Assessment of Amblyseius fallacis (Acari: Phytoseiidae) for biological control of tetranychid mites in an Ontario peach orchard

We introduced a mass-reared pyrethroid-resistant strain of the predatory phytoseiid mite Amblyseius fallacis (Garman) into an Ontario peach orchard in an attempt to control populations of the phytophagous mites Panonychus ulmi Koch and Tetranychus urticae Koch (Acari: Tetranychidae). Releases of 1,000 and 2,000 mites per tree were made, at three different times. The release of 2,000 mites per tree in June and in July resulted in significantly higher phytoseiid densities than was observed on control trees. However, densities of P. ulmi or T. urticae were not significantly affected by any release rate or by timing. The release of 1,000 A. fallacis per tree, or of any density in August, did not significantly increase phytoseiid abundance. In the following year, population dynamics of both phytoseiid and phytophagous mites were not significantly affected by the previous year's release. Amblyseius fallacis can be a useful predator in some fruit orchards. However, further research is necessary into the timing and rate of release, modified spray programmes, and with different crops, in order to clarify the role of this species for biological control in Ontario peach orchards.     

Factors influencing the acaricidal activity of flucycloxuron

Contact activities of flucycloxuron on immature stages of the two-spotted spider mite (Tetranychus urticae (Koch)) and the European red mite (Panonychus ulmi (Koch)) gradually decrease in the successive developmental stages. The levels of contact activity of flucycloxuron on larvae and protonymphs ofT. urticae andP. ulmi are of the same order. Deutonymphs ofT. urticae are less susceptible to contact activity than the similar stage ofP. ulmi. In adultT. urticae, the transovarial ovicidal activity was used as an indicator for cuticular penetration. More than 90% of the maximal penetration into adult mites occurs within 8 h. Reversibility of the transovarial activity was not observed after 24 h, but did occur after a subsequent 48 h stay on untreated leaves. The ovo-larvicidal activity of flucycloxuron onP. ulmi after treatment of apple leaves is strongly negatively influenced by leaf age, partly by lower retention of the spray liquid on the leaves. Leaf penetration was measured by application of flucycloxuron on leaf uppersides and assessment of the transovarial activity in mites (P. ulmi orT. urticae) infested on the undersides, one day after treatment. In this test system, leaf penetration was found to be strongly species dependent. Penetration was high in cucumber, moderate in French beans, cotton, roses and strawberry, but low in apple and pepper. Leaf penetration in French bean plants is drastically reduced at increasing leaf age. The overall positive effect of increase in relative air humidity on leaf penetration, is statistically highly significant (P=0.001) for French beans and almost significant (P=0.08) for cucumbers. WithT. urticae on French bean it was found that in this test flucycloxuron needs more than one day for maximal leaf penetration. Although in apple leaves penetration from uppersides was low, penetration from undersides was much higher. The surfactants Arkopal N 130, Silwet L-77 and X2-5309 enhance penetration from leaf under-sides.     

Field Releases of the Predatory Mite Neoseiulus fallacis (Acari: Phytoseiidae) in Canada, Monitored by Pyrethroid Resistance and Allozyme Markers

The predacious phytoseiid mite Neoseiulus fallacis (Garman) is an important agent for the biological control of spider mites in deciduous fruit orchards in North America and Canada. It would be helpful to monitor the fate of released individuals to improve the results of introductions of the predators in biological control trials. We have used two types of genetic markers, pyrethroid resistance and allozymes, for indirect estimation of the survival of N. fallacis introduced in an apple orchard in Ontario, Canada. Mite samples were submitted to toxicological tests. The polymorphism of four enzymes was studied in individual females using an isoelectric focusing technique. A mite sample was taken from the field, mass-reared in the laboratory, and selected for permethrin resistance. This strain was released on several apple trees treated with permethrin, and mite samples were collected from the same trees 10 to 90 days later. The genetic composition and the insecticide resistance level of this sample were compared to those of two other samples from trees where mites had not previously been released, either in the same orchard or in a neighboring block. A control susceptible strain was compared using mites collected earlier from trees on the same site but outside the present experiment. The mites collected from the release trees and those from the strain used for the releases were found to be genetically closely related, as judged from a small genetic distance, and from similar levels of insecticide resistance in both samples. The control samples from the nonrelease trees were genetically distant from these and displayed low resistance levels. These results indicate that the released genotypes established and persisted in the release trees for the period of the experiment. The utility of the two approaches in assessing the fate of released natural enemies is discussed.     

