- Email: [email protected]
Effects of temperature on the establishment potential of the predatory mite Amblyseius californicus McGregor (Acari: Phytoseiidae) in the UK
Journal of Insect Physiology 48 (2002) 593–599 www.elsevier.com/locate/jinsphys
Effects of temperature on the establishment potential of the predatory mite Amblyseius californicus McGregor (Acari: Phytoseiidae) in the UK A.J. Hart a,∗, J.S. Bale a, A.G. Tullett a, M.R. Worland b, K.F.A. Walters c a
School of Biosciences, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK b British Antarctic Survey, High Cross, Madingley Road, Cambridge CB3 0ET, UK c Central Science Laboratory, Sand Hutton, York YO41 1LZ, UK
Received 28 November 2001; received in revised form 10 March 2002; accepted 26 March 2002
Abstract Amblyseius californicus was introduced into the UK in the early 1990s as a biocontrol agent against glasshouse red spider mite Tetranychus urticae. This study investigated the effects of temperature on the establishment potential of A. californicus in the UK in the light of recent reports of their successful overwintering outside of glasshouse environments. The developmental thresholds were 9.9 and 8.6 °C respectively using simple and weighted linear regression. Using the day-degree requirement per generation calculated by weighted regression (143 day-degrees) in combination with climate data, it was estimated that up to seven generations would be possible annually outdoors in the UK. Non-diapausing adult females froze at ⫺22 °C, with 100% mortality after reaching their freezing temperature. Up to 90% of mites died before freezing after short exposures to low temperatures. Significant acclimation responses occurred; 90% of acclimated individuals survived 26 days exposure at 0 °C and 11 days at ⫺5 °C (acclimated mites were reared at 19 °C, 6L:18D followed by 1 week at 10 °C, 12L:12D). Non-diapausing adult females survived over 3 months outdoors in winter under sheltered conditions and oviposition was observed. The experimental protocol used in this study is discussed as a pre-release screen for the establishment potential of other Amblyseius species, and similar non-native biocontrol agents. 2002 Elsevier Science Ltd. All rights reserved. Keywords: Amblyseius californicus; Tetranychus urticae; Cold tolerance; Temperature; Biological control
1. Introduction Amblyseius californicus McGregor (Acari: Phytoseiidae), also known as Neoseiulus californicus, is a predatory mite originating from California and Florida, used extensively in biological control programs against red spider mite (Tetranychus urticae Koch, Tetranychidae) on a global scale. A. californicus is widely used in the Mediterranean region, particularly Italy and Spain where it is reported to occur naturally (Castagnoli and Simoni, 1991; Griffiths, 1999). Recently, A. californicus has been imported into the UK
Corresponding author. Tel.: +44-121-414-5925; fax: +44-121414-5562. E-mail address: [email protected] (A.J. Hart). ∗
as a biocontrol agent and released in greenhouses in a series of ‘limited release’ trials. However, due to inevitable escapes from these environments, it is now reported to be established outdoors in several areas of southern Britain, and its distribution could potentially expand (Jolly, 2000). Phytoseiid mites have long been used in the control of glasshouse pests including red spider mite and thrips (McMurtry and Croft, 1997), and their use is increasing owing to the pressure on growers to find alternatives to chemical pesticides (van Lenteren, 2000). Phytoseiulus persimilis was the first predatory mite to be used in the control of red spider mite in glasshouses in UK in the late 1960s (Hussey and Bravenboer, 1971). Since then, the use of this species has increased greatly and has become an example of successful biological control based on a thorough knowledge of the plant/pest/predator relationship (Caltagirone, 1981). The
0022-1910/02/$ - see front matter 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 2 - 1 9 1 0 ( 0 2 ) 0 0 0 8 7 - 2
594
A.J. Hart et al. / Journal of Insect Physiology 48 (2002) 593–599
use of other predatory mites has been increasing since the 1980s, including species from the genus Amblyseius (van Lenteren and Woets, 1988; van Lenteren, 1995). To assess the risk of establishment of a biocontrol agent such as A. californicus, various abiotic and biotic factors need to be considered (Bale and Walters, 2001; Hart et al., 2002). The most important abiotic determinant is temperature, which affects survival, reproduction and development (Bale and Walters, 2001). Successful establishment in a new habitat requires an adequate thermal budget (number of day-degrees above the developmental threshold) for development and reproduction, and the ability to survive harsh conditions that may be encountered during the winter season. In addition to the abiotic factors, availability of prey and in some cases, the ability to enter diapause, are also necessary considerations when assessing establishment potential. There is recent evidence that a naturalised strain of A. californicus has become established outdoors in some areas of the UK and possesses the ability to enter diapause in winter (Jolly, 2000). It is not known if commercial strains supplied by biocontrol companies can diapause immediately upon escape from release sites, but it is clear that the ability to diapause can be selected for relatively quickly, as all of the licensed releases of A. californicus in the UK have occurred within the last 10 years. A relationship between diapause and cold hardiness ability has been shown in a number of arthropods, with the diapausing stage having an increased level of cold tolerance compared to non-diapausing counterparts (see Denlinger, 1991). Diapause can be induced in A. californicus and related species by exposing individuals throughout their juvenile development to short day lengths and temperatures below 20 °C (Fitzgerald and Solomon, 1991; Morewood and Gilkeson, 1991; Veerman, 1992; Morewood, 1993). Not all species, strains or populations are capable of entering diapause, and some predatory mites can overwinter successfully without diapausing (Morewood, 1993). Although there are many studies on the biology and diapause of A. californicus and other phytoseiid mite species, few have investigated their cold hardiness (Morewood, 1992, 1993), and, in general, there is insufficient information for a comprehensive risk assessment of the establishment potential of these species in new environments. Overwintering survival is an important factor often overlooked in the licensing system for the release of non-native biocontrol agents in the UK. The aim of this study was to assess the overwintering ability and effects of temperature on a commercially available strain of A. californicus to assess the risk of establishment posed by individuals escaping from glasshouse environments.
