offensive-defensive interactions between herbivores and ... · plants are defended against...

Offensive-Defensive Interactions between Herbivores and Plants: Their Relevance in HerbivorePopulation Dynamics and Ecological TheoryAuthor(s): David F. RhoadesReviewed work(s):Source: The American Naturalist, Vol. 125, No. 2 (Feb., 1985), pp. 205-238Published by: The University of Chicago Press for The American Society of NaturalistsStable URL: http://www.jstor.org/stable/2461633 .Accessed: 12/03/2012 11:49

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

The University of Chicago Press and The American Society of Naturalists are collaborating with JSTOR todigitize, preserve and extend access to The American Naturalist.

http://www.jstor.org

http://www.jstor.org/action/showPublisher?publisherCode=ucpress

http://www.jstor.org/action/showPublisher?publisherCode=amsocnat

http://www.jstor.org/stable/2461633?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp

Vol. 125, No. 2 The American Naturalist February 1985

OFFENSIVE-DEFENSIVE INTERACTIONS BETWEEN HERBIVORES AND PLANTS: THEIR RELEVANCE IN HERBIVORE POPULATION

DYNAMICS AND ECOLOGICAL THEORY

DAVID F. RHOADES

Department of Zoology, University of Washington, Seattle, WA 98195

Submitted November 9, 1983; Accepted June 26, 1984

Many species of small herbivorous animals such as insects, voles, rabbits, and leaf-eating birds exhibit wide variation in population numbers through time. Populations of some species remain at low levels for a number of years, erupting to high levels at irregular intervals (Berryman and Baltensweiler 1981). Spruce budworm (Morris 1963), the African armyworm (Khasimuddin 1981a; Odiyo 1981), many species of bark beetles (Berryman 1979), and locusts (Kennedy 1956; Tsyplenkov 1978) typify these species. Populations of other species exhibit remarkably regular cycles (Berryman and Baltensweiler 1981) with periodicities of about 4 yr in microtine rodents (voles and lemmings; Keith 1974; Finerty 1980), and about 10 yr in muskrats, hares (Keith 1974; Finerty 1980), grouse and ptarmigan (Keith 1963), larch budmoth (Baltensweiler et al. 1977), Douglas-fir tussock moth (Brookes et al. 1978), and forest tent caterpillars (Hodson 1941). These population variations have been of great interest to theoretical ecologists and pest and game managers for many years. Traditionally, population variations have been attributed to a combination of direct weather-induced changes in mortality and natality of herbivores, together with effects of competition for food and varying levels of predation, parasitism, and disease (see Cold Spring Harbor Symposia on Quantitative Biology 1957). This is because population eruptions are sometimes correlated with unusual weather, and population collapses are sometimes associated with food depletion and increased levels of parasitism, predation, and disease. However, mathematical population models based on these classical factors have so far proved relatively unsuccessful in predicting population trends (Baltensweiler 1968; Fye 1974; Stehr 1974; Waters and Stark 1980). The correlation between eruptions and unusual weather is not clear-cut. Usually the correlation is only a partial one. It is also becoming clear that weather often acts indirectly rather than directly affecting herbivore mortality. Similarly, populations often collapse in the presence of an abundance of food, and in a way that cannot be accounted for by weather, predation, parasitism, or disease (see Rhoades 1983a).

Am. Nat. 1985. Vol. 125, pp. 205-238. ? 1985 by The University of Chicago. 0003-0147/85/2502-0007$02.00. All rights reserved.

206 THE AMERICAN NATURALIST

DEFENSIVE OFFENSIVE ADAPTATIONS ADAPTATIONS

|PRODUCERS ONS MER

FIG. 1.-Offensive-defensive coevolution between consumers and producers. Solid arrows = "lead to the evolution of"; dashed arrows = "selects against." Defensive adaptations refer to producers and offensive adaptations refer to consumers.

In these traditional models, plants have been viewed as passive participants. There is now strong evidence, however, that plants are far from passive in their interactions with herbivores. In this paper, I briefly review the evidence that plants are defended against herbivores, that plant nutritional quality to herbivores can vary as a function of physical stress of plants and degree of herbivory experienced by plants, and that defensive communication between plants exists. I postulate that acquired immunity to herbivores may be an important component of plant defense. Finally, I suggest that there may be two alternative strategies of herbivores to counter plant defensive systems. These strategies may explain why population levels of some species of herbivores are variable while those of others are relatively invariant, why some species of herbivores exhibit phase polymorphism, and together with other considerations, why it has proved difficult to demonstrate competition between herbivores.

DEFENSIVE ADAPTATIONS OF PLANTS AND COMPLEMENTARY OFFENSIVE ADAPTATIONS OF HERBIVORES

Offensive-Defensive Coevolution between Consumers and Producers

The existence of producer organisms has led to the evolution of consumer organisms that prey on them (fig. 1). Individual producer organisms with properties which render them less suitable as prey should be at a relative advantage compared to individual producers which lack these properties. Predation by consumers on producers should therefore lead to the evolution of defensive adaptations in producers (fig. 1). Similarly, individual consumers with properties that enable them to overcome prey defenses will be at an advantage. This should lead to the evolution of offensive adaptations in consumers that decrease the effectiveness of prey defenses. These offensive adaptations, in turn, feed back to select for amplified prey defenses, and so on. The evolutionary positive-feedback loop between offense and defense is probably largely controlled by the metabolic cost (Leonard 1977; McKey 1979; Rhoades 1979; McLaughlin and Shriner 1980; Mooney and Gulman 1982) of defensive and offensive adaptations (fig. 1). For each known defensive adaptation in producer organisms we can therefore expect to find complementary offensive adaptations in consumers and vice versa. Known and possible defensive adaptations of plants against herbivores and their com-

INTERACTIONS BETWEEN HERBIVORES AND PLANTS 207

DEFENSIVE ADAPTATIONS OFFENSIVE ADAPTATIONS OF HERBIVORES OF PLANTS

STEALTH OPPORTUNISM 10 11 12

COMMUNICATION SUPPRESS, EMIT BETWEEN 'COMMUNICATION' ICOUNTERS IGNALS' PLANTS --- -

L ? --

7 8 9 ACQUIRED SUPPRESS

L IMMUNITY_ 'RECOGNITION SURPRISE PRECOGNITION L _ _ _ _ _ _ L _EGI_ _EC_ _ __ I__N

5 ~~~~~6 r 4 | |SUPPRESS [fINDUCE | INDUCIBLE INDUCED DECREASED DEFENSES DEFENSES DEFENSES

I 2 3

CODENST VES 11IDETOXIFY [AVOID

5 MUCH EVIDENCE O SOME EVIDENCE [ LITTLE OR NO EVIDENCE LJ

FIG. 2.-Defensive adaptations of plants and complementary offensive adaptations of herbivores. Each known or postulated adaptation is referred to by number in the text.

plementary known and possible offensive adaptations in herbivores are displayed in figure 2. For each defensive attribute of plants, two alternative offensive tactics in herbivores are known or can be postulated (fig. 2), as in the following discussion.

Passive Plant Defense and Counteradaptations of Herbivores

During the last 30 years it has become clear that many of the chemical constituents of plants (plant secondary metabolites) protect plants against attack from pathogens and herbivores, and are used by plants to interfere with the growth or germination of competing plants (Rice 1974; Wallace and Mansell 1976; Harborne 1978; Rosenthal and Janzen 1979). With respect to plant-herbivore interactions, attention has been focused, until quite recently, on preformed constitutive (Levin 1971) defensive substances of plants (fig. 2.1) that deter and toxify herbivores. Some of these compounds, such as alkaloids and cyanogens, have a direct toxic action. Others, such as tannins and lignins, may act by reducing the digestibility of plant tissues (Feeny 1975, 1976; Rhoades and Cates 1976; Rhoades 1979; Swain 1979; but see Bernays 1981; Klocke and Chan 1982; Zucker 1983).

To circumvent the effects of plant defensive substances, herbivores have evolved detoxification mechanisms (fig. 2.2). The most widely recognized of these mechanisms is metabolic conversion of the defensive substance to a less toxic derivative, followed by excretion of the metabolite or its conjugate (Millburn 1978; Brattsten 1979; Dowd et al. 1983). In addition, dietary toxins are degraded within the gut of ruminant and nonruminant vertebrates by microbial activity


(Freeland and Janzen 1974; Allison and Cook 1981). The high gut pH observed in many phytophagous insects may be an adaptation to inhibit complex formation between plant tannins and proteins or other dietary constituents, to thus circumvent the digestibility-reducing action of tannins in plant tissues (Feeny 1970; Rhoades 1977; Berenbaum 1980). Some adapted herbivores have evolved refrac- tory metabolism in which target enzyme systems are relatively insensitive to plant toxins (Rosenthal et al. 1976; Vaughan and Jungreis 1977; Jaenike et al. 1983).

In addition to detoxifying plant defensive metabolites, most herbivores, to one degree or another, use these substances as cuing stimuli to aid in the recognition of their host plants (see Rhoades 1983a). Use of plant defensive chemicals has been carried a stage further by many herbivores which store them for their own defense (Rothschild 1973) or nutritionally benefit from their presence in the food (Rosenthal et al. 1977; Berenbaum 1981; Bernays and Woodhead 1982; McFarlane and Distler 1982).

Besides detoxifying and using plant defensive substances, herbivores also avoid them (fig. 2.3). Food selection by herbivores is determined largely by a general positive response to sugars, salts, amino acids, and other primary nutrients found in all plants, coupled with positive and negative responses to the specific secondary metabolites in host and nonhost plant species (see Rhoades 1983a). Within a host plant species, or within an individual host plant, herbivores often preferentially attack those plants or tissues which contain low levels of secondary metabolites (Jones 1972; Cates 1975; McKey 1979; Rhoades 1979; Schultz 1983a; Whitham 1983). The demonstration of deterrence of herbivore feeding or oviposition by secondary metabolites (fig. 2.3) has been one of the main methods used to establish the protective action of secondary metabolites to plants (Rhoades 1979).

The discussion to this point encompasses what can be termed the theory of passive plant defense against herbivores (see Fraenkel 1959; Ehrlich and Raven 1965; Feeny 1975, 1976; Rhoades and Cates 1976). Salient features can be summarized as follows. Each species of plant has evolved a unique set of defensive metabolites which deter attack from most herbivores except those few species which have broken through the defenses by counteradaptation. Each species of herbivore has evolved counteradaptations against the defenses of a limited number of plant species. Herbivores use the plant defensive systems to which they are adapted as host-finding cues and feeding stimulants, while avoiding plant species containing defensive systems to which they are not adapted and those tissues of their host species which contain levels of defensive substances sufficient to overload the accommodation mechanisms of the herbivores. Apart from containing defensive chemicals, the plants were regarded as passive, and their defenses were regarded as static. It is now becoming clear that plants are much less passive in their interactions with herbivores than originally conceived, and their defenses are far from static.

Active Plant Defense and Counteradaptations of Herbivores

A variety of influences stressful to plants promotes attack by phytophagous insects. These stressful influences include water deficit or surplus, unusual or


rapidly changing climatic conditions, nutrient-poor soils, competition from other plants, pollution, and damage during cultural operations (see Mattson and Addy 1975; Rhoades 1983a). White (1969, 1974, 1976) suggested that insects often preferentially attack physically stressed plants, and exhibit high fecundity and survival when feeding on them, because stress leads to increased concentrations of free amino acids and other soluble nitrogenous constituents in plant tissues. Compromised defensive systems in stressed plants are probably of greater importance, however (Rhoades 1979, 1983a; Wright et al. 1979; Waring and Pitman 1983). Thus, by attacking stressed plants, herbivores can minimize their exposure to plant defenses (fig. 2.3). For example, the oleoresin exudation pressure of pines experiencing a water deficit is reduced, resulting in increased bark beetle survival (Vite 1961). Similarly, defoliation causes decreased monoterpene production in the stems of grand firs, which promotes successful attack by the fir engraver beetle (Wright et al. 1979). In other cases, the effects of stress on plant defensive chemistry are less clear-cut. While some compounds decrease in concentration in tissues of stressed plants, other compounds increase. This may be due to reallocation, by stressed plants, of resources from costly but effective defensive systems to less costly but less effective defensive systems (Rhoades 1979, 1983a).

While most experimental work and theory concerning the effects of plant defenses on plant-herbivore interactions has, until recently, focused on preformed constitutive defenses, the dual importance of constitutive and inducible plant defenses has long been recognized by plant pathologists (Muller and Borger 1941; Muller 1956; Levin 1971; Horsfall and Cowling 1980). Inoculation with fungi (Suzuki 1980), bacteria (Goodman 1980), viruses (Hamilton 1980), and nematodes (McIntyre 1980) can increase resistance of plants to further challenges by the pathogens (see also Kuc 1982). In the case of attack by fungi, the mechanisms of induced resistance are fairly well understood. They involve the accumulation of low molecular weight antibiotic compounds (phytoalexins) and polymeric lignins, tannins, enzyme inhibitors, and agglutinins (Kuc and Caruso 1977; Suzuki 1980).

