The occurrence of entomopathogenic fungi in apple orchards and their biocontrol potential against Eriosoma lanigerum

)


INTRODUCTION
Apple, Malus domestica (Borkhausen), production is one the most economically important horticultural crop practices in temperate regions across the world (Harris et al. 2002;Stoeckli et al. 2008). However, apple production is subjected to attack by several insect pests that negatively affect productivity, resulting in significant crop damage (Beers et al. 2003). Eriosoma lanigerum (Hausmann) (Hemiptera: Aphididae) or woolly apple aphid (WAA) is a major crop pest that attacks apple trees globally, resulting in significant economic losses to the apple-producing industry (Blackman and Eastop 2000;Abu-Romman and Ateyyat 2014).
WAA is of serious concern to the main apple production areas of South Africa, as it has become widespread within apple orchards, with it being considered as one of the main insect pests that attack apple trees (Damavandian 2000;Heunis & Pringle 2003). Infestation of apple trees by root-based colonies of WAA induces formation of hypertrophic galls on the root system, and the presence of galls impedes the flow of plant nutrients and water throughout the plant (Brown et al. 1995). The root colonies of WAA are the main source of arboreal infestation, as the first instar nymphs, which are also known as crawlers, continuously migrate from the roots to the arboreal parts of the plant (Pringle et al. 1994;Mols and Boers 2001;Damavandian & Pringle 2007;Lordan et al. 2015). Crawlers are produced by adult female aphids which reproduce parthenogenetically and overwinter on the roots (Thwaite and Bower 1983;Pringle et al. 1994).
Arboreal colonies of WAA form densely packed colonies, covered by a white, waxy filamentous secretion which protects the colonies from unfavourable environmental conditions, attacks by parasites and predators, and from chemical insecticides (Smith 1999; Moss et al. 2006). Arboreal colonies mostly infest and develop on vulnerable parts of the apple trees such as on pruning cuts and spits that have resulted from heavy cropping (Nicholas et al. 2005). In cases of severe infestation, the presence of the WAA can result in yearly reduction of the quality and quantity of the fruit that is produced. This is due to the destruction of buds, shoots and fruit-bearing wood, which eventually affects production capacity (Brown and Schmitt 1990;Brown et al. 1995).
Management of WAA in South African apple orchards, and across the world, generally relies on the combined use of biological and chemical control with other alternative methods like the use of resistant apple cultivars (Pringle 1998;Timm 2003;Stoeckli et al. 2008;Gontijo 2011;Gresham 2013). For biological control, during the 20th century, the main parasitoid of WAA, Aphelinus mali (Heldemann) (Hymenoptera: Aphelinidae) was introduced from North America into apple orchards globally. In the South African apple-growing areas, the parasitoid was introduced in 1920 (Cross et al. 1999b;Short 2003;Dedryver et al. 2010). However, A. mali parasitises and controls mainly the arboreal colonies of WAA, and not the root colonies, which contributes to the current failure of the parasitoid to control the WAA (Heunis and Pringle 2003). Hence, the parasitoid alone was found to be ineffective at controlling and preventing outbreaks of WAA (Heunis and Pringle 2006). https://doi.org/10.17159/2254-8854/2023/a13728 The most common method of controlling WAA infestation in apple orchards has, for the past several decades, been the application of high doses of chemical insecticides (Vega et al. 2009;Bhatia et al. 2011). Chemical pesticides including vamidothion and other organophosphates, have been used since the 1960s to control WAA in apple orchards (Damavandian 2000;Nicholas 2000;Damavandian and Pringle 2002;Christians 2003). However, many apple growers globally have reported that vamidothion failed to provide effective control of WAA in their orchards, as different biotypes of WAA have been shown to have developed some level of resistance to the insecticide (Pringle et al. 1994;Nicholas 2000;Timm 2003). Pringle et al. (1994) also found a vamidothion-resistant biotype of WAA in South Africa, across two apple-producing areas in the Elgin and Grabouw area. The aphid has also developed some level of resistance against a wide array of chemical insecticides that were used in the past for its control (Christians 2003).
The development of resistance against chemical insecticides by pests such as WAA, as well as concerns regarding the deleterious effects of chemicals on the environment, have provided a strong drive for the development of microbial agents to be used in controlling insect pests (Inglis et al. 2001). Entomopathogenic fungi (EPF) have the potential to be used for biological control of sap-sucking insects such as the WAA, which cannot easily be controlled using chemical pesticides and other biological control means (Chandler et al. 1997;Wraight et al. 1998;Cross et al. 1999a;Pu et al. 2005). The potential of EPF for biological control was first noticed during the 19th century by Metchnikoff, in 1879, and by Krassilstschik, in 1888, when they mass-produced Metarhizium anisopliae (Metchn.) Sorokin, and tested the fungus for control of the wheat cockchafers and the sugar beet curculionid (Gillespie and Moorhouse 1989).
The fungal division Ascomycota contains the Hypocreales fungal species including the PARB clade of the M. anisopliae complex and the Beauveria species, of which some are commercially produced and used globally for biological control against a variety of agricultural pests in agroecosystems (De Faria and Wraight 2007;Quesada-Moraga et al. 2007;Hatting et al. 2019). Fungal species belonging to the M. anisopliae species complex, including the species Metarhizium brunneum Petch, Metarhizium robertsii (Metchnikoff) Sorokin, Metarhizium pinghaense Chen & Guo, and M. anisopliae are cosmopolitan soil-dwelling entomopathogens that have been well studied in terms of biological control (Rehner and Kepler 2017). Entomopathogenic fungi cause mycosis in many different taxa of arthropods and in almost every order of insects, with them having the ability to infect all the life stages of insects (Dedryver et al. 2010;Shahid et al. 2012). Stokwe (2016) explored the efficacy of entomopathogenic nematodes (EPN) and fungi as biological control agents of WAA, under South African conditions. Results from the study indicate that EPN cannot be used as a biocontrol agent against WAA, as the associated bacteria do not grow in the haemocoel of WAA (Stokwe andMalan 2016, 2017), however the use of commercial fungal isolates, derived from Beauveria bassiana (Bals.-Criv) Vuill. (Hypocreales: Cordycipitaceae) and Metarhizium anisopliae (Metchnikoff) Sorokin, showed the potential for use against this insect species, under laboratory conditions. No previous research has been conducted in testing the potential of South African fungi isolated from apple orchards, for the control of WAA.
The aim of this study was to survey for EPF in the local apple orchards of the Western Cape province, and to assess their potential as biological control agents of the WAA, under laboratory conditions. This was accomplished by the collection of soil samples from apple orchards and baiting for EPF using susceptible insects. All isolated fungi were identified to species level, using both morphological and molecular techniques. The susceptibility of female WAA against identified EPF isolates was screened, and the best candidates used in a dose-and timeresponse bioassay.

