Introduction

Black vine weevil (BVW), Otiorhynchus sulcatus F. (Coleoptera: Curculionidae) is a significant pest of ornamental and fruit crops. Larval feeding on roots and underground plant parts can stunt plant growth and with serious infestations, cause plant death. Although chemical insecticides can provide effective control (Cross and Burgess 1986), they have gradually been withdrawn due to environmental and safety concerns and increased the demand for biological alternatives such as entomopathogenic fungi (EPF). Several species of EPF are common in the soil environment but the introduction and persistence of high levels of colony forming units (CFU) in the soil has been shown to be critical in achieving good control of soil dwelling insect pests (Ansari et al. 2007; Parker et al. 2015). Pilz et al. (2011) showed that Metarhizium brunneum Petsch (Hypocreales: Clavicipitaceae) (until recently Metarhizium anisopliae var. anisopliae) applied to soil as fungal-colonised barley kernels was able to establish in the field and reduce populations of Diabrotica virgifera LeConte in maize, but the number of CFU of M. brunneum in the soil then declined if no insect host was present. Bruck (2005) also showed a gradual decline in the number of CFU of M. brunneum in a peat-based growing medium (GM) after inoculation. However, there has been little work on which soil factors are responsible for longevity of EPF populations (rhizosphere competence), and how this can be extended (Roberts and St Leger 2004). In experiments designed to understand the behaviour of M. brunneum in the soil after application, the highest numbers of CFU were found in the rhizosphere (Hu and St Leger 2002; Bruck 2005; Klingen et al. 2015).

Attempts have been made to supply nutrients to sustain EPF in the absence of insect hosts, either in axenic culture or in soil in the presence of competing microorganisms. EPF such as M. brunneum and Beauveria bassiana (Balsamo) Vuillemin (Hypocreales: Clavicipitaceae) produce extracellular chitinases to penetrate the chitin cuticle of insects (St Leger et al. 1986). The EPF Beauveria brongniartii (Saccardo) Petch, B. bassiana, Lecanicillium psalliotae (Treschew) Zare and M. brunneum have been successfully grown on axenic media containing chitin or chitosan (St Leger et al. 1986; Palma-Guerrero et al. 2007). Chitin-rich substrates including silkworm chrysalis, shellfish wastes and yeast extract, a source of fungal chitin, have been used to axenically culture several EPF (Suresh and Chandrasekaran 1998; Matsumoto et al. 2004; Ypsilos and Magan 2005; Mishra and Malik 2012; Rustiguel et al. 2012). Amendment of GM with chitinous material, crab meal, has been used to support the growth and persistence of inoculated M. brunneum for BVW control (Bruck 2005). Various other carbon and nitrogen sources including cornsteep solid, almond mesocarp, maize, millet, rice and wheat grains, whey, seeds, palm leaves, vegetable and coffee wastes, sugar cane bagasse, fish meal extract, feather powder and urea, have been used as nutrient sources for EPF in axenic culture (Mohan and Pillai 1982; Dorta et al. 1990; Lopez-Llorca et al. 1999; Ypsilos and Magan 2005; Santa et al. 2005; Soundarapandian and Chandra 2007; Sahayaraj and Namasivayam 2008; Kassa et al. 2008; Rodriguez Marcondes et al. 2008). Parker et al. (2015) found that B. bassiana and M. brunneum persisted longer in forest soil when applied as a millet formulation than as a peanut formulation. The persistence of B. bassiana and M. brunneum inocula and their pest control efficacy have been shown to be improved by the amendment of soil with composts, plant wastes and animal manures (Lopez-Llorca et al. 1999; Sabbour 2006), although not to our knowledge, spent mushroom compost (SMC).

