Full text of DOI:10.1111/j.1654-109X.2002.tb00554.x

Applied Vegetation Science 5: 247-254, 2002
© IAVS; Opulus Press Uppsala.
- IMPACT OF FREE-RANGE PIGS ON MOUNTAIN PASTURES IN THE SWISS JURA -
247
Impact of free-range pigs on mountain pastures in the Swiss Jura
Vittoz, Pascal & Hainard, Pierre
Institut d’écologie, Université de Lausanne, Bâtiment de biologie, 1015 Lausanne, Switzerland;
Fax +41216924265; E-mail pascal.vittoz@ie-bsg.unil.ch
Abstract. The objective of this experiment was to study the
impact of free-range pigs foraging on mountain pasture veg-
etation. Pigs (15 week-old, mean weight 50 kg, n = 25) were
enclosed from June to mid-September in a 2-ha enclosure in
the Jura Mountains (Switzerland) and fed with a mixture of
lactoserum and cereals. The enclosure contained four different
plant communities. Eutrophic pastures on deep soil were
largely overturned, but recolonization was quick and vegeta-
tion dominated by the original species. Mesotrophic pastures
were less damaged on stony soil but were completely de-
stroyed on deep soil, recovery was slow and was characterized
by a shift of plant species in a more eutrophic direction. Four
years were not sufficient for complete recovery. Oligotrophic
calcareous pastures on shallow stony soil were not damaged.
The result of this study was that extensive breeding of pigs in
mountain pastures are harmful to plant species in mesotrophic
pastures, and should be restricted to less sensitive plant com-
munities, with a rotation of two or three different sites.
The consequences on vegetation were rarely studied
(Braund et al. 1998) and never investigated in natural
mountain pastures.
The behaviour of pigs, and their influence on vegeta-
tion, can be compared with that of wild boars (Graves
1984). This species is common in the Jura Mountains,
and they often root pastures, especially at the end of the
winter. Baettig (1980) sampled stomach contents of
wild boars and found mainly roots, rhizomes, earth-
worms, larvae and bulbs. Rooting was observed on flat
areas and in depressions, where soil is deep and wet, but
not on stony and drier slopes. In the Jura mountains
most flat areas and depressions are fertilized pastures.
They have an impoverished, common flora, and wild
boar rooting is not a threat to rare pasture plants, which
mainly grow on shallow, stony soils (Vittoz 1998).
Until now, pig breeding in the Jura Mountains has
been limited to pigsties, sometimes with a very small
exterior enclosure, which quickly becomes completely
devoid of vegetation. Free-range meat can be sold at a
higher price and free-range breeding could become more
widespread in the future. It is, therefore, important to
know the impact of such a practice on the vegetation to
avoid long-term adverse consequences on plant com-
munities and plant species diversity. In this study, we
aim (1) to assess the consequences of pig rooting on the
vegetation on different soil conditions and (2) to monitor
its recovery over four years. We intend to be able to
suggest a sustainable method for free-range pig breeding.
Keywords: Disturbance; Management; Permanent plot; Pig
breeding; Plant conservation; Recolonization; Rooting.
Nomenclature: Lauber & Wagner (2000).
Introduction
Summer grazing in pastures is the only agricultural
exploitation above 1100 m a.s.l. in the Swiss Jura Moun-
tains. For centuries, dairy cows have been farmed and
cheese has been produced on the farms in this region.
Although young cattle raised for meat are replacing this
type of exploitation, cheese production still exists on
some mountain farms. Pigs are generally fed with
lactoserum, a by-product of cheese production, and are
kept in pigsties next to the farms. Two problems occur:
contamination of water with the liquid manure (Watson
et al. 1998; Schaffter 1999) and health disorders or
disturbed behaviour due to overcrowded pigsties (Stolba
& Wood-Gush 1981). To improve breeding conditions,
and simultaneously ease manure management, trials
with free-range pigs were carried out on artificial mead-
ows (Andresen & Redbo 1999; Andresen et al. 2001) or
in semi-natural environments (Wood-Gush et al. 1990).