The Transfer of Typhlodromus pyri on Grape Leaves for Biological Control of Panonychus ulmi (Acari: Phytoseiidae, Tetranychidae) in Vineyards in Ontario, Canada

The phytophagous mite Panonychus ulmi Koch has become a significant problem in Ontario vineyards. We attempted to introduce and establish populations of the predatory mite Typhlodromus pyri Scheuten for P. ulmi biological control. Grape leaves were transferred from a vineyard containing T. pyri in early summer 1998, by picking leaves from a donor vineyard and attaching them onto leaves in the release vineyard where T. pyri were extremely rare. Two release treatments were used with rates of 8.5 (1×) and 25.5 (3×) mobiles per vine. In the first season, T. pyri established in similar densities in both release treatments, which were significantly higher than controls. However, there were no differences among treatments in P. ulmi densities in 1998 as a result of predator release. During summer 1999, significantly fewer P. ulmi mite-days were observed in release plots compared to the control. Amblyseius fallacis (Garman) was common throughout the release vineyard in 1998 and in 1999, but appeared on the vines too late in the season to maintain low P. ulmi densities. T. pyri appeared to out-compete A. fallacis in 1999 because A. fallacis densities were significantly lower in plots where T. pyri had been released than in control plots. We conclude that T. pyri can be effective for P. ulmi biological control in Ontario vineyards and may be introduced by transferring leaves. In Europe, transferring prunings has been the standard method of inoculating T. pyri into new vineyards. Here we show that transferring leaves is another practical method.     

Amensalism via webs causes unidirectional shifts of dominance in spider mite communities

Competitive displacement is considered the most severe consequence of interspecific competition; if a superior competitor invades the habitat of an inferior species, the inferior species will be displaced. Most displacements previously reported among arthropods were caused by exotic species. The lack of investigation of displacement among native species may be due to their apparently harmonious coexistence, even if it is equivalent to an outcome of interspecific association. A seasonal change in the species composition of spider mites, from Panonychus ulmi to Tetranychus urticae, is observed in apple trees worldwide. Previous laboratory experiments have revealed amensal effects of T. urticae on P. ulmi via their webs. Using manipulation experiments in an orchard, we tested whether this seasonal change in species composition occurred as the result of interspecific competition between these spider mites. Invasion by T. urticae prevented an increase in P. ulmi densities throughout the experimental periods. Degree of overlap relative to the independent distribution on a leaf-surface basis (ω _S) changed from positive to negative with increasing density of T. urticae. T. urticae invasion drove P. ulmi toward upper leaf surfaces (competitor-free space). The niche adjustment by P. ulmi occurred between leaf surfaces but not among leaves. Our findings show that asymmetrical competition between T. urticae and P. ulmi plays an important role in this unidirectional displacement and that the existence of refuges within a leaf produces the apparently harmonious coexistence of the mites and obscures their negative association.     

Effects of acaricides,pyrethroids and predator distributions on populations of <Emphasis Type="Italic">Tetranychus urticae</Emphasis> in apple orchards

We sampled mites in three apple orchards in Nova Scotia, Canada, that had been inoculated with pyrethroid-resistant Typhlodromus pyri and had a history of Tetranychus urticae outbreaks. The objective of this study was to monitor populations of T. urticae and phytoseiid predators on the ground and in trees and to track dispersal between the two habitats. Pesticides were the chief cause of differences in mite dynamics between orchards. In two orchards, application of favourably selective acaricides (abamectin, clofentezine) in 2002, coupled with predation by T. pyri in trees and Neoseiulus fallacis in ground cover, decreased high T. urticae counts and suppressed Panonychus ulmi. By 2003 phytoseiids kept the tetranychids at low levels. In a third orchard, application of pyrethroids (cypermethrin, lambda-cyhalothrin), plus an unfavourably selective acaricide (pyridaben) in 2003, suppressed phytoseiids, allowing exponential increases of T. urticae in the ground cover and in tree canopies. By 2004 however, increasing numbers of T. pyri and application of clofentezine strongly reduced densities of T. urticae in tree canopies despite high numbers crawling up from the ground cover. Another influence on T. urticae dynamics was the distribution of the phytoseiids, T. pyri and N. fallacis. When harsh pesticides were avoided, T. pyri were numerous in tree canopies. Conversely, only a few N. fallacis were found there, even when they were present in the ground cover and on tree trunks. Low numbers were sometimes due to pyrethroid applications or to scarcity of prey. Another factor was likely the abundance of T. pyri, which not only competes with N. fallacis, but also feeds on its larvae and nymphs. The scarcity of a specialist predator of spider mites in trees means that control of T. urticae largely depends on T. pyri, a generalist predator that is not particularly effective in regulating T. urticae. The Canadian Crown's right to retain a non-exclusive, royalty-free licence in and to any copyright is acknowledged.     

