Grazing by collembola affects the outcome of interspecific mycelial interactions of cord-forming basidiomycetes

f u n g a l e c o l o g y 4 ( 2 0 1 1 ) 4 2 e5 5

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/funeco

Grazing by collembola affects the outcome of interspecific mycelial interactions of cord-forming basidiomycetes Timothy D. ROTHERAY, Matthew CHANCELLOR, T. Hefin JONES, Lynne BODDY* Cardiff School of Biosciences, Cardiff University, Cardiff, CF10 3AX, United Kingdom

article info

abstract

Article history:

While there is a plethora of studies on the effects of invertebrate grazing on mycelia,

Received 16 June 2009

including several studies on saprotrophic cord-forming basidiomycetes, there is little

Revision received 28 July 2010

information on the effects of grazing on mycelial interactions. The study compares the

Accepted 15 September 2010

progress and outcomes of interspecific mycelial interactions between Hypholoma fasciculare,

Available online 21 October 2010

Phallus impudicus, Phanaerochaete velutina and Resinicium bicolor when grazed by the collembola

Corresponding editor: Gareth Griffith

Folsomia candida or Protaphorura armata in agar culture and trays (24
24 cm) of non-sterile soil. In ungrazed systems results were broadly consistent with previous studies, though

Keywords:

there were few instances of deadlock, and a clear transitive (A > B > C) hierarchy could not be

Accelerated mycelial extension

discerned. Instead, there was an intransitive hierarchy (i.e., A > B, B > C but C > A). Addi-

Interspecific variation

tionally, in agar culture, there were considerable differences in combative ability of four

Mycelial morphology

different strains of H. fasciculare. Collembola grazing had major effects on mycelial interactions. F. candida grazing altered both the outcome and progression of half of the fungal interactions studied, while the less active P. armata had almost no discernable effects on fungal interactions. Grazing by F. candida affected mycelial extension rate in five of the interaction combinations, some increasing, others decreasing. In grazed systems of P. velutina interacting with H. fasciculare, extension rate of the former was much more rapid over the opponent than over soil. Not only did grazing affect mycelial interactions, but interactions affected grazer activity; H. fasciculare was grazed in areas where it was interacting with P. velutina mycelium, but less so elsewhere. By altering the outcome of mycelial interactions and mycelial extension rate, collembola grazing may alter the distribution of cord-forming fungi on the forest floor, and may also play a role in maintaining the diversity of fungal species. The differences in combative ability of different strains of a species imply an even more complex scenario in the natural world. ª 2010 Elsevier Ltd and The British Mycological Society. All rights reserved.

Introduction As many species of saprotrophic cord-forming fungi occupy similar spatial niches at the soilelitter interface, interactions are inevitable (Boddy 2000). Interspecific fungal interactions are highly aggressive and generally occur following gross

mycelial contact (Boddy 2000). Interactions result in either deadlock or some form of replacement (i.e. partial, complete, or mutual); these outcomes are variable and can be substantially affected by both biotic and abiotic variables (Chapela et al. 1988; Woods et al. 2005). Intraspecific genetic variability may also be

* Corresponding author. E-mail address: [email protected] (L. Boddy). 1754-5048/$ e see front matter ª 2010 Elsevier Ltd and The British Mycological Society. All rights reserved. doi:10.1016/j.funeco.2010.09.001

Grazing by collembola affects the outcome of mycelial interactions

important in determining fungal interaction progression and eventual outcome, although most studies investigating interactions of higher fungi have tended to concentrate on a single strain of each species. A few studies with Trichoderma species (Ascomycota) interacting with wood-decay fungi have involved multiple strains (see Philp et al. 1995; Wheatley et al. 1997) but most of these have focussed on strains of medical or biocontrol value (Walker et al. 1995; Vainio et al. 2001). Despite possible implications for both species abundance and diversity, the role played by fungal genetic variability in fungal fitness remains unknown. As well as encountering other fungi, non-unit-resourcerestricted fungi (sensu Cooke & Rayner 1984) growing out from resources are also more accessible to grazers. Various soil invertebrate taxa, including Nematoda, Collembola and Oligochaeta, are known to graze on fungal mycelia and can substantially alter fungal morphology and physiology (Ruess et al. 2000; Harold et al. 2005; Tordoff et al. 2006; Boddy & Jones 2008). Such effects are likely to alter fungal fitness and, therefore, their combativeness in interactions with other soil microorganisms, including fungi. The visible changes in morphology and physiology during fungal interactions are associated with biochemical changes, such as increased enzymatic activity and production of diffusible (DOC) and volatile (VOC) organic compounds (Griffith et al. 1994; Baldrian 2004; Hynes et al. 2007; Evans et al. 2008). This enhanced activity may also lead to lysis of hyphal compartments and nutrient leakage into the surrounding environment (Wells & Boddy 2002). Numerous plant species use chemicals emitted as a result of insecteherbivore action to attract parasitoids to the vicinity (Dicke 1994; Tentelier & Fauvergue 2007), and a comparable process may occur with fungi. Mycophagous grazers may use chemical cues arising from interaction activity to locate high quality resource patches while fungi may actively recruit grazers to provide a competitive advantage over an opponent. Orientation to fungal odours by invertebrates has been well documented (Swift & Boddy 1984; Hedlund et al. 1995; Guevara et al. 2000) and anecdotal evidence exists of attraction of fungus gnats to fungal interaction zones (Boddy et al. 1983). This study aims to: (i) elucidate the effects of invertebrate grazing on the morphology and competitive abilities of saprotrophic cord-forming basidiomycetes when interacting interspecifically; (ii) determine whether physiological and chemical changes in the mycelium, such as pigment production, may affect invertebrate behaviour (e.g. with preferential invertebrate grazing at interaction zone); (iii) determine whether different fungal strains of the same species affect fungal combativeness and invertebrate grazing. One set of experiments was performed in agar culture, and another set in trays of soil. Though soil trays are more realistic, agar culture allows many more combinations to be performed.

43

(Sigma, UK). Collembola were provided with dried baker’s yeast (Saccharomyces cerevisiae, Spice of Life Ltd., Cardiff, UK) weekly. Tub moisture was maintained with deionised water (DI). Experimental F. candida were selected using a stacked sieving system with sieves of known pore size, the larger sieves being uppermost (Nickel-Electro Ltd., Weston-superMare, UK). Collembola were added to the top sieve and allowed to self-sort by moving through the sieves for 5 min. Those of body diameter 250e400 mm were placed in fresh culture pots and left without food for 24 hr to evacuate any gut contents. These were then removed to experimental plates using an electric “pooter”.