Location and Method
Site description
This research took place under the aegis of the
European project AGREAUALP, which started in 1995,
and aimed to study the influence of mountain agricul-
ture on underground water and rivers (Bourjot unpubl.).
The Swiss contribution investigated the consequences
of free-range pigs on soils, underground water and grass
on limestone. The experiment took place at Les
Amburnex, a summer pasture in the western part of the

248
VITTOZ, P. &HAINARD, P.
Swiss Jura mountains (46∞32' N, 6∞14' E, 1330 m a.s.l.).
Annual rainfall is 1630 mm, with ca. 140 days of snow
on the ground; mean temperature is 3.5 ∞C (Bloesch &
Calame 1994), with 120-135 days of plant growth
(Schreiber 1977). Only limestone exists in the region,
but the high rainfall leads to acidification of deep soils.
Storm showers are frequent in summer, and vegetation
rarely suffers drought.
The experimental enclosure (2 ha) was situated next
to a farm, across a small coomb on the slopes of a hill,
with a general northwestern aspect. Cows have grazed
the pasture every summer for centuries (June to Septem-
ber). A previous study described plant communities in
the region (Vittoz 1998), from which five were present
in the experimental field (Fig. 1):
1. Eutrophic pasture on deep soil, occurring on rela-
tively flat areas, regularly fertilized;
2. Mesotrophic pasture on moderately deep and stony
soil, generally on a gentle slope, sometimes mixed
with small depressions of eutrophic vegetation;
3. Oligotrophic calcareous pasture on shallow stony
soils, situated on steeper slopes;
4. Mesotrophic acid pasture on deep soil, in small de-
pressions partly shaded by trees;
5. Spruce grove (Picea abies) with sparse ground veg-
etation, situated at the top of a hill was not studied in
this experiment as it was only used as a resting place
by the pigs and was not rooted.
Mesotrophic and oligotrophic pastures have a high
plant diversity, while eutrophic pasture diversity is much
lower (Thomet 1981; Vittoz 1998).
Fig. 1. Vegetation map of the experimental site and location of
the permanent plots.
Experimental period with pigs
Cows initially grazed the pasture for one week at the
end of May, to reduce grass height at the beginning of
the season. Twenty-five piglets (15 week-old, mean
weight 50 kg) were introduced to the 2 ha area on the 3rd
of June, 1996. They were free to use the pastures, the
spruce grove, the pigsty and three scattered shelters
(Fig. 1). A mixture of lactoserum (48% dry weight) and
a rich feed (14 MJ.kg–1, mainly cereals, poor in proteins
and phosphorus; Jost 1990) was given twice a day in a
trough next to the pigsty, in sufficient quantity to feed
them all (daily mean of 18 l of lactoserum and 2.4 kg of
feed per pig during the experiment). Pigs were slaugh-
tered in four groups between the 2nd and 14th of Sep-
tember, 1996 and had a mean weight of 117.5 kg.
In autumn 1996 the farmer decided to speed growth of
eutrophic pastures on deep soil, which are the most pro-
ductive areas. Lumps of overturned earth were returned to
allow surviving plants to grow and the surface was directly
sownwithamixtureofPoapratensis, Festucarubras.l., F.
pratensis, Lotus corniculatus, Agrostis stolonifera, Trifo-
lium repens, Cynosurus cristatus, Lolium perenne and
Phleum pratense. Cows again grazed the area from Au-
gust 1997 and pigs were restricted to a small enclosure
next to the pigsty. Unfortunately, the farmer increased the
enclosure size in 1998 and 2000 and the pigs again
overturned some permanent plots (1, 4, 5 and 6; Table 1).
Sampling and data analysis
Permanent plots (n = 13) were chosen in the spring
of 1996, distributed in the different vegetation types
(Fig. 1, Table 1; note that plot 3 is in a small depression
of eutrophic pasture in the middle of a mesotrophic
pasture). By the end of the experiment the pigs had not
rooted in the plots in mesotrophic pastures (7 and 9), but
only in the near vicinity. We therefore selected two
small, completely overturned plots (8 and 10) for the
study. These new plots were situated less than 2 m from

- IMPACT OF FREE-RANGE PIGS ON MOUNTAIN PASTURES IN THE SWISS JURA -
249
Table 1. Summary of the experiment, with relevé dates, plant cover (%) and the proportion (%) of overturned soil in autumn 1996.
