). Implicit in this concept of pest management is the maximal use of natural enemies for pest control. Much too often the importance of natural enemies in controlling pest populations has been ignored and is emphasized only by its absence when pesticides destroy the effectiveness of natural enemies. Pest resistance is one of the main problems facing pest management, yet natural enemy resistance to pesticides is nearly unknown. Comparing pests and natural enemies, there are more than 268 cases of insecticide resistance known for pest species (Brown 1976) but only 12 reported cases for natural enemies (Croft and Brown 1975, ‘Croft 1977). Three of these twelve resistant natural enemies are parasitoids which developed resistance as a result of laboratory selections — two braconids and an aphelinid. The remaining nine are predators reported to be resistant in the field - seven phytoseiid mites, an anthomyiid and a cocinellid. The proposed reasons for this major imbalance fall into four main categories: (1) Previous field and laboratory research has focused on pest Species rather than natural enemies. Perhaps 3 then, at least part of this skewed dichotomy is due to lack of study of natural enemies. Croft and Brown (1975) list five reasons for this differential emphasis and effort. (a) Greater attention is given to control of direct competitors rather than to conservation of bene- factors. (b) The assumption is often made that natural enemies respond to insecticides in the same way as pests and thus their study is unnecessary. (c) Much greater monetary resources are made avail- able for studying the responses of pests as compared to those of predators and parasites. (d) Predators and parasites are often more diffi- cult to rear or culture in the large numbers required for detailed experimentation. (e) There is a lack of standardized toxicological test methods for natural enemies similar to those developed for pest evaluations. (2) The appearance of a resistant pest species is often more apparent than that of a resistant natural enemy. Damage caused by resistant pest species makes them more visible and usually results in their investigation. Natural enemy abundance is quite dependent on host or prey levels and thus resistance development among natural enemies could be overlooked if host or prey species were not especially abundant. 4 (3) A third possibility is that natural enemy and pest species may actually differ in their ”intrinsic” abilities to develop resistance. Gordon (1961) hypo- thesized that "the extraordinarily high and generalized tolerance of the larval feeding stages of relatively poly— phagous holometabolous insects to contact insecticides is probably the result of selection for endurance of prolonged and varied biochemical stresses associated with a diversity of their natural food plants." Kreiger et al. (1971), working with lepidopterous larvae, found a correlation be- tween range of host plants and activity of aldrin epoxidase in midgut tissues (polyphagous > oligophagous > monophagous). They concluded that the detoxification of secondary plant substances is the chief function of the mixed function oxi- dase (MFO) system in the midgut of lepidopterous larvae. Brattsten and Wilkinson (1977) confirmed that in the south- ern armyworm.moth, Spodoptera eridma (a broadly polyphagous insect), MFO enzymes are induced by secondary plant sub- stances and that this induction "proceeds with enough Speed to provide the animal with increased protection against these potentially offensive dietary factors". Can this principle be extended to include natural enemies? Is there a natural ordering of MFO activity in insects (as reflected in inherent tolerances to certain in- secticides) related to the degree of biochemical and meta- bolic specialization (polyphagous > oligophagous >monophagous 5 > predators > parasites)? Plapp and Bull (1977) studied the toxicity of several insecticides to the tobacco bud- worm, Heliothis virenscens (F.), one of its predators, Chrysopa carnea (Stephens), and one of its parasites, Carpoletis sonorensis (Carlson). They determined the order of toxicity to organophosphates (detoxified mostly by oxidases) as parasite > predator > budworm. Certainly much more data is needed to determine if this principle may be extended to other pests, predators and parasites. Differences also exist between pest and natural enemy Species due to their different modes of life. Second- ary poisoning (from their prey or host) may complicate natural enemy resistance development. Additionally, pest species are often less mobile than natural enemy species (due to the need for natural enemies to search for hosts or prey, especially following insecticidal treatment), and thus natural enemies may come in contact with more in- secticide. Croft and Brown (1975) however, stated that "there is little experimental evidence published to date indicating that the behavior of these natural enemies con- fers a greater exposure to insecticides." (4) The most compelling hypothesis explaining the paucity of resistant natural enemies involves their density dependence upon their hosts or prey. In order for a natural enemy population to exist, sufficient hosts or prey must be present. In the presence of pesticide applications, natural 6 enemies are faced not only with the stress imposed by the selecting chemical but additionally with the stress of a limited food supply. Thus one might expect natural enemies to develop resistance only after their host or prey species have done so. It is significant to point out that in all known cases of natural enemy resistance, ”resistance in the principal prey (the pest) has preceded the resistance similarly developed by the predator" (Croft 1977; all cases of parasite resistance reported were the result of laboratory selections). Complicating this problem of natural enemy resistance development is the necessary simultaneous maintenance of both an adequate food supply and continued exposure to the selecting chemical. Commonly, as soon as a pest species develops resistance, and thus becomes abund- ant enough for its natural enemy to reproduce, the pesticide is changed to one that again controls the pest. Thus the natural enemy is faced with a food shortage as well as the stress of pesticide exposure. Among the documented cases of natural enemy resistance is one phytoseiid mite species, Amblyseius fallacis Carmen, which when combined with one of its principal prey species, Tetranychus urticae Koch, the two-spotted spider mite, presents an ideal model pest/natural enemy system. These two mite species are remarkably similar in most physio- logical, biological and behavioral characteristics as well 7 as in their magnitude and Spectrum of resistance to organo- phosphates (Croft 1977). In addition, they are quite com- parable in most ecological