Fungal isolates H. fasciculare (four fruit bodies isolates, from disparate locations; agar medium experiment only; Supplementary Table 1), Phallus impudicus (isolated from a cord), P. velutina (isolated from wood) and Resinicium bicolor (two wood isolates (1, 2) agar experiment only) were cultured on 2 % malt extract agar medium (MEA, 20 g l1 malt, Munton and Fison, UK; Lab M agar no. 2) and maintained in the dark at 20  C. H. fasciculare, P. impudicus and P. velutina were from the Cardiff University Culture Collection and the R. bicolor isolate from Steve Woodward, Aberdeen University. The first three species are extremely common in temperate deciduous woodland; R. bicolor is common in coniferous woodland though infrequent in deciduous woodland. These species were selected for study because they cover a broad spectrum of foraging patterns (Boddy 1999), and their interaction outcomes (Boddy 2000) and effect of collembola grazing when growing alone (Boddy & Jones 2008) have been well characterised.

Interactions in agar culture

Materials and methods

Fungi were paired opposite to each other in all of the 36 possible combinations. Inoculation was timed to ensure that when the two mycelia met each individual had colonised an equivalent area of agar medium. Three, 5 mm inoculum plugs were spaced equidistantly along a chord 2 cm from the edge of 2 % MEA plates (9 cm dia). Plates were incubated at 20  C in the dark and maintained in cardboard boxes within black polythene bags to reduce water loss. Twenty collembola were added at two points at the edge of the Petri dish, 10 on each mycelium, to each of five replicates per combination when the interacting fungi in a minimum of 50 % of the replicates had made contact for 2 d or more. Twenty collembola are a realistic field density (Petersen & Luxton 1982) and have effects on fungal morphology and activity when growing alone (Boddy & Jones 2008). Agar plate interactions were photographed with a Nikonª Coolpix 5700 digital camera mounted on a Kaiser RA1 camera stand (Kaiser, Germany) set at a height of 36.7 cm. Photography started on the day of collembola addition (t0) and then every 2 d until 14 d (ty), followed by every 4 d until 26 d (tx).

Collembola culturing

Determination of outcome of interactions in agar culture

Folsomia candida and P. armata were cultured in 0.6 l plastic tubs with pierced lids for aeration. Each tub contained 9:1 plaster of Paris (Minerva Dental Ltd., Cardiff, UK): activated charcoal

Outcome of interaction was classed as: (1) deadlock, where neither fungus gained territory of the other; (2) partial replacement, where one fungus extended at least 5 cm into

44

the territory of the opponent but did not reach the opposite side of the plate; (3) mutual replacement, as (2) but both opponents extended into the territory of the other; or (4) replacement, where the mycelium of one fungus was completely killed by the other and the former’s territory completely colonised. In agar, outcomes were determined by visual inspection and confirmed by reisolation from the underside of the agar onto 2 % MEA. Relative combative ability of different H. fasciculare strains was compared by employing a scoring system, following Crockatt et al. (2008). Each replicate was scored as: 2, total replacement of opponent; 1, partial replacement of opponent; 0, deadlock; 1, partial replacement by opponent; 2, total replacement by opponent. A cumulative score was attributed to each strain providing a combative index. A strain showing total replacement in all five replicates scored 10, while a species totally replaced in all replicates scored 10.

Soil tray preparation Soil (pH 4.5) collected from the top 5e10 cm mixed deciduous woodland in (SO517069) was air-dried for 7 d, and sieved through 4 mm then 2 mm mesh to remove large organic material and stones. The soil was then frozen for 24 hr to prevent population explosions of resident soil fauna (but minimising effects on microbial community) before rewetting with DI to attain a soil matric potential of ca. 0.012 MPa. The wet soil (200 g/tray) was evenly compacted to 4 mm depth into 24
24
2 cm non-vented lidded bioassay trays (NunceGibco, Paisley, UK), which were then weighed to 0.01 g and stored in stacks of 20, double-wrapped in black PVC bags to prevent desiccation (20  C  1  C, dark). Trays were used within 7 d of being made, and rewetted to original weight with DI every 7 d. During the experiment any contaminating flora were removed. No non-experimental fauna were observed.

T.D. Rotheray et al.

were weighed and deionised water was added weekly to maintain that weight. To half of the trays, collembola were added. For any given interaction, when 50 % or more of the fungi had met the opposing fungus, for a minimum of 2 d, 20 F. candida were added to each corner of the tray (80/tray). P. armata (80/tray) was used in two additional interactions; R. bicolor against H. fasciculare and P. impudicus against H. fasciculare. Methodology for culture and addition of P. armata was the same as for F. candida. Images of soil trays were captured as for agar plates but with the camera at a height of 47 cm and artificial illumination provided by two 1 000 W spot flood lamps (Gamma 7 Al, Gamma, Chicago USA), each set 60 cm either side of the stand and 120 cm above it. Photography commenced on the day of collembola addition, t0, and continued every 2 d for the first 11 d (t10), then every 4 d until 23 d (t22), every 10 d until 43 d (t42) and finally at 85 d (t84). Outcome of interactions was determined t84 after collembola addition (see below).

Determination of outcome of interactions in soil microcosms Extension rates of one mycelium growing over another were determined by linear regression analysis of extent by time from digital images. Mycelial extent was measured as the furthest visible extent of mycelium along four zones at equal angles through a 90 segment of the mycelium (Fig 1). Unless removed through grazing the mycelium at the two central measurements (30 and 60 ) was invariably in contact with the opposing mycelium. Cords severed by grazing and not visibly

Preparation of wood inoculum Wood blocks were cut from freshly-felled beech (Fagus sylvatica) timber (Coed Cymru Hardwood Sawmill, Wentwood, UK) into 2
2
1 cm blocks. Blocks with discolouration and knots were discarded. Groups of 20 blocks were doublewrapped in heat-sealed biohazard bags and autoclaved for two separate 1 hr sessions each being 24 hr apart. Sterile wood blocks were added to cultures of one isolate each of H. fasciculare, P. impudicus, P. velutina and R. bicolor, on 2 % MEA and incubated in the dark at 20  C for 3 months.

Inoculation of soil trays All four fungal species were interacted in every possible interspecific combination, with 20 soil trays/interaction (total 120 soil trays). Mycelium and agar were scraped off each wood block inoculum before addition to soil trays, to prevent growth of soil microbes on the agar medium. To each tray a wood block of the slower growing of the two fungal species was added 9 cm away from a corner along a diagonal between corners. A wood block colonised by an opponent was added, 9 cm away from the opposite corner, timed such that the fungi met when the mycelia were approximately 8 cm dia. Trays

Fig 1 e Measurement of mycelial radial extent in soil microcosms. The direction in which the four measurements (indicated with white single arrows) were taken at t0 and treach. Measurements were made on both species in every interaction. The length of the double-headed arrow (22.6 cm internal tray width) was applied to scale all images.

