Ants as Ecosystem Engineers in Natural Restoration of Human Made Habitats

Abstract

Three to four dominant seed-transporting ant species of different size categories
(Tetramorium caespitum, Lasius niger, Formica rufibarbis/Formica pratensis) on the
plateau of abandoned ore sedimentation basin (tailings containment) were studied as
pioneer and subsequent colonisers of this industrial waste deposit, from the viewpoint of
their functioning in plant seed dispersal. We examined the role of ants in primary
vegetation succession. Experiments of seed removal by ants with plant species found withinclose proximity of tailings were related to the succession. Ant activity generates a
considerable shift in the the quality of the colonised surface, as they collectively act as
ecosystem engineers.

Full text

Journal of Landscape Ecology (2013), Vol: 6 / No. 1. aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
18
ANTS AS ECOSYSTEM ENGINEERS IN NATURAL
RESTORATION OF HUMAN MADE HABITATS
PAVEL KOVÁŘ1, PAVEL VOJTÍŠEK2 & IRENA ZENTSOVÁ2
1Department of Botany & 2Institute of the Environment, Faculty of Science, Charles
University in Prague, Benátská 2, 128 01 Prague 2, Czech Republic
Received: 15th May 2013, Accepted: 17th July 2013
ABSTRACT
Three to four dominant seed-transporting ant species of different size categories
(Tetramorium caespitum, Lasius niger, Formica rufibarbis/Formica pratensis) on the
plateau of abandoned ore sedimentation basin (tailings containment) were studied as
pioneer and subsequent colonisers of this industrial waste deposit, from the viewpoint of
their functioning in plant seed dispersal. We examined the role of ants in primary
vegetation succession. Experiments of seed removal by ants with plant species found within
close proximity of tailings were related to the succession. Ant activity generates a
considerable shift in the quality of the colonised surface, as they collectively act as
ecosystem engineers.
Keywords: human-made habitats, industrial-waste deposits, abandoned tailings
containment, primary vegetation succession, plant species diversity, dispersal of plant
seeds, ant-plant interaction, myrmecochory, ecosystem engineers, bioturbation
INTRODUCTION
Ants often play an important role in the dynamics of plant communities acting as seed
dispersal agents (e.g. Hölldobler et Wilson 1990, Jolivet 1996, Kovář et Kovářová 1998).
Foraging workers are likely to take seeds as food into their nests, carrying them a distance
away from the parent plant. Although ants disperse seeds over relatively short distances,
advantages commonly associated with this dispersal strategy include avoidance of parent
competition and of density-dependent predation, as well as dispersal towards microsites
favorable to germination (Wolf et Debussche 1999, Kovář et al. 2001). Activity of
successionally alternating ant species during decades imply significant change of the
substrate by bioturbation during nest-building (Dostál et al. 2005, Frouz et Jílková 2008)
and dynamics in vegetation development (Vlasáková et al. 2009, Vojtíšek 2012). To build
and maintain mounds, ants transport large amounts of substrate from deeper layers (King
1981, Frouz 1996), through a process known as bioturbation (e.g., Meysman et al. 2006).
As a result, ants influence concentrations of chemical elements, soil texture, amount of soil
organic matter, activity and biomass of microfauna, pH as well as the other soil properties
(Czerwinski et al. 1971, Culver et al. 1983, Frouz et al. 1997). This phenomenon, which
reflects physical and chemical effects of the soil biota´s microdisturbations such as anthills
or molehills - having a positive effect on vegetation succession – is usually summarised in
the term ecosystem engineering (e.g., Jones 1994, Wright et Jones 2004, Jouquet et al.
2006, Sanders et van Veen 2011). It is apparently functional on spoil substrates enriched in

