Soil fauna and site assessment in beech stands of the Belgian Ardennes

Abstract

Soil fauna (macrofauna and mesofauna) were sampled in 13 beech forest stands of the Ardenne mountains (Belgium) covering a wide range of acidic humus forms. The composition of soil fauna was well correlated not only with humus form, but also with elevation, phytosociological type, tree growth, mineral content of leaf litter, and a few soil parameters such as pH and C/N ratio. The nature of the mechanisms that can explain these relationships is discussed in light of existing knowledge.

Full text

Soil fauna and site assessment in beech stands
of the Belgian Ardennes
Jean-François Ponge, Pierre Arpin, Francis Sondag, and Ferdinand Delecour
Abstract: Soil fauna (macrofauna and mesofauna) were sampled in 13 beech forest stands of the Ardenne mountains
(Belgium) covering a wide range of acidic humus forms. The composition of soil fauna was well correlated not only with
humus form, but also with elevation, phytosociological type, tree growth, mineral content of leaf litter, and a few soil
parameters such as pH and C/N ratio. The nature of the mechanisms that can explain these relationships is discussed in light of
existing knowledge.
Résumé : La faune du sol (macrofaune et mésofaune) a été échantillonnée dans treize peuplements de hêtre des Ardennes
belges, couvrant une gamme étendue de formes d’humus acides. La composition faunistique est bien corrélée, non seulement
avec la forme d’humus, mais aussi avec l’altitude, le groupement phytosociologique, la croissance des arbres, la
composition minérale de la litière de feuilles et quelques paramètres édaphiques tels que le pH et le rapport C/N. La nature des
mécanismes pouvant expliquer ces relations est discutée, à la lueur des connaissances actuelles.
Introduction
The assessment of site quality for the growth of forest stands
has been based mainly on ground vegetation (Rodenkirchen
1985) and soil features (Turvey and Smethurst 1985). When
the soil type does not change heavily, it has been observed that
strong discrepancies in forest productivity may be explained
by the rate at which litter disappears from the ground surface
(Delecour 1978). This rate, expressed by a coefficient calcu-
lated first by Jenny et al. (1949), was proposed as a site factor
for European beech (Fagus sylvatica L.) forests by Delecour
and Weissen (1981).
The disappearance of canopy litter from the ground surface
(improperly called decomposition) is strongly associated with
humus form, i.e., moder and mor humus are characterized by
a slower rate of disappearance of leaf litter than mull humus
(Van der Drift 1963). This phenomenon has been found to
result from the consumption of litter by fauna and microflora,
which vary in quantity and quality from one site to another
(Toutain 1987; Schaefer and Schauermann 1990; Muys and
Lust 1992).
We contrasted soil macro- and meso-fauna with other site
factors in 13 beech forests of the Belgian Ardennes, which
share the same parent rock and regional climate but strongly
differ in their productivity and humus form. In a previous paper
(David et al. 1993) we characterized mull humus as having a
higher diversity of macrofaunal groups than moder humus.
Nevertheless, the discriminative power of macrofauna was
poor in the moder group (from hemimoder to dysmoder),
where elaterid larvae (Insecta, Coleoptera) were one of the few
macrofaunal taxa present. We hypothesized that a more com-
plete study of soil fauna could allow to better discriminate
these sites.
Study sites
The sites were 13 beech (Fagus sylvatica) forest stands of mature
trees, where soils and plant communities had beenpreviously studied
in relation to forest productivity (Manil et al. 1953, 1963; Dagnelie
1956a, 1956b, 1957). They are typical of the forest cover of the Ar-
denne Mountains. The climate shares Atlantic and mountain features,
being characterized by abrupt changes in temperature, with a mean
annual temperature of 7.2°C and a mean annual rainfall ranging from
1000 to 1400 mm according to geographical location. These old Her-
cynian Mountains have been strongly eroded, culminating at 694 m
altitude. Rocks, ranging from Cambrian to Devonian age, are poor in
bases (schists, graywackes, quartzites). Phytosociological and soil
types are given in Table 1, together with elevation and geographical
location.
Material and methods
Soil fauna
Macrofauna was sampled by forcing a 30 ×30 cm steel frame into the
litter sensu lato and the first 5 cm of underlying soil. Three samples
were taken in each site in June 1989, then three others in October 1989.
Samples were placed in plastic bags and then transported to the labo-
ratory. Animals were extracted within 15 days by the dry-funnel
method. For soil-dwelling earthworms an additional sampling around
the same plots was done by watering a 50 ×50 cm area three times at
10-mm intervals with diluted formaldehyde as a repellent (2, 3, then
4‰ v/v), then digging the soil underneath down to 30 cm.
Mesofauna was sampled by forcing a 5 cm diameter steel cylinder
into the top 15 cm of soil (litter included), at the same dates as for
macrofauna, but with only 2 ×2 replicates. Samples were then pro-
cessed as mentioned above.
Given the poor efficiency of the dry-funnel method for enchytraeid
worms, these animals, together with other visible soil animals, were
Received April 22, 1997. Accepted September 5, 1997.
J.-F. Ponge1and P. Arpin. Museum national d’histoire
naturelle, Laboratoire d’écologie générale, 4, avenue du
Petit-Chateau, 91800 Brunoy, France.
F. Sondag. ORSTOM, Centre d’Ile-de-France, Laboratoire des
formations superficielles, 32, avenue Henri-Varagnat, 93143
Bondy Cedex, France.
F. Delecour.2Faculté des sciences agronomiques de Gembloux,
Science du sol, avenue Maréchal-Juin 27, 5030 Gembloux,
Belgium.
1Author to whom all correspondence should be addressed.
e-mail: Jean-Francois.Ponge@wanadoo.fr
2Present address: Chaussée de Charleroi 97, 5030 Gembloux,
Belgium.
Can. J. For. Res. 27: 2053–2064 (1997)
2053
© 1997 NRC Canada