Collection of soil samples
A total of 48 soil samples, with each sample comprising of five subsamples, were collected across six apple farms in the Grabouw, Elgin and Vyeboom areas located in the Western Cape province, South Africa. At each farm, two apple orchards were selected and from each orchard four soil samples were collected. Each apple orchard was divided into four quadrants and from each quadrant five apple trees were randomly selected, and soil subsamples collected from a depth of 15 cm under each tree canopy (Inglis et al. 2012). The five soil subsamples collected from each quadrant were pooled in clear labelled plastic bags and transported to the laboratory in a cooler box.

Baiting procedure
Each combined soil sample was mixed and sieved through a mesh sieve (4 mm) to remove the rock and leaf material and transferred to 1-l containers. For aeration, the 1-l containers were filled with soil samples leaving 10 cm space on top. All soil samples were baited for EPF after collection, and in the case of the soil samples that were dry, distilled water was used to moisten the soil to maintain enough humidity during the baiting process (Meyling and Eilenberg 2007;Goble et al. 2010). Ten larvae of both Galleria mellonella L. (Lepidoptera: Pyralidae) and Tenebrio molitor (Coleoptera: Tenebrionidae) were added to the surface of the soil samples in the plastic containers which were then closed and incubated in the dark at a controlled room temperature of ± 25 °C (Zimmermann 1986;Meyling 2007). Larvae of both G. mellonella and T. molitor used as EPF bait insects were reared in the laboratory following methods by Van Zyl and Malan (2015).
To ensure that the insects made good contact with the soil, the containers were inverted daily (Goble et al. 2010). After 7 and 14 days the dead larvae were removed. The dead larvae were surface sterilised by first being dipped in distilled water to remove the soil and then dipped first in 70% ethanol and then in distilled water. To isolate the EPF, each dead larva with symptoms of fungal infection, characterised by fungal growth or by hardening of the cadaver was placed on a Petri dish fitted with a filter paper that was moistened with distilled water (Goble et al. 2010). The Petri dishes were then placed in 2-l plastic containers with paper towels moistened with distilled water. The plastic containers were closed and incubated in a controlled room temperature of ± 25 °C, in the dark.
Every four days, overt mycosed insects were removed and the fungi were scraped from the surface of the larva cadaver and placed on Sabouraud dextrose agar plates (9 mm diam.) with 1 g of yeast (SDAY) plates, supplemented with 200 µl of penicillinstreptomycin, to prevent bacterial contamination. The fungal cultures on the agar plates were then incubated at a controlled temperature of ± 25 °C in a growth chamber and checked for fungal growth every four days. The fast-growing saprophytic fungi were discarded, and the contaminated plates were re-plated. Fungal plates were grouped according to morphological characteristics of growth patterns and spore structure.

