Search

Field stations are platforms for documenting patterns and processes in ecosystems and are critical for understanding how anthropogenic climate change reshapes nature. Although networks of field stations have been used to identify patterns at continental to global scales, these broad, sparsely distributed networks miss variation in climate change at local and regional scales. We propose that regional-scale research networks are essential for addressing the myriad of ecological and evolutionary challenges—including management and mitigation options—that cannot be answered by more broadly distributed networks or by individual field sites. We discuss our experiences leveraging natural areas throughout California at the Institute for the Study of Ecological and Evolutionary Climate Impacts. We then explore benefits and challenges of networking research at spatial scales congruent with regional patterns of climate variation and climate change, the challenges of sustained infrastructure and research support, and opportunities for future regional-scale research networks.

Historical data sets can be useful tools to aid in understanding the impacts of global change on natural ecosystems. Resampling of historically sampled sites (“snapshot resampling”) has often been used to detect long‐term shifts in ecological populations and communities, because it allows researchers to avoid long‐term monitoring costs and investigate a large number of potential trends. But recent simulation‐based research has called the reliability of resampling into question, and its utility has not been comprehensively evaluated. Here we combine long‐term empirical data sets with novel community‐level simulations to explore the accuracy of snapshot resampling of both population‐ and community‐level metrics under a variety of conditions. We show that snapshot resampling often yields spurious conclusions, but the accuracy of results increases when inter‐annual variability in the response variable is low or the magnitude of change through time is high. Snapshot resampling also generally performs better for community‐level metrics (e.g., species richness) as opposed to population‐level metrics pertaining to a single species (e.g., abundance). Finally, we evaluate strategies to improve the accuracy of snapshot resampling, including sampling multiple years at the end of the study, but these produce mixed results. Ultimately, we find that snapshot resampling should be used with caution, but under certain circumstances, can be a useful for understanding long‐term global change impacts.

Although ecological specialists exploit a relatively limited resource base, it is unclear whether specialization limits local population density. Here, we focus on the relationship between diet specialization and local population density of a phytophagous insect, the diamondback moth, Plutella xylostella L. (Lepidoptera: Plutellidae). We compared key life history traits and food plant use among five pairs of high- and low-density populations from widely separate geographical areas. Moths from populations that persist at high densities were more generalized in food plant use than moths from low-density populations. In particular, the oviposition preference and larval performance of moths from some high-density populations were less variable across a suite of food plants, suggesting that moths from high- density populations had a broader diet. In addition, low- density populations were less similar to each other, exhibiting opposing preferences for particular plant species. Hence diet breadth may explain some of the persistent differences in the population density of diamondback moths in the field, consistent with the idea that ecological specialization may be generally associated with population density.

We report results from a continuing, long-term field experiment addressing biotic responses to climatic change in grasslands. We focus on effects of summer precipitation (enhanced rainfall, drought, control) and winter ground temperatures (warming, control) on growth, reproduction and herbivory in St John's wort, Hypericum perforatum L. Both winter warming and summer rainfall regimes modified performance and interactions of H. perforatum, particularly those with herbivorous insects. Winter warming had positive effects, with earlier initiation of plant growth and reduced damage by gall-forming and sucking insects in spring, but also had strong negative effects on plant height, flowering, and reproduction. Summer drought reduced reproductive success, but even severe drought did not affect plant growth or flowering success directly. Rather, summer drought acted indirectly by modifying interactions with herbivorous insects via increased vulnerability of the plants to herbivory on flowers and capsules. Overall, the effects of summer precipitation were expressed mainly through interactions that altered the responses to increased winter temperatures, particularly as summer drought increased. The field site, in Oxfordshire, UK, is near the northern limit of distribution of the species, and the experiment tested probable responses of H. perforatum as climates shift towards those more typical of the current center of the distribution of the species. However, if climates do change according to the projected scenarios, then H. perforatum is unlikely to fare well near its northern boundary. Increased winter temperatures, particularly if accompanied by increased summer drought, will probably render this species even less abundant in England than at present.