Journal of Landscape Ecology (2013), Vol: 6 / No. 1
19
heavy metals – in abandoned tailings of our interest, too, where adjacent processes of the
soil improvement are facilitated in this way (Kovář 2004, Kovář et al. 2011, Štefánek et al.
2012).
Aim of the study
The main aim of the study is to compare plant cover on the tailings containment and the
influence of the ant activity on dispersal of plants, and facilitation of primary succession of
vegetation after approximately 10 years of spontaneous development. Periods of
comparison: the first one in 1998-2000 (sparse mosaic of lichens, herbs and shrubs;
Jarešová et Kovář 2004) and the second one in 2011-2012 (differentiated mosaic including
tree stands; Vojtíšek 2012).
Study site
An abandoned sedimentation basin with ore deposits is located on the left bank of the
Elbe river, close to the old surface quarry near Chvaletice in the northern part of Železné
hory hills, Czech Republic. Geographical location: 50°02' N, 15°26' E; altitude 200 m. Area
under spontaneous natural restoration: 40 ha. Habitat parameters: Waste sludge of Fe-Mn
ores with high sulphur and phenol contents were hydraulically transferred and deposited in
tailings containment (abandoned gradually in 1975-1980). The non-reclaimed basin has
remained largely treeless or non-vegetated and no manipulations were carried out after
abandonment (across the total surface area in 1984).
METHODS
Series of vegetation succession (in 1998 and 2011)
Physiognomically distinct vegetational types correlating with successional stages were
selected: (1) in 1998 – (a) mosaic of open areas with sporadic vegetation of mosses, lichens
and herbs, almost predominantly with the grass Calamagrostis epigejos), (b) mosaic of herb
stands with seedlings and saplings of aspen (Populus tremula) and birch (Betula pendula),
and (c) relatively tall (max. 3 m) stand of birch and aspen; (2) in 2011 – (A) open space
with cryptogams, (B) herb stand (grassland), (C) shrubby stand (seedlings of Betula
pendula, Populus tremula, Salix sp. div., Pinus sylvestris, Tilia platyphyllos, Cerassus
avium, Sarothamnus scoparius), and (D) the most developed community with trees,
mainly birch (Betula pendula) and aspen (Populus tremula), about 5 m tall.
Phytocoenological relevés in plant stands were recorded using seven-grade scale of
abundance (Braun-Blanquet 1928). The results of this analysis in more detailed description
were published in the previous study (Kovář et al. 2011).
Frequency of ant nests in different plant stands
Dominant ant species were studied in the surface plateau of the abandoned sedimentation
basin: Tetramorium caespitum (Linnaeus, 1758); Lasius niger (Linnaeus, 1758), and
Formica rufibarbis (Fabricius, 1793) – in 1998-2000; the last one was partly substituted by
Formica pratensis (Retzius, 1783) – in 2011. Each of these species represents a different
size category - Tetramorium being the smallest, Lasius being medium-sized, and Formica
the largest. Inside above-mentioned successional formations, 10 x 10 m quadrats were
randomly set, three in each vegetation type. The plots were fixed and marked, closely
prospected and every location of ant nest was recorded.

Journal of Landscape Ecology (2013), Vol: 6 / No. 1. aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
20
Ants and seed removal experiments
Ant activity supporting dispersal of plants and facilitating primary succession of
vegetation was studied with the help of repeated seed removal experiments (after the period
of approximately 10 years). Seeds were offered in especially constructed two-layer round
dishes, protecting seeds from other predators and from rain and the wind. The base of each
dish had a low edge to avoid problems with access, even by the smallest ant Tetramorium
caespitum. As the subject of this testing was only attractiveness of the seeds, not the
distance of their dispersal, seeds were exposed next to ant nests. In a growing season, seeds
were always offered from June until September when seeds were mature. The length of
exposure was eight hours (10.00 a.m. - 6.00 p.m.). Seeds of 20 plant species (1998-2000)
and more (2011-2012) collected around the ore deposits (just as they ripened) were used for
this experiment. In both periods, three size categories of dominant ants were tested in seed
removal experiments: (1) 1998-2000: Tetramorium caespitum, Lasius niger, Formica
rufibarbis, (2) 2011-2012: Tetramorium caespitum, Lasius niger, Formica pratensis.
Succession of ant species over ten years produced a two-fold increase in their diversity
(when Formica sanguinea and F. pratensis represent semi-parasitic species using the nest
occupation of the pioneers Formica cunicularia or F. rufibarbis.
Table 1: Parameters of particular plant species used in offer experiments (1998-2000)
Species
Life
Form
Length of
life
Type of
distribution
Family
Acetosella vulgaris
Th
A
ANIMa
Pgo
Agrostis tenuis
H
P
ANIMa+WIND
Gra
Anthoxanthum odoratum
H
P
ANIMa
Gra
Carex vulpina
H
P
ANIMa
Cyp
Cerastium vulgatum
H
P
UNSP
Car
Holcus lanatus
H
P
UNSP
Gra
Hypericum perforatum
H
P
WIND
Hyp
Chrysanthemum leucanthemum
H
P
UNSP
Com
Lotus corniculatus
H
P
UNSP
Leg
Luzula multiflora
H
P
ANIMe
Jce
Lychnis flos-cuculi
H
P
WINDc
Car
Plantago lanceolata
H
P
ANIMm
Pla
Poa pratensis
H
P
UNSPag+WIND
Gra
Potentilla argentea
H
P
UNSP
Ros
Rumex obtusifolius
H
P
ANIMa
Pgo
Trifolium dubium
Th
A
ANIMa
Leg
Tripleurospermum inodorum
Th
A
UNSP
Com
Vicia hirsuta
Th
A
UNSP+ANIM
Leg
Vicia sativa
Th
A
UNSP+ANIM
Leg
Vicia tetrasperma
Th
A
UNSP
Leg
Life form: H – hemicryptophyte, Th – therophyte. Length of life: P – perennials, A – annuals. Distribution:
UNSP – non-specified, ANIMe – zoochory, presence of elaiosome, ANIMa – epi-zoochory (hair, feathers),
ANIMm – epi-zooochory (sticky secretions). WIND – anemochory. Plant family: Pgo – Polygonaceae, Gra –
Gramineae, Cyp – Cyperaceae, Car – Caryophyllaceae, Hyp – Hypericaceae, Com – Compositae, Leg –
Leguminosae, Jce – Juncaceae, Pla – Plantaginaceae, Ros – Rosaceae. According to Grime et al. (1988), and the
last author’s observation.

