全 文 ：ACTA ECOLOGICA SINICA
Volume 27, Issue 3, March 2007
Online English edition of the Chinese language journal

Cite this article as: Acta Ecologica Sinica, 2007, 27(3), 813−836.


Received date: 2006-07-11; Accepted date: 2007-03-04
*Corresponding author. Vladimir Krivtsov, CECS, The Crew Building, King’s Buildings, University of Edinburgh, West mains Road, Edinburgh EH9 3JN,
Scotland, UK. E-mail: e96kri69@netscape.net
Copyright © 2006, Ecological Society of China. Published by Elsevier BV. All rights reserved.



RESEARCH PAPER

Some aspects of forest soil and litter ecology in the
Dawyck Cryptogamic Sanctuary with a particular
reference to fungi
V. Krivtsov1,*, A. Brendler2, R. Watling3,1, K. Liddell2, H.J. Staines2

1 Caledonian Mycological Enterprises, UK
2 SIMBIOS, Schools of Science & Engineering and Computing, University of Abertay Dundee, Bell Street, Dundee DD1 1HG, Scotland, UK
3 Royal Botanic Gardens Edinburgh, Edinburgh, Scotland, UK


Abstract: Here we report on ecology and biodiversity of fungi in a unique mycological sanctuary in Britain, where data on species
composition have been collected since 1994. To complement the biodiversity data by the information on the fungal ecological inter-
actions and their role in the overall ecosystem functioning, soil properties and the composition of forest litter and field layer, bacte-
rial population numbers and fungal biomass (in terms of ergosterol) were measured in 8 plots covered with different vegetations
(beech, birch, birch-oak-beech, grass) over a May–Aug. period, and the results were analysed by correlation analysis and stepwise
regression modelling together with data on protozoa and nematodes available from parallel research. The results highlighted the
complexity of factors influencing temporal dynamics and spatial variability of fungal biomass in soil and forest litter. Most of the
registered interactions appeared to be transient, and this should be taken into account while interpreting environmental observations.
Interpretation of the specific relationships is given and implications for further research and overall ecosystem functioning are dis-
cussed.

Key Words: beech; birch; forest; fungi; grass; indirect effects; leaf litter; multiple regression modelling; soil; bacteria; ergosterol;
glomalin



The importance of fungi in the decomposition of plant re-
mains, biogeochemical recycling in the soil/litter subsystem,
and overall ecosystem functioning is well recognised[1]. To date,
many field and laboratory studies have addressed various
important issues related to fungal/environmental interactions
and fungal role in the ecosystem structure and functioning,
including environmental aspects of fungal biodiversity and
succession using observations of fruit bodies[2－8] and molecu-
lar techniques[9], aspects of fungal involvement in the decom-
position of specific types of litter[10－12], ecological characteris-
tics of important species[5,13], plant/mycorrhizal interactions[14,15],
seasonal dynamics of fungal mycelium[16] and fruit body pro-
duction[7], differences in fungal mycelium abundance in dif-
ferent types of soil and litter[17], and specific interrelations
with other biota[18－32].
The Cryptogamic Sanctuary is part of the Heron Wood Re-
serve, and represents a unique wooded area specifically desig-
nated for cryptogamic, especially mycological, and ecological
research. Situated in the Scottish Borders with properties
handed for management and development by the Secretary of
State for Scotland to the Royal Botanic Garden, Edinburgh in
1968, it is exceedingly well documented. Intensive recording
has been undertaken since 1994. The previous research con-
ducted at the sanctuary has concentrated on ecosystem function-
ing and microbial interactions in forest litter and soil[28,29,33－37],
and influences of fungi, bacteria and litter composition on
forest soil and litter fauna[28,34,38－40]. There was also a model-
ling study of fungal fruiting patterns within a larger area of the
Dawyck Botanic Garden[41]. The present paper gives the sum-
mary of the fungal biodiversity observed within a long-term
V. Krivtsov et al. / Acta Ecologica Sinica, 2007, 27(3): 813–836

monitoring programme, and provides a focus on the dynamics
of bacterial and fungal biomass, and their ecological interac-
tions in the soil and forest litter during a summer period.
1 Materials and methods
1.1 Site description
Heron Wood (Peebleshire, Scotland, Ordnance Survey grid
reference NT(36)175 355, approximately 235 m ASL) is a
very old ornamental Beech (Fagus) plantation invaded by young
birch (Betula) saplings over twenty years ago, and offers a
wide diversity of micro-habitats. Within Heron Wood there are
scattered volunteer oaks, possibly reflecting the former history
of the site. Heron Wood is 7.5 hectares in size and is divided
into the Sanctuary, which attracts no management, and the
Reserve, where limited management is undertaken. The latter
is about 1/3rd of the whole designated area. In order to study
the diversity, careful monitoring is necessary and therefore
within the Sanctuary at Heron Wood a 10000 m2 plot has been
delineated. In each quarter of this Biodiversity Plot two
smaller plots, 100 m2 in size, have been designated, making 8
in all (a plan adopted from earlier mycodiversity studies un-
dertaken in the UK, including Scotland, by the Forestry Au-
thority and the British Mycological Society). Hence the layout
of the sampling site was in accordance with standard practice
used by the Forestry Commission, and covered a range of
woodland habitats (dominated by beech, birch and oak) and a
clear area covered with grass (dominated by Holcus lanatus).
1.2 Sampling and analysis
Composite samples of forest floor (consisting predomi-
nantly of forest litter, mainly remains of leaves in various stages
of decomposition; hence hereafter these samples will be re-
ferred to as ‘forest litter’ or ‘leaf litter’ interchangeably) and
soil cores (taken from the soil surface down to approximately
10–15 cm depth after removing litter) were collected monthly
(32 samples: 8 plots, and 4 replicates from each plot) using a
plastic sampling frame (10cm×15cm) and a manual soil corer.
The exact positions of sampling points were determined a
priori using a random number generator. In the laboratory the
samples were hand-sorted, and the litter composition was
determined on the % (weight by weight) basis. Moisture con-
tent was determined via weight loss after drying for 48 h at
80 °C.
For assessment of fungal biomass, subsamples of litter were
shredded and soil was sieved (mesh size 2 mm). Ergosterol[42]
was extracted using sonication [43]. Ergosterol concentration in
the extracts was assayed by HPLC (Spectra-Physics) using
hexane/ isopropanol (98:2v/v) as a mobile phase. The ergos-
terol peaks at 282 nm were registered using a UV absorbance
detector and quantified using an automatic integrator. Soil
glomalin content (easily extractable fraction) was measured
spectrophotometrically[44]. Extractions of both fungal bio-
markers were carried out on fresh samples.
Bacteria were assessed using light microscopy following
staining with DTAF[45]. Total microbial biomass (expressed as
C and N contents) was assessed by the difference in absorb-
ance at 280 nm measured on the extracts from fumigated and
unfumigated samples. [46]
Subsequent data handling and analyses were carried out us-
ing Microsoft Excel and Minitab software. Apart from the data
on fungi, bacteria and total microbial biomass reported here in
detail, the data set analysed also included the information on a
number of other variables studied in parallel projects using
standard methods, including litter composition, soil properties,
and microinvertebrates[29,35,36,40].
2 Results
2.1 Fungal biodiversity
It should be noted that in addition to the intensive short-
term investigation reported below in detail, we have also
carried out a long-term monitoring programme of fungal fruit-
ing. In the study, the three main trophic strategies of fungi
have been addressed and data (not shown) are available for all
groups. The three trophic strategies are the necrotrophic strat-
egy (characteristic of those fungi which utilise food materials
from living organisms and often bring about cell death), the
saprotrophic strategy (using material from dead organisms),
and the biotrophic strategy (characteristic of those fungi which
obtain nutrients from a liaison with one or more other organ-
isms). Over 200 different species of macromycetes have been
recorded in the plots and outside the plots covering a whole
spectrum of wood-rotters, litter rotters, necrotrophs and ecto-
mycorrhizal fungi.
Those plots dominated by beech canopy are characterised
by a suite of ectomycorrhizal agarics. There is a quartet of
members of the Russulaceae (Russulales), Russula fellea, R.
mairei and R. nigricans with the first two as well as Lactarius
blennius, also very common in these plots, specific to beech.
In smaller numbers yet well distributed are Cortinarius torvus
(Cortinariaceae, Cortinariales), Tricholoma columbetta, T.
sciodes and T. ustale (Tricholomataceae, Tricholomatales), the
first and last two species being found only with beech. In
contrast in the birch dominated plots the only putative ecto-
mycorrhizal agarics of note are Paxillus involutus (Paxillaceae,
Boletales, Basidiomycota) and Boletus chrysenteron (Boleta-
ceae, Boletales).
The lignicolous fungi occurring on all the tagged trees,
stumps and logs in addition to those throughout the reserve
have been catalogued over an eight year period and include
many species of importance to break-down of woody debris
viz. Schizopora paradoxa on oak, Gloeocystidellum porosum
on beech (both species belong to corticiaceous Polyporales;
Basidiomycota), and even include one species which may
have important effects on the quality of the air, viz. Phellinus
ferreus (Hymenochaetaceae, Hymenochaetales) — a white
V. Krivtsov et al. / Acta Ecologica Sinica, 2007, 27(3): 813–836