Aim Our aim in this study was to document the global biogeographic variation in the effects of soil microbes on the growth of Centaurea solstitialis (yellow starthistle; Asteraceae), a species that has been introduced throughout the world, but has become highly invasive only in some introduced regions. Location To assess biogeographic variation in plant–soil microbe interactions, we collected seeds and soils from native Eurasian C. solstitialis populations and introduced populations in California, Argentina and Chile. Methods To test whether escape from soil-borne natural enemies may contribute to the success of C. solstitialis, we compared the performance of plants using seeds and soils collected from each of the biogeographic regions in greenhouse inoculation/sterilization experiments. Results We found that soil microbes had pervasive negative effects on plants from all regions, but these negative effects were significantly weaker in soils from non-native ranges in Chile and California than in those from the non-native range in Argentina and the native range in Eurasia. Main conclusions The biogeographic differences in negative effects of microbes in this study conformed to the enemy-release hypothesis (ERH) overall, but the strong negative effect of soil biota in Argentina, where C. solstitialis is invasive, and weaker effects in Chile where it is not, indicated that different factors influencing invasion are likely to occur in large scale biogeographic mosaics of interaction strengths.

Interactions between herbivores and seed predators may have long-term consequences for plant populations that rely on persistent seed banks for recovery after unpredictable fires. We assessed the effects of browsing by deer and seed predation by rodents, ants and birds on the densities of seeds entering the seed bank of Ceanothus cuneatus var. rigidus, a maritime chaparral shrub in coastal California. Ceanothus produced many more seeds when protected from browsers in long-term experimental exclosures than did browsed plants, but the seed densities in the soil beneath browsed and unbrowsed Ceanothus were the same at the start of an intensive one-year study. The density of seeds in the soil initially increased in both treatments following summer seed drop: while densities returned to pre-drop levels within a few weeks under browsed plants, soil seed densities remained high for 5-8 months beneath unbrowsed plants. Rodent abundance (especially deer mice) was higher near unbrowsed plants than >30 m away, and rodents removed Ceanothus seeds from dishes in the experimental plots. At least in the short term, rodent density and rates of seed removal were inversely related to the intensity of browsing. Our data have management implications for maintaining viable Ceanothus populations by regulating the intensity of browsing and the timing, intensity and frequency of fires.

The Argentine ant is a common pest of mediterranean-type ecosystems worldwide, causing widespread extirpation of many native ant species. This study examines spatial and temporal patterns of invasion at a local scale, investigates the effects of this invader on native ant communities and assesses causal mechanisms for these spatial, temporal and numerical patterns, particularly in relation to local climate. Argentine ant workers were more active in the drier months of summer and fall, and abundance was correlated with cumulative precipitation the previous winter. Argentine ant worker abundance greatly increased in conjunction with EI Nino events. No similar relationship existed for native ants. This invasive ant affected native ant abundances differently by species: some species were able to tolerate its presence temporarily. Overall, this study provides a framework of factors to consider when devising control strategies for the invasive Argentine ant.

Spatial and temporal variation in critical life history traits differ markedly between two co-occurring, rare and endangered annual plants, Gilia tenuiflora arenaria and Chorizanthe pungens pungens, that have restricted, but overlapping geographic distributions in coastal habitats of central California, USA. Environmental variation was extreme during the 6-year study (1997–2002), with the greatest differences between an intense wet, warm El Nino year (1998), followed by a very dry, cool La Nina (1999). Both species have similar increases in adult survival and seed set with wetter, warmer weather in spring, but they differed in other traits: more Gilia germinated in years with high rainfall and temperature, while more Chorizanthe germinated in dry, cooler years; Gilia abundance increased with the number of years since the previous large El Nino event, while Chorizanthe abundance declined; and Gilia abundance was independent of the previous year’s seed set, while Chorizanthe density was directly related to the previous year’s seed set. The strong negative associations of Chorizanthe abundance with warm, wet weather and with time since the previous El Nino probably reflect the particular sequence of annual weather patterns from 1997 through 2002, especially the extremes in 1998 and 1999. Since it germinates readily in most years under a wide range of winter conditions and does not develop a long-lived seed bank, Chorizanthe seedling abundance reflects recent additions and depletions of its seed bank, rather than prevailing weather, per se. In contrast, Gilia seeds may remain in the seed bank for many years, until relatively rare winter conditions trigger significant germination. These species-specific demographies enhance persistence and coexistence of these species, but the mechanisms differ from each other and from those described for annuals in other highly variable environments.