SMC from button mushroom, Agaricus bisporus (Lange) Imbach cultivation is a source of fungal chitin, as well as other organic and inorganic nutrients (Gerrits 1988; Nitschke et al. 2011). SMC is widely used in GM (Chong et al. 1991) and as a soil amendment as a source of organic matter and inorganic nutrients (Stofella and Kahn 2001); globally over 30 Mt are produced annually (Gerrits 1988; Anon. 2014). It has not been established whether SMC is a suitable substrate for EPF, although mushroom wastes and SMC have been shown to sustain mycoparasites in the biocontrol of soil-borne fungal plant pathogens (Coventry et al. 2006; Trillas et al. 2006; Gupta et al. 2009; Lopez-Mondejar et al. 2012). SMC frequently contains fungi such as Trichoderma harzianum Rifai (Visscher 1988) which may compete with EPF, so that it requires pasteurisation before use. Colonising pasteurised substrates with EPF in enclosed incubation would reduce microbial competition before mixing into the soil or GM, but increase costs. However, SMC is often steamed at 60–70 °C for several hours before being emptied from mushroom sheds (Gerrits 1988), thereby avoiding the need for subsequent pasteurisation if it can be used promptly after emptying.

The aim of this study was to evaluate the effect SMC on M. brunneum and B. bassiana persistence and virulence towards BVW in a peat-based GM. The hypothesis that SMC would affect persistence and virulence was evaluated in replicated factorial design experiments. To test this hypothesis, M. brunneum and B. bassiana were added to the GM with or without SMC, both in short-term incubations (20 °C) and in longer-term cool winter glasshouse conditions using potted strawberry plants.

Materials and methods

EPF inocula

M. brunneum strain F52 was obtained as an axenic rice grain inoculum containing a manufacturer specified 2% w/w M. brunneum or 9 × 108 CFU g−1 (Met52® Novozymes Biologicals BioAg Group, Bagsvard, Denmark). B. bassiana strain ATCC 74040 was obtained as an axenic liquid inoculum containing a manufacturer specified 7.16% w/w B. bassiana or 2.3 × 107 CFU ml−1 (Naturalis L®, Belchim Crop Protection Ltd, Cambridgeshire, UK). To determine the actual numbers of CFU of M. brunneum or B. bassiana in EPF inocula or substrates used in the experiments, samples (1 g) were mixed with 9 ml sterile water in each of three replicate universal tubes. The resulting extracts were then serially diluted with sterile water so that dilutions of 1 to 10−5 were made. Aliquots (0.1 ml) of each dilution were pipetted into sterile Petri dishes containing potato dextrose agar (Oxoid, Fisher Scientific, 39 g l−1) + chlortetracycline (Sigma Aldrich, 2 ml l−1) + Triton X-100 (Fisher Scientific, 2 ml l−1), which were then incubated at 22 °C. The numbers of actively growing colonies of M. brunneum and B. bassiana were recorded with a binocular microscope after three, five and seven days, which were then used to calculate the number of CFU per g fresh weight of inoculum or substrate. To ensure that the colonies counted were M. brunneum or B. bassiana, the plate cultures were allowed to sporulate and then identified according to their macro- and micro-characteristics (Domsch et al. 1980) and by comparison with axenic cultures of similar age.

Substrates

Non-steamed SMC was obtained within one week of mushroom cultivation from G’s Fresh, May Farm, Littleport, Cambridgeshire, UK. The compost was originally prepared from wheat straw-based horse manure, poultry manure and gypsum, and was colonised by mushroom mycelium. Casing material, a peat and lime mixture used to cover the compost for mushroom production, was removed due to its alkalinity which reduces its suitability as a GM. GM (Bulrush Professional Multi-purpose Compost, Bulrush Peat Co., Magherafelt, Co. Derry, N. Ireland), based on about 60% v/v peat and about 40% v/v wood fibre, and containing proprietary three-month controlled release fertiliser and base fertiliser was used for the experiments. Substrates were analysed for pH, electrical conductivity and moisture content (Coventry et al. 2006).