Etudes comparatives de l'influence de la température et de la nourriture sur le développement deAmblyseius andersoni,Neoseiulus fallacis,Galendromus longipilus etTyphlodromus pyri [Acari: Phytoseiidae]

Résumé L'étude en laboratoire de la biologie du développement deAmblyseius andersoni, Neoseiulus fallacis, Galendromus longipilus etTyphlodromus pyri a permis de calculer les seuils de développement et le taux intrinsèque d'accroissement naturel à 20°C lorsque ces espèces sont nourries avecTetranychus urticae. Le taux de fécondité spécifique à l'age des 4 espèces a été déterminé en utilisant comme proie soitT. urticae, soitPanonychus ulmi. PourA. andersoni et pourT. pyri, P. ulmi constitue une nourriture meilleure queT. urticae, contrairement àN. fallacis etG. longipilus, dont le taux d'accroissement est supérieur quand ils se nourrissent deT. urticae. PourG. longipilus, P. ulmi constitue même une nourriture nettement suboptimale. Pour les femelles de toutes les espèces on a pu mettre en évidence la faculté d'adapter le taux de fécondité et la durée d'oviposition au nombre des proies disponibles. En termes de fécondité, cela permet à ces espèces de réaliser presque tout leur potentiel, indépendamment de la ration quotidienne de nourriture.      

Effect of pollen,natural prey and factitious prey on the development of <Emphasis Type="Italic">Iphiseius degenerans</Emphasis>

The predatory mite Iphiseius degenerans (Berlese) is commercially available as a biological control agent of thrips and spider mites in greenhouse crops. Developmental duration and immature survival of I. degeneransreared on nine types of food (almond pollen, apple pollen, castor bean pollen, plum pollen, sweet pepper pollen, Tetranychus urticaeKoch, Frankliniella occidentalis(Pergande), Ephestia kuehniella Zeller eggs and Artemia franciscana Kellogg cysts) and on three substrates (Multicel, sweet pepper leaf, and bean leaf) were determined in the laboratory. All experiments were carried out at 25 °C. On Multicel, mean developmental times on pollen ranged from 6.0 to 7.1 days, with the lowest value recorded on almond pollen and the highest on sweet pepper pollen. When reared on castor bean pollen significantly longer developmental times were obtained on a sweet pepper leaf compared to a bean leaf or Multicel. Developmental duration when offered T. urticaeon Multicel ranged between 6.1 and 6.9 days, on a bean leaf development was completed in 8.0 days. The longest developmental times on Multicel were recorded on Ephestia eggs (7.0 days) and on decapsulated Artemia cysts (7.5 days). No development beyond the protonymphal stage occurred in the absence of food or when encapsulated Artemia cysts or thrips larvae were offered on Multicel. On a sweet pepper leaf and a bean leaf, respectively 6.7 and 10.0% of the eggs reached adulthood when thrips larvae were provided as food; developmental times recorded here, were 9.0 and 8.3 days. Overall, immature mortality occurred mainly in the protonymphal stage and ranged from 0.0 to 36.7%. In conclusion, I. degenerans is able to feed on a variety of natural and unnatural foods, but thrips larvae and sweet pepper pollen are unfavourable food for immature development. This could compromise the establishment of this biological control agent when used against thrips in sweet pepper crops.     

Physiological effect of the host plant on the suitability ofTetranychus urticae as prey forPhytoseiulus persimilis [Acari: Tetranychidae,Phytoseiidae]

Weight gain of adult femalesPhytoseiulus persimilis Athias-Henriot was determined after they were fed eggs or adult females ofTetranychus urticae Koch reared on either Lima bean (Phaseolus vulgaris L.) or nightshade (Solanum douglasii Dunal). No significant difference was found when the predator was fed eggs ofT. urticae reared on either host. Predators were heavier when fed adult femaleT. urticae reared on bean. Apparently, a factor determining the prey's suitability is incorporated into its hemolymph or tissues and is not restricted to its gut content.      

