Insect-Pest Management and Control (1969)

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.

CHAPTER Ecological Basis for Control Since the late 1940's, many workers concerned with the control of crop pests in North America and elsewhere have strongly advocated a more fundamental approach to pest-control problems. The limitations of chemi- cal control were recognized, as well as the fact that control measures could not be significantly improved without ecological knowledge of the species to be controlled. As a consequence, much-needed emphasis has been placed on ecological studies basic to an understanding of the causes of pest outbreaks and of the requirements for long-term control of the pests. Discussed in this chapter are (1) the special nature of crop ecosystems (agroecosystems) in which pest outbreaks develop, (2) special attributes of pest populations at or near epidemic levels—levels that normally warrant control, (3) the bio- metrical techniques required to measure pest-population parameters, and (4) the practical and scientific application of results of long-term studies on the dynamics of insect-pest populations in agricultural ecosystems. THE ECOLOGICAL APPROACH The study of a pest population in a crop ecosystem, in relation to chemical, biological, cultural, physical, or integrated control practices, must be con- sidered in a study of fundamental relationships among host plant, pest popu- lation, and related biotic agents, soil, climate, and the management practices of man. Soils, host plants, and management practices, singly or combined, are relatively stable in the maintenance of crop stands in full production, but pest populations and related natural control factors—parasites, predators, diseases, 48

ECOLOGICAL BASIS FOR CONTROL 49 and weather—fluctuate dynamically from time to time and clearly contain clues to pest-population increases and decreases. An exception may be the distribution of an insect-resistant crop variety such as wheat resistant to the Hessian fly, Mayetiola destructor (Say) (see Chapter 6). Insight into the primary causes of pest outbreaks can therefore be gained through detailed studies of extrinsic and intrinsic factors acting on the pest during the rise and fall of local or widespread infestations. Such a study must be long-term, so that exact relationships between mortality factors and host population within and between generations can be elucidated. This approach may ultimately provide growers with a practical means of pest control and science with a fundamental explanation of the underlying biological phenomena. AGROECOSYSTEMS Ecosystems are self-sufficient habitats where living organisms and the non- living environment interact to exchange energy and matter in a continuing cycle. Descriptions of such systems have varied slightly in the past only be- cause of the different conceptions of the systems' space and time limits, the components involved, and the various approaches taken in their study. Agroecosystems, or crop systems, are special situations; variations here are caused mainly by the nature of the crop environment, the host plants and arthropod species that are present, space and time boundaries, and the distri- bution of a species in a crop area during the rise and fall of a local epidemic. These systems are relatively artificial but stable in nature. Homogeneity of crop conditions causes the artificiality, while regularity and uniformity of management practices create stability. The host plant and soil-substrate components may sometimes limit the development of pest populations, and some mortality factors, extrinsic and intrinsic to the populations, can limit pest numbers from time to time. Agroecosystems usually contain a few common or major species and numerous relatively rare or minor species. In apple orchards, for example, the ratio of major to minor species is approximately 6:150. In any outbreak, usually only one pest species, most often a major species, is present in high densities, and the other species are relegated to such low numbers that they cannot be effectively sampled. Consequently, the study of a pest outbreak in an agroecosystem is in reality the study of one pest species in relation to natural control factors, or in relation to man-made factors superimposed on natural control factors, and this constitutes a study of the variable compo- nents of the system under these conditions. Such a system is a much simpler study situation than is frequently envisaged, and it is ideally suited to the population-dynamics approach.