aspects, occupying nearly identical habitats throughout their life histories. They thus provide an outstanding Opportunity to investigate the development of resistance in a pest/natural enemy system as well as to test the hypothesis that the dependence of natural enemies upon their host or pest species is a primary deterrant to the development of resistance. In the present study, susceptible p0pulations of A. fallacis and T. urticae were separately selected for resistance to azinphosmethyl, a common broad-spectrum organo- phosphate. In the first part of the study, a non-limiting food source was supplied to each population, eliminating the above hypothesized deterrant to resistance development. In the second part of the study, the selection was repeated with the addition of small numbers of resistant genotypes to each population. This experiment was conducted to reflect the importance of the initial gene frequency (of the major gene responsible for resistance) in the first selection experiment. LITERATURE REVIEW Biological, Physical and Behavioral Comparisons of Tetranychus urticae Koch and Amblyseius fallacis Carmen. Basic Bionomics (see Table 1) Both T. urticae and A. fallacis have four develop- mental stages (egg, six-legged larvae, protonymph and deutonymph), each of the last three being followed by a quiescent resting stage (nymphochrysalis, deutochrysalis, teliochrysalis) (McMurtry et a1. 1970, van de Vrie et a1. 1972). Males of A. fallacis have been reported to have no deutonymphal stage (Ballard 1954) whereas males of T, urticae do, although it is shorter in duration than that of the female (Boudreaux 1963). Developmental and ovipositional periods and thus the intrinsic rate of increase for both species, vary greatly with temperature and to a lesser extent, relative humidity (McMurtry et al. 1970, van de Vrie et a1. 1972). Developmental times (in days) for A. fallacis range from 5.0 (Ballard 1954) to 5.8 (McClanahan 1968) at 26°C to 11.6 (McClanahan 1968) at 20°C. Depending on temperature, values for T. urticae range from six to ten days (van de Vrie et a1. 1972). Boudreaux (1954) listed an average preovipositional TAB LE 1 A Biological Comparison of T. urticae and A. fallacis Parameter T. urticae AlgfalIaEis SOurce Number of four four McMurtry et al. developmental 1970; van de stages Vrie et al.1972 developmental 6-10 days 5-11.6 days Ballard 1954; period McClanahan 1968; van de Vrie et al. 1972 preovipositional 1 day 1 day Boudreaux 1954; period (22-27°C) (26°C) Bravenboer 1959 duration of 10.8-26.3 22 days Caegle 1949; oviposition days Ballard 1954 eggs/female/day 2.5-5.6 2.2 Caegle 1949; Ballard 1954 intrinsic rate .2585 (22°C) .279 (25°C) Croft unpubl. of increase Wrensch and Young 1975 mode of arrhenotokous arrhenotok- Helle and Bolland reproduction partheno- ous parthen- 1967; Hansell genesis ogenesis et a1. 1964 # chromosomes Helle and Bolland ‘male 3 4 1967; Hansell female 6 8 et a1. 1964 mating necessary no yes Helle and Bolland for egg 1967; Rock et a1. production 1976 sex ratio varies varies Overmeer and average 2.5 2.0-5.0 Harrison 1969; Burrel and McCormich 1964; Smith and Newsom 1970; Lee 1972; Croft unpubl., Rock et a1. 1971; Ballard 1954 size 450 by 300 400 by 250 - (length by width) (microns) (microns) diapause fertilized fertilized McMurtry et al.1970; van de Vrie et a1. 1972. adult female adult female 10 period for A. fallacis of one day at 78°F (26°C). Bravenboer (1959) reported values for T. urticae of .5 days (27-33°C), 1 day (22-27°C), 2 days (18-22°C) and 5 days (13.5°C). Ballard (1954) reported 22 days as the average duration of oviposition for A, fallacis at 26°C. Caegle (1949) reported ovipositional periods (depending on temperature) of 10.8-26.3 days for T. urticae. Values of eggs per female per day were 2.2 (Ballard 1954, 26°C) for A. fallacis and 2.5-5.6 (Caegle 1949) for T, urticae. Croft (unpubl.) obtained a value of .279 for the intrinsic rate of increase of A, fallacis at 25°C. Wrensch and Young (1975) determined a value of .2585 for T. urticae at 22°C. Reproduction In both T, urticae (Helle and Bolland 1967) and A. fallacis (Hansell et al. 1964) reproduction is based upon arrhenotokous parthenogenesis, the male having the hap- loid number of three (T, urticae) or four (A, fallacis) chromosomes. In T. urticae, virgin females produce male progeny (Helle and Bolland 1967). In A. fallacis, however, mating is necessary for egg production (Rock et a1. 1976). This difference would be of significance only under low population densities in which male-female encounters were restricted. Experiments by Overmeer and Harrison (1969) indi- cated that the sex ratio 0f.Z- urticae averages near 2.5:l.0 (female to male), fluctuating with the genotype of the 11 female parent. Mitchell (1972) further showed that the sex ratio was the result of several alleles inherited from the female which respond to natural selection. He pointed out that this was especially interesting in view of data in- dicating that the sex ratio is quite important in dispersal (Overmeer and Harrison 1969, Mitchell 1970). The sex ratio of A. fallacis also seems to fluctuate with reported values falling in the 2.0-5.0 (female to male) range (2.0 by Burrell and McCormich 1964, Smith and Newsom 1970, Lee 1972; 3.0 by Croft unpubl.; 4.0 by Rock et a1. 1971 and 5.0 by Ballard 1954). Genetics and Mechanism of Resistance to Organophosphates Mechanism of Resistance The mode of action of organophosphorous insecticides involves the inhibition of an enzyme, cholinesterase, important in the transmission of nerve impulses (Brown 1969a). Cholinesterase normally acts by metabolizing acetylcholine, the chemical messenger linking nerve axons through their synapses. When Cholinesterase is inhibited, acetylcholine accumulates at the synapse, resulting first in an increase and then in cessation of nervous conduction. The result is paralysis and death of the organism. Two distinct mechanisms have been discovered for the resistance of T. urticae to organophosphates. The first was discovered by Smissaert (1964) and involves an altered 12 Cholinesterase molecule which is both less active and less susceptible to organophosphorous inhibition. This mech- anism has been reported in strains of T. urticae from the Netherlands (Smissaert 1964, V033 and Matsumura 1964), New Zealand (Ballantyne and Harrison 1967, Overmeer and Harrison 1969), Germany (Schulten 1968), Israel (Zahavi and Tahori 1970) and Great Britain (Cranham 1974). A second mechanism.was discovered for two US strains of T. urticae (Matsumura and Voss 1964, Herne and Brown 1969). In these strains, the Cholinesterase activity of the resistant strain was shown to be identical with the suscept- ible strain and resistance was attributed to a higher de- toxicative capacity of the resistant strain resulting from an increased