Journal of Landscape Ecology (2013), Vol: 6 / No. 1
21
Species diversity and effectiveness of myrmecochory
Species-area curves were constructed (2012) for plants by extending the sampling area
and counting the present plant species in each doubled quadrat (all this in four repetitions
on the (1) area with ant nests, and (2) area without ant nests of the abandoned tailings
containment. The potential argument is that differences in plant species diversity, observed
within both areas within the deposit plateau, were not primarily caused by presence/absence
of ant nests. Instead, these differences were attributed to different microhabitat conditions
that evolved the following transplantation control. Five plots 15 x 15 cm were placed
randomly in the area with an absence of ant nests, and the seeds of six relatively common
plant species in many places on the deposit were sown in them: Holcus lanatus, Vicia
hirsuta, Plantago lanceolata, Agrostis tenuis, Trifolium repens, Rumex acetosella. Their
seedlings were monitored during the growing season (2012).
Substrate properties
Six soil samples were taken from each of three layers (0-2 cm, 2-4 cm, and 4-8 cm,
respectively). The following characteristis of the substrate were analysed: incinerable C,
available P, pH and conductivity (measure of salinity) within all 4 vegetation types and/or
habitats (see above). All analytical procedures used were the same, and comparable with
previous studies in the locality (Kovář 2004; in details Vojtíšek 2012).
RESULTS
Ant species and ecological succession
Successional shift from three very weakly-defined clusters of vegetation in 1998
consisted of four relatively well differentiated types of habitats in 2012 (see Methods).
Results of the earlier term within the vegetation succession (Jarešová et Kovář 2004)
showed that lichen and moss species together with the annuals Vicia tetrasperma and V.
hirsuta occupy one semi-open environment. In contrast, woody species and other vascular
plants (Calamagrostis epigejos, Phragmites australis) represent the most advanced type of
plant stand.
It is possible to conclude that differences among vegetation relevés relate to vegetation
coverage. Since all three ant species studied in the period 1998-2000, namely Formica
rufibarbis, indicate their preference to occupy habitat mosaics with open surfaces.
However, the mutual relationship between ants and plants does not show significant
dependencies between an ant species and a particular plant species. Tetramorium caespitum
and Formica rufibarbis are closely associated with the presence of herbs, mostly of Vicia
species, and Calamagrostis epigejos. From nest numbers listed in Table 2, it is apparent
that Lasius niger is the most widespread ant of all the species in this locality. This species
builds its nests mostly in places with low growth of aspen and birch seedlings and saplings.
Tetramorium caespitum and Formica rufibarbis prefer surfaces with sparse vegetation.
The pattern of nests belonging to the aforementioned ant species during the 2011-2012
survey period, confirms the previous knowledge about their dependence on habitat and/or
vegetation types. An increasing proportion of tall woody stands slightly supressed typically
pioneer ant colonisers such as Formica rufibarbis or F. cunicularia. These species probably
became the target of attacks from successionally advanced species (Formica sanguinea,
F.pratensis) known as semi-parasitic invaders to a suitable phase of the ecosystem
development. This was the reason why attention was paid to behaviour of newly established
nests of Formica pratensis, apparently bound with periodic shadow by tree coverage and/or