hand sorted directly in special soil cores (5 ×5×15 cm), which were
taken in June 1989 for micromorphological purposes (two replicates
in each site), according to the method described by Ponge (1991).
Hand sorting was performed by dividing the cores into smallvolumes
of litter and soil, which were observed in ethyl alcohol under a dis-
secting microscope. Plant fragments as well as soil aggregates were
thoroughly comminuted, and all mesofauna and macrofauna were
recovered, thus allowing comparisons of extraction methods.
Table 2 indicates the animal groups that were identified and
counted, together with the methods used for their recovery.
Litter accumulation
The surface weight of litter layers, estimated just after main leaf fall,
was used to compare the different sites. The O horizon, i.e., the pure
or near pure organic matter accumulated at the top of the soil profile
(Delecour 1980; Brêthes et al. 1995; Jabiol et al. 1995), can be di-
vided into several horizons called OL (entire leaves), OF (fragmented
leaves), and OH (holorganic faecal material). These horizons are
called L, F, and H, respectively, in the classification of Green et al.
(1993), which assigns the term O horizon to wetland soils only. The
more rapidly litter disappears from the ground, the less important are
OF and OH horizons compared with OL horizon, which at the end of
autumn is mainly made of freshly fallen litter.We calculatedthe litter
accumulation index (LAI) as the ratio WOF+OH/WOL, where WOF+OH
and WOL are the areal weights of OF +OH and OL horizons, respec-
tively. For that purpose, these horizons were sampled in the study sites
at the end of November 1989, by forcing six 15 cm diameter stainless
steel cylinders through the topsoil. Samples were transported to the
laboratory and then dried in air-forced chambers at constant tempera-
ture (25°C) during a fortnight, before being weighed to the nearest
10–2 g. After this step, beech leaves were sorted and weighed sepa-
rately, in the OL horizon only.
Stand productivity
Following previous work on the same sites (Dagnelie 1956a, 1956b,
1957), a linear relationship was demonstrated between the mean total
height of codominant trees and the mean annual increment of wood
available for timber production. For instance, total heights of 25, 30,
and 35 m were associated with increments of 3.6, 5.4, and
7.4 m3⋅ha–1⋅year–1, respectively. Thus we used total height of adult
codominant trees as a productivity index. This height was measured
on six codominant trees growing in the vicinity of the sampling plot.
In some cases (sites 1, 5, 100) fewer individuals (3, 3, 2, respectively)
were used, because of the smaller size of the study site or timber
harvesting during previous years. The total height of each selected
tree was measured with a Suunto Hypsometer® compass to the near-
est 0.25 m.
Litter chemical analyses
Beech leaf litter and miscellaneous litter were separately analysed in
the OL samples used for the determination of the litter accumulation
index. For that purpose samples from the same site were bulked into
a composite sample, which was ground then dried overnight at 103°C,
in order to determine its dry mass. The ash content was measured by
calcinating 1 g of powdered dry litter in a muffle furnace at 550°C for
5 h. Total nitrogen was quantified by Kjeldahl digestion into a
Kjeltec® autoanalyser on a separate 200-mg subsample.Total carbon
was quantified by the Anstett method, using concentrated sulphuric
acid and potassium bichromate as oxidants and Mohr salts for titra-
tion, on a 100-mg subsample. Other elements (Ca, Mg, K, P, Fe) were
determined by high-frequency plasma emission photometry on the
ashed subsample after dissolution in hydrochloric acid and elimina-
tion of silica by hydrofluoric acid.
Humus form
Humus form was identified in each sampling plot in June 1989 while
taking samples for micromorphological studies (two replicates in
each site). Nomenclature was derived from Brêthes et al. (1995). Ac-
cording to this classification, the O horizon (litter sensu lato) and the
A horizon (organo mineral horizon underlying the O horizon) may
vary somewhat independently, transition forms between mull and
moder groups being called hemimoder (belonging to the moder
group), amphimull, and dysmull (belonging to the mull group) ac-
cording to the absence or presence of a crumbly structure in the A
horizon, combined with absence or presence of an OH horizon.
Soil chemical analyses
These analyses were performed separately on six replicate samples
taken in each site after collection of the O horizon as mentioned
above. The underlying A horizon was collected down to 5 cm depth
under the bottom of the O horizon, then air dried until analysis. Sam-
ples were sieved (<2 mm) and then homogenized. Water pH and po-
tassium chloride pH were measured on a 5-g subsample diluted with
deionized water (soil:water 1:1 w/w). A 50-g subsample was crushed
with pestle and mortar, then sieved (<200 µm) for further analyses.
Cation exchange capacity was measured on a 10-g subsample by
percolating the soil with 0.5 M calcium chloride until saturation of
exchange sites and then displacing calcium with 1 M potassium ni-
trate. Determination of calcium and chloride content was performed
in the filtrate by flame nitrous oxyde – acetylene atomic absorption
Site Locality Phytosociological type
Elevation
(m) Soil type
1 Saint-Hubert (CEA) Luzulo – Fagetum festucetosum 370 Dystric Cambisol
3 Saint-Hubert (UA) Luzulo – Fagetum festucetosum 465 Dystric Cambisol
4 Saint-Hubert (UA) Luzulo – Fagetum typicum 500 Dystric Cambisol
5 Saint-Hubert (UA) Luzulo – Fagetum vaccinietosum 505 Dystric Cambisol
16 Rienne (WA) Luzulo – Fagetum vaccinietosum 445 Dystric Cambisol
17 Rienne (WA) Luzulo – Fagetum typicum 430 Dystric Cambisol
22 Haut-Fays (AA) Luzulo – Fagetum typicum 400 Gleyic Cambisol
24 Haut-Fays (AA) Luzulo – Fagetum festucetosum 390 Dystric Cambisol
26 Willerzie (WA) Luzulo – Fagetum vaccinietosum 430 Leptic Podzol
28 Houdremont (WA) Luzulo – Fagetum festucetosum 375 Dystric Cambisol
40 Willerzie (WA) Luzulo – Fagetum vaccinietosum 385 Ferric Podzol
100 Saint-Hubert (CEA) Melico – Fagetum festucetosum 350 Dystric Cambisol
307 Saint-Hubert (CEA) Luzulo – Fagetum vaccinietosum 380 Leptic Podzol
Note: AA, Atlantic Ardenne; CEA, Central-Eastern Ardenne; UA, Upper Ardenne; WA, Western Ardenne.
Nomenclature of soil types follows FAO–UNESCO classification (Driessen and Dudal 1991).
Table 1. Geographical, vegetation, and soil features of the 13 investigated sites.
Can. J. For. Res. Vol. 27, 1997
2054
© 1997 NRC Canada

photometry and complexometry with a Technicon® autoanalyser, re-
spectively. Exchangeable cations (Ca, Mg, K, Na) were determined
on a 10-g subsample after displacement of sorbed cations with ammo-
nium nitrate. Potassium and sodium were determined on the filtrate
by flame emission photometry, calcium and magnesium by flame
atomic absorption photometry. Total carbon and nitrogen were deter-
mined with a CHN Carlo Erba® analyser on a 5-mg subsample. Total
bases (Ca, Mg, K, Na), iron, and manganese were determined on a 1-g
subsample after boiling with concentrated hydrochloric acid. Potas-
sium and sodium were determined by flame air – acetylene emission
photometry; magnesium, iron, and manganese were determined by
flame air – acetylene atomic absorption photometry; and calcium was
determined by flame nitrous oxyde – acetylene atomic absorption
photometry. Total phosphorus was determined on a 1-g subsample
with a Technicon® autoanalyser after treatment with concentrated
hydrogen peroxyde followed by boiling with perchloric acid.
Data analysis
Effects of season or extraction methods on animal densities were
tested by means of two-way ANOVA using sites as blocks (Sokal and
Rohlf 1995; Rohlf and Sokal 1995). To ensure the additivity of variance,
Code Animal group Method of recovering
ANTS Insecta, Hymenoptera MES, MIC
CANT Insecta, Coleoptera, Cantharidae, larvae MAC, MES, MIC
CATE Insecta, Lepidoptera, larvae MAC
CECI Insecta, Diptera, Cecidomyidae, larvae MES, MIC
CENT Myriapoda, Chilopoda MAC, MES, MIC
CERA Insecta, Diptera, Ceratopogonidae, larvae MAC, MIC
CHEL Chelicerae, miscellaneous MAC
CHIR Insecta, Diptera, Chironomidae, larvae MAC, MES, MIC
CLIC Insecta, Coleoptera, Elateridae, larvae MAC, MES, MIC
CMIS Insecta, Coleoptera, miscellaneous, larvae MAC, MES, MIC
COAD Insecta, Coleoptera, adults MAC, MES, MIC
COCH Insecta, Homoptera MES, MIC
COLL Insecta, Collembola MES, MIC
COPE Crustacea, Copepoda MIC
CURC Insecta, Coleoptera, Curculionidae, larvae MAC
DERM Insecta, Dermaptera MAC
DIPL Insecta, Diplura MAC, MES, MIC
DMIS Insecta, Diptera, miscellaneous, larvae MAC, MES, MIC
DOEM Insecta, Diptera, Dolichopodidae +Empididae, larvae MAC, MES, MIC
ENCH Annelida, Oligochaeta, Enchytraeidae MES, MIC
FANN Insecta, Diptera, Muscidae, larvae MAC
ISOP Crustacea, Isopoda MAC, MES, MIC
LIMN Insecta, Coleoptera, Limnobiidae, larvae MAC
LMIS Insecta, miscellaneous, larvae MAC
LUMB Annelida, Oligochaeta, Lumbricidae MAC, special extraction
MILL Myriapoda, Diplopoda MAC, MIC
MITE Acari, excl. Oribatida MES, MIC
MOLL Mollusca, Gastropoda MAC, MIC
OPIL Chelicerae, Phalangida MAC
ORIB Acari, Oribatida, miscellaneous MES, MIC
PAUR Myriapoda, Pauropoda MES, MIC
PHTH Acari, Oribatida, Phthiracaridae +Euphthiracaridae MES, MIC
PROT Insecta, Protura MES, MIC
PSEU Chelicerae, Pseudoscorpionida MAC, MES, MIC
PSOC Insecta, Psocoptera MES, MIC
PSYC Insecta, Diptera, Psychodidae, larvae MAC
RHAG Insecta, Coleoptera, Rhagionidae MAC
SCAT Insecta, Diptera, Scatopsidae, larvae MAC
SCIA Insecta, Diptera, Sciaridae, larvae MAC, MIC
SPID Chelicerae, Araneida MAC, MES
SYMP Myriapoda, Symphyla MES, MIC
THRI Insecta, Thysanoptera MES, MIC
TIPU Insecta, Diptera, Tipulidae, larvae MAC, MIC
TRIC Insecta, Trichoptera, larvae MAC
Note: MAC, extraction of macrofauna; MES, extraction of mesofauna; MIC, micromorphological
dissection. Zoological nomenclature is according to Dindal (1990). The methods that have been selected for
estimating densities are in bold.
Table 2. Coding and methods of recovery used for the different animal groups investigated.
Ponge et al. 2055
© 1997 NRC Canada

