J Insect Conserv (2012) 16:921–930
DOI 10.1007/s10841-012-9479-y
ORIGINAL PAPER
Resource use of specialist butterflies in agricultural landscapes:
conservation lessons from the butterfly Phengaris (Maculinea)
nausithous
Sergej H. D. R. Jansen • Milena Holmgren
Frank van Langevelde • Irma Wynhoff
•
Received: 4 October 2011 / Accepted: 9 March 2012 / Published online: 22 March 2012
Ó Springer Science+Business Media B.V. 2012
Abstract Most of the European grassland butterfly species
are dependent on species rich grasslands shaped by low
intensity farming. Conservation of these specialist species
in agricultural landscapes relies on knowledge of their
essential resources and the spatial distribution of these
resources. In The Netherlands, the dusky large blue Phengaris
(Maculinea) nausithous butterflies were extinct until their
reintroduction in 1990. In addition, a spontaneous recolonization of road verges in an agricultural landscape occurred in
2001 in the southern part of The Netherlands. We analyzed
the use of the essential resources, both host plants and host
ants, of the latter population during the summers of 2003 and
2005. First we tested whether the distribution of the butterflies
during several years could be explained by both the presence
of host plants as well as host ants, as we expected that the
resource that limits the distribution of this species can differ
between locations and over time. We found that oviposition
site selection was related to the most abundant resource.
While in 2003, site selection was best explained by the
presence of the host ant Myrmica scabrinodis, in 2005 it was
more strongly related to flowerhead availability of the host
plant. We secondly compared the net displacement of individuals between the road verge population and the reintroduced population in the Moerputten meadows, since we
expected that movement of individuals depends on the
S. H. D. R. Jansen I. Wynhoff (&)
Dutch Butterfly Conservation, PO Box 506, 6700 AM
Wageningen, The Netherlands
e-mail: Irma.Wynhoff@vlinderstichting.nl
S. H. D. R. Jansen M. Holmgren F. van Langevelde
Resource Ecology Group, Wageningen University, PO Box 47,
6700 AA Wageningen, The Netherlands
structure of their habitat. On the road verges, butterflies
moved significantly shorter distances than on meadows,
which limits the butterflies in finding their essential resources.
Finally we analyzed the availability of the two essential
resources in the surroundings of the road verge population.
Given the short net displacement distances and the adverse
landscape features for long-distance dispersal, this landscape
analysis suggests that the Phengaris population at the Posterholt site is trapped on the recently recolonized road verges.
These results highlight the importance of assessing the
availability of essential resources across different years and
locations relative to the movement of the butterflies, and the
necessity to careful manage these resources for the conservation of specialist species in agricultural landscapes, such as
this butterfly species.
Keywords Phengaris (Maculinea) nausithous
Myrmica Habitat management
Oviposition site selection Host specificity
Introduction
Most of the European grassland butterfly species are
dependent on open grasslands rich in plant species, which
are shaped by years of low intensity farming (Van Swaay
2002). These grasslands were common at the outskirts of
almost all villages and cities, but nowadays in the Northwest of Europe, they can only be found in nature reserves
and along linearly shaped landscape elements, such as road
verges, ditches and stream banks. This habitat reduction
and fragmentation have led to a drastic decline in butterfly
diversity, particularly among specialist species (Aviron
et al. 2007; Blair and Launer 1997; Poschlod et al. 2005;
Van Swaay 1990; Van Swaay et al. 2006, Wenzel et al.
123
922
2006). The remaining patches with vegetation resembling
the former extensively managed grasslands are usually
inhabited by low numbers of generalist butterfly species
while specialist species are rare (Duelli and Obrist 2003;
Kivinen et al. 2008).
One of the few specialist butterfly species that can
survive in small nature reserves and on linearly shaped
landscape elements is the dusky large blue Phengaris
(Maculinea) nausithous (European Union Habitats Directive App. II and IV, IUCN Red List of European Butterflies, Dutch National Red List; Van Swaay et al. 2011). In
the southern parts of The Netherlands, this butterfly species
used to be quite common on hay meadows and pastures. In
the 1960s, population numbers declined sharply to be
finally declared locally extinct in 1976 when the last population disappeared (Bos et al. 2006; Boeren et al. 2011).
The butterflies were reintroduced in the center of the
Netherlands in 1990 in the nature reserve Moerputten
(Fig. 1, Wynhoff 2001; Wynhoff et al. 2008), originating
from a Polish source population (Wynhoff 1998b). Independent from the reintroduction and rather unexpectedly, in
2001, a few individuals from small remnant stream bank
populations across the border with Germany spontaneously
recolonized several road verges close to the village of
Posterholt in the south of the Netherlands, which is at a
distance of about 90 km from the reintroduction site
(Fig. 1; Wynhoff et al. 2005).
The conservation of rare species like this one relies
strongly on a deep understanding of their essential resources
and the spatial distribution of these resources (Dennis et al.
2006; Vanreusel and Van Dyck 2007; Van Langevelde and
Wynhoff 2009). P. nausithous butterflies have a complex
Fig. 1 Location of four patches
with the host plant S. officinalis
for the study on P. nausithous
next to the village of Posterholt
(The Netherlands). On the small
map the location of Moerputten
and Posterholt in The
Netherlands. Black lines main
roads, white lines minor roads,
black areas houses, gray areas
forest, white areas agricultural
fields
123
J Insect Conserv (2012) 16:921–930
life cycle in which they engage in multitrophic interactions
with two host species, the plant Sanguisorba officinalis on
which butterflies deposit their eggs and young larvae feed,
and specific Myrmica ant species in which nests the latest
instar caterpillars spend 10 months feeding on grubs and
overwintering (Thomas 1984). Although the caterpillars of
P. nausithous are adopted by all Myrmica ant species, the
survival rate in the nests of different ant species varies
(Thomas et al. 1989; Munguira and Martı́n 1999; Stankiewicz and Sielezniew 2002; Tartally and Varga 2005;
Tartally et al. 2008). It has been suggested that the preferred
ant species for P. nausithous is Myrmica rubra, but in the
absence of this ant species the caterpillars are found in M.
scabrinodis nests (Pech et al. 2007; Tartally et al. 2008;
Witek et al. 2008).