Journal of Landscape Ecology (2013), Vol: 6 / No. 1. aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
22
relatively dense vegetation. Intraspecific competition among nests of this Formica species
probably caused the success of the one nest (ant hill) with very high numbers of individuals
and wide territorial range of activity (tens of metres). However, it is relatively weak at
competing with the other size categories of ants (Tetramorium caespitum, Lasius niger).
Namely Lasius niger seems to be highly mobile with the construction of new nests in this
industrial environment, and it exhibits a ruderal-type adaptive strategy in diverse
microbahitat mosaics. A total of 12 ant species were identified within the locality: Formica
cunicularia, F. rufibarbis, F. cinerea, F. sanguinea, F. pratensis, Lasius niger, L. flavus, L.
platythorax, Myrmica schencki, M. rubra, Tetramorium caespitum, Temnothorax
crassipinus (underlined names indicate the species recorded in 2012 against the year 1998).
Table 2: Average numbers of ant nests per quadrat (10 x 10 m2) related to vegetation
types: 1 – mosaic of open surface with cryptogams and sparse herb vegetation, 2 – mosaic
of herb stands, and seedlings or saplings of aspen and birch, 3 – woody stands with taller
growth of aspen and birch.
Vegetation
type
Tetramorium
caespitum
Lasius
niger
Formica
rufibarbis
1
2
3
total
5,00
4,33
4,33
13,67
4,67
8,67
7,00
20,33
1,33
0,00
0,00
1,33
Assessment of variability in plant species richness on the abandoned sedimentation
basins (Vaňková et Kovář 2004) demonstrated high heterogeneity of especially the ore-
washery deposits – such as in Chvaletice. Relatively distinct groups of plant taxa associated
with either the ash-slag or ore-washery deposits were distinguished. However, there are
numerous groups of species of wider ecological amplitude lacking a strict connection to the
type of industrial sediments (the bodies of deposits represent early stages of succession
and/or persist in early stages of succession for a long time). Transport of diaspores from the
surrounding landscape, mainly through zoochory and anemochory, is among the factors
governing the development of plant communities in the initial and/or young stages of
succession (Kovář 2004).
A typical S-shape in species-area curves was observed when number of plant species
plotted against increasing plots was demonstrated for the tailings surface (Fig. 1). However,
the steepness of the curve was considerably greater in the area colonised by the ant nests, in
comparison to the area without the ant nests and/or with only their sporadic incidence (the
highest number of nests in the frame of four quadrats 10 x 10 m monitoring was 30 in the
first case, and 9 in the second one). Low numbers of plant species were found in the interior
part of the abandoned sedimentation plateau in 1998 (less than 10), while in 2012 numbers
were approximately four-times higher. A similar relationship is observed for the areas with
extremely rare presence of ant nests in comparison with dense incidence of ant nests (Fig.
1).

Journal of Landscape Ecology (2013), Vol: 6 / No. 1
23
Fig. 1: Species-area relationsips on the abandoned sedimentation basin in Chvaletice,
in the surface plateau area with only rare occurrence of ant nests (left) and in the
surface plateau area with high density of ant nests (right) – data from 2011.
Experiments with removal and transportation of seeds by ants
Data from experiments with seed offered to particular ant colonies (1998-2000, Table 1)
were processed with the help of ANOVA. First, the species specific differences in seeds
collecting were tested, including test of significant differences. According to Tukey test,
Lasius niger showed the highest seed dispersal of the three ant species. However, there was
no significant difference in seed dispersal between Formica rufibarbis and Tetramorium
caespitum (Fig. 2). Figure 3 shows a different trend of seed collecting by Lasius niger
during the growing season (1999) compared with the two other ant species. Analogical
comparison at the same abandoned tailings containment 12 years later (Tetramorium
caespitum, Lasius niger,Formica pratensis) shows dominance in the seed dispersal activity
of the largest size category of ant species (Fig. 4). In the first case (Fig. 2) Lasius niger
dominates in gathering of seeds, in the second one (Fig. 4) Formica pratensis is the most
active species in the seed collection process (Table 3). In general, the results showed
significant differences in seed dispersal of several plant species related to collecting
effectiveness of ants.Ant workers of a particular species preferred certain plant species
(Jarešová et Kovář 2004). For example, Chrysanthemum leucanthemum was more preferred
than other plant species. In comparison, Hypericum perforatum and Vicia sativa were
collected in very small amounts.

Journal of Landscape Ecology (2013), Vol: 6 / No. 1. aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
24
±1.96*Std. Err.
±1.00*Std. Err.
Mean
a b a
NUMBER OF SEEDS GATHERED BY ANTS
(MEAN VALUES FOR ONE OFFER)
0
10
20
30
40
50
60
70
80
90
100
110
120
T. caespitum L. niger F. rufibarbis
Table 3: Two groups of plant seeds for experimental offering to ants according to
seasonal period (first group harvested in the second half of June and the first half of July –
left, second group harvested in the second half of July and the first half of August), 2011-
2012.
First group of seeds for offer
Second group of seeds for offer
Potentila argentea (Po.Arg)
Holcus lanatus (Ho.Lan.)
Rumex acetosella (Ru.Act)
Hieracium laevigatum (Hie.Lae.)
Holcus lanatus (Ho.Lan.)
Plantago lanceolata (Pla.Lanc.)
Calamagrostis epigejos (Cal.Ep)
Trifolium arvense (Tri.Arv.)
Trifolium dubium (Tri.Dub)
Calamagrostis epigejos (Cal.Ep.)
Poa pratensis (Poa)
Melilotus officinalis (Mel.Off.)
Trifolium repens (Tri.Rep.)
Senecio jacobaea (Sen.Jac.)
Vicia hirsuta (Vic.Hirs.)
Lotus corniculatus (Lot.Cor.)
Agrostis tenuis (Agr.Ten.)
Picris hieracioides (Pi.Hie.)
Plantago lanceolata (Pla.Lanc.)
Trifolium repens (Tri.Rep.)
Lychnis flos-cuculi (Lich.F.C.)
Vicia cracca (Vic.Cra.)
Lathyrus tuberosus (Lat.Tub.)
Cirsium arvense (Cir.Arv.)
Centaurium erythraea (Ce.Ery.)
Fig. 2: Comparison of average numbers of effective seed collecting by three ant species
from arranged offers (the graph includes all data from the growing seasons 1998-2000).