In the last two rows, R indicates that the plot was completely rooted.
Type of pasture
Eutrophic
Mesotrophic
Oligotrophic calcicolous Mesotrophic acid
Plot number
Area (m2)
1
20
2
16
3
4
4
16
5
1
6
1
7
15
8
1
9
4
10
1
11
9
12
20
13
16
14
6
15
9
Cow grazing, May 1996
30.05.96 (plant cover)
99
90
8
99
99
99
85
99
95
5
95
0
80
5
85
0
95
2
85
2
90
Arrival of piglets, 03 June 96
Proportion of plot overturned
Removal of pigs, 02-14.09.96
16.09.96 (plant cover)
15 100 100 100 100
100
100
100
85
2
1
7
2
98
1
95
1
85
98
95
90
3
Lumps turned back, October 96
Sowing, October 96
04-28.06.97 (plant cover)
Cow grazing, August 97
01.10.97 (plant cover)
40
95
50
55
30
35
99
15
99
25
90
98
98
95
85
20
95
R
99
98
97
98
95
90
98
98
R
95
R
99
98
R
99
98
98
80
95
90
60
85
95
95
95
98
80
90
90
21-26.09.98 (plant cover)
23-24.08.00 (plant cover)
99
99
98
95
85
90
the previous plots, on the same regular surface, and we
are confident that the original vegetation was similar.
All 15 plots were located in homogeneous vegetation,
with area varying between 1 and 20 m2 (Table 1).
Vegetation data were collected using the Braun-
Blanquet (1964) method. Cover-abundance values were
r (1-2 individuals);
+ (< 1% cover);
1 (1-5%),
2 (6-25%);
3 (26-50%);
n
yi1
yi2
n
D =
-
Â
n
(1)
i=1
y
y
Â
i=1
Â
i1
i2
i=1
where n is the total number of species that are present in
at least one of the relevés, and yi1 and yi2 correspond
respectively to the cover values of the species i in
relevés 1 and 2 (transformed with r Æ 1, + Æ 2, 1 Æ 3,
etc.). D varies between 0 for a pair of plots with the same
species in similar proportions and 2 with no common
species. We modified the equation to account for the
differences in total plant cover between the relevés
compared:
4 (51-75%).
Relevés were made before the arrival of the pigs (spring
1996, except plots 8 and 10), after the experiment (au-
tumn 1996), in spring and autumn 1997, autumn 1998
and autumn 2000 (Table 1).
n
yi1
yi2
-
n
Ê
D = log
Ë
ˆ
Ê
Ë
ˆ
¯
R
+1
Most of the species could be identified from late
spring to the beginning of the autumn. Species which
disappear in summer (Gagea lutea, Scilla bifolia, Cro-
cus albiflorus and Veratrum album ssp. lobelianum) or
develop later (Gentiana campestris and G. ciliata) were
removed from the relevés prior to analysis. A species
was considered ‘new’ for the plot if it appeared in one
relevé with a cover value of 1 or more, or in three
consecutive relevés with a cover value of +; it was
defined as ‘temporary’ if it disappeared before the final
relevé. An ‘increasing’ species was defined as an in-
crease of at least two cover classes during two consecu-
tive relevés and a ‘decreasing’ species as a reduction by
at least two cover classes during three consecutive
relevés. It was considered as ‘disappeared’ if it was still
absent in the final relevé.
¯Â
n
r
(2)
i=1
y
y
Â
i=1
Â
i1
i2
i=1
where R is the highest plant cover of both relevés (%)
and r the lowest. D is not modified for the same total
cover, and may double for very contrasting covers.
Detrended Correspondence Analysis (DCA, from
Wildi & Orlóci 1996) was used to characterize the
species composition of the plots using cover classes.
The default options of the program were used for the
analysis. To illustrate changes in species composition
and abundance in a given plot through time, vectors can
be drawn in ordination space by connecting, in chrono-
logical order, the points that represent the same plot.