We investigate patch selection strategies of hosts and parasitoids in heterogeneous environments. Previous theoretical work showed that when host traits vary among patches, coevolved populations of hosts and parasitoids make congruent choices (i.e., hosts and parasitoids preferentially select the same patches) and ex- hibit direct density dependence in the distribution of percent par- asitism. However, host-parasitoid systems in the field show a range of patterns in percent parasitism, while behavioral studies indicate that hosts and parasitoids can exhibit contrary choices (i.e., hosts avoid patches favored by the parasitoid). We extend previous theory by permitting life-history traits of the parasitoid as well as the host to vary among patches. Our analysis implies that in coevolutionarily stable populations, hosts preferentially select patches that intrinsically support higher host equilibrium numbers (i.e., the equilibrium num- ber achieved by hosts when both populations are confined to a single patch) and that parasitoids preferentially select patches that intrin- sically support higher parasitoid equilibrium numbers (i.e., the equi- librium number achieved by the parasitoids when both populations are confined to a patch). Using this result, we show how variation in life-history traits among patches leads to contrary or congruent choices or leads to direct density dependence, inverse density de- pendence, or density independence in the distribution of percent parasitism. In addition, we determine when populations playing the coevolutionarily stable strategies are ecologically stable. Our analysis shows that heterogeneous environments containing patches where the intrinsic rate of growth of the host and the survivorship rate of the parasitoid are low result in the coevolved populations exhibiting contrary choices and, as a result, promote ecological stability.

We explored consequences of spatial and temporal heterogeneity in herbivory on the survival, growth, and reproduction of the Californian native dune thistle, Cirsium occidentale, in coastal and inland sites, for 2 years. We assessed the relative impacts of insect and mammalian herbivores and compared the relative importance of herbivory in coastal and inland habitats and among locations with different microclimates across a coastal dune. Effects of insect and mammalian herbivores were tested with a combination of insecticidal spray and cage exclusion treatments in a factorial experiment at the coastal site. Mammalian herbivores strongly affected the population dynamics of C. occidentale in both years, and their effects were augmented by fungal infection (1991), herbivory by stem-borers (1990) and, to a lesser extent, by insect seed predators in both years. Mammals caused most plant deaths, but the mammal species responsible differed among sites. Rabbit herbivory altered the vegetative growth of coastal thistles and significantly modified other key aspects of Cirsium demography, including growth rate and timing of reproduction. Small, uncaged plants grazed by rabbits took at least 1 year longer to mature than did caged plants. Larvae of Pyrausta subsequalis were the only insects that killed established plants. In 1990 and 1991, the numbers of insects damaging seed heads before dispersal were low, but were sufficient to cause receptacle and seed damage. The number of mature, undamaged seeds (and percent successful seed production) was reduced significantly only for heads infested by fungi near the ocean in 1991: the fungus occurred in 37% of heads and caused a 77% reduction in mature seeds.

Defoliation by herbivores can reduce carbon assimilation, change plant water relations, and even shift the biotic structure of plant communities. In this study, we took advantage of a long-term deer exclosure experiment to examine the consequences of persistent deer herbivory on plant water relations and the xylem structure-function relationships in Ceanothus rigidus, a maritime chaparral shrub in coastal California. Browsed plants had thicker stems with many intertwined short distal twigs, and significantly higher sapwood-to-leaf area ratios than their non-browsed counterparts. Leaf area-specific hydraulic conductivity was similar in both browsed and non-browsed plants, but xylem area-specific conductivity was significantly lower in the browsed plants. Vessel diameters were equivalent in both plant groups, but the number of vessels on a transverse area basis was nearly 40 % lower in the browsed plants, accounting for their lower transport efficiency. Mid-day in situ water potentials and losses of hydraulic conductivity due to embolism were similar in both groups of plants but stomatal conductance was higher in the browsed shrubs in the early part of the growing season. We discuss our findings in the context of whole-plant ecophysiology, and explore the consequences of herbivory on hormonal signals, wood anatomy, and xylem function. © 2014 Springer-Verlag Berlin Heidelberg.