Journal of Landscape Ecology (2013), Vol: 6 / No. 1
25
Fig. 3: Box plots show average values of seed collecting by a particular ant species
during 4 months of a season (data from 1999).
Fig. 4: Comparison of average numbers of effective seed collecting by three ant species
(F – Formica pratensis, L – Lasius niger, T – Tetramorium caespitum) from arranged
offers (2011-2012).
Major behavioral similarity is noticeable between the ant species Tetramorium caespitum
and Formica rufibarbis, despite the great difference in their body size: both of them mostly
collected large diaspores of Luzula multiflora and Carex vulpina. Another major collecting
activity was noted for these two species with respect to seeds of Anthoxanthum odoratum
±1.96*Std. Err.
±1.00*Std. Err.
Mean
MONTH
NUMBER OF SEEDS GATHERED BY ANTS
(MEAN VALUES FOR ONE OFFER)
Tetramorium cae spitum
0
20
40
60
80
100
120
VI VII VIII IX Las ius nig er
VI VII VIII IX
Formic a rufibar bis
0
20
40
60
80
100
120
VI VII VIII IX

Journal of Landscape Ecology (2013), Vol: 6 / No. 1. aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
26
and Potentilla argentea. Lasius niger mostly collected Anthoxanthum odoratum and P.
argentea, followed by Trifolium dubium and Vicia hirsuta. For generalisation of these
results, we must to take into account that dispersal of plant diaspores can be, to a certain
extent, affected by the season when the seeds were offered.
Substrate properties
In general, basic substrate properties of studied deposits in Chvaletice are a long-term
outcome of the process of sulphidic mineralization which was described by Rauch (2004).
More ecotoxicological aspects in the locality is treated in wider publication (Kovář
2004). Sampling of the substrate covered all 4 types of habitats distinguished for the
purposes of studying ant-plant interactions in the abandoned tailings containment during
period of 2011-2012: open space with cryptogams (A), herb stand (B), shrubby stand (C)
and the most developed community with trees (D). Six soil samplings were taken from each
of the three layers (0-2 cm, 2-4 cm, and 4-8 cm, respectively). The following characteristis
of the substrate were analysed: incinerable C, available P, pH and conductivity (measure of
salinity). A very rough illustration of the extent of the above-mentioned variables, which
could result in significant obstacles for the continuity of the ant colonisation of the
substrate, is shown in Table 4.
Table 4: Scheme of pairs of habitat types on the surface of tailing containment in
Chvaletice which significantly differ in a particular soil feature: open space with
cryptogams (A), herb stand (B), shrubby stand (C) and the most developed community with
trees (D).
Depth of soil profile
0-2cm
2-4cm
4-8cm
Substrate parameter
Carbon
AD
CD
Phosphorus
AB
AD
CD
BC
Salinity
AB
AB
AD
AD
AD
BC
BC
CD
CD
CD
pH
AB
AB
AD
AD
AD
BC
CD
CD

Journal of Landscape Ecology (2013), Vol: 6 / No. 1
27
DISCUSSION
Density of occurrence of ant nests within habitats and/or vegetation types, on the
abandoned plateau of the ore washery sedimentation basin, shows a relatively high degree
of dependence of ant species on vegetation cover in comparison with a particular plant
species. The most widespread ant species Lasius niger distributes its colony on places
sparsely overgrown by juvenile shrubs and trees. A similar observation was published by
Frouz (1996), who monitored occurrence of this species on dumps formed during brown
coal mining in the Sokolov region (western Bohemia, Czech Republic). Under
environmental conditions comparable with those described in this work, Lasius niger was
also a dominant species. It reached the highest values of abundance and biomass in the
initial stages of vegetation succession on dumps covered with shrub or herb vegetation.
However, on the ore deposit, Lasius niger tended to avoid surfaces overgrown by only herb
vegetation - evidently due to extremely high temperatures on the barren ore deposit surface
(60°C). The moss and lichen layer occurring in the successional stage with tree seedlings
represents a compact but dynamic cover of the soil surface (Pohlová 2004, Hroudová et
Zákravský 2004). The cover which is overheated in sunny days protects the soil below.
The space between the compact substrate and the surface layer offers eligible interface
shelter for ant migration and food transport including plant diaspores. In experiments with
Lasius niger (Brian et al., 1965), this ant survived on very dry soils only if adequate shade
was provided. It is ppossible that aspen and birch growing on ore deposits fulfil this
function. The two remaining ant species (Tetramorium caespitum and Formica rufibarbis)
resisted desiccation more than Lasius niger. This tolerance could explain higher occurrence
of those two ant species in areas with sparse vegetation.
In general, Lasius niger was the most successful ant species in myrmecochorous dispersal
of plants on the surface of the ore deposit. In comparison, in heathlands of southern
England, Tetramorium caespitum exhibited the greatest interest in seeds (Brian 1964).
Lasius niger showed a stronger preference to aphid honeydew than other ant species. The
most efficient seed collecting ant in a semi-natural ecosystem of mountain pasture was also
Tetramorium caespitum, one of 12 ant species present in this ecosystem (Kovář et
Kovářová 1998). However, seedlings of Vicia tetrasperma or Vicia hirsuta were observed
relatively frequently under the installed shelters covering the open substrate of the ore
deposit, which ants inhabited with their colonies - both the Vicia species were included in
seed offer experiments, and consequences of myrmecochory of this type are described in
numerous papers (e.g., Wilson 1971, Hölldobler et Wilson 1990, Jolivet 1996, Agosti et al.
2000). Preference of this ant to a specific plant´s seeds could explain the onset of these
annual Vicia´s following fire in August 1994 (Štefánek 2004).
Brian et al. (1967) also noted higher activity of ants in seed collecting in June and July,
compared to August and September. It is explained by increased dietary requirements due
to nurturing big larvae of sexual individuals in these months. These findings are similar to
our results obtained from the abandoned ore deposit in Chvaletice. The increase was
significant only in July (together with Tetramorium caespitum), but increased seed
dispersal was also exhibited by other species. Major differences between the months might
have been suppressed by low offer of insect feed in this generally poor biotope on ore
deposit, resulting in greater seed collecting activity throughout the year. Another
explanation is that seeds also serve as construction material for ant nests (lack of typical
soil, humus and litter with dead plant material are among the limiting factors). This use of
seeds is mentioned by a number of authors (Gorb et Gorb 1995, Gorb et Gorb 2000, Gorb et
al. 1997). Functioning as a vector of organic litter transport, the ants play a very important