wood-rot species known to produce chloro hydrocarbons.
Litter rotters occur in all the plots including grassland, and
include a wide range of members of the genus Clitocybe,
Collybia, and especially Mycena (all Tricholomataceae, Tri-
cholomatales), the mycelial activity of the latter being very
evident as bleached areas on leaves (especially those sur-
rounding the basidiomes). Some of these litter rotters are high-
tly specific to the substrate that they colonize. The larger
mycota, which are necrotrophs, include Pholiota squarrosa
(Strophariaceae, Agaricales, Basidiomycota) on both birch and
pine, and Fistulina hepatica (Fistulinaceae, Fistulinales,
Basidiomycota) on oak. There is evidence in parts of the
Heron Wood that these have helped towards the colonization
of other saprotrophic taxa. There is little doubt that many of
these fungi are inter-related in the ecosystem and by studying
one component critically it will allow an entry point so that
the constituent elements can be conceptually isolated and
studied to attempt to construct a picture of the total mycota’s
activities.
Saprotrophic fungi are best adapted for decomposition of
woody debris and leaf litter, as they can produce enzymes
capable of breaking down recalcitrant substances[47]. In the
Heron Wood, fruit bodies of saprotrophic fungi are encoun-
tered in all types of plots. However, as evidenced by the re-
cords of basidiomes, the beech plots are dominated by ecto-
mycorrhizal fungi with few saprotrophs fruiting, whilst the
reverse is seen in the birch plots. This is perhaps favoured by
the fact that beech leaves once shed soon become compacted,
whilst the litter under the birches stays much less compacted
because of the herbaceous plants. Gadgil & Gadgil[48] from
experiments in pine woods have shown that there is suppres-
sion of saprotrophs by ectomycorrhizal roots.
2.2 Abiotic factors
2.2.1 Temperature and rainfall
The maximum temperatures recorded in May–August 2001
were in the range 15.5–18℃. There was, however, a decrease
in the maximum temperatures from May to June (15.5 ℃).
The minimum temperatures showed a gradual increase from
May (4.4 ℃) when the minimum temperature was the lowest
to July when the minimum temperature (10.1 ℃) was the
highest. The total monthly rainfall was the highest in July
(91.2 mm) and the lowest in June (55.1 mm). In July the total
rainfall increased again and then decreased in August. However,
the cumulative rainfall over 10 days before the samples were
collected was the highest in August and the lowest in May.
2.2.2 Soil physico-chemical properties
Soil Moisture Content⎯Fig. 1 shows changes in the mois-
ture content at different sites together with the results of the
oneway ANOVA. The twoway ANOVA (general linear model,
p<0.05) analysis showed no significant differences for site ×
month interactions.
Site 4 (the only grassland site) had higher moisture contents
than the other sites in all sampling months (except sites 5 and
6 in May). A decrease from May to June was observed in all
sites, and was especially notable in sites 5 and 6 where the soil
moisture content was the highest in May. The beech sites 1, 7
and 8 were the driest sites in June with approximately 24%–
27% moisture. From June to July the moisture content in-
creased in most of the sites except sites 2 and 6. A decrease is
shown from July to August in all sites except site 2.
Fig. 1 Moisture content in soil
The a and b labeled values showed significant differences (ANOVA) between each other; the values indicated with ab did not differ significantly from a and b
and between each other
V. Krivtsov et al. / Acta Ecologica Sinica, 2007, 27(3): 813–836

The changes in the average soil moisture content in the
sanctuary over the study period are shown in Fig. 2. The high-
est moisture content was observed in May (33.2 %). A notable
decrease is shown from May to June, whilst in July the mois-
ture content increased again. A decrease, however, was regis-
tered from July to August when the samples were the driest
(28.7 % moisture content).
Soil pH — pH in sites 1, 2 and 8 (dominated by beech) was
lower than that in the other sites in almost every sampling
month (Fig. 3). In sites 1 and 2 pH was lower than that in sites
3, 4, 6 and 7. The lowest pH was observed in site 2 in June
(pH 2.8), and the highest in site 7 in July (pH 3.8). A decrease
is shown in all sites from May to June, whilst in July the pH in
all sites increased again. From July to August a decrease was
observed in all sites except site 5.
The mean seasonal pH in soil from May to August is shown
in Fig. 4. A notable significant decrease was observed from
May to June, followed by an increase in July. However, there

Fig. 2 Average moisture content in soil
The a and b labeled values showed significant differences between each other with ANOVA; the value indicated with ab is not significantly different to the a-
and b-values
Fig. 3 pH in soil
The values indicated with the letters a, b and c are significantly different from each other (ANOVA); bc labeled values differ significantly to a-values and ab
indicated values differ significantly to c-values; the value indicated with abc is not significantly different to all values

Fig. 4 Average pH in soil
June values were significantly ( p < 0.05) different from May and July values
V. Krivtsov et al. / Acta Ecologica Sinica, 2007, 27(3): 813–836

was another decrease in pH in August. The twoway ANOVA
(general linear model, p < 0.05) analysis showed no significant
effect for site×month interactions.
Total Microbial Biomass ⎯ An increase in microbial bio-
mass was observed (Fig. 5) from May to June in most of the
sites studied (except in sites 3 and 5). The increase was par-
ticularly prominent in sites 2 and 7 with values of approxi-
mately 1.20 mg C/g dwt soil in May increasing to approxi-
mately 4–4.50 mg C/g dwt soil in June. In July soil microbial
biomass increased again (except sites 2 and 8), resulting in the
maximum values at site 7, with 5.69 mg C/g dwt soil and 0.91
mg N/g dwt soil. A decrease in the microbial biomass was
observed from July to August in sites 1, 2, 7 and 8 (beech,
beech-birch sites), whilst in sites 3, 4, 5 and 6 the values in-
creased. The lowest value was measured in site 5 in June (0.67
mg C/g dwt soil and 0.11 mg N/g dwt soil). The twoway
ANOVA (general linear model, p<0.05) analysis showed
significant differences for sites combined with months. The
ANOVA oneway sitewise analysis showed no significant differ-
ences between the 8 sites.
Fig. 6 shows the overall average microbial carbon and ni-
trogen content for the study period. A gradual increase in
carbon as well as nitrogen contents was observed from May to
July when they were the highest (4.32 mg C/g dwt soil and
1.42 mg N/g dwt soil). A decrease was observed from July to
August when the concentrations of carbon and nitrogen were
the lowest (1.23 mg C/g dwt soil 0.41 mg N/g dwt soil). It can
also be seen that there was quite considerable variation be-
tween the sampling sites, resulting in the large error bars.
Soil Organic Content and the Content of Roots ⎯ Changes
in the organic content of the soil are shown in Figs. 7–8. In
May, June and August the values were almost equal (approxi-
mately 14 %). In July, however, the average organic content of
the soil was at the highest concentration observed throughout
the study (17.1 %).
In most sites there was a notable increase in soil organic
content from June to July, when it was the highest in site 1
with a value of 21.9 %. From July to August a decrease was
observed in all sites (except site 2). In site 2 the soil organic
content did not show any considerable variation. The lowest
value of soil organic content was measured in site 8 in May
(10.9 %). The twoway ANOVA (general linear model, p <
Fig. 5 Microbial carbon content in soil
No significant differences between the sites were found by the oneway ANOVA

Fig. 6 Average microbial carbon and nitrogen contents in soil
The values indicated with different letters a, b, c show significant differences (ANOVA) between each other; error bars show standard deviations.
V. Krivtsov et al. / Acta Ecologica Sinica, 2007, 27(3): 813–836

0.05) analysis revealed significant differences for site×month
interactions.
The amount of roots found in the soil samples ranged be-
tween 0.5 mg roots/g and 11mg roots/g dwt soil. The variation
between the sampling sites was extremely high, and no
ANOVA analysis (oneway site and month and twoway site×
month) showed any significant differences (data not shown).
2.3 Fungal biomass in soil
Figs. 9 and 10 show the changes in soil ergosterol content
(a biomarker for live fungal biomass) for the study period. In
May, the mean monthly ergosterol content of the soil was the
lowest with a value of 7.90 µg ergosterol/g dwt soil. A small

Fig. 7 Average organic content in soil
Values indicated with different letters a and b showed significant differences between each other with ANOVA
Fig. 8 Organic content in soil
The a labeled values showed significant differences (ANOVA) to the values labeled with c and bc but not to the ab-values; the values indicated with c are
significantly different to a- and ab-values but not to the bc indicated value