The evidence of selection of oviposition sites by this
butterfly species based on plant and ant resources is controversial. There is some evidence indicating that females
select sites based on vegetation characteristics regardless of
the presence of host ant species (Thomas and Elmes 2001;
Musche et al. 2006, Fürst and Nash 2010), whereas other
studies suggest that Phengaris butterflies select for both
resources (Van Dyck et al. 2000; Wynhoff et al. 2008; Van
Langevelde and Wynhoff 2009; Van Dyck and Regniers
2010; Patricelli et al. 2011). It has been shown that oviposition site selection by this species depends on butterfly
density: visits of adult butterflies to plots with the host
plant but without the host ants especially occurred in years
with high butterfly densities (Wynhoff et al. 2008). Due to
differences in climate, management and the possible negative impact of the caterpillars on the nests of their host
ants, densities of butterflies and host ant nests can vary
J Insect Conserv (2012) 16:921–930
923
between years (Nowicki et al. 2009). The resource that
limits oviposition site selection of this butterfly could
therefore be different between areas and over time.
The extremely specialized interaction with both host
plants and host ants imposes challenging restrictions for the
survival of P. nausithous populations. Since the butterfly
population has been able to survive in fragmented agricultural landscapes, such as those found in the area surrounding Posterholt, where both host plants and host ants
are scarce and scatteredly distributed, we test whether the
occurrence of butterflies during several years can be
explained by both the presence of host plants as well as
host ants. We compare the use of these essential resources
of the spontaneously recolonized population with other
studies on oviposition site selection of P. nausithous.
When movement along linear elements is limited to
short distances, we also expect that individuals of
expanding populations on spatially restricted habitat, like
road verges, move differently, while searching for the
essential resources as compared to individuals of populations on meadows (Van Langevelde and Grashof-Bokdam
2011; Hovestadt and Nowicki 2008). We compare net
displacement data measured in Posterholt with the net
displacement data of the reintroduced Moerputten population. Finally we analyze the surroundings of Posterholt to
predict whether the spontaneously recolonized population
could find their essential resources given the spatial distribution of these resources in the surroundings.
cubes as bait to attract ants. Sugar baits were placed on a
concave glass covered by black plastic at the foot of a
host plant during the morning hours and checked after
2 h. We kept the attracted ants in alcohol for later identification in the laboratory. Baits without ants were left in
the field and were checked regularly every 2 h until the
end of the day.
It appeared that the butterfly flight area was limited to
road verges and banks of drainage canals that are covered
with their host plant. We identified four patches with host
plants (Fig. 1), 10–20 m wide and between 100 and 300 m
long including the recolonized site and located within the
expected dispersal range of the species (Van Langevelde
and Wynhoff 2009). We sampled the small population of
P. nausithous butterflies during their flight period between
July and August in 2003 and 2005 using the MRR method.
In 2003, weather conditions were comparatively warm for
this region (mean temperature in July 18.8 °C, in August
19.3 °C), dry (July 57 mm, August 22 mm rainfall) and
shiny (240 and 241 h of sunshine in July and August
respectively). In contrast, the weather during the flight
period of 2005 was cooler (mean temperature in July
17.7 °C, in August 16.2 °C), rainy (116 and 82 mm rainfall in July and August respectively) and with lower irradiance (July 163 h, August 193 h). These conditions depart
from the long term average conditions in The Netherlands
(period from 1971 to 2000) in July (17.4 °C, 70 mm
rainfall, 201 sunshine hours) and August (17.2 °C, 62 mm
rainfall, 198 h of sunshine).
Methods and materials
Flowerhead availability and ant species presence
Study area
Within the four selected host plant patches, we randomly
distributed 1-m2 plots (80 plots in 2003, with patch 1: 24
plots; patch 2: 22 plots; patch 3: 26 plots; patch 4: 8 plots;
77 plots in 2005, with patch 1: 35 plots; patch 2: 21 plots;
patch 3: 14 plots, patch 4: 7 plots; Fig. 1). At the beginning
of the flight period, we counted the total number of flowerheads present at the shoots of all host plants in each plot.
We used the number of flowerheads as a measure for
resource availability since the flowerheads provide a place
for oviposition and are the main source of nectar for this
butterfly species. We sampled the ants directly after the
flight period of the butterflies when most caterpillars are
still on the flowerheads. This reduces the effect of ant nest
disturbance on caterpillar survival. The ants were baited
with sugar cubes in the center of each plot in the same way
as described above. In the analyses, we used all plots with
Myrmica species present, but also the plots with other ant
species and the plots without ants. In these latter categories
caterpillars are expected to die because they are foraged
upon or because they will not be found and adopted by the
ants to survive the winter.