The effect of pyrethroid lambdacyhalothrin applications on the spatial distribution of phytophagous and predatory mites in apple orchards

The spatial distribution of three phytophagous mites,Panonychus ulmi (Koch),Tetranychus urticae Koch andAculus schlechtendali (Nalepa), and two predacious mites,Zetzellia mali (Ewing) andAmblyseius fallacis (Garman), and the effect of pyrethroid lambdacyhalothrin applications on mite spatial dispersion were investigated over a 3-year period in an apple orchard in Ontario. The index of dispersion and the slope of Taylor's power law were used to evaluate dispersion patterns of mites. Panonychus ulmi showed that between-tree spatial variation decreased with an increase of population densities, whereas between-leaf variation increased with population densities. With all other four species it appeared that between-tree variation is much greater than between-leaf variation at all field population density levels. The values ofb by Taylor's power law suggested that all five species of mites are aggregated, but that in generalP. ulmi andT. urticae (b=1.427–1.872) are more aggregated than their predators (b=1.254–1.393). Taylor's regression technique suggests that pyrethroid applications causedP. ulmi, T. urticae, Z. mali andA. fallacis to be less aggregated whileA. schlechtendali was more aggregated. The impact of changes in mite spatial distribution following pyrethroid applications on sampling plans is discussed.     

Functional response of Euseius finlandicus and Amblyseius andersoni to Panonychus ulmi on apple and peach leaves in the laboratory

The functional response of adult females of the predatory mites Euseius (Amblyseius) finlandicus and Amblyseius andersoni to larvae and adult females of the fruit tree red spider mite Panonychus ulmi was determined on apple and peach leaf disks in the laboratory at 25°C and 16:8 (L:D). For adult females of P. ulmi the predation efficiency of E. finlandicus was higher on peach than on apple, whereas that of A. andersoni was higher on apple than on peach. Efficiency of predation on larvae of P. ulmi by either predator did not differ significantly between apple and peach. On both plants, A. andersoni had a higher predation rate than E. finlandicus on larvae of P. ulmi. It is concluded that in the laboratory the host plant has a substantial effect on predation efficiency of A. andersoni and E. finlandicus when they preyed on adults but not when they preyed on larvae of P. ulmi.     

Neoseiulus fallacis: dispersal and biological control of Tetranychus urticae following minimal inoculations into a strawberry field

In three separate tests, 100 adult female Neoseiulus fallacis (Garman) (plus immatures) were released at five point locations across 1.6-m rows of strawberries to control twospotted spider mites, Tetranychus urticae Koch. Beginning in April, during 6–12 weeks, predators controlled pests locally and dispersed downwind by 20–30 m. Up to 100 m² around each release point was colonized, and the entire 2.5 ha field was covered by predators by September. Distances dispersed by N. fallacis were similar within and across rows, suggesting that dispersal was primarily by aerial rather than by ambulatory means. Factors that affected dispersal were temperature, wind direction, density of spider mites, and mowing and flailing of foliage. An exponential model of dispersal was fitted to the data. On average, the area dispersed by N. fallacis doubled every 70 degree-days. From these results, a strategy of minimum release is suggested. To establish N. fallacis over a field in a single season, ca. 100 adult females per 1–2 m of row can be released before 1 July, after T. urticae have achieved 2–5 female adults per leaf. Releases should be 50 m apart and to the upwind side of the field. Selective sprays may be needed to suppress spider mites until predators gain control and disperse over the field.     

Emission of volatile organic compounds by apple trees under spider mite attack and attraction of predatory mites

Emission rates of volatile organic compounds (VOCs) from Pirus malus L. subsp. mitis (Wallr.) var. Golden Delicious and var. Starking attacked by the phytophagous mite Panonychus ulmi Koch, and their attractiveness to the predatory mites Amblyseius andersoni Chant and Amblyseius californicus McGregor, were studied during three years. A large variability was found in the emission of individual VOCs depending on the infestation, the apple tree variety and the date. There were larger total VOC emission rates and larger total VOC leaf concentrations in apple trees attacked by phytophagous mites, especially in the var. Starking. In infested trees of this variety, there were also more predatory mites. An olfactometer assay showed that predatory mites preferentially chose branches infested by Panonychus ulmi (85% went to infested branches vs 15% to uninfested control branches) indicating that volatiles may be used as cues to find their prey.     