During the last 10 years we have recognized that plants also possess inducible defenses against herbivores (fig. 2.4). Plants can change their chemical properties to render their tissues less suitable for herbivore growth and development in direct response to herbivore damage or simulated herbivore damage. Much of the evidence for these plant responses has been reviewed by Rhoades (1979, 1983a; see also Berryman 1972; Raffa and Berryman 1982; Schultz and Baldwin 1982; Karban 1983; McNaughton and Tarrants 1983). Rapid responses, which affect current herbivores, and delayed responses, which protect plants against subsequent generations of herbivores, are known (Haukioja and Hakala 1975; Haukioja 1980; Rhoades 1983a). Responses do not appear to be restricted to plants of any particular taxonomic affinity or growth form. Systemic accumulation of leaf phenolic compounds following leaf damage has been observed in many species of trees. Local production of terpenes at the site of damage is a common response of conifers to attack by bark beetles. A variety of chemical responses has been observed in herbaceous plants. Whether damage-induced responses are caused by increased content of constitutive defensive chemicals or the accumulation of different substances is unknown in many cases, but both types of response have


been observed. For example, fiber (lignocellulose) in larch (Benz 1974, 1977) and wheatgrass (Higgins et al. 1977) or tanninlike substances in birch (Niemela et al. 1979), alder (Rhoades 1983b), sedge (Rhoades 1983a), oak (Schultz and Baldwin 1982), maple and poplar (Baldwin and Schultz 1983), or silica in grasses (McNaughton and Tarrants 1983) are all present at relatively high concentrations in the unattacked plants, with moderate increases stimulated by insect or mechanical damage. On the other hand, proteinase inhibitors in tomato (Nelson et al. 1983), coumestrol (an estrogenic compound) in alfalfa (Loper 1968), and juvabione-related compounds (potential insect juvenile hormone analogues) in firs (Puritch and Nijholt 1974), produced by the plants in response to insects or mechanical damage, are not present or are scarcely detectable in unattacked plants. I suggest that the concepts of "qualitative" and "quantitative" plant defensive systems proposed by Feeny (1975, 1976; see also Rhoades and Cates 1976; Rhoades 1979) may be useful in explaining this dichotomy. Quantitative defensive substances such as lignin, tannins, and silica are postulated to act in a dose-dependent manner, even against adapted herbivores. An attacked plant can therefore gain protection against subsequent attack from the same herbivore by increasing the concentration of these substances in its tissues. On the other hand, qualitative defensive substances such as specific enzyme inhibitors and animal hormone analogues, although providing protection from nonadapted herbivores at low concentration, are postulated to provide little protection against adapted herbivores even at very high concentrations. An attacked plant would therefore gain little protection against subsequent attack from the same consumer by increasing the concentration of qualitative defensive substances already present during the initial attack. In this case, the synthesis of novel substances, not present in the unattacked plant, would be a more effective defense. We can therefore expect that responses involving quantitative defensive substances usually should result from increases in concentration of substances present in significant quantities in unattacked plants, whereas responses involving qualitative defensive substances should usually be caused by substances not found in significant concentrations in unattacked plants. In other words, quantitative defenses should participate mainly in quantitative responses and qualitative defenses should participate mainly in qualitative responses.

The existence of inducible defensive responses in plants should select for adaptations in herbivores to suppress the responses (fig. 2.5). In addition, given that plant defensive systems are plastic, adaptations in herbivores that actively decrease levels of defense in plants (fig. 2.6) are likely. There is evidence for both of these tactics in herbivores. Many species of bark beetles mass-attack their host trees using aggregation pheromones (Coster and Johnson 1979; Wood 1982). In some conifers, exudation pressure of preformed resin (the constitutive defense) and local synthesis of resin at the site of attack (the induced defense) are so reduced by the weight of numbers of attacking beetles that the tree succumbs (Berryman 1969, 1972; Cates and Alexander 1982; Raffa and Berryman 1983). Thus, through concerted attack, bark beetles can reduce the effectiveness of host defenses far below that experienced by a single attacking beetle (fig. 2.6). Some insects, particularly bark beetles and other boring insects, are dependent on


ectosymbiotic relationships with viruses, mycoplasms, bacteria, fungi, protozoa, or nematodes to achieve successful attack. The symbionts infest and propagate in the plant, increase the nutritional quality of plant tissues to the insect, and in some cases, themselves become the main food source (Norris 1979; Whitney 1982). Leaf composting by leaf-cutting ants (Cherrett 1972) can be viewed as an extreme form of this tactic. Gall-forming and plant-sucking arthropods in general are known, or strongly suspected, to inject substances into plants that increase their nutritional quality (Miles 1968a, 1968b, 1978; Dieleman 1969; Carter 1973; Os- borne 1973; Markkula et al. 1976; Hori 1976; Hori and Endo 1977; Dropkin 1979; Norris 1979; Cornell 1983a, 1983b). For instance, the saliva of Hemiptera often contains polyphenoloxidase enzymes which appear to oxidatively neutralize plant defensive phenolic compounds and/or be involved in the production of auxin-like compounds in situ (fig. 2.6; Miles 1968a, 1968b, 1978). Sirex wood wasps inject a toxin and spores of a symbiotic fungus during oviposition in stems of Pinus radiata. The toxin blocks translocation of photosynthate from the leaves, curtail- ing synthesis by the plant of polyphenols and resins in the attack lesions. Exuda- tion of preformed resin into the lesions is also reduced. The fungus then rapidly invades the conductive tissue of the plant to induce transpirational stress and provide a suitable substrate for the developing Sirex larvae (fig. 2.6; Coutts 1968, 1969a, 1969b; Madden 1977). Some lepidopteran leaf-miners retard senescence of leaves in which they are feeding to produce "green islands" in the leaves. Mined areas of leaves and the caterpillars contain high levels of cytokinins (Engelbrecht et al. 1969; Engelbrecht 1971). Green islands have also been observed to develop around the feeding sites of other insects (see Osborne 1973; Kahn and Cornell 1983). The squash beetle Epilachna tredecimnota cuts a circular trench around an area

of squash leaf, so that only a few veins and pieces of lower epidermis hold the encircled leaf section in place. The beetle then feeds on the excised area. This behavior prevents the plant from mobilizing defensive substances into the feeding site (fig. 2.5; Carroll and Hoffman 1980). Alder sawflies (Eriocampa ovata) cut all the main veins of a leaf of red alder before feeding on it (Mackay and Wellington 1977). Many insect species girdle a stem before feeding on the distal part or attack a petiole before feeding on the leaf (see Rhoades 1983a). These behaviors probably prevent the mobilization of defensive chemicals into the feeding region (fig. 2.5), or cause the accumulation of nutrients in it, or both. There may also be less obvious ways in which herbivores suppress plant defensive responses (fig. 2.5), for instance by producing salivary components that interfere with the mechanisms which plants must possess in order to recognize damage. For instance, Miles (1968a) suggested that one of the functions of the salivary sheath secreted by phytophagous Hemiptera may be to reduce the chance of defensive responses by the plants.

It would be adaptive for plants to recognize the difference between mechanical damage and herbivore damage and also between damage by different herbivore species (fig. 2.7). Acquired immunity of plants against particular herbivore species is a further possibility (fig. 2.7). Effects of this kind are well established in interactions between plants and pathogens. In many cases, plant pathogens suc-


cessfully invade susceptible hosts not so much because they can detoxify or tolerate the defensive response of their host plant, but rather because they are not recognized as foreign by the plant and defensive responses are not induced (Sequeira 1980). To my knowledge, there is no direct evidence for differential defensive response of plants to mechanical versus herbivore damage, to damage by different herbivore species, or for acquired immunity in plants to particular herbivore species. There is, however, considerable evidence showing that regrowth rates of plants following grazing by grasshoppers and ungulates, or following mechanical clipping plus application of saliva or saliva components, can differ significantly from regrowth rates following clipping alone (Reardon et al. 1972, 1974; Dyer and Bokhari 1976; Dyer 1980; Capinera and Roltsch 1980; Detling and Dyer 1981; Detling et al. 1981). In some cases regrowth was reduced compared to clipped controls, in others it was stimulated, and in still others no difference was observed. These various effects on regrowth rates are possibly manifestations of underlying offensive-defensive interactions between plants and herbivores in which enhanced regrowth rates are associated with an induced increase in food quality caused by herbivore salivary secretions, whereas decreased regrowth rates are associated with an induced decrease in food quality controlled by the plant. In other words, herbivore saliva may contain factors which cause plants to allocate more resources to growth and less to defense (fig. 2.6). If the plant has evolved mechanisms to recognize the salivary factors of the particular herbivore, with or without prior conditioning from attack by that herbivore, the plant allo- cates less to growth and more to defense following an attack (fig. 2.7). Compari- sons of defensive commitment and regrowth rates of plants following mechanical damage with those following attack by a series of herbivores which vary in their degree of natural association with the plants, and following attack by herbivores on naive plants versus plants conditioned by previous attack, may prove re- vealing.

My interpretation of the effects of salivary constituents on plant regrowth following damage differs from that of Dyer et al. (1982). These investigators (see also Owen and Wiegert 1976, 1982) inferred, from cases in which grazing or treatment with salivary extracts stimulated plant growth, that plant-grazer interactions have a mutualistic component, at least at low levels of damage. As noted by McNaughton (1979) and Stenseth (1983), however, increased growth caused by grazing does not necessarily lead to increased plant reproductive fitness. To my knowledge, there is no evidence for enhanced reproductive fitness of grazed plants relative to ungrazed competitors. In addition, it is difficult to reconcile cases of decreased growth of plants caused by salivary and crop secretions (Detling and Dyer 1981), or decreased growth of plants grazed by herbivores compared to clipped controls (Capinera and Roltsch 1980), with plant-grazer mutualism. On the other hand, both stimulation and inhibition of plant growth by herbivores and their secretions can be readily understood if herbivores have evolved adaptations to manipulate plant nutritional quality to the advantage of the herbivores, and plants, in turn, have evolved mechanisms to recognize these adaptations and use them as defensive cues.

If recognition and acquired immunity are components of plant defensive adapta-


tion (fig. 2.7), we can expect adaptations, such as salivary secretions, to suppress recognition in some herbivores (fig. 2.8). Surprise would be an alternative tactic (fig. 2.9). All mobile herbivores which feed on many different individual plants during their lifetime possibly employ surprise to one degree or another, since there is little time for induction of defensive responses before departure of the herbivore. Herbivore species in which migration or dispersal are important features of their population dynamics, such as locusts (Uvarov 1977; Taylor 1979), spruce budworm (Clark et al. 1978), larch budmoth (Baltensweiler and Fischlin 1979), tent caterpillars (Wellington 1980), armyworms (Odiyo 1981), and voles (Thompson 1955; Finerty 1980), may also use this tactic by immigrating into populations of naive plants and then emigrating before, or as soon as, defensive responses are induced.

Advanced warning of attack (fig. 2.10) would be highly advantageous to plants. There is limited evidence that unattacked plants can detect and defensively respond to airborne substances emitted by nearby attacked plants. Rhoades (1983b) found that leaf quality, as measured in bioassays, of Salix sitchensis trees attacked by tent caterpillars and nearby unattacked control trees declined within 14.5 days of the initiation of attack. No evidence was found for root connections between the trees. Given that plants in general (Hanover 1972; Rasmussen 1972; Abeles 1973; Freeland 1980; Smith 1981), and willows in particular (Rasmussen 1970), emit volatile organic compounds into the atmosphere, and that in some cases the amounts and type of emissions can be changed by plant damage (Rhoades, in prep.), Rhoades (1983a) proposed that plants could receive pheromonal signals emitted by nearby attacked trees. To test this idea, Baldwin and Schultz (1983) confined individually potted sugar maple seedlings in two growth chambers. They then tore leaves on some of the plants in one chamber and compared the phenolic chemistry of the damaged plants, undamaged plants in the same chamber (communication controls) and undamaged plants in the separate chamber (true controls), for several days following damage. They found increased levels of leaf total phenolics and tannins in leaves of the damaged plants and in the communication controls compared to the true controls. Similar results were obtained with poplar. Their results support the hypothesis of pheromonal communication between plants in response to damage. Perry and Pitman (in press) found changes in foliage toxicity of Douglas-fir to western spruce budworm that occurred coincidentally with the appearance of a rapidly building natural population of budworm in the vicinity of their study trees. They suggest intertree communication as a possible explanation for this result.

If damage-induced pheromonal communication between plants is common we can expect some herbivores to have evolved adaptations that suppress it (fig. 2.11). This could occur as an indirect result of suppression of recognition of attack (fig. 2.8) or through herbivore secretions that block the release of communication substances from the wound. Alternatively, some herbivores may themselves emit pheromonal countersignals to confuse, mimic, or otherwise interfere with the signals generated by plants (fig. 2.12). For example, herbivores may produce a signal which causes plants to reallocate defensive substances from the attacked tissue to other tissues, leading to an increase in nutritional quality of the food.


Indeed, if plant pheromonal communication is common, it would be surprising if some herbivores did not exploit the system, especially considering that herbivores heavily utilize airborne chemical signals such as sex, alarm, and aggregation pheromones, many of which are closely related to or even derived from plant volatiles (Prokopy 1981; Roelofs 1981; Borden 1982; Wood 1982; Weldon 1983). An example of the reverse phenomenon, in which a plant interferes with a pheromonal communication system of an herbivore, is known in wild potato which repels aphids by releasing the alarm pheromone of the aphid (Gibson and Pickett 1983). It is therefore possible that offensive-defensive interactions between herbivores and plants are conducted, in part, at the pheromonal level. If so, interactions at the level of gustation, plant tissue chemistry, and catabolic detoxification, stressed in much recent research, may in many cases assume a secondary role.

I suggest that the combinations of herbivore offensive tactics 2, 5, 8, 11, or 3, 6, 9, 12 (fig. 2) constitute alternative attack strategies, which I term stealth and opportunism, respectively.