The ecological values based on light, soil humidity, pH
and nutrients for every Swiss species (Landolt 1977)
were used to calculate the mean ecological values for
each relevé. Linear regressions of these values against
the main DCA axes were used to explain the ecological
meaning.
To quantify the difference between the relevés and
the original vegetation, we used a modified Whittaker
(1952) percentage similarity, as employed in Legendre
& Legendre (1979):

250
VITTOZ, P. &HAINARD, P.
Results
seeds. Three plots in mesotrophic pasture (8, 10 and 15),
however, needed more time to recover. They reached
normal cover in 1998 (Table 1).
Rooting and recolonization
Pigs started to root as soon as they were released in
early June. Degradation was slow and locally restricted
in June and July, but increased strongly in August.
Initially the pigs were foraging in deep soil close to the
shelters and in the small depression of mesotrophic acid
pasture (almost completely overturned in mid-June). In
mid-September, at the end of the experiment, the eu-
trophic pastures on deep soil were more than 50%
damaged, with some areas completely overturned. A
large part of the mesotrophic pasture on stony soil was
also partly damaged (Table 1, Fig. 2). At this time, pigs
weighed more than 100 kg and were strong enough to
root through the stones. Plot 11, close to the pigsty and
the trough, was little overturned but was damaged by
trampling.
Pasture regeneration started slowly in spring 1997,
with poor germination of sown seeds in the eutrophic
pastures (Troxler unpubl.). However, most of the dam-
aged plots were completely covered by vegetation in
autumn 1997 (Table 1), due to the recovery of surviving
shoots and the later germination of naturally occurring
Species changes
Most of the species densities showed only small
variations during the experiment. However, some species
increased after the disturbance (e.g. Veronica serpylli-
folia ssp. humifusa, Carlina acaulis) and the most com-
mon change was the appearance of new species (e.g.
Cerastium fontanum ssp. vulgare, Poa trivialis, Veronica
chamaedrys). Lolium perenne (sown in the autumn of
1996) is new for Les Amburnex. New species, or those
that increased, were almost exclusively found in over-
turned or damaged plots (Table 2) and were still present
at the end of the experiment (except Poa trivialis which
disappeared in 1998 or 1999).
The completely overturned mesotrophic plots (8, 10
and 15) suffered a large decrease in species, especially
mesotrophic or oligotrophic species. At the end of the
experiment, recovery was not complete and some species
disappeared completely, such as Hieracium pilosella
(plots 8 and 15), Briza media (8), Agrostis capillaris
(10) and Carex montana (15). This was partly compen-
sated by new species (Table 2). However, other species
returned to their original population size before the end
of the monitoring period (e.g. Festuca rubra in plot 15,
Trifolium repens and Agrostis capillaris in plot 8). The
first relevé of plots 7 and 9 are used as a point of
reference for plots 8 and 10. However, we cannot be
sure that all species which disappeared during observa-
tion were present at the beginning of the experiment.
They are normally common species in this type of
vegetation, abundant enough in the reference plot to
indicate that they were likely to have been present in
plots 8 and 10 before pig rooting.
Vegetation changes
The DCA gives an insight into the individual changes
of the plots (Fig. 3). Following the mean ecological
indicator values of the relevés (Landolt 1977), the first
axis is negatively correlated with nutrients (highest r2 =
0.942), humidity (r2 = 0.813) and pH (r2 = 0.676). The
second axis is positively correlated with light conditions
(r2 = 0.260). Vectors show the changes of the plots
during the four years. We observed an interannual fluc-
tuation in the undamaged plots with short and irregular
vectors (2, 7, 9 and 12-14). Other plots had longer
vectors (3, 4, 5, 8, 10 and 15), indicating significant
changes in species composition. Plots 8, 10 and 15
showed a strong move to the left at the beginning, which
corresponds to the colonization of ruderal or eutrophic
Fig. 2. Proportion of overturned pastures at the end of the
experiment (16 September 1996).