Journal of Landscape Ecology (2013), Vol: 6 / No. 1. aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
28
role in balancing extremity of this environment (overheating of the soil surface, high
concentration of heavy metals in the substrate, salinisation etc.), improving the substrate
conditions and supporting life and/or successional restoration of the habitat (Agosti et al.
2000).
Differentiation of the areas on deposit surface plateau into parts (1) poor in presence of
ant nests and (2) rich in density of ants and their nests – correlated with plant species
diversity (1 – low, 2 – high) – leads to the question: What is primary? Is is enviroment
unfriendly for establishment of plant seedlings, or does the problem consist of poor
availability and/or dispersal of seeds by myrmecochory? The transplantation experiment
with sowing the seeds of the most common plant species extensively spread throughout the
deposit plateau, rather supported the second hypothesis. Germination success of seeds
(previously tested for germination ability in laboratory conditions) on that part of deposit
with and/or without rare ant nests, confirmed the capacity of local microhabitats to facilitate
the growth of plants (Table 5). It is apparent that very weak success in germination is
exhibited by Plantago lanceolata and Trifolium repens. However, at the end of the growing
season several seedlings of Trifolim repens appeared as well as Rumex acetosella, which
can be presented as an effect of dormancy. Plant species Vicia hirsuta, Rumex acetosella
and Agrostis tenuis, followed by Holcus lanatus were successful. However, Vicia hirsuta
suffered from increased mortality of seedlings (three weeks post-germination) later in the
season. A certain degree of self-thinning after succesful germination of seeds of tested
plants, had a positive effect in increasing survival of the remaining individuals to maturity
and achieving successful reproduction. Similar observations were made by Dauber etal.
(2008), Lenoir (2009), Veen et Olff (2011) and Chen et Li (2012).
Table 5: Degree of germination and survival success of chosen plant seeds sown into
the deposit surface area characterised by absence and/or rarity of ant nests, and low
plant species diversity.
Number of germinated and survived seeds of 60 in total
Number of plot/plant
species
Holcus
lanatus
Vicia
hirsuta
Plantago
lanceolata
Rumex
acetosella
Agrostis
tenuis
Trifolium
.repens
1
4(4)
7(0)
0
10(10)
7(6)
1(1)
2
1(0)
6(6)
3(3)
2(2)
5(4)
0
3
2(0)
4(0)
0
1(1)
10(7)
0
4
7(7)
4(1)
0
5(5)
0
1(1)
5
5(4)
10(10)
0
10(10)
9(7)
0
SUM
19(15)
31(16)
3(3)
28(28)
31(24)
2(2)
CONCLUSIONS
• Biodiversity: The results show that species diversity of ants increased moderately (rather,
dominant species has changed) and diversity of plants increased several times during one
decade.
• Plant-ant relationship: there is a strong dependence of plant species diversity on presence
of numerous ant nests.
• Role of myrmecochory: manipulative experiments with plant seed removal (offer of seeds
near ant nests) demonstrated plant dispersal.