The study area is located near the village of Posterholt (51
080 0000 N, 6 020 0000 E) within the valley of the river Roer,
in the Limburg province of The Netherlands (Fig. 1). This
area is at the margins of the distribution range of
P. nausithous in Northwestern Europe (Wynhoff 1998a).
The typical agricultural landscape is dominated by fields
with fodder crops (mainly wheat and corn), pastures and
small human settlements. After the new population of the
butterflies at Posterholt was discovered in 2001, in the
next 2 years we mapped all host plants, S. officinalis, in
the nature reserves, road verges and stream borders up to
a distance of 6,655 m from the recolonized site. This
investigation included all sites where the butterfly species
had been observed until its extinction in The Netherlands
(Bos et al. 2006). In the consecutive years the population
was monitored by transect counts and several markrelease-recapture (MRR) studies. We determined the
availability of host ants at 1,023 plots of 1-m2 around a
single host plant or a group of host plants by using sugar
123
924
Butterflies and oviposition
The MRR study allowed us to estimate population sizes
during the flight season of both 2003 and 2005. On the four
patches (Fig. 1), the butterflies were captured with a net
and then marked with an individual code number using a
fine-tipped Lumocolor overhead pen with permanent ink.
After marking the butterflies were released immediately.
For each butterfly captured, we measured and noted the
GPS-coordinates. Sampling was conducted on 22 days in
2003 and 24 days in 2005 under appropriate weather
conditions. In 2003, only during 4 single days in the flight
period no butterflies were captured and marked, while in
2005 we were able to mark them on each day. An oviposition was recorded when a female not only touched the
flowerhead with her abdominal tip but also squeezed the tip
between the flowerbud and then froze to release the egg. In
addition, in 2005, we collected 10–20 flowerheads in each
plot and searched for egg remnants as an indirect evidence
of oviposition. The flowerheads were collected after the
caterpillars had left their host plant about 4 weeks after the
end of the flight period. We estimated the minimal population size by Minimal Number Alive (MNA; Amler et al.
1999) and the total population size according to Jolly-Seber
(Begon 1979).
Butterfly net displacement
We used the GPS coordinates from the captures and
recaptures to calculate the net displacement of a butterfly
between captures. The data were analyzed to find differences of net displacement between years, gender and
number of days between recaptures. In this analysis the
data of the reintroduced population of P. nausithous in the
nature reserve Moerputten in 1990 were included. This
nature reserve consists of a central lake surrounded by tall
beds of Phragmites, Typha and tall Carex species. Around
these, moist forests are present nowadays where wet
meadows used to occur when the area was still in agricultural use. On the outer borders of the nature reserve,
partially within the forest, different types of grasslands are
found, of which the hay meadows with a high abundance of
S. officinalis are most important as habitat of Phengaris
butterflies. For a detailed description of the Moerputten site
see Wynhoff (1998b) and Wynhoff et al. (2008).
Statistical analysis
We analyzed the data on resource use of the 2 years separately, because the weather conditions during the flight
period and the butterfly population density were very different between the years. We checked the normality of the
data by testing the residuals for using either a parametric or
123
J Insect Conserv (2012) 16:921–930
a non-parametric test. An Anova test was used to assess for
differences in flowerhead availability between the plots
occupied by different ant species, namely M. rubra,
M. scabrinodis, M. ruginodis, a mix of Myrmica species,
non-Myrmica species and ant free plots. A t-test or Mann–
Whitney test was used to test for differences in flowerhead
availability between years and between plots where butterflies and ovipositions were observed versus the plots
where butterflies and ovipositions were not observed. The
Kruskal–Wallis test was used to test whether the number of
observed butterflies differed between plots occupied by
these categories of ant species. We used logistic regressions to model oviposition site selection (i.e. probability of
butterfly presence) and oviposition behavior (i.e. probability of oviposition presence) as a function of resource
availability expressed as the number of flowerheads and the
presence or absence of the different ant species for each
year separately. The mean net displacement of the butterflies was analyzed using a General Linear Model (GLM)
with a pairwise comparison according to the Sidak post hoc
test (Field 2005). In this analysis the number of days in
between capture and recapture was taken as covariable.
Results
In 2003, 38 individuals (sex ratio males:females = 0.41) of
P. nausithous were marked, of which 41.0 % were recaptured. In 2005 the population was larger. We captured 89
butterflies (sex ratio males:females = 0.71) and achieved a
recapture of 68.5 %. The estimated population size of
P. nausithous in 2005 was approximately 160 butterflies,
almost three times larger than in 2003 with 54 individuals.
In 2003, the total number of butterflies during the whole
flight period was not much higher than the minimal population size of 48 butterflies. This indicates that not many
butterflies have escaped our attention. In 2005, even though
more than half of the marked individuals have been
recaptured at least once, the estimated total population size
was higher than the minimal population size of 100. The
within-season daily population size dynamics is presented
in Fig. 2. In both years the butterflies were observed only
in patches 1 and 2 (Fig. 1).
Flowerhead and host ant availability
We found three Myrmica ant species in the 1-m2 plots:
M. rubra, M. scabrinodis and M. ruginodis (Table 1). In
2005, there were fewer plots with ant species present than
in 2003. This was especially the case with M. rubra. In
general, plots with M. ruginodis or with a mix of Myrmica
species appeared to be quite rare. The most common nonMyrmica ant species was Lasius niger.