Grazing by collembola affects the outcome of mycelial interactions

connected to the wood block were not included in measurements. At the end of the experiment wood blocks that had been reached by the mycelium of the opposing fungus were harvested. These blocks were cut in half, three wood chips taken from the freshly cut surface, and plated onto MEA and incubated for 7 d at 20  C, in the dark. If only the opposing species was re-isolated, the outcome was recorded as total replacement. If both fungal species were isolated the result was recorded as partial replacement. If only mycelium of the original inoculum was isolated the outcome was recorded as overgrowth by the opponent but not replacement. Results for each interaction were recorded as overgrowth, partial replacement or replacement, and expressed as a percentage of total reisolations. Where mycelium of one strain overgrew that of the opponent but did not reach the wood block, the outcome was reported as partial overgrowth.

Statistical analysis Statistical comparisons made between mean extension rates were as follows: (1) the difference in mycelial extent between grazed and ungrazed trays across all four measurements at the final time point (treach; the time when a mycelium reached the edge of the tray, which depended on species); (2) the difference in mycelial extent between grazed and ungrazed trays when growing toward (or over) the opponent mycelium (30 and 60 measurement) at the final time point (treach); (3) the difference in mycelial extent over time between grazed and ungrazed trays when growing toward (or over) the opponent mycelium; (4) the difference in mycelial extension over time between grazed and ungrazed trays when growing over soil (0 and 90 ); (5) the difference in mycelial extension over time when growing over soil compared to when growing over the opponent mycelium in grazed trays; and (6) the difference in mycelial extension over time when growing over soil compared to when growing over the opponent mycelium in ungrazed trays. This complete analysis was not always appropriate and a framework was developed to determine the suitable level of analysis for each species in each interaction (Fig 2).

Results Outcome of interactions in agar culture H. fasciculare: It was combative, several strains replacing R. bicolor, and being replaced only occasionally by P. velutina and R. bicolor 1. H. fasciculare 4 was, however, sometimes replaced by R. bicolor and P. velutina though not P. impudicus. There were marked differences between different strains of the same species. H. fasciculare 3, for example, replaced R. bicolor 1 whereas H. fasciculare 4 did not (Table 1). H. fasciculare 1 was the most combative of the four isolates, and H. fasciculare 4 was the weakest (Tables 1 and 2). Grazing increased combativeness in two isolates (H. fasciculare 2 and H. fasciculare 3) and reduced it in the others (Table 2). H. fasciculare grew as dense mycelium from the inoculum and overgrew opponents with the formation of cords (Supplementary Table 2), occasionally

45

emerging through the interaction zone from several discrete points (Fig 3A). H. fasciculare 4 often produced non-linear, apparently disordered, cords when overgrowing other species (Fig 3I). All H. fasciculare isolates produced yellow pigment although H. fasciculare 1 produced pigmentation (Fig 3A and B) only when overgrowing another species. All new growth of H. fasciculare isolates was white, with the exception of the cords of H. fasciculare 4 when interacting with P. impudicus, which developed a yellow colour (Fig 3I). P. impudicus: Although it was overgrown to varying degrees by H. fasciculare 1, 2 and 3, P. impudicus was strongly combative partially replacing, mutually replacing or deadlocking with all other species and strains (Table 1). In the presence of other species, the morphology of P. impudicus changed before contact occurred - dense cords originated from up to 1 cm behind the growing front (Fig 3K; Supplementary Table 2). This morphological response also occurred in self-pairings; when P. impudicus was substantially overgrown it occasionally broke through the opposing mycelium at the interaction zone, producing fast growing plumes (Fig 3G). P. impudicus mycelia darkened in some interactions but did not produce strong pigmentation (Supplementary Table 2). During interactions with R. bicolor, a zone of lysis was produced in the interaction zone as P. impudicus replaced R. bicolor (Supplementary Table 2; Fig 3K and O). Collembola burrowed within the lytic zones (Fig 3N). Grazing slightly increased the overall combativeness of P. impudicus (Table 2). P. velutina: It was less combative than H. fasciculare and P. impudicus though it occasionally replaced H. fasciculare 2 and 3 (Table 1). When overgrowing another mycelium, P. velutina formed finely branched cords often emerging from a narrow point of the interaction zone (Fig 3J). Combativeness of P. velutina increased slightly when grazed (Table 1), and rate of overgrowth increased when grazed cords of P. velutina and P. impudicus overgrew each other (Supplementary Table 2). A dark pigment permeated the medium when P. velutina was overgrown and when interacting with itself (Supplementary Table 2). R. bicolor was less combative than H. fasciculare and P. impudicus. R. bicolor 1 was more combative than R. bicolor 2. Combative ability of R. bicolor 1 was reduced when grazed whereas that of R. bicolor 2 increased slightly (Table 1). Following contact with mycelium of an opposing species, R. bicolor 1 rapidly produced dense aerial hyphae up to about 1 cm behind the interacting front (Fig 3L), but this was either less marked or absent in R. bicolor 2 (Supplementary Table 2). R. bicolor overgrew as dense, tightly packed, unbranched linear cords (e.g. Fig 3L). When interacting with H. fasciculare and P. velutina (Fig 3M), R. bicolor produced deep red pigment (Fig 3E, F and H), which was more pronounced in R. bicolor 1 than in R. bicolor 2 (Fig 3E, F but see H).

Outcome of interactions between extra-resource mycelia on soil In ungrazed systems there was never evidence of one mycelium replacing another on the surface of soil, but there was often overgrowth especially by P. velutina and P. impudicus (Table 3; Fig 4), to the extent that the wood block inoculum of an opponent was sometimes reached and interactions then took place within the wood (see below). In interactions

46

T.D. Rotheray et al.

(a) A comparison of the mycelial extent across all four measurements between grazed trays and ungrazed using a two-sample t-test (at Treach). The observation that fungal extension appeared to be different dependent on whether it was growing across (or toward) the opponent led to the subsequent test in all interactions.

(b) A comparison of the mycelial extent across (or towards) the opponent (30º and 60º measurements Fig 1) between grazed and ungrazed trays using a two-sample t test (at Treach). No significant difference

Significant difference

(d) The mycelial extent was measured at three equally (c) No further analysis

spaced time points between t0 and treach.

(f) Repeated Measures (RM) ANOVA co mparing the change in mycelial extent over time in grazed and ungrazed trays across all four measurements.

(g) Data examined to determine whether overall extension measurement mean was negative (mycelium reducing over time) or positive (mycelium extending over time) at three equally spaced time points between t0 and treach. Mean negative extension

(h) RM ANOVA was used to compare extension toward opponent in grazed and ungrazed trays using 30º and 60º measurements (Fig 1).

Mean positive extension

(i) Both the time points and replicates in which mycelium commenced overgrowing the opponent mycelium in the 0º and 90º measurements (Fig 1) were determined. The data were subsequently grouped into two sets (i) extension toward/over opponent mycelium and (ii) extension over soil sets. RM ANOVA was used to compare extension over opponent in grazed and ungrazed trays, and extension over soil in grazed and ungrazed trays.

