Xylobolus subpileatus, a specialized basidiomycete functionally linked to old canopy gaps

Abstract

Documenting succession in forest canopy gaps provides insights into the ecological processes governing the temporal dynamics of species within communities. We analyzed the fruiting patterns of a rare but widely distributed saproxylic macromycete, Xylobolus subpileatus, during the ageing of natural canopy gaps in oak forests. In one of the last remaining Quercus ilex L. old-growth forests (on the island of Corsica, western Mediterranean basin), we systematically recorded and conducted molecular analyses of X. subpileatus basidiomes in 80 dated natural canopy gaps representing a 45-year long sequence of residence time of tree logs on the forest floor. Xylobolus subpileatus fruited exclusively on Q. ilex logs. The probability of fruiting of X. subpileatus significantly increases during the process of wood decomposition to reach its maximum in the oldest gaps, approximately 40 years after treefall. In contrast, the abundance and the richness of saprobic and ectomycorrhizal fruitbodies decrease as canopy gaps age. Our results emphasize the high ecological specialization of X. subpileatus. They also highlight the imperative need to conserve the last patches of old-growth Mediterranean forests to secure the persistence of this endangered and functionally unique macromycete whose presence is highly dependent on old wood in advanced stages of decomposition.

Full text

1
Xylobolus subpileatus, a specialized basidiomycete
1
functionally linked to old canopy gaps
2
3
A. Taudiere
1,2
, J.-M. Bellanger
1
, P.-A. Moreau
3
, C. Carcaillet
2,4
, A. Christophe
1
, T. Læssøe
5
,
4
C. Panaïotis
6
and F. Richard
1
5
6
Affiliation and Address:
7
1/ UMR 5175 CEFE – CNRS - Université de Montpellier - Université Paul Valéry Montpellier -
8
EPHE - INSERM, Campus CNRS, 1919 Route de Mende, F-34293 Montpellier, France
9
2/ École Pratique des Hautes Études (EPHE) and PSL Research University Paris, France
10
3/ Département de Botanique, Faculté des Sciences Pharmaceutiques et Biologiques, Université
11
Lille - Nord de France B. P. 83, 59006 Lille Cedex, France.
12
4/ Laboratoire d'Écologie des Hydrosystèmes Naturels et Anthropisés (UMR5023 CNRS),
13
Université Lyon 1, 69622 Villeurbanne, France
14
5/ Natural History Museum of Denmark/Department of Biology, Universitetsparken 15, København
15
Ø, Denmark.
16
6/ Office de l’Environnement de la Corse, Conservatoire Botanique National de Corse. Avenue Jean
17
Nicoli, 20250 Corte, France.
18
* Corresponding authors: adrien.taudiere@cefe.cnrs.fr; franck.richard@cefe.cnrs.fr
19
Tel: +33 (0)4.67.61.32.62 Fax: +33 (0)4.67.61.41.21.38
20
Page 1 of 31

2
ABSTRACT: Documenting succession in forest canopy gaps provides insights into the ecological
21
processes governing the temporal dynamics of species within communities. We analyzed the
22
fruiting patterns of a rare but widely distributed saproxylic macromycete, Xylobolus subpileatus,
23
during the ageing of natural canopy gaps in oak forests. In one of the last remaining Quercus ilex
24
old-growth forests (on the island of Corsica, western Mediterranean basin), we systematically
25
recorded and conducted molecular analyses of X. subpileatus basidiomes in 80 dated natural canopy
26
gaps representing a 45 year-long sequence of residence time of tree logs on the forest floor.
27
X. subpileatus fruited exclusively on Q. ilex logs. The probability of fruiting of X. subpileatus
28
significantly increases during the process of wood decomposition to reach its maximum in the
29
oldest gaps, ca. 40 years after treefall. In contrast, the abundance and the richness of saprobic and
30
ectomycorrhizal fruitbodies decrease as canopy gaps age. Our results emphasize the high ecological
31
specialization of X. subpileatus. They also highlight the imperative need to conserve the last patches
32
of old-growth Mediterranean forests to secure the persistence of endangered and functionally
33
unique macromycetes whose presence is highly dependent on old wood in advanced stages of
34
decomposition.
35
Keywords: old-growth stand, Mediterranean forest, coarse woody debris, natural canopy gaps,
36
conservation
37
38
Page 2 of 31

3
RÉSUMÉ : L'étude des successions écologiques ayant lieu dans les ouvertures de canopée
39
forestière offre l'opportunité de mieux comprendre les processus écologiques gouvernant la
40
dynamique temporelle des espèces à l’intérieur des communautés. Nous avons analysé la
41
distribution des fructifications d'une espèce – rare mais largement distribuée – de macromycète
42
saproxilique, Xylobolus subpileatus, au cours du vieillissement de trouées naturelles dans les forêts
43
de chênes. Dans l'une des dernières forêts anciennes de chêne verts (Quercus ilex ; île de Corse,
44
ouest du bassin méditerranéen), nous avons échantillonné systématiquement et analysé
45
moléculairement les basidiomes de X. subpileatus dans 80 trouées datées qui représentent une
46
chronoséquence de 45 ans de temps de résidence au sol des troncs d'arbres. X. subpileatus ne
47
fructifie que sur les troncs de chêne vert (Q. ilex). Sa probabilité de fructification est positivement
48
corrélée à l'âge des trouées et atteint un maximum dans les plus vieilles trouées (environ 40 ans
49
après la chute de l'arbre). A contrario, l'abondance et la richesse des fructifications de champignons
50
décomposeurs et ectomycorhiziens décroissent avec le vieillissement de la trouée. Nos résultats
51
soulignent la spécialisation écologique poussée de X. subpileatus et mettent en avant le besoin
52
impératif de préserver les dernières parcelles de forêts anciennes méditerranéennes afin de protéger
53
cette espèce de macromycète fonctionnellement originale dont la présence dépend fortement de la
54
complétion du processus naturel de décomposition.
55
Mots clés: parcelle forestière ancienne, forêt méditerranéenne, débris ligneux grossiers, ouverture
56
naturelle de canopée, conservation
57
58
Page 3 of 31

4
INTRODUCTION
59
Identifying the factors that determine the tremendous concentration of biological diversity in
60
old-growth forests is a pivotal question for conservation ecologists, but also an inspiration for forest
61
managers to implement integrated management in cultivated ecosystems. During the past decades,
62
this issue has particularly interested mycologists, who have documented the diversity and
63
specificity of fungal guilds characterizing old-growth forests (e.g. Smith et al. 2002 and Junninen et
64
al. 2006 for ectomycorrhizal and wood-decaying fungi, respectively). Mycologists have also
65
addressed the broader issue of historical legacies of human practices on the diversity of fungal
66
communities in forests. In this regard, recent studies showed that anthropic disturbances alter the
67
dynamics of fungal communities at short (e.g. Paillet et al. 2010, Halme et al. 2013) and longtime
68
scales (Diedhiou et al. 2009).
69
Whatever ecological guild (plant mutualists, decayers) is considered, old-growth forests
70
accommodate particularly rich fungal diversity in temperate and boreal biomes. From a
71
belowground perspective, these ecosystems are dominated by long-lived ectomycorrhizal (hereafter
72
ECM) tree species (e.g. Fagaceae, Pinaceae) that accumulate species-rich (Goodman and Trofymow
73
1998) and dynamic (Izzo et al. 2005) ECM fungal communities, which tend to differ in composition
74
from those in managed ecosystems (Smith et al. 2002). Furthermore, because they accumulate
75
large-diameter woody debris in all stages of decay, old-growth forests are considered as unique in
76
the amount of suitable habitats for rich communities of wood-inhabiting fungi (Brazee et al. 2014),
77
including species that are infrequent (Stokland and Larsson 2011) or particularly demanding in
78
terms of their ecological requirements (Berglund et al. 2011).
79
In old-growth forests, biotic (e.g., insect outbreaks, root-pathogenic fungi, dwarf mistletoe; Worrall
80
et al. 2005) and abiotic (e.g., wind and snow; McCarthy et al. 2001) disturbances provoke canopy
81
openings of various sizes, by causing the fall of trees, singly or in groups (review: McCarthy et al.
82
2001). The physiognomy of forest ecosystems driven by canopy gap dynamics is thus typically
83
Page 4 of 31

