Comparative studies of two species of predacious phytoseiid mites (Acarina: Phytoseiidae), with special reference to their responses to the density of their prey

Imperfect diet choice reduces the performance of a predatory mite

Two mutually unexclusive hypotheses prevail in the theory of nutritional ecology: the balanced diet hypothesis states that consumers feed on different food items because they have complementary nutrient and energy compositions. The toxin-dilution hypothesis poses that consumers feed on different food items to dilute the toxins present in each. Both predict that consumers should not feed on low-quality food when ample high-quality food forming a complete diet is present. We investigated the diet choice of Phytoseiulus persimilis, a predatory mite of web-producing spider mites. It can develop and reproduce on single prey species, for example the spider mite Tetranychus urticae. A closely related prey, T. evansi, is of notorious bad quality for P. persimilis and other predator species. We show that juvenile predators feeding on this prey have low survival and do not develop into adults. Adults stop reproducing and have increased mortality when feeding on it. Feeding on a mixed diet of the two prey decreases predator performance, but short-term effects of feeding on the low-quality prey can be partially reversed by subsequently feeding on the high-quality prey. Yet, predators consume low-quality prey in the presence of high-quality prey, which is in disagreement with both hypotheses. We suggest that it is perhaps not the instantaneous reproduction on single prey or mixtures of prey that matters for the fitness of predators, but that it is the overall reproduction by a female and her offspring on an ephemeral prey patch, which may be increased by including inferior prey in their diet.

Artificial selection for timing of dispersal in predatory mites yields lines that differ in prey exploitation strategies

Dispersal is the main determinant of the dynamics and persistence of predator–prey metapopulations. When defining dispersal as a predator exploitation strategy, theory predicts the existence of a continuum of strategies: from some dispersal throughout the predator–prey interaction (the Milker strategy) to dispersal only after the prey had been exterminated (the Killer strategy). These dispersal strategies relate to differences in prey exploitation at the population level, with more dispersal leading to longer predator–prey interaction times and higher cumulative numbers of dispersing predators. In the predatory mite Phytoseiulus persimilis, empirical studies have shown genetic variation for prey exploitation as well as for the timing of aerial dispersal in the presence of prey. Here, we test whether artificial selection for lines that differ in timing of dispersal also results in these lines differing in prey exploitation. Six rounds of selection for early or late dispersal resulted in predator lines displaying earlier or later dispersal. Moreover, it resulted—at the population level—in predicted differences in the local predator–prey interaction time and in the cumulative numbers of dispersers in a population dynamics experiment. We pose that timing of dispersal is a heritable trait that can be selected in P. persimilis, which results in lines that show quantitative differences in local predator–prey dynamics. This opens ways to experimentally investigate the evolution of alternative prey exploitation strategies and to select for predator strains with prey exploitation strategies resulting in better biological control.

Estimating intrinsic growth rates of arthropods from partial life tables using predatory mites as examples

The intrinsic rate of natural increase of a population (r_m) has been in focus as a key parameter in entomology and acarology. It is considered especially important in studies of predators that are potential biological control agents of fast-growing pests such as mites, whiteflies and thrips. Life-table experiments under controlled laboratory conditions are standard procedures to estimate r_m. However, such experiments are often time consuming and may critically depend on the precise assessment of the developmental time and the fecundity rate early in the reproductive phase. Using selected studies of predatory mites with suitable life-table data, we investigated whether and how measurements of growth rates can be simplified. We propose a new method for estimating r_m from partial life tables, in which the researcher can choose a level of precision based on a stand-in measure of relative error. Based on this choice, the procedure helps the researcher to decide when a life-table experiment can be terminated. Depending on the chosen precision, significant amounts of experimental time can be saved without seriously compromising the reliability of the estimated growth parameter.

Oviposition-site shift in phytophagous mites reflects a trade-off between predator avoidance and rainstorm resistance

Predators can reduce prey population densities by driving them to undertake costly defences. Here, we report on a remarkable example of induced antipredator defence in spider mites that enhances the risk to rainstorms. Spider mites live on the undersides of host plant leaves and usually oviposit on the leaf undersurface. When they are threatened by predatory mites, they oviposit on three-dimensional webs to avoid egg predation, although the cost of ovipositing on webs has not yet been clearly determined. We prepared bean plants harbouring spider mite (Tetranychus kanzawai) eggs on either leaf surfaces or webs and exposed them to rainstorms outdoors. We found that fewer eggs remained on webs than on leaf surfaces. We then examined the synergistic effect of wind and rain by simulating both in the laboratory. We conclude that ovipositing on webs comes at a cost, as eggs are washed off the host plants by wind and rain. This may explain why spider mite populations decrease drastically in the rainy season, although they inhibit leaf undersides where they are not directly exposed to rainfall.