Journal of Landscape Ecology (2013), Vol: 6 / No. 1
29
• Survival of seedlings established after the ant transport of seeds into ant nests confirmed
the role of ants as eco-engineers in restoration.
• Selected parameters of the substrate in abandoned tailings containment show differences
among four habitat types (successional stages).
• Extreme soil conditions are balanced in areas inhabited by ants which facilitate the
acceleration of vegetation succession.
ACKNOWLEDGEMENTS
We thank to P. Pech (Dept. Biology, Faculty of Science, University of Hradec Králové)
and J. Frouz (Inst. Environment, Faculty of Science, University of Prague) for
determinantion of ant species. The work has been supported by the grant no. 206/93/2256
of the Grant Agency of the Czech Republic “Biotic interactions during vegetation
succession on toxic substrates”, grant no. 200/1997/B/BIO of the Grant Agency of Charles
University and by the research project MSM 113100004, chapter “Ecological plasticity and
taxonomical variability of expansive plant species“ of the Czech Ministry of Education.
REFERENCES
AGOSTI, D., MAJER, J. D., ALONSO, L. E., SCHULTZ, T. R. (EDS.), 2000. Ants. Standard
methods for measuring and monitoring biodiversity. - Smithsonian Institution Press,
Washington and London.
BRAUN-BLANQUET, J., 1928: Pflanzensoziologie. - Springer Verlag, Berlin.
BRIAN, M. V., 1964: Ant pattern and density in southern English heath. - Journal of Animal
Ecology, 33: 451-461
BRIAN, M. V., HIBBLE, J., STRADLING, D. J., 1965. Ant pattern and density in southern
English heath. - Journal of Animal Ecology, 34: 545-555.
BRIAN, M. V., ELMES, G., KELLY, A. F., 1967. Population of the Tetramorium caespitum
Latreille. - Journal Animal Ecology, 36: 337-342.
CHEN Y.-W., LI, X.-R., 2012. Spatio-Temporal Distribution of Nests and Influence of Ant
(Formica cunicularia Lat.) Activity on Soil Property and Seed Bank after Revegetation in
the Tengger Desert. - Arid Land Research and Management, 26(4): 365-378.
CZERWINSKI, Z., JAKUBCZYK, H., PETAL, J., 1971. Influence of ant hills on the meadow
soils. - Pedobiologia, 11: 277-285.
CULVER, D.C., BEATTIE, A.J., 1983. Effect of ant mounds on soil chemistry and vegetation
patterns in a Colorado montane meadow. – Ecology, 64: 485–492.
DAUBER, J., NIECHOJ, R., BALTRUSCHAT, H. ET AL., 2008. Soil engineering ants increase
grass root arbuscular mycorrhizal colonization. - Biology and Fertility of Soils, 44(5): 791-
796.
FROUZ, J., 1996. Ants (Hymenoptera: Formicidae) in dump after brown coal excavations
and the surrounding within the Sokolovsko Region. - Collect. Distr. Museum, Most, Ser.
Nat., 18: 45-51. [in Czech]
FROUZ, J., ŠANTRŮČKOVÁ, H., KALČÍK J., 1997: The effect of wood ants (Formica
polyctena Forest.) on the transformation of phosphorus in a spruce plantation. –
Pedobiologia, 41: 437–447.