5
characterized by complex mosaics of small-scale patches of vegetation that are in different stages of
84
response to disturbance (Worrall et al. 2005; Oldeman 2012). From its creation to its complete
85
closure, a canopy gap is the scene of marked shifts in abiotic conditions (light availability,
86
temperature and moisture; Rajala et al. 2012). These changes, taken together with the existence of
87
contrasted life-history strategies of fungi adapted to the different stages of wood decomposition,
88
drive the succession of organisms within gaps, from early- to late-successional species (Jönsson et
89
al. 2008). For mycorrhizal fungal guilds, gap-driven dynamics have so far been poorly investigated.
90
In contrast, the succession of saproxylic fungi on fallen logs has been accurately characterized,
91
particularly in boreal and temperate forests. During the process of wood decay, composition shifts
92
occur within these communities, from early- to late-successional assemblages successively
93
dominated by ascomycete and basidiomycete species (e.g. Lindblad 1998; Rajala et al. 2012), with
94
succession scenarios partially determined by log size and architecture (Heilmann-Clausen and
95
Christensen 2003).
96
The wood-inhabiting stereoid basidiomycete Xylobolus subpileatus (Fig. 1a) is a widely distributed
97
macromycete recorded from 14 countries from all continents but Africa and Antarctica (Global
98
Biodiversity Information Facility, 12 July 2016). This fungus is rare throughout Europe, and in
99
some countries is considered by mycologists to be endangered (Austria, Hongrie; Papp 2011) or
100
critically endangered (Czech Republic; Kotlaba 1986). Although knowledge about this rare species
101
is still fragmentary, it is known to be ecologically specialized. First, X. subpileatus has been
102
recorded only on dead oak wood (Quercus spp.), with a marked preference for trees long dead
103
(Long 1915; Parmasto 2001). Second, X. subpileatus selectively hydrolyses lignin, provoking
104
characteristic heart-rotting honeycomb-like patterns of decay (Fig. 1b), where medullary ray
105
parenchyma and early-wood vessels are not readily decayed and remain between pockets of
106
degraded material (Blanchette 1984). The ecological specialization of X. subpileatus, and the
107
Page 5 of 31

6
consequences of this specialization for its fine-scale spatiotemporal distribution in forest
108
ecosystems, have never been investigated.
109
In the Mediterranean basin, remaining primary vegetation covers less than 5% of its original extent
110
(Myers et al. 2000). Old-growth stands are extremely rare in this region because of millennia of
111
intensive and uninterrupted human use of forest (Quezel and Médail 2003). In the mountainous
112
island of Corsica (Fig. 2a), human activities over ca. 5 000 years have dramatically reduced both the
113
area and the quality of natural forests (Reille 1992) and likely altered the occurrence of wildfires
114
(Leys et al. 2014), which are the main large disturbance in Mediterranean forests. However, only
115
few small and quasi-inaccessible patches of old-growth forests persist, where both the long-term
116
continuity of forest cover and the absence of logging at the large scale of the complete forest cycle
117
(Panaïotis et al. 1998) are attested by historical maps and textual archives on forests. The most
118
emblematic old-growth forest, that found in the Fango (Fig. 2b), has been continuously monitored
119
for macromycetes since 1999 (Richard et al. 2004). In this exceptionally well-preserved Quercus
120
ilex old-growth stand, it is possible to accurately date forest canopy gaps using the understory shrub
121
Phillyrea latifolia as a disturbance marker based on its capacity to spontaneously re-sprout
122
immediately after the canopy opening (Panaïoitis et al. 1995). In this study, we took advantage of
123
the discovery of X. subpileatus in the Fango permanent transect during the British Mycological
124
Society field trip in Corsica (2013) to determine the ecological requirements of this species. Unique
125
opportunities for such a study are offered by this site, where a large variety of usable woody
126
substrates co-occur, originating from various species, with wood of various ages and dimensions
127
remaining on the forest floor until its complete decay.
128
Using a synchronic analysis performed on 80 natural canopy gaps of known age, we
129
described the fine-scale distribution of X. subpileatus basidiomes on the basis of (i) the identity of
130
the plant forming the colonized log, (ii) the dimensions of wood fragment and (iii) the age of the
131
corresponding canopy gap (time since treefall). We hypothesized that the ecological specialization
132
Page 6 of 31

7
of X. subpileatus results in preferential fruiting on Q. ilex, on the largest logs of this species and in
133
the oldest canopy gaps. We finally compared the temporal patterns of X. subpileatus fruiting with
134
those of ECM and saprobic fungal communities in the same gaps, using the dataset in Richard et al.
135
(2004). We predicted that the abundance and the richness of ECM and saprobic fruitbodies in gaps
136
decrease as gaps age, while X. subpileatus basidiomes show the opposite pattern. We aimed at
137
testing the role of canopy gaps, from their initiation until the very end of the process of succession,
138
in providing suitable habitat for highly specialized and rare organisms in forest ecosystems little
139
disturbed by humans.
140
MATERIAL AND METHODS
141
Study site
142
The study site is situated in the Fango forest (42°20′N; 8°49′E), in the northwestern part of the
143
island of Corsica (Fig. 2). The Fango valley has been a Man and Biosphere (MAB) reserve since
144
1973, as it contains rare stands of old-growth Quercus ilex (holm oak) forests (Panaïotis et al. 1995;
145
Quézel and Médail 2003). This forest covers a 4 318-ha area on Hercynian granite with enclaves of
146
volcanic rhyolites. Soils are alocrisols (Richard et al. 2009) with mull humus overlying (i) a thick
147
organic layer with a slightly acidic pH ranging from 5.7 to 6.4 and a C:N ratio ranging from 24 to
148
28 and, (ii) a poorly fissured rhyolitic bedrock (Richard et al. 2009). The climate is subhumid with a
149
mean annual rainfall of 750 mm and an average annual temperature of 14.6°C at 192 m above sea
150
level (asl). Temperatures range from 3.5°C (mean January minima) to 29.9°C (mean July maxima).
151
In Corsica, a high diversity of wood-decaying fungi develop in Q. ilex wood, and commonly fruit
152
on standing logs (e.g. Hyphodontia quercina, Inonotus dryadeus, Phellinus erectus), branches (e.g.
153
Abortiporus biennis, Hexagonia nitida, Meriulopsis corium, Polyporus cf. tuberaster) and fallen
154
logs (e.g. Hyphodontia quercina, Abortiporus biennis; Pieri and Rivoire 1992, 1994).
155
Characteristics of the Q. ilex old-growth stand
156
Page 7 of 31