50 INSECT-PEST MANAGEMENT AND CONTROL PEST POPULATIONS Because, in the control of crop pests, concern is chiefly with the species as it behaves at the population level, this is the level at which to seek solutions to pest-control problems. Pest populations are composed of groups of individuals, in one or more stages, of a single species; the groups occupy a particular habitat and have characteristics peculiar to the group rather than to the individual. Major population characteristics, such as density, birth and death rates, age distribution, biotic potential, and growth, are biological attributes of the popu- lation as a whole and are meaningful only because the life stages of the popula- tion—the egg, larva, pupa, and adult—are interdependent and interrelated. Be- cause these stages are the dynamic units of the population, and collectively have definite organization and structure, they can be described and expressed as statistical functions. Thus, to obtain fundamental information on the population dynamics of a species, it must be realized that individuals in one stage are not isolated from individuals in other stages and that they cannot survive independently of the population; in other words, there is no sharp break between any two stages in a population. Therefore, eradication of a pest by controlling only the damaging stage is difficult, because some of the individuals usually survive to assure continuity of the species. Control studies have often been short-term and concerned mainly with the damaging stage of a pest population, usually the larval stage, and entomolo- gists have therefore become accustomed to think of the stage and not the population as the ultimate unit in pest control. But it is necessary to know not only what effect chemicals, parasites, predators, resistant crop varieties, and any other control measures, singly or combined, have on a given stage, or even on a limited number of stages of a pest population, but also the effect of each on all stages and indirectly on population trend. However effective a factor may be against a given stage of a pest species, its value is limited if it does not account for significant decreases between generations. To know that a factor kills 95% of the eggs or larvae of a pest tells little if the extent of the effect of such a factor on population regulation is not known. POPULATION MEASUREMENT A study of the primary causes of pest outbreaks includes the measurement of population changes from generation to generation and the study of the factors responsible for these changes. A quantitative approach to such studies

ECOLOGICAL BASIS FOR CONTROL 51 is now possible through the use of recently developed sophisticated sampling techniques that provide for the collection of data of the required quantity and quality. Multivariate statistical techniques can be used for the analysis of the action and interaction of the many population parameters that are measured, and systems analysis and allied computer techniques permit realis- tic and precise analyses of whole systems and not just fragments of them. These techniques are specifically suited to handle both the magnitude and kind of complexity found in agroecosystems. Their use should provide the key to a precise understanding of these ecosystems and to the intelligent and practical manipulation, for control, of mortality factors or husbandry practices in crop habitats. In pest-population studies, the first concern is with the interaction of pest stages and their mortality factors within and between generations. For knowledge of this, a population-dynamics study has priority and is basic to the correct interpretations of the action of mortality factors, singly or com- bined, in the population regulation of the species. Specifically, basic require- ments for the adequate measurement of the parameters involved are (1) that confidence limits be biometrically established for all population data collect- ed; (2) that all stages of the pest population, and related mortality factors, be measured and appraised in terms of a common unit; and (3) that such data be collected for several generations of a species, during the rise and fall of an infestation. To analyze collected data, and to determine their long-term population significance, a series of life tables should be prepared. The life table is a useful numerical aid or device used in the study of insect populations to record in a systematic fashion those facts basic to the age distribution of mortality. Con- ventionally, it consists of data and calculations arranged in a series of columns, from left to right, under the following symbols: x — the age interval. Convenience usually dictates that these intervals correspond to development stages of the insect. 1 - the number alive (1) at the beginning of the age interval noted in the x column. d F — the factor (F) responsible for the death of individuals (dx) within each age interval. dx - the number dying (d) within the age interval stated in the x column. 100<7 ~ percentage mortality (dx as a percentage of 1^). lOOd IN, - the percentage of generation mortality (d as a percentage N, — the number of eggs observed in the present generation. N - the number of eggs observed in the next generation.

52 INSECT-PEST MANAGEMENT AND CONTROL An example of a life table for one generation of the eye-spotted bud moth, Spilonota ocellana (Denis & Schiffermiiller), on apple in Quebec, is illustrated below: d* 100», 100 Parasites 30 30 30 Predators 10 10 10 Larvae 60 Frost 55 92 55 Pupae 5 Parasites 3 60 3 Adults 2 (Sex ratio = 50:50;* eggs per o = 100) Eggs (W2) 100 Population trend (index) = Index: Unity = population constant > 1 = population increase < 1 = population decrease From these tables may be determined: (1) the initial density and survival rate within each stage of the pest population, (2) the mortality factors pres- ent in each stage and their effect, and (3) the survival rate after application of insecticide to each stage. More important, from such a series it is possible to determine biometrically the stage and the factors in the stage that are most responsible for the increases and decreases in pest numbers between and with- in generations and that warrant further experimental and field investigation. A population-dynamics study also reveals information of immediate practical value: emergence dates and duration of economically important stages, of use in chemical control; the degree of crop injury in proportion to pest density, of use to the grower; and forecasts of densities of pest popula- tions, based on the index of population trend, of use to the grower in planning spray programs. Finally, this approach to the measurement of pest-population parameters should result in equations that describe observed data. The use of such equations, rather than those that presume certain modes of action, offers the best hope of expanding an understanding of animal processes. Simply stated, such equations are predictive equations in which the dependent variable may be population density N at some future time t + 1, and the independent variables are density at an earlier time t, along with weather, predation, para- sitism, diseases, and other factors that determine the rate of change in the population between t and r + 1. MULTIFACTOR STUDIES Detailed studies carried out from 1953 to 1968 on the population dynamics of 11 Canadian crop insects have provided basic information for the practical