Compatibility of Bifenazate and Phytoseiulus persimilis for Management of Twospotted Spider Mites in North Carolina Staked Tomatoes

The twospotted spider mite (TSSM, Tetranychus urticae Koch) is a key pest of tomatoes (Solanum lycopersicon L. [Solanales: Solanaceae]) in North Carolina, and its management has relied principally on synthetic acaricides. Augmentative biological control of TSSM is a commonplace and effective management strategy in greenhouses worldwide, but in field-grown vegetable crops biocontrol of TSSM is poorly developed. We conducted small-plot field experiments in 2016 and 2019 to test the ability of the predatory mite, Phytoseiulus persimilis Athias-Henriot, to establish, disperse, and reduce TSSM in staked tomatoes, as well as test their compatibility with a commonly used reduced-risk acaricide, bifenazate (Acramite). Across both years, the most effective treatment for reducing TSSM was the combination of P. persimilis + Acramite. Phytoseiulus persimilis successfully established both years, but its population growth and dispersal were greater in 2016 than 2019. Acramite alone significantly reduced TSSM pressure in 2019 and P. persimilis alone did not reduce TSSM pressure below the control in either year. However, results in 2016 were influenced by the eventual dispersal of P. persimilis into all experimental plots, despite efforts to hinder their movement with corn barriers. Results in 2019 were influenced by the late-season invasion of TSSM into the trial, which delayed P. persimilis releases and influenced their establishment and growth. This study is the first to demonstrate the utility and limitations of P. persimilis in staked field tomatoes, which is a promising option for growers to manage TSSM in tomatoes in the southeast United States.

Founder effects on trans-generational dynamics of closed inbreeding lineages of the predatory mite Phytoseiulus persimilis

Both close inbreeding and distant outbreeding may reduce fitness below the level of individuals with intermediate parental relatedness. In the haplodiploid plant-inhabiting predatory mite Phytoseiulus persimilis, which is patchily distributed within and among host plants, fitness is indeed reduced in the short term, i.e. by a single generation of inbreeding. However, in the medium to long term (multiple generations), distant out-breeding should provide for favorable demographic founder effects in isolated populations. We tested this prediction in isolated experimental lineages founded by females mated to a sibling (close inbreeding), a male from the same population (intermediate relatedness) or a male from another population (distant outbreeding) and monitored lineage growth and persistence over four generations. Cross-generationally, lineages founded by distantly outbred females performed the best, i.e. produced the most descendants. However, this was solely due to superior performance from the F₂ generation onwards, whereas in the F₁ generation, lineages founded by females mated to males from their own population (intermediate relatedness) performed the best, as predicted from short-term in- and out-breeding depression effects. At the genetic level, this result was most likely due to distantly outbred founders introducing higher allelic variability and lower homozygosity levels, counterbalancing inbreeding depression, which inevitably occurs in isolated lineages, from the F₂ generation onwards.

The response of three species of phytoseiid mite (Acari: Phytoseiidae) to synthetic pyrethroid pesticides in the laboratory and the field

The Kanzawa spider mite, Tetranychus kanzawai Kishida, is a major pest in tea fields [Camellia sinensis (L.) O. Kuntze] in Japan. However, recently, there have been some instances where acaricides are no longer applied as a result of the low occurrence of T. kanzawai in tea fields in Japan. In the period of 2015-2017, surveys of predatory mites in the study tea field detected Amblyseius eharai Amitai and Swirski, Phytoseiulus persimilis Athias-Henriot, Euseius sojaensis (Ehara), Amblyseius obtuserellus Wainstein and Begljarov, and Typhlodromus vulgaris Ehara in tea fields, but not Neoseiulus womersleyi (Schicha), indicating that a major change in the composition of the phytoseiid mite population had occurred. In laboratory studies, we confirmed the ability to avoid synthetic pyrethroid insecticides of the major beneficial mites in tea fields, A. eharai and P. persimilis, but not of E. sojaensis, a predatory mite whose population declined heavily after pesticide application. Attempts are made in this study to associate the decrease in T. kanzawai frequency in Japan with changes in pesticide used, method of spraying, and composition of the phytoseiid mite population. By continuing the method of pesticide spraying ('partial surface'), which leaves refugia in the leaf layer with sub-lethal dosages of pesticide, phytoseiid mites are aided to evade pesticides, resulting in maintenance of the composition of the phytoseiid mite populations in terms of diversity and abundance. Maintaining the diversity and abundance of Phytoseiidae may have contributed to the stabilization of the T. kanzawai population at low densities in Japanese tea fields.