SMC was pasteurised at 65 °C for 2 h in 354 ml glass jars (microcosms) each containing 250 ml (90 g) SMC. After cooling to ambient, the SMC in the microcosms was inoculated with either M. brunneum or B. bassiana. Met52 granules were added to substrates (SMC or GM) at the recommended rate of 0.5 g l−1 (Anon. 2016) producing an initial substrate CFU count of 6.5 ± 0.5 × 106 CFU l−1. Naturalis L was applied to substrates at a rate of 1 ml l−1, which in preliminary tests (data not shown) produced a similar initial substrate CFU count to that used for Met52. Microcosms of pasteurised SMC without EPF inoculum were also prepared. The microcosms had loosely fitting metal lids allowing gas exchange but minimising desiccation and contamination of the substrate.

The microcosms were incubated at 20 ± 1 °C in darkness for 18 days before the number of CFU of M. brunneum or B. bassiana in the substrates were determined. Samples of the M. brunneum and B. bassiana inocula used in each experiment were analysed for CFU with the same procedure. The incubated and/or EPF colonised SMC samples were used for the following experiments within two weeks.

Microcosm experiment 1: persistence of EPF in peat-based GM

GM (250 ml, 90 g) containing the following additives were filled into microcosms described above (n = 3): (a) none, control (b) uninoculated fresh SMC, 10% v/v (c) M. brunneum Met52 inoculum, 0.5 g l−1 (d) B. bassiana Naturalis L inoculum, 1 ml l−1 (e) SMC colonised with M. brunneum, 10% v/v (f) SMC colonised with B. bassiana, 10% v/v. The microcosms were then kept in an incubator at 21 ± 1 °C in darkness. Samples of the microcosm contents were analysed for the populations of M. brunneum and B. bassiana CFU after 0, 11, 24, 41, 78 and 189 days as described above.

Microcosm experiment 2: control of BVW larvae

Microcosms containing the same treatments as in microcosm experiment 1 were prepared. The following treatments where EPF inocula were added to SMC at the time of filling of microcosms were also prepared: (g) SMC, 10% v/v, inoculated with M. brunneum Met52 0.5 g l−1 (h) SMC, 10% v/v, inoculated with B. bassiana Naturalis L 1 ml l−1.

Between six and ten late instar BVW larvae, obtained from infested strawberry (Fragaria × ananassa Duchesne ex Rozier) plants were inserted around the surface of the substrate in each microcosm to a depth of 25 mm. Small pieces of carrot as food for the larvae and moisture were added to the surface of the substrate and then replaced or replenished weekly. Microcosms were kept at 20 ± 1 °C in darkness for three weeks, the contents were then removed and sorted and the viability of the larvae assessed. Samples of substrate were removed from 25 mm depth after 0, 5, 8, 11 and 21 days and analysed for the populations of M. brunneum and B. bassiana CFU as described above. The experiment consisted of four similar consecutive batches of microcosms, each with three replicate microcosms of each treatment (a) to (h).

Glasshouse strawberry pot experiments

Two similar glasshouse strawberry pot experiments with a duration of 125 days were set up on 3 October 2014 and 5 October 2015. GM containing treatments (a) to (h) described above were prepared and filled into 1.5 l plastic pots (size 14T, Plantpak, Waalwijk, Netherlands) with a single strawberry plant (cv. Flamenco) grown in 7 cm square pots containing a peat-based GM. Four second or third instar BVW larvae, obtained from infested strawberry plants, were inserted in the substrate at a depth of 40 mm at opposite sides of the pot, 30 mm from the edge. Similar pots were prepared without BVW larvae. There were ten replicates of each treatment, and 20 replicate pots of treatment (a). The pots were placed in plastic saucers and spaced on a glasshouse bench in a randomised block design, each block containing a single replicate of each of treatments (b) to (h) and two replicates of treatment (a). The pots were regularly watered to maintain a moisture content of 72 ± 3% w/w in the substrates. The glasshouse minimum air temperature was 2 ± 0.5 °C. Ventilation was set at an air temperature of 20 °C. The pots were lit continuously with high intensity sodium discharge grow lamps (Master Son-T PIA Plus 400 W, Philips) at 4.25 W m−2.