ECOLOGICAL BASIS FOR CONTROL S3 application of ecological principles in the management and control of the pests. Some of these studies covered as many as 18 generations of the pest species. Sequential measurements were made of the stages and mortality factors in each generation of the pest and were transferred to life tables. For each generation, the estimated numbers of eggs, larvae, pupae, and adults were obtained by measurements made under natural conditions, that is, in crop fields untreated with insecticide. All estimates were based on a standard sample unit and conformed to predetermined confidence limits. Rates of predation, kills by weather and other factors, and percentages of parasitism were derived from frequent field observations or from extensive rearings of host material. Sex ratio of adults was obtained from the pupae, and degree of fecundity was obtained from adults reared under field conditions in the absence of natural mortality factors. Adult mortality was based on an indirect measurement representing the difference between the initial egg population observed in the field and the number of eggs expected on the basis of egg potential of adults of the previous generation. A population increase, de- crease, or equilibrium was measured, as illustrated previously, as the number of eggs laid in one generation divided by the number laid in the preceding generation. When a sufficient number of life tables, replicated in time and space, had been compiled for a species, a simple correlation analysis was carried out to show which stage (critical age interval), and which factor (key factor) within that stage, contributed most variation to population trend. Species studied were as follows: Species Origin Generations Studied Critical Age Intervals Key Factorsa North America 6 Fruit-tree leaf roller, Archips argyrospilus (Walker) Diamondback moth, Europe Plutella maculipennis (Curtis) Imported cabbageworm, Europe Pieris rapae (Linnaeus) Pistol casebearer, North America Coleophoro malivorella (Riley) Eye-spotted bud moth, North America Spilonota ocellana (Denis & Schiffermuller) European corn borer, Europe Ostrinia nubilalis (Hu•bner) Colorado potato beetle, North America Leptinotarsa decemlineata (Say) 18 18 Pupa, adult Adult Larva Larva Larva Adult Larva Parasitism, migration Weather Disease Parasitism Weather Migration Food supply

54 INSECT-PEST MANAGEMENT AND CONTROL Species Origin Winter moth, Europe Operophtera brumata (Linnaeus) Oystershell scale, Europe Lepidosaphes ulmi (Linnaeus) Apple leaf miner, Europe Lithocolletis blancardella (Fabricius) Birch leaf miner, Europe Fenusa pusilla (Lepeletier) Generations Critical Age Studied Intervals Key Factors" Pupa Parasitism Egg, adult Predation, parasitism Pupa, adult Predation, weather Larva, adult Predation, migration "All the factors except weather are density-dependent; weather is density-independent. KEY REGULATING FACTORS Analyses of life tables for the species listed showed that the critical age interval could occur in any stage of development and that only one or two key mortality factors (or agents) within the critical stage accounted for regu- lation, or changes in population trend. These factors varied from generation to generation; this accounts in part for their importance as population regula- tors. Mortality factors other than those in the critical age interval were low and constant and did not contribute significantly to population changes. Key factors that affected eggs and larvae in six of the species were of twofold importance: they regulated the population, and they accounted for high kills before or during economically important stages. Key factors in the adult stages of five of the species varied and caused lower mortalities, yet their effect on population regulation was just as great. In 9 of the 11 species studied, the key factors were density-dependent and therefore regulatory; that is, they acted severely against the population when pest density was high, and less severely when density was low. Results con- firmed what had been expected: as a rule, biotic factors were density- dependent and physical factors were density-independent. The population of each species increased and decreased at the same time in all areas of the species' range. For instance, the eye-spotted bud moth was equally abundant in southwestern Quebec and 300 miles east (lies aux Coudres), and was then reduced by frost (-29°C) in both areas at the same time. The life tables of this species showed only this key factor. Hence, frost apparently regulates populations of this species in all major apple regions of Quebec.