Confirming insect susceptibility to fungi
To determine the pathogenicity of the isolated fungi, one fungal agar plate was selected from each of the morphologically similar groups of fungal culture plates. The conidia, 2-3 week-old, were scraped from the growth medium plates using a sterile scalpel. The collected fungal spores were transferred into sterile 1.5-ml micro-centrifuge tubes, containing 1.0 ml sterilised distilled water and a single drop of 0.05% v/v Tween 20. The microcentrifuge tubes were closed and vortex-mixed for 60 s to produce a homogenous conidial suspension (of T. molitor were dipped in each conidial (Goble et al. 2010). Five final instars of G. mellonella and five final instars solution for a period of 5 s. This was used to confirm insect susceptibility to the collected fungal isolates against insects in general. The dipped larvae were then transferred to Petri dishes (90 mm diam.), each fitted with a piece of filter paper moistened with distilled water. The Petri dishes were placed in 2 l plastic containers that were then closed with a lid and incubated at a controlled room temperature of ± 25 °C in the dark. After two to three days, the Petri dishes were checked for dead larvae, which were removed from the Petri dishes and surface sterilised. The cadavers were surface sterilised by first dipping in 70% ethanol, thereafter in distilled water, followed by being dipped in 5% bleach and three times in distilled water to conclude the process (Lacey and Solter 2012). Each cadaver was placed on a paper towel to dry, after which it was positioned on a SDAY plate which was sealed with Parafilm and incubated at room temperature for a period of three to four days to verify the death of the larvae by overt mycosis.

Identification of isolated EPF
DNA was extracted from the 12 selected morphologically different EPF isolates using the Zymo Research Quick-DNA fungal/bacterial miniprep kit (Zymo Research Corporation), according to the supplied manufacturer protocol. The polymerase chain reaction (PCR) process was undertaken for the molecular identification process for the selected EPF strains, using the KAPA2G ReadyMix PCR Kit (Labotec). The DNA extract of each of the selected EPF from the DNA extraction process was used for the process. Internal transcribed spacer (ITS) primers, ITS 1 [forward primer (5'-TCCGTAGGTGAACCTGCGG-3')] and ITS 4 [reverse primer (5'-CTCCTCCGCTTATTGATATGC-3')], and the elongation factor-1alpha gene (EF-1α) sequencing primers, as well as EF1F [forward primer (5'-GTCGGTGGTATCGACAAGCGT-3')] and EF2R [reverse primer (5'-AGCATGTTGTCGCCGTTGAAG-3')], were used for the PCR procedure. The PCR thermocycle conditions consisted of the first denaturing step being undertaken at 94 °C for 10 min, followed by 36 cycles of 94 °C for 30 s, at 56 °C for 30 s (for the ITS primers), or at 53 °C for 30 s (for the EF primers) and then at 72 °C for 1 min. The PCR reaction was completed with a final extension at 72 °C for 7 min, after which cooling off took place, with the substance being held at 4 °C. The thermocycle conditions were adapted from Abaajeh and Nchu (2015).
The PCR products were visualised on an agarose gel using ethidium bromide. The un-purified PCR products were sent to the Central Analytical Facility (CAF), DNA Sequencing Unit, Department of Genetics at Stellenbosch University, for post-PCR clean-up and sequencing reaction. DNA sequences were aligned and edited using the software program, CLC Main Workbench, version 6. Sequences were blasted against the NCBI (National Centre for Biotechnology Information) GenBank database for species identification.