carboxyesterase activity. Herne and Brown (1968) suggested that the difference between the European and the American strains may be due to the former having been selected with demeton and oxydemetonmethyl, whereas the latter were selected with parathion and malathion. Accord- ing to this hypothesis, resistance of the Strain of T. urticae used in the present Study is assumed to be due to a higher detoxicative ability of the resistant strain over the susceptible strain. Motoyama et a1. (1972) discovered an identical mech- anism for azinphosmethyl resistance in A, fallacis. Again, the resistant strain was found to degrade the organophos- phate faster than the susceptible strain, resulting in less 13 inhibition of the cholinesterase activity. When bimolecu- 1ar rate constants were compared, no difference was found between S and R strains, indicating that a modified cholinesterase was not associated with resistance. Further studies by Motoyama et a1 (1977) have indicated that the exact mechanism of azinphosmethyl resistance in A. fallacis involves glutathione S-transferases. Genetics of Resistance Most cases of strong resistance to an insecticide depend upon allelism in a single major gene (Brown 1969b). During the development of resistance, however, several modifying genes may be important. Independent of the mechanism involved, the resistance of T. urticae to both malathion (Taylor and Smith 1956) and parathion (Schulten 1966, Ballantyne and Harrison 1967, Herne and Brown 1969) has been shown to be due to a single major dominant gene, with or without modifiers. Croft et a1. (1976) determined that the resistance of A, fallacis to azinphosmethyl is due principally to a single allele which exhibits partial dmminance. They hypo- thesized that partial dominance may be a characteristic of resistance to azinphosmethyl as suggested by the study of Dittrich (1972). Dittrich (1972) demonstrated complete .dominance for parathion, incomplete dominance for paraoxon, and no dominance for oxydemetonmethyl in a demeton-parathion selected strain of T, urticae. 14 Previous Selection Experiments with T. urticae and A, fallacis Two-spotted spider mites, Tetranychus urticae, are infamous for their ability to develop resistance to a wide variety of pesticides. Resistance is especially prevalent in greenhouses where more intensive control programs and greater numbers of spider mite generations are present (Helle 1965). Numerous laboratory selection experiments with organophosphorous compounds and T. urticae have been carried out (Watson 1956, Hanson 1958, Helle 1959, Saba 1960, Watson and Naegele 1960, Abul-Hab and Stafford 1961, Dittrich 1963, Helle 1965, Overmeer 1966) with common results; most often a resistant strain is produced. In highly inbred strains, however, slower resistance develop- :ment has been encountered. McEnroe and Harrison (1968) hypothesized a slow response to selection in an "internally ,balanced” strain (highly inbred) and a "rapid response to selection when strains are outcrossed". This hypothesis was experimentally confirmed by McEnroe and Naegele (1968) and McEnroe and Kot (1968). The first reported case of OP-resistance in A, fallacis was by Motoyama et a1. (1970). Since then, numerous reports have indicated both a high magnitude and widespread distribution of resistance throughout the mid- western and eastern deciduous fruit-growing regions of North America (Croft and Brown 1975, Croft et a1. 1976, Croft and Nelson 1972). One of the few successful selection 15 experiments with A, fallacis has been performed by Croft and Meyer (1973). They produced a carbaryl-azinphosmethyl resistant strain through the selection and hybridization of two strains; one carbaryl resistant and one azinphos- methyl resistant. Croft (1972) and Croft and Meyer (1973) commented on the lack of success in laboratory selection attempts and suggested that "greater success might be realized by periodically introducing genetic variability (wild genotypes) during laboratory selection experiments". MATERIALS AND METHODS The alternative experiments of the present study are outlined in Figure 1. In Experiments 1 and 3 (the form of Experiment 1 was repeated in Experiment 2 with a different strain of susceptible A, fallacis) susceptible populations of Tetranychus urticae and Amblyseius fallacis were separately selected for resistance to azinphosmethyl under Similar greenhouse conditions. Azinphosmethyl was chosen as the selecting chemical for two reasons: (1) azinphosmethyl is one of the primary broadspectrum organophosphorous insecticides used in tree fruit pest management and (2) the genetics of resistance to azinphos- methyl has been determined for A, fallacis (Motoyama et a1. 1972, Croft et a1. 1976). Crucial to these experiments is the provision of a non-limiting food supply to each species, thus eliminating one of the hypothetical deterrants to the development of resistance. Resistant T. urticae were supplied to the A, fallacis as prey, thus allowing survival through chemical selections. In Experiments 4 and 5, the same two susceptible populations were again selected for resistance to azinphos- -methyl. Here, however, small numbers of resistant adult ‘females of each Species were added to the first (P) and 16 17 Figure 1. Outline of the experimental cells of Experiments 1-5. Figure 2. Diagram of Experiments 4 and 5. 18 Experiment 1,2 susceptible A. fallacis with non-Iimiting food supply Experiment 3 susceptible T. urticae with Hon-Iimiting food supply Experiment 4 A. fallacis plus smaII numBers of resistant A. fallacis Experiment 5 T. urticae plus smaII numbers of resistant T. urticae 800 susceptible adult females 800 surviving adult females 800 survivors 800 survivors rue——+u1<——+ru 20 resistant adult females selection 20 resistant adult females selection 2 l selection F 3 l selection 19 second (F1) generations (see Figure 2). Experiments 4 and 5 provide a comparison of the "innate" ability of the two populations to develop resistance under similar conditions, since any differences in the initial gene frequence (of the major gene responsible for resistance) will be negated by flooding the system with resistant genotypes. A com— parison of Experiments 2 and 3 vs. 4 and 5 should thus indicate the importance of the initial gene frequencies in the development of resistance. The origins of both T. urticae and A, fallacis strains used in the experiments are listed in Table 2. The Rose Lake susceptible (S) Strain was collected in 1974 from a weed-groundcover habitat which had no known history of pesticide application (Croft et a1. 