Fig. 9 Average ergosterol content in soil
The values labeled with the letters a and b showed significant differences between each other with ANOVA; the value indicated with ab is not significantly
different to the a- and b-values
V. Krivtsov et al. / Acta Ecologica Sinica, 2007, 27(3): 813–836

increase was registered from May to June and the content
continued to increase from June to July, when it was the high-
est with a value of 13.61 µg ergosterol/g dwt soil. In August,
however, the ergosterol content showed a slight decrease.
In most sites an increase was observed from May to June.
In July the ergosterol contents were the highest in most sites.
A decrease was observed in August (except sites 2 and 7). The
highest ergosterol content in almost all months was registered
in the beech sites (i.e. sites 1 and 8). It should be noted, how-
ever, that in site 1 the values decreased dramatically from July
to August. There was a great variability noticed between the
four sampling points of each site as well as between the 8 sites,
hence the large error bars. The twoway ANOVA (general
linear model, p<0.05) analysis showed no significant differ-
ences for site×month interactions.
2.4 Soil glomalin (biomarker for AM fungi) content
Changes in glomalin content in soil are shown in Figs. 11
and 12. The highest mean glomalin content (3.44 mg/g dwt
soil) was observed in May. A notable decrease in the values
was observed from May to June, and continued in July, when
the content was the lowest with the concentration of 2.11 mg/
g dwt soil. From July to August an increase was observed.
In sites 5 and 6 the highest glomalin levels were registered
in May, with the values in site 6 reaching 4.44 mg/g dwt soil.
In all sites a considerable decrease from May to June could be
seen. In July there was a decrease in most of the sites (except
sites 1, 2 and 7), and glomalin content was the lowest in site 6
(1.17 mg/g dwt soil). From July to August the glomalin con-
tent increased most sites. There was a great variability be-
tween the four sampling points of each site as well as between
the 8 sites, resulting in the large error bars. The twoway
ANOVA (general linear model, p<0.05) analysis showed no
significant differences for site×month interactions.
2.5 Soil bacteria
Figs. 13 and 14 show the data on soil bacterial abundance.
It should be noted that the oneway ANOVA analysis for site
and month effects showed no significant differences. However,
the twoway ANOVA analysis showed significant site × month
Fig. 10 Ergosterol content in soil
The a labeled values showed significant differences (ANOVA) to the values labeled with c and bc but not to the ab-values; the values indicated with c are
significantly different to a- and ab-values but not to the bc indicated value

Fig. 11 Average glomalin content in soil
The values labeled with the letters a and b showed significant differences between each other with ANOVA; the value indicated with ab is not significantly
different to the a- and b-values; error bars show standard deviations
V. Krivtsov et al. / Acta Ecologica Sinica, 2007, 27(3): 813–836


Fig. 12 Glomalin content in soil
The a labeled values showed significant differences (ANOVA) to the values labeled with c and bc but not to the ab-values; the values indicated with c are
significantly different to a- and ab-values but not to the bc indicated value; error bars show standard deviations
Fig. 13 Bacterial counts in soil
Error bars show standard deviations

Fig. 14 Average bacterial counts in soil
Error bars show standard deviations
V. Krivtsov et al. / Acta Ecologica Sinica, 2007, 27(3): 813–836

interactions.
In Fig. 14 it is evident that in June and July soil bacteria
population was the most abundant. In most of the sites (except
sites 4, 7, 8) an increase in the bacterial numbers was regis-
tered from May to June (Fig. 13) and a decrease (except sites
1 and 5) from July to August. Sites 7 and 8 particularly dif-
fered in comparison with the other sites. In these sites a de-
crease from May to June and a notable increase from June to
July were observed. In site 4 (which is the only grassland site)
the bacterial abundance showed only relatively limited varia-
tion. Notable differences, however, were observed in sites 1, 2
and 3 between May and June, and in site 7 between June and
July. The highest bacterial numbers were found in site 2 in
June (2.01×108 bacteria/g dwt soil) and the lowest in site 3 in
August (1.26×108 bacteria/g dwt soil).
2.6 Forest litter
2.6.1 Forest litter components
The components found in the collected Forest Floor sam-
ples (mainly consisting of forest litter) were distinguished into
beech leaves, birch leaves, oak leaves, needles, grass, moss,
lichens, fern, wood (i.e. dead wood, twigs, bark of decaying
logs, etc.), roots with attached soil, seeds, cones and fragments.
Beech leaves and seeds were most common in the beech sites
1, 2 and 8 as well as in the mixed forest site 7. The highest
values for birch leaves were found in sites 5 and 6 (i.e. sites
dominated by birch). In site 4, which is the only grass-
dominated site, the highest amount of grass and the lowest
mass of beech leaves, seeds and wood were found. Some birch
leaves and moss were also collected in site 4, the former being
blown there by wind. In sites 3 and 5 the highest amount of
wood was found, and in sites 5 and 6 the most birch leaves were
collected. Moss was very common in sites 3, 4, 5, 6 and 8.
2.6.2 Litter moisture content
Fig. 15 shows changes in the moisture content of the Forest
litter samples. In May in all sites (except sites 2–3) the lowest
moisture content was observed, and hence the mean seasonal
moisture content had the lowest value in May (Fig. 15-16). An
increase from May to June was registered in all sites. In July
there was a considerable increase in sites 1, 7 and 8 again,
with the moisture content decreasing in all other sites. High
values of moisture content were measured in all sites (except
site 7) in August, when it was the highest in site 4 (84.45 %).
The lowest moisture content was observed in site 7 in May
(40.58%).
The mean overall moisture content increased from May to
June (Fig. 16). In July, the value decreased slightly, and there
was another considerable increase in August. In some sites
(particularly in July) the error bars showed a large variation
between the samples. The twoway ANOVA analysis for sites
combined with months showed no significant differences for
moisture content.
2.6.3 Fungal biomass in forest litter
A decrease in the ergosterol content was observed in all
sites (except site 8) from May to June (Fig. 17). In sites 1, 2, 4
and 7 the decrease continued in July, when the lowest ergos-
terol content is shown in site 2. In August, the values in-
creased again (except sites 5 and 8), particularly dramatically
in sites 2, 3 and 4. The highest ergosterol content was ob-
served in site 4 in August (with one of the samples exceeding
650 µg ergosterol/g dwt forest litter).
Fig. 15 Moisture content in forest litter
The a and b indicated values showed significant differences (ANOVA) between each other but they were not significant different to ab-values; error bars show
standard deviations
V. Krivtsov et al. / Acta Ecologica Sinica, 2007, 27(3): 813–836

The mean overall changes in the forest litter ergosterol con-
tent are shown in Fig. 18. A decrease was registered from May
to June, and gradual increase was observed from June through
till August, when the mean seasonal ergosterol content was the
highest. It can also be seen that the variation between the
sampling sites was extremely high, hence the large error bars.
The ANOVA analysis (with Tukey’s honesty test) showed
no significant differences in ergosterol levels between the sites,
and the twoway ANOVA analysis for site×month interactions
revealed no significant differences for the ergosterol content.
2.6.4 Bacteria in forest litter
In all sites (except site 8) a notable increase in bacterial
numbers was observed from May to June (Fig. 19). Also in all
sites, however, a decrease (especially high in sites 3, 4, 5 and

Fig. 16 Average moisture content in forest litter
The differently labeled values (a,b,c) indicate significantly different values (ANOVA); error bars show standard deviations
Fig. 17 Ergosterol content in forest litter

Fig. 18 Average ergosterol content in forest litter
The values labeled with the letters a and b showed significant differences between each other with ANOVA; the value indicated with ab is not significantly
different to the a- and b-values; the error bars show standard deviation
V. Krivtsov et al. / Acta Ecologica Sinica, 2007, 27(3): 813–836

6) was shown from June to July. Later in August the bacterial
numbers increased again in most sites. Considerable changes
in bacterial numbers were registered in the birch sites (sites 3
and 6) and the grassland site (site 4). The most dramatic in-
creases were in site 4 (grassland) from May to June and from
July to August. The changes in the birch-beech mixed sites (2
and 7) and the beech sites (1 and 8) were respectively lower
than those in the birch sites. The numbers of bacteria were the
hig-hest in site 4 in August (1.93×109 bacteria/g dwt forest
litter) and the lowest in site 1 in August (2.29×108), indicating
a great variability in bacterial abundance in this month.
Overall numbers of bacteria in forest litter showed a notable
increase from May to June (Fig. 20). A decrease, however,
was observed from June when the bacterial numbers were the
highest (1.05×109 bacteria/g dwt forest litter) to July when the
numbers were the lowest (4.90×108 bacteria/g dwt forest litter,
respectively).
After the decrease in July the bacteria population increased
again in August as can be seen in the graph. There was a great
variability between the 8 plots shown through the large error
bars. The combined ANOVA analysis for bacteria versus site×
month showed significant differences (p =0.000).
3 Discussion
During the period of this research, the values of most variables
showed considerable changes. These changes, presumably, reflect
the dynamics of biological community over the study period.
Superimposed on these seasonal changes was variability owing
to biological interactions and habitat characteristics. The de-
tails of the patterns observed for a key selection of the vari-
ables studied are discussed below with the help of correlation
analysis (Tables 1–2) and stepwise regression modelling (Ap-
pendixes 1–2).
3.1 Soil pH
Fig. 19 Bacteria counts in forest litter
The values labeled with the letters a and d are significantly different (ANOVA) from each other; the ab labeled value is significantly different to d and cd labeled
values but not to a and abcd labeled value; the abc-values are significantly different to d-value but not to all other values; the bcd indicated value is significantly
different to a-value but not to the others; the cd-value differs significantly to a- and ab-values; The error bars show standard deviations