ECOLOGICAL BASIS FOR CONTROL 55 The implication of such findings is that, if the successful manipulation of key factors for population regulation is subsequently made possible through sound studies of the components involved, the results would have very broad application. Findings regarding crop-plant insects could be considerably modified, the extent of modification depending on the resistance (including tolerance) of the crop varieties on which the insects are feeding. Resistance is especially well-studied in the European corn borer, Ostrinia nubilalis (Hiibner), where larval population reductions in excess of 60% have often been recorded on resistant compared with susceptible varieties in the same tests. Resistant varieties can also have extensive effects on insect and protozoan parasites and have influenced efficacy of chemical control where such relationships have been studied. Therefore, the ecological interrelationships of crop varieties, which can be changed yearly by the grower, can be extensive. Bio tic Population densities of 7 of the 11 pests were regulated by insect parasites, predators, or diseases. The number of hosts attacked by parasites or predators depended not only on host density, but also on the density of the agent and frequency of attack. Thus, where these key biotic agents are removed by pesticides, an upsurge in the pest population occurs. By simulating responses involved in predation, it was shown that the number of prey consumed de- pended on prey density, rate of discovery, hunger level, and time spent by the predator in consuming prey. Similarly, a functional response was implicit in the investigations of all 7 species. Since none of the parasites or predators regulating the populations of these pests had been introduced to control them, it is apparent that under certain circumstances resident biotic agents can effectively control major crop pests. The winter moth and birch leaf miner are recent introductions to Canada. Considerable resistance to their establishment in that country was offered by resident parasites and predators: mortalities of 98 percent or less in a generation at first permitted some population increase, but higher mortalities subsequently reduced populations. In-depth studies of the components involved in the manipulation of these beneficial biotic agents are a logical next step, because the importance of these species in the suppression of pests cannot be questioned. A capsule virus, i.e., a granulosis virus encapsulated in a crystalline con- tainer, was the key factor that suppressed field populations of the imported cabbageworm. The death rate from the disease caused by the virus was an increasing function of host density. The practical application of this informa- tion is under study in control of the cabbageworm in large crop areas of southwestern Ontario. Disease-producing organisms such as this virus are

56 INSECT-PEST MANAGEMENT AND CONTROL usually density-dependent, because the denser the population of the host, the more easily transmission from environment or from diseased insects to healthy ones occurs, and in dense populations the microorganisms are able to increase at a more rapid rate and in greater numbers than do the host insects. Abiotic Weather was a key factor in two of the species studied, and food supply in one. Weather is a key mortality factor independent of density. Under its favorable influence during several seasons, or, in the case of multivoltine species, during a single season, a pest may build up to outbreak numbers and fluctuate at epidemic levels. Thus, in a local situation, for instance, near the distributional limits of a species, climatic factors may obscure the action of density-dependent agents that operate concurrently but at a lower level of mortality. These agents may well be regulative in more-typical parts of the range. As an example, in eastern Canada the diamondback moth survives only in the spring and summer and fluctuates in numbers according to the type of weather that prevails during the flight period; in South Africa, its numbers appear to be regulated by density-dependent factors throughout the year, because the weather there is rarely adverse. A much different situation prevails for the bud moth: frost regulates the species over long periods and over large areas. In relation to regulating factors, it should be pointed out that the simple categorizing of factors as density-dependent or density-independent does not per se determine the effect of a key factor on population trend, but rather it is how sensitive the pest is to variations in the key factor that is important. In the case of the bud moth, the further very important action of frost in selection of frost-resistant strains of the species, and thus on subsequent density of the pest, cannot be discounted. In field populations of the Colorado potato beetle at Ottawa, Canada, intraspecific competition for food was the key factor. Mortality agents are few in potato fields, and in most years the available food supply governs the number of progeny reaching adulthood. Upon depletion of the food, the partly fed larvae simply starve or, having limited powers of dispersal, wander about ineffectually until they die. Migration Migration was a key mortality factor in three of the species studied and generally resulted from overcrowding; that is, it was density-dependent. Migrating insects successfully re-established themselves only in new areas where food sources and breeding conditions were suitable; heavy mortality occurred where food was scarce or inferior. At high population densities, foliage-feeders may decrease their food supply with time, so that food