8
This study was carried out within a district of about 1 500 ha, named Perticato, extending from 90
157
to 1 619 m asl (Capu a u Ghjallichiccia; Fig. 2b), and where a permanent transect was established in
158
1994, devoted to the study of forest dynamics (Panaïotis et al. 1995). The vegetation at the study
159
site is a complex mosaic of tall shrubby patches dominated by evergreen Mediterranean species
160
(‘macchia’) and of old-growth forest stands, mainly located in the lower parts of valleys. In forest
161
patches, large Q. ilex trees dominate a 7-m-high, species-poor and dense understory layer that
162
persists under the oak canopy. This understory is composed of ‘macchia’ species, i.e. Phillyrea
163
latifolia, Erica arborea, Arbutus unedo and scattered individuals of Fraxinus ornus, Cistus
164
salviifolius and C. monspeliensis (Richard et al. 2009).
165
Textual forest archives unambiguously attest that the district of Perticato has not been exploited for
166
wood and livestock (cows and goats) since 1827 (Panaïotis et al. 1995). For this reason, the long-
167
term expression of forest dynamics generated a high density of natural canopy gaps of about 100 m
2
168
each, which occur when old Q. ilex stems (170 ± 46 years) break and fall down (Panaïotis et al.
169
1997).
170
Dating of canopy gaps
171
Canopy gaps are small-scale disturbances involving either a part of a single tree (hereafter partial
172
canopy gaps), the entire canopy of a single tree (hereafter single canopy gaps), or falls of several
173
adjacent individuals (hereafter multiple canopy gaps; McCarthy et al. 2001). Our study included
174
falls of either trees (Q. ilex) or large shrub individuals (Phillyrea latifolia, Erica arborea, Arbutus
175
unedo and Fraxinus ornus).
176
The residence time of tree stems on the ground, i.e. the dating of fall events, was estimated using
177
the synchronic method developed by Panaïotis and coauthors (1995) based on the high sprouting
178
speed and rate of Phillyrea latifolia, which produces new aerial and basal stem sprouts during the
179
first spring after canopy opening. In each sampled canopy gap, we then accurately dated the age of
180
Page 8 of 31

9
Phillyrea latifolia stem sprouts to assess the date of the tree fall event (Panaïotis et al. 1995 and
181
Fig. 1c-e). In 11 of the 80 dated gaps, the dynamics of P. latifolia re-sprouting attested to a
182
succession of canopy-opening events and made uncertain the assignation of the different parts of
183
fragmented logs to a given event. All analyses were carried out with and without these 11 gaps
184
(Table S1 and Fig. S1). As the results are consistent whether or not these gaps are taken into
185
account, we included these gaps in the final analysis presented here, and their age was set using the
186
date of the most recent event recorded (conservative dating regarding the observed dynamics of
187
X. subpileatus).
188
Sampling of Xylobolus subpileatus basidiomes
189
The survey was conducted in April 2015 in a 25-ha area where a 6 400-m
2
(160 m × 40 m)
190
permanent transect was established in 1994 (Panaïotis et al. 1997; Fig. 1c-e and Fig. 2). All canopy
191
gaps present in the 25-ha area were visited and dated using the method of Panaïotis et al. (1995),
192
while the appearance of basidiomes of Xylobolus subpileatus and Stereum hirsutum was
193
systematically recorded. In each gap where X. subpileatus was suspected to be present, basidiomes
194
morphologically assigned to X. subpileatus were collected and stored for subsequent microscopic
195
description, molecular analyses and final identification. We also collected two S. hirsutum
196
specimens for molecular verification of their taxonomic identification.
197
In addition, in all Q. ilex canopy gaps located in the
permanent transect of 6 400 m
2
(Fig. 2c),
198
supplementary measurements were performed to refine the determination of the ecological
199
requirements of X subpileatus. On each Q. ilex stem, we measured the surface covered by
200
X. subpileatus basidiomes, and the dimensions (length, diameter at each tip, and volume) of the
201
stem.
202
Taxonomic assignment of X. subpileatus basidiomes
203
Page 9 of 31

10
In order to confirm the taxonomic identity and homogeneity of the collected fungal material, 16
204
voucher collections assigned to Xylobolus subpileatus were randomly selected out of 22 colonized
205
logs, microscopically analyzed, and compared to five reference collections assigned to genus
206
Xylobolus by expert field mycologists (T. Læssøe, Denmark; G. Trichies, France; J.-P. Vidonne,
207
France). Two voucher collections of the phylogenetically related species Stereum hirsutum were
208
also sampled for taxonomic verification and comparative purposes. All samples are kept in the
209
personal herbarium of J-M Bellanger at the CEFE (Montpellier, France) and one representative
210
collection of the Corsican X. subpileatus is deposited at the LIP herbarium (Faculté des Sciences
211
Pharmaceutiques et Biologiques, Université Lille 2, France), under the code LIP0401115.
212
All collections were then subjected to molecular-genetic analyses. Briefly, DNA was extracted and
213
the internal transcribed spacer (ITS) locus was amplified from dried specimens as described by
214
Loizides et al. (2016). PCR amplification and sequencing were performed with the primers ITS1F
215
and ITS4 or ITS4B (Gardes and Bruns 1993). We obtained 15 validated sequences (14/22 sequences
216
for X. subpileatus and 1/2 sequences of S. hirsutum). Sequences were deposited in GenBank under
217
the accession numbers listed in Table 1. The analyzed dataset includes these sequences as well as a
218
subset of publically available ITS sequences of Xylobolus spp. (i.e. X. subpileatus, X. frustulatus,
219
X. apricans and X. annosus) and other Stereaceae in GenBank, selected by BLAST. Phylogenetic
220
analyses were all performed online at www.phylogeny.lirmm.fr (Dereeper et al. 2008) and on the
221
CIPRES Science Gateway, as described in Loizides et al. (2016). The tree depicted in figure S1
222
corresponds to the 50% majority-rule consensus phylogram resulting from the Bayesian Inference
223
(BI) of phylogeny. A Maximum Likelihood (ML) analysis was also conducted. Branch supports
224
were assessed using the Bayesian posterior probabilities (BPP, as percentages) and the Shimodaira–
225
Hasegawa version of the approximate likelihood-ratio test (SH-aLRT), respectively.
226
Analysis of fruiting patterns in canopy gaps of the permanent transect
227
Page 10 of 31

11
To compare the temporal dynamics of X. subpileatus occurrence with the general fruiting patterns
228
of fungal communities in canopy gaps, we re-analyzed the dataset presented in Richard et al.
229
(2004). In the aforementioned study performed at the same site, basidiomes of soil macromycetes,
230
including decomposers and ectomycorrhizal fungi, had been mapped at a 0.1-m accuracy during the
231
period 1999-2003 in ten gaps included in the permanent transect (Richard et al. 2004). Here, we
232
used this dataset to describe the temporal dynamics of basidiome production and fungal species
233
richness in ten dated canopy gaps of the permanent transect. These gaps ranged from 3 to 23 years
234
in age. Briefly, basidiome abundances and species richness were calculated in a circle of 5 m in
235
radius around the fallen tree trunk where the area covered by X. subpileatus basidiomes was
236
assessed.
237
Statistical analyses
238
The effect of the age of canopy gaps on the presence of X. subpileatus and S. hirsutum basidiomes
239
was tested using a General Linear Model (function glm, R-Development Core Team, 2016).
240
Confidence intervals were calculated using the function confint.glm from the MASS R packages
241
(Venables and Ripley, 2002). Correlation between gap ages and basidiome richness and numbers
242
were tested using Pearson correlation tests.
243
RESULTS
244
Taxonomic assignment of analyzed basidiomes
245
The phylogenetic analysis of our ITS sequence dataset indicates that, as currently defined,
246
Xylobolus is polyphyletic, with three distinct lineages nested within Stereaceae (marked by asterisks
247
in Fig. S2): X. apricans, X. spectabilis, and a weakly supported clade encompassing almost all the
248
other Xylobolus sequences, including those of our two reference collections of S. hirsutum, type
249
species of that genus. The latter lineage is labeled “Xylobolus” in figure S1. However, a taxonomic
250
revision of the genus, clearly beyond the scope of the present work, would be necessary to stabilize
251
Page 11 of 31