STEALTH, OPPORTUNISM, AND HERBIVORE POPULATION DYNAMICS

Although all herbivore populations fluctuate to some degree, population levels of some species are much more variable and rise to much higher levels than others. Differences in generation times and intrinsic rates of population increase among herbivore species are obviously involved to some degree. Invertebrate and small vertebrate herbivores often have shorter generation times and higher reproduction rates than large vertebrate herbivores. This probably contributes to the higher variability of population numbers in the former group by allowing their populations to respond rapidly to favorable changes in food quality and other influences. There are, however, great differences in population variability even among invertebrate and small vertebrate species. In natural systems, most species of small herbivores have low, relatively stable populations but a few species show wide population swings (Berryman and Baltensweiler 1981; Schultz 1983b). Schultz (1983b) estimated that fewer than 10% of the species listed in the Cana- dian Forest Survey of Lepidoptera (Prentice 1962, 1963) exhibit periodic or occasional outbreaks. Despite their small number, these pest species often constitute the most abundant and destructive herbivores in natural environments. I suggest that species with low, relatively constant populations may be stealthy, whereas those with variable populations may be opportunistic (fig. 2, table 1), or at least may employ the latter strategy part of the time.

Population Dynamics of Stealthy Herbivores

Inhibition of plant defensive response is an important component of the proposed stealthy strategy. Since defensive responses are more likely to be induced as amount of damage increases, we can expect stealthy herbivores to display adaptations that minimize their impact on plant fitness. Thus, stealthy herbivores should attack tissues that are of relatively low value to their host plants, such as


TABLE 1

PROPOSED CHARACTERISTICS OF STEALTHY AND OPPORTUNISTIC HERBIVORE SPECIES OR PHASES IN ADDITION TO THOSE OF FIGURE 2

Stealthy Opportunistic

Invariant population levels Variable population levels Conservative use of the host plant Profligate use of the host plant

Attack tissues of low value to the host Attack tissues of high value to the host High efficiency of conversion of ingested Low ECI

material (ECI)t Low rate of conversion of ingested material (RCI) Variable* RCI High digestibility of food (AD)t Low AD Low feeding rate (consumption index, CI)t Variable* CI Low body temperature High body temperature Low metabolic rate High metabolic rate

Constant levels of parasitism and predation Variable** levels of parasitism and predation Constant growth, survival, and reproductive rates Variable* growth, survival, and reproductive

rates Low maximum intrinsic rate of population increase High I'max

(rmax)

Low maximum fecundity (bmax) High bmax Nonmigratory Migratory and nonmigratory Sexual reproduction Sexual and parthenogenetic reproduction Mutual interference Mutual facilitation

Single eggs Clustered eggs Solitary Gregarious Territorial Colonial Regular spatial dispersion of damage Contagious spatial dispersion of damage

* High during population increase, low during population decrease. ** Low during population increase, high during population decrease. t Nomenclature follows Waldbauer (1968).

mature leaves as opposed to young leaves and growing tips, and they should efficiently convert plant biomass into herbivore biomass. They should also avoid their fellows when foraging or ovipositing since a plant experiencing attack from a single herbivore is less likely to respond defensively than one attacked by many. In other words, mutual interference should occur between stealthy herbivores that attack the same type of tissue. This should result in a solitary habit; the wide dispersal of single eggs by invertebrates (see Stamp 1980); a regular (even) dispersion (Southwood 1978) of damage among individual host plants; and territorial defense of the resource in some cases (table 1). Territoriality is common among invertebrate and vertebrate folivores, particularly in butterflies, grasshoppers, and primates (Hladik 1978; Montgomery 1978; Baker 1983). The stealthy strategy would lead to low and relatively constant populations because as herbivore populations rose above a certain level, rapid plant responses, acting on the attacking generation of herbivores, should lower their fecundity and increase their mortality. Besides directly affecting herbivores, the plant responses should also increase their susceptibility to parasitism and predation (see Price et al. 1980; Schultz 1983a; Rhoades 1983a). Population levels of stealthy herbivores are thus viewed as being under the influence of rapid negative feedback from changes in plant nutritional quality and consumers at higher trophic levels.


Population Dynamics of Opportunistic Herbivores

For opportunistic herbivore species, two substrategies are suggested. First, they preferentially attack plants whose defenses are compromised by physical stress (fig. 2.3). After the period of stress has passed, the plants regain their defensive capabilities, responses are induced, and the herbivore populations collapse under the influence of direct nutritional effects and increased susceptibility to parasites and predators caused by poor food (Rhoades 1983a). This results in eruptive population dynamics, with the timing of outbreaks correlated with climatic or other events stressful to plants. Outbreaks of locusts (see below), spruce budworm (Morris et al. 1958; see also Rhoades 1983a), and bark beetles (Berryman 1978) are in this category. If the period of plant stress is sustained, this results in continued high populations of herbivores and total destruction of the plant resource in some cases, leading to herbivore population collapse from lack of food. Strong migratory or dispersal ability should be characteristic of the herbivores to enable them to locate stressed hosts. Alternatively, opportunists invade a portion of their geographical range occupied by naive plants which have not been immunized by recent attack (fig. 2.9). High-quality food leads to a rapid and large increase in herbivore population. The plants ultimately increase their chemical defenses and the herbivores emigrate. In time the plants lose their immunity and the herbivores return. This results in cyclic population dynamics with the periodicity dependent on the induction and relaxation times of acquired immunity and associated changes in plant defensive commitment. Cycles of grouse, ptarmigan (Keith 1963), voles, lemmings, snowshoe hares (Finerty 1980; Bryant 1981), larch budmoth (Baltensweiler and Fischlin 1979), and Urania moths (Smith, in press) may result from this mechanism.

Possibly, opportunists emit countersignals (fig. 2.12) that cause increased plant nutritional quality (fig. 2.6) until recognition (fig. 2.7) occurs. Alternatively, these signals may delay recognition and the onset of defensive responses. In contrast to stealthy herbivores, opportunists can be expected to display adaptations which stress their host plants (e.g., bark beetles and Sirex), and to attack tissues of high value to the plant. Mass attack, to overload plant defense mechanisms, may be an important component of the opportunistic strategy. Aggregation would also lead to numerical amplification of countersignals. In other words, mutual facilitation between opportunists is probable. We can therefore expect opportunists to display adaptations which promote aggregation, such as egg clustering (e.g., spruce budworm, Douglas-fir tussock moth, gypsy moth, tent caterpillars, fall webworm, fall cankerworm, armyworms, Urania moths, and locusts), aggregation pheromones (Prokopy 1981), and coloniality. A contagious (clumped) spatial dispersion (Southwood 1978) of damage and a migratory habit are also to be expected (table 1). Migratory behavior, however, is not a necessary requirement in species with variable populations if they undergo transition between stealthy and opportunistic phases (see Douglas-fir tussock and gypsy moths discussed below).

Rapid population build-up of opportunists would enable them to take the most advantage of transitory circumstances that impair plant defensive capability, thus selecting for high maximum fecundity and high maximum intrinsic rates of popula-


tion increase in opportunists as opposed to stealthy herbivores. Growth rates, survival, and reproductive rates of opportunists should therefore be high during population eruption resulting from high-quality food and low during population collapse, whereas these properties should be relatively constant for stealthy herbivores. Similarly, mortality of opportunists from parasitism and predation should be low during population increase as the result of satiation and low susceptibility caused by high-quality food of the herbivores, whereas it should be high during collapse. On the other hand, stealthy herbivores should exhibit relatively constant rates of parasitism and predation (table 1). Populations of opportunists are thus viewed as being under the influence of positive feedback resulting from mutual facilitation and predator satiation during population increase, and delayed negative feedback from plant defensive responses and consumers at higher trophic levels during population decrease.

Watt (1965) analyzed data for Macrolepidoptera gathered by the Canadian Forest Insect Survey and found that although there were far fewer gregarious species (n = 32) than solitary species (n = 520), gregarious species had much higher average abundances and far more variable population levels than solitary species. This finding supports the present analysis, but whether the same trend will be found for solitary and gregarious herbivore species other than Canadian forest Macrolepidoptera remains to be seen. Insects that feed colonially are generally more successful when raised en masse than when raised singly or in small groups on their food plant, whereas the converse is true for solitary species (Wellington 1957; Iwao 1968; Way and Cammell 1970; Shiga 1976; Peters and Barbosa 1977). These observations are consistent with the hypothesis of mutual facilitation between opportunists and mutual interference between stealthy herbivores.

Host-Plant Specificity of Stealthy and Opportunistic Herbivores

Many species of herbivorous insects with variable populations such as gypsy moth, fall cankerworm, spruce budworm, fall webworm, locusts, and some tent caterpillar species appear to be dietary generalists in that they have been recorded attacking many different plant species (see Prentice 1962, 1963; Uvarov 1977; Gerardi and Grimm 1979). Watt (1964), however, found that in Canadian forest Macrolepidoptera there is no clear-cut correlation between the degree of dietary generalism and variability of population numbers. If anything, he found that, on average, generalist species have slightly more variable populations than special- ists. Arguments can be advanced in favor of both dietary specialization and generalism in stealthy and opportunistic herbivores. On one hand, from the "Jack-of-all-trades-master-of-none" hypothesis we might expect specialization to be evolutionarily favored in both groups since close association with few host species is likely to lead to the evolution of more-effective offensive adaptations than association with many hosts. On the other hand, opportunists should also be under selection for broad host range because this would provide them with more opportunities to exploit. The advantage to evenly dispersing their damage among plants should also select for broad host range in stealthy herbivores. Therefore,


although the relationship between dietary specialization and population variability remains an open question, I suggest that little correlation will be found.

PHASE CHANGES IN HERBIVORES AND THEIR RELATIONSHIP TO POPULATION STABILITY

Striking changes in herbivore "quality" are often observed during population fluctuations. In extreme cases of this phenomenon discrete phases or ecotypes characteristic of endemic or epidemic populations have been recognized. I suggest that phase changes in herbivores may be due to facultative transition between the strategies of stealth and opportunism.

Locusts and Armyworms: A Population Model

The most widely recognized and distinct examples of phase polymorphism are found in locusts (Schistocerca gregaria, Locusta migratoria, Nomadacris sep- temfasciata, Locustana pardalina, Chortoicetes terminifera, and other species; Uvarov 1966). At low population densities, the solitaria phase occurs. In this phase, each locust lives well separated from other individuals. Nymphs hatching from egg pods laid by solitary females are cryptically colored (light gray, fawn, or green; Uvarov 1966). Hatching is asynchronous and the nymphs scatter upon emergence (Kennedy 1956). The gregaria phase occurs at high population density. Adults of this phase are more active than solitaria adults. They aggregate during feeding and egg laying to produce vast fields of egg pods which hatch synchronously, and they take flight in cohesive migratory swarms (Nolte 1974; Uvarov 1966, 1977). Adults of the two phases differ morphometrically (Uvarov 1966). Gregaria nymphs are much darker than solitaria nymphs, because of the presence of melanin in their cuticle. Together with yellow or orange markings, this dark coloration makes them more conspicuous than solitaria nymphs. Gregaria nymphs are more active than solitaria nymphs, they grow faster, and have a marked tendency to aggregate when feeding and resting (Kennedy 1956; Nolte 1974; Uvarov 1966, 1977). Most investigators have reported higher metabolic rates in gregaria than in solitaria (Butler and Innes 1936; Uvarov 1966).

Intermediate forms occur both in the field and in the laboratory (Kennedy 1956). In the laboratory, phase change can be induced by raising nymphs under crowded or solitary conditions. The transition begins within the first generation, but con- tinuous exposure to the appropriate conditions for two or more generations is more effective (Kennedy 1956; Uvarov 1966). These changes result from pheno- typic plasticity, since they occur without the intervention of selection. Transition to the gregaria form and aggregation of locusts are stimulated by an airborne "gregarization pheromone" produced in the feces of crowded nymphs (Nolte 1963, 1977; Ellis and Gillett 1968; Gillett 1968; Gillett and Phillips 1977). Nolte (1974, 1977) claimed that he had identified this pheromone as 2-methoxy-5- ethylphenol in a number of locust species and that this compound caused changes in chromosome morphology (increased chiasma frequencies) in the dividing cells of testes of exposed nymphs, but his findings have been vigorously challenged (Dearn 1974a, 1974b; Gillett 1983). Gillett and Phillips (1977) found evidence for a solitarizing pheromone produced in the feces of adults.


Solitaria and gregaria differ in so many morphological, physiological, and behavioral ways that they were regarded as different species until Uvarov (1921, 1928) proposed his phase theory. The phase theory explained the taxonomic puzzle of the sudden appearance and disappearance of destructive pests, which could not be found except during periods of high abundance. It did not explain why populations of locusts sometimes rise to a density sufficient to trigger the appearance of the destructive form (Key 1950).

Population explosions of locusts are of the irregular, eruptive type (Key 1950; Kennedy 1956; Uvarov 1977). Outbreaks are initiated from restricted areas within the geographic range of the solitaria form which itself exhibits a narrower range of occurrence than that invaded by the gregaria form (Tsyplenkov 1978; Kennedy 1956; Uvarov 1977). Typically these outbreak foci are river deltas, floodplains, lake margins, and other areas likely to experience wide seasonal and annual variation in soil moisture, and in which the effects of local or distant deviations in rainfall are likely to be magnified (Symmons 1978; Tsyplenkov 1978). At intervals, which may vary between 3 and 20 or more yr, the populations of locusts undergo a dramatic increase associated with the appearance of the gregaria phase. Migra- tory marching and flight by gregarized nymphs and adults from the outbreak foci lead to progressive invasion, by subsequent generations, of areas far removed from the point of origin of the outbreak. After a number of years, which may vary between 2 or 3 and 15 or more, the outbreak subsides, the locusts become solitarized, and disperse back to their restricted habitat. It is generally accepted that variations in degree of parasitism and predation are not responsible for population surges and collapses (Tsyplenkov 1978; Uvarov 1977).