1976). Mites were maintained on units similar to those described by McMurtry and Scriven (1964) and Hoying (1976) until their use in the experiments. The MCnroe S and Garden 3 strains were also collected from areas (Monroe Co., MI on soybean and Lansing, M1 on beans and cucumber respectively) which had no known history of pesticide application. The resistant (R) A, fallacis strain was collected in September 1976 from apple leaves and groundcover obtained from a commercial apple orchard near Belding, MI. For the ‘previous ten years, azinphosmethyl supplemented occasion- ally with diazinon had been the principal broadspectrum OP compound applied for pest control (see Croft et a1. 1976 for ‘method of collection). 20 TABLE 2 Origins of Mite Strains Used in the Experiments Strain Origin and History Susceptible A, fallacis Rose Lake S Rose Lake State Game Area (near Lansing, MI) collected from weed-ground cover habitat (Croft et a1. 1976) Monroe S Monroe Co., MI collected from soybean August 1976 Garden 8 Lansing, MI collected from beans and cucumber August 1976 Resistant A, fallacis Belding R commercial apple orchard near Belding, MI Sept.1976 Susceptible T. urticae New Zealand S derived from the susceptible LN4 strain, Lincoln College, Canterbury, New Zealand (courtesy R.A. Harrison) Resistant T. urticae Greenhouse R greenhouse T. urticae from Michigan State University, E. Lansing, MI 21 The susceptible T, urticae strain was obtained from a susceptible New Zealand colony of T, urticae (derived from the LN4 strain) courtesy of Dr. R.A. Harrison (Lincoln College, Canterbury, New Zealand). The resistant T, urticae strain was collected from a Michigan State University greenhouse (E. Lansing, MI) in which numerous insecticidal treatments (organophosphates and others) had been made for the previous ten years. Both T. urticae and A, fallacis populations were reared and selected on 16-day (an 18-day cycle was used, with two days added to the beginning of each cycle, during the winter months, Nov.-March) generation cycles as shown in Table 3. Each cycle was initiated with the hand trans- fer (using a fine camel hair brush) of 200 adult female mites to each of four fresh lima bean plants (Phaseolus limensis). The lima bean plants used for the A. fallacis populations were infested with an abundant prey population (resistant T. urticae) one day prior to each transfer. On day eight the leaves from the initial four plants were transferred to eight fresh plants in order to insure good plant condition (plants deteriorate under continuous T. urticae feeding in about ten days) and an adequate food supply. On day twelve the dried leaves from the four initial plants were removed (active life stages crawled onto the fresh plants), the resistance level of the popu- lations were determined, and the plants were surveyed to 22 TABLE 3 Events in the l6-day Generation Cycle Day # Event 12 l4 l6 transfer 200 mated adult females to each of four fresh lima bean plants (A. fallacis plants were previously infested with T. urticae) transfer leaves from the initial four plants to eight fresh plants to insure good plant con- dition and adequate food supply. remove dried dead leaves of four initial plants; slide-dip to determine present resistance levels; survey plants to deter- mine pre—spray population levels. spray seven plants to run-off at levels predicted to give approx- imately 75% mortality; leave eighth plant in case of excessive mortality. assess post-spray population levels; initiate next generation by transferring 200 adult female survivors to each of four new plants. 23 determine approximate pre-spray population levels (a des- cription of the survey technique follows). On day four- teen the plants were sprayed to run-off (using a compressed air knapsack sprayer) at a concentration calculated to give approximately 75% mortality. Wettable-powder insecticide (GuthionR 50-WP) was used for both resistance level deter- minations and spraying operations. In order for any residual action to occur, the plants were left for 48 hours (day 16) before post-spray population levels were deter- mined. Surviving adult females were then removed and used to initiate the next generation cycle. The slide-dip method of Anon. (1968) as modified by Croft et a1. (1976) was used to determine resistance levels for both T, urticae and A. fallacis populations. Twenty adult females were placed on their backs on Permacel BrandR filament tape which was affixed by Scotch BrandR double-Stick tape to a microscope slide. Adult females were used once distension of their abdomen indicated that oviposition had begun. Six-hundred mites (one-hundred at each of five con- centrations and a control) were dipped in toxicant solu- tions of 50-WP insecticide dissolved in distilled water. Slides were dipped for five seconds, blotted on paper, and allowed to dry for fifteen minutes. Slides were then held ,for 48 hours at 25°C and 95% RH before mortality was deter- mined by failure to exhibit leg or mouthpart movement when 24 mites were lightly prodded with a camel hair brush. Results were plotted on logarithmetic-probability paper; LC50 and slope were calculated using a computer. Spray dosages predicted to give approximately 75% mortality were determined from the plotted dosage-mortality curves. Each lima bean plant contained approximately 100 basal leaves (trifoliets were removed to keep the plants under vegetative control). A random survey technique was used in order to determine pre-spray population levels. Ten uniformly distributed basal leaves were sampled from each plant (chosen in an X pattern) and appropriate con- versions were made to determine the total number of mites present. 48 hours after spraying (day 16), the total plant was surveyed to determine post-spray survival. Populations of A, fallacis and T. urticae were reared and selected in adjacent greenhouse rooms in order to maintain nearly equal environmental conditions. Temperature records were maintained to check that similar temperature ranges (65-85°F) were present. Mites were reared on plants placed in large water trays in order to minimize dispersion. RESULTS AND DISCUSSION Experiment 1 In Experiment 1, the Rose Lake susceptible Strain of A. fallacis was selected with azinphosmethyl through seven generations (see Table A-l). The change in the dosage-mortality lines with selection is Shown in Figure 3, with the change in LC50 (50% lethal concentration) plotted in Figure 4. As seen in Figure 4, the LC50 did not in- crease with selection, but in fact, decreased somewhat. The Rose Lake susceptible strain was a highly inbred strain derived originally from a limited number of indivi- duals. The high Slope of 4.391 (of the dosage-mortality line per decade) indicates a narrow genetic base with res- pect to azinphosmethyl tolerance. Limited genetic vari- ability was hypothesized to explain the inability of the Rose Lake strain to respond to azinphosmethyl selection. This hypothesis is tested in Experiment 2. Experiments 2 and 3 The same selection format was followed in Experi- ments 2 (A, fallacis) and 3 (T. urticae). In Experiment 2, however, the A. fallacis parental population (generation number 0) was derived from equal numbers of three suscept- ible populations (300 mites each from the Rose Lake S, 25 26 Monroe 8 and Garden S strains). Dosage-mortality lines for the composite strain as well as for the three suscept- ible strains from.which it was derived are shown in Figure 5. The increased genetic variability available to the composite strain decreased the slope of the parental population (of Experiment 2 versus 1) from 4.391 to 1.780. Figure 6 shows the change in the dosage- mortality lines with selection for Experiment 2. As shown in Table A-2, the LC50 increased by a factor of 23.87 (.00107 to .02549 % A.I.) in 14 selections. The susceptible T. urticae strain used in Experiment 3 was obtained from New Zealand (derived from the LN4 strain, courtesy of Dr. R.A. Harrison, Lincoln College, New Zealand) and was found to have an LC50 of .03039 (% A.I.) with a slope of 2.200. The slopes of the two susceptible populations used in starting Experiments 2 and 3 compare quite favorably (1.739 for A, fallacis and 2.200 for T, urticae). The parental (generation 0) L050 levels for the two species, however, differ by a factor of approximately 30 (.00107 for A. fallacis and .03029 fol/T. urticae, % A.I.). Questions as to whether (1) this reflects the true "intrinsic" susceptibility of the two Species and (2) whether this difference in toxicant ranges used in the two experiments will differentially affect the ability of the two species to acquire resistance, remain as yet un- answered. 27 Figure 7 shows the change in dosageemortality lines for Experiment 3. The LC50 of the T. urticae population increased by a factor of 20.41 (.03029 to .61080 % A.I.) in 22 selections. Figure 8 shows the rela- tive change in LC50 (with the LC50 of the parental strain used as a base) with selection for the two species in Experiments 2 and 3 (data in Table A-6). In Figures 9 and 10, the population sizes prior to selection and the percent mortality after selection, respectively, are compared for Experiments 2 and 3 (data in Tables A-2 and A-3). The figures demonstrate that population sizes prior to selection (average of 4658 for A. fallacis and 4779 for T. urticae) and percent mortality after selection (average of 71.0% for A, fallacis and 76.1% for T. urticae) were similar for the two experiments. Experiments 4 and 5 Experiments 4 (A. fallacis) and 5 (T. urticae) test the importance of the initial gene frequencies (of the genes responsible for resistance development) in Experiments 2 and 3. The addition of resistant individuals to each population prior to selection should negate any differences in initial gene frequency. Thus, any differences in the two species, important in the development of resistance (other than the initial gene frequency) should result in differential rates at which resistance is acquired. 28 Figures 11 and 12 show the change in dosage- mortality lines for the two populations. Again the LCSO'S of the parental populations (.0153 for A. fallacis and .5048 for T. urticae) differ by a factor of approximately 30. In addition, the distance between the LCSO'S of the parental (line 00) and the resistant (line R) populations differ for the two species (.46512/.00092=506.67 for A, fallacis; .61847/.03029=20.42 for T, urticae). Two methods of comparing the change in LC50 for the two species are contrasted in Figures 13 and 14. In Figure 13, the ”X"-fold change in LC50 (using the parental LC50 as a base) is plotted. In this figure, resistance development of the A, fallacis population appears to have surpassed the T. urticae population. In Figure 14, the fractional change in LC50 with respect to the LC50 of the resistant strain (LC =0.00; LC =1.00) is 50(parental) 50(R) plotted as a function of selection. Here it is seen that the T. urticae population approaches the L050 of the resistant strain much more quickly than the A. fallacis population. Great care must be taken in interpreting these results. The two resistant populations used in adding resistant genotypes into Experiments 4 and 5 were obtained from quite different sources (see Table 2) than the suscept- ‘ ible populations used in all five experiments. Incom— patibilities between strains of both T. urticae (Helle and 29 Pieterse 1965) and A. fallacis collected from different areas is quite common. In addition, the differences in the resistant/susceptible LC50 ratios for the two species further complicates comparisons. Further studies are presently under way in our laboratory to resolve these problems. Figures 15 and 16 show population sizes prior to selection and percent mortality after selection, respectiv- ely, for the two experiments. Again, comparable ranges are present for the two species. Table 4 shows a summary of Experiments 1-5. In Experiment 1, the Rose Lake susceptible strain failed to acquire resistance through 7 selections. Experiments 2 and 3 demonstrate that under the present selection regime (similar population sizes, mortality levels, environmental conditions, unlimited food supplies, etc.) both species acquired an appreciable level of resistance, although the A. fallacis population did so somewhat faster. Experiments 4 and 5 fail to resolve the question of whether the two species differ in their abilities to acquire resistance when differences in initial gene frequency (of the major gene responsible for resistance) are negated. Predicted Versus Actual Selection Mortalities Dosage-mortality lines obtained using the slide-dip -method were used to calculate spray concentration levels to give approximately 75% mortality during selections. Using 30 m mmdoncoouo «.mn Comm 0 Hm.om mmmmo. mmomo. m vcwame 362 mmowuun .M m m waaeflom m mouaoz m ampumo ..IIIIII I m.~o mwoq m mo.am oqmqo. «mooo. m 0x64 omom mwomaamm .< q H.0n muse Nu H¢.om woman. mmomo. m vamHmoN 362 mmowuup .M m m accuse m moucoz I o.an mmoq «a mw.m~ mwmmo. noaoo. m oxmq omom mwomaamm .< N ¢.mn Humm 5 He. ooooo. oqaoo. m oqu omom mwomaamm .« H cowuooaom aoauooaom \mnwm mcoau . . . \sun con“ -omamm onus an A H < AV -Hmuuoz -manaom mo swamno one; onus ammo owmum>< mwmuo>< Honasz CHOMnx Hmcwm HmwuHcH Anvaflmnum mmwooam ucoaaummxm mnH mucoawummxm mo kumaabm < é mam<fi 31 the actual spray concentrations on the x-axis of the dosage-mortality line and reading across to the y-axis (see Figure 17 ) yielded a predicted mortality level. Actual mortality levels 48 hours after spraying were cal- culated as previously described (see Materials and Methods). Actual/predicted probit mortality ratios (probit ratios were used in order to obtain a linear relationship between dosage and mortality) were calculated for each selection in which the appropriate data were available (see Table A-8). Probit ratios were pooled for each species (Experiments 1,2,4 for A. fallacis and 3,5 for T. urticae) with the results summarized in Table 5. Figure 16 graphs the probit mortality ratio for each species broken down by experiment number. An F-test of probit ratios showed probit mortality means were significantly different for the two species using a 99.5% confidence level (1.091 1’ .105 for A. fallacis and 1.219 1* .134 for T. urticae). These results should be of use in more accurately determining spray concentrations to give desired spray mortality (using the same spray methods). 