Preparation of conidial suspensions
Selected EPF isolates were screened for their pathogenicity against the WAA, by harvesting conidia from 2-3-week-old surface cultures, grown on Saboraud Dextrose Agar (SDA) plates, by scraping with a sterile blade. Collected fungal spores were suspended in 20 ml of sterile distilled water supplemented with 0.05% v/v Tween 20, in McCartney (28 ml) glass bottles. The bottles containing conidia were sealed, and vortex-mixed for 2 min to produce a homogenous suspension. The conidial suspension was poured into a sterile 100-ml glass beaker, through an organza fabric, to remove the fungal hyphae and mycelium, before being poured back into the McCartney bottle, vortex-mixed for 2 min, and used as the conidial stock (Mathulwe et al. 2022). One ml of the conidial stock was transferred into a McCartney (28 ml) glass bottles containing 9 ml of sterile distilled water, and vortex mixed for 3-4 min to produce a homogenous suspension. To determine the conidial concentration of the suspensions, 100 µl of the suspension was pipetted on both ends of a haemocytometer under a coverslip and the total number of spores was counted using a Zeiss Axiolab 5 light microscope. An average number of spores was determined, after counting the total number of spores on both ends of the haemocytometer, and to quantify the conidial concentration per unit volume, methods by Inglis et al. (2012) were followed. Serial dilutions were conducted to obtain the desired conidial concentrations.
The conidial viability of the produced diluted suspensions for each EPF isolate was checked by spread plating 100 µl of the suspension on three SDA plates, at a conidial concentration of 10 7 conidia/ml (Ekesi et al. 2002). The percentage germination of the conidia on the plates was checked 24 h post incubation in the dark at a controlled room temperature of ±25 °C, by 100 spore count from each plate. The conidia were counted as viable if they developed a germ tube, and those without a germ tube were counted as non-viable. The average number of viable conidia were calculated from the three plates, and only conidial suspensions with percentage viability of >85% were used in the screening bioassay (Mathulwe et al. 2022).

Screening of entomopathogenic fungi
Collected EPF isolates were screened for their pathogenicity against the WAA. All isolates were tested at a standard infective conidial concentration of 1 × 10 7 conidia/ml, and insects in the control treatment were treated with sterile distilled water supplemented with 0.05% v/v Tween 20. A total of 60 adult female WAA (12 insects in five 24-well bioassay plates) per treatment were used to test the pathogenicity of each EPF isolate. Each insect was placed in an alternate well of a 24-well bioassay plate, which was fitted with a filter disc paper, and inoculated using 50 µl of the produced conidial suspension. To prevent settling of conidia at the bottom of the inoculum, the conidial suspensions were regularly vortex mixed for 20-30 s during the inoculation process (Mathulwe et al. 2022). To prevent escape of insects from the wells, each plate was closed with a glass cover. The bioassay plates were transferred to 2-l plastic containers fitted with moistened paper towels and incubated at controlled temperature of ± 25 °C. Insect mortality was recorded five days post treatment. To determine the cause of insect mortality, insect cadavers were surface sterilised using 70% ethanol and sterile distilled water. Insect cadavers were placed on water agar and incubated at a temperature of ± 25 °C. Four to five days following incubation, overt mycosis was recorded. The bioassay trials were repeated for each EPF isolate using fresh conidia-inoculum, on a different date.

Dose-response bioassays
The two most effective fungal isolates, M. robertsii (6EIKEN) and M. pinghaense (5HEID), based on the screening bioassays, were used. Conidial suspensions were prepared following the same procedure detailed above. The conidial viability of suspensions was conducted following the same process detailed above. Four conidial concentrations for both M. robertsii (6EIKEN) and M. pinghaense (5HEID), and a control treatment were tested (0, 10 4 , 10 5 , 10 6 and 10 7 conidia/ml). The same number of insects (n = 60) were used, following the same experimental methodology as detailed above for each concentration used. The mortality of WAA and the cause of mortality was assessed five days post-treatment for each conidial concentration following the same procedure outlined above, to determine the lethal concentration required to kill 50% and 90% of the colony, LC 50 and LC 90 . The experiment was repeated for each EPF isolate and conidial concentration using fresh inoculum, on a different date.