12
names, species limits and phylogenetic boundaries in Xylobolus. Nevertheless, all 14 samples
252
attributed to Xylobolus that we collected in Corsica nested within that lineage, as a strongly
253
supported monophyletic subclade that was previously not represented in public databases.
254
Collections in this clade are 100% identical at the ITS locus, and include LIP0401115, a sample
255
collected in the same area during the 2013 annual foray of the British Mycological Society, and
256
taxonomically assigned to X. subpileatus by the specialist of our consortium T. Læssøe. Because
257
this species apparently displays a wide distribution and as no morpho-anatomical features observed
258
in Corsican collections contradict the original diagnosis, we thus apply this binomial to the
259
subclade, at least provisionally, even though no sequence from the North American type material is
260
currently available to support our interpretation. The analysis also attested the identity of two
261
collections of Stereum hirsutum (Fig. S2, AT44 and AT67), a common and widespread species with
262
broad ecological and morphological ranges, often confused with other Stereaceae species.
263
Characteristics of analyzed canopy gaps
264
In all, we sampled 86 canopy gaps of about 100 m² each (Panaïotis et al. 1995) within a forest patch
265
of 25 ha in area (Fig. 2c). Our sampling included 4, 48 and 34 multiple, single and partial multiple
266
treefall canopy gaps, respectively, with a large majority of gaps created by Q. ilex (n = 65, 76%).
267
The analysis also included canopy openings in ‘macchia’ patches, and mostly logs of Arbutus unedo
268
(11, 13%) and Erica arborea (8, 9%), with very few individuals of Phillyrea latifolia (1) and
269
Fraxinus ornus (1).
270
The date of treefall could be estimated in 80 (93%) of the 86 analyzed canopy gaps, of which 59
271
were created by Q. ilex treefalls. Our dataset included openings ranging from 1 (tree fallen in 2014)
272
to 45 years old in age, with 14.6 years on average (Fig. 1c-e and Fig. S3). The temporal distribution
273
of treefall events in the 25-ha area showed three main periods of gap creation corresponding to [1-
274
5], [12] and [30-40] year-old logs (see Fig. S3).
275
Page 12 of 31

13
Distribution of Xylobolus subpileatus on plant species
276
Basidiomes of X. subpileatus (Fig. 1a) were detected in 22/80 dated canopy gaps and only in gaps
277
made of Q. ilex logs (22/59). The typical honeycomb-like decomposition pattern of decayed wood
278
was frequently observed in the vicinity of basidiomes (Fig. 1b). Xylobolus subpileatus fruited in
279
gaps from 7- to 40-years old (mean: 23.5 ± 10.5; Fig. 3a). Basidiomes were found in 11% of canopy
280
gaps less than 20 years old, whereas 67% of gaps ranging from 20 to 45 years old contained logs
281
colonized by this species. At the end of the wood-decaying process (40-year-old canopy gaps),
282
X subpileatus basidiomes were almost systematically present (84% of sampled logs). The fruiting
283
probability significantly increased with age of gaps, gaining 7% of its value per year of log
284
residence on the ground (P < 0.05 by one-way ANOVA; Table 2), and ranged from 18.5 to 83.4% in
285
7- and 40-year-old gaps, respectively (Fig. 3a).
286
Within the permanent transect of 6 400m², we found 35 logs distributed across 16 Q. ilex-created
287
gaps. The analysis of the fine-scale distribution of basidiomes on Q. ilex logs revealed the presence
288
of the fungus on logs of diameters from 15 to 80 cm (median: 27 cm), of length from 1.25 to 14 m
289
(median: 5.91 m), and of volume from 0.02 to 2.36 m
3
(median: 0.21 m
3
; Table S2). The log volume
290
was not significantly correlated with the occurrence of X. subpileatus basidiomes (p = 0.84 using
291
glm) or with the surface covered by basidiomes (p = 0.72 using lm). On the contrary, both
292
occurrence (p = 0.037 using glm) and surface of X. subpileatus (p = 0.036 using lm) basidiomes
293
were correlated with the age of the gap. The surface covered by X. subpileatus basidiomes varied
294
from 0.13 to 1.01 m
2
(median: 0.20 m
2
, Fig. S4). The fungus covered from 0.24 to 18.25% (median:
295
7.00%) of the log surface (Table S2).
296
In contrast, basidiomes of Stereum hirsutum were recorded in Q. ilex logs in 6/80 canopy gaps of
297
recent origin (mean: 8.3 ± 3.1 years; Fig. 3b). By contrast with X. subpileatus, the probability of
298
S. hirsutum basidiome occurrence tended to decrease with increasing gap age (Fig. 3b). However,
299
Page 13 of 31

14
this tendency was not significant, probably owing to the low number of records of this species
300
(Table 2).
301
Dynamics of abundance and diversity of ectomycorrhizal and saprobic basidiomes
302
within gaps
303
In our analysis, the abundance of both ectomycorrhizal and saprobic basidiomes (dataset: cf.
304
Richard et al. 2004) in canopy openings significantly decreased with gap age (Fig. 4, all p < 0.05
305
using Pearson correlation tests). The production of basidiomes is high in the youngest canopy gaps
306
colonized by X. subpileatus (41.4 and 29.8 basidiomes per 100 m
2
for ECM and saprobic
307
macromycetes in 7-year-old gaps, respectively; Fig. 4). On the contrary, at the average residence
308
time of X. subpileatus (23.5 years), the production of basidiomes is very low (with 18 and 0.4
309
basidiomes per 100 m
2
for ECM and saprobic macromycetes in 23-year-old gaps, respectively;
310
Fig. 4a).
311
Ectomycorrhizal species richness markedly decreased during gap ageing (Fig. 4b). Relating to the
312
patterns of occurrence of X. subpileatus, species richness decreased from the youngest canopy gaps
313
colonized by X. subpileatus (with 20.9 and 7.3 species per 100 m
2
for ECM and saprobic
314
macromycetes in 7-years old gaps, respectively; Fig. 4b) to the average time of residence of
315
colonized logs (with 9.6 and 0.4 species per 100 m
2
for ECM and saprobic macromycetes in 23-
316
year-old gaps, respectively; Fig. 4b).
317
318
319
DISCUSSION
320
In this study, we used a unique experimental design to characterize the ecology of
321
X. subpileatus along a 40-year-long sequence of forest canopy gap dynamics. In accordance with
322
previous studies (Papp 2011), X. subpileatus fruited exclusively on oak logs in the Fango forest.
323
Page 14 of 31