data were previously transformed into log(x+1). All means given for
each site were calculated using log-transformed data.
Sites were ordinated according to their faunal composition using
correspondence analysis (Greenacre 1984). Active variates were
mean densities of the different animal groups in the 13 studied sites.
Data were reweighted to a unit standard deviation and focused around
a mean of 10 by using the transformation x→(x–m)/s+10, where
mis the mean and sis the standard deviation for each variate, respec-
tively. In this way the different animal groups have a similar mass and
similar total variance, thus allowing factorial coordinates to be di-
rectly interpreted in terms of their contribution to factorial axes.Each
variate was associated with a conjugate, varying in an opposite sense
(x′=20 – x). Thus each animal group will be represented by two
points, one indicating higher densities for this group, the other lower
densities. Passive variates, describing environmental conditions, were
added, in order to measure their degree of relationship with this ordi-
nation, which was based on faunal composition only. Passive data
were reweighted and focused in a similar way. Correlation coeffi-
cients between factorial axes and variates or between variates were
calculated on transformed data according to the product–momentfor-
mula of Pearson and were tested by the Student t-test method (Sokal
and Rohlf 1995).
Results
Choice of methods for recovering animals
Most macrofaunal groups were sampled on a much wider sur-
face than mesofauna, given the lower density and patchiness
of these animals in the soil (Macfadyen 1957). Enchytraeid
worms were recovered by dissecting litter and humus samples
at a high magnification. This was also the case for copepods,
phthiracarid mites, miscellaneous mites, pauropods, Symphyla,
Protura, cecidomyid, ceratopogonid, chironomid, sciarid, mis-
cellaneous fly larvae, cochineals, and booklice. In all these
cases the advantage of direct counting against active extraction
of animals was evident; thus we judged it preferable to choose
the first method, despite the poorer number of replicates (two,
versus four for active extraction of mesofauna). For miscella-
neous oribatid mites and springtails, which were collected in
high numbers both by dry funnels and by direct counting, an
ANOVA was performed on June samples (two replicates for
each method in each of the 13 sites). Extraction by the dry-funnel
method furnished more animals than direct counting for oriba-
tid mites (p< 0.0001), but differences between methods were
insignificant for springtails (p=0.17). The methods chosen for
the different animal groups are indicated in Table 2.
Seasonal influences
Densities of three macrofaunal groups were significantly af-
fected by season, with more animals in November than in June,
i.e., spiders, adult beetles, and pseudoscorpions, with
p=0.003, 0.03, and p< 0.0001, respectively (two-way ANOVA).
Only two mesofaunal groups were significantly affected, with
more animals in June than in November, i.e., springtails and
miscellaneous oribatid mites, with p=0.006 and 0.01, respec-
tively. Given that significant differences were few, we decided
to pool the data from the two sampling periods into a compos-
ite mean for each study site.
Ordination of sites according to faunal composition
Correspondence analysis of faunal data helped to ordinate sites
according to their faunal composition. The first axis extracted
25% of the total variance. Examination of the position of sites and
zoological groups along this axis (Fig. 1) and of faunal densi-
ties (Table 3) showed a progressive shift from macrofauna-
dominated to enchytraeid-dominated sites, with the exception
of some macrofaunal groups such as click-beetle larvae
(CLIC), Diplura (DIPL), and cochineals (COCH). On the posi-
tive side of axis 1 only enchytraeid (ENCH) and click-beetle
(CLIC) densities were significantly correlated with axis 1 co-
ordinates. On the negative side, limnobiid larvae (LIMN), sca-
topsid larvae (SCAT), dolichopodid-empidid larvae (DOEM),
milliped (MILL), Trichoptera larvae (TRIC), cantharid larvae
(CANT), woodlice (ISOP), earthworm (LUMB), pseudoscor-
pion (PSEU), rhagionid larvae (RHAG), chironomid larvae
(CHIR), mollusc (MOLL), and muscid larvae (FANN) densi-
ties, were all significantly correlated with axis 1 coordinates.
All these groups were significantly correlated between them,
indicating that the global trend depicted by axis 1 was a com-
munity gradient.
We may nevertheless question whether groups placed in an
intermediate position, i.e., not far from the origin, (i) do not vary
to a great extent between sites, (ii) are influenced by factors
other than this community gradient, or (iii) are more abundant
in sites placed in an intermediary position (such as sites 3, 17,
22, 24) than in sites placed far from the origin on the positive
Fig. 1. Ordination of 13 sites and 44 zoological groups, used as
main variates, according to their coordinates along axis 1 of
correspondence analysis. Coding of sites and zoological groups is
according to Tables 1 and 2, respectively. The position of the origin
is indicated by an arrow. Codes for zoological groups belonging to
macrofauna are in bold. Variates significantly correlated with axis 1
coordinates are indicated by rectangular bordering.
Can. J. For. Res. Vol. 27, 1997
2056
© 1997 NRC Canada