ECOLOGICAL BASIS FOR CONTROL 57 becomes scarce when the heaviest feeding occurs and most of the repro- ductive material is being laid down in the insect's body. A proportionately greater reduction in egg substrate and a corresponding decrease in the egg load may follow. This increases the ability of females to fly and the distance they travel. Maximum egg loads are developed at low population densities, and females are forced to lay their eggs in feeding areas. Apparently, this type of migration is essentially a homeostatic response that, in the case of crop species, adjusts the pest population in advance to the capacity of the crop habitat. Insecticides applied to control the injurious stage of the pest can have no real effect on the total population if they are not applied in areas from which the insect has migrated. Otherwise, control may be more efficient if based on naturally spreading pathogens directed against the larvae, or on the release of sterile males or sex attractants directed against adults. POSSIBLE REGULATING FACTORS Other mortality factors may be of significance in population regulation; they act in a density-dependent manner against certain crop pests but have not yet been shown to play a critical role. Vigor An example of the action of increased population density on insect vigor is that of the western tent caterpillar, Malacosoma pluviale (Dyar). In a favorable environment, the colonial habits of the larvae promote survival of the weaker individuals, because the stronger adults migrate as usual while the weaker individuals remain and oviposit nearby. Thus, progressively weaker colonies are produced locally by an increasingly sluggish resident population, and their numbers swell to the point where they far outnumber the few stronger colonies that remain. Ultimately, the colonies become too sluggish to reproduce and are suddenly eliminated by the first season of in- clement weather. Colonies of intermediate vigor are also lost, because their weaker members succumb and those that remain are not sufficiently numerous to provide the silk and mass of insulating bodies required for survival in the now more rigorous environment. The disappearance of the less viable colonies from the population increases the relative number of active colonies. The resulting improvement in population quality leads to an increase in density and thus completes the cycle. The action of vigor in this way suggests population regulation by a homeostatic device that manifests itself in the absence of extrinsic mortality factors.

58 INSECT-PEST MANAGEMENT AND CONTROL Fecundity The influence of population density on fecundity was shown to be signifi- cant in spruce budworm, Choristoneura fumiferana (Clemens), populations. Below a certain population pressure, maturing larvae fed solely on the current year's foliage of the host tree, but at higher densities they were forced to feed on old foliage, which caused undernourishment of late-instar larvae and a marked reduction in fecundity of the adults. Therefore, reduced fecundity was an increasing function of population density. Fecundity is greatly in- fluenced by crop varieties on which insects feed. When a crop variety is insect-resistant, known effects on insects range in various examples from a very slight effect to entire suppression of the production of young or eggs. Behavior In some grasshoppers, the interaction between individuals of a species leads to the emigration of part of the population. The change from the solitary to the gregarious form is apparently initiated by encounters between individuals, and the encounters are more frequent as the size of the popula- tion increases. Repeated encounters are habit-forming, and individuals learn to aggregate. Gregarious nymphs are more active and excitable than solitary nymphs, and the wings of gregarious adults are stronger. When the gregarious tendency has intensified to the point where ever-enlarging assemblies of these forms occur, whole swarms take off in a sustained migratory flight. The re- maining population then reverts to the solitary nonmigratory phase. Similarly, self-adjustment of numbers is obvious in highly evolved social insects such as the termites, wasps, bees, and ants. The queen alone is respon- sible for reproduction and regulates her oviposition according to the quantity and quality of the food she receives and the density of the colony. If crowding occurs, population balance is restored by egg-eating, fratricide, or expulsion of supernumerary members. Competition Laboratory studies of the cabbage looper, Trichoplusia ni (Hiibner), showed that stress of competition for food and space in a crowded environment has a deterrent effect on population growth. Larvae of the looper have preferred feeding sites and are habitually aggressive. Stress of competing for limited feeding space led to reduced food consumption, sporadic feeding, excessive dissipation of energy, and eventually to increased susceptibility to disease. Larvae often died from infection that normally would be nonfatal.