Fig. 20 Average bacteria counts in forest litter
The values labeled with different letters (a,b) showed significant differences in the ANOVA analysis; the Error bars show standard deviations
V. Krivtsov et al. / Acta Ecologica Sinica, 2007, 27(3): 813–836

Soil chemistry and its pH largely depend on the geochemi-
cal characteristics of parent material, and are also influenced
by vegetation, e.g. litter properties[49]. The acid soil in the
whole investigated area is typical for deciduous tree vegeta-
tion, such as birch and beech trees, which produce acidic leaf-
litter[50]. Hence the different vegetations could be a contribut-
Table 1 Correlations between soil variables
Pearson Correlation Value
Organic
content
Moisture
content
pH Gloma-
lin
Ergos-
terol
Root
content
Micr.
nutrient
content
Bacteria Micro-bial
feeding
nematodes
Plant
feeding
nematodes
Moisture content 0.478 0.369 0.241
pH − 0.208
Glomalin 0.352 0.369 − 0.208 0.229 0.297
Ergosterol 0.565 0.229 0.379
Root content
Ratio of soil
particles (>2mm/
<2 mm)
0.225 − 0.204 0.195
Mircobial
nutrient content
0.332 0.379 0.479 − 0.338
Microbial feeding
nematodes
0.372 0.387 0.479
Plant feeding
nematodes
0.241 0.297 − 0.338
Other
invertebrates
0.372 0.313 0.207 0.224 0.219 0.199 0.296
Total nematodes 0.402 0.362 0.280 0.885 0.410
Only significant (p<0.05) correlations are given; negative correlations are listed in bold

Table 2 Correlations between leaf litter variables
Microbial feeders Plant feeders Other invertebrates Total nematodes Bacteria Ergosterol
Beech − 0.211 − 0.202 − 0.300
Wood
Seeds − 0.228 − 0.226 − 0.327
Roots − 0.174 0.314
Moss 0.178 0.177
Oak
Grass 0.582 0.420 0.331 0.556 0.175
Lichens − 0.178
Fern 0.244
Birch 0.505
Fragments − 0.174 − 0.245 − 0.290
Moisture content 0.445 0.292 0.195 0.471 0.335
Ergosterol 0.448 0.406 0.305
Flagellates 0.299 0.194
Ciliates 0.521 0.209 0.270
Amoebae 0.490 0.453 0.352
Bacteria 0.418 0.389 0.177
Microbial feeding nematodes 0.655 0.575 0.265 0.418 0.448
Plant feeding nematodes 0.655 0.442 0.389 0.406
Other invertebrates 0.575 0.442 0.177 0.305
Total nematodes 0.265
Only significant ( p < 0.05) correlations are given; negative correlations are listed in bold
V. Krivtsov et al. / Acta Ecologica Sinica, 2007, 27(3): 813–836

ing factor to the changes observed in pH[51]. Sites 1, 2 and 8,
which are mostly associated with beech, were more acidic
than the other sites, dominated by birch, oak or grass, which is
in line with the research conducted elsewhere[50].
The temporal, as well as the sitewise changes in the pH lev-
els, could also have been influenced by other factors, particu-
larly by the microbial content of the soil. Considering the
monthly stepwise regression models, several interactions were
found for pH in the soil, including those with microbial nutri-
ent content (for August and combined dataset), organic con-
tent (for combined dataset) and bacteria (for June and July
data). In addition, the results of correlation analysis suggest
that local changes in pH might have been influenced by fungi.
Fungi have the ability to alter the pH of their surroundings
during growth. They tend to lower ambient pH through cation
exchange[52], and changes in the amount of fungal mycelium
may therefore correspond to changes in the pH of the upper
sections of the soil profile. The Pearson correlation table (Ta-
ble 1) shows that in this study the glomalin content of soil
correlates significantly negatively with pH at a p-value <0.02.
That means there is a higher abundance of glomalin (bio-
marker for arbuscular mycorrhizal fungi) when the soil pH is
more acidic. Therefore, changes in the biomass of arbuscular
mycorrhizal fungi are reflected in changes of the soil pH.
It is also known that some fungi are able to grow slowly at
pH extremes. In nature mycorrizhal fungi develop symbiosis
with plant roots in soils of pH 3, and some wood-decaying
basidiomycetes commonly tolerate similar pH values in the
wood[53]. pH of the soil in the Heron Wood Reserve throughout
all the sites and months was around pH 3. This suggests that
fungi present in the Reserve may particularly be well repre-
sented by mycorrhizal fungi and wood decaying basidiomy-
cetes, which is also supported by the results of our long-term
monitoring studies.
3.2 Moisture content
Water availability is one of the most important environmental
factors affecting the growth of fauna and flora in an ecosys-
tem[54]. It is essential for transport and metabolism in all cells.
Hence the results show significant correlations (p<0.05) of
moisture content with the soil and forest litter biota, most
notably in forest litter where the moisture content is affected
more rapidly by abiotic factors than that in soil. In forest litter,
positive correlations of moisture content with bacteria and
ergosterol content were observed, while in soil there was a
positive correlation of moisture content with glomalin. The
importance of moisture to the dynamics of microorganisms
was further supported by the results of stepwise regression
models. The moisture content was among the important pre-
dictors of the microbial nutrient content in May and in the
combined data, and of the organic content in May, in July and
in the combined data, which is in line with the contemporary
microbiological theory[54].
3.3 Soil organic content and total microbial biomass
The soil type and climate determine the type of vegetation
in a particular area, but soil characteristics are at the same
time moulded by the vegetation, particularly the mass and
nature of the soil organic matter[55]. Hence there were pro-
nounced differences in soil organic content and total microbial
biomass between the soil samples extracted at Heron Wood
throughout the 8 sites representing different vegetations. The
highest organic content present in the soil was measured in
July. Peaks of temperature and rainfall were also recorded in
July, thus suggesting that the production of organic matter
might have depended on the abiotic parameters.
Organic matter derived by decomposition of plant litter,
animal carcasses, faeces and microbial cells may be termed
the cellular fraction. A second component of the soil organic
matter, humus, is a mixture of highly heterogeneous complex
polymeric substances, which forms colloidal particles often
physically associated with the inorganic soil minerals. These
particles, in turn, determine certain chemical and physical
characteristics of soil, e.g. moisture conditions or nutrient
availability[55]. The statistical relationships revealed that the
soil organic content may, therefore, have reflected local differ-
ences in the abundance and composition of these components.
The values of soil moisture and microbial nutrient content
were positively correlated (p-value< 0.05) to the soil organic
content. Considering the stepwise regression models (see
Appendix 2), organic content was also interrelated to the ratio
of soil particles (> 2 mm/ < 2 mm for July and combined data),
which leads to the suggestion that organic content is related to
the particle size composition of the soil. Glomalin and ergos-
terol (fungal biomarkers) increased when the organic content
increased and showed significant correlation (p < 0.05) with
the soil organic content. Positive significant (p < 0.05) correla-
tion was also found between soil organic content and total
nematodes, microbial feeders and other invertebrates (Krivtsov
et al., unpublished). Both fungi and invertebrates utilise or-
ganic matter directly and therefore compete for the same
limiting resources and affect each other’s growing poten-
tial[1,56]. The correlation might also have reflected the fact that
fungi (dead and living) constituted a significant proportion of
the organic matter.
3.4 Content of roots in soil
The soil system is interspersed with a net of dead, living
and growing roots of the different forest vegetation. Roots are
mainly responsible for the water and soluted nutrient supply of
the plants, and are often involved in symbiosis with bacteria as
well as Mycorrhizas, which helps most vascular plant taxa tap
scarce N and P[1,56]. Hence the root content was expected to be
related to variables like moisture content, bacteria population
and growth of mycorrizhal fungi. Stepwise regression models
showed interactions of root content with ergosterol contents
(fungal biomarker) in soil in May and in combined data (Ap-
V. Krivtsov et al. / Acta Ecologica Sinica, 2007, 27(3): 813–836