or on the negative side of axis 1. The case of groups such as
ants (ANTS), copepods (COPE), earwigs (DERM), miscella-
neous insect larvae (LMIS), psychodid larvae (PSYC), and
booklice (PSOC) cannot be accounted for, since they are
scarce and present in a low number of sites. On the contrary,
oribatid mites (ORIB) and sciarid larvae (SCIA), placed not
far from the origin, are abundant and present everywhere. The
first group proved to be significantly more abundant in some
sites than in others (F=3.56, df =12/39, p=0.0013), and the
second group did not significantly differ between sites (F=
1.17, df =12/13, p=0.39). Examination of the mean densities
of oribatid mites in the 13 sites (Table 3) showed that these
animals were very abundant in sites located on both sides of
axis 1. Thus their distribution did not follow the global trend
exhibited by the first axis of correspondence analysis (case ii).
Sciarid larvae were rather evenly distributed (case i). We did not
register the third postulated case, i.e., zoological groups charac-
teristic of sites placed in an intermediary position by the analysis.
Explanatory value of site features
Elevation, together with phytosociological type and humus form,
proved to discriminate the studied sites, ordinated according to
100 1 28 3 22 24 17 16 307 5 40 26 4
LIMN 17 21 6.3 0.5 2.5 0.5 0.8
SCAT 1.6 3.1 0.9 0.5 0.5
DOEM 1 200 160 780 160 26 160 100 100 34 28 3.7 5.2 5.2
MILL 54 24 16 2.2 2.9 2
TRIC 0.9 1.6
CANT 0.7 3.3 5.2 0.5 0.9 0.5 0.5 0.5
ISOP 42 0.5 2 0.5 0.5
LUMB 61 2 0.5 1.3 1.6
PSEU 17 20 25 10 8.3 4.8 6.7 12 9.5 8.8 3.8 0.7 8.2
RHAG 6.6 24 44 12 16 12 6 8.8 1.7 7.3 5.5 0.7 1.3
CHIR 800 570 980 2 500 690 19 44 65 19 1 100 34
MOLL 6 0.5 0.5
FANN 1.3 1.3 14 0.5 0.5 0.5 1.6 0.7
CENT 100 99 160 63 21 7.8 69 55 20 12 98 54 16
OPIL 0.5 0.5 0.5
CHEL 5.6 0.7 0.5
CERA 400
TIPU 2.2 0.9 0.5 2.5 2.5 0.5 0.5 0.7 1.6 1.6 0.5
PROT 1 500 19 19
PAUR 19 5 400 27 27 44 2 400
CECI 800 400 48 570 27 1 300 1 100 27 1 700 19 19
CMIS 34 30 47 37 23 43 32 14 29 39 39 34 25
DERM 1.3 0.7
SPID 11 8.6 8.7 53 9.4 2.9 4.8 8 19 2.3 5.5 13 10
LMIS 0.5 0.5
SCIA 1 500 1 700 5 800 1 700 65 1 800 3 700 3 900 2 000 19 17 000 980 4 200
ORIB 55 000 150 000 110 000 80 000 43 000 44 000 62 000 94 000 210 000 87 000 72 000 100 000 57 000
COAD 15 8.9 11 27 5.5 22 4.4 13 9.3 18 17 10 9.6
COPE 19 19 19
ANTS 28
DMIS 19 19 27 19 27 19 39
THRI 21 4.6 3.7 21 26 3.7 3.7 21
PSYC 1.1
PSOC 27
CATE 0.5 0.8 0.5 0.5
MITE 13 000 18 000 33 000 32 000 19 000 11 000 24 000 38 000 32 000 14 000 29 000 29 000 28 000
CURC 0.5
COLL 42 000 70 000 59 000 88 000 38 000 51 000 57 000 71 000 100 000 68 000 45 000 81 000 120 000
COCH 27 19 34
PHTH 2 900 8 200 20 000 7 300 29 000 5 100 43 000 24 000 33 000 15 000 27 000 9 600 16 000
SYMP 400 19 19 980 39 27 1 100 570 19
DIPL 15 0.7 97 2.9 23 38 87 21 25 36 98 140 50
CLIC 3.1 6.7 41 81 38 6.4 12 20 18 130 120 84 39
ENCH 12 000 89 000 100 000 150 000 180 000 170 000 46 000 130 000 79 000 400 000 410 000 380 000 800 000
Table 3. Mean densities per square metre of zoological groups in the 13 investigated sites, ordinated according to axis 1 of correspondence
analysis.
Ponge et al. 2057
© 1997 NRC Canada

axis 1 of correspondence analysis (Fig. 2, Table 4). Elevation
was significantly and positively correlated with axis 1 (r=
0.65, p< 0.05), thus increasing from site 100 to site 4. Along
this community gradient humus form varied from dysmull to
dysmoder, i.e., from rapid to slow disappearance of litter
(Brêthes et al. 1995). Oligomull was indistinguishable from
dysmull, and amphimull, hemimoder, and eumoder were
placed in an intermediary position, being indistinguishable
from each other. The phytosociological type varied from
Melico – Fagetum festucetosum, with a rich ground flora and
highly productive, which is characteristic of lowland sites
(Thill et al. 1988), to Luzulo – Fagetum vaccinietosum, much
poorer in ground flora and weakly productive, which is mostly
established on tablelands and sunny slopes. Soil types did not
express a good relationship with axis 1, contrary to humus
forms and phytosociological types.
Total height of codominant trees was significantly corre-
lated with axis 1 (r=–0.56, p< 0.05), together with pH in H2O
(r=–0.75, p< 0.01), pH in KCl (r=–0.71, p< 0.01), and C/N
ratio of the A horizon (r=0.88, p< 0.01). Thus the community
gradient from site 100 to site 4 was characterized by a bulk
decrease in the height of trees and soil pH, and an increase in
C/N ratio (Fig. 3, Table 4). No significant correlation was
found for axis 1 with the litter accumulation index (LAI) and
surface weight of OF +OH horizons.
Among total soil elements, only manganese was signifi-
cantly correlated with axis 1 (r=–0.87, p< 0.01), its content
in the top 5 cm of the A horizon, decreasing from site 100 to
site 4 (Fig. 4, Table 4). No significant correlation was found
with cation exchange capacity nor exchangeable bases.
The richness of litter in mineral matter (ash content) was
significantly correlated with axis 1, for both total litter and
beech leaf litter (r=–0.86, p< 0.01 and r=–0.77, p< 0.01,
respectively), decreasing from site 100 to site 4 (Fig. 5, Table 4).
At the elemental scale the same trend was depicted by iron,
calcium, and magnesium, for both total litter and beech leaf litter.
Discussion
The fauna of investigated sites was clearly varying in the same
sense as soil fertility, this feature being expressed not only by
pH and C/N ratio of the A horizon (Brady 1984), but also
Fig. 2. Ordination of 13 sites and some additional variates
(elevation, humus forms, soil types, phytosociological types),
according to their coordinates along axis 1 of correspondence
analysis. Variates significantly correlated with axis 1 coordinates
are indicated by rectangular bordering.
Fig. 3. Ordination of 13 sites and some additional variates (pH and
C/N ratio in the A horizon, litter accumulation, height of trees)
according to their coordinates along axis 1 of correspondence
analysis. Variates significantly correlated with axis 1 coordinates
are indicated by rectangular bordering.
Can. J. For. Res. Vol. 27, 1997
2058
© 1997 NRC Canada