Substrate samples were taken at the start of the glasshouse experiments and from the pots after 59 and 109 days and assessed for populations of M. brunneum and B. bassiana CFU. At the end of the experiments, before any adult BVW had emerged, the substrate was removed and inspected for the numbers of viable and dead BVW larvae and pupae. The remaining substrate was carefully removed and washed from the roots and surface moisture removed from the plants. The plants were then weighed before and after drying at 70 °C.

Statistical analysis

For analyses of EPF CFU in samples taken from different GM treatments and at different time points in the above microcosm and glasshouse experiments, we used one- and two-way ANOVA. A log10 transformation of numbers of EPF CFU in GM samples was used to homogenize the variances in the data. Numbers of surviving BVW larvae and pupae and final plant weights were initially analysed separately for glasshouse experiments 1 and 2 by ANOVA. Since there was no significant effect of the blocking structure on any of the measurements in either experiment, the treatment means were then included in a combined ANOVA of both experiments 1 and 2. The effect of treatments and time sampling points was assessed by conducting two-sided t-tests on means obtained from the ANOVA analyses. Results were analysed by GenStat Version 13.1. Throughout, α ≤ 0.05 was used to determine statistical significance.

Results

Properties of SMC, GM and EPF inocula

The GM and SMC used in the experiments had similar pH values and moisture contents but the SMC had a much higher electrical conductivity (Table 1). Addition of 10% v/v SMC to GM resulted in a ×2 increase in the electrical conductivity compared with non-amended GM.

Table 1 Mean properties (± SE) of spent mushroom compost (SMC) and peat-based growing medium (GM) used in the experiments; n = 5
Full size table

The average EPF populations in the axenic inoculum products determined were 1.28 (± 0.73) × 109 CFU g−1 for M. brunneum in Met52 and 3.37 (± 1.24) × 108 CFU ml−1 for B. bassiana in Naturalis L. These values are similar to the manufacturer product specifications stated earlier. The mean EPF populations in the colonised SMC used in the incubated microcosm and glasshouse strawberry pot experiments were 3.98 (± 1.88) × 108 CFU g−1 for M. brunneum and 5.97 (± 1.75) × 108 CFU g−1 for B. bassiana.

Incubated microcosms

As expected, no M. brunneum or B. bassiana could be detected in the SMC or GM used in the experiments without the addition of Met52 or Naturalis L. Treatments (a) GM only and (b) GM + SMC were therefore removed from the statistical analysis of numbers of CFU. In microcosm experiment 1, there were differences in CFU between the other GM amendment treatments (F3,24 = 39.52; p < 0.001), the time sampling points (F5,24 = 11.13; p < 0.001), and the interaction of GM treatment × time of sampling (F15,24 = 3.92; p = 0.001). The population of B. bassiana CFU in GM declined rapidly after the addition of Naturalis L, and could not be detected after 41 days (Fig. 1). Where B. bassiana was added with 10% v/v colonised SMC, the population of CFU declined more gradually and was still detectable after 189 days although lower than at the start (t24 = 2.54; p = 0.018). Conversely, there was no significant reduction in the population of M. brunneum CFU after addition of Met52 granules or 10% v/v colonised SMC to GM (t24 = 0.787 or 1.593; p = 0.439 or 0.124) (Fig. 1).

Fig. 1
figure 1

Mean population (± SE) of entomopathogenic fungi (EPF; Mb, M. brunneum Met52; Bb, B. bassiana Naturalis L) in growing medium (GM) with and without 10% v/v spent mushroom compost (SMC) in microcosms, n = 3. At 100%, EPF rates per l GM were 0.5 g for Mb and 1 ml for Bb. No EPF were detected in microcosms containing uninoculated GM, with or without SMC, or after 41 days in microcosms containing GM + 100% Bb without SMC