Exposure-time-response bioassays
The two effective isolates, M. robertsii (6EIKEN) and M. pinghaense (5HEID), were used for the exposure -time-response bioassays. Conidial suspensions were prepared following the procedure described above. Insects were exposed to the EPF isolates at a concentration of 10 7 conidia/ml, over five different periods (1, 2, 3, 4, and 5 days). Insect mortality was recorded after each time period, from each bioassay plate. The same inoculation and experimental procedure outlined above was followed, to calculate the lethal time required to kill 50% and 90% of the colony at a concentration of 10 7 conidia/ml over time, LT 50 and LT 90 values.

Statistical analysis
Analysis of the obtained data was done using the statistical software, STATISTICA Version 13.5.0.17 (TIBCO Software Inc. 2018). Data was analysed using ANOVA, and the normality of the residuals was checked using normal probability plots. To check the assumption that the variances were homogeneous, the Levene's test was used. LSD (least significant difference) tests and post-hoc tests at 95% confidence intervals were conducted. For the exposure-time-response bioassays, a Logit analysis was conducted to calculate the LT 50 and LT 90 values. A Probit analysis was conducted, using the statistical software NCSS 2020 version 20.0.2 (Hintze 2007) and IBM SPSS statistics data editor, version 28, to determine the lethal concentration dose.

Molecular identification of EPF
The information related to the various fungal isolates that were collected across the six farms in the Grabouw, Elgin and Vyeboom areas is indicated (Table 1) (Figure 1). Molecular identification of the isolated EPF using ITS region indicated that the 13 selected morphologically different indigenous isolates belonged to four different EPF species. The EPF species included an isolate of Cordyceps fumosorosea (Wize) Kelper, Shrestha and Spatafora, an isolate of Purpureocillium lilacinum (Thom) Luangsa-ard, Hou-braken, Hywel-Jones and Samson, two isolates of B. bassiana and nine isolates belonging to the M. anisopliae species-complex (Table 1). The EF-1α gene region identified the Metarhizium species as falling in the PARB clade of the M. anisopliae species complex. From the EF-1α region, two were M. robertsii, two were M. pinghaense and five of the isolates were M. brunneum (Table 1). One fungal isolate, not known to be an EPF, namely Aspergillus tamarii (Kita) Centralbl. Bakteriol. (MT340979), was also collected during the survey.
The data fitted the model well, and a positive correlation was observed between the fungal concentration and insecticidal activity for both EPF isolates (Figure 4). The same amount of time, measured in days, was required to kill 50% of the E. lanigerum, for both M. robertsii (6EIKEN) (LT 50 = 2.12) and M. pinghaense (5HEID) (LT 50 = 2.05), at a concentration of 10 7 conidia/ml. To kill 90% of the E. lanigerum colonies, the M. robertsii required less time (LT 90 = 4.19) than did M. pinghaense (LT 90 = 4.45).