by mineral richness of leaf litter (Mangenot and Toutain 1980)
and tree growth (Dagnelie 1957). We may nevertheless ask to
what extent the faunal composition here was determined by
site conditions. The possibility of feedback loops between
fauna and site conditions should not be overlooked, too, except
for some features such as elevation, which are not placed under
biological control.
In the Belgian Ardennes, altitude has been locally considered
100 1 28 3 22 24 17 16 307 5 40 26 4
Elevation (m) 350 370 375 465 400 390 430 445 380 505 385 430 500
Phytosociological type
Melico – Fagetum festucetosum +
Luzulo – Fagetum festucetosum +++ +
Luzulo – Fagetum typicum ++ +
Luzulo – Fagetum vaccinietosum +++++
Soil type
Dystric Cambisol ++++ +++ + +
Gleyic Cambisol +
Leptic Podzol ++
Ferric Podzol +
Humus form
Oligomull +
Dysmull ++ +
Amphimull ++
Hemimoder +
Eumoder ++++++
Dysmoder ++ ++++
Height of trees (m) 37 42 38 36 36 39 37 37 26 37 24 31 35
Litter accumulation index
(LAI) 1.1 3.1 14.1 9.8 10.4 7.3 6.0 9.1 10.3 8.7 9.0 9.4 7.2
OF+OH (kg⋅m–2) 0.8 2.3 9.0 8.7 6.9 4.3 4.8 5.7 8.4 5.9 7.7 7.7 6.6
Soil analyses (A horizon)
pH in water 4.3 3.8 3.6 3.6 3.7 3.6 3.4 3.3 3.6 3.5 3.1 3.4 3.6
pH in KCl 3.6 3.1 3.0 2.8 3.1 3.0 2.7 2.6 2.8 2.9 2.0 2.5 2.9
C/N 14.5 14.2 14.9 16.6 16.5 16.6 18.3 18.9 19.5 17.4 19.8 18.9 17.8
Total Ca (%) 0.19 0.11 0.02 0.03 0.05 0.04 0.10 0.09 0.02 0.11 0.06 0.04 0.02
Total Mg (%) 0.12 0.10 1.17 0.11 0.24 0.18 0.11 0.11 0.11 0.13 0.04 0.11 0.19
Total K (%) 0.25 0.21 0.21 0.25 0.24 0.25 0.23 0.24 0.30 0.26 0.13 0.23 0.28
Total Na (%) 0.08 0.11 0.06 0.11 0.07 0.09 0.11 0.19 0.07 0.12 0.05 0.08 0.10
Total Fe (%) 8.0 8.6 10.2 9.1 4.9 5.4 4.6 5.5 7.9 9.4 1.4 4.8 6.9
Total Mn (%) 0.24 0.16 0.15 0.11 0.04 0.08 0.02 0.01 0.11 0.07 0.00 0.01 0.04
CEC (mequiv.⋅100 g–1) 13.1 7.5 11.1 10.6 10.2 10.6 15.2 16.1 8.6 9.1 21.4 15.0 13.0
Exchangeable Ca (mequiv.⋅100 g–1) 3.48 0.25 0.29 0.72 0.22 0.14 0.98 0.26 0.11 0.09 0.86 0.36 0.30
Exchangeable Mg (mequiv.⋅100 g–1) 0.61 0.17 0.33 0.21 0.16 0.15 0.39 0.29 0.25 0.13 0.69 0.30 0.29
Exchangeable K (mequiv.⋅100 g–1) 0.44 0.25 0.23 0.34 0.29 0.24 0.46 0.31 0.31 0.18 0.57 0.35 0.32
Exchangeable Na (mequiv.⋅100 g–1) 0.13 0.07 0.08 0.06 0.05 0.07 0.10 0.12 0.11 0.08 0.19 0.14 0.10
Litter analyses
Ashes in total litter (%) 6.7 5.4 3.9 3.7 4.2 4.0 3.6 3.1 4.2 2.9 3.7 2.7 3.2
N in total litter (%) 1.4 1.6 1.6 1.7 1.5 1.5 2.0 1.4 1.1 1.4 1.7 1.5 2.2
C/N in total litter 31.9 24.3 30.5 31.0 32.4 36.7 27.4 32.4 44.7 36.4 29.5 36.0 24.0
Ca in total litter (%) 1.22 0.60 0.52 0.44 0.42 0.45 0.44 0.37 0.50 0.39 0.55 0.37 0.42
Mg in total litter (%) 0.14 0.06 0.08 0.04 0.05 0.05 0.06 0.05 0.07 0.05 0.08 0.05 0.06
K in total litter (%) 0.32 0.31 0.21 0.32 0.27 0.30 0.52 0.26 0.34 0.33 0.42 0.31 0.31
Fe in total litter (mg⋅kg–1) 930 1100 540 390 610 520 540 440 560 310 330 220 350
Ashes in beech leaf litter (%) 8.2 5.0 4.3 4.5 4.7 4.7 4.0 3.3 4.4 3.2 4.1 3.3 3.8
N in beech leaf litter (%) 1.4 1.6 1.6 1.7 1.5 1.5 2.0 1.4 1.1 1.4 1.8 1.6 1.5
C/N in beech leaf litter (%) 29.8 27.6 29.8 26.2 30.1 31.5 23.3 31.8 38.9 34.1 25.5 30.5 29.8
Ca in beech leaf litter (%) 1.84 0.72 0.62 0.56 0.58 0.60 0.50 0.41 0.68 0.51 0.63 0.46 0.54
Mg in beech leaf litter (%) 0.18 0.05 0.08 0.04 0.04 0.04 0.05 0.04 0.07 0.04 0.07 0.05 0.05
K in beech leaf litter (%) 0.22 0.2 0.14 0.21 0.14 0.18 0.36 0.15 0.22 0.19 0.39 0.25 0.18
Fe in total litter (mg⋅kg–1) 730 640 540 470 430 530 530 420 410 330 310 250 360
Table 4. Vegetation and soil features of the 13 investigated sites, ordinated according to axis 1 of correspondence analysis.
Ponge et al. 2059
© 1997 NRC Canada

as the most prominent regional factor influencing stand pro-
ductivity, humus, and phytosociological type (Dagnelie 1957;
Manil et al. 1963; Delecour and Prince-Agbodjan 1975; Thill
et al. 1988). Higher altitude means colder climate and higher
precipitation, in a geographic zone (the Ardenne mountains),
where the regional climate is harsher and more rainy than in
any other part of Belgium (average annual temperature 7°C,
average annual rainfall 1100 mm). This may have conse-
quences on the level of biological activity, but also on the
leaching of mineral elements during periods of low biological
activity (winter), upland sites being thus impoverished com-
pared with lowland sites. In addition, erosion progressively
enriched lowland sites in nutrients at the expense of upland
sites (Duchaufour 1995). Combined with climate effects of
altitude (Manil et al. 1963), higher elevation (upland sites) also
means harder parent rocks than along slopes (Thill et al. 1988),
and even more than along rivers (lowland sites, the more typi-
cal being site 100, located along the river Masblette). This
geomorphological effect of altitude may affect the cycling of
nutrients through differences in mineral weathering (Gaiffe
and Bruckert 1990). Because of synergistic effects of climate,
erosion, and rock hardness upland sites will be thus charac-
terized by poorer availability of mineral elements for organ-
isms, when compared with lowland sites.
In the litter compartment of the beech ecosystem, the avail-
ability of elements to litter-consuming animals is related
to mineral richness of beech and total litter, sites with a mull
fauna (negative side of axis 1) having richer beech and total
litter than sites with a moder fauna (positive side of axis 1). It
should be highlighted that this effect of litter richness concerns
more metals (iron) and alkaline earths (calcium, magnesium)
than main nutrients such as nitrogen, potassium, and phospho-
rus, or the C/N ratio, contrary to literature data on plant litter
decomposition (Melillo et al. 1982) and palatibility of leaf lit-
ter to saprophagous animals (Hendriksen 1990). The high cal-
cium requirements of most earthworm (Piearce 1972),
milliped (Reichle et al. 1969; Carter and Cragg 1976), and
woodlice (Krivolutzky and Pokarzhevsky 1977) species, all
typical of the negative side of axis 1 (mull side), may never-
theless explain the absence of these groups in sites with a
poorer Ca content of litter (moder side). But here possible
feedback loop effects, which reinforce this selective process,
must be considered, (i) through the cycling of mineral ele-
ments by fauna and (ii) through the phenolic content of beech
foliage. Woodlice, millipeds, and earthworms have been con-
sistently demonstrated to increase the leaching of nutrients
from decaying leaf litter (Anderson et al. 1983; Morgan et al.
1989), thus increasing their availability to plants (Haimi and
Einbork 1992). Sulkava et al. (1996) demonstrated that at low
Fig. 4. Ordination of 13 sites and some additional variates
(exchangeable and total bases in the A horizon) according to their
coordinates along axis 1 of correspondence analysis. Variates
significantly correlated with axis 1 coordinates are indicated by
rectangular bordering. A plus or a minus sign means higher or
lower values, respectively.
Fig. 5. Ordination of 13 sites and some additional variates (mineral
content of litter) according to their coordinates along axis 1 of
correspondence analysis. Variates significantly correlated with
axis 1 coordinates are indicated by rectangular bordering. A plus or
a minus sign means higher or lower values, respectively.
Can. J. For. Res. Vol. 27, 1997
2060
© 1997 NRC Canada