Full size image

In microcosm experiment 2, there were differences in CFU between the EPF and SMC amendment treatments in the GM (F5,30 = 3.30; p = 0.017), the time sampling points (F4,30 = 5.56; p = 0.002), and the interaction of GM treatment × time of sampling (F20,30 = 2.82; p = 0.005). Where B. bassiana was added to GM as Naturalis L liquid inoculum without SMC or with colonised SMC, the number of CFU did not change significantly during the experiment (t30 = 1.21 or 0.06; p = 0.236 or 0.953). Similarly, the addition of M. brunneum to GM, either as Met52 granules or as colonised SMC, did not result in a significant change in the number of CFU during the experiment (t45 = 0.95 or 0.05; p = 0.347 or 0.960). Addition of M. brunneum or B. bassiana with uncolonised SMC to GM resulted in an initial increase in CFU followed by stable numbers in the substrate, i.e., differences between days 0 and 5 were significant (t30 = 2.85 or 4.27; p = 0.008 or < 0.001) but no significant differences between days 5 and 21 (t30 = 0.79 or 0.25; p = 0.436 or 0.804) (Fig. 2).

Fig. 2
figure 2

Mean population (± SE) of entomopathogenic fungi (EPF; Mb, M. brunneum Met52; Bb, B. bassiana Naturalis L) in growing medium (GM) with and without 10% v/v spent mushroom compost (SMC) across four batches of microcosms; n = 3 in each batch. At 100%, EPF rates per l GM were 0.5 g for Mb and 1 ml for Bb. No EPF were detected in microcosms containing uninoculated GM, with or without SMC

Full size image

At the end of incubated microcosm experiment 2, the majority of live BVW were retrieved as late instar larvae although a small number of larvae had pupated due to the warm incubation period. Of the retrieved dead BVW larvae, 82% were visibly colonised with mycelium of M. brunneum or B. bassiana. There were differences between GM amendment treatments in the percentage of retrieved live larvae and pupae (F7,24 = 5.27; p < 0.001). Addition of EPF with 10% v/v colonised SMC to GM resulted in fewer live BVW larvae at the end of the experiment than the untreated or SMC treated GM (t24 > 2.09; p < 0.047) (Fig. 3). The addition of M. brunneum to GM resulted in fewer live BVW larvae and pupae when applied in colonised SMC than when Met52 granules were used alone at ×10 rate (0.5 g l−1) (t24 = 3.14; p = 0.004). Addition of Met52 granules at 0.05 g l−1 to GM with SMC at the time of potting resulted in a final percentage of live BVW larvae and pupae, that was intermediate between and not significantly different to those obtained with the above two methods of applying M. brunneum to GM, (t24 < 1.71; p > 0.100). The effects of addition of B. bassiana to GM, either alone at ×10 rate (Naturalis L at 1 ml l−1) or with SMC at the time of potting at 0.1 ml l−1, on the percentage of live BVW larvae and pupae, were not significant when compared with the untreated control GM (t24 < 1.46; p > 0.157).

Fig. 3
figure 3

Mean final percentage (± SE) of live black vine weevil larvae and pupae in growing medium (GM) with and without 10% v/v spent mushroom compost (SMC) and entomopathogenic fungi (EPF) across four batches of microcosms initially containing 6–10 larvae, n = 3 in each batch. Any of the eight bars with the same letter are not significantly different (α = 0.05). At 100%, EPF rates per l GM were 0.5 g for M. brunneum Met52 and 1 ml for B. bassiana Naturalis L

Full size image

Glasshouse strawberry pot experiments

The minimum, average and maximum glasshouse air temperatures were 3.8, 12.0 and 24.6 °C in experiment 1 and 1.9, 12.1 and 22.0 °C in experiment 2. At the times of the final assessments, the roots of the strawberry plants had grown through the potting substrates and were visible at the sides and base of the pots.

No EPF CFU were found in the strawberry pots containing only GM or GM with 10% v/v SMC. Treatments (a) GM only and (b) GM + SMC were therefore removed from the statistical analysis of numbers of CFU. There were differences in CFU between the other GM amendment treatments (F5,18 = 5.83; p = 0.002) and the interaction of GM treatment × time of sampling (F10,18 = 4.01; p = 0.005). In pots where EPF and SMC were combined with GM at the time of potting of strawberry plants, the initial populations of EPF CFU were lower than in the other EPF treatments (t18 > 3.77; p < 0.001) (Fig. 4). However, by day 59 of the experiments, these populations had increased and were not significantly different to those of the other EPF treatments (t18 < 1.07; p > 0.298). Where Naturalis L was added to GM without SMC, the B. bassiana CFU population declined, and by the end of the experiments was significantly lower than in the other EPF treatments (t18 > 2.59; p < 0.019) (Fig. 4).