An association between the timing of the initiation of locust outbreaks and drought, or in some cases excessive precipitation, has been known for many years (see Tsyplenkov 1978; White 1976). Many workers have assumed that this association is caused by direct effects on locust mortality and natality, but it now appears that the influence of weather on locust populations is mainly indirect by somehow altering the habitat of the solitary form (Tsyplenkov 1978; White 1976). White (1976), as part of his general hypothesis to explain the association between insect outbreaks and unusual weather, has suggested that stress of locust food plants (mainly graminoids and forbs), as a result of deviant rainfall, leads to increased food quality of the plants resulting in an outbreak. In White's scheme, outbreaks are thus most likely to be initiated from restricted areas with unpredict- able soil moisture contents, which most of them are. White (1976) does not address phase transformation, however, and his hypothesis does not explain why locust outbreaks continue long after weather stress has passed, why swarms can invade and increase in areas of unstressed plants, and why outbreaks ultimately collapse.

I suggest the following scheme which combines White's stress hypothesis with concepts of alternative attack strategies of herbivores and attack-induced changes in defensive posture of plants, developed earlier in this paper. The combination of morphological, physiological, and behavioral attributes which constitute the solitaria phase are adaptations which enable these insects to attack their food plants using the stealth strategy. This strategy is employed at low population densities by


locusts, and it is the mode characteristic of nonpest grasshopper species. Low metabolic rates and a sedentary habit permit efficient conversion of plant to insect biomass and minimal damage to food plants per unit converted. Behavioral dispersion minimizes damage to individual food plants. This may be combined with adaptations, such as salivary secretions, to suppress recognition of attack by the food plants. Territoriality, exhibited by some grasshopper species, is to be expected in solitaria. Populations are held in check by rapid plant defensive responses induced when the total populations of locusts and other stealthy herbivores rise above a critical level. Plant defensive responses exert direct effects on mortality and natality of solitaria and also increase their susceptibility to parasites and pathogens.

Following any event which leads to a high population density of solitaria they can change into the opportunistic gregaria form, and an outbreak will result, if the local food plants have not been recently attacked by gregaria. The most important triggering event is water stress to the host plants in the outbreak foci which compromises plant defenses, leading to rapid population build-up and gregarization of the locusts. Concentration of scattered solitaria by convergent airflows may be another mechanism whereby outbreaks are initiated (Waloff 1966, 1972). Following food depletion in the foci, the gregarious locusts migrate en masse into surrounding areas of unstressed food plants. Given that these unstressed food plants have not been recently attacked by gregaria, the gregarious locusts, acting in concert, can induce favorable food quality in the plants. This may be accomplished by mass attack which overloads rapid response capability in the plants (cf. bark beetles) and maintains food quality near the constitutive level (alternative A). Alternatively, gregaria may produce offensive pheromones in their saliva or emit them into the air which cause an increase in food quality above the constitutive level (alternative B). This may be accomplished by causing the plants to reduce the concentration of defensive metabolites in their aerial parts, or to increase the levels of nutrients in them, or both. The airborne offensive pheromone and the gregarization pheromone (discussed above) may be one and the same. High metabolic rate and high levels of locomotory activity lead to inefficient conversion of plant to insect biomass and high damage to the food plants per unit converted. The opportunistic strategy is successful only on naive plants. After a short period, perhaps after a single mass attack, the plants become resistant. This would account for the marching behavior of gregarious nymphs (Uvarov 1977) which constantly exposes them to unattacked plants. With respect to alternative A (above), acquired resistance may be manifest as powerful delayed defensive responses in the remaining portions of attacked plants. In alternative B, which I favor, resistance may be due to a recognition process involving alteration of the receptivity of the plants to the offensive pheromones of the locusts following previous attack, such that increased nutritional quality is no longer stimulated by the pheromones. Indeed, following immunization, the offensive pheromone may stimulate decreased nutritional quality.

As plant populations become immunized, the swarm migrates into naive plant populations. The swarm increases in size as a result of high fecundity, non-


diapausing eggs, and gregarization of resident solitaria in invaded areas (see Key 1950). This leads to progressive invasion of areas remote from the original outbreak foci (see Uvarov 1977). The outbreak subsides when the swarm is unable to locate naive food-plant populations. At this time solitarization and dispersal to the solitaria habitat occur. After a number of years (I suggest 3 to 4 yr) the plants lose their acquired resistance and again become vulnerable to invasion by gregaria. Thus, there is potential for cyclicity in the population dynamics of locusts, with the cycle period dependent on the time span of induction plus relaxation of acquired resistance. I further suggest that the periodicity is rarely manifest because the average time span between triggering events is long compared to the induction-relaxation time. This scheme has similarities to that proposed by Berry- man and others for the initiation and propagation of bark beetle epidemics (see Berryman 1978). At low population densities, the beetles persist in individual trees weakened by lightning strike, disease, ice and wind storms, and other factors, but are unable to successfully attack healthy trees. At high population density, however, the beetles can overcome healthy trees. So, any event which leads to the formation of a large local population, such as temporary weather stress of the trees or damage from cultural operations, can result in a self- sustaining outbreak of beetles that progressively invades areas of healthy trees far removed from the site of initiation.

The population ecology and phase transformation of armyworms (ground- feeding noctuid moths) share many striking similarities with those of locusts. The African armyworm Spodoptera exempta is a well-studied example. Outbreaks are of the eruptive type (Khasimuddin 1981a; Odiyo 1981). African armyworms feed mainly on grasses and sedges (Odiyo 1981). At low densities the larvae are mainly green in color, solitary, sluggish, inconspicuous, and have low feeding rates. During population outbreaks, the larvae are predominantly black in color, gregarious, active, conspicuous, developmentally synchronized, highly voracious, and exhibit marching behavior (Faure 1943 b; Rose 1975; Khasimuddin 1981 a, 1981 b; Odiyo 1981). The gregarious phase grows faster than the solitary phase (Rose 1975). Transition between forms can occur in one generation. When raised under crowded conditions gregarious-phase larvae are produced, whereas when raised singly, solitary-phase larvae result (Khasimuddin 1981 a). Dark-colored, active, fast-growing forms of other species of noctuid moth larvae can be induced by crowding (Faure 1943a; Matthee 1947; Iwao 1962). Eggs of armyworms are laid in masses (Rose 1975). Adults from gregarious larvae tend to migrate readily, while moths from solitary larvae usually oviposit in the same general area that they were raised in, to produce endemic populations (Odiyo 1981). During outbreak years, migration of adults leads to progressive shifts in the position of outbreak epicen- ters at monthly intervals (the generation time; Odiyo 1981). Rose (1975) found restricted outbreaks in two areas 150 km apart, in both of which heavy out-of- season hailstorms had occurred 2 to 3 wk before the outbreaks were reported. This suggests that food-plant stress may be important in initiating outbreaks of the African armyworm. I suggest the same general scheme to explain population outbreaks and phase polymorphism in armyworms as that suggested for locusts.


Incipient Phase Polymorphism

The appearance of dark-colored forms at high population density in the field or when raised under crowded conditions in the laboratory is not unique to locusts (Orthoptera:Acrididae) and armyworms (Lepidoptera:Noctuidae). It has been observed in many other herbivorous insects in the orders Lepidoptera and Or- thoptera, namely, larch budmoth (Tortricidae; Day and Baltensweiler 1972; Bal- tensweiler et al. 1977); species of Plusiidae (Long 1953); Saturniidae (Long 1953; Iwao 1968); Notodontidae, Geometridae, Sphingidae (Iwao 1968; Futuyma et al. 1981; Mitter and Futuyma 1983); species of Noctuidae not normally classified as armyworms (Long 1953; Iwao 1968); katydids (Tettigoniidae; Chopard 1949); walkingstick insects (Phasmidae; Key 1957); and a cricket (Gryllidae; Fuzeau- Braesch 1960, 1968). The dark crowded larvae usually grow faster, are more active, and have lower mortality than the pale solitary forms when raised on their food plants. Interestingly, the species displaying this incipient phase polymorphism are, where known, those whose populations are prone to erupt to high levels. Usually they lay their eggs in clusters and the larvae have a colonial or semicolonial habit at least part of the time.

The low-density forms are pale-colored and relatively inactive whereas the high-density forms are dark-colored and active possibly because they are stealthy and opportunistic, respectively. The stealthy strategy is to minimize damage suffered by the plant per unit of plant biomass converted to insect biomass, to avoid inducing defensive responses in the plant. In other words, stealthy forms or species are ECI (table 1) maximizers. This strategy requires low levels of physical activity and low metabolic rate. Low metabolic rate is promoted by low body temperature so the low-density forms are pale-colored possibly to reduce incident radiation load and enable them to readily maintain a low body temperature. On the other hand, the strategy of opportunists is "hit and run." This strategy requires high levels of physical activity and a high metabolic rate. Opportunists may attempt to maximize the rate of conversion of plant material to insect biomass (RCI) rather than the efficiency of conversion (see Smith 1976). Higher rates of conversion are promoted by higher body temperatures so the high-density forms are dark-colored possibly to increase absorption of solar radiation and help maintain high body temperature (table 1).

To date, the nutritional ecology of herbivores and their various nutritional indices such as growth rates, efficiencies of conversion, and food digestibility have been examined largely in terms of the degree of food-plant specialization of the herbivores (monophagous, oligophagous, polyphagous) and the growth form (herb, shrub, tree) of their food plants (Scriber and Feeny 1979; Scriber and Slansky 1981; Scriber 1983). The present analysis suggests that the attack strategies of herbivores may also be important, perhaps of overriding importance, in determining these parameters. In general, we should expect innocuous species, with low, relatively invariant population levels, to exhibit high ECI, food digestibility (AD), and low feeding rates (CL), and RCI. On the other hand, species with variable populations (pest species) should exhibit low ECI, AD, and variable CI and RCI (see table 1).


Other striking changes in herbivore "quality" besides color changes have been observed between high- and low-density populations, and during population increases and declines, for insects, voles, and grouse (see Haukioja and Hakala 1975; Rhoades 1983a). Usually, the forms predominant in increasing populations are more fecund, more active, and have lower mortality than those characteristic of declining populations. These changes can be interpreted as merely the result of changing food quality during increase and decline but a component of incipient phase polymorphism is probable.

In cases of locusts, armyworms, and other insects previously discussed, where transition between light and dark morphs can be induced in one or a few generations by crowding, it is clear that the changes are the result of plasticity rather than differential survival of light or dark genotypes. The same is probably generally true of phase change and other changes in herbivore quality during population fluctuations.

Douglas-Fir Tussock Moth, Gypsy Moth, and Other Nonmigratory Pest Species: A Population Model

The association between herbivore population variability and migratory ability has been emphasized in this paper. However, the females of some tree-feeding species of insects with variable population levels have reduced wings or, even if they possess developed wings, they rarely fly. This is true of Douglas-fir tussock moth (Brookes et al. 1978); gypsy moth (Gerardi and Grimm 1979); many other species of Lymantriidae (Browne 1968); and fall cankerworm (Geometridae; Mit- ter and Futuyma 1983). Similar to most insects with variable populations, these insects deposit their eggs in masses. The flightless habit possibly arose as part of an alternating stealthy-opportunistic attack strategy as in the following scheme. At low density, the insects are stealthy. Upon hatching from the egg mass, the larvae immediately disperse by crawling or silk ballooning to lead a solitary life utilizing adaptations to suppress recognition and minimize plant damage as previously discussed. There comes a point when the plants have lost their acquired immunity to the opportunistic form, upon which the insects switch to that strategy. Instead of dispersing, the opportunistic larvae maintain loose aggregations and utilize adaptations to increase the nutritional quality of the food plants. An outbreak erupts, the plants become resistant, and the outbreak subsides. Surviv- ing insects revert to the stealthy form, reestablishing an endemic population. Induction of the opportunistic form and rapid population build-up may be aided by physical stress of the food plants (see V. I. Benkevitch [1961-1964], M. G. Khanislamov et al. [1962] as summarized in Campbell et al. [1978], Bess [1947], Spurr et al. [1947], Campbell and Sloan [1977] for evidence of association between outbreaks of gypsy moth with weather and other stresses of host trees; see Brookes et al. [1978] for similar effects on Douglas-fir tussock moth). This scheme predicts a fundamental cyclicity in the population dynamics of the insects, which is determined by the induction and relaxation times of acquired immunity, but which is modulated in amplitude and exact timing by physical stress of the host plants. The population dynamics of Douglas-fir tussock moth match this pattern remarkably well. In the western United States, outbreaks of tussock moth have


occurred on average every 8 to 9 yr (Mason and Luck 1978). Stage (1978) found that the timing of outbreaks can be predicted if it is assumed that there is a minimum waiting time of 7 yr from the date of the previous outbreak, to which is added a short random time period in which a critical event occurs, with an annual probability of 0.3. The waiting time is probably induction plus relaxation of acquired host resistance and the critical event is probably physical stress of the trees. A tendency for female Sightlessness to evolve in insects with sessile stealthy and opportunistic phases would be expected since this allows increased allocation of resources to egg production. This tendency would be reinforced in species which attack trees, since large food-plant size would allow effective within-plant dispersion of stealthy-phase larvae, and long food-plant life span may allow adaptation over many generations of the insects to the properties of individual trees if females oviposit on the same trees on which they fed as larvae (see Alstad and Edmunds 1983).