to medium moisture the structure of soil animal communities
determined the extent of N mineralization. Thus the availabil-
ity of mineral elements for vegetation may be increased or
decreased according to composition of the soil fauna (Scheu
and Parkinson 1994). This in turn may affect the mineral com-
position of the beech foliage (Toutain and Duchaufour 1970).
The phenolic content of tree foliage has been demonstrated to
influence the palatibility of leaves to earthworms (Satchell and
Lowe 1967), a lower content in phenolics being associated
with higher palatability. Thus the phenolic content of litter
may directly affect some animal groups through their food
preferences. It also determines soil-forming and microbial
processes, a higher phenolic content of tree foliage and litter
increasing the leaching of bases during periods of low biologi-
cal activity and making proteins harder to decay through com-
plexing processes (Davies 1971). Conversely, the production
of phenolics and other secondary plant metabolites increases
in nutrient-poor conditions (Kuiters 1990), thus self-reinforcing
the process.
We can now examine the influence of soil chemistry and
humus form on soil animals, and conversely, their influence
on these conditions. Observations on the distribution of soil
animals in varying site conditions proved that beside consid-
erable variation from species to species, some zoological
groups in bulk are seemingly correlated with soil and humus
properties. Less acid soils, with mull humus forms, were found
to be characterized by a richer and more abundant sapro-
phagous macrofauna, especially earthworms, molluscs, wood-
lice, and millipeds (Bornebusch 1930; Van der Drift 1962;
Abrahamsen 1972b; Petersen and Luxton 1982; David 1987;
Herlitzius 1987; Staaf 1987; Schaefer and Schauermann 1990;
Schaefer 1991; Ponge and Delhaye 1995), like in the present
study. The above-mentioned association of chironomid fly lar-
vae with mull humus (negative side of axis 1) has already been
registered by Healey and Russel-Smith (1971). The associa-
tion of Nematocera fly larvae families (Rhagionidae,
Dolichopodidae, Empididae, Chironomidae, as representative
in our samples) with less acid soils has been already estab-
lished by Herlitzius (1987). In the case of enchytraeids, litera-
ture data indicate that species richness decreases when acidity
increases and unincorporated organic matter accumulates
(moder or mor humus), the opposite trend being observed with
total abundance, because of the dominance of Cognettia
sphagnetorum in raw humus (Abrahamsen 1972a; Healy
1980; Petersen and Luxton 1982), thus confirming our results
on this group as a whole. A similar phenomenon has been
observed by Bornebusch (1930) on click-beetle (Elateridae)
larvae in beech forests of Denmark, the density of Athous sub-
fuscus increasing dramatically in raw humus (in fact dys-
moder), which is confirmed by our observations on beech
forests in Belgium (David et al. 1993).
The direct action of soil chemistry on soil animals is diffi-
cult to evidence, because of multiple interactions with trophic
and habitat features, although it has been suspected following
community studies (Ponge 1993; Healy 1980), and studies on
the sensitivity of animals to acidity and osmolarity of soil so-
lutions (Laverack 1961; Jaeger and Eisenbeis 1984; Heungens
and Van Daele 1984). Experimental liming has been found to
be detrimental to enchytraeid species living in acid conditions
(Abrahamsen 1983; Huhta et al. 1986), the contrary being true
for earthworms (Huhta 1979; Toutain et al. 1987; Robinson
et al. 1992). These results should nevertheless be accepted
with caution, because in the short term, abrupt changes in soil
conditions following lime (or acid) application act only on
existing species. Results such as those of Robinson et al.
(1992), Muys and Lust (1992), and Rundgren (1994), who in
some sites did not observe any increase in earthworm densities
following liming, could be explained by the absence of acid-
intolerant species in the vicinity of experimental sites. This
introduces the problem of the time lapse needed for slow eco-
logical processes such as the adaptation of communities to
changing environmental conditions (Burges 1960). Results
from synchronic studies on humus dynamics (Bernier and
Ponge 1994) indicated that the course of shifts from moder to
mull humus could be conditioned by the activity of some bur-
rowing and acid-tolerant earthworm species such as Lum-
bricus terrestris. Other mull inhabitants may colonize the soil
profile only several decades after it has begun to be trans-
formed by this burrowing species. Thus the need for conditions
prevailing in mull humus forms, expressed by a lot of sapro-
phagous and even predaceous groups (pseudoscorpions,
dolichopodid-empidid, and rhagionid larvae), is probably the
result of multiple interactions involving feeding, behavioural,
and physico-chemical requirements of soil animals.
The action of soil fauna on soil chemical properties is better
known, mainly through their building of humus forms (Ku-
biëna 1955; Bal 1970; Hole 1981) and their above-mentioned
action on nutrient cycling. It has been experimentally verified
that the introduction of lacking animal groups, without any
further change in environmental conditions, may definitely
change site quality (Bal 1982; Scheu and Parkinson 1994).
These experiments concerned only the introduction of earth-
worm species, followed by the appearance of mull humus
forms as the result of their burrowing activity. Here we may
ask whether the appearance of dysmoder humus form (moder
humus with a thick OH horizon) can be determined not only
by the absence of zoological groups comprising litter-consuming
and burrowing species, but also by high densities of animals
such as enchytraeids, which we have found in huge amounts
in sites placed on the positive side of axis 1 (Table 3).
Enchytraeids have been suspected as having a detrimental in-
fluence not only on decomposition of organic matter (Wolters
1988) but also on earthworm populations (Haukka 1987) when
they reach high densities. Conversely, other authors found
them contributing significantly to mineralization processes
(Sulkava et al. 1996), thus giving a contrasted landscape con-
cerning the role of these animals in litter decomposition and
soil-forming processes.
Beside acidity (pH in H2O and KCl) and C/N ratio, manga-
nese was unexpectedly the only soil nutrient for which the
content was significantly correlated with axis 1. Free and ex-
changeable acidity and C/N ratio can be considered as in-
volved in feedback loops in the course of humification
processes (Ulrich 1986); thus they are causes as well as con-
sequences of the building of humus forms. The manganese
content of the topsoil, which is also involved in many biologi-
cal processes, has been found to be associated with humus
type, and like iron, it is much higher in mull than in moder
humus (Duchaufour and Rousseau 1959; Toutain and Védy
1975). Together with the C/N ratio, manganese is highly cor-
related with vitality of forest trees (Van Straalen et al. 1988).
Manganese, as well as iron, oxidizes phenolic acids, thus
Ponge et al. 2061
© 1997 NRC Canada