pendix 2). In forest litter significant correlation (p < 0.05) of
root content with bacteria was recorded. However, Pearson
correlation analysis for soil data showed no significant corre-
lation of root content with any other important variable.
The values of glomalin, which arguably represent the amount
of arbuscular mycorrhizal fungi, from May to July, seemed to
be related to the measured root content (not shown). Both
variables showed a peak in May and a gradual decrease till
July. August patterns were, however, different, and hence
Pearson correlation tests showed no significant correlations.
The high value of roots in May is suggested to be the result of
spring growing of the roots.
It should be noted that large variation of the soil root con-
tent was observed between the samples. It is known that roots
are not continuously spread in various sizes in the soil[57]. As a
result it could be difficult to measure their content with the
soil cores used. This uncertainty might have reflected the lack
of recorded significant relationships.
3.5 Bacteria
3.5.1 Soil
The temporal changes in the number of bacteria present de-
pend on various biotic and abiotic factors in soil, such as the
abundance of other soil biota, nutrient content, soil moisture
content, temperature and soil texture that predetermine the
water permeability of the soil. The Pearson correlation analysis
for bacteria (performed on the overall dataset) showed no
significant correlation of soil bacteria with any single tested
factor. It can, therefore, be suggested that over the period of
the study different factors were important for the soil bacteria.
However, stepwise regression models for bacteria showed
variable interactions of soil bacteria with other soil variables
(Appendix 2). In June and July bacteria showed interactions
with pH. Comparing Figs. 4 and 14 it can be observed that
bacterial numbers increased when pH decreased. In August,
stepwise regression modelling showed interactions of bacteria
with ergosterol, amoebae, ciliates and the collective variable
´other invertebrates`. Relationships with microinvertebrates
suggest prey-predator interactions as bacteria are a food re-
source for protozoa (such as amoebae, ciliates) and other
invertebrates. Relationships with fungi may indicate attraction
to the same resource. The combined monthly data showed
interactions of bacteria with soil organic content and glomalin.
The former could be a food resource for bacteria, whilst the
latter may indicate the attraction of both types of decomposers
to the organically-rich hot spots. These relationships provide
important information on the complexity of the factors influ-
encing bacterial dynamics in soil.
Differences in the population abundance between the sites
were not significant (oneway ANOVA test) as well as differ-
ences between the months. Hence it can be suggested that in
Heron Wood during the warm period of the year the popula-
tion of soil bacteria is relatively stable in the whole area, and
the bacterial dynamics may be more affected by the overall
soil composition and conditions in this area than by sitewise
differences of the vegetation. As was already mentioned, during
the study period the soil was not so much affected by changes
in temperature and rainfall because of the covering forest litter
layer, and this buffering effect might have led to the relative
uniformity in the overall numbers of soil bacteria.
3.5.2 Forest litter
In contrast with soil, bacteria in forest litter were much
more abundant, (compare Figures 13-14 with Figures 19-20).
One reason could be the various stages of decomposition
present in the litter[58]. This would lead to greater amounts of
nutrients present to be used by bacteria and other organisms,
allowing them to exist in higher numbers without becoming
limited by nutrients. An increase in available nutrients and
numbers of bacteria would lead to higher amounts of other
organisms to be able to live in the litter through increased
variety of prey[56]. Increased numbers of organisms in litter
were also noted for nematodes and other invertebrates
(Krivtsov et al., unpublished).
In forest litter, as the uppermost layer in the soil system, the
bacteria numbers may be expected to show strong dependence
on climate factors such as temperature and moisture. In this
study, the results appear to suggest that bacterial numbers
were lower in months of hotter and dryer weather. In May,
when the moisture content was the lowest and the maximal
temperature was one of the highest, the least abundant bacteria
population was found. The stepwise regression models con-
firm the relation of bacteria numbers with litter moisture as
they show relationships of bacteria with the moisture content
in May and June and in the combined model. The importance
of moisture content might have been particularly pronounced
in the grassland site (4), where there is no protection from
trees and bushes. For example in the months with the highest
moisture content (the highest total rainfall, 10 days before
sampling), i.e. August and July, bacteria were the most abun-
dant and showed higher numbers than in the forest sites.
The Pearson correlation analysis showed significant corre-
lation (p<0.05) between bacteria and the moisture content of
forest litter, thus supporting considerations already made.
Correlation analysis also showed correlation of bacteria with a
number of forest litter components. Bacteria were significantly
(p<0.05) positively correlated with roots, moss and grass, and
significantly negatively correlated with beech, seeds and fra-
gments. These relationships are also apparent from the com-
parative inspection of the graphs displaying changes in litter
composition and the bacterial abundance in the different sites.
Hence in the beech and birch-beech sites (sites 1, 2, 7, 8)
where beech leaves, seeds and fragments were the most abun-
dant, the bacterial population was the lowest during all months
of the study. A reason could be that beech leaves contain more
cellulose than birch leaves and grass, and are not so easy for
V. Krivtsov et al. / Acta Ecologica Sinica, 2007, 27(3): 813–836

bacteria to decompose In addition, beech leaves are known to
contain some substances toxic for bacteria. By contrast, in
sites 4, 5 and 6 (grassland and birch forest), where the highest
amounts of grass, roots and moss were found, the bacterial
population was at its highest, which was confirmed by the
correlation of bacteria with roots, moss and grass.
Significant (p < 0.05) positive correlations between bacteria
and other biota were shown for microbial and plant feeding
nematodes, and other invertebrates (such as tardigrades and
enchytraeid worms). Bacteria are preyed upon by microbial
feeding nematodes, and hence large numbers of bacteria form
optimal conditions for microbial feeders. Plant feeding nema-
todes are suggested to feed on roots, grass and plant litter and
they are found in high numbers in the areas that are preferred
by bacteria. In addition, nematodes, tardigrades and enchy-
traeid worms stimulate bacteria by their excretions and facili-
tate the decomposition; both processes would lead to greater
food resources for bacteria.
3.6 Fungi
In an ecosystem, fungal growth is affected by a number of
abiotic and biotic factors. Severe stresses may be tolerated by
fungi, which either possess appropriate physiological charac-
teristics or can adapt to the stresses through a temporary al-
teration in their development. Hence, a combination of water
availability, temperature, pH, nutrient availability and other
fauna and flora of the system will affect the fungal biomass
levels[53], and the latter may, therefore, be expected to show
significant relationships with the important variables influenc-
ing the fungal growth and losses. In the research presented
here, a number of significant interactions involving the fungal
biomarkers examined were revealed by the Pearson correla-
tion analysis (Tables 1–2) and stepwise regression modelling
(Appendixes 1–2).
3.6.1 Fungi in soil and forest litter
Ergosterol levels in forest litter were dramatically higher
(approximately 30 times) than the values measured in soil.
However, the ergosterol levels in forest litter revealed no
significant differences between the sites studied, whilst for the
soil levels significant differences were shown by the analysis
of variance.
Access to water is one of the factors important for fungal
growth. As the water availability in soil did not show dramatic
changes, there was always enough moisture for fungal growth.
Consequently, ergosterol was not significantly related to the
soil moisture content. By contrast, in forest litter, where mois-
ture content is more changeable than that in the soil, ergosterol
content was significantly related (p < 0.05) to the moisture
content. An explanation for the significant correlation of forest
litter ergosterol with grass and moss might also relate to the
water availability, because grass and moss require areas with
high moisture. Furthermore, the importance of moisture con-
tent for forest litter biota might partly have been responsible
for the revealed correlations between ergosterol, microbial and
plant feeding nematodes, and other invertebrates.
Considering fungal levels in the 8 sites studied, the grass-
land (site 4) and the birch forest sites (sites 3, 5 and 6) showed
the highest forest litter ergosterol contents, particularly in
August. In these sites grass, birch leaves and moss contents of
the litter samples were the highest. In soil, however, the fungi
present were more abundant in the beech forest sites (sites 1
and 8).
3.6.2 Fungal Interactions with other biota
Significant correlations (p<0.05) of ergosterol contents in
soil and litter and the glomalin content in soil with various
groups of nematodes and other invertebrates considered in this
project may have resulted from a combination of direct and
indirect interactions. Some nematodes are known to feed on
fungal hyphae[56], but only little evidence exists that fungal
feeding nematodes utilise arbuscular mycorrhizal fungi hy-
phae in soil. Whilst nematodes feed on fungi, some specialised
fungi also trap, kill and colonise nematodes. Hence nematode
killing fungi could influence the numbers of nematodes as
well. A number of other meiofaunal groups (such as enchy-
traeid worms) also feed on fungi[56], which explains the sig-
nificant correlation of fungal biomarkers (p<0.05) with ‘other
invertebrates’.
It should be noted that the data on microbial feeding nema-
todes included both fungal and bacterial feeding nematodes.
Hence, the significant correlation between microbial feeding
nematodes and fungal biomass might, at least partly, have
resulted from a direct interaction. An indirect component of
this relationship may, however, have reflected the fact that
fungi were leaching nutrients from plant material that was
used by bacteria preyed upon by microbial feeding nematodes.
The results of stepwise regression modelling revealed that
ergosterol in forest litter (combined model) and glomalin in
soil (combined model) showed significant interactions with
bacteria. Both relationships were negative, which might have
reflected the competition between alternative decomposition
pathways.
3.6.3 Comparison of relationships revealed by fungal
biomarkers
A point of interest relates to differences and similarities in
the interactions exhibited by soil ergosterol and glomalin
contents, which could be regarded as proxies for total live and
arbuscular mycorrhizal fungi, respectively[42,59]. Glomalin and
ergosterol were significantly correlated (p<0.05) by data of the
whole sampling period, but there were also notable site-spe-
cific differences in separate months, which leads to the sug-
gestion that fungi, other than arbuscular mychorrizal, have
considerably contributed to the ergosterol variability in some
months. Both fungal biomarkers showed significant relation-
ships to tardigrades and enchytraeid worms. Glomalin was
significantly negatively related to pH and showed also signifi-
V. Krivtsov et al. / Acta Ecologica Sinica, 2007, 27(3): 813–836