Mycelia sometimes appeared to be extending at different rates over the opponent and over the soil within grazed and ungrazed treatments. To test this the following step was used.

(j) RM ANOVA comparing extension over opponent and over soil within both grazed and ungrazed trays.

Fig 2 e The framework applied to refine application of statistical tests in analysis of overall extension change during grazed and ungrazed interactions. Where data violated normality a ManneWhitney U-test (instead of the t-test) or a two-way ANOVA on ranked data (instead of the RM ANOVA) was used. When assumptions were violated a Scheirer Ray Hare test replaced two-way ANOVA.

between R. bicolor and P. velutina, but not other combinations, cords of the latter grew to both sides of the R. bicolor wood block forming a complete ring comprising a thick cord in ungrazed systems (Fig 4A and B). In some interactions fungi produced pigments and exudates at the point of contact with the opponent (Fig 4J and K). R. bicolor produced a red pigment along the interaction line when interacting with H. fasciculare and small globules of exudate were

visible where R. bicolor cords met those of P. velutina (Fig 4J and K). There was evidence of preferential grazing by F. candida at the interaction zone, for example with H. fasciculare when interacting with either P. velutina or P. impudicus (Fig 4F, G and I). In grazed systems R. bicolor mycelia were always completely grazed away by F. candida (e.g. Fig 4D) whereas P. armata apparently had little impact (Fig 5E). P. velutina was sometimes heavily grazed, but not to extinction, especially in

R. bicolor 2 Ungrazed

Grazed

R. bicolor 1

P. velutina

Ungrazed

Grazed

Ungrazed r 100

Hf1

r 100

r 100

MR 80 PR 20

Hf2

pr 100

pr 100

Hf3

pr 60 D 40

D 20 pr 80 pr 50 D 50

MR 60 PR 20 pr 20 pr 100

r 11 PR 89

r 22 PR 78

Hf4

PR 100

PR 100

pr 10 PR 40 R 50

Pi

pr 100

pr 100

pr 100

d 11 MR 11 PR 22 R 56 pr 100

Pv

MR 80 D 20

MR 80 pr 20

pr 20 r 80

Rb1

D 100

D 100

Grazed

P. impudicus

H. fasciculare 4

H. fasciculare 3

H. fasciculare 2

Ungrazed

Grazed

Ungrazed

Grazed

Ungrazed

Grazed

Ungrazed

Grazed

r 100

PR 100

PR 100

D 100

D 100

D 100

D 100

D 100

D 100

MR 75 R 25 MR 22 r 33 R 44 MR 20 D 80

MR 75 r 25 r 80 R 20

MR 100

MR 40 PR 60 PR 33 D 67

D 100

D 100

D 100

D 100

D 100

D 100

D 25 PR 50 pr 25

pr 100

R 100

D 100

D 100

D 100

Grazing by collembola affects the outcome of mycelial interactions

Table 1 e Outcome of interactions 12 weeks after addition of Folsomia candida as grazing treatment

MR 20 pr 40 r 40

Outcomes were mutual replacement (MR), partial replacement (PR), total replacement (R) and deadlock (D). For outcomes designated PR and R, upper case signifies that replacement was by the fungus listed at the head of the column whereas lowercase signifies replacement by the fungus listed in the row. Numbers refer to percentage of replicates exhibiting a given response. There were 4e10 replicates/interaction. Hf ¼ H. fasciculare, Pi ¼ P. impudicus, Pv ¼ P. velutina, Rb ¼ R. bicolor.

47

48

T.D. Rotheray et al.

Table 2 e Combative scores of H. fasciculare isolates against other species in agar 12 weeks after addition of Folsomia candida as grazing treatment R. bicolor 2 Ungrazed H. fasciculare 1 H. fasciculare 2 H. fasciculare 3 H. fasciculare 4 Inverse column score

Ungrazed Grazed Ungrazed Grazed Ungrazed Grazed Ungrazed Grazed

R. bicolor 1

Grazed

Ungrazed

10

Grazed

1 10

3

5

1 L22

L6.5

0 1 0 5.5

L6.4

0 5

0 3 L5

Grazed 3

2.5 1.1

3.5

Ungrazed 4

2.5

5

P. impudicus

10

3

3

Grazed

10

4 10

5

Ungrazed

0

10

L18

P. velutina

10 L8

1

5 1

Total row score 25 23 11.5 16.5 6.9 13.5 L13.5 L19

Total replacement in all replicates scored 10 points, partial replacement five points and deadlock or mutual replacement zero points. Scores are based on the outcome of all replicates. Score in the right column is for the isolates listed in the rows. Scores in the bottom row relate to the species listed at the head of the columns.

interactions with R. bicolor (Fig 5A). H. fasciculare was usually only sparsely grazed, except when interacting with P. velutina where it was completely removed in grazed systems by 64 d (Fig 5G). In three (of six) interaction combinations, grazing by F. candida altered the extent of overgrowth by mycelia on the soil surface (Table 3). For example, the mutual overgrowth of P. velutina and R. bicolor in ungrazed systems shifted to complete overgrowth by P. velutina when grazed (Table 3; Fig 5A and B). There was greater growth of P. impudicus over R. bicolor when grazed than ungrazed. Interactions involving H. fasciculare were little changed when grazed except when interacting with P. velutina, where F. candida grazing shifted the balance in favour of the latter (Table 3). There were no differences in mycelial overgrowth between ungrazed systems and those grazed by P. armata (Table 3).

Extension rate of mycelial opponents in soil microcosms In five interaction combinations the change in mycelial radial extent was significantly different between grazed and ungrazed treatments (Table 4, Figs 5 and 6). Of these, in three combinations, R. bicolor against H. fasciculare, P. impudicus and P. velutina, the presence of F. candida resulted in a decrease in mycelial extent of R. bicolor (Fig 5). In the other two interactions, F. candida grazing of P. velutina interacting with either H. fasciculare or P. impudicus, both grazed and ungrazed mycelia increased in size over time (Fig 6). The radial extension of P. velutina (for all four extension measurements taken per replicate) was significantly (P  0.05) less in grazed than in ungrazed treatments when interacting with P. impudicus but not significantly (P > 0.05) different when interacting with H. fasciculare (Fig 6A and B). P. velutina growth over H. fasciculare was slightly but not significantly (P > 0.05) faster in grazed systems than in ungrazed systems (Fig 6C). When over soil and interacting with H. fasciculare, however, mycelial extension of P. velutina was significantly (P  0.05) faster in ungrazed trays (Fig 6E).