alleviating allelopathic and complexing processes due to
small-molecule aromatic compounds (Lehmann et al. 1987).
If we try to synthesize all these relationships into a common
scheme, the following hypothetical sequence can be consid-
ered as most realistic, at least in the present stage of our knowl-
edge. Altitude, given the specificity of the studied zone (the
Ardenne Mountains), can be considered as determining a lot
of site features that may drive the soil system towards one or
the other of two poles: a mull pole, better expressed in lowland
sites, with more animal groups, especially saprophagous
macrofauna, and better growth of trees, and a dysmoder pole,
better expressed in upland sites, with fewer animal groups,
mostly enchytraeids, and poorer growth of trees. Mechanisms
of the action of site conditions upon soil fauna (and the reverse)
may involve in first the content of leaf litter in metals and
alkaline earths, which proved better correlated with faunal
abundance and diversity than richness of the soil in these ele-
ments. If this hypothesis is true, then mull and dysmoder, sta-
bilized by numerous feedback loops involving vegetation,
decomposers, and humus profiles (Perry et al. 1989), should
act as steady-state positions for ecological conditions prevail-
ing in beech ecosystems of the Ardenne Mountains. In this
case the number of intermediate conditions should be less than
expected if the sites had been randomly scaled between these
two poles. This may be observed along axis 1, where sites 1,
100, and 28 (mull pole) are clearly isolated from the rest of the
sample. Unfortunately, the total number of sites of the mull
type was not high enough for properly testing the significance
of this pattern over the whole range of investigated sites.
References
Abrahamsen, G. 1972a. Ecological study of Enchytraeidae (Oligo-
chaeta) in Norwegian coniferous forest soils. Pedobiologia,
12: 26–82.
Abrahamsen, G. 1972b. Ecological study of Lumbricidae (Oligo-
chaeta) in Norwegian coniferous forest soils. Pedobiologia,
12: 267–281.
Abrahamsen, G. 1983. Effects of lime and artificial acid rain on the
enchytraeid (Oligochaeta) fauna in coniferous forest. Holarct.
Ecol. 6: 247–254.
Anderson, J.M., Ineson, P., and Huish, S.A. 1983. Nitrogen andcation
mobilization by soil fauna feeding on leaf litter and soil organic matter
from deciduous woodlands. Soil Biol. Biochem. 15: 463–467.
Bal, L. 1970. Morphological investigation in two moder humus pro-
files and the role of thesoil faunain their genesis. Geoderma, 4: 5–36.
Bal, L. 1982. Zoological ripening of soils. PUDOC, Wageningen,
Netherlands.
Bernier, N., and Ponge, J.F. 1994. Humus form dynamics during the
sylvogenetic cycle in a mountain spruce forest. Soil Biol. Bio-
chem. 26: 183–220.
Bornebusch, C.H. 1930. The fauna of forest soil. Forstl. Forsogsvaes.
Dan. 11.
Brady, N.C. 1984. The nature and properties of soil. 9th ed. Macmil-
lan, New York.
Brêthes, A., Brun, J.J., Jabiol, B., Ponge, J.F., and Toutain, F. 1995.
Classification of forest humus forms: a French proposal. Ann. Sci.
For. 52: 535–546.
Burges, A. 1960. Time and size as factors in ecology. J. Ecol.
48: 273–285.
Carter, A., and Cragg, J.B. 1976. Concentrations and standing crops
of calcium, magnesium, potassium and sodium in soil and litter
arthropods and their food in an aspen woodland ecosystem in the
Rocky Mountains (Canada). Pedobiologia, 16: 379–388.
Dagnelie, P. 1956a. Recherches sur la productivité des hêtraies d’Ar-
denne en relation avec les types phytosociologiques et les facteurs
écologiques. I. Recherche d’un critère de station utilisable dans les
hêtraies d’Ardenne. Bull. Inst. Agron. Stat. Rech. Gembloux,
24: 249–284.
Dagnelie, P. 1956b. Recherches sur la productivité des hêtraies d’Ar-
denne en relation avec les types phytosociologiques et les facteurs
écologiques. II. Utilisation d’un critère de station dans les hêtraies
d’Ardenne. Bull. Inst. Agron. Stat. Rech. Gembloux,
24: 369–410.
Dagnelie, P. 1957. Recherches sur la productivité des hêtraies d’Ar-
denne en relation avec les types phytosociologiques et les facteurs
écologiques. III. Interprétation des résultats. Bull. Inst. Agron.
Stat. Rech. Gembloux, 25: 44–94.
David, J.F. 1987. Relations entre les peuplements de Diplopodes et
les types d’humus en forêt d’Orléans. Rev. Ecol. Biol. Sol,
24: 515–525.
David, J.F., Ponge, J.F., and Delecour, F. 1993. The saprophagous
macrofauna of different types of humus in beech forests of the
Ardenne (Belgium). Pedobiologia, 37: 49–56.
Davies, R.I. 1971. Relation of polyphenols to decomposition of or-
ganic matter and to pedogenetic processes. Soil Sci. 111: 80–85.
Delecour, F. 1978. Facteurs édaphiques et productivité forestière.
Pédologie, 28: 271–284.
Delecour, F. 1980. Essai de classification pratique des humus.
Pédologie, 30: 225–241.
Delecour, F., and Prince-Agbodjan, W. 1975. Étude de la matière
organique dans une bio-toposéquence de sols forestiers ardennais.
I. Distribution du carbone et de l’azote dans les fractions hu-
miques. Bull. Rech. Agron. Gembloux, 10: 135–150.
Delecour, F., and Weissen, F. 1981. Forest-litter decomposition rate
as a site factor. Mitt. Forstl. Bundes-Versuchsanst. Wien,
140: 117–123.
Dindal, D.L. 1990. Soil biology guide. John Wiley & Sons, New York.
Driessen, P.M., and Dudal, R. 1991. The major soils of the world.
Agricultural University of Wageningen, Wageningen, Nether-
lands, and Catholic University of Leuven, Belgium.
Duchaufour, P. 1995. Pédologie. Sol, végétation, environnement.
Masson, Paris, France.
Duchaufour, P., and Rousseau, L.Z. 1959. Les phénomènes d’intoxi-
cation des plantules de résineux par le manganèse dans les humus
forestiers. Rev. For. Fr. 11: 835–847.
Gaiffe, M., and Bruckert, S. 1990. Origine paléoécologique de l’apti-
tude des calcaires jurassiques à la fracturation. Conséquences tec-
toniques, pédogénétiques et écologiques. Bull. Soc. Neuchâtel.
Sci. Nat. 113: 191–206 +2 inlet plates.
Green, R.N., Trowbridge, R.L., and Klinka, K. 1993. Towards a taxo-
nomic classification of humus forms. For. Sci. Monogr. 29.
Greenacre, M.J. 1984. Theory and applications of correspondence
analysis. Academic Press, London, United Kingdom.
Haimi, J., and Einbork, M. 1992. Effects of endogeic earthworms on
soil processes and plant growth in coniferous forest soil. Biol.
Fertil. Soils, 13: 6–10.
Haukka, J.K. 1987. Growth and survival of Eisenia fetida (Sav.) (Oli-
gochaeta: Lumbricidae) in relation to temperature, moisture and
presence of Enchytraeus albidus (Henle) (Enchytraeidae). Biol.
Fertil. Soils, 3: 99–102.
Healey, I.N., and Russel-Smith, A. 1971. Abundance and feeding
preferences of fly larvae in two woodland soils. In Organismes du
sol et production primaire, Proceedings of the 4th Colloquium
Pedobiologiae, 14 Sept. – 19 September 1970, Dijon, France. In-
stitut national de la recherche agronomique, Paris, France.
pp. 177–191.
Healy, B. 1980. Distribution of terrestrial Enchytraeidae in Ireland.
Pedobiologia, 20: 159–175.
Hendriksen, N.B. 1990. Leaf litter selection by detritivore and geo-
phagous earthworms. Biol. Fertil. Soils, 10: 17–21.
Can. J. For. Res. Vol. 27, 1997
2062
© 1997 NRC Canada