2. Materials and methods 2.1. Rearing of A. californicus A strain of A. californicus, originally imported from USA, supplied by Syngenta Bioline (formerly Novartis BCM) was reared at 26 °C with a photoperiod of 18L:6D (Treatment 1) under quarantine conditions at the University of Birmingham. Populations to be used for experimentation were established on artificial arenas: black tiles (approximately 20 × 12cm2) ringed by moist filter paper and Oecotak were placed on sponge blocks in containers half filled with water to act as a barrier to the mites. A. californicus were reared on the black tiles and fed each day with T. urticae. Dwarf French beans (Phaseolus vulgaris L.) were grown and infested with red spider mite and used as a source of prey for the predators. French bean leaves were added daily and removed the following day and any eggs laid by A. californicus were used to set up new cultures for experimentation as and when required. Populations of A. californicus were also reared under potential diapause-inducing conditions in incubators maintained at 19 °C and 6L:18D photoperiod (Treatment 2) (Fitzgerald and Solomon, 1991; Jolly, 2000). Adult female A. californicus were placed on artificial arenas as described previously and oviposited on to French bean leaves. The eggs were allowed to hatch and the nymphs developed into adults under these conditions, with excess red spider mite provided as prey. Fifty gravid adult females reared under these conditions were selected and placed on individual 1 cm2 arenas created by dividing up the tiles with strips of moist filter paper and lines of Oecotak and monitored daily over a 7 day period for oviposition at 19 °C, 6L:18D. Failure to oviposit was used as an indicator of the successful induction of an adult reproductive diapause. To obtain acclimated populations of A. californicus, females that had been reared under diapausing conditions (Treatment 2) were transferred to incubators maintained at 10 °C with a 12L:12D photoperiod for 7 days (Treatment 3). 2.2. Development Incubators were set at a range of temperatures (13.5, 14.5, 19, 20, 22.5, 25.5 and 30.5 °C) with a photoperiod of 18L:6D. Temperatures of the incubators were monitored throughout the experiments using Tinytalk dataloggers. Female A. californicus were maintained at 25 °C on French bean leaves infested with red spider mite and eggs laid by A. californicus were collected over a 24 h period. Individual eggs were placed on 1 cm2 artificial arenas which were created by sub-dividing black tiles with strips of moist filter paper and Oecotak, placed in containers half filled with water (as described previously) and kept in incubators at the seven tempera-
A.J. Hart et al. / Journal of Insect Physiology 48 (2002) 593–599
tures. Between 61 and 89 individuals were reared at each temperature. The arenas were examined daily for egg hatch, after which emerging A. californicus nymphs were monitored for development and fed daily with excess red spider mite eggs and nymphs. The various developmental stages of the mites (larva, protonymph, deutonymph and adult) were identified by observing nymphal moults, which were indicated by the change from six to eight legs from larva to deuteronymph, and the appearance of cast skins on the arena surface for subsequent stages. Simple and weighted linear regression (for methods see Hart et al., 1997, 2002) were used to analyse the data and provide estimates of the developmental threshold and day-degree requirements for this species. Temperature data from a Birmingham meteorological centre were used in combination with the values for development to provide estimates of the number of generations possible in central areas of the UK under outdoor conditions. 2.3. Overwintering and cold hardiness 2.3.1. Measurement of supercooling point Supercooling points (SCP) of A. californicus from Treatments 1 and 3 were assessed by a differential scanning calorimeter (DSC) at the British Antarctic Survey, Cambridge. Groups of 3–5 individuals were placed in small pans, which were sealed and placed inside the DSC. A total of 34 mites from each treatment were cooled to their supercooling point at a rate of 1 °C min1 , with the onset of the freezing exotherm (the SCP) recorded on a print-out at the end of each experiment. 2.3.2. Lower lethal temperature Thirty adult female A. californicus from Treatments 1 and 3 were placed individually into small plastic Beem capsules (approximately 1 cm in length) with a piece of moist filter paper. Ten capsules were contained within each of three test tubes held in a rack in a low temperature programmable alcohol bath. The mites were exposed singly within each replicate test tube because of their small size, thus ensuring that every individual could be recovered at the end of the experiment. Thirty mites were cooled at 1 °C min-1 to a series of seven subzero temperatures ranging between ⫺10 and ⫺22 °C and held at the exposure temperature for 1 min. After warming at 1 °C min-1 to room temperature, the mites were placed individually onto numbered arenas delineated on a black tile with moist filter paper and Oecotak, and placed in containers half filled with water in an incubator maintained at 15 °C, 18L:6D photoperiod. Survival of these mites was assessed 24 h later. The data collected from the lethal temperature experiments were analysed by probit analysis and plotted as the 10, 50 and 90% mortality levels ( ± 95% fiducial limits) against temperature.
595
2.3.3. Lethal time A similar procedure was followed for the lethal time experiments, with adult female A. californicus from Treatments 1 and 3 placed individually in small Beem capsules with a piece of moist filter paper. Between 7 and 22 batches of three Petri dishes each containing 10 capsules with one mite per capsule were held at ⫺5 , 0 and +5 °C in temperature-controlled incubators for varying periods of time (between 1 and 96 days). Three Petri dishes containing 30 mites in total were removed at regular intervals and survival assessed 24 h later as for the lethal temperature experiment. Probit analysis was used to estimate the time taken to kill 10 , 50 and 90% of the population at the three exposure temperatures. 2.3.4. Field exposures Adult female A. californicus reared under potentially diapausing conditions but without acclimation (Treatment 2) were individually contained in Beem capsules with moist filter paper. Ten capsules were placed in a Petri dish and three Petri dishes of 10 capsules formed one batch. Between three and six batches of mites were sealed inside secure plastic containers. These containers were placed in a field location, that was sheltered from wind and precipitation, together with a Tinytalk data logger to record temperature during the course of the experiment. Containers were placed in the field at various times during the winter, and batches of mites were removed from the containers at intervals during field exposure. Mortality was assessed as for the laboratory experiments. A second field trial was set up to run concurrently, placing mites from Treatment 2 in small glass vials (approximately 3 cm high and 1.5 cm diameter) with a ventilated lid, together with agar on the bottom (2 ml), a piece of filter paper and a French bean leaf disc infested with red spider mite added at intervals as prey. Five adult female A. californicus were placed in each vial, and six vials formed one batch of 30 mites. Between seven and eight batches of mites were put into plastic containers at the field site and batches of 30 mites (six vials) were removed at intervals through the winter and examined for survival and reproduction.