Outcome of interactions within wood inocula in soil microcosms H. fasciculare often replaced or partially replaced R. bicolor in wood blocks, though was itself often replaced by P. velutina

(Table 5). P. velutina also sometimes replaced R. bicolor. In other interactions there was no evidence of replacement even if mycelial overgrowth on soil had occurred. There was some evidence of changes in the outcome of interaction in wood blocks following grazing: F. candida grazing enhanced the ability of P. velutina to replace H. fasciculare. In the presence of P. armata, H. fasciculare was less able to replace R. bicolor (Table 5).

Discussion The outcomes of ungrazed interations in agar and soil were similar to those reported in previous studies of the same species (Dowson et al. 1988). Importantly, outcomes on agar varied depending on strain (only single strains were studied on soil), H. fasciculare 1 being more combative than the other three strains. Such intraspecific variation in combativeness has also recently been reported in the non-cord-forming basidiomycete, Hericium coralloides (Crockatt et al. 2008). Thus, care must be taken when extrapolating to species based on results for single strains. This study provides the first report of effects of invertebrate grazing on the outcome of mycelial interactions. Effects, however, depended on grazer, F. candida altering both the outcome and progression of half of the mycelial interactions studied, while P. armata had almost no discernable effects on fungal interactions. Thus, outcomes of interactions in the field, where there are many different species of collembola and other invertebrates, may differ considerably from those in the simplified laboratory microcosms. P. armata is less active than F. candida, with a longer generation time. Thus, whilst an equivalent biomass of the two collembola species was used, the extent of grazer activity may have been lower with P. armata. Frequently the most abundant collembola species under natural conditions (Petersen & Luxton 1982), P. armata affects mycelium of other species of fungi (Hedlund et al. 1991), and may affect the basidiomycetes in the field, if density of grazers is higher. Previous studies have revealed effects of collembola grazing on extension rates of individual mycelia including both reduction and increase of extension rate (e.g. Bengtsson & Rundgren 1983; Kampichler et al. 2004; Bretherton et al. 2006;

Grazing by collembola affects the outcome of mycelial interactions

49

Fig 3 e Morphological and pigment changes during interspecific mycelial interactions in agar, either ungrazed or grazed by Folsomia candida. Hf, Hypholoma fasciculare; Pi, Phallus impudicus; Pv, Phanerochaete velutina; Rb, Resinicium bicolor. Numbers discriminate different strains. Images AeM are 9 cm diam, while in N and O the interaction zone is magnified (2 cm diam).

50

T.D. Rotheray et al.

Table 3 e Outcomes of interactions between extra-resource mycelia on the surface of soil after 50e56 d Fungal species

R. bicolor (Rb1) P. velutina P. impudicus

H. fasciculare (Hf1)

H. fasciculare (Hf1)

P. impudicus

P. velutina

Ungrazed

Grazed by F. candida

Ungrazed

Grazed by P. armata

Ungrazed

Grazed by F. candida

Ungrazed

Grazed by F. candida

P 100 o 70 M 30 o 90 M 10

P 100 o 100

P 100 e

P 100 e

P 100 M 100

O 100 M 100

O 100

M 100

o 100

o 100

o 100

Outcomes with upper case letters were mutual overgrowth (M), partial overgrowth (P), total overgrowth (O), by the fungus at the head of the column, whereas lowercase signifies overgowth by the fungus listed in the row. Numbers refer to percentage of replicates exhibiting a given response. There were 8e20 replicates/interaction.

Tordoff et al. 2006, 2008). In the present study, there were several examples of increases and one example of decrease in extension rate in grazed systems of interspecifically interacting mycelia. The effect varied depending on where the mycelium was growing (i.e. across soil or over the opponent mycelium). For example, grazing by F. candida accelerated P. velutina extension rate when the mycelium was growing over H. fasciculare, compared to growing over soil. Differences in extension rate in different mycelial regions have previously been observed: Mortierella isabellina, for example, exhibited accelerated growth away from the region of grazing in areas where collembola (P. armata) were excluded (Hedlund et al. 1991). The accelerated growth of mycelium across an opponent mycelium only occurred in P. velutina, and only when interacting with H. fasciculare. This suggests that the interacting fungi respond differentially to one another and that grazing is a further complicating factor. In contrast, grazing by F. candida on R. bicolor had a negative impact on mycelial extension rate. The frequent severing of R. bicolor cords by F. candida in all interactions suggests that R. bicolor was a preferred food source. How R. bicolor survives in the field when it is so heavily damaged by F. candida in microcosm studies remains an unanswered question. The microcosms lacked the presence of collembola predators but, in the field, high population densities of F. candida may attract predators such as Acarina, Coleoptera, such as Staphylinidae and Carabidae, thus limiting mycelial damage through grazer population control (Hopkin 1997). In coniferous forests, in which it is more commonly found (Kirby et al. 1990; Woods et al. 2006), R. bicolor cords may be less palatable as mycelium may contain inhibitory compounds such as polyphenols and terpenes (Rayner & Boddy 1988; Bardgett 2005). Although studies suggesting that fungi utilise resource-derived feeding deterrents are scarce, changes in substrate can lead to substantial changes in fungal emissions of volatile organic compounds (Bruce et al. 2000). Thus, the type of wood colonised may play a role in enabling fungi to withstand grazing. P. impudicus and H. fasciculare were usually resilient to grazing with few, if any, morphological differences between the mycelia of grazed and ungrazed interactions. This contrasts with H. fasciculare growing singly where extension rate was reduced (Harold et al. 2005; Tordoff et al. 2006) and there were small regions of intense grazing (Kampichler et al. 2004). Provision of food choice may have minimised the impact of collembola on H. fasciculare except when interacting with P. velutina. Here, H. fasciculare mycelium was completely removed by F. candida.

Such complete removal has not been seen with H. fasciculare growing alone (Kampichler et al. 2004; Harold et al. 2005) suggesting that the interaction with P. velutina may have altered the palatability of H. fasciculare. Initially, H. fasciculare was only grazed in areas where P. velutina had overgrown it. This further suggests that interacting with P. velutina increased the palatability of H. fasciculare but, as grazing took place through P. velutina mycelium, that there was active searching for H. fasciculare mycelium by the collembola. Collembola prefer cords of low vitality in some fungi (Kaneda & Kaneko 2004), although preference for actively growing hyphae has also been reported in other species (Moore et al. 1985). Though intense grazing of mycelia might be expected to impact negatively on the outcome of interspecific mycelial interactions, R. bicolor retained possession of more wood block resources when extra-resource mycelia were destroyed than when ungrazed during interactions with both H. fasciculare and P. velutina. This may, however, reflect reduced combativeness of the opponent when grazed rather than a directly positive effect of grazing on R. bicolor. Grazing on one interacting species may also have an indirect effect on the combativeness of the opponent fungus. For example, the accelerated growth of P. velutina when grazed limited the ability of H. fasciculare to overgrow the P. velutina mycelium. With the exception of the interaction with P. velutina, H. fasciculare was generally tolerant of F. candida grazing. This may provide a niche advantage, allowing H. fasciculare to proliferate in areas of high mycophagy, where more palatable species would be unable to compete. Collembola densities in soil can be very high (Hopkin 1997) but H. fasciculare is ubiquitous in forest soils (Rayner & Boddy 1988). Whether the abundance of H. fasciculare can be attributed, even in part, to grazing tolerance is yet to be confirmed. Previous studies have noted that site of grazing within the mycelium depends on fungal species (Tordoff et al. 2006). With interacting mycelia additional ‘types’ of grazing site are available. In particular, the interaction zone comprises dead and dying hyphae, leaked hyphal content, pigments, volatile and diffusible compounds, and mycelium with different morphology (e.g. Boddy 2000; Wells & Boddy 2002; Evans et al. 2008; Woodward & Boddy 2008). Such sites may be nutrient rich (Wells & Boddy 2002) and contain attractive or inhibitory compounds. Grazing of H. fasciculare beneath P. velutina has already been noted. There was also concentrated grazing in the interaction zone with P. impudicus though this did not alter the