- IMPACT OF FREE-RANGE PIGS ON MOUNTAIN PASTURES IN THE SWISS JURA -
251
Table 2. Species changes in the plots: N = new; T = temporary (new but disappeared at the end); >> = increase; << = decrease;
✝ = disappeared. Bold plot numbers are little damaged or undamaged plots during the experiment, and hatched species were sown in
autumn 1996.
Type of pasture
Eutrophic
Mesotrophic
Oligotrophic calcicolous Mesotrophic acid
Plot number
No. of species in May 1996
No. of species in August 2000
1
20
17
2
27
28
3
19
20
4
16
24
5
15
18
6
17
20
7
36
35
8
9
33
33
10
27
11
40
41
12
42
43
13
36
39
14
33
42
15
34
32
21
Eutrophic species
Cerastium fontanum ssp. vulgare
Poa trivialis
N
T
T
T
N
>>
N
N
N
N
N
N
N
N
T
T
N
N
Lolium perenne
N
Festuca pratensis
Poa pratensis
<<
N
N
N
N
N
>>
>>
N
Veronica chamaedrys
Veronica serpyllifolia ssp.humifusa
Dactylis glomerata
Phleum pratense
N
N
N
>>
>>
N
N
N
Plantago major
>>
<<
N
N
>>
Poa supina
<<
Trifolium pratense
Trifolium repens
<< <<
Capsella bursa-pastoris
Cardamine pratensis
Urtica dioica
N
<<
<<
N
Bellis perennis
N
Mesotrophic species
Agrostis capillaris
Festuca rubra s.l.
Plantago media
<<
N
<<
<<
✝
✝
<<
T
<<
T
✝
N
Poa alpina
>>
<<
✝
Cynosurus cristatus
Luzula campestris
Lotus corniculatus
Plantago atrata
<<
<<
N
N
Cirsium acaule
Leontodon hispidus
Prunella vulgaris
Ranunculus carinthiacus
N
N
✝
Oligotrophic species
Briza media
✝
>>
✝
<<
✝
<<
>>
T
Carlina acaulis
>>
Hieracium pilosella
Hieracium lactucella
Nardus stricta
✝
Potentilla crantzii
Medicago lupulina
Carex caryophyllea
Thymus pulegioides
Carex montana
<<
N
<<
N
✝
N
Carex sylvatica
Euphrasia rostkoviana
Veronica officinalis
N
<<
fluctuating from 0.7 to 1. Almost every plot had a
different variation and it is difficult to conclude a gen-
eral trend over the observed period. Clearly the original
community was not achieved and plots 3, 8, 10 and 15
still had high values in autumn 2000. One factor in our
assessment of the responses of plots 8 and 10 may have
been due to the use of the first relevé of plot 7 and 9 as
the point of reference. However, we can hypothesize
that only a small part depends on this difference because
the plots are geographically very close in identical situ-
ations, and plot 15 showed a similar reaction.
species and the reduction of mesotrophic and oligo-
trophic ones (Table 2). A slow trend towards the starting
point can be identified, but it is far from being reached
for plots 8 and 15 at the end of the monitoring. Plot 11
was intermediate, located in a slope next to the pigsty
(Fig. 1). It was not rooted but was damaged by trampling
and fertilized by dung.
Percentage similarity shows similar results (Fig. 4).
The undamaged plots had low values, fluctuating around
0.5, while the overturned plots started with high values
in September 1996, decreasing in spring 1997 and then

252
VITTOZ, P. &HAINARD, P.
Fig. 3. Detrended Correspondence Analysis of the relevés. Vectors drawn on the ordination represent the magnitude and the
direction of the changes in species composition and the abundance from the first relevé (spring 1996, before the experiment) to
the last relevé for each plot (broken and grey lines are to improve legibility).
turned plots on eutrophic pastures and in overturned
plots on mesotrophic pastures. The first group (plots 2,
7, 9, and 12-14) have low percentage of similarities
(Fig. 3) and short arrows in the DCA graph (Fig. 4). No
significant vegetation change occurred during the ex-
periment, but probably a natural interannual fluctuation.
Varying meteorological conditions, favouring different
species may explain these fluctuations (e.g. Austin et al.
1981; van der Maarel & Sykes 1997) but sampling
variations may also play a role.