ECOLOGICAL BASIS FOR CONTROL 59 Genetic Factors Too little is known of the effect of genetic factors in the insect on population regulation. To what extent do such factors cause mortality? There is no doubt that susceptibility of a species to mortality factors is related in some degree to the number of individuals with poor genotypes within that species. A poor genetic constitution, for example, one that lacks somatic vigor or viability, may be expressed in many ways. For instance, eggs are too few in number, of too low viability, or too exposed to predators; a weak larva falls from a branch and is killed; a slow-moving caterpillar is overtaken by a predator; or a mature larva, like that of the eye-spotted bud moth, fails to build a sufficiently pro- tective shelter on apple and is killed by frost. Genetic drift or Sewell Wright effect also is certainly a factor in insects such as grasshoppers, in which the population may contract for a period into restricted colonies. These adverse situations increase the susceptibility of a species to elimination. However, strongly elusive individuals with a high reproductive capacity, and efficiency, have the opposite effect, that is, population survival within the generation and possible population increase from generation to generation. Furthermore, variability, the outward evidence of a large gene pool, is also evidence of possible genetic change, and change is inherent in population dy- namics. The selective presence of key factors is therefore directly or indirectly connected with the genetics of a species and thus with population change. USE OF MATHEMATICAL MODELS Rapid progress in the physical and chemical sciences has been brought about by an intimate feedback between theory and adequate mathematical models based on experiment and measurement. If the science of ecology is to become more than merely descriptive, or at best correlational, similar methods must be used; and mathematical models, based on sound multi- factor studies of the type reported in this chapter, must be developed for natural insect populations. The time is past when purely theoretical deduc- tions can further the understanding of natural populations, when qualitative conclusions about natural populations are sufficient, and when disconnected bits of information, however quantitative, about the effects of a factor on a population can be accepted as an understanding of that population in the absence of data on other factors and their interaction. The quantification of all population parameters for pest species is needed, since without math- ematics we can scarcely begin to think about entities that have more than a few variables. Mathematics provides a quantitative description of events that

60 INSECT-PEST MANAGEMENT AND CONTROL occur during the rise and fall of a pest population and permits a calculation of the consequences of purposeful alterations of certain population para- meters. Thus, there are two main reasons for the study of natural populations of crop pests by the use of life tables and mathematical models: the desire of the economic entomologist to learn how to manage pest populations on a scientific and optimum rather than on an ad hoc basis; and the desire of the population theorist to gain field data to test his theories and to bring to them the new conceptual strength that is possible only through the integration of theory and data. Mathematical models are really mathematical statements that make biological sense, that attempt to mimic numerical changes taking place in natural populations, and by which quantitative predictions can be made. Such models, based on life-table data and key-factor information, will be used to predict pest-population density at least one generation in advance. This information will be extremely helpful in deciding the necessity for control and in the planning, if necessary, of the logistics of large and often complex operations involved in the application of insecticides or the utiliza- tion of the sterile-male technique. The comparison of observed population density or survival rate of an insect pest with the values predicted by a model makes possible the calculation of the proportion of variance in the system that is explained by the model. This is the only scientific method of demon- strating how much or how little is understood about the population dynamics of a species. Explanations of population behavior, stress, vigor, and other factors that fail to show how the models for these factors fit the observed facts, and the degree of predictability achieved, are unaccept- able. Models will also be used to assess the form and degree of interaction by mortality factors within and between stages of a pest species, and their effect on population trend. Mathematical models based on field populations, and developed by multiple regression techniques, provide estimates of these inter- actions and the variance that interaction contributes. Under controlled conditions, where individual variables are studied separately, interactions cannot be measured. A third use for such models will be to calculate and test optimum tactics and strategy for pest management. As adequate models become available, many approaches should be possible, within mathematical constraints dictated by economics, that would permit the manipulation of key factors regulating population change. For example, it should be possible to lower the mean population density of a pest and reduce the frequency with which it escapes control by natural enemies or threatens its food supply; to determine when applied control measures are necessary and what control will interfere the least with natural enemies and other species; to determine weak spots in the life cycle of a pest or suggest new methods of control; and to integrate