32 TABLE 5 A Comparison of Predicted and Actual Mbrtalities Following Spray Application for T. urticae and A. fallacis (see Appendix, TaBIe A-8) A. fallacis T. urticae number of samples 21 21 average predicted 56.9 42.0 mortality (%) range high 81.9 70.9 low 25.0 15.3 average predicted 5.1942 4.7989 mortality (probits) average actual 72.6 76.1 mortality (%) range high 88.7 94.6 low 55.8 43.6 average actual 5.5584 5.7534 mortality (probits) average ratio (probits) 1.091* 1.206* range high .957 .981 low 1.337 1.504 standard deviation .10530 .13431 95% confidence (1.04164, 1.13988) (1.15826, 1.28019) intervals *significantly different at .005 level. CONCLUSIONS Numerous references (McEnroe and Harrison 1968, McEnroe and Kot 1968, McEnroe and Naegele 1968, Croft 1972, Croft and Meyer 1973) have been made to the import- ance of genetic variability (i.e. a wide genetic pool) in selection experiments. This principle is demonstrated in a comparison of Experiments 1 and 2 (A. fallacis). The Rose Lake susceptible strain failed to increase its tolerance to azinphosmethyl through seven selections in Experiment 1. When the Monroe and Garden susceptible strains were combined with the same Rose Lake strain (decreasing the slope from 4.391 to 1.780), however, a resistance factor of 23.87 was achieved in 14 generations. Great difficulty was encountered in obtaining a susceptible T, urticae strain for use in Experiments 3 and 5. The strain finally used in these experiments (New Zealand S) had a higher LCSO of .03029 as compared to a value of .00107 (% A.I.) for the susceptible A, fallacis population with which it was compared in Experiments 2 and 4. As mentioned previously, it is unclear whether the different toxicant ranges present in Experiments 2 (A. fallacis) and 3 (T. urticae) differentially favored 33 34 either population in the development of resistance. Biological, physiological and behavioral charac- teristics of A. fallacis and T. urticae are almost identi- cal. The two major exceptions are that (l) fertilization is necessary for egg production in A, fallacis (unferti- lized T. urticae produce male progeny) and (2) A. fallacis and T. urticae differ in their mode of food uptake (pre- dator versus herbivore, respectively). The first of these two differences (fertilization and egg production) would be significant only under low population densities which were not present in these experiments. Two effects may arise from the second major difference between the two species - (l) the possibility of secondary poisoning exists for A. fallacis and (2) A. fallacis might be subject to greater toxicant exposure due to more extensive searching for its prey. Azinphosmethyl is a contact insecticide. Since the mites were exposed to the chemical for only two days, direct contact with the toxicant would seem to be more sig- nificant to the predator than the possibility of secondary poisoning. The claim that A. fallacis would encounter more toxicant in searching for its prey deserves consideration. Although prey populations (T, urticae) were maintained at high levels throughout the experiments, it was observed that the mobility of A, fallacis was greater than that of the more sedentary T. urticae. 35 The results of the actual versus predicted mortal- ity comparison are quite interesting in this context. The slide-dip technique does not allow for differences in mobility since all mites are held stationary throughout testing. One might then expect that the actual/predicted mortality ratio would be higher for A. fallacis since mites were free to move about in the actual selection experiments. In fact, the ratio was somewhat higher for T. urticae. Although other factors may be at work here, this is an indication that mobility is probably not a signi- ficant factor in the present study. An interpretation of Experiments 4 and 5 is quite difficult in view of the many factors involved in these experiments. Parental LCSO'S for the two species differ by a factor of 30 as in Experiments 2 and 3. Additionally, the distance between the dosage-mortality of the parental and resistant strains (from which the small numbers of resistant mites were added) is quite different for the two species (see Table A-7). Ideally, comparisons should be made between the two species over identical toxicant ranges. Although Experiment 4 is at present incomplete, it can be seen that the two species did not respond identic- ally in Experiments 4 and 5. The susceptible T, urticae strain was able to attain the LC50 of the resistant T. urticae strain in 6 generations. After 6 generations, the susceptible A, fallacis strain had failed to do so, 36 although it had made significant progress towards the LC50 of the resistant A, fallacis strain (a 51.63-fold increase in L050) and in fact had shown a greater absolute increase in LC50 level than did the T. urticae strain. The Belding R strain (A, fallacis) used in Experiment 4 had been in- tensively selected with organophosphates for approximately 10 years prior to use in the experiment. Possibly the in- ability of the susceptible A. fallacis strain to attain the LC50 of the resistant strain was due to genetic incompatib- ilities between the two strains. A future study mentioned previously will circumvent this problem by repeating Experiments 4 and 5 using the resistant strains developed in Experiments 2 and 3 for introducing resistant genotypes into the two susceptible populations. The main purpose of Experiments 1-5 was to test the hypothesis that the major obstacle in resistance develop- ment in a specific natural enemy (A, fallacis) is a food limitation following pesticide selection. This added stress (in addition to the direct chemical stress) upon the natural enemy, hypothetically limits natural enemy resist- ance development as compared to that achieved in pest species. Although Experiments 4 and 5 have failed to concretely eliminate the question of the initial gene frequency, the results of Experiments 2 and 3 do indicate that a natural enemy can develop resistance as quickly as a similar pest species (23.87-fold resistance in the natural enemy in 14 37 generations versus 20.21-fold resistance in the pest in 22 generations). These results may have practical signi- ficance in the applied aspect of pest-management. If economic thresholds for pest species were raised to maximum levels, the additional prey available for natural enemies (following pesticide selection) would allow maximum resistance development in the natural enemy. LIST OF REFERENCES 38 Abul-Hab, J.K., Stafford, E.M. 1961. Studies of the eggs of a strain of two-spotted spider mite, Tetranychus telarius, resistant to parathion. J. Econ Ent 54:591-5 Anonymous. 