Plants quickly recolonized the eutrophic pastures
(Table 1, plots 1 and 3-6), although they were almost
completely damaged by rooting (Fig. 2). These pas-
tures, which are fertilized every year, are the most used
by cows. Competitive species, adapted to grazing and
trampling (Grime et al. 1988), quickly recolonized the
bare soil either from surviving shoots, rhizomes or root
fragments (e.g. Poa trivialis, P. pratensis, Taraxacum
officinale, Veronica serpyllifolia ssp. humifusa) or from
the seed bank (e.g. Dactylis glomerata, Trifolium
pratense; Grime et al. 1988; Oberdorfer 1990). The
other new species in these plots are ruderal with large,
persistent seed banks (e.g. Cerastium fontanum ssp.
vulgare, Capsella bursa-pastoris, Plantago major,
Urtica dioica; Grime et al. 1988).
Discussion
Rooting
During this experiment, pigs had a clear preference
for rooting deep soil (Fig. 2). Two reasons may explain
this behaviour (1) the soil is easier to root because it is
softer and wetter (Andresen et al. 2001) or (2) it contains
the highest stock of food. Crocus bulbs, a favourite food
of wild boars (Baettig 1980), are mostly present under
these conditions (Vittoz 1998) and earthworms are more
abundant and larger in fertilized meadows (Baubet et al.
1997). This preference for soft or humid soils was also
observed in pigs in semi-natural environments by Wood-
Gush et al. (1990) and in wild and feral swine by Graves
(1984) in the USA. Although pigs received food in
sufficient quantity they certainly rooted for foraging and
started to damage more difficult soils after they had
exploited most of the easiest and richest soils. The
rooting of stony slopes seems contrary to wild boar
habits (Baettig 1980), and shows that pig density was
probably too high during this experiment. Rain was
quite regular during this summer and wetter soil condi-
tions do not explain this change of behaviour. The
change of temperature does not explain this change
either, because rooting activity is independent of air
temperature (Andresen & Redbo 1999).
Finally, the long arrows and high percentage simi-
larities of the overturned mesotrophic pastures (plots 8,
10 and 15) illustrate the important changes observed in
these plots (Figs. 3 and 4) with an increase of nutrients,
pH and water ecological values (Landolt 1977). This
may correspond to the temporary increase of nitrate
observed in the soil after the destruction of the vegeta-
Species recolonization and vegetation changes
Vegetation composition evolved in three distinguish-
able ways (Figs. 3 and 4): in undamaged plots, in over-

- IMPACT OF FREE-RANGE PIGS ON MOUNTAIN PASTURES IN THE SWISS JURA -
253
pilosella, H. lactucella) grow too slowly and lack a seed
bank to efficiently recolonize disturbed areas (Grime et
al. 1988). The recolonization of disappeared species is
only possible from adjacent areas and will be better if
surviving plants are not too far away (Walker et al.
1997). Rooting in mesotrophic pasture on stony soil
never exceeded a few square meters and recovery should
be possible in a reasonable time. But it will take much
longer for the recovery of the mesotrophic acid pasture,
a large proportion of which was overturned and similar
vegetation was absent in the close vicinity.
Free-range pigs and plant conservation
This experiment took place in a pasture with no
endangered species, but species on the Swiss Red List
(Daphne cneorum, Genista pilosa, Nigritella nigra,
Senecio integrifolius; Landolt 1991) are in the neigh-
bouring oligotrophic pastures on stony soil (Vittoz 1999).
As no influence was observed on this type of pastures,
increasing the numbers of free-range pigs should not
threaten these species, at least as long as the area of
eutrophic pasture on deep soil is large enough for pigs to
root for the whole summer.
Mesotrophic acid pasture on deep soil, the most
damaged vegetation community in this experiment, is
quite rare in the Jura mountains (Vittoz 1998). It was
much more common up to the 19th century, occupying
most of the pasture on deep soil (Béguin 1972). These
pastures changed to eutrophic pastures after the wide-
spread use of tractors and fertilizer, except in inaccessi-
ble areas such as small depressions in the middle of
rocks or trees. It is therefore important to prevent pigs
using these areas, because many plant species are re-
stricted to this vegetation type.