J Insect Conserv (2012) 16:921–930
925
100
Number of butterflies
80
60
40
20
0
15-7
20-7
25-7
30-7
4-8
date
Fig. 2 Daily population size of P. nausithous in the flight period of
2003 (white bars) and 2005 (gray bars) at the Posterholt site in The
Netherlands. The daily population size was estimated according to
Jolly Seber. Error bars represent the standard errors
Table 1 Ant species distribution at the Posterholt location in 2003
and 2005 (N: Number of plots and % of total)
Ant species
2003
N
2005
%
N
%
No-ants
14
17.5
25
32.4
M. rubra
18
22.5
11
14.3
M. scabrinodis
26.0
20
25.0
20
M. ruginodis
8
10.0
2
2.6
Non-Myrmica
17
21.3
19
24.7
Mix-Myrmica
3
3.8
–
–
80
100.0
77
100.0
Total
The mean number of flowerheads per plot in 2005 (mean
number of flowerheads ± SE: 94.54 ± 6.16) was almost
twice as high as in 2003 (mean number of flowerheads ± SE: 44.70 ± 2.68, t = -7.423, p \ 0.001). In
2003 the flowerhead availability was equally distributed
between the different ant species groups (Anova:
F5,69 = 1.219, p = 0.310, Fig. 3). In 2005, there was a significantly lower number of flowerheads in the plots with
M. rubra and a significantly higher number of flowerheads in
the plots with the ant M. ruginodis, whereas the other groups did
not differ in number of flowerheads (Anova: F4,71 = 1.219,
p = 0.008, significant differences between groups according to
the Sidak post hoc test are indicated by letters).
Oviposition site selection
In 2003 the butterflies were observed in 22 plots (27.5 %,
n = 80). The mean number of flowerheads in the occupied
plots was not significantly higher when compared to the
Fig. 3 The distribution of flowerheads of S. officinalis over the
different ant species in 2003 and 2005 at the Posterholt location. Data
were collected from 80 plots in 2003 and 77 plots in 2005 within four
patches (Fig. 1). Error bars represent the standard errors of the mean.
Letters indicate significant differences
plots where no butterflies were seen (t test on log-transformed number of flowerheads: t = -1.295, df = 73,
p = 0.199). The total number of butterflies counted on
plots with M. scabrinodis was significantly higher when
compared to the plots with other ant species (Kruskal–
Wallis test: v2number of butterflies = 11.175, p = 0.048,
n = 80). In a logistic regression analysis the oviposition
site selection, expressed as the probability of observing a
123
926
J Insect Conserv (2012) 16:921–930
Table 2 Results of the logistic regression analyses of the presence and absence of P. nausithous in 2003 and 2005
Year
2003
2005
Number of cases
80
77
% predicted correctly
72.5
Variable
bi
71.1
SE
Wald
Sig.
bi
SE
Wald
Sig.
-1.117
0.468
5.682
0.017
0.012
0.004
8.115
0.004
Constant
-1.838
0.437
17.647
\0.001
Ant presence M. scabrinodis
1.162
0.557
4.349
0.037
Ant presence Non Myrmica species
1.157
0.539
4.604
0.032
Number of flowerheads
It was tested whether the ant species and the number of flowerheads had an effect on the probability of observing butterflies. Only factors with
significant effect are presented. The regression coefficients (b), their standard errors (SE), the Wald-statistics and their significance levels are
given. The probability of observing a butterfly is given by eregression equation/(1 ? eregression equation)
butterfly in a plot, showed to be positively related to the
presence of the ant M. scabrinodis and the non-Myrmica
ants. These factors correctly predict 72.5 % of the observations (Table 2).
In 2005 the butterflies occupied 40 plots (51.9 %,
n = 77). This year, butterflies were present in plots with a
significantly higher flowerhead availability compared to
the plots where butterflies were absent (Mann–Whitney
test: Z = -3.123, p = 0.002, n = 77). Interestingly, we
found no significant influence of the ant species on the
presence and absence of the butterflies in our plots during
this second year. The total number of butterflies showed to
be equally distributed over the different ant species groups
(Kruskal–Wallis test: v2number of butterflies = 8.911,
p = 0.063, n = 77). The logistic regression analysis of the
oviposition site selection in 2005 showed that the habitat
selection was only positively related to the number of
available flowerheads. This factor predicts 71.1 % of the
observations correctly (Table 2).
Mean net displacement
In 2005, the distance between butterfly captures was larger,
though not significantly different from 2003, whereas the
reintroduced butterflies on the meadows of the Moerputten
flew much longer distances (GLM, F2,264 = 35.931,
p \ 0.001, Fig. 4). Male and female butterflies did not
differ in the mean net displacement between captures
(F1,264 = 0.206, p = 0.65). We found no interaction
between sex and site (F2,264 = 0.164, p = 0.849). The
distance between captures increased with the number of
days in between (F1,264 = 13.158, p \ 0.001).
Oviposition behavior
In 2003, oviposition was observed on 5 plots (6.3 %,
n = 80) and in 2005 on 14 plots (18.2 %, n = 77). In
both years there was no significant difference in the
number of flowerheads between the plots where oviposition was observed and the plots with no oviposition (for
2003: t test on log transformed number of flowerheads,
t = 0.659, df = 73, p = 0.516; for 2005: Mann–Whitney
test: Z = -1.219, p = 0.223, n = 77). The logistic
regression analyses for the years 2003 and 2005 also
showed that the oviposition behavior, expressed as the
probability of observing oviposition in a plot, was not
related to the flowerhead availability or the ant species
distribution. These results were confirmed when the
regressions for the presence and absence of oviposition
were run only on the plots where butterflies have been
present.
123
Fig. 4 Mean net displacement of marked individuals of P. nausithous. Roer 2003 and Roer 2005: the Posterholt population in 2003
and 2005 resp., Moerputten 1990: the reintroduced population in the
nature reserve Moerputten in 1990. Error bars represent the standard
errors of the mean. Letters indicate significant differences
J Insect Conserv (2012) 16:921–930
In 2005, only 10 % of the marked butterflies captured in
one patch were recaptured in another patch. The mean net
displacement distance was 75 ± 6 m and the maximum net
displacement was as far as 375 m (Jansen et al. 2006). In
2003 the mean net displacement was even less: 43 ± 8 m.