1968. Test Methods for Resistance in Insects of Agricultural Importance: First Conference. Bull Entomol. Soc. Am. 14: 31-7. Ballard, R.C. 1954. The biology of the predacous mite T hlodromus fallacis (Carmen) (Phytoseiidae) at 78g F. 0510 J. Sci. 54:175-9. Ballantyne, G.H., Harrison, R.A. 1967. Genetic and biochemical comparisons of organophosphate resist- ance between Strains of Spider mites (Tetran chus species: Acari). Entomol. Exp. Appl. 15:231-9. Bartlett, B.R. 1956. Natural predators. Can selective insecticides help to preserve biological control? Agr. Chem. 11:42-44. Boudreaux, H.B. 1963. Biological aspects of some phyto- phagous mites. Ann. Rev. Ent. 8:137-54. Brattsten, L.B., Wilkinson, C.F. 1977. Herbivore—plant interactions: mixed-function oxidases and second- ary plant substances. Science 196:1349-52. Bravenboer, L. 1959. De chemische en biologische bestrijding van de Spintmijt Tetranychus urticae Koch. (In Dutch: English summary). Diss. Landbouwhogesch. Wageningen, 85 pp. Brown, A.W.A. 1969a. Insect resistance: part II mechanisms of resistance. Farm Chem. Oct. 1969. 1969b. Insect resistance: Part III development and inheritance of resistance. Farm Chem. Nov. 1969. Burrell, R.W., McCormick, W.J. 1964. Typhlodromus and Amblyseius (Acarina: Phytoseiidaé) as predators on orchard mites. Ann. Entomol. Soc. Am. 57:483-7. Caegle, L.R. 1949. Life history of the two-spotted Spider mite. Tech. Bul. Virginia Agr. Exp. Sta. 113:1-31. 39 Cranham, J.E. 1974. Rep. E Malling Res. Stn. 1973. 159 pp. Croft, B.A. 1972. Resistant natural enemies in pest management systems. Span. 15:19-22. 1977. Resistance in arthropod predators and parasites. In: Watson, D. L. Brown, A. W. A. Pest Management &_Insect Resistance, 377- 93. Academic Press. , Brown, A.W.A. 1975. Responses of arthropod natural enemies to insecticides. Ann. Rev. Entomol. 20:285-335. , Hoying, S. A. 1976. Organophosphorous- res1stance and its inheritance in the predaceous mite Amblyseius fallacis. J. Econ. Ent. 69: 64- 68. , Meyer, R.H. 1973. Carbamate and organophos- phorous-resistance patterns in populations of Amblyseius fallacis. Environ. Ent. 4:691-5. , Nelson, E.E. 1972. Toxicity of apple orchard pesticides to Michigan populations of Amblyseius fallacis. Environ. Ent. 4:895-8. Dittrich, V. 1963. Investigation on OP resistance of two genetically differing populations of Tetranychus urticae K. Ent. Exp. Appl. 6:10-20. , 1972. Phenotypic expression of gene OP for resistance in two-spotted spider mites tested with various organophosphates. J. Econ. Ent. 65:1248- 55. Doutt, R. L. Smith, R. F. 1969. The pesticide syndrome - diagnosis and suggested prophylaxis. In: Huffaker, C. B. ed. Biological control. Plenum.P—ess, N.Y. Geier, P.W., Clark, L.R. 1961. An ecological approach to pest control. Proc. Tech. Meeting Inter Union Conserv. Nature Nat. Res. 8th, 1960, Warsaw, pp. 10-18. Gordon, H.T. 1961. Nutritional factors in insect resist- ance to chemicals. Ann. Rev. Ent. 6:27-54. Hansell, R.I.C., Mollison, M.M., Putman, W.L. 1964. A cytological demonstration of arrhenotoky in three méges of the family Phytoseiidae. Chromosoma 15: 5 -7. 40 Hoying, S.A. 1976. Taxonomic, distributional, and hybridization comparisons between Typhlodromus longipilus Nesbitt and Typhlodromus occidentalis Nesbitt (Acarina: Phytoseiidae) M.S. Thesis. Michigan State Univ. 52pp. Helle, W. 1959. The occurrence of resistance to organic phosphate compounds in the two spotted spider mite at Aalsmeer. Tijdschrift P1 ziekten 65:107-15. . 1965. Resistance in the Acarina: mites. Advan. Acarology 2:71-93. , Bolland, H.R. 1967. Karyotypes and sex deter- mination in spider mites (Tetranychidae). Genetica 38:43-53. , Pieterse, A.H. 1965. Genetic Affinities Between Adjacent Populations of Spider Mites (Tetranychus urticae Koch). Ent. Exp. Appl. 8:305-8. Herne, D.H.C., Brown, A.W.A. 1969. Inheritance and bio- chemistry of OP resistance in a New York strain of the two-spotted spider mite. J. Econ. Ent. 62: 205-9. Krieger, R.I., Feeny, P.P., Wilkinson, C.F. 1971. Detoxi- fication enzymes in the guts of caterpillars: an evolutionary answer to plant defenses? Science 172:579-81. Lee, T.D. 1972. Biological studies of Amblyseius fallacis and Typhlodromus longipilus (Acarina: Phytoseiidie), with particular emphasis on diapause. Ph.D. Thesis, Rutgers Univ., N.J. Luckmann, W.H., Metcalf, R.L. 1975. The pest-management concept. T2: Metcalf, R.L., LuCkmann, W.H., eds. 1975. Introduction to insect pest management. Wiley & Sons 587 pp. Matsumura, F., Voss, G. 1964. Mechanism of malathion and parathion resistance in the two-spotted spider mite, Tetranychus urticae. J. Econ. Ent. 57:911-7. McGlanahan, R.J. 1976. Influence of temperature on the reproductive potential of two mite predators of the two-spotted spider mite. Can. Entomol. 100: 549-56. 41 McEnroe, W.D., Harrison, R. 1968. The role of linkage in the response of an arrhenotokous Spider mite, Tetranychus urticae, to directed selection. Ann. Entomol. Soc. Am. 61:1239-47. , Kot, J. 1968. Evolution of Organophosphorous resistance and fitness in a hybrid swarm of the two- spotted spider mite, Tetran chus urticae. Ann. Entomol. Soc. Am. 61:I255-9. , Naegele, J.A. 1968. The co-adaptive process in an organophosphorous-resistant strain of the two- spotted spider mite, Tetranvchus urticae. Ann. Entomol. Soc. Am. 61:1055-9. McMurtry, J.A., Huffaker, C.B., Van de Vrie, M. 1970. Ecology of tetranychid mites and their natural enemies: a review I. Tetranychid enemies: their biological characters and the impact of spray practices. Hilgardia 40:331-90. , Scriven, G.T. 1964. 1964. Insectary production of phytoseiid mites. J. Econ. Ent. 58:282-4. Metcalf, R.L. 1975. Insecticides in pest management. In: Metcalf, R.L., Luckmann, W.H., eds. 1975. Introdfic- tion to insect pest management. Wiley & Sons 587pp. 'Mitchell, R. 1970. An analysis of dispersal in mites. Amer. Nat. 104:425-31. . 1972. The sex ratio of the spider mite Tetranychus urticae. Ent. Exp. Appl. 15:299-304. Mbtoyama, N., Dauterman, W.C., Rock, G.C. 1977. Toxicity of O-alkyl analogues of azinphosmethyl and other insecticides to resistant and susceptible predaceous mites, Amblyseius fallacis. J. Econ. Ent. 70:475-6. , Rock, G.C., Dauterman, W.C. 1970. Organophosphor- ous resistance in an apple orchard population of T§phlodromus (Amblyseius) fallacis. J. Econ. Ent. , N.C., Rock, G.C., Dauterman, W.C. 1972. Studies on the mechanism of azinophosmethyl resistance in the predaceous mite, NeoSeiulus fallacis. Pesticide Biochem. Physiol. 