The breeding of free-range pigs is an improvement
for the health of the animals and for the quality of the
meat (Sather et al. 1997; Parriaux & Schaffter unpubl.).
More widespread free-range production is desirable when
pigs are bred during the summer on mountain farms that
have cheese production. Nose-ringing could be a good
way to inhibit rooting and minimize pasture damage,
but it does induce behaviour suggesting reduced welfare
(Horrell et al. 2001). Otherwise, the following precau-
tions are necessary to avoid vegetation damage:
Fig. 4. Variation of the percentage similarity (Legendre &
Legendre 1979) between each relevé and the initial relevé
(spring 1996) of the plot. Note that plot 8 corresponds to the
first relevé of plot 7 and plot 10 corresponds to the first
relevé of plot 9.
•
enclosures should include mostly eutrophic pastures
on deep soil, they should not be located on
mesotrophic pastures on deep soil or any other deli-
cate environments (e.g. wetlands);
tion, due to pig faeces or the decay of uprooted, dead
material in the absence of living plants to take up nutri-
ents (Parriaux & Schaffter unpubl.). However, it is more
probably due to the decrease of the oligotrophic and
mesotrophic species in the plots (Table 2). These spe-
cies (e.g. Briza media, Nardus stricta, Hieracium
•
•
rotation of two or three enclosures (one each sum-
mer), to allow complete vegetation recovery;
low density of pigs (ca. five pigs per ha).

254
VITTOZ, P. &HAINARD, P.
Flora. Veröff. Geobot. Inst. Eidg. Tech. Hochsch. Stift.
The breeding of free-range pigs in accordance with
Rübel Zür. 64: 1-208.
these recommendations is a good compromise between
conservation of rare vegetation types and economical
needs of agriculture. This would ensure a sustainable
use of mountain pastures.
Landolt, E. 1991. Plantes vasculaires menacées en Suisse.
Listes rouges nationale et régionales. OFEFP, Bern, CH.
Lauber, K. & Wagner, G. 2000. Flora Helvetica. Flore illustrée
de la Suisse. Haupt, Bern, CH.
Legendre, L. & Legendre, P. 1979. Ecologie numérique. Tome 2:
La structure des données écologiques. Masson, Paris, FR.
Oberdorfer, E. 1990. Pflanzensoziologische Excursionsflora.
6th ed., Ulmer, Stuttgart, DE.
Sather, A.P., Jones, S.D.M., Schaefer, A.L., Colyn, J. &
Robertson, W.M. 1997. Feedlot performance, carcass com-
position and meat quality of free-range reared pigs. Can. J.
Anim. Sci. 77: 225-232.
Acknowledgements. We are grateful to P. Kissling, H.
Bruelheide, N. Andresen, W. Strahm and two anonymous
reviewers for their helpful comments on the manuscript.
References
Schaffter, N. 1999. Impact de l’exploitation des alpages sur la
qualité des eaux. Thèse EPFL, Lausanne, CH.
Schreiber, K.-F. 1977. Niveaux thermiques de la Suisse.
Département fédéral de justice et police, Bern, CH.
Stolba, A. & Wood-Gush, D.G.M. 1981. The assessment of
behavioural needs of pigs under free-range and confined
conditions. Appl. Anim. Ethol. 7: 388-389.
Thomet, P. 1981. Die Pflanzengesellschaften der Schweizer
Juraweiden und ihre Beziehung zur Bewirtschaftungs-
intensität. Tätichkeitsber. Naturf. Ges. Baselland 31: 243-
368.
van der Maarel, E. & Sykes, M.T. 1997. Rates of small-scale
species mobility in alvar limestone grassland. J. Veg. Sci.
8: 199-208.
Vittoz, P. 1998. Flore et végétation du Parc jurassien vaudois:
typologie, écologie et dynamique des milieux. Thèse,
Université de Lausanne, CH.
Andresen, N. & Redbo, I. 1999. Foraging behaviour of grow-
ing pigs on grassland in relation to stocking rate and feed
crude protein level. Appl. Anim. Behav. Sci. 62: 183-197.