Effects of predator-mediated apparent competition on the population dynamics of Tetranychus urticae on apples

The two spotted spider mite, Tetranychus urticae Koch, is an important herbivore pest of apple trees in Northwest China. This spider mite and another less damaging spider mite, Eotetranychus pruni Oudemans, are attacked by a common and often effective phytoseiid predator, Euseius finlandicus (Oudemans). Functional relationships were studied in the field to evaluate the impact of E. pruni and E. finlandicus on T. urticae. The results from this study showed that the predator-mediated apparent competition strongly affected the population dynamics of T. urticae. The addition of the apparent competitor E. pruni alone had little impact on T. urticae densities. Although the release of the predator E. finlandicus alone could result in reduction in T. urticae densities, the greatest reduction in T. urticae densities occurred in plots where both the predator E. finlandicus and apparent competitor E. pruni were released. In apple orchards, the early introduction of both the apparent competitor E. pruni and predator E. finlandicus would evidently increase the population size of the predator E. finlandicus and consequently significantly enhance the control of T. urticae populations. It is concluded from the study that the predator and apparent competitor release might be an appropriate control for the target species.     

Response of predatory mites with different rearing histories to volatiles of uninfested plants

The behavioural response of the predatory mite Phytoseiulus persimilis to volatiles from several host plants of its prey, spider mites in the genus Tetranychus, was investigated in a Y-tube olfactometer. A positive response to volatiles from tomato leaves and Lima bean leaves was recorded, whereas no response was observed to volatiles from cucumber leaves, or leaves of Solanum luteum and Solanum dulcamara.Different results were obtained for predators that differed in rearing history. Predators that were reared on spider mites (Tetranychus urticae) on Lima bean leaves did respond to volatiles from Lima bean leaves, while predators that had been reared on the same spider mite species but with cucumber as host plant did not respond to Lima bean leaf volatiles. This effect is compared with the effect of rearing history on the response of P. persimilis to volatile allelochemicals of prey-infested plant leaves.     

Sublethal effects of esfenvalerate residues on pyrethroid resistant Typhlodromus pyri (Acari: Phytoseiidae) and its prey Panonychus ulmi and Tetranychus urticae (Acari: Tetranychidae)

The effect of residues of esfenvalerate on oviposition of the resistant strain of the predatory mite Typhlodromus pyri and its main prey, European red mite Panonychus ulmi and two-spotted spider mite Tetranychus urticae, were investigated. T. pyri showed a significant linear reduction in oviposition after 24h in the presence of increasing levels of esfenvalerate residue applied at the field rate. Furthermore, when given a choice, T. pyri preferred to lay eggs on residue-free surfaces. Of the two prey species, only P. ulmi showed significant avoidance of increasing levels of residues of the field rate concentration of esfenvalerate, as measured by runoff mortality, however both P. ulmi and T. urticae, when given a choice, showed a preference for esfenvalerate-free surfaces. As with the predatory mite T. pyri, both prey species showed a significant linear reduction of oviposition with increasing esfenvalerate residues and a preference to lay eggs on esfenvalerate-free surfaces. Esfenvalerate residues as high as 15X field rate were not repellent to pyrethroid-resistant T. pyri. The possible effects of these sublethal effects on predator-prey dynamics and implications for integrated mite control programmes in apple orchards are discussed.     

Odour-mediated responses of a predatory mirid bug and its prey, the two-spotted spider mite

It has been shown that many natural enemies of herbivorous arthropods use herbivore induced plant volatiles (HIPVs) to locate their prey. Herbivores can also exploit cues emitted by plants infested with heterospecifics or conspecifics. A study was conducted to test whether green bean HIPVs as well as odours emitted directly by spider mites influenced the orientation behaviour of the predatory mirid bug, Macrolophus caliginosus and its prey, Tetranychus urticae in a Y-tube olfactometer. Our results show that both spider mites and M. caliginosus preferred spider mite infested green bean plants to uninfested plants. For M. caliginosus this response was mediated by HIPVs whereas for T. urticae it was mediated through a composite response to both HIPVs and odours emitted directly by the conspecifics (and their associated products). The results may be of use in practical biocontrol situations, through e.g., plant breeding for improved HIPV production, conditioning of mass-reared predators to appropriate cues, and employment of “push–pull-strategies” by using HIPVs.     

Which predatory mite can control both a dominant mite pest,Tetranychus urticae,and a latent mite pest,Eotetranychus asiaticus,on strawberry?