Biological and life table parameters of Typhlodromus laurentii and Iphiseius degenerans (Acari, Phytoseiidae) fed on Panonychus citri and pollen of Oxalis pes-caprae under laboratory conditions

Typhlodromus laurentii and Iphiseius degenerans are two generalist phytoseiid mites, broadly spread in the Mediterranean area, especially in citrus orchards. In the present work we report results on various biological and life table parameters of the two phytoseiids, fed on pollen of Oxalis pes-caprae and various stages of the tetranychid Panonychus citri. Iphiseius degenerans had the shortest post embryonic development (6.53 days), the highest oviposition rate (1.83 eggs/female/day) and the shortest mean time between eggs laid (0.55 day) on Oxalis pollen, whereas the two food types did not influence these parameters in T. laurentii. However, Oxalis pollen showed a positive effect on the survivorship of the latter phytoseiid, with a median life time (LT50) of 44.51 days, which was two times longer than that registered on prey with the same phytoseiid, and on both food types with I. degenerans. This latter species had a better performance on the pollen (rm = 0.243, λ = 1.275, Ro = 22.88, DT = 2.85) than on prey (rm = 0.182, λ = 1.199, Ro = 17.43, DT = 3.81). On the other hand, the pollen influenced the net reproductive rate (25.43 females/female) of T. laurentii positively but showed the same effect as the prey on the other demographic parameters. Our results improve knowledge on the feeding behaviour of the above mentioned phytoseiids on two food sources that could represent the main possibility to maintain a consistent population of these predators during winter. Moreover, both phytoseiids were shown to be good biocontrol candidates of P. citri populations.

Potential lethal and non-lethal effects of predators on dispersal of spider mites

Predators can affect prey dispersal lethally by direct consumption or non-lethally by making prey hesitate to disperse. These lethal and non-lethal effects are detectable only in systems where prey can disperse between multiple patches. However, most studies have drawn their conclusions concerning the ability of predatory mites to suppress spider mites based on observations of their interactions on a single patch or on heavily infested host plants where spider mites could hardly disperse toward intact patches. In these systems, specialist predatory mites that penetrate protective webs produced by spider mites quickly suppress the spider mites, whereas generalist predators that cannot penetrate the webs were ineffective. By using a connected patch system, we revealed that a generalist ant, Pristomyrmex punctatus Mayr (Hymenoptera: Formicidae), effectively prevented dispersal of spider mites, Tetranychus kanzawai Kishida (Acari: Tetranychidae), by directly consuming dispersing individuals. We also revealed that a generalist predatory mite, Euseius sojaensis Ehara (Acari: Phytoseiidae), prevented between-patch dispersal of T. kanzawai by making them hesitate to disperse. In contrast, a specialist phytoseiid predatory mite, Neoseiulus womersleyi Schicha, allowed spider mites to escape an initial patch, increasing the number of colonized patches within the system. Our results suggest that ants and generalist predatory mites can effectively suppress Tetranychus species under some conditions, and should receive more attention as agents for conservation biological control in agroecosystems.

Consumption rate, functional response and preference of the predaceous mite Iphiseius degenerans to Tetranychus urticae and Eutetranychus orientalis

The functional response of females of the phytoseiid mite, Iphiseius degenerans (Berlese), to increasing densities of females of its prey, Tetranychus urticae Koch and Eutetranychus orientalis Klein, on bean leaves, were studied under laboratory conditions. Our results indicated that the predator consumed significantly more items of E. orientalis than of T. urticae at all densities treatments. Daily consumption of the predator increased with increasing prey density until a plateau was reached-maximum number of prey consumed was ca. 4 for T. urticae and ca. 12 for E. orientalis. A Type II functional response was determined by a logistic regression model. The highest estimated value a (instantaneous rate of attack) and the lowest value of T ( h ) (handling time) were found for the predator feeding on E. orientalis. Prey selection was evaluated by simultaneously presenting both prey species to the predator in various ratios and at increasing densities. I. degenerans showed a higher predation rate and higher preference for E. orientalis at all the ratios and prey densities tested. This may be due to the smaller size or the inactivity of E. orientalis and the inability of the predator to cope with the webbing of T. urticae. Our results suggest that I. degenerans can be considered a suitable biological control candidate based on its preference for E. orientalis in the Mediterranean region.