cant correlation with plant feeding nematodes, whereas ergos-
terol showed significant correlation with the microbial nutrient
content, microbial feeders and total numbers of nematodes,
and a negative correlation with the ratio of soil particles (>2
mm/ < 2mm).
Considering the above, it appears that, along with certain
similarities, there were remarkable differences between the
patterns of relationships exhibited by two fungal biomarkers
— glomalin and ergosterol. It should be noted that apart from
arbuscular mycorrhizal fungi, the fungal component of forest
soil also includes ectomycorrhizal, saprotrophic and parasitic
fungi[52]. These groups could be expected to react differently
to local changes in biotic and abiotic parameters, and the
stepwise regression modelling appears to be useful in reveal-
ing this differential pattern. It should also be noted that the
usage of glomalin as a proxy for AM fungi has recently been
questioned. Hence the results presented here might be of use
for the reassessment of the biochemical and ecological roles of
this compound.
3.7 Differences between statistical analysis for separate
months and overall dataset
Stepwise regression modelling applied to the results of this
research provided useful information on functional interrela-
tions among various ecosystem components. It should be
noted that the models were generally stronger (i.e. in terms of
the percent of the total variance explained) for separate
monthly datasets (except models for microbial nutrient con-
tent), and not for the overall dataset combining all data. Fur-
thermore, the pattern of relationships revealed that it differed
between the months studied. For example, in July glomalin in
soil showed significant relationships with pH and C-content,
while in August it was significantly related to the soil organic
content and amoebae. It is also worth pointing out that some
of the relationships appeared to change sign between the sam-
pling dates. For example, the ergosterol content in soil was
shown to be significantly positively related to glomalin in
May and August, demonstrating an overall significant rela-
tionship on the combined dataset. In June and July, however,
the relationship between these two variables was negative.
It therefore appears that the monthly sampling provided 4
separate snapshots of the ecosystem state, and the pattern of
the relationships recorded in the regression models was con-
siderably different for each month. Considering the complex-
ity of the multivariate interactions studied, this is not surpris-
ing. In a natural ecosystem different processes could become
prominent at different stages of a seasonal cycle, and the
changes of patterns observed might, therefore, have reflected
temporal changes in the overall ecosystem functioning.
4 Concluding remarks
The ecological interactions registered in this research were
indicative of the specific conditions of the study, and may,
therefore, prove useful for future reference. The results of this
study highlighted the complexity of factors influencing tempo-
ral dynamics and spatial variability of fungal biomass in forest
soil and litter. Most of the registered interactions appeared to
be transient, i.e. manifested only at certain parts of the re-
search period. The complexity and transience of ecological
relationships of fungi and other forest soil and litter biota are
in line with our previous studies[28,29,34,35,41], and should be
borne in mind while interpreting environmental datasets.
Analyses showed that the dynamics of microbial commu-
nity in beech and birch plots appears to be considerably dif-
ferent. In particular, the abundance of arbuscular mycorrhizal
fungi (as indicated by levels of easily extractable glomalin)
was greater in birch plots, corresponding to a more dense
grass cover in this habitat. Total soil fungal biomass (as indi-
cated by ergosterol measurements), however, was higher in
beech plots, thus suggesting a notable difference exhibited by
different members of fungal community (e.g. saprotrophic and
arbuscular mycorrhizal, as opposed to ectomycorrhizal fungi).
On the other hand, the abundance of soil bacteria was not
significantly different between beech and birch plots. In forest
litter, however, the pattern was reversed: fungal levels did not
appear to differ much, whilst bacterial abundance in birch
plots was considerably higher.
Based on the patterns observed, we hypothesised that de-
composition in beech-dominated habitats is delayed (i.e. in
comparison with birch-dominated) because of the differences
in the extent to which these plants create and maintain a litter
and humus layer[60]. Decomposition rates and patterns are
dependent upon chemical composition of litter[61]. As the leaf
litter produced by beech trees is of lower quality (e.g. for
microflora and microfauna) compared with birch[62], the de-
composition in beech-dominated habitats occurs somewhat
deeper in the soil profile, and may be carried out with some
considerable involvement of ectomycorrhizal fungi.
This hypothesis also appears to be supported by the fungal
fruiting records. Despite a relative scarcity of the grass cover
(and hence lesser importance of arbuscular mycorrhizal fungi),
beech-dominated habitats feature an impressive number of
ectomycorrhizal fungi, and the higher abundance and biodi-
versity of these fungi in beech plots (i.e. in comparison with
birch plots) appear to correspond to the higher soil ergosterol
content found in the beech-dominated habitats.
References
[ 1 ] Carroll G C, Wicklow D T. The fungal community. Its organi-
sation and role in the ecosystem. 2nd ed. New York: Marcel
Dekker, 1992.
[ 2 ] Watling R, Gregory N M. Larger fungi from Kashmir. Nova
Hedwigia, 1980, 32: 493–564.
V. Krivtsov et al. / Acta Ecologica Sinica, 2007, 27(3): 813–836