3. Results 3.1. Development of A. californicus The development of A. californicus from egg lay to adult is shown in Fig. 1 with temperature plotted against rate of development (1/days to adult). Using simple linear regression the developmental threshold was estimated to be 9.9 °C with a day-degree requirement of 123.5 day-degrees per generation. When weighted linear regression was applied to the data, a lower estimated
596
A.J. Hart et al. / Journal of Insect Physiology 48 (2002) 593–599
Fig. 1. Development of A. californicus from egg to adult at different temperatures, with lines fitted by simple linear regression (solid line) and weighted linear regression (dotted line).
developmental threshold of 8.6 °C was obtained with an increased day-degree requirement of 142.9 day-degrees per generation. The annual number of generations of A. californicus that would be possible at outdoors sites in central England was estimated using the day-degree per generation values obtained from the developmental threshold calculations (weighted linear regression) in combination with mean daily temperatures from 1991 to 2000 supplied by the University of Birmingham. Approximately seven generations per year are possible in the UK, with up to six generations occurring during a ‘summer’ period between 1st April and 30th September. 3.1.1. Induction of diapause Oviposition was observed in 100% of the 50 adult female mites selected from the population reared at 19 °C and 6L:18D (Treatment 2). Therefore, it was assumed that no individuals of this strain had entered diapause under these conditions.
Fig. 2. LTemp10, 50 and 90 ( ± 95%fiducial limits) of adult female A. californicus reared at 26 °C, 18L:6D (Treatment 1) and at 19 °C, 6L:18D followed by a period of acclimation for 7 days at 10 °C, 12L:12D (Treatment 3).
approximately 90% of the sample died above the SCP, indicating extensive pre-freeze mortality. 3.2.3. Lethal time The lethal times at the 10, 50 and 90% mortality levels, after exposure to ⫺5, 0 and +5 °C for increasing periods of time for adult female A. californicus reared under two regimes are shown in Fig. 3. A strong acclimation response was apparent at the 50 and 90% mortality levels at ⫺5 and 0 °C, with 10% of mites from Treatment 3 surviving for 11 days at ⫺5 °C and 26 days at 0 °C, compared with Treatment 1 values of 8 days at ⫺5 °C and 15 days at 0 °C respectively. There was prolonged survival of A. californicus from both Treatments
3.2. Cold tolerance and overwintering 3.2.1. Supercooling point The SCPs of adult female A. californicus reared under Treatment 1 (⫺21.6 ± 0.3°C, range ⫺17.7 to ⫺24.5 °C) were not significantly different from those of mites reared under Treatment 3 (⫺22.2 ± 0.4°C range ⫺17.7 to ⫺25.8 °C). No mites were alive after freezing. 3.2.2. Lower lethal temperature Lethal temperatures at the 10, 50 and 90% mortality levels were plotted against temperature (Fig. 2). There was a significant acclimation response (shown by nonoverlapping fiducial limits) for mites reared under Treatment 3 surviving exposure to lower temperatures compared with Treatment 1. The increase in mortality from 10 to 90% occurred over a narrow range of temperatures (⫺12.8 to ⫺15.1 °C for Treatment 1 and ⫺16.3 to ⫺19.3 °C for Treatment 3). For both treatments, up to
Fig. 3. Lethal time ( ± 95%fiducial limits) of adult female A. californicus reared at 26 °C, 18L:6D (Treatment 1) and at 19 °C, 6L:18D with a period of acclimation for 7 days at 10 °C, 12L:12D (Treatment 3) after exposure to ⫺5, 0 and +5 °C for increasing periods of time.
A.J. Hart et al. / Journal of Insect Physiology 48 (2002) 593–599
597
°C, with a maximum of 14.2 °C and a minimum of 0.8 °C. In the sheltered conditions of the trial, the temperature did not fall below 0 °C. 4. Discussion
Fig. 4. Mortality of adult female A. californicus, after being reared at 19 °C, 6L:18D, and placed outdoors on the dates shown without prey.
1 and 3 after exposure to +5 °C, with a significant acclimation response at the 10, 50 and 90% mortality levels. Approximately 10% of mites from Treatment 3 survived almost 3 months at this temperature. 3.2.4. Field exposure Field mortality in samples without prey is shown in Fig. 4. There was an increase in mortality over time reaching 100% after about 100 days. This trend was observed in four different experiments, placed in the field at various intervals throughout the winter period. In the concurrent field trial, in which A. californicus were provided with prey, approximately 10% of mites were alive after 112 days exposure at the end of February (Fig. 5). During the course of the experiments, eggs were laid by a number of A. californicus adult females, with egg hatch and nymphal development observed. The mean temperature recorded during the field experiment was 7.0
Fig. 5. Mortality of adult female A. californicus, after being reared at 19 °C, 6L:18D, and placed outdoors on the dates shown with provision of prey.
Development of A. californicus was approximately linear within the range of temperatures used in the experiment, with a development threshold of 9.9 °C, 1 °C higher than that estimated by Castagnoli and Simoni (1991) when using simple linear regression. Also, the strain used in this study developed from egg to adult at a slower rate (8.1 days at 25 °C) compared with that studied by Castagnoli and Simoni (5.8 days at 25 °C). The slower rate of development observed in this study may be attributable to strain differences, or minor differences in rearing conditions, although the same prey (T. urticae) was provided in both the cases. Applying weighted linear regression analysis to the data as outlined by Hart et al. (1997) resulted in a lower estimate of the developmental threshold (8.6 °C). The line is weighted towards the temperatures with least variation which in this case occurred at the lower temperatures used (13.5 and 14.5 °C). Linear methods were suitable for these data, as the mites were not reared at extremes of temperature where curvilinearity may become apparent. A. californicus has a short generation time and was estimated to be able to complete up to six generations in the summer in central areas of the UK, and in some years, a full generation during the winter months would be possible. This suggests that the eggs laid by nondiapausing adult female A. californicus at the start of winter might be able to develop through to adulthood in some years, provided that suitable prey were available and temperatures were not severe. A. californicus had a low mean SCP (⫺21.6 °C for adult females reared under non-diapause-inducing conditions) which was comparable to that of other predatory mites; for example, adult females of P. persimilis with a mean SCP of ⫺22.5 °C and A. cucumeris of ⫺20.7 °C (Morewood, 1992). As in the study of Morewood, rearing under diapause-inducing and acclimation regimes had no significant effect on their SCP. Some pre-freeze mortality was evident in A. californicus, with mortality increasing from 10 to 90% between ⫺16.3 and ⫺19.3 °C with mites from Treatment 3 (reared at 19 °C, 6L:18D followed by a period of acclimation for 7 days at 10 °C, 12L:12D), whereas the SCP range of mites from the same treatment was ⫺17.7 to ⫺25.8 °C. This data suggests that A. californicus may possess a higher level of cold tolerance than might have been predicted by knowledge of its native climate. Results from the lethal time experiments confirmed the relatively strong cold hardiness of this species, with individuals from