Effect of temperature on the life-history traits of Neoseiulus californicus (Acari: Phytoseiidae) fed on Panonychus ulmi

The developmental rate and reproductive biology of Neoseiulus californicus, a generalist predator on spider mites and small insects, was investigated in the laboratory at five constant temperatures: 15, 20, 25, 30, and 34°C. The European red mite, Panonychus ulmi, an important pest in Korean apple orchards, was used as prey. Mean developmental time and adult longevity were inversely related to temperature from 15 to 30°C. Lifetime fecundity was greatest at 25°C, whereas daily fecundity was highest at 30°C. The sex ratio (female to male) was highest (0.77) at 25°C and lowest (0.67) at 34°C. Survivorship during immature development varied from 74.3 to 92.9%, with the lowest rate at 34°C. Life table parameters were analyzed and pseudo-replicates for the generation time (t ( G )), the intrinsic rate of natural increase (r (m)), finite rate of increase (λ), net reproductive rate (R (0)), and doubling time (t ( D )) were generated using the Jackknife method. Generation time (t ( G )) was lowest (10.7 days) at 34°C, R (0) was highest (49.2) at 25°C, and both r (m) (0.29) and λ (1.34) were highest at 30°C. In conclusion, the development and adult life-history traits obtained for N. californicus fed on P. ulmi indicated significant potential for biological control.

Effect of trichomes on the predation of Tetranychus urticae (Acari: Tetranychidae) by Phytoseiulus macropilis (Acari: Phytoseiidae) on tomato, and the interference of webbing

The two-spotted spider mite, Tetranychus urticae Koch, is an important pest of tomato in different parts of the world. Biological control of this pest on this crop has not been very successful. Phytoseiulus macropilis (Banks) has been used commercially for the control of T. urticae on different crops, but no information has been published on its potential to control T. urticae on tomato. The objective of this work was to compare the performance of a Brazilian population of P. macropilis on tomato with its performance on other plant species, relating the observed variation to the respective types and densities of trichomes. It has been hypothesized that the presence of the webbing produced by T. urticae could help the predator to avoid contact with trichomes and consequently to improve its performance on tomato plants. This hypothesis was also evaluated. Phytoseiulus longipes Evans was included in the work as a control, given that it has been reported to be a promising predator of Tetranychus species on tomato. The study was conducted under laboratory conditions. It was found that the performance of P. macropilis was similar to that of P. longipes and that trichomes hampered the locomotion as well as prey consumption and oviposition rate of both predators; that the presence of webbing resulted in higher levels of prey consumption and of predator oviposition; and that the presence of webbing eliminated the negative effect of trichomes of eggplants and partially eliminated the negative effect of trichomes of the 'cerasiforme' tomato variety. The observed density of trichomes in 'Carmem' one of the most common tomato varieties grown in Brazil, seems not to have interfered significantly with the prey consumption and the oviposition rate of P. macropilis. The results suggest that the latter is a promising predator of T. urticae on tomato. Complementary studies are warranted, to further evaluate the potential of P. macropilis for use as a biological control agent of this pest.

Suitability of the predatory mites Iphiseiodes zuluagai and Euseius concordis in controlling Polyphagotarsonemus latus and Tetranychus bastosi on Jatropha curcas plants in Brazil

One of the most promising plant species for biofuel production in Brazil is the physic nut Jatropha curcas. Major phytosanitary problems include the attack of two pest mite species, the broad mite Polyphagotarsonemus latus and the spider mite Tetranychus bastosi. Owing to pesticide-related problems, there is an increasing demand for sustainable environmental-friendly control methods such as biological control. In this study we evaluated the suitability of the predatory mite species Iphiseiodes zuluagai and Euseius concordis in controlling P. latus and T. bastosi on J. curcas. The number of T. bastosi killed by I. zuluagai was lower than the number of P. latus consumed. Euseius concordis preyed upon both T. bastosi and P. latus but the number of prey killed was always lower in comparison with I. zuluagai. However, P. latus and T. bastosi are suitable for the development of I. zuluagai and E. concordis as oviposition of both predators did not differ in relation to prey species. The preference of I. zuluagai for leaves of plants infested by either P. latus or T. bastosi, combined with the higher values for predation obtained by this predatory mite when fed on P. latus, compared to those values obtained by E. concordis, suggests that I. zuluagai can be more efficient than E. concordis in reducing populations of P. latus and T. bastosi under field conditions. Furthermore, we report here on the first record of predatory mites associated with P. latus and T. bastosi on native J. curcas plants in Brazil. In conclusion, we emphasize the crucial importance of predatory mites as agents of natural biological control of mite pests on J. curcas in small farms.