Grazing by collembola affects the outcome of mycelial interactions

51

Fig 4 e Mycelial morphology on soil during interspecific interactions in (B, F, J, K) ungrazed systems, systems grazed by (A, C, E, G, H, I) Folsomia candida or by (D) Protophorura armata. Hf, Hypholoma fasciculare; Pi, Phallus impudicus; Pv, Phanerochaete velutina; Rb, Resinicium bicolor. Scale bar (10 cm) on image H applies to images AeH. Scale bar (2 cm) on image I applies to image I alone. Images J and K show pigment and exudate production respectively during interactions.

outcome of any interaction. The apparent increased density of hyphae and networking (Rotheray et al. 2008) at the mycelial margin in P. impudicus mycelium does, however, indicate that it was not entirely unaffected. The resilience of P. impudicus to

F. candida grazing when growing alone has been established (Tordoff et al. 2006) although, unlike H. fasciculare, P. impudicus always remained resilient when interacting with other mycelia. Deadlock in all interactions indicates that, relative to the

52

T.D. Rotheray et al.

4 2

Radial extension (cm)

A

B

Overall extension with H. fasciculare

Mycelial extension toward opponent

4 2

0 -2 -4 -6

0 -2 -4

Grazed Ungrazed

-8 -10 0

4

8

12

16

20

24

28

32

-6

Grazed

-8

Ungrazed

-10

36

0 4 Radial extension (cm)

C

D

Overall extension with P. impudicus

12

16

20

24

28

32

36

2

0

0

-2

-2

-4

-4

-6

Grazed

-6

Grazed

-8

Ungrazed

-8

Ungrazed

-10 0

Radial extension (cm)

8

Mycelial extension toward opponent

4

2

-10

E

4

4

8

12

16

20

24

28

F

Overall extension with P. velutina

2

0

32

4

8

12

16

20

24

28

32

Mycelial extension toward opponent

2 1

1 0

0

-1

-1

-2

Grazed

-2

Grazed

-3

Ungrazed

-3

Ungrazed

-4

-4 0

2

4 Day

6

8

0

2

4

6

8

Day

Fig 5 e Mycelial radial extent over time on soil for Resinicium bicolor interacting with (A, B) Hypholoma fasciculare, (C, D) Phallus impudicus and (E, F) Phanerochaete velutina, in ungrazed systems (closed symbols) and systems grazed by Folsomia candida (closed symbols). (A, C, E) are mean ± standard error of the mean of all four (0 , 30 , 60 , 90 ) measurements (Fig 1), and (B, D, F) are the mean of 30 and 60 measurements. Resinicium bicolor interacting with H. fasciculare (A) Scheirer Ray Hare c2 [ 0.999, P < 0.001; (B) Two-way ANOVA on ranked data, F3,88 [ 15.55, P < 0.001; R. bicolor interacting with P. impudicus (C) RM ANOVA F2.859,74.345 [ 4.803, P [ 0.005; (D) RM ANOVA F3.013,36.154 [ 6.461, P < 0.001; R. bicolor interacting with P. velutina, (E) SRH c2 [ 0.762, P [ 0.238; (F) Two-way ANOVA on ranked data, F3,104 [ 2.14, P [ 0.1; critical values are for time 3 treatment interaction.

Table 4 e Analysis of mycelial extent in grazed and ungrazed systems. Data are presented for each interaction combination with comparisons carried out for each species in all interactions H. fasciculare (Hf1) grazed by F. candida

R. bicolor (Rb 1) P. velutina P. impudicus

H. fasciculare (Hf1) grazed by P. armata

P. impudicus grazed by F. candida

P. velutina grazed by F. candida

Species in column

Species in row

Species in column

Species in row

Species in column

Species in row

Species in column

Species in row

t22 ¼ 1.08 P ¼ 0.291 W ¼ 375 P ¼ 0.351 W ¼ 442.5 P ¼ 0.387

t22 ¼ L6.14 P < 0.001 t40 ¼ 2.16 P ¼ 0.037 t35 ¼ 0.034 P ¼ 0.734

t26 ¼ 0.61 P ¼ 0.549 e

W ¼ 200 P ¼ 0.9085 e

t22 ¼ 1.08 P ¼ 0.291 t36 ¼ 0.74 P ¼ 0.463

t12 [ L3.61 P [ 0.004 t36 [ L4.65 P < 0.001

t26 ¼ 1.32 P ¼ 0.1

t26 [ L3.07 P [ 0.005

W ¼ 85 P ¼ 0.398

t16 ¼ 0.61 P ¼ 0.55

Where data were not normal, a ManneWhitney U-test (W) was applied. The change in mycelial extent was measured from the time of collembola addition (t0) to the point at which one of the interacting mycelia reached the wood block of the opposing fungus (treach). Figures in bold are significant values at P  0.05.