[ 3 ] Keizer P J, Arnolds E. Succession of ectomycorrhizal fungi in
roadside verges planted with Common Oak (Quercus robur L.)
in Drenthe, the Netherlands. Mycorrhiza, 1994, 4: 147–159.
[ 4 ] Watling R. Assessment of fungal diversity—macromycetes,
the problems. Canadian Journal of Botany, 1995, 73: S15–S24.
[ 5 ] Watling R. Pulling the threads together: Habitat diversity. Bio-
diversity and Conservation, 1997, 6: 753–763.
[ 6 ] Watling R. A fungus sanctuary and reserve. Mycophile, 2001,
42: 1, 6–7.
[ 7 ] Watling R. The relationships and possible distributional pat-
terns of boletes in Southeast Asia. Mycological Research, 2001,
105: 1440–1448.
[ 8 ] Heilmann-Clausen J. A gradient analysis of communities of ma-
crofungi and slime moulds on decaying beech logs. Mycologi-
cal Research, 2001, 105: 575–596.
[ 9 ] Gardes M, Bruns T D. Community structure of ectomycorrhi-
zal fungi in a Pinus muricata forest: Above and below ground
views. Canadian Journal of Botany, 1996, 74: 1572–1583.
[10] Singh S P, Pande K, Upadhyay V P, Singh J S. Fungal commu-
nities associated with the decomposition of a common leaf lit-
ter (Quercus leucotrichophora a Camus) along an elevational
transect in the Central Himalaya. Biology and Fertility of Soils,
1990, 9: 245–251.
[11] Becker M, Kraepelin G, Lamprecht I. Microcalorimetric inves-
tigations of microbial activities decomposition of needle litter
under laboratory conditions. Thermochimica Acta, 1991, 187:
15–25.
[12] Cornejo F H, Varela A, Wright S J. Tropical forest litter decom-
position under seasonal drought — nutrient release, fungi and
bacteria. Oikos, 1994, 70: 183–190.
[13] Watling R. Macrofungi Associated with British Willows. Pro-
ceedings of the Royal Society of Edinburgh, Section B- Bio-
logical Sciences, 1992, 98: 135–147.
[14] Mason P A, Last F T, Pelham J, Ingleby K. Ecology of some
fungi associated with an aging stand of birches (Betula pendula
and Betula pubescens). Forest Ecology and Management, 1982, 4:
19–39.
[15] Danielson R M, Visser S. Host response to inoculation and
behavior of introduced and indigenous ectomycorrhizal fungi
of Jack Pine grown on oil sands tailings. Canadian Journal of
Forest Research, 1989, 19: 1412–1421.
[16] Golovchenko A V, Polyanskaya L M. Seasonal dynamics of
population and biomass of microorganisms in the soil profile.
Eurasian Soil Science, 1996, 29: 1145–1150.
[17] Polyanskaya L M, Sveshnikova A A, Vladychenskii A S,
Zvyagintsev D G. Microbial biomass pools in brown and black
brown soils of Southwest Tien Shan. Microbiology, 1995, 64:
451–458.
[18] Ohtonen R, Munson A, Brand D. Soil microbial community
response to silvicultural intervention in coniferous plantation
ecosystems. Ecological Applications, 1992, 2: 363–375.
[19] Sidorova I, Velikanov L L. The influence of higher Basidiomy-
cetes on the myco and mycrobiota structures in soils and litters
of forest ecosystems. I. The influence of Basidiomycetes on
quantity of fungi and bacteria. Mikologiyai Fitopatologiya,
1997, 31: 20–26.
[20] Zhang Q, Zak J C. Potential physiological activities of fungi
and bacteria in relation to plant litter decomposition along a
gap size gradient in a natural subtropical forest. Microbial
Ecology, 1998, 35: 172–179.
[21] Griffiths B S, Ritz K, Ebblewhite N, Dobson G. Soil microbial
community structure: Effects of substrate loading rates. Soil
Biology & Biochemistry, 1999, 31: 145–153.
[22] Moller J, Miller M, Kjoller A. Fungal bacterial interaction on
beech leaves: influence on decomposition and dissolved or-
ganic carbon quality. Soil Biology & Biochemistry, 1999, 31:
367–374.
[23] Okoh A I, Babalola G O, Olaniran A O. Aerobic heterotrophic
bacterial and fungal communities in the topsoil of Omo Bio-
sphere Reserve in southwestern Nigeria. Biotropica, 2000, 32:
208–212.
[24] Kshattriya S, Sharma G D, Mishra R R. Fungal succession and
microbes on leaf litters in 2 degraded tropical forests of North-
east India. Pedobiologia, 1994, 38:125–137.
[25] Alekhina L K, Polyanskaya L M, Dobrovol′skaya T G. Popula-
tion dynamics of microorganisms in the soils of the Central
Forest Reserve (model experiments). Eurasian Soil Science,
2001, 34: 88–91.
[26] Griffiths B S, Ritz K, Wheatley R, Kuan H L, Boag B, Chris-
tensen S, Ekelund F, Sorensen S J, Muller S, Bloem J. An ex-
amination of the biodiversity ecosystem function relationship
in arable soil microbial communities. Soil Biology & Bio-
chemistry, 2001, 33: 1713–1722.
[27] Maraun M, Scheu S. Changes in microbial biomass, respira-
tion and nutrient status of beech (Fagus sylvatica) leaf litter
processed by millipedes (Glomeris marginata). Oecologia,
1996, 107: 131–140.
[28] Krivtsov V, Liddell K, Garside A, Bezginova T, Salmond R,
Thompson J, Griffiths B, Staines H J, Watling R, Brendler A,
Palfreyman J W. Some aspects of complex interactions involv-
ing soil mesofauna: Analysis of the results from a Scottish
woodland. Ecological Modelling, 2003, 170: 441–452.
[29] Krivtsov V, Griffiths B, Salmond R, Liddell K, Garside A,
Bezginova T, Thompson J, Staines H J, Watling R, Palfreyman
J W. Some aspects of interrelations between fungi and other
biota in forest soil. Mycological Research, 2004, 108: 933–946.
[30] Mikola J, Sulkava P. Responses of microbial feeding nematodes
to organic matter distribution and predation in experimental
soil habitat. Soil Biology & Biochemistry, 2001, 33: 811–817.
[31] Perez-Moreno J, Read D J. Nutrient transfer from soil nema-
todes to plants: a direct pathway provided by the mycorrhizal
mycelial network. Plant Cell and Environment, 2001, 24:
1219–1226.
V. Krivtsov et al. / Acta Ecologica Sinica, 2007, 27(3): 813–836

[32] Pouyat R V, Parmelee R W, Carreiro M M. Environmental
effects of forest soil lnvertebrate and fungal densities in oak
stands along an urban rural land use gradient. Pedobiologia,
1994, 38: 385–399.
[33] Krivtsov V, Garside A, Thompson J, Bezginova T, Salmond R,
Liddell K, Griffiths B, Staines H J, Watling R, Palfreyman J W.
Interrelations between soil nematodes, bacteria, fungi and pro-
tozoa in the Dawyck cryptogamic sanctuary′ in winter. In: Ryss
A Y ed. IV International Nematology Symposium, Moscow:
Moscow University, 2001, 64–65, 173.
[34] Krivtsov V, Liddell K, Bezginova T, Salmond R, Garside A,
Thompson J, Palfreyman J W, Staines H J, Watling R, Brendler
A, Griffiths B. Ecological interactions of heterotrophic flagel-
lates, ciliates and naked amoebae in forest litter of the Dawyck
Cryptogamic Sanctuary (Scotland, UK). European Journal of
Protistology, 2003, 39: 183–198.
[35] Krivtsov V, Walker S J J, Staines H J, Watling R, Burt Smith G,
Garside A. Integrative analysis of ecological patterns in an un-
tended temperate woodland utilising standard and customised
software. Environmental Modelling & Software, 2004, 19:
325–335.
[36] Krivtsov V, Liddell K, Bezginova T, Salmond R, Staines H J,
Watling R, Garside A, Thompson J A, Griffiths B S, Brendler
A. Forest litter bacteria: Relationships with fungi, microfauna,
and litter composition over a winter spring period. Polish Jour-
nal of Ecology, 2005, 53(3): 383–394.
[37] Walker S J J, Watling R, Staines H J, Garside A, Knott D,
Palfreyman J W, Krivtsov V. Modelling an Untended Scottish
Forest Ecosystem Utilising Standard and Customised Software.
Pages Accepted in IEMSS, 2002.
[38] Bezginova T, Thompson J, Staines H J, Salmond R, Krivtsov V,
Garside A, Griffiths B, Liddell K, Watling R, Palfreyman J.
Interrelations between leaf litter composition and nematodes in
the Heron Wood Reserve′, Peebleshire, Scotland. In: Ryss A Y
ed. IV International Nematology Symposium, Moscow, June
2001. 62–63. Moscow: Moscow University, 2001.
[39] Thompson J, Brendler A, Staines H, Salmond R, Bezginova T,
Palfreyman J, Krivtsov V. Comparison of two methods for
nematode extraction. In: A.Y. Ryss ed. IV International Nema-
tology Symposium, Moscow: Moscow University, 2001, 79–
80.
[40] Krivtsov V, Garside A, Bezginova T, Thompson J, Palfreyman
J W, Salmond R, Liddell K, Brendler A, Griffiths B S, Watling
R, Staines H J. Ecological study of the forest litter meiofauna
of a unique Scottish woodland. Animal Biology, 2006.
[41] Krivtsov V, Watling R, Walker S J J, Knott D, Palfreyman J W,
Staines H J. Analysis of fungal fruiting patterns at the Dawyck
Botanic Garden. Ecological Modelling, 2003, 170: 393–406.
[42] Scheu S, Parkinson D. Changes in bacterial and fungal biomass
C, bacterial and fungal biovolume and ergosterol content after
drying, remoistening and incubation of different layers of cool
temperate forest soils. Soil Biology & Biochemistry, 1994, 26:
1515–1525.
[43] Ruzicka S, Norman M D P, Harris J A. Rapid ultrasonication
method to determine ergosterol concentration in soil. Soil Bi-
ology and Biochemistry, 1995, 27: 1215–1217.
[44] Wright S F, Upadhyaya A. A survey of soils for aggregate sta-
bility and glomalin, a glycoprotein produced by hyphae of ar-
buscular mycorrhizal fungi. Plant and Soil, 1998, 198: 97–107.
[45] Nannipieri P, Alef K. Methods in Applied Soil Microbiology
and Biochemistry. London: Academic Press Limited, 1995.
[46] Nunan N, Morgan M A, Herlihy M. Ultraviolet absorbance
(280 nm) of compounds released from soil during chloroform
fumigation as an estimate of the microbial biomass. Soil Biol-
ogy and Biochemistry, 1998, 30: 1599–1603.
[47] Ingold G, Hudson H J. The biology of fungi. 6th ed. London:
Chapman and Hall, 1993.
[48] Gadgil R L, Gadgil P D. Suppression of litter decomposition by
mycorrhizal roots of Pinus radiata. New Zealand Journal of
Forest Science, 1975, 5: 33–41.
[49] Vestin J L K, Nambu K, van Hees P A W, Bylund D, Lundström
U S. The influence of alkaline and non alkaline parent material
on soil chemistry. Geoderma, 2006.
[50] Hagen-Thorn A, Callesen I, Armolaitis K, Nihlgard B. The im-
pact of six European tree species on the chemistry of mineral
topsoil in forest plantations on former agricultural land. Forest
Ecology and Management, 2004, 195(3): 373–384.
[51] Cassagne N, Bal Serin M C, Gers C, Gauquelin T. Changes in
humus properties and collembolan communities following the
replanting of beech forests with spruce. Pedobiologia, 2004,
48(3): 267–276.
[52] Deacon J W. Introduction to modern mycology. Oxford: Black-
well Scientific Publications, 1984.
[53] Cooke R C, Whipps J M. Ecophysiology of fungi. Oxford:
Blackwell Scientific Publications, 1993.
[54] Miles J. Soil in the ecosystem. In: Fitter A H, Atkinson D, Read
D J, Usher M B eds. Ecological Interactions in Soil, Oxford,
London, Edinburgh, Boston, Palo Alto, Melbourne: Blackwell
Scientific Publications, 1985, 407–427.
[55] Swift M J, Heal O W, Anderson J M. Decomposition in terres-
trial ecosystems. Oxford: Blackwell Scientific, 1979.
[56] Coleman D C, Crossley D A Jr. Fundamentals of soil ecology.
London: Academic Press, 1996.
[57] Fitter A H, Atkinson D, Read D J, Usher M B. Ecological
interactions in soil. Oxford: Blackwell Scientific Publications,
1985.
[58] Gallardo A, Merino J. Control of leaf litter decomposition rate
in a Mediterranean shrubland as indicated by N, P and lignin
concentrations. Pedobiologia, 1999, 43: 64–72.
[59] Wright S F, Upadhyaya A. Quantification of arbuscular my-
corrhizal fungi activity by the glomalin concentration on hy-
phal traps. Mycorrhiza, 1999, 8: 283–285.
V. Krivtsov et al. / Acta Ecologica Sinica, 2007, 27(3): 813–836