DISCUSSION
From the study, six different EPF species were obtained across the six apple farms. Collected EPF were all hypocrealean fungal species, which are among the well-known EPF species that are common components of the soil environment microbiota with a cosmopolitan distribution (Chandler et al. 1997;Zimmermann 2007). Among the collected EPF isolates were C. fumosorosea, B. bassiana and M. anisopliae species complex isolates, which are well-known to cause mycosis within various taxa of arthropod, and virtually in all insect orders (Chase et al. 1986; Zimmermann 1986; Meyling and Eilenberg 2007). Similar observations to those that were made in the current study were made in a study conducted by Abaajeh and Nchu (2015), in which fungal isolates belonging to six entomopathogenic fungal species were also successfully isolated from soil samples collected in the Western Cape province, using similar baiting techniques, where collected EPF species included both M. robertsii and P. lilacinum. Purpureocillium lilacinum, when previously used against the cotton aphid, Aphis gossypii Glover (Hemiptera: Aphididae), was observed to have a negative effect on the reproduction of this aphid species when present as an endophyte of the plants (Castillo Lopez et al. 2014). Testing of the efficacy of P. lilacinum isolate collected during the current study against the WAA was however not possible due to the short survival period of the fungus when grown an artificial media, SDA.
Strains of almost all the collected EPF species have also been commercially developed as mycoinsectides for biological control against various insect pest species across agroecosystems (De Faria and Wraight 2007;Asensio et al. 2003;Shi and Feng 2004). Some of the fungal strains are currently commercialised for use against other insect pests in South Africa (Hatting et al. 2019). Three of the isolated M. anisopliae species complex isolates collected during this survey, namely M. robertsii, M. pinghaense and M. brunneum, are known to be of interest as entomopathogens in agricultural ecosystems (Rehner and Kepler 2017;Hatting et al. 2019). In South Africa, five commercial products containing B. bassiana are currently registered for the management of insect pests, with the products including BB plus WP®, manufactured by Biological Control Products SA (Pty) Ltd, is registered for use against both aphids and spider mites (De Faria and Wraight 2007). All isolates collected during the current study are also known to affect all the life stages of their host, as they produce toxic metabolites such as beauvericin, oosporein and insecticidal cyclic peptides, known as destruxins (Inglis et al. 2001;Zimmermann 2007;Dedryver et al. 2010;Shahid et al. 2012).
Screening of the pathogenicity of the South African EPF isolates at a standard conidial concentration of 10 7 conidia/ml against the WAA, showed the efficacy of local EPF isolates and their potential as possible biological control agents of the WAA. It was observed in this study that the exposure of WAA to the local Metarhizium isolates, M. pinghaense, M. robertsii and M. brunneum, resulted in higher percentage mortality of the insects, relative to when the insects were treated using C. fumosorosea and B. bassiana. This is an indication that the WAA was more susceptible to infection by the Metarhizium isolates, relative to other local fungal isolates. Similarly, Mathulwe et al. (2022) when screening South African EPF and EPN isolates for their pathogenicity against the obscure mealybug, Pseudococcus viburni (Hemiptera: Pseudococcidae), under laboratory conditions, showed the potential of the same EPF isolates used in the current study, M. robertsii (6 EIKEN) and M. pinghaense (5HEID), as possible biological control agents against the mealybug, with an average percentage mortality of > 90%, induced by both EPF isolates, tested at an infective concentration 10 7 conidia/ml. Their study also showed high efficacy of two South African M. brunneum isolates (DO30 and 3GREY), when used against the insect pest. This indicates that these local South African Metarhizium isolates are effective when used against sucking insects and can further be evaluated for their potential as possible biological control agents of the insects under field conditions. Erler et al. (2014), showed a high efficacy of an isolate of M. brunneum (F52) when evaluating the field efficacy of the strain against pear psylla, Cacopsylla pyri L. (Hemiptera: Psyllidae), an insect pest of pears that shows some level of resistance against insecticides. In their study, M. brunneum F52 was found to be effective against different stages of the pear psylla and was also comparable in efficacy to the standard chemical pesticides used for the management of the insect pest. In other countries, M. brunneum has already been successfully commercialised against a variety of insect pests (Rehner and Kepler 2017). Metarhizium pinghaense, in a study conducted by Kirubakaran et al. (2018), was found to have some lethal effects when used against the rice leafroller, Cnaphalocrocis medinalis (Gunee) (Lepidoptera: Pyralidae), which is a destructive pest of rice crops. Stokwe (2016) also showed the potential of two South African-registered commercial EPF products derived from B. bassiana and M. anisopliae to be effective against the WAA, resulting in > 70% mortality under laboratory conditions, with no significant difference between the isolates.
Further analyses of the exposure time and the dose-response bioassays, showed no significant difference between the M. pinghaense and M. robertsii isolates. The results suggest that the Metarhizium isolates have similar effect on the WAA colonies, and when applied under field and semi-field conditions they might perform in similar manner, depending on their tolerance to various environmental variables such as changes in temperature, ultra-violet light and humidity. Both  Metarhizium pinghaense (5HEID), generated using the IBM SPSS statistics data editor, version 28, software program M. pinghaense and M. robertsii were also observed to require a short amount of time of about two and four days, to cause 50% and 90% infection and mortality of the WAA, under laboratory conditions, respectively.
In conclusion, the results obtained in the current study provide an insight into the diversity of EPF species across the agricultural lands of the Western Cape province, South Africa. The obtained results of the screening bioassays conducted in the current study also provides useful information by giving an insight into the potential of local EPF isolates as possible biological control agents, which can be used for the control of WAA in apple orchards, specifically M. brunneum, M. robertsii and M. pinghaense. The results from the study also broaden the range of possible biological control agents that can be used in WAA management, since various biotypes of the WAA are welldocumented to have developed some level of resistance against a wide range of chemical insecticides, and currently only a few WAA management techniques are available, most of which have proven to be unsuccessful. Further research into the performance of the three fungal species against WAA, under both glasshouse and field conditions, as well as into their persistence under field conditions, should be undertaken.