Grazing by collembola affects the outcome of mycelial interactions

A Radial extension (cm)

8

2

0

0

Radial entension (cm)

2

6

Radial extension (cm)

6

8

0

6

6

8

Grazed Ungrazed

4

4

2

2

0

0 2

4

6

6

0

8

2

4

F8

Mycelial extent over soil

6

8

6

8

6

8

Mycelial extent over soil

6

Grazed

Grazed

Ungrazed

Ungrazed

4

4

2

2

0

0 2

4

6

8

Mycelial extent in grazed trays

6

Over soil

8

0

4

2

H8

Over soil Over opponent

4

2

4

Mycelial extent in grazed trays

6

Over opponent

2 0 0

Radial extension (cm)

4

Mycelial extent over opponent

Ungrazed

0

I

2

D8

Grazed

8

Radial extesnion (cm)

4

Mycelial extent over opponent

8

0

G

4

2

0

E

Ungrazed

Ungrazed

4

Mycelial extent of P. velutina in interactions with P. impudicus Grazed

6

Grazed

0

C

B8

Mycelial extent of P. velutina in interactions with H. fasciculare

6

53

2

4

6

8

J

Mycelial extent in ungrazed trays

8 6

2

8

Over opponent

4

Mycelial extent in ungrazed trays

6

Over soil

4

0

Over soil Over opponent

4 2

2 0

0 0

2

4 Day

6

8

0

2

4 Day

6

8

Fig 6 e Mycelial radial extent over time on soil for Phanerochaete velutina interacting with (A, C, E, G, I) Hypholoma fasciculare and (B, D, F, H, J) Phallus impudicus. Measurements (see Fig 1) are mean ± standard error of the mean of: (A, B) mycelial extent in grazed and ungrazed systems across all four extension measurements (Fig 1) (A, F2.443,92.850 [ 2.147, P [ 0.112; B, F2.051,151.785 [ 24.451, P < 0.001); (C, D) mycelial extent in grazed and ungrazed systems over opponent (C, F2.443,92.850 [ 2.147, P [ 0.112; D, F2.125,84.981 [ 5.611, P [ 0.004); (E, F) mycelial extent in grazed and ungrazed systems over soil (E, F4,140 [ 212.849, P < 0.001; F, F2.534,73.490 [ 25.626, P < 0.001); (G, H) mycelial extent over the soil and over the opposing mycelium within grazed systems (G, RM ANOVA F2.601,46.819 [ 22.264, P < 0.001; H, F3,51 [ 3.651, P [ 0.018); (I, J) mycelial extent over soil and over the opposing mycelium within ungrazed systems (I, F3.141,56.541 [ 32.797, P < 0.001; J, F1.751,26.262 [ 9.835, P < 0.001). All reported critical values are for RM ANOVA time 3 treatment interaction.

54

T.D. Rotheray et al.

Table 5 e Outcomes of interactions after 80e84 d from wood block reisolations Fungal species

H. fasciculare (Hf1)

H. fasciculare (Hf1)

P. impudicus

P. velutina

Ungrazed

Grazed by F. candida

Ungrazed

Grazed by P. armata

Ungrazed

Grazed by F. candida

Ungrazed

Grazed by F. candida

R. bicolor (Rb1)

P 40 O 60

P 20 O 80

R 29 o 71

O 100

O 100

R 20 P 20 O 60

O 100

P. velutina

r 70 R 10 P 10 O 10 o 100

r 100

R 42 P 29 O 29 e

e

M 100

M 100

o 100

o 100

o 100

P. impudicus

Outcomes were mutual overgrowth (M), partial replacement (P), total replacement (R) and overgrowth (O). For outcomes designated PR, R and O, uppercase signifies replacement was by the fungus listed at the head of the column whereas lowercase signifies replacement by the fungus listed in the row. Numbers refer to percentage of replicates exhibiting a given response. There were four to 10 replicates per interaction treatment.

species against which it was paired, P. impudicus was effective in defence but not in attack. This contrasts with a previous study (Dowson et al. 1988) where P. impudicus was a poor competitor against a similar range of species. The species and strains used in the present study were, however, different, the microcosms larger and the experimental period shorter.

Conclusions Saprotrophic fungi play a critical role in ecosystem function making recalcitrant nutrients available for continued plant primary productivity. These nutrients tend to be retained in mycelia, but are probably released during mycelial interactions (J.M. Wells & L. Boddy unpub.). Further, invertebrate grazing of the microbial communities (including fungi) can increase the rate of carbon mineralisation (Bardgett et al. 1993; Cole et al. 2000). The present study indicated that collembola may also play a role in determining fungal species combativeness during interactions with potential effects for species assemblages and diversity, and consequently may also affect decomposition processes. The fungal response to collembola grazing during interspecific interactions appears to vary depending on the combination of grazer species and the identity of the interacting fungi. Grazing clearly adds another variable affecting the outcome of interspecific mycelial interations to those already known, for example, abiotic environmental conditions (Boddy 2000), inoculum size (Holmer & Stenlid 1993, 1997) and species composition (Holmer & Stenlid 1993; Boddy 2000; Wald et al. 2004).

Acknowledgements We thank the Natural Environment Research Council for provision of a studentship (TDR), and the Cardiff University Mycology Research Group for useful comments on the manuscript.

Supplementary data Supplementary data associated with this article can be found in the online version at doi:10.1016/j.funeco.2010.09.001.

references

Baldrian P, 2004. Increase of laccase activity during interspecific interactions of white-rot fungi. FEMS Microbiology Ecology 50: 245e253. Bardgett RD, 2005. The Biology of Soil. Oxford University Press, Oxford. Bardgett RD, Whittaker JB, Frankland JC, 1993. The diet and food preferences of Onychiurus procampatus (Collembola) from upland grassland soils. Biology and Fertility of Soils 16: 296e298. Bengtsson G, Rundgren S, 1983. Respiration and growth of a fungus, Mortierella isabellina, in response to grazing by Onychiurus armatus (Collembola). Soil Biology & Biochemistry 15: 469e473. Boddy L, 1999. Saprotrophic cord-forming fungi: meeting the challenge of heterogeneous environments. Mycologia 91: 13e32. Boddy L, 2000. Interspecific combative interactions between wood-decaying basidiomycetes. FEMS Microbiology Ecology 31: 185e194. Boddy L, Coates D, Rayner ADM, 1983. Attraction of fungus gnats to zones of intraspecific antagonism on agar plates. Transactions of the British Mycological Society 81: 149e151. Boddy L, Jones TH, 2008. Interactions between Basidiomycota and invertebrates. In: Boddy L, Frankland JC, Van West P (eds), Ecology of Saprotrophic Basidiomycetes. Academic Press, Amsterdam, pp. 209e235. Bretherton S, Tordoff GM, Jones TH, Boddy L, 2006. Compensatory growth of Phanerochaete velutina mycelial systems grazed by Folsomia candida (Collembola). FEMS Microbiology Ecology 58: 33e44. Bruce A, Wheatley RE, Humphris SN, Hackett CA, Florence MEJ, 2000. Production of volatile organic compounds by Trichoderma in media containing different acids and their effect on selected wood decay fungi. Holzforschung 54: 481e486. Chapela IH, Boddy L, Rayner ADM, 1988. Structure and development of fungal communities in beech logs four and a half years after felling. FEMS Microbiology Ecology 53: 59e70. Cole L, Bardgett RD, Ineson P, 2000. Enchytraeid worms (Oligochaeta) enhance mineralization of carbon in organic upland soils. European Journal of Soil Science 51: 185e192. Cooke RC, Rayner ADM, 1984. Ecology of Saprotrophic Fungi. Longman, London. Crockatt ME, Pierce GI, Camden RA, Newell PM, Boddy L, 2008. Homokaryons are more combative than heterokaryons of Hericium coralloides. Fungal Ecology 1: 40e48. Dicke M, 1994. Local and systemic production of volatile herbivore-induced terpenoids e their role in plant-carnivore mutualism. Journal of Plant Physiology 143: 465e472.