There are two main differences between the proposed tussock moth scheme and that previously outlined for locusts and armyworms. First, in the tussock moth case, epidemics erupt from and collapse to endemic populations, rather than proceeding under the influence of migration. Second, in the tussock moth case, the average time span between events stressful to the food plants is short compared to plant defensive induction-relaxation time, whereas in the case of locusts and armyworms the opposite was proposed. Thus, for tussock moth the fundamental periodicity of the cycle is revealed, whereas for locusts and armyworms it is normally masked. The interaction between the timing of events stressful to food plants and the time span of induction plus relaxation of acquired resistance to herbivore attack may explain other cases of eruptive and cyclic herbivore population behavior (see Rhoades 1983a, fig. 3, and discussion). It is interesting to note that major outbreaks of tussock moth have been recorded mainly at interior sites with extreme continental climates in the western United States (Beckwith 1978), where host stress would be expected to be more intense and more frequent than in coastal regions with mesic maritime climates. Major outbreaks have not been reported in coastal areas, although low-density endemic populations of tussock moth are present. An alternative explanation for the geographical distribution of Douglas-fir tussock moth outbreaks is that there are genetic differences in defensive responses between plants in the two regions. Based partly on the work of Haukioja and coworkers (Haukioja 1980), I suggested that the propensity of herbivore outbreaks to occur in regions of harsh climate (high latitude and al- titude) was due to recent climatic warming which enabled herbivores to invade the cooler regions of their range. Because they had been partly protected from the herbivores by their physical location, the plants in the cooler regions were suggested to have evolved weak rapid-defensive responses but strong delayed responses, and this promoted herbivore population instability (see Rhoades 1983a, pp. 191-193).

Parthenogenesis and Population Stability

J. C. Schultz (personal communication) has suggested that parthenogenesis may be particularly common in herbivorous insect species with variable popula-


tion levels. Among such species, parthenogenesis occurs frequently in fall cankerworm (Mitter et al. 1979; Mitter and Futuyma 1983), pest species of sawflies (Koehler 1956; Wagner and Benjamin 1981), and aphids (Way 1973), and occa- sionally in gypsy moth (Gerardi and Grimm 1979) and locusts (Kennedy 1956; Uvarov 1966). Facultative or obligate parthenogenesis also occurs in many other herbivorous species of Orthoptera, Homoptera, Coeloptera, Lepidoptera, and Diptera (Cockayne 1938; White 1973; Went 1982). It is conceivable that thelytokous (female-producing) parthenogenesis may be more common among opportunistic species or phases than among stealthy herbivores since by dispensing with males, opportunists could achieve a more rapid population build-up than with a sexually reproducing population (table 1). Facultative thelytokous parthenogenesis, combined with facultative winglessness, would be an ideal combination of adaptations for opportunists. These are precisely the characteristics, together with aggregation during population build-up, shown by many pest aphid species (Bonnemaison 1968; Dixon 1973; Way 1973).

CONCLUDING REMARKS

The concept of alternative herbivore attack strategies, developed here, explains variability and invariance in herbivore population levels. It also explains phase polymorphism in herbivores and suggests that qualitative changes observed in many herbivore species during population fluctuations are incipient phase transi- tions. Developed to account for the properties of small herbivore species, these ideas may also be applicable to large herbivores. For instance, sloths may be slothful, howler monkeys may howl, and koalas may be sedentary (see Montgom- ery 1978), because they are stealthy.

It can be shown by simple arithmetic that for two species of organisms with the same mean reproductive rate, when arithmetically averaged over several generations, but with different variances in reproductive rate from one generation to the next, the species with lower variance will leave more progeny in the long term (Gillespie 1974; Root and Kareiva 1983). In an invariant environment we would therefore expect the evolution of invariant reproductive rates across generations for all species. The continuing existence of herbivore species with highly variable population levels and reproductive rates occurs in spite of this leveling process, and is probably maintained by variations in the environment which create opportunities for opportunists to exploit.

Herbivore population dynamics are probably better understood in terms of varying quality and quantity of both plants and herbivores (see Haukioja and Hakala 1975; Denno and McClure 1983) than in terms of only quantitative changes in amount of vegetation and numbers of herbivores. In natural systems, herbivores rarely reach outbreak levels. They usually consume only a small fraction of the net primary production (Hairston et al. 1960; Mattson and Addy 1975; Fox and Morrow 1983). Even in grasslands, which tend to be heavily utilized compared to most plant communities, consumption usually ranges only from 25% to 50% (Dyer et al. 1982). Thus, in terms of food quantity, herbivores must seldom compete with each other. In addition, aggregation with conspecifics during feeding clearly benefits many herbivores. (Besides cases previously discussed see Teale 1949;


Long 1955; Kalkowski 1958; Ghent 1960; Iwao 1968; Way and Cammell 1970; Shiga 1976; Hikada 1977; Kalin and Knerer 1977; Peters and Barbosa 1977; Weyh and Maschwitz 1978; Fitzgerald and Edgerley 1979; Capinera 1980; Haukioja 1980; Tsubaki 1981; and references therein.) Positive interactions in terms of food choice and reproductive success are also common between herbivore species. They occur between bark beetle species (Wood and Bedard 1977; Birch et al. 1980; Borden 1982); between boring beetles and defoliators (Staley 1965; Nichols 1968; Dunbar and Stephens 1976; Schultz and Allen 1977; Berryman and Wright 1978); between insects inhabiting Heliconia bracts (Seifert and Seifert 1976); between a folivorous grasshopper and a stem-girdling beetle (Lewis 1979); between stem-girdling beetles and other insects (Mares et al. 1977); among ungulate species (Bell 1971; McNaughton 1976); between prairie dogs and bison (Coppock et al. 1983); and possibly between snail species (Hershey 1983). Therefore, that simple competition theory based largely on changes in amount of resources has proved of little value in explaining co-occurrence and other patterns of herbivores (Hairston et al. 1960; Lawton and Strong 1981; Strong 1983) is not surprising.

On the other hand, the evidence that herbivore attack can lead to decreased plant nutritional quality, caused by plant defensive response, is substantial and growing rapidly. These changes in quality can be induced by low levels of damage (see Rhoades 1983b). There is also evidence that individual herbivores and aggregations of herbivores can actively induce favorable changes in food-plant quality. These changes in food quality can be expected to affect both the inducing species and other herbivore species. Therefore, although negative interactions between herbivores mediated through decreased food quantity are probably rare, both negative and positive interactions mediated through changes in food quality are probably common. I therefore suggest that the concept of competition among herbivores is of limited value because, by definition, it encompasses only negative interactions. Interactions among herbivores are more comprehensively described in terms of the dual, complementary concepts of interference and facilitation. Within species of stealthy herbivores and among stealthy species that feed on the same tissue, interactions are mediated through interference, whereas the interactions within and among species of opportunistic herbivores are mediated through facilitation during population increase, and delayed interference during population decrease. Stealthy herbivores should have little effect on the population dynamics of opportunists. On the other hand, opportunists can be expected to facilitate or interfere with population growth of stealthy herbivores that feed on the same tissue, during population increase or decrease of the opportunists, respectively. Interactions among opportunists that feed on different tissues and effects of opportunists on stealthy herbivores that feed on different tissues should depend on the degree to which reallocation of defensive metabolites occurs between attacked and nonattacked tissues.

There is a clear analogy between the opportunism-stealth dichotomy of attack strategies and the r-K continuum of life history strategies developed by MacAr- thur and Wilson (1967). In many ways, opportunism and stealth can be considered to be special cases of r- and K-selected life histories, respectively (cf. table 1 with Pianka 1970, table 1; see also Southwood 1976). The main difference is that in the


present scheme individual stealthy and opportunistic herbivores both strongly interact with other herbivores, in the first case by interference, and in the second case by facilitation and interference. In r-K theory, individual K-strategists compete (interfere) strongly with other organisms at the same trophic level, but r- strategists interact with them minimally.

In the same way that changes in plant quality are proving to be important in herbivore-plant interactions, changes in prey behavior (quality) may also be important in predator-prey interactions in general. Symmetrical and asymmetrical facilitation are common among predators, for example, wolf packs, mixed-species foraging flocks of birds, and army ant-bird associations. According to Potts (1981), primary outbreaks of crown-of-thorns starfish preying on coral polyps are probably caused by aggregations of adults from normal low-density populations, or enhanced larval recruitment, following abnormal storms and other distur- bances. Antonelli and Kazarinoff (1983) have found that the population dynamics of the starfish can be successfully modeled if they include a quadratic "coopera- tive" term in the predator equation. This underlines the importance of aggregation and suggests mutual facilitation within aggregations of the starfish. By analogy with plant-herbivore systems, I suggest that the population dynamics of crown-of- thorns starfish can be understood in terms of transition by the starfish between stealthy and opportunistic attack strategies, in response to stress of coral polyps which lowers their defensive capacity. I further suggest that induced defensive responses of coral will prove important in the population dynamics of the starfish.

Facilitation between producer species may also be common (e.g., see Connell and Slatyer 1977). In my view, the present controversy concerning the importance of competition in structuring communities (see Lewin 1983a, 1983b; Salt 1984) is a direct result of the narrow focus of competition theory on negative influences among organisms resulting from mutually induced decrease in resource quantity. If community structure is determined by a combination of interference, facilitation, changes in resource quantity and quality, and alternative strategies of organisms at each trophic level, the observation that communities do not conform to the patterns predicted by competition theory is to be expected. This viewpoint is far different from that expressed by some participants in the current competition controversy (see Lewin 1983a, 1983b), who would have us believe that ecological generalizations are of little value, and that communities are structured largely by random factors.

SUMMARY

A consideration of known defensive attributes of plants, and others that can be reasonably postulated, leads to the description of two alternative offensive strategies of herbivores, termed stealth and opportunism, to counter plant defenses. Stealthy herbivores display adaptations to minimize damage and defensive responses of their food plants. On the other hand, opportunists take advantage of circumstances, such as physical stress and loss of acquired resistance of food plants, that impair plant defensive capability. In addition, opportunists themselves display adaptations, such as mass-attack behavior, that stress food plants.


Herbivore species with low, relatively invariant population levels may be stealthy, whereas those with variable population levels may be opportunistic. Phase changes in herbivores may be the result of transition between the two strategies. It is suggested that interactions within and among species of herbivores are better understood in terms of interference and facilitation than in terms of competition.

ACKNOWLEDGMENTS

I thank Nancy E. Beckage, Alan A. Berryman, Mel I. Dyer, Peter W. Miles, and unidentified reviewers for many valuable suggestions and Lynn Erckmann for the illustrations. This work was supported by National Science Foundation grant BSR-8307581 to G. H. Orians and D. F. Rhoades and by Whitehall Foundation grant 83548 to D. F. Rhoades.

LITERATURE CITED

Abeles, F. B. 1973. Ethylene in plant biology. Academic Press, New York. Allison, M. J., and H. M. Cook. 1981. Oxalate degradation by microbes of the large bowel of

herbivores: the effect of dietary oxalate. Science 212:675-676. Alstad, D. N., and G. F. Edmunds, Jr. 1983. Adaptation, host specificity, and gene flow in the black

pineleaf scale. Pages 413-426 in R. F. Denno and M. S. McClure, eds. Variable plants and herbivores in natural and managed systems. Academic Press, New York.

Antonelli, P. L., and N. D. Kazarinoff. 1983. Starfish predation of a growing coral reef community. J. Theor. Biol. 107:667-684.

Baker, R. R. 1983. Insect territoriality. Annu. Rev. Entomol. 28:65-89. Baldwin, I. T., and J. C. Schultz. 1983. Rapid changes in tree leaf chemistry induced by damage:

evidence for communication between plants. Science 221:277-279. Baltensweiler, W. 1968. The cyclic population dynamics of the grey larch tortrix, Zeraphera griseana

Huber (=Semasia diniana Guenee) (Lepidoptera: Tartricidae). Pages 88-97 in T. R. E. Southwood, ed. Insect abundance. Blackwell, Oxford.

Baltensweiler, W., G. Benz, P. Bovey, and V. Delucchi. 1977. Dynamics of larch bud moth populations. Annu. Rev. Entomol. 22:79-100.

Baltensweiler, W., and A. Fischlin. 1979. The role of migration for the population dynamics of the larch bud moth, Zeiraphera diniana GN. (Lep. Tortricidae). Mitt. Schweiz. Entomol. Ges. 52:259-271.

Beckwith, R. C. 1978. Geographic range. in M. H. Brookes, R. W. Stark, and R. W. Campbell, eds. The Douglas-fir tussock moth: a synthesis. US Dep. Agric. For. Serv. Sci. Education Agency Tech. Bull. 1585.

Bell, R. H. V. 1971. A grazing system in the Serengeti. Sci. Am. 225(1):86-93. Benz, G. 1974. Negative Ruckkoppelung durch Raum- und Nahrungs-Konkurrenz sowie zyklische

Veranderung der Nahrungsgrundlage als Regelprinzip in der Populationsdynamik des gruen Larchenwicklers, Zeiraphera diniana (Guenee) (Lep. Tortricidae). Z. Angew. Entomol. 76:196-228. 1977. Eucarpia/IOBC working group breeding for resistance to insects and mites. Bull. SROP, 1977/3. (Report 1st meeting held at Wageningen, The Netherlands, 7 to 9 Dec. 1976, pages 155-159).

Berenbaum, M. 1980. Adaptive significance of midgut pH in larval Lepidoptera. Am. Nat. 115:138- 146. 1981. Effects of linear furanocoumarins on an adapted specialist insect (Papilio polyxenes). Ecol. Entomol. 6:345-351.

Bernays, E. A. 1981. Plant tannins and insect herbivores: an appraisal. Ecol. Entomol. 6:353-360.


Bernays, E. A., and S. Woodhead. 1982. Plant phenols utilized as nutrients by a phytophagous insect. Science 216:201-202.

Berryman, A. A. 1969. Responses of Abies grandis to attack by Scolytus ventralis (Coleoptera: Scolytidae). Can. Entomol. 101:1033-1041. 1972. Resistance of conifers to invasion by bark beetle-fungus associations. BioScience 22:598-602. 1978. A synoptic model of the lodgepole pine/mountain pine beetle interaction and its potential application in forest management. Pages 98-105 in D. L. Kibbee, A. A. Berryman, G. D. Amman, and R. W. Stark, eds. Theory and practice of mountain pine beetle management in lodgepole pine forests. Forest, Wildlife and Range Experiment Station, University of Idaho, Moscow, Idaho. 1979. Dynamics of bark beetle populations: analysis of dispersal and redistribution. Bull. Soc. Entomol. Suisse 52:227-234.