Fig. 4
figure 4

Mean population (± SE) of entomopathogenic fungi (EPF; Mb, M. brunneum Met52; Bb, B. bassiana Naturalis L) in growing medium (GM) with and without 10% v/v spent mushroom compost (SMC) following potting of strawberry plants in two experiments; n = 3 in each experiment. At 100%, EPF rates per l GM were 0.5 g for Mb and 1 ml for Bb. No EPF were detected in pots containing uninoculated GM, with or without SMC

Full size image

In both of the glasshouse strawberry pot experiments, small natural infestations of BVW larvae were recorded in pots that were not artificially infested. Of pots that did not have an initial artificial infestation of larvae, 37% in glasshouse experiment 1 and 3% in glasshouse experiment 2 had larvae at the end. It can be assumed that similar percentages of artificially infested pots also had a similar natural infestation of BVW larvae since they were randomly positioned among pots without an artificial infestation.

At the end of the glasshouse strawberry pot experiments, the majority of live BVW were retrieved as late instar larvae although a small number of pupae were also retrieved. Based on the initial artificial and subsequent natural infestations of larvae, 89% of dead BVW larvae had disintegrated and could not be found although 11% of dead larvae and pupae were retrieved. Of these retrieved dead larvae and pupae, 76% were visibly colonised with mycelium of M. brunneum or B. bassiana. There were significant differences between GM amendment treatments in the percentage of retrieved live larvae and pupae (F7,16 = 2.99; p = 0.033). Addition of 10% v/v SMC or B. bassiana as Naturalis L at 1 ml l−1 to GM did not significantly affect the final percentage of live BVW larvae and pupae in pots compared with the untreated GM control (t16 = 0.758 or 0.155; p = 0.460 or 0.879) (Fig. 5). Compared with the untreated control, addition of SMC with EPF inocula to GM at a rate of 0.05 g l−1 (Met52) or 0.1 ml l−1 (Naturalis L) resulted in reductions in the percentage of live BVW larvae and pupae (t16 > 2.08; p < 0.048), which were not significantly different to that obtained with adding Met52 alone to GM at 0.5 g l−1 (t16 < 0.43; p > 0.673).

Fig. 5
figure 5

Mean final number (± SE) of live black vine weevil (BVW) larvae and pupae in growing medium (GM) with and without 10% v/v spent mushroom compost (SMC) and entomopathogenic fungi (EPF) across two strawberry pot experiments; n = 10 and 20 for untreated controls in each experiment. There were low natural background BVW larvae infestations of pots in both experiments. Values are the means of pots with and without an initial artificial infestation of four BVW larvae. Any of the eight bars with the same letter are not significantly different (α = 0.05). At 100%, EPF rates per l GM were 0.5 g for M. brunneum Met52 and 1 ml for B. bassiana Naturalis L

Full size image

Across all treatments in the two glasshouse experiments, dry weight and fresh weight of strawberry plants were closely correlated (R2 = 0.928 and 0.882; p < 0.001). The effects of BVW larvae on plant dry weight were assessed by noting which uninfested plants did not have a natural infestation of larvae, either dead or alive at the end. However, it is possible that some of these plants had a natural infestation but due to disintegration of dead larvae, these were not detected at the end of the experiment. Overall, the presence of BVW larvae resulted in a reduction in plant dry weight (mean values 9.0 and 10.5 g with and without larvae present) (F1,16 = 5.19; p = 0.037). In the absence or presence of BVW larvae, there were no significant effects of SMC or EPF treatments on plant dry weight compared with plants growing in unamended GM (F7,16 < 2.27; p > 0.097).