Herlitzius, H. 1987. Decomposition in five woodland soils : relation-
ships with some invertebrate populations and with weather. Biol.
Fertil. Soils, 3: 85–89.
Heungens, A., and Van Daele, E. 1984. The influence of some acids,
bases and salts on the mite and Collembola population of a pine
litter substrate. Pedobiologia, 27: 299–311.
Hole, F.D. 1981. Effects of animals on soil. Geoderma, 25: 75–112.
Huhta, V. 1979. Effects of liming and deciduous litter on earthworm
(Lumbricidae) populations of a spruce forest, with an inoculation
experiment on Allolobophora caliginosa. Pedobiologia, 19: 340–345.
Huhta, V., Hyvönen, R., Koskenniemi, A., Vilkamaa, P., Kaasalainen, P.,
and Sulander, M. 1986. Response of soil fauna to fertilization and
manipulation of pH in coniferous forests. Acta For. Fenn. 195.
Jabiol, B., Brêthes, A., Ponge, J.F., Toutain, F., and Brun, J.J. 1995.
L’humus sous toutes ses formes. École nationale du génie rural des
eaux et des forêts, Nancy, France.
Jaeger, G., and Eisenbeis, G. 1984. pH-dependent absorption of solu-
tions by the ventral tube of Tomocerus flavescens (Tullberg, 1871)
(Insecta, Collembola). Rev. Ecol. Biol. Sol, 21: 519–531.
Jenny, H., Gessel, S.P., and Bingham, F.T. 1949. Comparative study
of decomposition rates of organic matter in temperate and tropical
regions. Soil Sci. 68: 419–432.
Krivolutzky, D.A., and Pokarzhevsky, A.D. 1977. The role of soil
animals in nutrient cycling in forest and steppe. In Soil organisms
as components of ecosystems. Edited by U. Lohm and T. Persson.
Ecol. Bull. 25: 253–260.
Kubiëna, W.L. 1955. Animal activity in soils as a decisive factor in
establishment of humus forms. In Soil zoology. Edited by
D.K. McE. Kevan. Butterworths, London, United Kingdom.
pp. 73–82.
Kuiters, A.T. 1990. Role of phenolic substances from decomposing
forest litter in plant–soil interactions. Acta Bot. Neerl.
39: 329–348.
Laverack, M.S. 1961. Tactile and chemical perceptions in earth-
worms. II. Responses to acid pH solutions. Comp. Biochem.
Physiol. 2: 22–34.
Lehmann, R.G., Cheng, H.H., and Harsh, J.B. 1987. Oxidation of
phenolic acids by soil iron and manganese oxides. Soil Sci. Soc.
Am. J. 51: 352–356.
Macfadyen, A. 1957. Animal ecology. Aims and methods. Pitman,
London, United Kingdom.
Mangenot, F., and Toutain, F. 1980. Les litières. In Actualités
d’écologie forestière. Sol, flore, faune. Edited by P. Pesson.
Gauthier-Villars, Paris, France. pp. 3–59.
Manil, G., Collin, E., Evrard, R., and Gruber, R. 1953. Les sols for-
estiers de l’Ardenne. Le plateau de Saint-Hubert-Nassogne. I.
Étude pédologique. Bull. Inst. Agron. Stat. Rech. Gembloux, 21:
43–68 +addendum.
Manil, G., Delecour, F., Forget, G., and El Attar, A. 1963. L’humus,
facteur de station dans les hêtraies acidophiles de Belgique. Bull.
Inst. Agron. Stat. Rech. Gembloux, 31: 1–114.
Melillo, J.M., Aber, J.D., and Muratore, J.F. 1982. Nitrogen and lig-
nin control of hardwood leaf litter decomposition dynamics. Ecol-
ogy, 63: 621–626.
Morgan, C.R., Schindler, S.C., and Mitchell, M.J. 1989. The effects
of feeding by Oniscus asellus (Isopoda) on nutrient cycling in an
incubated hardwood forest soil. Biol. Fertil. Soils, 7: 239–246.
Muys, B., and Lust, N. 1992. Inventory of the earthworm communi-
ties and the state of litter decomposition in the forests of Flanders,
Belgium, and its implications for forest management. Soil Biol.
Biochem. 24: 1677–1681.
Perry, D.A., Amaranthus, M.P., Borchers, J.G., Borchers, S.L., and
Brainerd, R.E. 1989. Bootstrapping in ecosystems. BioScience,
39: 230–237.
Petersen, H., and Luxton, M. 1982. A comparative analysis of soil
fauna populations and their role in decomposition processes.
Oikos, 39: 287–388.
Piearce, T.G. 1972. The calcium relations of selected Lumbricidae. J.
Anim. Ecol. 41: 167–188.
Ponge, J.F. 1991. Food resources and diets of soil animals in a small
area of Scots pine litter. Geoderma, 49: 33–62.
Ponge, J.F. 1993. Biocenoses of Collembola in Atlantic temperate
grass–woodland ecosystems. Pedobiologia, 37: 223–244.
Ponge, J.F., and Delhaye, L. 1995. The heterogeneity of humus pro-
files and earthworm communities in a virgin beech forest. Biol.
Fertil. Soils, 20: 24–32.
Reichle, D.E., Shanks, M.H., and Crossley, D.A., Jr. 1969. Calcium,
potassium, and sodium content of forest floor arthropods. Ann.
Entomol. Soc. Am. 62: 57–62.
Robinson, C.H., Piearce, T.G., Ineson, P., Dickson, D.A., and
Nys, C. 1992. Earthworm communities of limed coniferous soils:
field observations and implications for forest management. For.
Ecol. Manag. 55: 117–134.
Rodenkirchen, H. 1985. Site diagnosis by phytosociological indica-
tion in soil amelioration experiments. In Abstracts of the First
IUFRO Workshop on Qualitative and Quantitative Assessment of
Forest Sites with Special Reference to Soil, 10 September –
15 September 1984, Birmensdorf, Switzerland. Edited by
W. Bosshard. Swiss Federal Institute of Forestry Research, Bir-
mensdorf, Switzerland. pp. 24–25.
Rohlf, F.J., and Sokal, R.R. 1995. Statistical tables. 3rd ed. Freeman,
New York.
Rundgren, S. 1994. Earthworms and soil remediation: liming of
acidic coniferous forest soils in southern Sweden. Pedobiologia,
38: 519–529.
Satchell, J.E., and Lowe, D.G. 1967. Selection of leaf litter by Lum-
bricus terrestris.In Progress in soil biology. Edited by O. Graff
and J.E. Satchell. Vieweg, Braunschweig, Germany, and North-
Holland Publishing Company, Amsterdam, Netherlands. pp. 102–119.
Schaefer, M. 1991. Fauna of the European temperate deciduous for-
est. In Ecosystems of the world. VII. Temperate deciduous forests.
Edited by E. Röhrig and B. Ulrich. Elsevier, Amsterdam, Nether-
lands. pp. 503–525.
Schaefer, M., and Schauermann, J. 1990. The soil fauna of beech
forests: comparison between a mull and a moder soil. Pedobiolo-
gia, 34: 299–314.
Scheu, S., and Parkinson, D. 1994. Effects of invasion of an aspen
forest (Canada) by Dendrobaena octaedra (Lumbricidae) on plant
growth. Ecology, 75: 2348–2361.
Sokal, R.R., and Rohlf, F.J. 1995. Biometry. The principles and prac-
tice of statistics in biological research. 3rd ed. Freeman, New York.
Staaf, H. 1987. Foliage litter turnover and earthworm populations in
three beech forests of contrasting soil and vegetation types. Oe-
cologia, 72: 58–64.
Sulkava, P., Huhta, V., and Laakso, J. 1996. Impact of soil faunal
structure on decomposition and N mineralisation in relation to
temperature and moisture in forest soil. Pedobiologia,
40: 505–513.
Thill, A., Dethioux, M., and Delecour, F. 1988. Typologie et poten-
tialités forestières des hêtraies naturelles de l’Ardenne Centrale.
Institut pour l’encouragement de la recherche scientifique dans
l’industrie et l’agriculture, Brussel, Belgium.
Toutain, F. 1987. Activité biologique des sols, modalités et li-
thodépendance. Biol. Fertil. Soils, 3: 31–38.
Toutain, F., and Duchaufour, P. 1970. Étude comparée des bilans
biologiques de certains sols de hêtraie. Ann. Sci. For. 27: 39–61.
Toutain, F., and Védy, J.C. 1975. Influence de la végétation forestière
sur l’humification et la pédogénèse en milieu acide et en climat
tempéré. Rev. Ecol. Biol. Sol, 12: 375–382.
Toutain, F., Diagne, A., and Le Tacon, F. 1987. Effets d’apports
d’éléments minéraux sur le fonctionnement d’un écosystème for-
estier de l’Est de la France. Rev. Ecol. Biol. Sol, 24: 283–300.
Turvey, N.D., and Smethurst, P.J. 1985. Variations in wood density
of Pinus radiata D. Don across soil types. Can. J. For. Res. 15: 43–49.
Ponge et al. 2063
© 1997 NRC Canada

Ulrich, B. 1986. Natural and anthropogenic components of soil acidi-
fication. Z. Pflanzernaehr. Bodenkd. 149: 702–717.
Van der Drift, J. 1962. The soil animals in an oak-wood with different
types of humus formation. In Progress in soil zoology. Edited by
P.W. Murphy. Butterworths, London, United Kingdom.
pp. 343–347.
Van der Drift, J. 1963. The disappearance of litter in mull and mor in
connection with weather conditions and the activity of the macro-
fauna. In Soil organisms.EditedbyJ. Doeksen and J. Van der Drift.
North-Holland Publishing Company, Amsterdam, Netherlands.
pp. 125–133.
Van Straalen, N.M., Kraak, M.H.S., and Denneman, C.A.J. 1988.
Soil microarthropods as indicators of soil acidification and forest
decline in the Veluwe area, the Netherlands. Pedobiologia,
32: 47–55.
Wolters, V. 1988. Effects of Mesenchytraeus glandulosus (Oligo-
chaeta, Enchytraeidae) on decomposition processes. Pedobiolo-
gia, 32: 387–398.
Can. J. For. Res. Vol. 27, 1997
2064
© 1997 NRC Canada