Background: Biological invasions are fundamentally biogeographic processes that occur over large spatial scales. Interactions with soil microbes can have strong impacts on plant invasions, but how these interactions vary among areas where introduced species are highly invasive vs. naturalized is still unknown. In this study, we examined biogeographic variation in plant-soil microbe interactions of a globally invasive weed, Centaurea solstitialis (yellow starthistle). We addressed the following questions (1) Is Centaurea released from natural enemy pressure from soil microbes in introduced regions? and (2) Is variation in plant-soil feedbacks associated with variation in Centaurea's invasive success? Methodology/Principal Findings: We conducted greenhouse experiments using soils and seeds collected from native Eurasian populations and introduced populations spanning North and South America where Centaurea is highly invasive and noninvasive. Soil microbes had pervasive negative effects in all regions, although the magnitude of their effect varied among regions. These patterns were not unequivocally congruent with the enemy release hypothesis. Surprisingly, we also found that Centaurea generated strong negative feedbacks in regions where it is the most invasive, while it generated neutral plant-soil feedbacks where it is noninvasive. Conclusions/Significance: Recent studies have found reduced below-ground enemy attack and more positive plant-soil feedbacks in range-expanding plant populations, but we found increased negative effects of soil microbes in range-expanding Centaurea populations. While such negative feedbacks may limit the long-term persistence of invasive plants, such feedbacks may also contribute to the success of invasions, either by having disproportionately negative impacts on competing species, or by yielding relatively better growth in uncolonized areas that would encourage lateral spread. Enemy release from soil-borne pathogens is not sufficient to explain the success of this weed in such different regions. The biogeographic variation in soil-microbe effects indicates that different mechanisms may operate on this species in different regions, thus establishing geographic mosaics of species interactions that contribute to variation in invasion success. Fil: Andonian, Krikor. University of California; Estados Unidos Fil: Hierro, Jose Luis. Consejo Nacional de Investigaciones Científicas y Técnicas. Instituto de Ciencias de la Tierra y Ambientales de La Pampa. Universidad Nacional de La Pampa. Facultad de Ciencias Exactas y Naturales. Instituto de Ciencias de la Tierra y Ambientales de La Pampa; Argentina Fil: Khetsuriani, Liana. Georgian Academy of Sciences. Institute of Botany; Georgia Fil: Becerra, Pablo. Universidad Católica de Chile; Chile Fil: Janoyan, Grigor. American University of Armenia; Armenia Fil: Villarreal, Diego. Universidad Nacional de La Pampa. Facultad de Ciencias Exactas y Naturales; Argentina Fil: Cavieres, Lohengrin. Universidad de Concepción; Chile Fil: Fox, Laurel R.. University of California; Estados Unidos Fil: Callaway, Ragan M.. University of Montana; Estados Unidos

Parasitoid behavior may be influenced directly by the plants on which their hosts feed (Vet & Dicke, 1992), but it is usually assumed that aspects of parasitoid demography (e.g., growth, survival and sex ratio) only respond indirectly to plants, via changes in properties of their hosts (Godfray, 1994). A large body of literature shows that sex ratios of parasitoids respond directly to such host properties as body size or population density, and to environmental factors such as temperature (King, 1993; Godfray, 1994). In this paper, we demonstrate that demographic responses, particularly sex ratios of parasitoids, are properties that may be affected directly by their hosts’ foodplants. We assessed the direct effects of plant properties on sex allocation in two different herbivore-parasitoid systems, and explored indirect effects of plants on wasp fecundity and survival.

The pressure to escape natural enemies may shape how herbivorous insects use their plant resources. On wild crucifers, ovipositional preferences of the diamondback moth (Plutella xylostella; DBM) were similar to searching preferences of its main parasitoid, an ichneumonid wasp (Diadegma insulare). But on cultivated crucifers, these species had opposite preferences. In addition, DBM ovipositional preferences did not correlate with growth or reproduction on several foodplants. We interpret these patterns as evidence of evolution for use of enemy-free space in agricultural systems.

We manipulated soil fertility and insect attack for two species of Eucalyptus in natural stands of subalpine woodland on shallow, infertile granitic soils. E. pauciflora and E. stellulata responded in similar ways to simultaneous insecticide and fertilizer treatments. Eliminating herbivorous insects produced the largest changes - improved plant growth, increased leaf N and P, and reduced leaf specific density. Fertilizer regime modified some leaf properties, but had little effect on tree growth. E. stellulata trees were initially shorter than E. pauciflora, but grew faster without herbivores; by the end of the experiment both species were the same size when herbivores were removed. Foliage N and P levels increased most in trees with the most balanced fertilizer addition (NPK), and increased in all trees protected from insects, regardless of fertilizer regime. In this system, herbivorous insects exacerbated the effects of nutrientpoor soils, and may affect dominance of Eucalyptus species in mature forests.