15
Using the dating of canopy openings, we showed that the fruiting of X. subpileatus was tightly
324
correlated with the age of gaps, but not with the size and volume of logs laying on the ground.
325
Xylobolus subpileatus preferentially reproduces in closed canopy gaps
326
Our analysis of the temporal distribution of X. subpileatus basidiomes provided evidence of the
327
fruiting of X. subpileatus on late stages of decay of fallen logs. Fruitbodies were never detected on
328
logs within the seven years following treefall, and the probability of presence increased by 7% per
329
year of residence on the ground until the complete decaying of the substrate (Fig. 3a). The strong
330
affinity of X. subpileatus, at its sexual stage, for highly decayed wood sheds some light on the
331
ecological niche of this rare species. Its increased probability of fruiting as gaps age may be a
332
response to changes in both biotic and abiotic environments at ground level in old-growth forest
333
patches (Oldeman 2012). In terms of ecological strategies, the pattern of X. subpileatus fruiting
334
occurrence may reflect niche specialization toward highly decayed logs, exceptional competitive
335
abilities, or both.
336
First, the temporal pattern of X. subpileatus fruiting may indicate its affinity for deep shade
337
environments. From the time of the treefall to the complete canopy recovery by adjacent trees and
338
understory shrubs, the amount of light entering into the forest gap progressively decreases, inducing
339
changes in the microclimatic conditions (moisture, temperature, daily variability, etc.) for organisms
340
living on logs (Messier et al. 1999; McCarthy et al. 2001). Sun exposure affects composition of
341
saproxylic fungal communities in temperate forests, with some species fruiting in fully exposed
342
large gaps, while others require canopy-closed, multilayered openings (Brazee et al. 2014). In the
343
context of the seasonally dry and variable Mediterranean climate, sun exposure in new gaps and dry
344
microclimatic conditions may contribute to restricting the fruiting of X. subpileatus to logs in deep
345
shade.
346
Page 15 of 31

16
Second, the temporal distribution of X. subpileatus basidiomes might reflect a high degree of
347
functional specialization towards decomposition of lignin to complete its biological cycle
348
(Blanchette 1984). Indeed, during the decay process, physical and chemical properties of wood
349
gradually change. In particular, water and lignin contents increase with loss of cellulose and with
350
decrease in density (Rajala et al. 2012), as a result of the functional succession of saproxylic species
351
on the log. During this process, early dominant soft-rot ascomycetes are progressively replaced by
352
white- and brown-rot species of basidiomycetes (Rayner and Boddy 1988). Regarding the catabolic
353
abilities of the corresponding species, this species succession reflects the turnover from cellulose
354
decayers to organisms that are efficient in degrading polyphenols, including lignin (Daniel and
355
Nilsson, 1998). From this perspective, the very late dominance of fruiting X. subpileatus on Q. ilex
356
logs may primarily be a positive response to the emergence of substrates with high concentrations
357
of lignin in canopy gaps at the end of the decay process. It is likely that increasing lignin
358
concentration acts as a chemical filter for many saproxylic fungi, but not for X. subpileatus (Otjen
359
and Blanchette 1984), and as a driver of the dynamics of ecological succession on logs.
360
Third, the dynamics of X. subpileatus sexual reproduction may reflect a highly competitive strategy
361
regarding ecological succession (Grime et al. 1977; Last et al. 1987). Indeed, fungal interspecific
362
interactions become more complex during the wood-decay process. Based on a large corpus of
363
basidiome records in temperate and boreal forests, but also using the sequencing of mycelia in
364
decaying wood, it has been established that species richness increases during the decaying process
365
(Kubartová et al. 2012), and interspecific fungal interactions increase with time within guilds (i.e.
366
among saproxylic species) and among guilds (i.e. between saproxylic and ectomycorrhizal species)
367
of fungi (Rajala et al. 2012; Ottoson et al. 2015). In other words, in late phases of decay, white-rot
368
fungi not only compete among themselves, but also with rich communities of ectomycorrhizal
369
delignifying species, that become the most dominant ecological strategy in wood (Rajala et al.
370
2012). In the system we studied, it is likely that the temporal pattern of X. subpileatus occurrence in
371
Page 16 of 31

17
gaps reflects its high competitive abilities in particularly rich fungal communities (Richard et al.
372
2004). The use of next-generation sequencing methods in this exceptional set of dated gaps should
373
provide complementary insight not only into the ecological niche of this Xylobolus species, but also
374
into that of ectomycorrhizal communities in Mediterranean ecosystems.
375
Implications for the conservation of fungal diversity
376
In the present study, we compared the temporal fruiting pattern of X. subpileatus to (i) that of the
377
related species Stereum hirsutum (Fig. 3) and (ii) those of the ECM and saprobic fruiting
378
communities of macromycetes in the same study site (Fig. 4; Richard et al. 2004).
379
We found evidence for contrasted requirements of X. subpileatus and S. hirsutum regarding the age
380
of canopy gaps enabling their fruiting. The two species never co-occurred on any wood logs
381
investigated here. However, our dataset does not provide evidence of strict niche segregation with
382
respect to gap age. Although S. hirsutum was not detected in gaps > 12 yrs old, X. subpileatus was
383
recorded to fruit in gaps as young as 7 years, allowing a 5-year-long overlap (Fig. 3). The different
384
optima of the two species may reflect their preferences not only for wood at different stages of
385
decay but also for different parts of the decaying log (e.g. sapwood vs heartwood). Before
386
concluding about possible mechanisms of niche differentiation between these two close relatives,
387
additional work is needed to accurately describe the occupancy pattern of their respective mycelia
388
in logs throughout the period from treefall to the end of the wood-decay process.
389
From a conservation perspective, and taking into account only basidiome occurrence, our findings
390
highlight the complementarity of canopy gaps of uneven age in our study site in maintaining
391
populations of these two species. Further, our results support the exceptional capacity of old-growth
392
forests to allow the expression of contrasted ecologies at very fine scales, due to natural dynamics
393
of gaps (Fournier et al. 2012). Indeed, in old-growth forests more than any other environment, the
394
stochasticity of disturbance regimes governs the temporal distribution of treefall hazards, allowing
395
Page 17 of 31

18
the persistence of all stages of wood decay at the scale of forest patches, which in turn allows a
396
continuous beginning of new ecological successions (Wilson 1999).
397
The probability of occurrence of X. subpileatus basidiomes on the one hand, and the fruiting
398
patterns of both litter saprobic and ectomycorrhizal macromycetes on the other hand, show opposite
399
patterns during gap closure (Fig. 4). Comparable case studies are lacking, particularly in the
400
Mediterranean area, mostly because long-term surveys based on permanent plots are scarce (Quézel
401
and Médail 2003). In the best-documented study of succession of wood-inhabiting fungi, early
402
peaks of species richness were observed a few years after treefall (Lindhe et al. 2004), but no clear
403
pattern emerges from the available reports that consider the entire sequence of wood decay (e.g.
404
Berglund et al. 2005). In contrast, the occurrence of red-listed species of wood-inhabiting fungi
405
have been reported to increase with time in northern European forests (Heilmann-Clausen and
406
Christensen 2005). Our results provide supplementary evidence, in the Mediterranean bioclimatic
407
region, that species richness is an unsatisfactory proxy for the presence of the most threatened
408
elements of fungal diversity, in the same guild (Heilmann-Clausen and Christensen 2005), or in
409
different guilds (Fig. 4). The studies that examine this question, both the previous studies we cite
410
and our own, are all based on fruitbody records. One remaining question is thus whether (and to
411
what degree) the observed fruiting patterns reflect the distribution of resident mycelia. A perfect
412
match is unlikely, in particular for mycorrhizal fungi. In a previous study conducted at the same
413
site, we showed that reproductive (fruitbodies) and vegetative (soil mycelia) mycorrhizal
414
communities differed markedly in structure and composition (Richard et al. 2005).
415
One important implication of the present study is the demonstration that it is possible to accurately
416
date X. subpileatus fruiting with respect to the entire forest cycle. A fine-scale analysis of patch
417
dynamics (Panaïotis et al. 1997, 1998) showed that oak individuals that created natural treefalls
418
averaged 170 ± 46 years in age. In this context, X. subpileatus shows a maximum of occurrence 40
419
Page 18 of 31