Berryman, A. A., and W. Baltensweiler. 1981. Population dynamics of forest insects and the management of future forests. Proc. IUFRD World Congr. 17:423-430.

Berryman, A. A., and L. C. Wright. 1978. Defoliation, tree condition and bark beetles. Pages 81-87 in M. H. Brookes, R. W. Stark, and R. W. Campbell, eds. The Douglas-fir tussock moth: a synthesis. US Dep. Agric. For. Serv. Sci. Education Agency Tech. Bull. 1585.

Bess, H. A. 1947. Relation of forest and site to gypsy moth abundance. Harv. For. Bull. 22:7-26. Birch, M. C., P. Svihra, T. D. Paine, and J. C. Miller. 1980. Influence of chemically mediated behav-

ior on host tree colonization by four cohabiting species of bark beetles. J. Chem. Ecol. 6:395-414.

Bonnemaison, L. 1968. L'effet de groupe chez les aphides. Colloq. Int. Cent. Natl. Rech. Sci. 173:213-236.

Borden, J. H. 1982. Aggregation pheromones. Pages 74-139 in J. B. Mitton and K. B. Sturgeon, eds. Bark beetles of North American conifers. University of Texas Press, Austin.

Brattsten, L. B. 1979. Biochemical defense mechanisms in herbivores against plant allelochemicals. Pages 200-270 in G. A. Rosenthal and D. H. Janzen, eds. Herbivores: their interaction with secondary plant metabolites. Academic Press, New York.

Brookes, M. H., R. W. Stark, and R. W. Campbell. 1978. The Douglas-fir tussock moth: a synthesis. US Dep. Agric. For. Serv. Sci. Education Agency Tech. Bull. 1585.

Browne, F. G. 1968. Pests and diseases of forest plantation trees. Clarendon, Oxford. Bryant, J. P. 1981. Phytochemical deterrence of snowshoe hare browsing by adventitious shoots of

four Alaskan trees. Science 213:889-890. Butler, C. G., and J. M. Innes. 1936. A comparison of the rate of metabolic activity in the solitary and

migratory phases of Locusta migratoria. Proc. R. Soc. Lond. B. 119:296-304. Campbell, R. W., L. C. Levitan, E. Sobecki, and M. F. Tardiff. 1978. Population dynamics of the

gypsy moth: an annotated bibliography. US Dep. Agric. For. Serv. Gen. Tech. Rep. NE- 48:1-124.

Campbell, R. W., and R. J. Sloan. 1977. Release of gypsy moth populations from innocuous levels. Environ. Entomol. 6:323-330.

Capinera, J. L. 1980. A trail pheromone from the silk produced by larvae of the range caterpillar Hemileuca olivae (Lepidoptera: Saturniidae) and observations on aggregation behavior. J. Chem. Ecol. 6:655-664.

Capinera, J. L., and W. J. Roltsch. 1980. Response of wheat seedlings to actual and simulated migratory grasshopper defoliation. J. Econ. Entomol. 73:258-261.

Carroll, C. R., and C. A. Hoffman. 1980. Chemical feeding deterrent mobilized in response to insect herbivory and counter adaptation by Epilachna tredecimnota. Science 209:414-416.

Carter, W. 1973. Insects in relation to plant disease. Wiley, New York. Cates, R. G. 1975. The interface between slugs and wild ginger: some evolutionary aspects. Ecology

56:391-400. Cates, R. G., and H. Alexander. 1982. Host resistance and susceptibility. Pages 212-263 in J. B.

Mitton and K. B. Sturgeon, eds. Bark beetles in North American conifers. University of Texas Press, Austin.


Cherrett, J. M. 1972. Chemical aspects of plant attack by leaf-cutting ants. Pages 13-24 in J. B. Harborne, ed. Phytochemical ecology. Academic Press, New York.

Chopard, L. 1949. L'6tat actuel de la question des phases chez les insects. Annee Biol. 25:105-109. Clark, W. C., D. P. Jones, and C. S. Holling. 1978. Patches, movements, and population dynamics in

ecological systems. Pages 385-432 in J. H. Steele, ed. Spatial patterns in plankton communities. Plenum, New York.

Cockayne, E. A. 1938. The genetics of sex in Lepidoptera. Biol. Rev. 13:107-132. Cold Spring Harbor Symposia on Quantitative Biology. 1957. Population studies: animal ecology and

demography. Long Island Biological Association. Cold Spring Harbor Symp. Quant. Biol. 22. Connell, J. H., and R. D. Slatyer. 1977. Mechanisms of succession in natural communities and their

role in community stability and organization. Am. Nat. 111: 1119-1144. Coppock, D. L., J. E. Ellis, J. K. Detling, and M. I. Dyer. 1983. Plant-herbivore interactions in a

North American mixed-grass prairie. II. Responses of bison to modification of vegetation by prairie-dogs. Oecologia 56:10-15.

Cornell, H. V. 1983a. The secondary chemistry and complex morphology of galls formed by the Cynipinae (Hymenoptera): why and how. Am. Midl. Nat. 110:225-234. 1983b. Why and how gall wasps form galls: cynipids as genetic engineers? Bull. R. Entomol. Soc. Lond. 9:53-58.

Coster, J. E., and P. C. Johnson. 1979. Dispersion patterns of Dendroctonusfrontalis and its predator Thanasimus dubius: influence of behavioral chemicals. Mitt. Schweiz. Entomol. Ges. 52:309-322.

Coutts, M. P. 1968. Rapid physiological change in Pinus radiata following attack by Sirex noctilio and its associated fungus, Amylostrereum sp. Aust. J. Sci. 30:275-277. 1969a. The mechanism of pathogenicity of Sirex noctilio on Pinus radiata I. Effects of the symbiotic fungus Amylostrereum sp. (Thelephoraceae). Aust. J. Biol. Sci. 22:915-924. 1969b. The mechanism of pathogenicity of Sirex noctilio on Pinus radiata II. Effects of S. noctilio mucus. Aust. J. Biol. Sci. 22:1153-1161.

Day, K. R., and W. Baltensweiler. 1972. Change in proportion of larval colourtypes of the larchform Zeiraphera diniana when reared on two media. Entomol. Exp. Appl. 15:287-298.

Dearn, J. M. 1974a. Phase transformation and chiasma frequency variation in locusts I. Schistocerca gregaria. Chromosoma 45:321-338. 1974b. Phase transformation and chiasma frequency variation in locusts II. Locusta migratoria. Chromosoma 45:339-352.

Denno, R. F., and M. S. McClure, eds. 1983. Variable plants and herbivores in natural and managed systems. Academic Press, New York.

Detling, J. K., and M. I. Dyer. 1981. Evidence for potential growth regulators in grasshoppers. Ecology 62:485-488.

Detling, J. K., C. W. Ross, M. H. Walmsley, D. W. Hilbert, C. A. Bonilla, and M. I. Dyer. 1981. Examination of North American bison saliva for potential plant growth regulations. J. Chem. Ecol. 7:239-246.

Dieleman, F. L. 1969. Effects of gall midge infestation on plant growth and growth regulating substances. Entomol. Exp. Appl. 12:745-749.

Dixon, A. F. G. 1973. Biology of aphids. Edward Arnold, London. Dowd, P. F., C. M. Smith, and T. C. Sparks. 1983. Detoxification of plant toxins by insects. Insect

Biochem. 13:453-468. Dropkin, V. H. 1979. How nematodes induce disease. Pages 219-238 in J. G. Horsfall and E. B.

Cowling, eds. Plant disease. Vol. 4. Academic Press, New York. Dunbar, D. M., and G. R. Stephens. 1976. The bionomics of the twolined chestnut borer. Pages 73-83

in J. F. Anderson and H. K. Kaya, eds. Perspectives in forest entomology. Academic Press, New York.

Dyer, M. I. 1980. Mammalian epidermal growth factor promotes plant growth. Proc. Natl. Acad. Sci. USA 77(8):4836-4837.

Dyer, M. I., and V. G. Bokhari. 1976. Plant-animal interactions: studies of the effects of grasshopper grazing on blue grama grass. Ecology 57:762-772.

Dyer, M. I., J. K. Detling, D. C. Coleman, and D. W. Hilbert. 1982. The role of herbivores in


grasslands. Pages 255-295 in J. R. Estes, R. J. Tyrl, and J. N. Brunken, eds. Grasses and grasslands. University of Oklahoma Press, Norman.

Ehrlich, P. R., and P. H. Raven. 1965. Butterflies and plants: a study in coevolution. Evolution 18:586-608.

Ellis, P. E., and S. Gillett. 1968. Social aggregation and an airborne gregarising factor in locusts. Colloq. Int. Cent. Natl. Rech. Sci. 173:173-183.

Engelbrecht, L. 1971. Cytokinin activity in larval infected leaves. Biochem. Physiol. Pflanz. 162:8-27. Engelbrecht, L., V. Orban, and W. Heese. 1969. Leaf-miner caterpillars and cytokinins in the green

islands of autumn leaves. Nature 223:319-321. Faure, J. C. 1943a. The phase changes of the lesser armyworm (Laphygma exigua (Hubn.)). Farming

S. Afr. 18:69-78. 1943b. Phase variation in the armyworm Laphygma exempta (Walk.). Sci. Bull. Dep. Agric. S. Afr. 234.

Feeny, P. P. 1970. Seasonal changes in oak leaf tannins and nutrients as a cause of spring feeding by winter moth caterpillars. Ecology 51:565-581. 1975. Biochemical coevolution between plants and their insect herbivores. Pages 3-19 in L. E. Gilbert and P. H. Raven, eds. Coevolution of animals and plants. University of Texas Press, Austin. 1976. Plant apparency and chemical defense. Rec. Adv. Phytochem. 10: 1-40.

Finerty, J. P. 1980. The population ecology of cycles in small mammals. Yale University Press, New Haven, Conn.

Fitzgerald, T. D., and J. S. Edgerly. 1979. Specificity of trail markers of forest and eastern tent caterpillars. J. Chem. Ecol. 5:564-574.

Fox, L. R., and P. A. Morrow. 1983. Estimates of damage by herbivorous insects on eucalyptus trees. Aust. J. Ecol. 8:139-147.

Fraenkel, G. S. 1959. The raison d'ftre of secondary plant substances. Science 129:1466-1470. Freeland, W. J. 1980. Insect flight times and atmospheric hydrocarbons. Am. Nat. 116:736-742. Freeland, W. J., and D. H. Janzen. 1974. Strategies in herbivory by mammals: the role of plant

secondary compounds. Am. Nat. 108:269-289. Futuyma, D. J., S. L. Leipertz, and C. Mitter. 1981. Selective factors affecting clonal variation in fall

cankerworm, Alsophila pometaria (Lepidoptera:Geometridae). Heredity 47:161-172. Fuzeau-Braesch, S. 1960. Etude biologique et biochimique de la pigmentation chez un insecte =

Gryllus bimaculatus de Geer (Orthopteres, Gryllidae). Bull. Biol. Fr. Belg. 94:527-629. 1968. L'effet de groupe chez le grillon: niveaux et relativity de ses manifestations. Colloq. Int. Cent. Natl. Rech. Sci. 173:115-125.

Fye, R. E. 1974. Populations defined and approaches to measuring populations, density, dispersal and dispersion. Pages 46-61 in F. G. Maxwell and F. A. Harris, eds. Proc. of the summer institute on biological control of plant insects and diseases. University of Mississippi Press, Jackson.

Gerardi, M. H., and J. K. Grimm. 1979. The history, biology, damage and control of the gypsy moth Porthetria dispar (L.). Associated University Presses, London.

Ghent, A. W. 1960. A study of the group-feeding behaviour of larvae of the jack pine sawfly, Neodiprion pratti banksianae Roh. Behaviour 16:110-148.

Gibson, R. W., and J. A. Pickett. 1983. Wild potato repels aphids by release of aphid alarm pheromone. Nature 302:608-609.

Gillespie, J. H. 1974. Natural selection for within-generation variance in offspring number. Genetics 76:601-608.

Gillett, S. D. 1968. Airborne factors affecting the grouping behaviour of locusts. Nature 218:782-783. . 1983. Primer pheromones in the desert locust. Anim. Behav. 31:221-230.

Gillett, S. D., and M. L. Phillips. 1977. Faeces as a source of a locust gregarisation stimulus. Effects on social aggregation and on cuticular colour of nymphs of the desert locust, Schistocerca gregaria (Forsk.). Acrida 6:279-286.

Goodman, R. N. 1980. Defenses triggered by previous invaders: bacteria. Pages 305-317 in J. G. Horsfall and E. B. Cowling, eds. Plant disease. Vol. 5. Academic Press, New York.

Hairston, N. G., F. E. Smith, and L. B. Slobodkin. 1960. Community structure, population control and competition. Am. Nat. 94:421-425.


Hamilton, R. I. 1980. Defenses triggered by previous invaders: viruses. Pages 279-303 in J. G. Horsfall and E. B. Cowling, eds. Plant disease. Vol. 5. Academic Press, New York.

Hanover, J. W. 1972. Factors affecting the release of volatile chemicals by forest trees. Mitt. Forstl. Bundes-Versuchsanst. Wien 97:625-645.

Harborne, J. B. 1978. Biochemical aspects of plant and animal coevolution. Academic Press, New York.

Haukioja, E. 1980. On the role of plant defenses in the fluctuation of herbivore populations. Oikos 35:202-213.

Haukioja, E., and T. Hakala. 1975. Herbivore cycles and periodic outbreaks. Formulation of a general hypothesis. Rep. Kevo Subarctic Res. Stat. 12:1-9.