After biological control of Tetranychus urticae using Phytoseiulus persimilis, a latent mite pest, Eotetranychus asiaticus, was found on strawberries growing in a plastic greenhouse in western Japan. To determine whether the release ofP. persimilis, an exotic natural enemy of T.urticae, enhanced the occurrence of the latent pest, the efficiency of P. persimilis in controlling E.asiaticus was compared with the efficiency of two indigenous phytoseiid mites, Amblyseius californicus and A. womersleyi, under laboratory conditions. Phytoseiuluspersimilis and A. californicus attacked not onlyT. urticae but also E. asiaticus. However, the predators produced very few eggs and their offspring rarely reached adulthood when fed on E. asiaticus, so their potential as control agents for the latent mite appears to be low. In contrast, A.womersleyi feeding on E. asiaticus reproduced as well as when fed on T. urticae, and exterminated both species of spider mite. This suggests that A. womersleyihas greater potential as a biological control agent in crops where both species may occur. This revised version was published online in August 2006 with corrections to the Cover Date.     

Prey preference of the phytoseiid miteTyphlodromus pyri 1. Response to volatile kairomones

Using a Y-tube olfactometer, a study has been made of the response of females of the predatory miteTyphlodromus pyri Scheuten (Acarina: Phytoseiidae) to volatile kairomones of three prey species: the European red spider mite (Panonychus ulmi (Koch)), the two-spotted spider mite (Tetranychus urticae Koch) and the apple rust mite (Aculus schlechtendali (Nalepa)).Predators that had been reared onT. urticae responded only to the volatile kairomone ofP. ulmi. In contrast, when reared onVicia faba L. pollen, they responded to the kairomones of all three prey species. Pollen-reared predators, offered a choice between kairomones of two different prey species, prefer theP. ulmi kairomone to those ofA. schlechtendali orT. urticae.The difference in response between predators reared onV. faba pollen andT. urticae seems to be caused by the low carotenoid content ofV. faba pollen. Predators that had been reared onV. faba pollen mixed with crystalline -carotene behaved in a way similar to conspecific that had been reared on the carotenoid-rich prey miteT. urticae. Obviously, pollen-rearedT. pyri females are in need of carotenoids, which can be obtained from, e.g.,P. ulmi, T. urticae orA. schlechtendali. This may explain why pollen-reared predators respond to more prey species thanT. urticae-reared predators.WhyT. pyri females need carotenoids has not been established. The only known function of carotenoids in mites is involvement in diapause induction. However, as pollen-rearedT. pyri enter reproductive diapause under short-day conditions, they either extract sufficient amounts of carotenoids fromV. faba pollen, or do not need carotenoids for diapause induction.Apart from the effect of dietary requirements on prey selection, food deprivation also affects the predator's response to kairomones. All the data mentioned above have been obtained for predators that had been starved for 20 h. Predators that had been reared onT. urticae and starved for 48 h before the experiment did respond to the volatile kairomone ofT. urticae in contrast to predators from the same culture that had been starved for 20 h. Thus foraging decisions byT. pyri are affected by both starvation time and specific hunger for carotenoids.     

Effect of the width of the herbicide strip on mite dynamics in apple orchards

Herbicide strips are used in apple orchards to promote tree growth and survival, to increase yield and to reduce the risk of rodent damage to tree bark. However, herbicide strips, particularly wider ones, may cause problems including soil erosion, reduced organic matter, leaching of nitrates into ground water and increased incidence of plant diseases and pests, including two-spotted spider mites, Tetranychus urticae Koch. In this 2 year study we monitored mite dynamics in apple trees and used sticky bands on tree trunks to determine rates of T. urticae immigration into Nova Spy apple trees in plots with wide (2 m) or narrow (0.5 m) herbicide strips. Use of wider herbicide strips promoted two risk factors that could trigger outbreaks of tetranychid mites. First, concentrations of leaf N in apple trees were higher and those of P and K were lower with the wide strips. Such changes in nutritional quality of leaves would increase the potential for more rapid population growth of T. urticae, and to a lesser extent, the European red mite, Panonychus ulmi (Koch). Second, there were higher rates of T. urticae immigration from the ground cover vegetation into the trees. In 2006, and for most of 2007, densities of T. urticae were higher with wide herbicide strips, whereas densities of P. ulmi were not enhanced. However, by late August to early September in 2007, densities of both tetranychids were lower with wide herbicide strips. This is because both risk factors were counterbalanced, and eventually negated, by the enhanced action of phytoseiid predators, mostly Typhlodromus pyri Scheuten. From July through September 2006, ratios of phytoseiids to tetranychids were always several-fold lower with wide herbicide strips but in 2007, from mid-July onwards, predator–prey ratios were usually several-fold higher with wide strips. However, this numerical response of phytoseiids to prey density can only occur where the pesticide program in orchards is not too harsh on phytoseiids. Hence the impact of width of herbicide strip is contingent on the composition and size of the phytoseiid complex and the impact of pesticides on predation.