598
A.J. Hart et al. / Journal of Insect Physiology 48 (2002) 593–599
Treatment 3 surviving for more than 11 days at ⫺5 °C and almost 3 months at +5 °C under laboratory conditions. Non-diapausing P. persimilis has also been shown to exhibit considerable cold hardiness suggesting that this feature may not be confined to A. californicus (Morewood, 1993). Significant acclimation responses were observed in both lethal temperature and lethal time exposures, despite the fact that individuals of this strain had not entered diapause. Non-diapausing A. californicus survived without prey for over 3 months under sheltered winter field conditions. Mites would be likely to overwinter in microhabitat sites protected from frosts, such as under the bark of trees, or in the structural framework of glasshouses. The results from the field experiments, therefore, give a realistic indication of the likely survival times of equivalent natural populations of A. californicus in central England. Provision of prey increased the period of survival by approximately 10%, with a small number of mites still alive at the end of February when the experiment was completed. This suggests that access to prey may be important for the longer-term survival of non-diapausing mites, which continue to feed and develop in winter. T. urticae is known to overwinter in glasshouses and survive soil sterilisation (van Lenteren, 2000), hence, these individuals may form a source of prey for non-diapausing A. californicus which remain in glasshouses, or those which recolonise after previous escapes into the surrounding locale. There was a further important observation made in the field experiments. During the course of the field exposures, eggs were laid in both trials by both starved and fed adult females, and in the case of fed mites, the eggs hatched and active larvae were observed. Thus, because of the short generation time of A. californicus, it may not be essential for non-diapausing individuals that enter the winter to survive until the following spring, because the population can be sustained by their progeny. Development from egg to adult occurs at moderately low temperatures, suggesting that provided they have access to prey, progeny of overwintering nondiapausing females could survive and develop, contributing to a new summer ‘wild’ population of A. californicus in the UK. Of equal importance is the recent discovery in A. californicus of differences between strains in the ability to diapause (Jolly, 2000) with 0% of an American strain entering diapause, 16% of a Spanish strain, and over 95% of a UK ‘wild’ strain. Clearly, selection pressure to diapause is strong in this species, suggesting that careful screening of the diapausing ability of different source populations of future non-native biocontrol agents should be carried out prior to licensing and release. Similar experiments to those described for A. californicus have been carried out using other non-native arthropod species, including. Macrolophus caliginosus Hart et
al. (2002), which has also been recently released in the UK. M. caliginosus is a non-native mirid used for the biocontrol of glasshouse whitefly (Trialeurodes vaporariorum Westwood), and has also been observed outside glasshouses in winter. Studies by Hart et al. (2002) have demonstrated that this species would be able to survive UK winters under certain conditions and, possibly, become established. The data from M. caliginosus and A. californicus indicate that a protocol based on thermal requirements and tolerances for development and survival could be used to assess the establishment potential of a range of non-native biocontrol agents prior to their release in the UK.
Acknowledgements We are grateful to Syngenta Bioline (formerly Novartis BCM) for the supply of Amblyseius californicus and to Barbara Russell for technical help. This work was carried out as part of a project funded by the Department for Environment, Food and Rural Affairs (DEFRA).
References Bale, J.S., Walters, K.F.A., 2001. Overwintering biology as a guide to the establishment potential of non-native arthropods in the UK. In: Atkinson, D., Thorndyke, M. (Eds.), Animal Developmental Ecology. BIOS Scientific Publishers Ltd, Oxford, pp. 343–354. Caltagirone, L.E., 1981. Landmark examples in biological control. Annual Review of Entomology 26, 213–232. Castagnoli, M., Simoni, S., 1991. Influenza della temperatura sull’incremento delle popolazioni di Amblyseius californicus (McGregor) (Acari: Phytoseiidae). Redia LXXIV, 621–640. Denlinger, D.L., 1991. Relationship between cold hardiness and diapause. In: Lee, R.E., Denlinger, D.L. (Eds.), Insects at Low Temperature. Chapman and Hall, New York, pp. 174–198. Fitzgerald, J.D., Solomon, M.G., 1991. Diapause induction and duration in the phytoseiid mite Typhlodromus pyri. Experimental and Applied Acarology 12, 135–145. Griffiths, D.A., 1999. In: Albajes, R., Gullino, M.L., van Lenteren, J.C., Elad, Y. (Eds.), Biological Control of Mites. Kluwer Academic Publishers, Dordrecht, pp. 217–234. Hart, A.J., Bale, J.S., Fenlon, J., 1997. Developmental threshold, daydegree requirements and voltinism of the aphid predator Episyrphus balteatus (Diptera, Syrphidae). Annals of Applied Biology 130, 427–437. Hart, A.J., Tullett, A.G., Bale, J.S., Walters, K.F.A., 2002. Effects of temperature on the establishment potential in the UK of the nonnative glasshouse biocontrol agent Macrolophus caliginosus. Physiological Entomology 27, 112–123. Hussey, N.W., Bravenboer, L., 1971. Control of pests in glasshouse culture by the introduction of natural enemies. In: Huffaker, C.B. (Ed.), Biological Control. Plenum Press, New York, pp. 195–216. Jolly, R.L., 2000. The predatory mite Neoseiulus californicus: its potential as a biological control agent for the fruit tree spider, Panonychus ulmi. The BCPC Conference at Brighton, Pest and Diseases 1, 487–490. van Lenteren, J.C., 1995. In: Dent, D. (Ed.), Integrated Pest Management. Chapman and Hall, London, pp. 311–343.
A.J. Hart et al. / Journal of Insect Physiology 48 (2002) 593–599
van Lenteren, J.C., 2000. A greenhouse without pesticides: fact or fantasy? Crop Protection 19, 375–384. van Lenteren, J.C., Woets, J., 1988. Biological control and integrated pest control in greenhouses. Annual Review of Entomology 33, 239–269. McMurtry, J.A., Croft, B.A., 1997. Life-styles of phytoseiid mites and their roles in biological control. Annual Review of Entomology 42, 291–321. Morewood, W.D., 1992. Cold hardiness of Phytoseiulus persimilis Athias-Henriot and Amblyseius cucumeris (Oudemans) (Acarina: Phytoseiidae). Canadian Entomologist 124, 1015–1025.
599
Morewood, W.D., 1993. Diapause and cold hardiness of phytoseiid mites (Acarina: Phytoseiidae). European Journal of Entomology 90, 3–10. Morewood, W.D., Gilkeson, L.A., 1991. Diapause induction in the thrips predator Amblyseius cucumeris (Acarina: Phytoseiidae) under greenhouse conditions. Entomophaga 36, 253–263. Veerman, A., 1992. Diapause in phytoseiid mites: a review. Experimental and Applied Acarology 14, 1–60.