The two smaller patches at the Posterholt location where no
P. nausithous butterflies were observed were within the
mean net displacement range measured in the reintroduced
Moerputten population (226 ± 16 m) and therefore within
the measured net displacement range of the species. Mean
net displacement values obtained by MRR studies underestimate dispersal distance and dispersal rate due to the
method itself (Hovestadt et al. 2011). Even when taking
that into consideration, the estimated values for the Posterholt population are low (Nowicki et al. 2005c).
Resource availability at landscape scale
We found ant species in only 45 % of the investigated plots
(n = 1,023) with (groups of) host plants in the landscape
around Posterholt. 67 % of these plots were occupied by a
Myrmica species. The most abundant host ant species in
these plots was M. rubra (35 %), while M. scabrinodis
(19 %) and M. ruginodis (11 %) were quite rare. At the
patches where we found P. nausithous, both density and
diversity of Myrmica ants was high while not more than
35 % of the plots were occupied by the potential predator
L. niger. At locations without P. nausithous, less ant species were found. Sometimes L. niger was the dominating
species, but we also found that M. rubra or M. ruginodis
comprised most of the local ant fauna. Figure 5 shows that
with increasing distance from the population at Posterholt,
Fig. 5 Sites with S. officinalis at the distance to the Posterholt
population of P. nausithous. Highest line Plot with only S. officinalis,
middle line plot with S. officinalis and ants of any species, lowest line
plot with S. officinalis and Myrmica ants
927
only a limited number of plots contained the two essential
resources for the butterflies.
Discussion
Specialist butterflies, such as P. nausithous, have to make
choices on where to go for nectar and on which host plants
to deposit their eggs in order to ensure high survival of
their offspring, preferably in nests of their host ant species.
The availability and the distribution of these essential
resources are expected to influence the butterfly behavior.
When both resources are abundant butterflies are more
likely to be found in plots with both host plants and ants,
whereas when a specific resource is limited and competition for it is high, butterflies might be forced to stay also in
plots with only one resource (Wynhoff et al. 2008). We
found that the flowerhead availability and the presence of
the ant M. scabrinodis were both important factors in
determining the distribution of P. nausithous butterflies.
The availability of these resources differed between the
2 years of the study and this seemed to have affected the
oviposition site selection made by the butterflies; 2003 was
a dry year with relatively low host plant flowerhead
availability and high ant presence, while 2005 had relatively high flowerhead availability and low ant presence. In
addition, in 2005 the population size of P. nausithous with
approximately 160 butterflies was almost three times as
high as in 2003. In our analysis, oviposition site selection
defined in terms of the probability that a butterfly was
present, seems to be best explained by the most abundant
resource: in 2003 site selection was explained by the
presence of the ant M. scabrinodis, while in 2005 site
selection was positively influenced by the number of host
plant flowerheads. These site selection changes are found
only in relation to butterfly distribution since during the
2 years we found no influence of the two resources on the
oviposition behavior which may have been prevented by
the low number of observations.
Most Phengaris species show geographical differences
in host ant specificity (Als et al. 2002, 2004; Steiner et al.
2003; Stankiewicz et al. 2005). In contrast P. nausithous is
dependent on only M. rubra in most of Europe, with some
populations at the edge of the butterfly distribution area
using M. scabrinodis as a single host (Munguira and Martı́n
1999; Tartally et al. 2008; Witek et al. 2008). While for the
reintroduced population in the centre of The Netherlands a
relation with M. rubra could be shown (Wynhoff et al.
2008, Van Langevelde and Wynhoff 2009), we were not
able to confirm this for the Posterholt population. It is
possible that this is due to the low occurrence of the typical
host ant species at the study site. The population might
survive in the nests of the second best host ant species,
123
928
namely M. scabrinodis, with a reduced survival probability
for the caterpillars (Thomas et al. 1989). Unfortunately, we
do not have information about the local host ant specificity
of the nearby German populations. Populations in the
Westerwald region (appr. 100 km from Posterholt) are
more numerous on sites with more M. rubra ants captured
on baits (Dierks and Fischer 2009), indicating that this is
most likely the most important host ant. It is only possible
to collect data about this by excavating the ant nests in the
beginning of the flight period in search for pupae, which is
highly undesirable in our small, spontaneously recolonized
population.
The Posterholt population is rather small in comparison
with other populations of this butterfly species (Binzenhöfer and Settele 2000; Nowicki et al. 2005a). In the
published literature only a few sites are known which host
populations smaller than 150 butterflies during the entire
flight period, but these occur on meadows and not on road
verges (Nowicki et al. 2005b). Usually, populations are
larger. Nowicki et al. (2005a) describes a metapopulation
of the species consisting of more than 50 local patches
where all together more than 50.000 butterflies thrive.
Although the Posterholt population might be small, it is the
largest population within a circle of at least 50 km. During
several field visits to the neighboring populations in Germany, we never found more than 7 butterflies on the same
location. Due to the lack of data it is not possible to estimate a reliable population size, but is obvious that the
number of butterflies in the German populations is below
that of the Dutch population (Boeren et al. 2011). Clearly,
the Posterholt population is not a sink population connected to another site with a high number of butterflies that
might feed this small population with dispersing individuals. Since the nearest populations are at distances of 5,500
and 6,300 m, they are within the dispersal range of the
species. However, for a regular exchange of individuals the
distances are probably too large (Van Langevelde and
Wynhoff 2009; Hovestadt et al. 2011).