Discussion

This work shows that SMC enables EPF to persist in a GM in the absence of an insect host, possibly by providing an alternative nutrient source such as chitin. This resulted in improved efficacy of EPF inoculum compared with using EPF inoculum without SMC. For M. brunneum, SMC reduced the amount of EPF inoculum needed to obtain the same BVW larvae kill by at least tenfold, and/or improved the kill of BVW larvae. B. bassiana was only effective in killing BVW larvae when incorporated in GM with SMC. M. brunneum persisted better than B. bassiana in the absence of SMC, possibly due to the nutrition provided by rice grain in the Met52 formulation. The initial increases in numbers of CFU of M. brunneum and B. bassiana following addition to GM with uncolonised SMC were probably due to the nutrition provided by the SMC. Where inoculum was added as colonised SMC there was no further increase above the initial level of CFU. Mixing the EPF inoculum with SMC at the time of blending with GM is preferable to using SMC colonised with EPF in GM since it does not require production and storage of colonised SMC, or registration of new SMC substrate formulations of EPF. The cost of axenic EPF inoculum is typically > $50 kg−1 (e.g., http://www.evergreengrowers.com; http://www.growershouse.com). At a recommended rate of a commercial formulation of M. brunneum, Met52 of 500 g m−3, this adds > $25 m−3 to the cost of a GM. However, by incorporation with SMC, the effective rate and cost of EPF inoculum can be reduced by 90%. Bruck and Donohue (2007) showed that for a granular formulation of M. brunneum F52, an application rate of 600 g m−3 GM provided slightly better control of BVW than 300 g m−3. The level of BVW control they achieved corresponded with that obtained here with a similar, recommended Met52 rate of 500 g m−3 GM (Anon. 2016).

In cool temperate soil over one year, rhizosphere competence of Norwegian isolates of Beauveria pseudobassiana Rehner & Humber and B. bassiana was better than that of an exotic isolate M. brunneum resulting in greater long term control of BVW larvae (Klingen et al. 2015). Parker et al. (2015) found that the persistence of B. bassiana isolates was better than that of M. brunneum isolates following inoculation in Canadian forest soil. When EPF were applied with SMC in incubated microcosms here, M. brunneum was more effective in controlling BVW larvae in the short-term and sustained higher populations of CFU in the longer-term than B. bassiana. However, in glasshouse experiments at lower temperature, both EPF sustained similar populations of CFU and were equally effective in BVW larvae control in the longer-term.

Klingen et al. (2015) found that B. pseudobassiana, B. bassiana and M. brunneum were all effective in controlling BVW larvae at 18 °C whereas at 6 °C, only the Beauveria species had efficacy, although the level of control was poor. Amendment of soil or GM with SMC may improve the biocontrol efficacy of EPF at low temperatures, which has been a significant challenge (Klingen et al. 2015).

The strawberry plant material used for the glasshouse experiments was raised in a peat-based GM that was not protected against BVW. After repotting into larger pots containing GM with EPF and/or SMC treatments, the original potting medium in the centre of the pot may have allowed some BVW survival, particularly from a natural infestation which was not necessarily placed in the surrounding GM. The inclusion of EPF + SMC in the strawberry propagation GM may provide a greater level of protection against BVW throughout the crop.

The chitin content of mycelium of A. bisporus is around 5% w/w on a dry matter basis (Nitschke et al. 2011) so it is possible that the fungal chitin in SMC provided a selective nutrient source for EPF in the same way as insect cuticle chitin. There are also nutrient sources for EPF in SMC other than mushroom mycelium since both M. brunneum and B. beauveria grew well on sterilised compost before mushroom cultivation, although growth on pasteurised compost without mushroom mycelium was poor (Fitzgerald and Noble 2013). Further work is needed using different sources of SMC to determine if there is a relationship between the content of chitin and/or other nutrient sources and the growth of EPF. Further work is also needed to determine if effective and economic control of BVW and other soil dwelling pests can be achieved in the field at low application rates of SMC + EPF inoculum. However, this work has already demonstrated that SMC can provide a benefit in the biocontrol of BVW larvae and potentially other soil-dwelling pests in containerised plant production, thereby valorising this by-product.