Effects of temperature on the establishment potential of the predatory mite Amblyseius californicus McGregor (Acari: Phytoseiidae) in the UK A.J. Hart a,∗, J.S. Bale a, A.G. Tullett a, M.R. Worland b, K.F.A. Walters c a
School of Biosciences, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK b British Antarctic Survey, High Cross, Madingley Road, Cambridge CB3 0ET, UK c Central Science Laboratory, Sand Hutton, York YO41 1LZ, UK
Received 28 November 2001; received in revised form 10 March 2002; accepted 26 March 2002
Abstract Amblyseius californicus was introduced into the UK in the early 1990s as a biocontrol agent against glasshouse red spider mite Tetranychus urticae. This study investigated the effects of temperature on the establishment potential of A. californicus in the UK in the light of recent reports of their successful overwintering outside of glasshouse environments. The developmental thresholds were 9.9 and 8.6 °C respectively using simple and weighted linear regression. Using the day-degree requirement per generation calculated by weighted regression (143 day-degrees) in combination with climate data, it was estimated that up to seven generations would be possible annually outdoors in the UK. Non-diapausing adult females froze at ⫺22 °C, with 100% mortality after reaching their freezing temperature. Up to 90% of mites died before freezing after short exposures to low temperatures. Significant acclimation responses occurred; 90% of acclimated individuals survived 26 days exposure at 0 °C and 11 days at ⫺5 °C (acclimated mites were reared at 19 °C, 6L:18D followed by 1 week at 10 °C, 12L:12D). Non-diapausing adult females survived over 3 months outdoors in winter under sheltered conditions and oviposition was observed. The experimental protocol used in this study is discussed as a pre-release screen for the establishment potential of other Amblyseius species, and similar non-native biocontrol agents. 2002 Elsevier Science Ltd. All rights reserved. Keywords: Amblyseius californicus; Tetranychus urticae; Cold tolerance; Temperature; Biological control
1. Introduction Amblyseius californicus McGregor (Acari: Phytoseiidae), also known as Neoseiulus californicus, is a predatory mite originating from California and Florida, used extensively in biological control programs against red spider mite (Tetranychus urticae Koch, Tetranychidae) on a global scale. A. californicus is widely used in the Mediterranean region, particularly Italy and Spain where it is reported to occur naturally (Castagnoli and Simoni, 1991; Griffiths, 1999). Recently, A. californicus has been imported into the UK
Corresponding author. Tel.: +44-121-414-5925; fax: +44-121414-5562. E-mail address: [email protected] (A.J. Hart). ∗
as a biocontrol agent and released in greenhouses in a series of ‘limited release’ trials. However, due to inevitable escapes from these environments, it is now reported to be established outdoors in several areas of southern Britain, and its distribution could potentially expand (Jolly, 2000). Phytoseiid mites have long been used in the control of glasshouse pests including red spider mite and thrips (McMurtry and Croft, 1997), and their use is increasing owing to the pressure on growers to find alternatives to chemical pesticides (van Lenteren, 2000). Phytoseiulus persimilis was the first predatory mite to be used in the control of red spider mite in glasshouses in UK in the late 1960s (Hussey and Bravenboer, 1971). Since then, the use of this species has increased greatly and has become an example of successful biological control based on a thorough knowledge of the plant/pest/predator relationship (Caltagirone, 1981). The
0022-1910/02/$ - see front matter 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 2 - 1 9 1 0 ( 0 2 ) 0 0 0 8 7 - 2
594
A.J. Hart et al. / Journal of Insect Physiology 48 (2002) 593–599
use of other predatory mites has been increasing since the 1980s, including species from the genus Amblyseius (van Lenteren and Woets, 1988; van Lenteren, 1995). To assess the risk of establishment of a biocontrol agent such as A. californicus, various abiotic and biotic factors need to be considered (Bale and Walters, 2001; Hart et al., 2002). The most important abiotic determinant is temperature, which affects survival, reproduction and development (Bale and Walters, 2001). Successful establishment in a new habitat requires an adequate thermal budget (number of day-degrees above the developmental threshold) for development and reproduction, and the ability to survive harsh conditions that may be encountered during the winter season. In addition to the abiotic factors, availability of prey and in some cases, the ability to enter diapause, are also necessary considerations when assessing establishment potential. There is recent evidence that a naturalised strain of A. californicus has become established outdoors in some areas of the UK and possesses the ability to enter diapause in winter (Jolly, 2000). It is not known if commercial strains supplied by biocontrol companies can diapause immediately upon escape from release sites, but it is clear that the ability to diapause can be selected for relatively quickly, as all of the licensed releases of A. californicus in the UK have occurred within the last 10 years. A relationship between diapause and cold hardiness ability has been shown in a number of arthropods, with the diapausing stage having an increased level of cold tolerance compared to non-diapausing counterparts (see Denlinger, 1991). Diapause can be induced in A. californicus and related species by exposing individuals throughout their juvenile development to short day lengths and temperatures below 20 °C (Fitzgerald and Solomon, 1991; Morewood and Gilkeson, 1991; Veerman, 1992; Morewood, 1993). Not all species, strains or populations are capable of entering diapause, and some predatory mites can overwinter successfully without diapausing (Morewood, 1993). Although there are many studies on the biology and diapause of A. californicus and other phytoseiid mite species, few have investigated their cold hardiness (Morewood, 1992, 1993), and, in general, there is insufficient information for a comprehensive risk assessment of the establishment potential of these species in new environments. Overwintering survival is an important factor often overlooked in the licensing system for the release of non-native biocontrol agents in the UK. The aim of this study was to assess the overwintering ability and effects of temperature on a commercially available strain of A. californicus to assess the risk of establishment posed by individuals escaping from glasshouse environments.