The distribution of the essential resources for these
butterflies can only be found along the linearly shaped
landscape elements, mainly road verges. We found that the
butterflies of the Posterholt road verge population move
over only very short distances compared to individuals of
the reintroduced population in the meadows of the Moerputten nature reserve, based on net displacement distances
derived from our MRR studies. Hovestadt and Nowicki
(2008) also found that similarly obtained net displacement
distances for the butterfly species Phengaris teleius are site
specific, most likely because of differences in habitat patch
heterogeneity. In contrast, the average distances of movements were several 100 m in populations that were at least
ten times larger than the Posterholt road verge population
and occurred in large extended habitats (Settele 1998;
123
J Insect Conserv (2012) 16:921–930
Binzenhöfer and Settele 2000; Stettmer et al. 2001). If the
butterflies from the studied patches 1 and 2 would move as
far as those in the Moerputten nature reserve, they would
have been able to also colonize patches 3 and 4, where both
resources are available. These findings support the
hypothesis that populations in linearly shaped habitat suffer
from low dispersal rates which might result in small local
populations with high extinction risk (Van Langevelde and
Grashof-Bokdam 2011). The Posterholt population did
indeed not colonize more patches at close distance, such as
patches 3 and 4. Nowicki et al. (2005a) also reports such
very high site fidelity. This high site fidelity may cause
serious problems in the conservation of the species, especially when its habitat is highly fragmented and restricted
to linearly shaped landscape elements (Van Langevelde
and Wynhoff 2009; Van Langevelde and Grashof-Bokdam
2011). Our results suggest that the road verge population
might be limited in their movements due to the shape of
their habitat. If so, mobility is a landscape-specific trait
rather than being defined only by species characteristics.
In the surroundings of the Posterholt population, it
appeared that dispersing butterflies have to bridge two
large gaps without any of the resources in between suitable plots (Fig. 5). The meadows where the historical
populations of P. nausithous used to occur have lost both
their species rich vegetation and their diverse ant communities. This makes them unsuitable as habitat for the
butterflies, even though the sites are now protected within
the borders of nature reserves. Given the short net displacement distances and the adverse landscape features for
long-distance dispersal, this landscape analysis suggests
that the Phengaris population at the Posterholt site is
trapped on the recently recolonized road verges. The historical meadow habitat is currently completely disappeared and even within nature reserves no sites were
found for the species to colonize. Future population
growth of P. nausithous in the Posterholt area could be
achieved by increasing nest densities of the different host
ant species (Anton et al. 2008) as this seems to be a
limiting resource in some years. In addition, potential
habitat could be expanded by promoting areas with host
plants. Until severe management activities to reestablish
habitat with a high density of host plants and host ants
have been taken place, no other locations than road side
verges and stream side vegetation will be available for the
butterflies. By adjusting the management and mowing
regime of the road verges and drainage canals where this
butterfly occurs, and with the help of agri-environmental
schemes, in the surrounding borders of agricultural fields,
both host plant densities as well as nest densities of the
different host ant species could be increased and refuge
areas for the ants can be created (Johst et al. 2006; Grill
et al. 2008; Nowicki et al. 2007).
J Insect Conserv (2012) 16:921–930
Nowadays, linearly shaped elements in the agricultural
landscape are frequently mown and have a rather short
vegetation height. Less frequent management schemes
allowing the growth of taller vegetation provide butterflies
and ants with shelter and food resources (Wynhoff et al.
2011). They can facilitate the dispersal of specialist butterflies and other species along these landscape elements
(Söderström and Hedblom 2007). Due to their low edgearea ratio, however, linearly shaped landscape elements
have a high sensitivity to disturbances due to land use.
Specialist species such as these butterflies occurring only
on such landscape elements, like road verges, are especially vulnerable for these disturbances. Nonetheless, if a
network of a high density of interconnected landscape
elements with suitable habitat is available, a high persistence for P. nausithous populations can be achieved.
Nearby nature reserves should also be included in this
network. Eventually the expansion of this patchy habitat
network will connect the Dutch and German P. nausithous
populations, creating a metapopulation throughout the
whole Roer valley.
Acknowledgments We thank Dutch Butterfly Conservation and
Wageningen University for facilitating this work. Jan Boeren, Jacqeline King and Mark Grutters assisted with the fieldwork. Rebekka
Eckelboom introduced us to the German sites. Nicoliene Peet identified the ants. M.H. thanks the Dutch NWO Meervoud programme
836.05.021.