Comparison of thread-cutting behavior in three specialist predatory mites to cope with complex webs of Tetranychus spider mites

Anti-predator defenses provided by complex webs of Tetranychus mites can severely impede the performance of generalist predatory mites, whereas this may not be true for specialist predatory mites. Although some specialist predatory mites have developed morphological protection to reduce the adverse effects of complex webs, little is known about their behavioral abilities to cope with the webs. In this study, we compared thread-cutting behavior of three specialist predatory mites, Phytoseiulus persimilis, Neoseiulus womersleyi and N. californicus, exhibited inside the complex web of T. urticae. No major difference was observed among them in the basic pattern of this behavior, using chelicerae and palps, and in the number of silken threads severed while moving inside the web. These results and observations suggest that each predator species cut many sticky silken threads to move inside the complex web without suffering from serious obstruction.

The predatory mite Phytoseiulus persimilis adjusts patch-leaving to own and progeny prey needs

Integration of optimal foraging and optimal oviposition theories suggests that predator females should adjust patch leaving to own and progeny prey needs to maximize current and future reproductive success. We tested this hypothesis in the predatory mite Phytoseiulus persimilis and its patchily distributed prey, the two-spotted spider mite Tetranychus urticae. In three separate experiments we assessed (1) the minimum number of prey needed to complete juvenile development, (2) the minimum number of prey needed to produce an egg, and (3) the ratio between eggs laid and spider mites left when a gravid P. persimilis female leaves a patch. Experiments (1) and (2) were the pre-requirements to assess the fitness costs associated with staying or leaving a prey patch. Immature P. persimilis needed at least 7 and on average 14+/-3.6 (SD) T. urticae eggs to reach adulthood. Gravid females needed at least 5 and on average 8.5+/-3.1 (SD) T. urticae eggs to produce an egg. Most females left the initial patch before spider mite extinction, leaving prey for progeny to develop to adulthood. Females placed in a low density patch left 5.6+/-6.1 (SD) eggs per egg laid, whereas those placed in a high density patch left 15.8+/-13.7 (SD) eggs per egg laid. The three experiments in concert suggest that gravid P. persimilis females are able to balance the trade off between optimal foraging and optimal oviposition and adjust patch-leaving to own and progeny prey needs.

The effects of prey patchiness, predator aggregation, and mutual interference on the functional response of Phytoseiulus persimilis feeding on Tetranychus urticae (Acari: Phytoseiidae, Tetranychidae)

The spatial distributions of two-spotted spider mites Tetranychus urticae and their natural enemy, the phytoseiid predator Phytoseiulus persimilis, were studied on six full-grown cucumber plants. Both mite species were very patchily distributed and P. persimilis tended to aggregate on leaves with abundant prey. The effects of non-homogenous distributions and degree of spatial overlap between prey and predators on the per capita predation rate were studied by means of a stage-specific predation model that averages the predation rates over all the local populations inhabiting the individual leaves. The empirical predation rates were compared with predictions assuming random predator search and/or an even distribution of prey. The analysis clearly shows that the ability of the predators to search non-randomly increases their predation rate. On the other hand, the prey may gain if it adopts a more even distribution when its density is low and a more patchy distribution when density increases. Mutual interference between searching predators reduces the predation rate, but the effect is negligible. The stage-specific functional response model was compared with two simpler models without explicit stage structure. Both unstructured models yielded predictions that were quite similar to those of the stage-structured model.