ECOLOGICAL BASIS FOR CONTROL 61 into an optimum form of population management all factors and practices that can be manipulated for pest control. Finally, if dependent or independent variables well correlated with population changes are known, increased insight into the operation of the components of such factors can be obtained in the laboratory, where models can be refined from a descriptive to a more explanatory basis. The feedback now established between computer and theory is proving to be most effective in the development of models that will become more and more descriptive. All ecologists agree that present models are only refined deductive models and are therefore subject to the same criticisms. However, if the limitations are recognized, the deductive-inductive approach to the building of models should be effective in suggesting unsuspected relations of pest-population parameters that warrant further experimental and field investigation. PRACTICAL APPLICATION Results of ecological studies on crop-pest populations have practical appli- cation in: (1) The prediction of the population level of each pest from generation to generation. This information, based on trend index data, helps to guide the growers in effective advance planning of spray programs and in more intelligent use of pest-control measures. (2) The feedback of information on the economic thresholds of a pest and on the crop damage caused by the active immature stage. For example, it is now known that five bud moth larvae, or fewer, per 100 leaf clusters on apple do not con- stitute an economically important population. Hence, sprays are omitted at these densities of the insect, because foliage or fruit damage is below the 5% economic level generally tolerated by the industry. In the spring, chemical control of the bud moth is omitted if winter lows have been -21°F or below, because such lows kill 90 to 95% of the overwintering larvae. However, these lows scarcely affect larvae of the pistol casebearer, which overwinter near the bud moth and receive similar protection, and the omitting of spray applica- tions against the bud moth favors the survival of parasites that control and regulate populations of the casebearer. (3) The utilization of information gained through ecological studies to govern, when possible, the use of insecticides as a possible substitute for other density-independent control factors, such as weather, in integrated pest management practices. The use of a resistant variety instead of insecticides can often achieve the same results, with obvious advantages. With even a low level of resistance, the persistent and cumulative effect can give control when combined with other ecological factors. The substitution of a tolerant for a susceptible variety can

62 INSECT-PEST MANAGEMENT AND CONTROL permit increase in parasites, with decreased economic level of damage. The complex interrelationships will require investigations in each specific case. For the elimination of risks inherent in agricultural pursuits, crop-pest con- trol measures must be integrated according to scientific ecological principles. Pest-control technology has become too complex for reliance on ad hoc emer- gency measures. In the future, not only will entomologists need to provide quantitative data on key factors, important in the dynamics of pest species and their control, but other workers, such as pesticide chemists, economists, and mathematicians, who are also charged with responsibilities in the develop- ment of pest-control measures, will have to provide equally detailed field data. On the basis of all the data, stored in computers for the service of growers, it will be possible to determine, with confidence, interrelationships of key fac- tors from all sources involved in the management of crop-pest species. Thus, the integration of all components important in pest-control programs will provide the accurate prediction of pest outbreaks, time, kind, and intensity; the economic threshold levels tolerable for each crop pest; the type of pesti- cide, virus, bacterium, or other control needed to keep pests at, or below, economic-threshold levels; the required dosage levels of chemicals or patho- gens used, and the number and timing of applications; and the most suitable harvest date for crops, as determined by the rate of breakdown of pesticides in or on plants or in the soil. BIBLIOGRAPHY Campbell, I. M. 1963. Discussion, p. 94. In E. J. LeRoux (ed.). Population dynamics of agricultural and forest insect pests. Mem. Entomol. Soc. Can. 32. Cheng, H. H., and E. J. LeRoux. 1968. Biology and dynamics of the birch leaf miner, Fenusa pusilla (Lepeletier) (Hymenoptera: Tenthredinidae) on blue birch, Betula caerulea grandis Blanchard, in Quebec. (In Press) Clark, L. R., P. W. Geier, R. D. Hughes, and R. F. Morris, 1967. The ecology of insect populations in theory and practice. Methuen & Co., London. 232 pp. Embree, D. G. 1965. The population dynamics of the winter moth in Nova Scotia, 1954-1962. Mem. Entomol. Soc. Can. 46. 57 pp. Harcourt, D. G. 1963. Major mortality factors in the population dynamics of the dia- mondback moth, Plutella maculipennis (Curt.) (Lepidoptera: Plutellidae), pp. 55- 56. In E. J. LeRoux (ed.). Population dynamics of agricultural and forest insect pests. Mem. Entomol. Soc. Can. 32. Harcourt, D. G. 1966. Major factors in survival of the immature stages otPieris rapae (L.). Can. EntomoL 98:653-662. Holling, C. S. 1959. Some Characteristics of simple types of predation and parasitism. Can. Entomol. 91:385-398. Holling, C. S. 1966. The functional response of invertebrate predators to prey density. Mem. Entomol. Soc. Can. 48. 86 pp. Jaques, R. P. 1962. Stress and nuclear polyhedrosis in crowded populations of Tricho- plusia ni (Hiibner). J. Insect Pathol. 4:1-22.