References
Als TD, Nash DR, Boomsma JJ (2002) Geographical variation in
host-ant specificity of the parasitic butterfly Maculinea alcon in
Denmark. Ecol Entomol 27:403–414
Als TD, Vila R, Kandul NP, Nash DR, Yen SH, Mignault AA,
Boomsma JJ, Pierce NE (2004) The evolution of alternative
parasitic life histories in large blue butterflies. Nature 432:
386–390
Amler K, Bahl A, Henle K, Kaule G, Poschlod P, Settele J (1999)
Populationsbiologie in der Naturschutzpraxis. Verlag Eugen
Ulmer GmbH & Co, Stuttgart
Anton C, Musche M, Hula V, Settele J (2008) Myrmica host-ants
limit the density of the ant-predatory large blue Maculinea
nausithous. J Insect Conserv 12:511–517
Aviron S, Kindlmann P, Burel F (2007) Conservation of butterfly
populations in dynamic landscapes: the role of farming practices
and landscape mosaic. Ecol Model 205:135–145
Begon M (1979) Investigating animal abundance. Edward Arnold,
London
Binzenhöfer B, Settele J (2000) Vergleichende autökologische
Untersuchungen an Maculinea nausithous (Bergstr.; 1779) und
Maculinea teleius (Bergstr.; 1779) (Lep.: Lycaenidae) im
nördlichen Steigerwald. UFZ-Bericht 2:1–98
Blair RB, Launer AE (1997) Butterfly diversity and human land use:
species assemblages along an urban gradient. Biol Conserv
80:113–125
Boeren J, Eckelboom R, Wynhoff I (2011) Het Donker pimpernelblauwtje in het Nederlandse en Duitse Roerdal. Nat Hist
Maandbl 100(10):189–198
929
Bos F, Bosveld M, Groenendijk D, van Swaay C, Wynhoff I, De
Vlinderstichting (2006) De dagvlinders van Nederland, verspreiding en bescherming (Lepidoptera: Hesperioidea, Papilionoidea). Nederlandse fauna 7, Leiden, Nationaal Natuurhistorisch
Museum Naturalis, KNNV Uitgeverij & European Invertebrate
Survey—Nederland
Dennis RLH, Shreeve TG, Van Dyck H (2006) Habitats and
resources: the need for a resource-based definition to conserve
butterflies. Biodivers Conserv 15:1943–1966
Dierks A, Fischer K (2009) Habitat requirements and niche selection
of Maculinea nausithous and M. teleius (Lepidoptera: Lycaenidae) within a large sympatric metapopulation. Biodivers Conserv
18:3663–3676
Duelli P, Obrist MK (2003) Regional biodiversity in an agricultural
landscape: the contribution of semi-natural habitat islands. Basic
Appl Ecol 4:129–138
Field A (2005) Discovering statistics using SPSS, 2nd edn. SAGE
Publications, London
Fürst MA, Nash DR (2010) Host ant independent oviposition in the
parasitic butterfly Maculinea alcon. Biol Lett 6:174–176
Grill A, Cleary DFG, Stettmer C, Bräu M, Settele J (2008) A mowing
experiment to evaluate the influence of management on the
activity of host ants of Maculinea butterflies. J Insect Conserv
12(6):617–627
Hovestadt T, Nowicki P (2008) Investigating movement within
irregularly shaped patches: analysis of mark-release-recapture
data using randomization procedures. Isr J Ecol Evol 54:137–154
Hovestadt T, Binzenhöfer B, Nowicki P, Settele J (2011) Do all interpatch movements represent dispersal? A mixed kernel study of
butterfly mobility in fragmented landscapes. J Anim Ecol 80:
1070–1077
Jansen SHDR, Holmgren M, Wynhoff I (2006) Blues in the Roer
Valley—habitat selection and behaviour of Maculinea nausithous butterflies. Proc Neth Entomol Soc Meet 17:45–50
Johst K, Drechsler M, Thomas J, Settele J (2006) Influence of
mowing on the persistence of two endangered large blue
butterfly species. J Appl Ecol 43:333–342
Kivinen S, Luoto M, Heikkinen RK, Saarinen K, Ryttäri T (2008)
Threat spots and environmental determinants of red-listed plant,
butterfly and bird species in boreal agricultural environments.
Biodivers Conserv 17:3289–3305
Munguira ML, Martı́n J (eds) (1999) Action plan for the Maculinea
butterflies in Europe. Nature and environment, no 97. Council of
Europe Publishing, Strasbourg
Musche M, Anton C, Worgan AW, Settele J (2006) No experimental
evidence for host ant related oviposition in a parasitic butterfly.
J Insect Behav 19:631–643
Nowicki P, Witek M, Skórka P, Settele J, Woyciechowski M (2005a)
Population ecology of the endangered butterflies Maculinea
teleius and M. nausithous, and its implications for conservation.
Popul Ecol 47:193–202
Nowicki P, Richter A, Glinka U, Holzschuh A, Toelke U, Henle K,
Woyciechowski M, Settele J (2005b) Less input same output:
simplified approach for population size assessment in Lepidoptera. Popul Ecol 47:203–212
Nowicki P, Settele J, Thomas J, Woyciechowski M (2005c) A review
of population structure of Maculinea butterflies. In: Settele J,
Kühn E, Thomas JA (eds) Studies in the ecology and conservation of butterflies in Europe, species ecology along a European
Gradient: Maculinea butterflies as a model, vol 2. Pensoft
Publishers, Sofia, pp 144–149
Nowicki P, Pepkowska A, Kudlek J, Skórka P, Witek M, Settele J,
Woyciechowski M (2007) From metapopulation theory to
conservation recommendations: lessons from spatial occurrence
and abundance patterns of Maculinea butterflies. Biol Conserv
140:119–129
123
930
Nowicki P, Bonelli S, Barbero F, Balletto E (2009) Relative
importance of density-dependent regulation and environmental
stochasticity for butterfly population dynamics. Oecologia 161:
227–239
Patricelli D, Barbero F, La Morgia V, Casacci LP, Witek M, Balletto
E, Bonelli S (2011) To lay or not to lay: oviposition of
Maculinea arion in relation to Myrmica ant presence and host
plant phenology. Anim Behav 82:791–799
Pech P, Fric Z, Konvička M (2007) Species-specificity of the
Phengaris (Maculinea)—Myrmica host system: Fact or myth?
(Lepidoptera: Lycaenidae; Hymenoptera: Formicidae). Sociobiology 50:983–1003
Poschlod P, Bakker JP, Kahmen S (2005) Changing land use and its
impact on biodiversity. Basic Appl Ecol 6:93–98
Settele J (1998) Metapopulationsanalyse auf Rasterdatenbasis. Teubner Verlag, Leipzig
Söderström B, Hedblom M (2007) Comparing movement of four
butterfly species in experimental grassland strips. J Insect
Conserv 11:333–342
Stankiewicz A, Sielezniew M (2002) Host specificity of Maculinea
teleius Bgstr. and M. nausithous Bgstr. (Lepidoptera: Lycaenidae). The new insight. Annales Zoologici 52:403–408
Stankiewicz AM, Sielezniew M, Švitra G (2005) Myrmica schenki
(Hymenoptera: Formicidae) rears Maculinea rebeli (Lepidoptera: Lycaenidae) in Lithuania: new evidence for geographical
variation of host-ant specificity of an endangered butterfly.