Species with known demographies may be used as proxies, or approximate models, to predict vital rates and ecological properties of target species that either have not been studied or are species for which data may be difficult to obtain. These extrapolations assume that model and target species with similar properties respond in the same ways to the same ecological factors, that they have similar population dynamics, and that the similarity of vital rates reflects analogous responses to the same factors. I used two rare, sympatric annual plants (sand gilia [Gilia tenuiflora arenaria] and Monterey spineflower [Chorizanthe pungens pungens]) to test these assumptions experimentally. The vital rates of these species are similar and strongly correlated with rainfall, and I added water and/or prevented herbivore access to experimental plots. Their survival and reproduction were driven by different, largely stochastic factors and processes: sand gilia by herbivory and Monterey spineflower by rainfall. Because the causal agents and processes generating similar demographic patterns were species specific, these results demonstrate, both theoretically and empirically, that it is critical to identify the ecological processes generating observed effects and that experimental manipulations are usually needed to determine causal mechanisms. Without such evidence to identify mechanisms, extrapolations among species may lead to counterproductive management and conservation practices. Resumen: Las especies con demografias conocidas pueden ser utilizadas como modelos aproximados para predecir tasas vitales y propiedades ecologicas de especies que no han sido estudiadas o cuyos datos no pueden ser obtenidos facilmente. Estas extrapolaciones asumen que el modelo y la especie blanco con propiedades similares responden de la misma forma a los mismos factores ecologicos, que pueden tender dinamicas poblacionales similares y que la similitud de las tasas vitales refleja respuestas analogas a los mismos factores. Utilice dos plantas anuales simpatricas y raras Gilia tenuiflora arenaria y Chorizanthe pungens pungens para probar estos supuestos experimentalmente. Las tasas vitales de estas especies son similares y estan correlacionadas estrechamente con la precipitacion, y agregue agua y/o evite el acceso de herbivoros a las parcelas experimentales. Su supervivencia y reproduccion fueron influidas por diferentes factores y procesos, en gran parte estocasticos: G. t. arenaria por la herbivoria y C. p. pungens por la precipitacion. Debido a que los agentes y procesos causales que generaron patrones demograficos similares fueron particulares para cada especie, estos resultados demuestran, tanto teorica como empiricamente, que es critico identificar los procesos ecologicos que generan efectos observados y que se requieren manipulaciones experimentales para determinar mecanismos causales. Sin tal evidencia para la identificacion de mecanismos, las extrapolaciones a otras especies pueden conducir a practicas de manejo y conservacion contraproducentes.

germinating were similar in the two habitats. Burning reduced the number of native species and seeds that germinated, and to a lesser extent the number of exotic species and seeds that germinated from the seed bank. In the field survey, intact chaparral areas had greater numbers of native species and greater native cover than roads, whereas road areas had greater numbers and cover of exotic species. Several species were found only in field surveys; in particular, few chaparral shrub seeds germinated from the seed bank. Similarity between the seed bank and vegetation was low in both chaparral and road sites. The results suggest that although many native species of chaparral plants are present in the seed bank, planting of some species may help accelerate recovery in disturbed areas.

Total phenols and condensed tannins in leaves of seven species of Eucalyptus ranged from 4–40% and 0–27% respectively of the leaf dry weight. The concentrations of these compounds were variable but usually high in young and older leaves throughout the growing season, and typically increased during winter, but no other trends with season or leafage were apparent. This pattern of seasonal variation in concentrations of total phenols and condensed tannins is different to that studied for other plant species. This difference may be related to repeated production of new leaves by Eucalyptus during the growing season, and the probability that these leaves will be attacked by herbivorous insects.

Recent theory about interactions between plants and their herbivores focuses on properties of individual plants that affect their resistance to herbivores, and it extrapolates from these individual properties to those of whole communities. In this paper I question three major assumptions of this approach—costs of defenses, basic differences between classes of defenses, and step-wise coevolution—and recast the theory from a different community perspective. I propose that major differences in defenses between plants with very different life-history characteristics arise from differences in community structure, especially the numbers of herbivore species involved in the interactions. In particular, because of the diffuse pattern of herbivory on persistent plants, a new model of coevolution—diffuse coevolution—becomes appropriate when many species are involved. Complex interacting assemblages have special properties that cannot be derived just by summing up all of the simple interactions that occur.