19
years after canopy openings, i.e. about 210 (170 + 40) years since the beginning of the long period
420
without human intervention affecting the forest stand. In the Mediterranean basin, such temporal
421
continuity of ‘naturalness’ is particularly rare because of the ancient and particularly high human
422
pressure in the region, and we may predict that potentially suitable habitats for X. subpileatus are
423
limited to very few locations (Quezel and Medail 2003) where the species could be sought within
424
multi-secular old-growth forest.
425
From a management perspective, the ecological requirements of X. subpileatus highlight the high
426
value of the remaining old-growth Q. ilex forests for macromycete conservation. The long-term
427
persistence of this ecologically unique species requires implementing management practices that
428
allow (i) the long-term continuity of forest cover in Q. ilex stands and (ii) the accumulation and the
429
residence on forest soil of large woody debris until its complete decay. Such scenarios rarely occur
430
in Mediterranean Q. ilex forests, which are mostly managed as short-rotation coppices to produce
431
fuelwood.
432
Conclusion
433
Because they are among the most threatened organisms in Europe, saproxylic fungi have fueled an
434
intense debate about how they should be conserved. In Mediterranean forest, saproxylic biota
435
constitute a major conservation issue, not only because they depend on the most important forestry
436
product, i.e. dead wood, but also because forest managers consider their preferred habitat to
437
engender high fire risk (large woody debris are the main fuel to sustain fire during days). We here
438
demonstrate that the conditions for fruiting of Xylobolus subpileatus are associated with forests long
439
preserved from human exploitation, where large woody debris take decades to be mineralized by
440
long and complex chains of organisms. The optimal conditions for this rare species occur only at the
441
end of this long process. In these old-growth stands, only large Quercus ilex logs are suitable
442
habitats for this late-successional fungal species, because oak logs last long enough on the forest
443
Page 19 of 31

20
floor to enable the accomplishment of a complete turnover of the composition of wood-decaying
444
fungi. Therefore, management practices that preserve old-growth stands and their natural dynamics
445
from fire risk are essential to secure the regional persistence of this endangered and functionally
446
unique species, whose current distribution is restricted to a few hectares of preserved forest patches
447
in Corsica.
448
Author Contributions
449
AT and FR originally formulated the idea, developed the methodology and performed statistical
450
analyses. AT, FR, CP, TL and AC generated data. J-MB and P-AM performed phylogenetic
451
analyses. FR and AT wrote the initial manuscript. All authors contributed to the final version of the
452
manuscript.
453
Acknowledgments
454
This study is dedicated to the memory of Justin Smith, who accompanied the discovery survey in
455
summer 2013 and died from sudden arrhythmic death syndrome, on March 3, 2014. We sincerely
456
thank Basile Zorabiniu for his precious help during field survey. We are grateful to Doyle McKey
457
for checking the English. This study was funded by the Observatoire de REcherche Méditerranéen
458
de l’Environnement (OSU OREME, UMS3282 CNRS and UMS223 IRD.) All the experiments
459
performed in this study comply with the current French regulations.
460
Page 20 of 31

21
References
461
Berglund, H., Edman, M., and Ericson, L. 2005. Temporal variation of wood-fungi diversity in boreal old-462
growth forests: implications for monitoring. Ecological Applications 15(3): 970–982. doi:10.1890/04-463
0628. 464
Berglund, H., Hottola, J., Penttilä, R., Siitonen, J. 2011. Linking substrate and habitat requirements of wood-465
inhabiting fungi to their regional extinction vulnerability. Ecography 34: 864–875. doi:10.1111/j.1600-466
0587.2010.06141.x. 467
Blanchette, R.A. 1984. Screening wood decayed by white rot fungi for preferential lignin degradation. 468
Applied and Environmental Microbiology 48(3): 647–653. 469
Brazee, N.J., Lindner, D.L., D’Amato, A.W., Fraver, S., Forrester, J.A., and Mladenoff, D.J. 2014. 470
Disturbance and diversity of wood-inhabiting fungi: effects of canopy gaps and downed woody debris. 471
Biodiversity and Conservation 23(9): 2155–2172. doi:10.1007/s10531-014-0710-x. 472
Daniel, G., and Nilsson, T. 1998. Developments in the study of soft rot and bacterial decay. In Forest 473
products biotechnology, Taylor & Francis Ltd. A. Bruce & J.W. Palfreyman, London - Bristol. pp. 37–474
62. 475
Dereeper, A., Guignon, V., Blanc, G., Audic, S., Buffet, S., Chevenet, F., Dufayard, J.-F., Guindon, S., Lefort, 476
V., Lescot, M., Claverie, J.-M., and Gascuel, O. 2008. Phylogeny.fr: robust phylogenetic analysis for the 477
non-specialist. Nucleic Acids Res. 36(Web Server issue): W465-469. doi:10.1093/nar/gkn180. 478
Diedhiou, A.G., Dupouey, J.-L., Buée, M., Dambrine, E., Laüt, L., Garbaye, J. 2009. Response of 479
ectomycorrhizal communities to past Roman occupation in an oak forest. Soil Biology and Biochemistry 480
41: 2206–2213. doi:10.1016/j.soilbio.2009.08.005 481
Fournier, T.P., Brossier, B., Gauthier, O., and Carcaillet, C. 2012. Post-Fire dynamics and spatial 482
heterogeneity of woody debris associated with Pinus halepensis in an oak—pine Mediterranean 483
ecosystem. Ecoscience 19(4): 356–363. doi:10.2980/19-4-3502. 484
Gardes, M., and Bruns, T.D. 1993. ITS primers with enhanced specificity for Basidiomycetes - application to 485
the identification of mycorrhizae and rusts. Molecular Ecology 2(2): 113–118. doi:10.1111/j.1365-486
294X.1993.tb00005.x. 487
Page 21 of 31