[60] Wardle D A. Communities and ecosystems. Linking the above-
ground and belowground components. Princeton: Princeton
University Press, 2002.
[61] McClaugherty C A, Pastor J, Aber, J D, Melillo J M. Forest
litter decomposition in relation to soil nitrogen dynamics and
litter quality. Ecology, 1985, 66(1): 266–275.
[62] Melillo J M, Aber J D, Muratore J F. Nitrogen and lignin con-
trol of hardwood leaf litter decomposition dynamics. Ecology,
1982, 63(3): 621–626.
[63] Griffiths B S, Bardgett R D. Interactions between microbe
feeding invertebrates and soil microorganisms. In: van Elsas J
D, Trevors J T, Wellington E M H eds. Modern soil microbiol-
ogy, New York: Marcel Dekker, 1998. 165–182.
[64] Krivtsov V. Study of cause and effect relationships in the for-
mation of biocenoses. Russian Journal of Ecology, 2001, 32:
230–234.
[65] Krivtsov V, Corliss J, Bellinger E, Sigee D. Indirect regulation
rule for consecutive stages of ecological succession. Ecological
Modelling, 2000, 133: 73–81.
[66] Krivtsov V, Liddell K, Salmond R, Garside A, Thompson J,
Bezginova T, Griffiths B, Staines H J, Watling R, Palfreyman J
W. Analysis of microbial interactions in forest soil. In: Shumny
V K, Kolchanov N A, Fedotov A M eds. Information technolo-
gies application to problems of biodiversity and dynamics of
ecosystems in North Eurasia, Novosibirsk: Institute of Cytol-
ogy and Genetics, 2002. 322–331.
[67] Krivtsov V, Bellinger E, Sigee D. Water and nutrient budgeting
of Rostherne Mere, Cheshire, UK. Nordic Hydrology, 2002, 33:
391–414.
[68] Kurihara Y, Kikkawa J. Trophic relations of decomposers. In:
Kikkawa J, Anderson D J eds. Community ecology: pattern
and process, Melbourne: Blackwell Scientific Publications, 1986.
126–160.
[69] Whiffen L K, Midgley D J, McGee P A. Polyphenolic com-
pounds interfere with quantification of protein in soil extracts
using the Bradford method. Soil Biology and Biochemistry,
39(2): 691–694.
[70] Okoh I A, Badejo M A, Nathaniel I T, Tian G. Studies on the
bacteria, fungi and springtails (Collembola) of an agroforestry
arboretum in Nigeria. Pedobiologia, 1999, 43: 18–27.
[71] Wright S F, Starr J L, Palineanu I C. Changes in aggregate
stability and concentration of glomalin during tillage management
transition. Soil Society of America Journal, 1999, 63: 1825–
1829.

V. Krivtsov et al. / Acta Ecologica Sinica, 2007, 27(3): 813–836

Appendix 1 Stepwise regression models (leaf litter variables)
Ergosterol content Bacteria
Item
May June July Aug. Combined May June July Aug. Combined
Constant − 87.16 188.6 108.6 475.9 270.6 130.1 − 1063 414.7 280.1 75.84
Beech dry
Birch dry
Oak dry
Grass dry 33.0
(9.75)
22.7
(6.27)
Moss dry 7.3
(2.79)
16.3
(3.71)
15.5
(2.47)
9.9
(2.04)
Wood dry
Seeds dry 3.0
(2.54)

Roots dry 10.2
(4.10)

Lichens dry − 525.0
( − 2.08)

Fragments dry − 2.30
( − 2.65)

Moisture content 7.3
(6.49)
3.26
(3.59)
7.7
(2.22)
32.5
(8.49)
6.0
(2.39)
Ergosterol
Bacteria − 0.062
( − 2.23)

Flagellates 0.00135
(3.57)
Ciliates 0.58
(4.07)

Amoebae 0.129
(3.38)
0.0284
(4.47)

Other
Invertebrates
0.141
(3.01)
0.107
(3.47)

Total nematodes 0.0090
(4.16)
− 0.0043
( − 2.39)

R² (%) 61.67 36.57 55.05 46.53 27.30 14.52 74.09 35.88 78.78 46.97
V. Krivtsov et al. / Acta Ecologica Sinica, 2007, 27(3): 813–836


Appendix 2 Stepwise regression models (soil variables)
Total microbial biomass Ergosterol content Glomalin content
Item
May June July Aug. Combined May June July Aug. Combined May June July Aug. Combined
Constant 0.08430 0.7001 3.085 − 4.865 − 1.0643 2.856 ⎯ − 1.595 2.480 1.824 − 0.0804 1.978 6.8845 − 1.733 4.2755
Ratio of soil
particles (>2
mm / <2mm)
0.435
(4.78)
− 0.179
( − 2.73)
Content of
Roots
2.59
(3.24)
2.5
(2.36)

pH 2.81
(3.33)
1.09
(2.24)
− 1.89
( − 2.17)
− 1.15
( − 3.19)
Soil moisure
content
0.035
(2.29)
− 0.109
( − 3.57)
− 0.314
( − 3.42)
0.106
(5.85)
Soil organic
content
0.210
(2.73)
0.231
(4.81)
0.90
(3.59)
0.89
(6.45)
0.261
(7.11)
0.345
(8.99)

Microbial
nutrient content
0.78
(2.89)
0.37
(2.34)

Ergosterol 0.071
(2.73)

Glomalin − 0.25
( − 2.20)
1.20
(2.61)
2.91
(3.90)
0.79
(2.07)

Bacteria − 0.0058
( − 2.31)
Flagellates − 0.0008
( − 2.23)
0.00005
(2.53)

Ciliates
Amoebae − 0.0026
( − 3.43)
− 0.00107
(2.16)
− 0.001
( − 2.78)

Other
invertebrates

Total
nematodes
0.00069
(4.13)
0.00278
(2.91)

R² (%) 15.29 33.32 36.21 26.97 41.64 35.56 68.54 33.60 44.42 63.55 17.60 24.53 73.63 27.43
Coefficients for significant predictors are followed (in brackets) by T ratios. For each model, R² is given in terms of the percent of the total variance explained
V. K
rivtsov et al. / Acta Ecologica Sinica, 2007, 27(3): 813–836
V. Krivtsov et al. / Acta Ecologica Sinica, 2007, 27(3): 813–836




Appendix 2 (continuation)
Soil organic content Soil bacteria
Item
May June July Aug Combined May June July Aug Combined
Constant − 6.503 9.210 0.6678 7.172 5.088156 ⎯ 472.0 − 85.25 107.77 114.1
Ratio of soil particles
(>2 mm/ <2mm)
− 0.87
( − 2.17)
− 0.78
( − 5.23)
− 8.7
( − 2.33)

Content of roots
pH − 2.11
( − 2.71)
− 107.0
( − 3.34)
65.0
(3.50)

Soil moisure content 0.480
(8.41)
0.411
(5.52)
0.399
(10.27)

Soil organic content 2.60
(2.84)
Microbial nutrient
content
6.10
(2.73)
3.81
(4.53)

Ergosterol 0.354
(6.58)
0.162
(4.00)
1.74
(2.68)

Glomalin 1.20
(5.08)
2.14
(8.99)
− 6.1
( − 2.11)
Bacteria
Flagellates
Ciliates − 0.130
( − 2.58)

Amoebae 0.0052
(3.13)
0.00294
(3.57)
0.00199
(2.51)
0.179
(3.24)
0.057
(4.99)

Other invertebrates − 0.278
( − 2.86)

Total nematodes 0.00106
(4.66)

R² (%) 89.67 29.85 75.85 76.81 69.29 27.11 49.00 63.74 7.11

V. K
rivtsov et al. / Acta Ecologica Sinica, 2007, 27(3): 813–836