To test the prediction that P. rapae egg densities increase with N fertilizer in large-scale systems as they do in model systems with potted plants, we used field experiments with Brassica oleracea var. acephala L. (collards and kale) planted in pots or large field plots, and treated with different levels of nitrogen fertilizer. In small-scale field experiments with potted kale and collards, egg densities were significantly higher on plants with high N than those with low N. But in larger scale experiments with field-grown collards, average seasonal P. rapae egg densities were not significantly correlated with leaf N content. These differences among experiments did not depend on the magnitude of the difference in foliage N levels.

Factors influencing cannibalism rates in the freshwater insect predator Notonecta hoffmanni were analyzed in a series of field and laboratory experiments. Starvation was re- jected as a cause of nymphal mortality. Cannibalism was related to both the relative and absolute abundances of alternative food, the cannibal's age, and the presence of refuges for the vulnerable nymphs, although the predator's state of hunger also influenced cannibalism rates. Cannibalism occurred when there was physical proximity between different age-classes even if altemative food were abundant. Long-term feeding history had no effect on canni- balism, which is a rapid response to immediate food availability. The population consequences of cannibalism seem analogous to those of dominance and spacing behavior, causing a decline in population size before animals are actually killed by the limiting resource, and increasing the chances that at least some relatively well-fed individuals will survive and reproduce success- fully.

Levels of insect attack and yields of leaf essential oils in Eucalyptus vary widely within and among species. We tested the hypotheses that 1) metabolic cost of oil detoxification increases with increasing oil yield, resulting in lower herbivore growth rates and, consequently, 2) in lower herbivore damage to plants. Distribution of insect damage, eggs, immature insects and adults and feeding rates, growth and survivorship of insects do not support the hypotheses, although a threshold level of oil may be necessary to influence herbivorous insects. Herbivorous beetles tested do not detoxify essential oils. Levels of leaf nitrogen, rather than oil content, explained differences in insect feeding and growth.

The freshwater hemipteran predator, Notonecta hoffmanni, has one generation per year. Eggs are first laid in late winter and begin hatching in April; most nymphs hatch in mid-June. A short period of heavy mortality of instar I nymphs in early summer is followed by a more gradual decline in the numbers of older individuals until autumn. The mortality of young nymphs shifts the age-distribution towards older individuals early in summer and increases the chances for the first nymphs hatched in spring to survive and breed successfully. Three alternative hypotheses about the causes of nymphal mortality are proposed and I argue that the primary cause is a reduction in food availability resulting in increased cannibalism. Reduction in spatial refuges also influences cannibalism rates. Field and laboratory experiments show that delayed reproduction by young adults is related to food availability. I suggest that a second generation of N. hoffmanni is possible if sufficient food were available when young adults mature in August.

Many herbivorous insects have generalized diets over the species' entire geographical ranges but they function as specialists with restricted diets in local communities. Local feeding specialization can be produced by biochemical, behavioral, ecological, and evolutionary processes. Much evidence is incompatible with the widely held assumptions that diet breadth is a species characteristic and that specialization among herbivorous insects implies greater efficiency and less niche overlap.

The experiments on Eucalyptus trees reported here demonstrate that allelopathic effects can be mediated by insects grazing on foliage. We show that the allelochemical nature of insect frass suppresses germination, growth and survival of herb layer species, that plant species vary in their tolerance of this material, and that as a result, the structure and composition of associated, herbaceous understory plant communities are markedly affected by frass fall.

Our previously published experiments on allelopathic effects of insect frass in Eucalyptus communities (Silander et al. 1983) have been criticized on the grounds that our estimates of annual frass production were exscessive (Ohmart 1985). However, we spanned the entire array of estimates of frass fall available from eucalypt communities, and we demonstrated allelopathic effects at even the lowest levels suggested by Ohmart. We suggest that average values of frass fall per hectare are irrelevant because they ignore both large scale variation among sites in time and space, and small scale variation in patchiness of frass accumulation. At the ecological scales relevant to germinating seeds, frass concentrations in local pockets may be considerably higher than the averages calculated for the entire ground surface.