2. Materials and methods 2.1. Rearing of A. californicus A strain of A. californicus, originally imported from USA, supplied by Syngenta Bioline (formerly Novartis BCM) was reared at 26 °C with a photoperiod of 18L:6D (Treatment 1) under quarantine conditions at the University of Birmingham. Populations to be used for experimentation were established on artificial arenas: black tiles (approximately 20 × 12cm2) ringed by moist filter paper and Oecotak were placed on sponge blocks in containers half filled with water to act as a barrier to the mites. A. californicus were reared on the black tiles and fed each day with T. urticae. Dwarf French beans (Phaseolus vulgaris L.) were grown and infested with red spider mite and used as a source of prey for the predators. French bean leaves were added daily and removed the following day and any eggs laid by A. californicus were used to set up new cultures for experimentation as and when required. Populations of A. californicus were also reared under potential diapause-inducing conditions in incubators maintained at 19 °C and 6L:18D photoperiod (Treatment 2) (Fitzgerald and Solomon, 1991; Jolly, 2000). Adult female A. californicus were placed on artificial arenas as described previously and oviposited on to French bean leaves. The eggs were allowed to hatch and the nymphs developed into adults under these conditions, with excess red spider mite provided as prey. Fifty gravid adult females reared under these conditions were selected and placed on individual 1 cm2 arenas created by dividing up the tiles with strips of moist filter paper and lines of Oecotak and monitored daily over a 7 day period for oviposition at 19 °C, 6L:18D. Failure to oviposit was used as an indicator of the successful induction of an adult reproductive diapause. To obtain acclimated populations of A. californicus, females that had been reared under diapausing conditions (Treatment 2) were transferred to incubators maintained at 10 °C with a 12L:12D photoperiod for 7 days (Treatment 3). 2.2. Development Incubators were set at a range of temperatures (13.5, 14.5, 19, 20, 22.5, 25.5 and 30.5 °C) with a photoperiod of 18L:6D. Temperatures of the incubators were monitored throughout the experiments using Tinytalk dataloggers. Female A. californicus were maintained at 25 °C on French bean leaves infested with red spider mite and eggs laid by A. californicus were collected over a 24 h period. Individual eggs were placed on 1 cm2 artificial arenas which were created by sub-dividing black tiles with strips of moist filter paper and Oecotak, placed in containers half filled with water (as described previously) and kept in incubators at the seven tempera-
A.J. Hart et al. / Journal of Insect Physiology 48 (2002) 593–599
tures. Between 61 and 89 individuals were reared at each temperature. The arenas were examined daily for egg hatch, after which emerging A. californicus nymphs were monitored for development and fed daily with excess red spider mite eggs and nymphs. The various developmental stages of the mites (larva, protonymph, deutonymph and adult) were identified by observing nymphal moults, which were indicated by the change from six to eight legs from larva to deuteronymph, and the appearance of cast skins on the arena surface for subsequent stages. Simple and weighted linear regression (for methods see Hart et al., 1997, 2002) were used to analyse the data and provide estimates of the developmental threshold and day-degree requirements for this species. Temperature data from a Birmingham meteorological centre were used in combination with the values for development to provide estimates of the number of generations possible in central areas of the UK under outdoor conditions. 2.3. Overwintering and cold hardiness 2.3.1. Measurement of supercooling point Supercooling points (SCP) of A. californicus from Treatments 1 and 3 were assessed by a differential scanning calorimeter (DSC) at the British Antarctic Survey, Cambridge. Groups of 3–5 individuals were placed in small pans, which were sealed and placed inside the DSC. A total of 34 mites from each treatment were cooled to their supercooling point at a rate of 1 °C min1 , with the onset of the freezing exotherm (the SCP) recorded on a print-out at the end of each experiment. 2.3.2. Lower lethal temperature Thirty adult female A. californicus from Treatments 1 and 3 were placed individually into small plastic Beem capsules (approximately 1 cm in length) with a piece of moist filter paper. Ten capsules were contained within each of three test tubes held in a rack in a low temperature programmable alcohol bath. The mites were exposed singly within each replicate test tube because of their small size, thus ensuring that every individual could be recovered at the end of the experiment. Thirty mites were cooled at 1 °C min-1 to a series of seven subzero temperatures ranging between ⫺10 and ⫺22 °C and held at the exposure temperature for 1 min. After warming at 1 °C min-1 to room temperature, the mites were placed individually onto numbered arenas delineated on a black tile with moist filter paper and Oecotak, and placed in containers half filled with water in an incubator maintained at 15 °C, 18L:6D photoperiod. Survival of these mites was assessed 24 h later. The data collected from the lethal temperature experiments were analysed by probit analysis and plotted as the 10, 50 and 90% mortality levels ( ± 95% fiducial limits) against temperature.
595
2.3.3. Lethal time A similar procedure was followed for the lethal time experiments, with adult female A. californicus from Treatments 1 and 3 placed individually in small Beem capsules with a piece of moist filter paper. Between 7 and 22 batches of three Petri dishes each containing 10 capsules with one mite per capsule were held at ⫺5 , 0 and +5 °C in temperature-controlled incubators for varying periods of time (between 1 and 96 days). Three Petri dishes containing 30 mites in total were removed at regular intervals and survival assessed 24 h later as for the lethal temperature experiment. Probit analysis was used to estimate the time taken to kill 10 , 50 and 90% of the population at the three exposure temperatures. 2.3.4. Field exposures Adult female A. californicus reared under potentially diapausing conditions but without acclimation (Treatment 2) were individually contained in Beem capsules with moist filter paper. Ten capsules were placed in a Petri dish and three Petri dishes of 10 capsules formed one batch. Between three and six batches of mites were sealed inside secure plastic containers. These containers were placed in a field location, that was sheltered from wind and precipitation, together with a Tinytalk data logger to record temperature during the course of the experiment. Containers were placed in the field at various times during the winter, and batches of mites were removed from the containers at intervals during field exposure. Mortality was assessed as for the laboratory experiments. A second field trial was set up to run concurrently, placing mites from Treatment 2 in small glass vials (approximately 3 cm high and 1.5 cm diameter) with a ventilated lid, together with agar on the bottom (2 ml), a piece of filter paper and a French bean leaf disc infested with red spider mite added at intervals as prey. Five adult female A. californicus were placed in each vial, and six vials formed one batch of 30 mites. Between seven and eight batches of mites were put into plastic containers at the field site and batches of 30 mites (six vials) were removed at intervals through the winter and examined for survival and reproduction.
3. Results 3.1. Development of A. californicus The development of A. californicus from egg lay to adult is shown in Fig. 1 with temperature plotted against rate of development (1/days to adult). Using simple linear regression the developmental threshold was estimated to be 9.9 °C with a day-degree requirement of 123.5 day-degrees per generation. When weighted linear regression was applied to the data, a lower estimated
596
A.J. Hart et al. / Journal of Insect Physiology 48 (2002) 593–599
Fig. 1. Development of A. californicus from egg to adult at different temperatures, with lines fitted by simple linear regression (solid line) and weighted linear regression (dotted line).
developmental threshold of 8.6 °C was obtained with an increased day-degree requirement of 142.9 day-degrees per generation. The annual number of generations of A. californicus that would be possible at outdoors sites in central England was estimated using the day-degree per generation values obtained from the developmental threshold calculations (weighted linear regression) in combination with mean daily temperatures from 1991 to 2000 supplied by the University of Birmingham. Approximately seven generations per year are possible in the UK, with up to six generations occurring during a ‘summer’ period between 1st April and 30th September. 3.1.1. Induction of diapause Oviposition was observed in 100% of the 50 adult female mites selected from the population reared at 19 °C and 6L:18D (Treatment 2). Therefore, it was assumed that no individuals of this strain had entered diapause under these conditions.