Grazing by collembola affects the outcome of mycelial interactions

Dowson CG, Rayner ADM, Boddy L, 1988. The form and outcome of mycelial interactions involving cord-forming decomposer basidiomycetes in homogenous and heterogenous environments. New Phytologist 109: 423e432. Evans JA, Eyre CA, Rogers HJ, Boddy L, Muller CT, 2008. Changes in volatile production during inter-specific interactions between four wood rotting fungi growing in artificial media. Fungal Ecology 1: 57e68. Griffith GS, Rayner ADM, Wildman HG, 1994. Interspecific interactions, mycelial morphogenesis and extracellular metabolite production in Phlebis radiata (Aphyllophorales). Nova Hedwigia 59: 331e344. Guevara R, Rayner ADM, Reynolds SE, 2000. Orientation of specialist and generalist fungivorous ciid beetles to host and non-host odours. Physiological Entomology 25: 288e295. Harold S, Tordoff GM, Jones TH, Boddy L, 2005. Mycelial responses of Hypholoma fasciculare to collembola grazing: effect of inoculum age, nutrient status and resource quality. Mycological Research 109: 927e935. Hedlund K, Boddy L, Preston CM, 1991. Mycelial responses of the soil fungus, Mortierella isabellina, to grazing by Onychiurus armatus (Collembola). Soil Biology & Biochemistry 23: 361e366. Hedlund K, Rayner ADM, Reynolds SE, 1995. Fungal odour discrimination in two sympative species of fungivorous collembolans. Functional Ecology 9: 869e875. Holmer L, Stenlid J, 1993. The importance of inoculum size for the competitive ability of wood decomposing fungi. FEMS Microbiology Ecology 12: 169e176. Holmer L, Stenlid J, 1997. Competitive hierarchies of wood decomposing basidiomycetes in artificial systems based on variable inoculum sizes. Oikos 79: 77e84. Hopkin SP, 1997. Biology of the Springtails. Oxford University Press, Oxford. Hynes J, Muller CT, Jones TH, Boddy L, 2007. Changes in volatile production during the course of fungal mycelial interactions between Hypholoma fasciculare and Resinicium bicolor. Journal of Chemical Ecology 33: 43e57. Kampichler C, Rolschewski J, Donnelly DP, Boddy L, 2004. Collembolan grazing affects the growth strategy of the cord-forming fungus Hypholoma fasciculare. Soil Biology & Biochemistry 36: 591e599. Kaneda S, Kaneko N, 2004. The feeding preference of a collembolan (Folsomia candida Willem) on ectomycorrhiza (Pisolithus tinctorius (Pers.)) varies with mycelial growth condition and vitality. Applied Soil Ecology 27: 1e5. Kirby JJH, Stenlid J, Holdenrieder O, 1990. Population structure and responses to disturbance of the basidiomycete Resinicium bicolor. FEMS Microbiology Ecology 36: 591e599. Moore JC, St. John TV, Coleman DC, 1985. Ingestion of vesiculararbuscular mycorrhizal hyphae and spores by soil microarthropods. Ecology 66: 1979e1981. Petersen H, Luxton M, 1982. A comparative analysis of soil fauna populations and their role in decomposition processes. Oikos 39: 287e388.

55

Philp RW, Bruce A, Munro AG, 1995. The effect of water soluble Scots pine (Pinus sylvestris L.) and Sitka spruce [Picea sitchensis (Bong.) Carr.] heartwood and sapwood extracts on the growth of selected Trichoderma species. International Biodeterioration and Biodegradation 35: 355e367. Rayner ADM, Boddy L, 1988. Fungal Decomposition of Wood: its biology and ecology. John Wiley and Sons, Chichester. Rotheray TD, Jones TH, Fricker MD, Boddy L, 2008. Grazing alters network architecture during interspecific mycelial interactions. Fungal Ecology 1: 124e132. Ruess L, Zapata EJG, Dighton J, 2000. Food preferences of a fungalfeeding Aphelenchoides species. Nematology 2: 223e230. Swift MJ, Boddy L, 1984. Animal microbial interactions during wood decomposition. In: Anderson JM, Rayner ADM (eds), Invertebrate Microbe Interactions. Cambridge University Press, Cambridge, pp. 89e131. Tentelier C, Fauvergue X, 2007. Herbivore-induced plant volatiles as cues for habitat assessment by a foraging parasitoid. Journal of Animal Ecology 76: 1e8. Tordoff GM, Boddy L, Jones TH, 2006. Grazing by Folsomia candida (Collembola) differentially affects mycelial morphology of the cord-forming basidiomycetes Hypholoma fasciculare, Phanerochaete velutina and Resinicium bicolor. Mycological Research 110: 335e345. Tordoff GM, Boddy L, Jones TH, 2008. Species-specific impacts of collembola grazing on fungal foraging ecology. Soil Biology & Biochemistry 40: 434e442. Vainio EJ, Lipponen K, Hantula J, 2001. Persistence of a biocontrol strain of Phlebiopis giganten in conifer stumps and its effect on within-species genetic diversity. Forest Pathology 31: 285e295. Wald P, Pitkanen S, Boddy L, 2004. Interspecific interactions between the rare tooth fungi Creolophus cirrhatus, Hericium erinaceus and H. coralloides and other wood decay species in agar and wood. Mycological Research 108: 1447e1457. Walker GM, McCleod AH, Hodgson VJ, 1995. Interactions between killer yeasts and pathogenic fungi. FEMS Microbiology Letters 127: 213e222. Wells JM, Boddy L, 2002. Interspecific carbon exchange and cost of interactions between basidiomycete mycelia in soil and wood. Functional Ecology 16: 153e161. Wheatley RE, Hackett C, Bruce A, Kundzweicz A, 1997. Effect of substrate composition on production of volatile organic compounds from Trichoderma spp. inhibitory to wood decay fungi. International Biodeterioration and Biodegradation 39: 199e205. Woods CM, Woodward S, Pinard MA, Redfern DB, 2006. Colonisation of Sitka spruce stumps by decay-causing hymenomycetes in paired inoculations. Mycological Research 110: 854e868. Woods CM, Woodward S, Redern DB, 2005. In vitro interactions in artificial and wood-based media between fungi colonizing stumps of Sitka spruce. Forest Pathology 35: 213e229. Woodward S, Boddy L, 2008. Interactions between saprotrophic fungi. In: Boddy L, Frankland JC, van West P (eds), Ecology of Saprotrophic Basidiomycetes. Elsevier, Amsterdam, pp. 125e141.