The working model for interpreting revolutionary interactions has itself evolved over the past 10 yr. The original "stepwise" model included two species, each evolving primarily in response to selective pressures exerted by the other; this model implied a perpetual "arms race" in which each species developed more specialized adaptations to the properties of the other. With increased understanding of ecological and evolutionary patterns, an alternative model of "diffuse coevolution" has been invoked. Diffuse coevolution implies that many species, on the same or different trophic levels, may simultaneously exert selective pressures on one another and be affected by changes in other component members. However, no appropriate metaphor has emerged that succinctly describes diffuse coevolution, and conveys the notion of evolutionary changes rippling out in many directions following changes in a component species. Despite its broad acceptance, the concept of diffuse coevolution has been more successful in drawing attention to the complexity of biotic components in evolutionary interactions than in producing clear predictions of outcomes. The strongest cases for stepwise revolutionary interactions, described in the book Coevolution (Futuyma and Slatkin 1983), are those involving a close physical intimacy between organisms, such as endoparasitic helminths and their vertebrate hosts (Holmes 1983). Many of Price's (1980) examples for specialization and numbers of species of herbivorous insects could be construed similarly. Price, in fact, argues that the most speciose families of herbivorous insects, such as cecidomyiids (gall formers), weevils and agromyzids (both with many endoparasites), and aphids, have the most intimate associations with their hosts and are the most specialized species. Therefore, host plant specialists are clustered in particular herbivore groups. Even though specialization increases the potential for stepwise coevolution, many of the host plants involved are used by diverse arrays of consumers, all of which may exert selective pressures on the plants. Coevolution must be evaluated in the context of the larger community. Effects of any change in plant properties may ripple through the entire community. All herbivores feeding on a plant must cope with all the defrnsive properties of that plant, even if only one herbivore species was the dominant selective agent driving a given change. Competitive relations among the plants may shift if the new defense alters resource allocation, herbivory rates, or the plant's ability to cope with herbivory. And the herbivores' interactions with their own natural enemies can be affected by changes in their own population dynamics or in the plant properties to which those enemies may themselves respond. From this viewpoint, a plant property is truly defensive even if it reduces potential herbivory against species not originally responsible for its selection. While stepwise coevolution requires that genetic lineages of interacting species continue to coexist (Futuyma and Slatkin 1983), in diffuse coevolution it is not necessary that lineages of all the main protagonists remain part of the system. The concept of diffuse coevolution also encompasses selective pressures exerted by other trophic levels, including the herbivores' own natural enemies. Many herbivorous insects have evolved morphological, physiological, or behavioral means of reducing the success of their natural enemies. Several papers argue that parasitoids influence the evolution of feeding specializations among herbivores (e.g., Lawton 1986), and Bernays and Graham (1988) rightly point out that predators might also affect the array of plants used by herbivores. Bernays and Graham (1988) argue, however, that generalized predators may lead herbivores to select plants that the predators are less likely to visit, potentially resulting in herbivore feeding specializations. But this argument does not take into account the behavioral flexibility shown by many generalized predators. And though Bernays and Graham (1988) emphasize the lability in plant use by herbivorous insects over a few generations, their argument implies that generalized predators are even less labile than herbivores. In fact, generalized predators (and parasitoids) typically have very labile searching behavior; though sometimes restricted by habitat type or plant architecture, they may be quick to modify searching even among different plant species if prey densities decline locally (e.g., Arthur 1966, Hassell 1978, Lawton 1986). The abilities of predators to change search patterns also typically is much more rapid than when herbivores shift host plant species. Because of these very different time scales

SUMMARY (1) Growth rate and size at moult increased with prey density for Notonecta hoffmanni Hungerford nymphs raised at various densities of their mosquito larvae prey. (2) The functional response of different sized instar IV nymphs was type-2, and large predators ate more at high prey densities than did small predators. (3) The feeding rate of individual predators did not vary in any regular way during the instar. Predator size at the start of the instar was a good predictor of the predators' functional responses for both long- and short-term experiments. Short-term experiments were good predictors of the long-term functional responses: the relationship between these two responses remains linear over a wide range of prey densities. These facts allow us to predict the functional response over the whole instar from a knowledge of predator weight. (4) Variation in predator size within one instar (IV) was as important as differences in the predators' size in different instars in determining the predator's feeding rate.