Fig. 2. LTemp10, 50 and 90 ( ± 95%fiducial limits) of adult female A. californicus reared at 26 °C, 18L:6D (Treatment 1) and at 19 °C, 6L:18D followed by a period of acclimation for 7 days at 10 °C, 12L:12D (Treatment 3).
approximately 90% of the sample died above the SCP, indicating extensive pre-freeze mortality. 3.2.3. Lethal time The lethal times at the 10, 50 and 90% mortality levels, after exposure to ⫺5, 0 and +5 °C for increasing periods of time for adult female A. californicus reared under two regimes are shown in Fig. 3. A strong acclimation response was apparent at the 50 and 90% mortality levels at ⫺5 and 0 °C, with 10% of mites from Treatment 3 surviving for 11 days at ⫺5 °C and 26 days at 0 °C, compared with Treatment 1 values of 8 days at ⫺5 °C and 15 days at 0 °C respectively. There was prolonged survival of A. californicus from both Treatments
3.2. Cold tolerance and overwintering 3.2.1. Supercooling point The SCPs of adult female A. californicus reared under Treatment 1 (⫺21.6 ± 0.3°C, range ⫺17.7 to ⫺24.5 °C) were not significantly different from those of mites reared under Treatment 3 (⫺22.2 ± 0.4°C range ⫺17.7 to ⫺25.8 °C). No mites were alive after freezing. 3.2.2. Lower lethal temperature Lethal temperatures at the 10, 50 and 90% mortality levels were plotted against temperature (Fig. 2). There was a significant acclimation response (shown by nonoverlapping fiducial limits) for mites reared under Treatment 3 surviving exposure to lower temperatures compared with Treatment 1. The increase in mortality from 10 to 90% occurred over a narrow range of temperatures (⫺12.8 to ⫺15.1 °C for Treatment 1 and ⫺16.3 to ⫺19.3 °C for Treatment 3). For both treatments, up to
Fig. 3. Lethal time ( ± 95%fiducial limits) of adult female A. californicus reared at 26 °C, 18L:6D (Treatment 1) and at 19 °C, 6L:18D with a period of acclimation for 7 days at 10 °C, 12L:12D (Treatment 3) after exposure to ⫺5, 0 and +5 °C for increasing periods of time.
A.J. Hart et al. / Journal of Insect Physiology 48 (2002) 593–599
597
°C, with a maximum of 14.2 °C and a minimum of 0.8 °C. In the sheltered conditions of the trial, the temperature did not fall below 0 °C. 4. Discussion
Fig. 4. Mortality of adult female A. californicus, after being reared at 19 °C, 6L:18D, and placed outdoors on the dates shown without prey.
1 and 3 after exposure to +5 °C, with a significant acclimation response at the 10, 50 and 90% mortality levels. Approximately 10% of mites from Treatment 3 survived almost 3 months at this temperature. 3.2.4. Field exposure Field mortality in samples without prey is shown in Fig. 4. There was an increase in mortality over time reaching 100% after about 100 days. This trend was observed in four different experiments, placed in the field at various intervals throughout the winter period. In the concurrent field trial, in which A. californicus were provided with prey, approximately 10% of mites were alive after 112 days exposure at the end of February (Fig. 5). During the course of the experiments, eggs were laid by a number of A. californicus adult females, with egg hatch and nymphal development observed. The mean temperature recorded during the field experiment was 7.0
Fig. 5. Mortality of adult female A. californicus, after being reared at 19 °C, 6L:18D, and placed outdoors on the dates shown with provision of prey.
Development of A. californicus was approximately linear within the range of temperatures used in the experiment, with a development threshold of 9.9 °C, 1 °C higher than that estimated by Castagnoli and Simoni (1991) when using simple linear regression. Also, the strain used in this study developed from egg to adult at a slower rate (8.1 days at 25 °C) compared with that studied by Castagnoli and Simoni (5.8 days at 25 °C). The slower rate of development observed in this study may be attributable to strain differences, or minor differences in rearing conditions, although the same prey (T. urticae) was provided in both the cases. Applying weighted linear regression analysis to the data as outlined by Hart et al. (1997) resulted in a lower estimate of the developmental threshold (8.6 °C). The line is weighted towards the temperatures with least variation which in this case occurred at the lower temperatures used (13.5 and 14.5 °C). Linear methods were suitable for these data, as the mites were not reared at extremes of temperature where curvilinearity may become apparent. A. californicus has a short generation time and was estimated to be able to complete up to six generations in the summer in central areas of the UK, and in some years, a full generation during the winter months would be possible. This suggests that the eggs laid by nondiapausing adult female A. californicus at the start of winter might be able to develop through to adulthood in some years, provided that suitable prey were available and temperatures were not severe. A. californicus had a low mean SCP (⫺21.6 °C for adult females reared under non-diapause-inducing conditions) which was comparable to that of other predatory mites; for example, adult females of P. persimilis with a mean SCP of ⫺22.5 °C and A. cucumeris of ⫺20.7 °C (Morewood, 1992). As in the study of Morewood, rearing under diapause-inducing and acclimation regimes had no significant effect on their SCP. Some pre-freeze mortality was evident in A. californicus, with mortality increasing from 10 to 90% between ⫺16.3 and ⫺19.3 °C with mites from Treatment 3 (reared at 19 °C, 6L:18D followed by a period of acclimation for 7 days at 10 °C, 12L:12D), whereas the SCP range of mites from the same treatment was ⫺17.7 to ⫺25.8 °C. This data suggests that A. californicus may possess a higher level of cold tolerance than might have been predicted by knowledge of its native climate. Results from the lethal time experiments confirmed the relatively strong cold hardiness of this species, with individuals from
598
A.J. Hart et al. / Journal of Insect Physiology 48 (2002) 593–599