Journal of Landscape Ecology (2013), Vol: 6 / No. 1. aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
30
GORB, E., GORB, S., 1995. Removal rates of seeds of five myrmecochorous plants by the
ant Formica polyctena (Hymenoptera: Formicidae). - Oikos 73: 367-374.
GORB, E., GORB S., 2000. Effects of seed aggregation on the removal rates of elaiosome-
bearing Chelidonium majus and Viola odorata seeds carried by Formica polyctena ants. -
Ecological Research, 15: 187-192.
GORB, S., GORB, E., SINDLAROVSKAJA, YU., 1997. Interaction between the non-
myrmecochorous herb Galium aparine and the ant Formica polyctena. - Plant Ecology,
131: 215-221.
HÖLLDOBLER, B., WILSON, E. O., 1990. The Ants. - Belknap Press, Cambridge,
Massachusetts.
HROUDOVÁ, Z., ZÁKRAVSKÝ, P., 2004. The influence of the moss layer on soil surface
microclimate in an abandoned ore-washery sedimentation basin. - IN: KOVÁŘ P. (ED.):
Natural Recovery of Human-Made Deposits in Landscape (Biotic Interactions and
Ore/Ash-Slag Artificial Ecosystems) - Academia, Prague, p. 235-247.
JAREŠOVÁ, I., KOVÁŘ, P., 2004. Interactions between ants and plants during vegetation
succession in the abandoned ore-washery sedimentation basin in Chvaletice. - IN: KOVÁŘ
P. (ED.): Natural Recovery of Human-Made Deposits in Landscape (Biotic Interactions and
Ore/Ash-Slag Artificial Ecosystems) - Academia, Prague, p. 300-310.
JONES, C.G., LAWTON, J.H., SHACHAK, M., 1994. Organisms as ecosystem engineers. –
Oikos, 69: 373–386.
JOUQUET, P., DAUBER, J., LAGERLOF, J., LAVELLE, P., LEPAGE, M., 2006. Soil
invertebrates as ecosystem engineers:intended and accidental effects on soil and feedback
loops. - Applied Soil Ecology, 32: 153–164.
KING, T.J., 1981. Ant-hills and grassland history. - J. Biogeogr., 8: 329–334.
KOVÁŘ, P. (ED.), 2004. Natural Recovery of Human-Made Deposits
in Landscape (Biotic
Interactions and Ore/Ash-Slag Artificial Ecosystems) - Academia, Prague.
KOVÁŘ, P., KOVÁŘOVÁ, M., 1998. Ant herbivory - a significant factor in population
dynamics of Veronica and other temperate plant species? - Thaiszia-J.Bot., 8: 141-146.
KOVÁŘ, P., KOVÁŘOVÁ, M., DOSTÁL, P., HERBEN, T., 2001. Vegetation of anthills in a
mountain grassland: effects of mound history and of dominant ant species. - Plant Ecology,
156: 215-227.
KOVÁŘ, P., ŠTEFÁNEK, M., MRÁZEK, J., 2011. Responses of vegetation stages with woody
dominants to stress and disturbance during succession of abandoned tailings in cultural
landscape. – Journal of Landscape Ecology, 4(2): 35-48.
MEYSMAN, F. J. R., MIDDELBURG, J. J., HEIP, C. H. R., 2006. Bioturbation: a fresh look
at Darwin's last idea. – Trends in Ecology & Evolution, 21(12): 688-695.
LENOIR, L., 2009. Effects of ants on plant diversity in semi-natural grasslands. -
Arthropod-Plant Interactions, 3(3): 163-172.
POHLOVÁ, R., 2004. Changes on microsites of the moss Ceratodon purpureus and lichens
Peltigera didactyla and Cladonia sp. div. in the abandoned sedimentation basin in
Chvaletice. - IN: KOVÁŘ P. (ED.): Natural Recovery of Human-Made Deposits in
Landscape (Biotic Interactions and Ore/Ash-Slag Artificial Ecosystems) - Academia,
Prague, p. 222-234.

Journal of Landscape Ecology (2013), Vol: 6 / No. 1
31
RAUCH, O., 2004. Genesis and characteristics of orewaste sulphate soils at Chvaletice. -
IN: KOVÁŘ P. (ED.): Natural Recovery of Human-Made Deposits in Landscape (Biotic
Interactions and Ore/Ash-Slag Artificial Ecosystems) - Academia, Prague, p. 45-58.
SANDERS, D., VAN VEEN, F. J. F., 2011. Ecosystem engineering and predation: the multi-
trophic impact of two ant species. – Journal of Animal Ecology, 80(3): 569-576.
ŠTEFÁNEK, M., 2004. Secondary succession after fire on an abandoned ore-washery
sedimentation basin – different trajectories (A comparison with primary succession). - IN:
KOVÁŘ, P. (ED.): Natural Recovery of Human-Made Deposits in Landscape (Biotic
Interactions and Ore/Ash-Slag Artificial Ecosystems) - Academia, Prague, p. 248-266.
ŠTEFÁNEK, M., KOVÁŘ, P., DLOUHÁ, V., 2012. Role of fire episode,leaf litter
decomposition and mulching effects in restoration of the surface soil crust microecosystem
on abandoned tailings containment. – Journal of Landscape Ecology, 5(3): 57-69.
VAŇKOVÁ, J., KOVÁŘ, P., 2004. Plant species diversity in the biotopes of un-reclaimed
industrial deposits as artificial islands in landscape. - IN: KOVÁŘ P. (ED.): Natural
Recovery of Human-Made Deposits in Landscape (Biotic Interactions and Ore/Ash-Slag
Artificial Ecosystems) - Academia, Prague, p. 30-45.
VEEN, G. F., OLFF, H., 2011. Interactive effects of soil-dwelling ants, ant mounds and
simulated grazing on local plant community composition. - Basic and Applied Ecology,
12(8): 703-712.
VLASÁKOVÁ, B., RAABOVÁ, J., KYNCL, T., DOSTÁL, P., KOVÁŘOVÁ, M., KOVÁŘ, P.,
HERBEN, T., 2009. Ants speed up succession from grassland towards forest. - Journal of
Vegetation Science, 20: 577-587.
VOJTÍŠEK, P., 2012. Ants and primary vegetation succession on abandoned industrial-
waste deposits. – MSc Thesis (deposited in Library of the Institute of Environment, Faculty
of Science, Charles University). [in Czech]
WILSON, E. O., 1971. The Insect Societies. - Harvard Univ. Press, Cambridge,
Massachusetts.
WOLF, A., DEBUSCHE, M., 1999. Ants as seed dispersers in Mediterranean old-field
succession. - Oikos, 84: 443-452.
WRIGHT, J.P, JONES, C.G., 2004. Predicting effects of ecosystem engineers on patch-scale
species richness from primary productivity. Ecology, 85: 2071–2081.