Influence of prey on developmental performance, reproduction and prey consumption of Neoseiulus californicus (Acari: Phytoseiidae)

The Spical strain of the predatory mite Neoseiulus californicus (McGregor) is used as a biological control agent, but little is known about its preferred prey and host plants in Japan. Here we studied the development, reproduction and prey consumption of the Spical strain when fed on eggs of five different spider mite species deposited on both their laboratory-rearing plant and cherry, on which all five spider mite species developed well. The developmental periods of immature N. californicus females and males were significantly affected by the prey species they fed on, but not by the plants. No difference was found between males and females. The developmental period was shorter on eggs of two Tetranychus species than on eggs of Panonychus ulmi. Immature females had a higher predation rate than immature males. Preoviposition period, oviposition period and the number of eggs laid per female were not significantly affected by either the plants or the type of prey eggs. The postoviposition period and total adult longevity were shorter on eggs of P. ulmi than of the other four prey species, but there was no effect of plant substrate. The postoviposition period of the Spical strain was much longer than that of other N. californicus strains or other predatory mite species: the postoviposition period of the Spical strain was more than three times longer than the oviposition period, accounting for more than 75% of the total adult longevity. This suggests that the females need multiple mating to reach full egg load, but this remains to be tested. Total consumption by N. californicus adults was lower for eggs of P. ulmi than for eggs of the other four species, apparently because of the shorter postoviposition period when fed on eggs of P. ulmi. The intrinsic rates of natural increase (r(m)) on the rearing plant did not differ among prey species, whereas those on cherry were significantly different: the value was higher on Tetranychus urticae eggs than on eggs of other species. Only when N. californicus fed on T. urticae eggs, the r(m)-values were significantly different between the rearing plant and cherry (higher on cherry). Thus, the Spical strain of N. californicus could feed on eggs of all five spider mite species, deposited on a variety of plants with similar r(m)-values, suggesting that it could be successfully used to control spider mites in orchards and various crop fields of Japan.

Influence of diet on life table parameters of Iphiseius degenerans (Acari: Phytoseiidae)

Abstract. Studies on the reproduction, longevity and life table parameters of Iphiseius degenerans (Berlese) were carried out under laboratory conditions of 25 +/- 1 degree C, 75 +/- 5% RH and 16L:8D h. As food sources for the predatory mite, Ricinus communis L. pollen, all stages of the spider mite Tetrranchus urticae Koch, Frankliniella occidentalis (Pergande) larvae, and Ephestia kuehniella Zeller eggs were selected. All diets were accepted as food by the adult mites. Female longevity ranged from 29.5 to 42.4 days, the highest value was recorded on a diet of Ephestia eggs. The highest percentage of females escaping the experimental arena was observed on the diet consisting of thrips larvae. The highest oviposition rate (1.9 eggs/female.day) was recorded when the predator was fed on spider mites on an artificial substrate. For other diets, oviposition rates ranged from 1.0 to 1.3 eggs/female.day. The intrinsic rate of natural increase (r(m)) of I. degenerans varied between 0.015 and 0.142 females/female x day. The diet consisting of castor bean pollen resulted in the highest population growth whereas the diet on spider mites brushed off onto a bean leaf arena resulted in the slowest population growth. This can be explained by the inability of the predator to cope with the webbing of T. urticae, and the high escape rate of the progeny when reared on spider mites. The percentage of females in the offspring ranged from 40 to 73%.

Spatial scale of aggregation in three acarine predator species with different degrees of polyphagy

Aggregative responses by the predatory mites, Phytoseiulus persimilis, Typhlodromus occidentalis, and Amblyseius andersoni (Acari: Phytoseiidae), to spatial variation in the density of mobile stages of Tetranychus urticae (Acari: Tetranychidae) were studied over different spatial scales on greenhouse roses. Significant spatial variations in prey numbers per leaflet, per leaf, per branch or per plant were present in all experimental plots. None of the predator species responded to prey numbers per plant, and all searched randomly among plants. Within a plant, the oligophagous P. persimilis searched randomly among branches, but aggregated strongly among leaves within a branch and among leaflets within a leaf. The narrowly polyphagous T. occidentalis searched randomly among leaflets within a leaf and amond leaves within a branch, but aggregated strongly among leaflets or leaves within a plant. The boradly polyphagous A. andersoni searched randomly among leaflets within a leaf, a branch or a plant, and among leaves within a branch or a plant, but distributed themselves more often on branches with lower prey densities. Thus, specialist predators aggregate strongly at lower spatial levels but show random search at higher spatial levels, whereas generalist predators show random search at lower spatial levels but aggregate at higher spatial levels. This is the first empirical evidence demonstrating the relation between the degree of polyphagy and the spatial scale of aggregation. It is also concluded that both the prey patch size (i.e. grain) and predator foraging range (i.e. extent) are important for analyzing spatial scales of predator aggregation. The importance of studying spatial scale of aggregation is also discussed in relation to predator-prey metapopulation dynamics.