Treatment 3 surviving for more than 11 days at ⫺5 °C and almost 3 months at +5 °C under laboratory conditions. Non-diapausing P. persimilis has also been shown to exhibit considerable cold hardiness suggesting that this feature may not be confined to A. californicus (Morewood, 1993). Significant acclimation responses were observed in both lethal temperature and lethal time exposures, despite the fact that individuals of this strain had not entered diapause. Non-diapausing A. californicus survived without prey for over 3 months under sheltered winter field conditions. Mites would be likely to overwinter in microhabitat sites protected from frosts, such as under the bark of trees, or in the structural framework of glasshouses. The results from the field experiments, therefore, give a realistic indication of the likely survival times of equivalent natural populations of A. californicus in central England. Provision of prey increased the period of survival by approximately 10%, with a small number of mites still alive at the end of February when the experiment was completed. This suggests that access to prey may be important for the longer-term survival of non-diapausing mites, which continue to feed and develop in winter. T. urticae is known to overwinter in glasshouses and survive soil sterilisation (van Lenteren, 2000), hence, these individuals may form a source of prey for non-diapausing A. californicus which remain in glasshouses, or those which recolonise after previous escapes into the surrounding locale. There was a further important observation made in the field experiments. During the course of the field exposures, eggs were laid in both trials by both starved and fed adult females, and in the case of fed mites, the eggs hatched and active larvae were observed. Thus, because of the short generation time of A. californicus, it may not be essential for non-diapausing individuals that enter the winter to survive until the following spring, because the population can be sustained by their progeny. Development from egg to adult occurs at moderately low temperatures, suggesting that provided they have access to prey, progeny of overwintering nondiapausing females could survive and develop, contributing to a new summer ‘wild’ population of A. californicus in the UK. Of equal importance is the recent discovery in A. californicus of differences between strains in the ability to diapause (Jolly, 2000) with 0% of an American strain entering diapause, 16% of a Spanish strain, and over 95% of a UK ‘wild’ strain. Clearly, selection pressure to diapause is strong in this species, suggesting that careful screening of the diapausing ability of different source populations of future non-native biocontrol agents should be carried out prior to licensing and release. Similar experiments to those described for A. californicus have been carried out using other non-native arthropod species, including. Macrolophus caliginosus Hart et
al. (2002), which has also been recently released in the UK. M. caliginosus is a non-native mirid used for the biocontrol of glasshouse whitefly (Trialeurodes vaporariorum Westwood), and has also been observed outside glasshouses in winter. Studies by Hart et al. (2002) have demonstrated that this species would be able to survive UK winters under certain conditions and, possibly, become established. The data from M. caliginosus and A. californicus indicate that a protocol based on thermal requirements and tolerances for development and survival could be used to assess the establishment potential of a range of non-native biocontrol agents prior to their release in the UK.
Acknowledgements We are grateful to Syngenta Bioline (formerly Novartis BCM) for the supply of Amblyseius californicus and to Barbara Russell for technical help. This work was carried out as part of a project funded by the Department for Environment, Food and Rural Affairs (DEFRA).
References Bale, J.S., Walters, K.F.A., 2001. Overwintering biology as a guide to the establishment potential of non-native arthropods in the UK. In: Atkinson, D., Thorndyke, M. (Eds.), Animal Developmental Ecology. BIOS Scientific Publishers Ltd, Oxford, pp. 343–354. Caltagirone, L.E., 1981. Landmark examples in biological control. Annual Review of Entomology 26, 213–232. Castagnoli, M., Simoni, S., 1991. Influenza della temperatura sull’incremento delle popolazioni di Amblyseius californicus (McGregor) (Acari: Phytoseiidae). Redia LXXIV, 621–640. Denlinger, D.L., 1991. Relationship between cold hardiness and diapause. In: Lee, R.E., Denlinger, D.L. (Eds.), Insects at Low Temperature. Chapman and Hall, New York, pp. 174–198. Fitzgerald, J.D., Solomon, M.G., 1991. Diapause induction and duration in the phytoseiid mite Typhlodromus pyri. Experimental and Applied Acarology 12, 135–145. Griffiths, D.A., 1999. In: Albajes, R., Gullino, M.L., van Lenteren, J.C., Elad, Y. (Eds.), Biological Control of Mites. Kluwer Academic Publishers, Dordrecht, pp. 217–234. Hart, A.J., Bale, J.S., Fenlon, J., 1997. Developmental threshold, daydegree requirements and voltinism of the aphid predator Episyrphus balteatus (Diptera, Syrphidae). Annals of Applied Biology 130, 427–437. Hart, A.J., Tullett, A.G., Bale, J.S., Walters, K.F.A., 2002. Effects of temperature on the establishment potential in the UK of the nonnative glasshouse biocontrol agent Macrolophus caliginosus. Physiological Entomology 27, 112–123. Hussey, N.W., Bravenboer, L., 1971. Control of pests in glasshouse culture by the introduction of natural enemies. In: Huffaker, C.B. (Ed.), Biological Control. Plenum Press, New York, pp. 195–216. Jolly, R.L., 2000. The predatory mite Neoseiulus californicus: its potential as a biological control agent for the fruit tree spider, Panonychus ulmi. The BCPC Conference at Brighton, Pest and Diseases 1, 487–490. van Lenteren, J.C., 1995. In: Dent, D. (Ed.), Integrated Pest Management. Chapman and Hall, London, pp. 311–343.
A.J. Hart et al. / Journal of Insect Physiology 48 (2002) 593–599
van Lenteren, J.C., 2000. A greenhouse without pesticides: fact or fantasy? Crop Protection 19, 375–384. van Lenteren, J.C., Woets, J., 1988. Biological control and integrated pest control in greenhouses. Annual Review of Entomology 33, 239–269. McMurtry, J.A., Croft, B.A., 1997. Life-styles of phytoseiid mites and their roles in biological control. Annual Review of Entomology 42, 291–321. Morewood, W.D., 1992. Cold hardiness of Phytoseiulus persimilis Athias-Henriot and Amblyseius cucumeris (Oudemans) (Acarina: Phytoseiidae). Canadian Entomologist 124, 1015–1025.
599
Morewood, W.D., 1993. Diapause and cold hardiness of phytoseiid mites (Acarina: Phytoseiidae). European Journal of Entomology 90, 3–10. Morewood, W.D., Gilkeson, L.A., 1991. Diapause induction in the thrips predator Amblyseius cucumeris (Acarina: Phytoseiidae) under greenhouse conditions. Entomophaga 36, 253–263. Veerman, A., 1992. Diapause in phytoseiid mites: a review. Experimental and Applied Acarology 14, 1–60.