1:205-15 Mrak, E. (Chairman). 1969. Report of the secretary's commission on pesticides and their relationship to environmental health. U.S. Department of Health, Education and Welfare, Washington, D.C. 42 National Academy of Sciences (NAS). 1969. Insect-pest management and control. Pub. 1695. Washington, D.C. Overmeer, W.P.J. 1966. Intersterility as a consequence of insecticide selection in Tetranychus urticae Koch (Acarina: Tetranychidae), Nature 20 : , Harrison, R.A. 1969. Notes on the control of the sex ratio in populations of the two-spotted spider mite Tetranychus urticae Koch (Acarina: Tetranychidaé): ‘New zeaIandFJ. of Sci. 12:920-8. Plapp, F.W., Bull, D.L. 1977. Toxicity and selectivity of some insecticides to Chrysopa carnea, a preda- tor of the tobacco budworm. (unpub.). Rock, G.C., Monroe, R.J., Yeargan, D.R. 1976. Demonstra- tion of a sex pheromone in the predaceous mite Neoseiulus fallacis. Environ. Ent. 5:264-6. , Yeargan, D.R., Rabb, R.L. 1971. Diapause in the phytoseiid mite Neoseiulus (T.) fallacis. J. Insect Physiol. 17:165I-9 Saba, F. 1960. Uber Entwicklung und Rukgang der Giftresistenz bei Tetranychus urticae Koch und deren Abhangigkeit von derWirtspranze. Dissertation, Universitat, Bonn. Schulten, G.G.M. 1966. Genetics of resistance to parathion and demeton-S-methyl in Tetranychus urticae Koch (Acarina). Genetica 37:207-17} , 1968. Publs R Trop Inst. Amsterdam 57:1. Smissaert, H.R. 1964. Cholinesterase inhibition in spider mites susceptible and resistant to organophosphates. Science 143:129-31. Smith, J.C., Newsom, L.D. 1970. ‘The biology of Amblyseius fallacis (Acarina: Phytoseiidae) at various temperature and photOperiod regimes. Ann. Entomol. Soc. 63:460-2. Smith, R.F. 1970. Pesticides: their use and limitations in pest management. TE: Rabb, R.L., Guthrie, F.E. eds. Concepts of pest management. North Carolina State Univ., Raleigh. Watson, D.L. 1956. Some genetic aspects of resistance to parathion in the two-spotted spider mite, Tetranychus telarius. Thesis, Cornell Univ. Wrensch, Taylor, 43 , D.L., Naegele, J.A. 1960. The influence of selection pressure on the development of resistance in populations of Tetranychus telarius (L). J. Econ. Ent. 53:80-4. D.L., Young, S.S.Y. 1975. Effect of quality of resource and fertilization status on some fitness traits in the two-spotted spider mite. Tetranychus urticae Koch Oecologia (Berlin) 18: B.A., Smith, F.F. 1956. Transmission of resistance between strains of two-spotted Spider mites. J. Econ. Ent. 49:858-9. Van de Vrie, M., McMurtry, J.A., Huffaker, C.B. 1972. Voss, G. Zahavi, Ecology of tetranychid mites and their natural enemies. III. Biology, ecolOgy and pest status and host—plant relations of tetranychids. Hilgardia 41:343-432. , Matsumura, F. 1964. Resistance to organo- phosphorous compounds in the two-Spotted Spider mites: two different mechanisms of resistance. Nature 202:319-20. M., Tahori, A.S. 1970. Sensitivity of Acetyl- cholinesterase in Spider Mites to Organophosphorous compounds. Biochem. Pharmacol. 19:219-225. APPENDIX A FURTHER FIGURES 44 FIGURE 3 Dosage-Mertality Lines for Experiment 1: Selection of an A, fallacis population through 7 generations. 45 A.H.c Fzmummmu "co. moqwoo fiooo. _ doooo. —r on d «5 11mm Juom Jump 11mm AlIJUlHOW 1N3383d 46 FIGURE 4 Change in LC50 with Time for Experiment 1 fh mmmzaz zo_emH mo. um unmoamaawwmi 76 «.mn o.mn Hmmm owmum>< «q.o . u u o~m.~ ooooo. ooao. u n N¢.N «.mw m.Hw omo. «Ha.m Nmooo. mmao. mo¢¢ o no.m o.¢n m.Hm mmo. Ham.e mmooc. mmao. qwmm m «N.m m.mo 0.00 nmo. moo.m NNHoc. «owe. comm q «o.¢ n.0w o.mn mmo. qNN.m moaoo. mnao. NnHm m sam.m o.~w 0.55 ouo. non.~ Hoooo. Hoao. moo¢ N - H.~w - «no. - - - mama AHEV H wm.a a w.mn qmo. Hmm.q oqaoo. mqmo. : Amv o mumDVm aMwwme mmwmwszwomwwmwv A.H.< NV AwooH\nHv muwm nmnabz Hmsuo¢ uowpmum hmumm mmon omuq onuq cowumanmom cowumuoomu *«IHSU Amflomaamm .flv H ucoaanmmxm How muma aoauooamm HI< m4muHHmuHoz couowpwum paw macaw .omoq ou mofiamaM¥ Hm>uH mo. um unmofimaamam« o.H5 m.mm ance mwmuo>< swm.NN m.mo H.w¢ coo. New.a oqmmo. mqmq. mmom qa n ¢.¢m . oom. . u u mma5 ma mo.m o.~5 H.¢q ooq. Nmo.H mqmmo. «qu. aqom NH u N.qw u ooq. u u u mmHo HH mm.m c.m5 <.ow ooq. mom.H 5cho. m5wo. mnNm 0H ¥5w.o m.mn ~.mm oom. 5HN.H wowao. Hmom. 5m~m m Ho. «.05 5.m¢ oom. moo.a «HNHO. omom. HOH¢ w um.¢ m.~5 5.0m mNH. mmm.H NM5oo. ONNH. mmom 5 ¢H.m ¢.H5 N.5¢ mNH. wa.H wHwoo. qwma. mwom 0 mm. H.¢5 o.~m m5c. Hmm.m m~¢oo. mH5o. mmm¢ m 5N.m 0.05 «.m0 050. 5¢5.H 5HNoo. memo. mmmm q *mw.m w.mm m.~m oqo. mq5.H mmmoo. 05mo. ¢¢am m oo.m m.wo m.N5 oqo. amm.a mmooo. «mac. momq N em. ¢.oo H.mo mso. Hm~.~ Hoaoo. acme. ommm Aamv H *H¢.w H.0m H.mm oao. mM5.H 5oaoo. w5Hc. u Amv o Axvsua Axvsufi AwOOH\nHV A.H.< NV Awooa\nHv wumdvm lamuuoz IHMuHoz Hm>oq onwm Hmnaaz «suwsu Hm5u0< uowpmnm hmuam macaw omoq omoq cowumaamom cowumuoamu Amfiomaamm aflv N quEHHmme How mama cofluooamm Nn< mqmoH mo. um unmoHMchHm¥ H.05 ~.om m55q mwmum>< *mm.NN o.m¢ m.nH oo.m NoN.m mowHo. mHom.oH mMNq NN «om.oN m.55 m.mN oo.¢ mmm.N ¢MNo¢. omo5.o Hm5N HN H5.o 5.o5 H.mN oo.¢ N5w.m Hm5mm. ommm.m Hmmq oN 05.0 o.m5 o.Nm oo.¢ wmm.m oomNm. 5oN¢.m mmmm mH «w¢.¢¢ 5.Hw 5.om om.¢ N5o.m HNqom. No5m.o HNmN mH mw.m u w.m¢ mN.m mom.N oomNN. mHm5.m oqu 5H «mw.om o.mo 5.¢N oo.¢ oN¢.H oooN5. oHoH.NH 5nmm oH «mw.H¢ u w.o¢ om.N mHo.N 5N¢oH. m5m5.N u mH u n u oo.N n u u u «H mumnom uMWWMwW nflwwmwm AwmwMMmHv A.H.< NV AwooH\nHv mNHm Honsoz sanso Hanuo< HUHoon hwumm mQOHm omUH omoH cowumHamom cowumumcmo Hemsafluaouo m-< mqmoH mo. um unmoHMchHm * *qo.o . - u Nmo.m NHmoq. on5.5 u *¥*m n.50 o.¢m mmos mwmum>< - 5.om u oo. . n u m¢o¢ o sow.5 5.mm m.mm om. omw.H oq5qo. oom5. mon m om. N.m5 5.H¢ oq. «@5.H N5Hmo. 5on. moon q mN.¢ m.m5 o.mN om. om5.H «5Noo. mNH5. omom m n s n oH. s a n a N Ho. - - so. own. Nquo. ammo. - Ammo H Nm.m w.¢o H.¢m No. mHH.N 5mHoo. NHmo. oqqq AHmv o N5. n - u mmH.N Nmooo. mnHo. . Amy oo HHVNuH HNVNHH HwooHHHHV A.H.< so Hwo0H\nHo mumsvm uHmuuoz uHmuuoz Hm>oH ONHm Honasz *sano Hmsuo< uoHpmum mmuom maon omUH omoH :OHumHnmom coHumuocmo q ucmEHuoaxm How mama GOHuoonm ¢u< m4m<fl 81 maoHumumcmw Hm vow m oucH oopom omoHuHD am ucmumHmom Mom maHH muHHmuuozuomeoa kuHHmuuo: wouoHvon paw mmon .omoH ou mmHHmm<4HH Ho>mH no. um pamonHame sm5.HH u n . omm.N 5¢wHo. m5om.oH u «sax N.m5 5.Nm oomm mwmum>< «om.HH : u u dmm.m mmmmo. wmmm.oH u o «mm.5H 5.mm 5.HN oo.o 5oH.m «doom. ooo¢.w m5mm m wo.¢ q.oo m.Nm oo.o HHm.¢ 55o¢¢. No¢¢.5 mmH5 q ¥N5.mm m.om m.om om.¢ mmH.N ooomm. mm¢m.m mw¢m m «Nw.MH w.oo m.om oo.m 5Nm.H Nqomq. 5m5N.5 mNom N mH. m.m5 o.oN om.H 5Hm.H me5H. momw.N NN¢m Ava H mm.m m.Hm «.ms om. N¢¢.H oNHNo. msos. HOHo HHHV o «mm.m u u . ooN.N mNomo. moon. I Amv oo unwavm uMwWMwW -MWWMWm AwmwMMmHv A.H.< NV “wooH\nHV oNHm Hmnabz *«ano Hm5u0< uoHpoum mmuam wQOHm omoH omUH nowumHsmom GoHuwHoaoo m quEHuomxm pom mama aoHuooHom mn< mHm