Hershey, A. E. 1983. Competition among two co-existing snails: a laboratory experimental analysis. Bull. Ecol. Soc. Am. 64:89.

Higgins, K. M., J. E. Browns, and B. A. Haws. 1977. The black grass bug (Lapbops hesperius Uhler): its effect on several native and introduced grasses. J. Range Manage. 30:380-384.

Hikada, T., ed. 1977. Adaptation and speciation in the fall webworm. Kodansha, Tokyo. Hladik, C. M. 1978. Adaptive strategies of primates in relation to leaf-eating. Pages 373-395 in G. G.

Montgomery, ed. The ecology of arboreal folivores. Smithsonian Institution, Washington, D.C.

Hodson, A. C. 1941. An ecological study of the forest tent caterpillar, Malacosoma disstria Hbn. in northern Minnesota. Univ. Minn. Agric. Exp. Stn. Minn. Tech. Bull. 148:1-55.

Hori, K. 1976. Plant growth-regulating factor in the salivary glands of several heteropterous insects. Comp. Biochem. Physiol. 53B:435-438.

Hori, K., and M. Endo. 1977. Metabolism of ingested auxins in the bug Lygus disponsi: conversion of indole-3-acetic acid and gibberellin. J. Insect Physiol. 23:1075-1080.

Horsfall, J. G., and E. B. Cowling. 1980. Plant disease. Vol. 5. Academic Press, New York. Iwao, S. 1962. Studies on the phase variation and related phenomena in some lepidopterous insects.

Mem. Coll. Agric. Kyoto Univ. 84. 1968. Some effects of grouping in lepidopterous insects. Colloq. Int. Cent. Natl. Rech. Sci. 173:185-212.

Jaenike, J., D. A. Grimaldi, A. E. Sluder, and A. L. Greenleaf. 1983. cx-amanitin tolerance in mycophagous Drosophila. Science 221:165-166.

Jones, D. A. 1972. Cyanogenic glycosides and their function. Pages 103-124 in J. B. Harborne, ed. Phytochemical ecology. Academic Press, New York.

Kahn, D. M., and H. V. Cornell. 1983. Early leaf abscission and folivores: comments and considerations. Am. Nat. 122:428-432.

Kalin, M., and G. Knerer. 1977. Group and mass effects in diprionid sawflies. Nature 267:427-429. Kalkowski, W. 1958. Investigations on territorial orientation during ontogenetic development in

Hyponomeuta evonymellus L., Lepidoptera, Yponmeutidae, Part II. Folia Biol. Krakow 6:245-263.

Karban, R. 1983. Induced responses of cherry trees to periodical cicada oviposition. Oecologia 59:226-231.

Keith, L. B. 1963. Wildlife's ten-year cycle. University of Wisconsin Press, Madison. 1974. Some features of population dynamics in mammals. Proc. Int. Congr. Game Biol., 11th. 1973:17-58.

Kennedy, J. S. 1956. Phase transformation in locust biology. Biol. Rev. 31:349-370. Key, K. H. L. 1950. A critique on the phase theory of locusts. Q. Rev. Biol. 25:363-407.

. 1957. Kentromorphic phases in three species of phasmatodea. Aust. J. Zool. 5:247-284. Khasimuddin, S. 1981 a. Behavioral ecology of the African armyworm, Spodoptera exempta (Walker):

observations on population processes during a high-density outbreak. Insect Sci. Appl. 1:143-146. 1981 b. Phase variation and "off season" survival of the African armyworm, Spodoptera exempta (Walker) (Lepidoptera:Noctuidae). Insect Sci. Appl. 1:357-360.

Klocke, J. A., and B. G. Chan. 1982. Effects of cotton condensed tannin on feeding and digestion in the cotton pest Heliothis zea. J. Insect Physiol. 28:911-915.


Koehler, W. 1956. Acantholyda nemoralis Thoms. in the Silesian forests. Rocz. Nauk Lesnych. 15:3- 194.

Kuc, J. 1982. Induced immunity to plant disease. BioScience 32:854-860. Kuc, J., and F. L. Caruso. 1977. Activated coordinated chemical defense against disease in plants.

Am. Chem. Soc. Symp. Ser. 62:78-89. Lawton, J. H., and D. R. Strong, Jr. 1981. Community patterns and competition in folivorous insects.

Am. Nat. 118:317-338. Leonard, K. J. 1977. Selection pressures and plant pathogens. Ann. N.Y. Acad. Sci. 287:207-222. Levin, D. A. 1971. Plant phenolics: an ecological perspective. Am. Nat. 105:157-181. Lewin, R. 1983a. Santa Rosalia was a goat. Science 221:636-639.

. 1983b. Predators and hurricanes change ecology. Science 221:737-740. Lewis, A. C. 1979. Feeding preference for diseased and wilted sunflower in the grasshopper

Melanopus differentialis. Entomol. Exp. Appl. 26:202-207. Long, D. B. 1953. Effects of population density on larvae of Lepidoptera. Trans. R. Entomol. Soc.

Lond. 104:543-585. 1955. Observations on sub-social behavior in two species of lepidopterous larvae, Pieris brassicae L. and Plusia gamma L. Trans. R. Entomol. Soc. Lond. 11:421-437.

Loper, G. M. 1968. Effect of aphid infestation on the coumestrol content of alfalfa varieties differing in aphid resistance. Crop Sci. 8: 104-106.

MacArthur, R. H., and E. 0. Wilson. 1967. The theory of island biogeography. Princeton University Press, Princeton, N.J.

McFarlane, J. E., and M. H. W. Distler. 1982. The effect of rutin on growth, fecundity and food utilization in Acheta domesticus (L.). J. Insect Physiol. 28:85-88.

McIntyre, J. L. 1980. Defenses triggered by previous invaders: nematodes and insects. Pages 333-343 in J. G. Horsfall and E. B. Cowling, eds. Plant disease. Vol. 5. Academic Press, New York.

Mackay, P. A., and W. G. Wellington. 1977. Notes on the life history and habits of the red-backed sawfly Eriocampa ovata (Hymenoptera: Tenthredinidae). Can. Entomol. 109:53-58.

McKey, D. 1979. The distribution of secondary compounds within plants. Pages 56-133 in G. A. Rosenthal and D. H. Janzen, eds. Herbivores: their interaction with secondary plant metabolites. Academic Press, New York.

McLaughlin, S. B., and D. S. Shriner. 1980. Allocation of resources to defense and repair. Pages 407- 431 in J. G. Horsfall and E. B. Cowling, eds. Plant disease. Vol. 5. Academic Press, New York.

McNaughton, S. J. 1976. Serengeti migratory wildebeest: facilitation of energy flow by grazing. Science 191:92-94. 1979. Grazing as an optimization process: grass-ungulate relationships in the Serengeti. Am. Nat. 113:691-703.

McNaughton, S. J., and J. L. Tarrants. 1983. Grass leaf silicification: natural selection for an inducible defense against herbivores. Proc. Natl. Acad. Sci. USA 80:790-791.

Madden, J. L. 1977. Physiological reactions of Pinus radiata to attack by woodwasp, Sirex noctilio F. (Hymenoptera:Siricidae). Bull. Entomol. Res. 67:405-426.

Mares, M. A., F. A. Enders, J. M. Kingsolver, J. L. Neff, and B. B. Simpson. 1977. Prosopis as a niche component. Pages 123-149 in B. B. Simpson, ed. Mesquite. Dowden, Hutchinson & Ross, Stroudsburg, Pa.

Markkula, M., S. Laurema, and K. Tiittanen. 1976. Systemic damage caused by Trioza apicaulis on carrot. Pages 153-155 in T. Jermy, ed. The host plant in relation to insect behavior and reproduction. Plenum, New York.

Mason, R. R., and R. F. Luck. 1978. Population growth and regulation. Pages 41-47 in M. H. Brookes, R. W. Stark, and R. W. Campbell, eds. The Douglas-fir tussock moth: a synthesis. US Dep. Agric. For. Serv. Sci. Education Agency. Tech. Bull. 1585.

Matthee, J. J. 1947. Phase variation in the lawn caterpillar (Spodoptera abyssinia Guen.). J. Entomol. Soc. S. Afr. 10:16-23.

Mattson, W. J., and N. D. Addy. 1975. Phytophagous insects as regulators of forest primary production. Science 190:515-522.


Miles, P. W. 1968a. Insect secretions in plants. Ann. Rev. Phytopathol. 6:137-164. 1968b. Studies on the salivary physiology of plant-bugs. Experimental induction of galls. J. Insect Physiol. 14:97-106. 1978. Redox reactions ot hemipterous saliva in plant tissues. Entomol. Exp. Appl. 24:534-539.

Millburn, P. 1978. Biotransformation of xenobiotics by animals. Pages 35-73 in J. B. Harborne, ed. Biochemical aspects of plant and animal coevolution. Academic Press, New York.

Mitter, C., and D. J. Futuyma. 1983. An evolutionary-genetic view of host-plant utilization by insects. Pages 427-459 in R. F. Denno and M. S. McClure, eds. Variable plants and herbivores in natural and managed systems. Academic Press, New York.

Mitter, C., D. J. Futuyma, J. C. Schneider, and J. D. Hare. 1979. Genetic variation and host plant relations in a parthenogenetic moth. Evolution 33:777-790.

Montgomery, G. G., ed. 1978. The ecology of arboreal folivores. Smithsonian Institution, Washing- ton, D.C.

Mooney, H. A., and S. L. Gulman. 1982. Constraints on leaf structure and function in reference to herbivores. BioScience 32:198-206.

Morris, R. F., ed. 1963. The dynamics of epidemic spruce budworm populations. Mem. Entomol. Soc. Can. 31:1-332.

Morris, R. F., C. A. Miller, D. 0. Greenbank, and D. G. Mott. 1958. The population dynamics of the spruce budworm in eastern Canada. Proc. 10th Int. Congr. Entomol. 4:137-149.

Muller, K. 0. 1956. Einige einfache versuche zum nachweis von phytoalexinen. Phytopathol. Z. 27:237-254.

Muller, K. O., and H. Borger. 1941. Experimentelle untersuchungen uber die phytophora-resistenz der kartoffel. Arb. Biol. Reichsanst. Land-Forstwirtsch., Berlin-Dahlem 23:189-231.

Nelson, C. E., M. Walker-Simmons, D. Makus, G. Zuroske, J. Graham, and C. A. Ryan. 1983. Regulation of synthesis and accumulation of proteinase inhibitors in leaves of wounded tomato plants. Am. Chem. Soc. Symp. Ser. 208:103-122.

Nichols, J. 0. 1968. Oak mortality in Pennsylvania. A ten-year study. J. For. 66:681-694. Niemela, P., A. M. Aro, and E. Haukioja. 1979. Birch leaves as a resource for herbivores. Damage-

induced increase in leaf phenols with trypsin-inhibiting effects. Rep. Kevo Subarctic Res. Stn. 15:37-40.

Nolte, D. J. 1963. A pheromone for melanization of locusts. Nature 200:660-661. 1974. The gregarization of locusts. Biol. Rev. 49:1-14. 1977. The action of locustol. J. Insect Physiol. 23:899-903.

Norris, D. M. 1979. How insects induce disease. Pages 239-255 in J. G. Horsfall and E. B. Cowling, eds. Plant disease. Vol. 4. Academic Press, New York.

Odiyo, P. 0. 1981. Development of the first outbreaks of the African armyworm, Spodoptera exempta (Walk.), between Kenya and Tanzania during the "off-season" months of July to December. Insect Sci. Appl. 1:305-318.

Osborne, D. J. 1973. Mutual regulation of growth and development in plants and insects. Pages 33-42 in H. F. van Emden, ed. Insect/plant relationships. Blackwell, London.

Owen, D. F., and R. C. Wiegert. 1976. Do consumers maximize plant fitness? Oikos 27:488-492. . 1982. Grasses and grazers: is there a mutualism? Oikos 38:258-260.

Perry, D. A., and G. B. Pitman. Genetic and environmental influences in host resistance to'herbivory: Douglas fir and the western spruce budworm. J. Appl. Entomol. (in press).

Peters, T. M., and P. Barbosa. 1977. Influence of population density on size, fecundity and development rate of insects in culture. Annu. Rev. Entomol. 22:431-450.

Pianka, E. R. 1970. On r- and K-selection. Am. Nat. 104:592-597. Potts, D. C. 1981. Crown-of-thorns starfish: man-induced pest or natural phenomena. Pages 55-86 in

R. L. Kitching and R. E. Jones, eds. The ecology of pests (some Australian case histories). CSIRO, Melbourne.

Prentice, R. M. 1962. Forest Lepidoptera of Canada. Vol. 2. Bull. 128. Canada Dep. Forestry, Ottawa. 1963. Forest Lepidoptera of Canada. Vol. 3. Publ. 1013. Canada Dep. Forestry, Ottawa.

Price, P. W., C. E. Bouton, P. Gross, B. A. McPheron, J. N. Thompson, and A. E. Weis. 1980. Interactions among three trophic levels: influence of plants on interactions between insect herbivores and natural enemies. Ann. Rev. Ecol. Syst. 11:41-65.


Prokopy, R. J. 1981. Epideictic pheromones that influence spacing patterns of phytophagous insects. Pages 181-213 in D. A. Nordlund, R. L. Jones, and W. J. Lewis, eds. Semiochemicals: their role in pest control. Wiley, New York.

Puritch, G. S., and W. W. Nijholt. 1974. Occurrence of juvabione-related compounds in grand fir and Pacific silver fir infested by balsam wooly aphid. Can. J. Bot. 52:585-587.

Raffa, K. F., and A. A. Berryman. 1982. Accumulation of monoterpenes and associated volatiles following inoculation of grand fir with a fungus transmitted by the fir engraver, Scolytus ventralis (Coleoptera: Scolytidae). Can. Entomol. 114:797-810.