Myrmecologische Nachrichten 7:51–54
Steiner FM, Sielezniew M, Schlick-Steiner BC, Höttinger H,
Stankiewicz A, Górnicki A (2003) Host specificity revisited:
new data on Myrmica host ants of the lycaenid butterfly
Maculinea rebeli. J Insect Conserv 7:1–6
Stettmer C, Binzenhöfer B, Hartmann P (2001) Habitatmanagement
und Schutzmassnahmen für die Ameisenbläulinge Glaucopsyche
teleius und Glaucopsyche nausithous. Teil 1: Populationsdynamik. Ausbreitungsverhalten und Biotopverbund. Natur und
Landschaft 76:278–287
Tartally A, Varga Z (2005) Myrmica rubra (Hymenoptera: Formicidae): the first data on host-ant specificity of Maculinea
nausithous (Lepidoptera: Lycaenidae) in Hungary. Myrmecologische Nachrichten 7:55–59
Tartally A, Rákosy L, Vizauer T-Cs, Goia M, Varga Z (2008)
Maculinea nausithous exploits Myrmica scabrinodis in Transylvania: unusual host ant species of a myrmecophilous butterfly in
an isolated region. Sociobiology 51:373–380
Thomas JA (1984) The behaviour and habitat requirements of
Maculinea nausithous (the Dusky Large Blue Butterfly) and M.
teleius (the Scarce Large Blue) in France. Biol Conserv
28:325–347
Thomas JA, Elmes GW (2001) Food-plant niche selection rather than
the presence of ant nests explains oviposition patterns in the
myrmecophilous butterfly genus Maculinea. Proc R Soc Lond B
268:471–477
Thomas JA, Elmes GW, Wardlaw JC, Woyciechowski M (1989) Host
specificity among Maculinea butterflies in Myrmica ant nests.
Oecologia 79:452–457
123
J Insect Conserv (2012) 16:921–930
Van Dyck H, Regniers S (2010) Egg spreading in the ant-parasitic
butterfly, Maculinea alcon: from individual behaviour to egg
distribution pattern. Anim Behav 80:621–627
Van Dyck H, Talloen W, Oostermeijer JGB, Feenstra V, Van der
Hidde A, Wynhoff I (2000) Does the presence of ant nests matter
for oviposition to a specialized myrmecophilous Maculinea
butterfly? Proc R Soc Lond B Biol Sci 267:861–866
Van Langevelde F, Grashof-Bokdam CJ (2011) Modelling the effect
of intersections in linear habitat on spatial distribution and local
population density. Int J Geogr Inf Sci 25:367–378
Van Langevelde F, Wynhoff I (2009) What limits the spread of two
congeneric butterfly species after their reintroduction: quality or
spatial arrangement of habitat? Anim Conserv 12:540–548
Van Swaay C (1990) An assessment of the changes in butterfly
abundance in The Netherlands during the 20th Century. Biol
Conserv 52:287–302
Van Swaay C (2002) The importance of calcareous grasslands for
butterflies in Europe. Biol Conserv 104:315–318
Van Swaay C, Warren M, Loı̈s G (2006) Biotope use and trends of
European butterflies. J Insect Conserv 10:189–209
Van Swaay C, Maes D, Collins S, Munguira ML, Šašić M, Settele J,
Verovnik R, Warren M, Wiemers M, Wynhoff I, Cuttelod A
(2011) Applying IUCN criteria to invertebrates: how red is the
Red List of European butterflies? Biol Conserv 144:470–478
Vanreusel W, Van Dyck H (2007) When functional habitat does not
match vegetation types: a resource-based approach to map
butterfly habitat. Biol Conserv 135:202–211
Wenzel M, Schmitt T, Weitzel M, Seitz A (2006) The severe decline
of butterflies on Western German calcareous grasslands during
the last 30 years: a conservation problem. Biol Conserv 128:
542–552
Witek M, Śliwińska EB, Skórka P, Nowicki P, Wantuch M, Vrabec
V, Settele J, Woyciechowski M (2008) Host ant specificity of
large blue butterflies Phengaris (Maculinea) (Lepidoptera:
Lycaenidae) inhabiting humid grasslands in East-central Europe.
Eur J Entomol 105:871–877
Wynhoff I (1998a) The recent distribution of the European Maculinea
species. J Insect Conserv 2:15–27
Wynhoff I (1998b) Lessons from the reintroduction of Maculinea
teleius and M. nausithous in the Netherlands. J Insect Conserv
2:47–57
Wynhoff I (2001) At home on foreign meadows: the reintroduction of
two Maculinea butterfly species. Dissertation, Wageningen
Agricultural University
Wynhoff I, Van Swaay C, Boeren J (2005) Overleven op de wegberm:
het Donker pimpernelblauwtje in Limburg. Nat Hist Maandbl
94:145–149
Wynhoff I, Grutters M, Van Langevelde F (2008) Looking for the
ants: selection of oviposition sites by two myrmecophilous
butterfly species. Anim Biol 58:371–388
Wynhoff I, Van Gestel R, Van Swaay C, Van Langevelde F (2011)
Not only the butterflies: managing ants on road verges to benefit
Phengaris (Maculinea) butterflies. J Insect Cons 15:189–206
Milena Holmgren