ECOLOGICAL BASIS FOR CONTROL 63 Knipling, E. F., and J. U. McGuire, Jr. 1966. Population models to test theoretical ef- fects of sex attractants used for insect control. U.S. Dep. Agr. Inform. Bull. 308. 20pp. Knipling, E. F., and J. U. McGuire, Jr. 1968. Population models to appraise the limita- tions and potentialities of Trichogramma in managing host insect populations. U.S. Dep. Agr. Tech. Bull. 1387. 44 pp. LeRoux, E. J. 1961. Effects of "modified" and "commercial" spray programs on the fauna of apple orchards in Quebec. Ann. Entomol. Soc. Quebec. 6:87-121. LeRoux, E. J., Editor, 1963. Population dynamics of agricultural and forest insect pests. Mem. Entomol. Soc. Can. 32. 103 pp. LeRoux, E. J. 1964a. The application of ecological principles to orchard entomology in Canada. Can. Entomol. 96:348-356. LeRoux, E. J. 1964b. Ecological considerations in chemical control. Insect population problems. Bull. Entomol. Soc. Amer. 10:70-74. LeRoux, E. J., R. O. Paradis, and M. Hudon. 1963. Major mortality factors in the popu- lation dynamics of the eye-spotted bud moth, the pistol casebearer, the fruit-tree leaf roller, and the European corn borer in Quebec, pp. 67-82. In E. J. LeRoux (ed.). Population dynamics of agricultural and forest insect pests. Mem. Entomol. Soc. Can. 32. LeRoux, E. J., and C. Reimer. 1959. Variation between samples of immature stages, and of mortalities from some factors, of the eye-spotted bud moth, Spilonota ocellana (D. & S.) (Lepidoptera: Olethreutidae), and the pistol casebearer, Coleophora mali- vorella, Riley (Lepidoptera: Coleophoridae), on apple in Quebec. Can. Entomol. 91: 428^149. Morris, R. F., Editor. 1963. On the dynamics of epidemic spruce budworm populations. Mem. Entomol. Soc. Can. 31. 332 pp. Paradis, R. O., and E. J. LeRoux. 1965. Recherches sur la biologic et la dynamique des populations naturelles d'Archips argyrospilus (Wlk.) (Lepiodoptre'res: Tortricidae) dans le sudouest du Quebec. Mem. Entomol. Soc. Can. 43. 77 pp. Pottinger, R. P., and E. J. LeRoux. 1968. The biology and dynamics of Lithocolletis blancardella Fabr. (Lepidoptera: Tineidae) on apple in Quebec. (In Press). Samarasinghe, S., and E. J. LeRoux. 1966. Biology and dynamics of the oystershell scale, Lepidosaphes ulmi (L.) (Homoptera: Coccidae) on apple in Quebec. Ann. Entomol. Soc. Quebec. 11:206-292. Ullyett, G. C. 1947. Mortality factors in populations ofPlutella maculipennis (Curtis) (Tineidae: Lep.) and their relation to the problem of control. S. Afr. Dep. Agr. For. Entomol. Mem. 2:77-202. Uvarov, B. P. 1928. Locusts and grasshoppers. Imperial Bureau of Entomology, London. 352 pp. Watt, K. E. F. 1963. Mathematical population models for five agricultural crop pests. Mem. Entomol. Soc. Can. 32:83-91. Watt, K. E. F., Editor. 1966. Systems analysis in ecology. Academic Press, New York. 276 pp. Watt, K. E. F. 1968. Ecology and resource management. McGraw-Hill Book Company, Inc., New York. 450 pp. Wellington, W. G. 1960. Qualitative changes in natural populations during changes in abundance. Can. J. Zool. 38:289-314. Wynne-Edwards, V. C. 1962. Animal dispersion in relation to social behaviour. Oliver and Boyd, London. 653 pp.