. 1983. The role of host plant resistance in the colonization behavior and ecology of bark beetles (Coleoptera: Scolytidae). Ecol. Monogr. 53:27-49.

Rasmussen, R. A. 1970. Isoprene: identified as a forest-type emission to the atmosphere. Environ. Sci. Technol. 8:667-671.

. 1972. What do hydrocarbons from trees contribute to air pollution? J. Air Pollut. Contr. Assoc. 22:537-543.

Reardon, P. O., C. L. Leinweber, and L. B. Merrill. 1972. The effect of bovine saliva on grasses. J. Anim. Sci. 34:897-898. 1974. Responses of sideoates grama to animal saliva and thiamine. J. Range Manage. 27: 400-401.

Rhoades, D. F. 1977. The antiherbivore chemistry of Larrea. Pages 135-175 in T. J. Mabry, J. H. Hunziker, and D. R. Difeo, Jr., eds. Creosote bush. Dowden, Hutchinson & Ross, Strouds- burg, Pa. 1979. Evolution of plant chemical defense against herbivores. Pages 3-54 in G. A. Rosenthal and D. H. Janzen, eds. Herbivores: their interaction with secondary plant metabolites. Academic Press, New York. 1983a. Herbivore population dynamics and plant chemistry. Pages 155-220 in R. F. Denno and M. S. McClure, eds. Variable plants and herbivores in natural and managed systems. Academic Press, New York. 1983b. Responses of alder and willow to attack by tent caterpillars and webworms: evidence for pheromonal sensitivity of willows. Am. Chem. Soc. Symp. Ser. 208:55-68.

Rhoades, D. F., and R. G. Cates. 1976. Toward a general theory of plant antiherbivore chemistry. Rec. Adv. Phytochem. 10:168-213.

Rice, E. L. 1974. Allelopathy. Academic Press, New York. Roelofs, W. L. 1981. Attractive and aggregating pheromones. Pages 215-235 in P. A. Nordlund, R. L.

Jones, and W. J. Lewis, eds. Semiochemicals: their role in pest control. Wiley, New York. Root, R. B., and P. M. Kareiva. 1983. The search for resources by cabbage butterflies (Pieris rapae):

ecological consequences and adaptive significance of Markovian movements in a patchy environment. Ecology 65:147-165.

Rose, D. J. W. 1975. Field development and quality changes in successive generations of Spodoptera exempta (Wlk.), the African armyworm. J. Appl. Ecol. 21:727-739.

Rosenthal, G. A., D. L. Dahlman, and D. H. Janzen. 1976. A novel means of dealing with L-

canavanine, a toxic metabolite. Science 192:256-258. Rosenthal, G. A., and D. H. Janzen, eds. 1979. Herbivores: their interaction with plant secondary

metabolites. Academic Press, New York. Rosenthal, G. A., D. H. Janzen, and D. L. Dahlman. 1977. Degradation and detoxification of

canavanine by a specialized seed predator. Science 196:658-660. Rothschild, M. 1973. Secondary plant substances and warning colouration in insects. Pages 59-83 in

H. F. van Emden, ed. Insect/plant relationships. Blackwell, London. Salt, G. W., ed. 1984. A round table on research in ecology and evolutionary biology. Am. Nat.

122:583-705. Schultz, D. E., and D. C. Allen. 1977. Characteristics of sites with high black cherry mortality due to

bark beetles following defoliation by Hydria prunivorata. Environ. Entomol. 6:77-81. Schultz, J. C. 1983a. Habitat selection and foraging tactics of caterpillars in heterogeneous trees.

Pages 61-90 in R. F. Denno and M. S. McClure, eds. Variable plants and herbivores in natural and managed systems. Academic Press, New York. 1983b. Impact of variable plant defensive chemistry on susceptibility of insects to natural enemies. Am. Chem. Soc. Symp. Ser. 208:37-54.


Schultz, J. C., and I. T. Baldwin. 1982. Oak leaf quality declines in response to defoliation by gypsy moth larvae. Science 217:149-151.

Scriber, J. M. 1983. Evolution of feeding specialization, physiological efficiency, and host races in selected Papilionidae and Saturniidae. Pages 373-412 in R. F. Denno and M. S. McClure, eds. Variable plants and herbivores in natural and managed systems. Academic Press, New York.

Scriber, J. M., and P. P. Feeny. 1979. The growth of herbivorous caterpillars in relation to degree of feeding specialization and to growth form of their foodplants (Lepidoptera:Papilionidae and Bombycoidea). Ecology 60:829-850.

Scriber, J. M., and F. Slansky, Jr. 1981. The nutritional ecology of immature insects. Annu. Rev. Entomol. 26: 183-211.

Seifert, R. P., and F. H. Seifert. 1976. A community matrix analysis of Heliconia insect communities. Am. Nat. 110:461-483.

Sequeira, L. 1980. Defenses triggered by the invader: recognition and compatibility phenomena. Pages 179-200 in J. G. Horsfall and E. B. Cowling, eds. Plant disease. Vol. 5. Academic Press, New York.

Shiga, M. 1976. Effect of group size on the survival and development of young larvae of Malacosoma neustria testacea Motschulsky (Lepidoptera, Lasiocampidae) and its role in the natural population. Kontyu (Tokyo) 44(4):537-553.

Smith, C. C. 1976. When and how much to reproduce: the trade-off between power and efficiency. Am. Zool. 16:763-774.

Smith, N. G. Periodic migrations and population fluctuations by the neotropical day-flying moth Urania fulgens through the isthmus of Panama. Pages 135-147 in E. G. Leigh, Jr., A. S. Rand, and D. M. Windsor, eds. The ecology of a tropical forest. Smithsonian Institution, Washington, D.C. (in press).

Smith, W. H. 1981. Air pollution and forests. Springer-Verlag, New York. Southwood, T. R. E. 1976. Bionomic strategies and population parameters. Pages 26-48 in R. M. May,

ed. Theoretical ecology. Blackwell, London. 1978. Ecological methods. Chapman & Hall, London.

Spurr, S. H., E. W. Littlefield, and H. A. Bess. 1947. Ecology of susceptible and resistant forest types and stands. Harv. For. Bull. 22:27-39.

Stage, A. R. 1978. Modeling probability of outbreak occurrence and stand involvement. Pages 59-61 in M. H. Brookes, R. W. Stark, and R. W. Campbell, eds. The Douglas-fir tussock moth: a synthesis. US Dep. Agric. For. Serv. Sci. Education Agency Tech. Bull. 1585.

Staley, J. M. 1965. Decline and mortality of red and scarlet oaks. For. Sci. 11:2-17. Stamp, N. E. 1980. Egg deposition patterns in butterflies: why do some species cluster their eggs

rather than deposit them singly? Am. Nat. 115:367-380. Stehr, F. W. 1974. Release, establishment and evolution of parasites and predators. Pages 124-136 in

F. G. Maxwell and F. A. Harris, eds. Proc. of the summer institute of biological control of plant insects and diseases. University of Mississippi Press, Jackson.

Stenseth, N. C. 1983. Grasses, grazers, mutualism and coevolution: a comment about handwaving in ecology. Oikos 41:152-153.

Strong, D. R., Jr. 1983. Natural variability and the manifold mechanisms of ecological communities. Am. Nat. 122:636-645.

Suzuki, H. 1980. Defenses triggered by previous invaders: fungi. Pages 319-332 in J. G. Horsfall and E. B. Cowling, eds. Plant disease. Vol. 5. Academic Press, New York.

Swain, T. 1979. Tannins and lignins. Pages 657-682 in G. A. Rosenthal and D. H. Janzen, eds. Herbivores: their interaction with plant secondary metabolites. Academic Press, New York.

Symmons, P. M. 1978. The prevention of plagues of the red locust. Acrida 7:55-78. Taylor, R. A. J. 1979. Simulation model of locust migratory behaviour. J. Anim. Ecol. 48:577-602. Teale, E. W. 1949. The insect world of J. Henri Fabre. Dodd, Mead, New York. Thompson, D. Q. 1955. The 1953 lemming emigration at Point Barrow, Alaska. Arctic 8:37-45. Tsubaki, Y. 1981. Some beneficial effects of aggregation in young larvae of Pryeria sinica Moore

(Lepidoptera: Zygaenidae). Res. Popul. Ecol. 23:156-167.


Tsyplenkov, E. P. 1978. Harmful Acridoidea of the USSR. U.S.D.C. Natl. Tech. Inf. Serv. Springfield, Va.

Uvarov, B. P. 1921. A revision of the genus Locusta L. (=Pachytylus Fieb.), with a new theory as to the periodicity and migrations of locusts. Bull. Entomol. Res. 12:135-163. 1928. Locusts and grasshoppers. Imperial Bureau of Entomology, London.

. 1966. Grasshoppers and locusts. Vol. 1. Cambridge University Press, Cambridge.

. 1977. Grasshoppers and locusts. Vol. 2. Centre for Overseas Pest Research, London. Vaughan, G. L., and A. M. Jungreis. 1977. Insensitivity of lepidopteran tissues to ouabain: physiologi-

cal mechanisms for protection from cardiac glycosides. J. Insect Physiol. 23:585-589. Vite, J. P. 1961. The influence of water supply on oleoresin exudation pressure and resistance to bark

beetle attack in Pinus ponderosa. Contrib. Boyce Thompson Inst. Plant Res. Prof. Pap. 21:37-66.

Wagner, M. R., and D. M. Benjamin. 1981. Allelochemics and nutritional indices for larch sawfly, Pristiphora erichsonii (Hartig): a specialist feeding on Larix spp. J. Chem. Ecol. 7:165-174.

Waldbauer, G. P. 1968. The consumption and utilization of food by insects. Pages 229-288 in J. W. L. Beament, J. E. Treherne, and V. B. Wigglesworth, eds. Advances in insect physiology. Vol. 5. Academic Press, New York.

Wallace, J. W., and R. L. Mansell, eds. 1976. Biochemical interaction between plants and insects. Plenum, New York.

Waloff, Z. 1966. The upsurges and recessions of the desert locust plague: an historical survey. Anti- Locust Mem. 8: 1- 111. 1972. The plague dynamics of the desert locust Schistocerca gregaria (Forsk.). Pages 363-369 in Proceedings of the international study conference on the current and future problems of Acridology, London, 1970. Centre for Overseas Pest Research, London.

Waring, R. H., and G. B. Pitman. 1983. Physiological stress in lodgepole pine as a precursor for mountain pine beetle attack. Z. Ang. Entomol. 96:265-270.

Waters, W. E., and R. W. Stark. 1980. Forest pest management: concept and reality. Annu. Rev. Entomol. 25:479-509.

Watt, K. E. F. 1964. Comments on fluctuations of animal populations and measures of community stability. Can. Entomol. 96:1434-1442. 1965. Community stability and the strategy of biological control. Can. Entomol. 97:887-895.

Way, M. J. 1973. Population structure in aphid colonies. Pages 76-84 in A. D. Lowe, ed. Perspectives in aphid biology. Entomology Society of New Zealand, Aukland.

Way, M. J., and M. Cammell. 1970. Aggregation behaviour in relation to food utilization by aphids. Pages 229-247 in A. Watson, ed. Animal populations in relation to their food resources. Blackwell, Oxford.

Weldon, P. J. 1983. The evolution of alarm pheromones. Pages 309-312 in D. Muller-Schwarze and R. M. Silverstein, eds. Chemical signals in vertebrates. Vol. 3. Plenum, New York.

Wellington, W. G. 1957. Individual differences as a factor in population dynamics: the development of a problem. Can. J. Zool. 35:293-323. 1980. Dispersal and population change. Pages 11-24 in A. A. Berryman and L. Safranyik, eds. Dispersal of forest insects: evaluation, theory and management implications. Washington State University Coop. Ext. Serv., Pullman.

Went, D. F. 1982. Egg activation and parthenogenetic reproduction in insects. Biol. Rev. 57:319-344. Weyh, R., and V. Maschwitz. 1978. Trail substance in larvae of Eriogaster lanestris L. Naturwis-

senschaften 65:64. White, M. J. D. 1973. Animal cytology and evolution. Cambridge University Press, Cambridge. White, T. C. R. 1969. An index to measure weather-induced stress of trees associated with outbreaks

of psyllids in Australia. Ecology 50:905-909. 1974. A hypothesis to explain outbreaks of looper caterpillars with special reference to populations of Selidosema suavis in a plantation of Pinus radiata in New Zealand. Oecologia 16:279-301. 1976. Weather, food and plagues of locusts. Oecologia 22:119-134.

Whitham, T. G. 1983. Host manipulation of parasites: within-plant variation as a defense against


rapidly evolving parasites. Pages 15-41 in R. F. Denno and M. S. McClure, eds. Variable plants and herbivores in natural and managed systems. Academic Press, New York.

Whitney, H. S. 1982. Relationships between bark beetles and symbiotic organisms. Pages 183-211 in J. B. Mitton and K. B. Sturgeon, eds. Bark beetles in North American conifers. University of Texas Press, Austin.

Wood, D. L. 1982. The role of pheromones, kairomones and allomones in the host selection and colonization behavior of bark beetles. Annu. Rev. Entomol. 27:411-446.

Wood, D. L., and W. D. Bedard. 1977. The role of pheromones in the population dynamics of the western pine beetle. Proc. Int. Congr. Entomol. 15:643-652.

Wright, L. C., A. A. Berryman, and S. Gurusiddaiah. 1979. Host resistance to the fir engraver beetle. 4. Can. Entomol. 111:1255-1262.

Zucker, W. F. 1983. Tannins: does structure determine function? An ecological perspective. Am. Nat. 121:335-365.

offensive-defensive interactions between herbivores and ... · plants are defended against...

Documents