22
Goodman, D.M., and Trofymow, J.A. 1998. Comparison of communities of ectomycorrhizal fungi in old-488
growth and mature stands of Douglas-fir on southern Vancouver Island. Northwest Science 72: 91–93. 489
Grime, J.P. 1977. Evidence for the existence of three primary strategies in plants and its relevance to 490
ecological and evolutionary theory. The American Naturalist 111(982): 1169. doi:10.1086/283244. 491
Halme, P., Ódor, P., Christensen, M., Piltaver, A., Veerkamp, M., Walleyn, R., Siller, I., Heilmann-Clausen, J. 492
2013. The effects of habitat degradation on metacommunity structure of wood-inhabiting fungi in 493
European beech forests. Biological Conservation 168, 24–30. doi:10.1016/j.biocon.2013.08.034. 494
Heilmann-Clausen, J., and Christensen, M. 2003. Fungal diversity on decaying beech logs–implications for 495
sustainable forestry. Biodiversity & Conservation 12(5): 953–973. 496
Heilmann-Clausen, J., and Christensen, M. 2005. Wood-inhabiting macrofungi in Danish beech-forests – 497
conflicting diversity patterns and their implications in a conservation perspective. Biological 498
Conservation 122(4): 633–642. doi:10.1016/j.biocon.2004.10.001. 499
Izzo, A., Agbowo, J., and Bruns, T.D. 2005. Detection of plot-level changes in ectomycorrhizal communities 500
across years in an old-growth mixed-conifer forest. New Phytologist 166(2): 619–630. 501
doi:10.1111/j.1469-8137.2005.01354.x. 502
Jönsson, M.T., Edman, M., and Jonsson, B.G. 2008. Colonization and extinction patterns of wood-decaying 503
fungi in a boreal old-growth Picea abies forest. Journal of Ecology 96(5): 1065–1075. 504
doi:10.1111/j.1365-2745.2008.01411.x. 505
Junninen, K., Similä, M., Kouki, J., and Kotiranta, H. 2006. Assemblages of wood-inhabiting fungi along the 506
gradients of succession and naturalness in boreal pine-dominated forests in Fennoscandia. Ecography 507
29(1): 75–83. doi:10.1111/j.2005.0906-7590.04358.x. 508
Kotlaba, F. 1986. Stereum frustulatum–pevník rozpraskanỳ, jeho ekologie a zeměpisné rozšíření v 509
Československu. Česká Mykol 40(3): 129–141. 510
Kubartová, A., Ottosson, E., Dahlberg, A., and Stenlid, J. 2012. Patterns of fungal communities among and 511
within decaying logs, revealed by 454 sequencing. Molecular Ecology 21(18): 4514–4532. 512
doi:10.1111/j.1365-294X.2012.05723.x. 513
Page 22 of 31

23
Last, F.T., Dighton, J., and Mason, P.A. 1987. Successions of sheathing mycorrhizal fungi. Trends in Ecology 514
& Evolution 2(6): 157–161. doi:10.1016/0169-5347(87)90066-8. 515
Leys, B., Finsinger, W., and Carcaillet, C. 2014. Historical range of fire frequency is not the Achilles’ heel of 516
the Corsican black pine ecosystem. Journal of Ecology 102(2): 381–395. doi:10.1111/1365-2745.12207. 517
Lindblad, I. 1998. Wood-inhabiting fungi on fallen logs of Norway spruce: relations to forest management 518
and substrate quality. Nordic Journal of Botany 18(2): 243–255. doi:10.1111/j.1756-519
1051.1998.tb01877.x. 520
Lindhe, A., Åsenblad, N., and Toresson, H.-G. 2004. Cut logs and high stumps of spruce, birch, aspen and 521
oak – nine years of saproxylic fungi succession. Biological Conservation 119(4): 443–454. 522
doi:10.1016/j.biocon.2004.01.005. 523
Loizides, M., Bellanger, J.-M., Clowez, P., Richard, F., and Moreau, P.-A. 2016. Combined phylogenetic and 524
morphological studies of true morels (Pezizales, Ascomycota) in Cyprus reveal significant diversity, 525
including Morchella arbutiphila and M. disparilis spp. nov. Mycological Progress 15(4): 1–28. 526
doi:10.1007/s11557-016-1180-1. 527
Long, W.H. 1915. A honeycomb heart-rot of oaks caused by Stereum subpileatum. Journal of Agricultural 528
Research 5(10): 421–430. 529
McCarthy, J. 2001. Gap dynamics of forest trees: A review with particular attention to boreal forests. 530
Environmental Reviews 9(1): 1–59. doi:10.1139/er-9-1-1. 531
Messier, C., Doucet, R., Ruel, J.-C., Claveau, Y., Kelly, C., and Lechowicz, M.J. 1999. Functional ecology of 532
advance regeneration in relation to light in boreal forests. Canadian Journal of Forest Research 29(6): 533
812–823. 534
Myers, N., Mittermeier, R.A., Mittermeier, C.G., Da Fonseca, G.A., and Kent, J. 2000. Biodiversity hotspots 535
for conservation priorities. Nature 403(6772): 853–858. doi:10.1038/35002501. 536
Oldeman, R.A.A. 2012. Forests: elements of silvology. Springer Science & Business Media. 537
Otjen, L., and Blanchette, R.A. 1984. Xylobolus frustulatus decay of oak: patterns of selective delignification 538
and subsequent cellulose removal. Applied and Environmental Microbiology 47(4): 670–676. 539
Page 23 of 31

24
Ottosson, E., Kubartova, A., Edman, M., Jonsson, M., Lindhe, A., Stenlid, J., and Dahlberg, A. 2015. Diverse 540
ecological roles within fungal communities in decomposing logs of Picea abies. FEMS Microbiology 541
Ecology 91(3). doi:10.1093/femsec/fiv012. 542
Paillet, Y., Bergès, L., Hjältén, J., óDor, P., Avon, C., Bernhardt-Römermann, M., Bijlsma, R.-J., De Bruyn, 543
L., Fuhr, M., Grandin, U., Kanka, R., Lundin, L., Luque, S., Magura, T., Matesanz, S., Mészáros, I., 544
Sebastià, M.-T., Schmidt, W., Standovár, T., Tóthmérész, B., Uotila, A., Valladares, F., Vellak, K., and 545
Virtanen, R. 2010. Biodiversity differences between managed and unmanaged forests: meta-analysis of 546
species richness in Europe. Conservation Biology 24(1): 101–112. doi:10.1111/j.1523-547
1739.2009.01399.x. 548
Panaïotis, C., Carcaillet, C., and M’Hamedi, M. 1997. Determination of the natural mortality age of holm 549
oak (Quercus ilex L.) stand in Corsica (Mediterranean Island). Acta Oecologica 18(5): 519–530. 550
doi:10.1016/S1146-609X(97)80038-0. 551
Panaïotis, C., Loisel, R., and Paradis, G. 1995. Dating natural gaps in the holm oak forest (Quercus ilex L.) in 552
Fango MAB Reserve (Corsica) by reading rings of maquis components. Annals of Forest Science 52: 553
477–487. 554
Panaïotis, C., Loisel, R., and Roux, M. 1998. Analyse de la réponse de la végétation aux trouées naturelles 555
dans une futaie âgée de Quercus ilex L. en Corse (île Méditerranéenne). Canadian Journal of Forest 556
research 28(8): 1125–1134. doi:10.1139/x98-070. 557
Papp, V. 2011. Adatok a Xylobolus Nemzetség Magyarországi El\hofordulásáról. Mikológiai Közlemények, 558
Clusiana 50(2). 559
Parmasto, E. 2001. Fungi as indicators of primeval and old-growth forests deserving protection. In Fungal 560
conservation, issues and solutions, Cambridge University Press. pp. 81–88. 561
Pieri, M., and Rivoire, M. 1992. Contribution à l'inventaire des Aphyllophoromycetideae et 562
Heterobasidiomycetes de la Corse - I. Bull. Feder. Assoc. Myc. Medit. 2 p. 5-10 563
Pieri, M., and Rivoire, M. 1994. Contribution à l'inventaire des Aphyllophoromycetideae et 564
Heterobasidiomycetes de la Corse - II. Bull. Feder. Assoc. Myc. Medit. 2 p. 15-19 565
Page 24 of 31