Andresen, N., Ciszuk, P. & Ohlander, L. 2001. Pigs on grass-
land: animal growth rate, tillage work and effects in the
following winter wheat crop. Biol. Agric. Hortic. 18: 327-
343.
Austin, M.P., Williams, O.B. & Belbin, L. 1981. Grassland
dynamics under sheep grazing in an Australian Mediterra-
nean type climate. Vegetatio 47: 201-211.
Baettig, M. 1980. Contribution à la biologie et écologie du
sanglier (Sus scrofa L.) dans le canton de Vaud. Conser-
vation de la faune, Section protection de la nature et des
sites de Vaud, CH.
Baubet, E., Touzeau, C. & Brandt, S. 1997. Les lombriciens
dans le régime alimentaire du sanglier (Sus scrofa) en
montagne. Mammalia 61: 371-383.
Vittoz, P. 1999. La flore du Parc jurassien vaudois et de ses
environs. Bull. Cercle Vaud. Bot. 28: 61-98.
Walker, B.H., Langridge, J.L. & Mcfariane, F. 1997. Resil-
ience of an Australian savanna grassland to selective and
non-selective perturbations. Aust. J. Ecol. 22: 125-135.
Watson, C.A., Anssems, E., Kuhne, B., Scholzel, Y. & Edwards,
S.A. 1998. Assessing the nitrogen pollution risk from
outdoor pig systems. In: Petchey, A., d’Arcy, B. & Frost,
A. (eds.) Diffuse pollution and agriculture II, pp. 230-235.
SAC, Aberdeen, UK.
Whittaker, R.H. 1952. A study of summer foliage insect
communities in the Great Smoky Mountains.Ecol. Monogr.
22: 1-44.
Wildi, O. & Orlóci, L. 1996. Numerical exploration of com-
munity patterns. A guide to the use of MULVA-5. SPB,
Amsterdam, NL.
Béguin, C. 1972. Contribution à l’étude phytosociologique et
écologique du Haut Jura. Beitr. Geobot. Landesaufn.
Schweiz 54: 1-190.
Bloesch, B. & Calame, F. 1994. L’air du temps. In: Capt, G.,
Jean-Petit-Matile, O. & Reymond, J. (éd.)Le Parc jurassien
vaudois, pp. 35-45. 24 Heures, Lausanne, CH.
Braun-Blanquet, J. 1964. Pflanzensoziologie. Grundzüge der
Vegetationskunde. 3rd ed., Springer-Verlag, Wien, AT.
Braund, J.P., Edwards, S.A., Riddoch, I. & Buckner, L.J.
1998. Modification of foraging behaviour and pasture
damage by dietary manipulation in outdoor sows. Appl.
Anim. Behav. Sci. 56: 173-186.
Ellenberg, H., Weber, H.E., Düll, R., Wirth, V., Werner, W. &
Paulißen, D. 1991. Zeigerwerte von Pflanzen in Mittel-
europa. Scripta Geobot. 18: 1-248.
Graves, H.B. 1984. Behavior and ecology of wild and feral
swine (Sus scrofa). J. Anim. Sci. 58: 482-492.
Grime, J.P., Hodgson, J.G. & Hunt, R. 1988. Comparative
plant ecology. A functional approach to common British
species. Unwin Hyman, London, UK.
Horrell, R.I., A’Ness, P.J., Edwards, S.A. & Eddison, J.C.
2001. The use of nose-rings in pigs: consequences for
rooting, other functional activities, and welfare. Anim.
Welf. 10: 3-22.
Wood-Gush, D.G.M., Jensen, P. & Algers, B. 1990. Behav-
iour of pigs in a novel semi-natural environment. Biol.
Behav. 15: 62-73.
Received 17 April 2001;
Revision received 22 March 2002;
Accepted 10 May 2002.
Coordinating Editor: H. Bruelheide.
Jost, M. 1990. Le porc d’alpage. Rev. Suisse Agric. 22: 242-
244.
Landolt, E. 1977. Ökologische Zeigerwerte zur Schweizer