25
Quézel, P., and Médail, F. 2003. Ecologie et biogéographie des forêts du bassin méditerranéen. Elsevier, 566
Paris, France. 567
Rajala, T., Peltoniemi, M., Pennanen, T., and Mäkipää, R. 2012. Fungal community dynamics in relation to 568
substrate quality of decaying Norway spruce (Picea abies [L.] Karst.) logs in boreal forests. FEMS 569
Microbiology Ecology 81(2): 494–505. doi:10.1111/j.1574-6941.2012.01376.x. 570
Rayner, A.D., and Boddy, L. 1988. Fungal decomposition of wood. Its biology and ecology. John Wiley & 571
Sons Ltd. New York. 602 p. 572
Reille, M. 1992. New pollen-analytical researches in Corsica: the problem of Quercus ilex L. and Erica 573
arborea L., the origin of Pinus halepensis Miller forests. New Phytologist 122: 359–378. 574
doi:10.1111/j.1469-8137.1992.tb04241.x 575
Richard, F., Moreau, P.-A., Selosse, M.-A., and Gardes, M. 2004. Diversity and fruiting patterns of 576
ectomycorrhizal and saprobic fungi in an old-growth Mediterranean forest dominated by Quercus ilex L. 577
Can. J. Bot. 82(12): 1711–1729. doi:10.1139/b04-128. 578
Richard, F., Selosse, M.-A., and Gardes, M. 2009. Facilitated establishment of Quercus ilex in shrub-579
dominated communities within a Mediterranean ecosystem: do mycorrhizal partners matter? FEMS 580
Microbiology Ecology 68(1): 14–24. doi:10.1111/j.1574-6941.2009.00646.x. 581
Smith, J.E., Molina, R., Huso, M.M., Luoma, D.L., McKay, D., Castellano, M.A., Lebel, T., and Valachovic, 582
Y. 2002. Species richness, abundance, and composition of hypogeous and epigeous ectomycorrhizal 583
fungal sporocarps in young, rotation-age, and old-growth stands of Douglas-fir (Pseudotsuga menziesii) 584
in the Cascade Range of Oregon, U.S.A. Canadian Journal of Botany 80(2): 186–204. doi:10.1139/b02-585
003. 586
Stokland, J.N., and Larsson, K.-H. 2011. Legacies from natural forest dynamics: Different effects of forest 587
management on wood-inhabiting fungi in pine and spruce forests. Forest Ecology and Management 588
261(11): 1707–1721. doi:10.1016/j.foreco.2011.01.003. 589
Wilson, E.O. 1999. The diversity of life. WW Norton & Company. New York. 424p. 590
Page 25 of 31

26
Worrall, J.J., Lee, T.D., and Harrington, T.C. 2005. Forest dynamics and agents that initiate and expand 591
canopy gaps in Picea–Abies forests of Crawford Notch, New Hampshire, USA. Journal of Ecology 592
93(1): 178–190. doi:10.1111/j.1365-2745.2004.00937.x. 593
Table 1: GenBank accession numbers of ITS sequences generated for the present work.
594
Gap
Species
Accession number
12
Xylobolus
subpileatus
KX578072
19
X.
subpileatus
KX578074
20
X. subpileatus
KX578075
38
X. subpileatus
KX578080
45
X.
subpileatus
KX578077
52
X.
subpil
eatus
KX578076
57
X. subpileatus
KX578083
59
X. subpileatus
KX578086
60
X.
subpileatus
KX578078
62
X.
subpileatus
KX578073
63
X. subpileatus
KX578079
64
X. subpileatus
KX578085
68
X.
subpileatus
KX578082
75
X.
subpileatus
KX578084
67
Stereum hirsutum
KX578081
Table 2:
Summary of the
generalized linear model of the presence of basidiomes of Stereum hirsutum
595
and Xylobolus subpileatus as a function of gap age (n = 59 logs; 7 with S. hirsutum, 22 with
596
X. subpileatus).
597
Stereum hirsutum
Xylobolus subpileatus
Estimate ± SE
p
Estimate ± SE
p
(Intercept)
-1.03 ± 0.654
0.112
-1.85 ± 0.550
<0.001
Age
-0.09 ± 0.053
0.0936
0.07 ± 0.024
<0.001
Page 26 of 31

27
Fig. 1: (a) Reproductive structure (basidiome) of Xylobolus subpileatus on Quercus ilex logs. (b)
598
Typical honeycomb decomposition pattern of decayed wood. View of decaying logs from the forest
599
edge in 1- (c), 5- (d) and 20-year-old (e) canopy gaps.
600
Fig. 2: Location of the study site in the northwestern Mediterranean (a), and more precisely on the
601
island of Corsica (b), and detailed pattern of the sampling area in red (c). The green area in panel b
602
delimits the Fango UNESCO Man and Biosphere reserve. Circles (
◦
)
symbolize Quercus ilex logs
603
and crosses (×) gaps created by other woody species. Dark grey, light grey and open circles
604
respectively indicate the presence of Xylobolus subpileatus,
the presence of Stereum hirsutum and
605
the absence of both species on gaps made of Quercus ilex logs. The grey rectangle in (c) represents
606
the area where surface of X. subpileatus fruitbodies and Q. ilex logs were measured.
607
Fig. 3: Probability of fruiting of (a) Xylobolus subpileatus and (b) Stereum hirsutum on tree logs as
608
a function of the number of years since tree fall within the 25-ha area (n = 59; 22 logs with
609
X. subpileatus, 7 with S. hirsutum). The distribution of presence/absence is represented by violin
610
plots: the white dot represents the median, the black rectangle the interquartile range and the grey
611
envelope is a kernel density estimation of the distribution. Black lines depict the linear model and
612
the associated grey shades delimit 95% confidence intervals of the probability of fruiting according
613
to time since treefall.
614
Fig. 4: Number of fruitbodies (a) and species richness per 100 m
2
(b) of saproxylic (in red) and
615
ectomycorrhizal fungi (in green) as a function of the age of the 10 canopy gaps. Black lines
616
represent the linear model of the probabilities (grey shades 95% CI) of presence of Xylobolus
617
subpileatus (cf. Fig. 3a).
618
Page 27 of 31

Fig. 1: (a) Reproductive structure (basidiome) of Xylobolus subpileatus on Quercus ilex logs. (b) Typical
honeycomb decomposition pattern of decayed wood. View of decaying logs from the forest edge in 1- (c), 5-
(d) and 20-year-old (e) canopy gaps.
Page 28 of 31

Fig. 2: Location of the study site in the northwestern Mediterranean (a), and more precisely on the island of
Corsica (b), and detailed pattern of the sampling area in red (c). The green area in panel b delimits the
Fango UNESCO Man and Biosphere reserve. Circles (◦) symbolize Quercus ilex logs and crosses (×) gaps
created by other woody species. Dark grey, light grey and open circles respectively indicate the presence of
Xylobolus subpileatus, the presence of Stereum hirsutum and the absence of both species on gaps made of
Quercus ilex logs. The grey rectangle in (c) represents the area where surface of X. subpileatus fruitbodies
and Q. ilex logs were measured.
Page 29 of 31

Page 30 of 31

Page 31 of 31