Population density and food analysis of Bombina variegata and Rana graeca in mountainous riverine ecosystems of northern Pindos (Greece)

Full text

129
INTRODUCTION
The mainland area of northwestern Greece hosts a
high biodiversity, including threatened, rare and en-
demic plants and animals. A large part of this area is
protected as a national park (Northern Pindos Na-
tional Park, Natura Network, site codes GR1310003
and GR2130001), originally designated as such for
the protection of the brown bear –one of the most
endangered mammalian species in Europe (Dafis et
al., 1996). Regarding amphibians, 14 species are known
to occur in northwestern Greece (Schneider et al.,
1984; Malakou et al., 1986; Denoël & Schabetsberger,
2003; Denoël, 2004; Sotiropoulos, 2004; Bisa, 2006),
but there are many gaps in our knowledge on their
ecology and population status.
Nowadays, it is fairly known that among vertebra-
tes, amphibians are subjected to a higher extinction
risk and 32.9% of extant species are considered
threatened worldwide (IUCN, 2004; Stuart et al.,
2004). As records on the decline of amphibian popu-
lations all over the world are increasing (Kiesecker et
al., 2001; Lips et al., 2004; Pounds et al., 2006), it is
urged that conservation measures need to be taken
for their protection. However, the lack of data on a-
bundance, life history and ecology of amphibian po-
pulations is the main obstacle for successful conser-
vation.
The aim of this study was to determine the popu-
lation density of Bombina variegata and Rana graeca
from permanent riverine habitats and to explore their
feeding ecology. In order to achieve this goal, we
conducted a study in water bobies with low human
impact located in the Northern Pindos National Park,
where the two species coexist. We focused on the
feeding habits of these species and we investigated
the spatial differences in the trophic resources, in an
attempt to investigate the food partitioning patterns
and the niche segregation mechanisms. Moreover, in
Population density and food analysis
of Bombina variegata and Rana graeca in mountainous
riverine ecosystems of northern Pindos (Greece)
RIKA BISA*, SPYROS SFENTHOURAKIS,
STELLA FRAGUEDAKIS-TSOLIS and BASIL CHONDROPOULOS
Section of Animal Biology,Department of Biology,University of Patras,
Panepistimioupoli 265 00 Rio,Greece
Received: 2 May 2007 Accepted after revision: 20 June 2007
Abundance and diet of Bombina variegata and Rana graeca were investigated from August 2004
to August 2005 in two permanent mountain water bodies (Zesto River and Prioni Nazeti Stream)
situated in the National Park of Northern Pindos (Greece). At Zesto, we recorded high popula-
tion densities for both anuran species, while at Nazeti population density was high only for R.
graeca. In the diet of B. variegata and of R.graeca, a large variety of prey taxa was identified in
the stomach contents, suggesting that they are opportunistic predators. However, at both sites,
the study species principally consumed five prey groups, i.e. ants, spiders, flies, terrestrial bee-
tles and aquatic insect larvae. A comparison of prey size consumed by adults revealed that B.var-
iegata feeds on smaller prey than R.graeca. Furthermore, intraspecific differences based on prey
size were obtained for adults and sub-adults of R.graeca. In this study we found that prey size is
a possible factor of niche partitioning for anuran species.
Key words: Anura, riverine habitats, density, diet, northern Pindos.
* Corresponding author: tel.: +30 6942 878712, fax: +30
2610 969267, e-mail: bisarika@hotmail.com
Journal of Biological Research-Thessaloniki 8: 129 – 137, 2007
J. Biol. Res.-Thessalon. is available online at http://www.jbr.gr
Indexed in: WoS
(Web of Science, ISI Thomson)
, CAS
(Chemical Abstracts Service)
and DOAJ
(Directory of Open Access Journals)

the case of R.graeca for which data were available, we
examined differences in the trophic spectrum of sub-
adults and adults.
Bombina variegata (Linnaeus, 1758) has a wide
distribution in the central and eastern Europe, while
R.graeca Boulenger, 1891 occurs mostly in the riveri-
ne mountain areas of the Balkan Peninsula (Arnold
& Ovenden, 2004). The snout-vent length of mature
specimens of B.variegata ranges from 30 to 52 mm.
Its biology, geographical distribution and life history
are well known (Barandun, 1990; Rafinska, 1991; Ba-
randun et al., 1997; Szymura, 1998; Buschmann, 2002;
Nürnberger et al., 2003) and studies concerning its
diet have revealed similarities to many other anurans
in terms of being opportunistic predators (Kuzmin,
1990). The body length of adults of R.graeca ranges
from 37 to 70 mm (Asimakopoulos, 1992). Studies
concerning the biology and ecology of R.graeca are
scarce, most likely due to the limited geographical di-
stribution and its preference for highland habitats.
MATERIALS AND METHODS
Study area
We conducted our research in two water bodies of
the Northern Pindos National Park: Zesto river and
Prioni Nazeti stream (Fig. 1). The climate of the
study area is Mediterranean-continental Mediter-
ranean with average winter temperatures of 0.5ÆC
(February) and summer temperatures of 25.9ÆC (Ju-
ly). Total annual precipitation is on average 1404 mm
(meteorological station: Metsovo 1159 m a.s.l., period
1981-1999).
Zesto river (39Æ53′ - 39Æ54′E, 21Æ07′ - 21Æ08′N)
is a permanent mountain river with a total length of
about 2.5 km. It flows through an area whose eleva-
tion ranges from 1600 to 1350 m a.s.l. The river is sit-
uated in one of the two cores of the park, named
Valia Kalda. A few streams feed into this river before
its flowing into Arkoudorema, a tributary of the Aoos
river. The dominant trees along the river banks are
the black pine (Pinus nigra) and the common beech
(Fagus sylvatica), whereas other characteristic plants
are the box-tree (Buxus sempervirens) and the bracken
fern (Pteridium aquilinum). Except for the two studied
taxa, other amphibians recorded during this study at
this water body were 13 adults and larvae of Triturus
alpestris, a single individual of Rana balcanica and
larvae of Salamandra salamandra and Bufo bufo.
Prioni Nazeti stream (39Æ56′E, 21Æ01′ - 21Æ02′N)
is located at the buffer zone of the National Park and
is a tributary of Aoos river. Prioni Nazeti stream
flows from 1300 to 1040 m a.s.l. and is surrounded by
slopes covered by a mixed coniferous forest. Apart
from the black pine, other trees of the bank vegeta-
tion are the common beech, the willow (Salix sp.), the
juniper (Juniperus sp.) and the European hornbeam
(Carpinus betulus). Regarding amphibians at this wa-
ter body, we also recorded five adults of B.bufo, lar-
vae of S.salamandra and one individual of R.balcani-
ca. For simplicity, we will hereafter refer to the study
sites as “Zesto” and “Nazeti”, respectively.
130 Rika Bisa et al. — Population density and food analysis of Bombina variegata and Rana graeca
FIG. 1. Geographical location of the two study sites and the nearest human settlement (Vovousa) at the study area.

Estimation of population density
Data were collected monthly from August 2004 to
August 2005, except for the hibernation period of
amphibians, which in the area in question lasts from
late October to early April. We visited both sites in
late October but we did not record any activity of the
study animals. Due to prolonged winter and severe
snowfall we could not approach the study sites before
May and we could not sample from Zesto in May
2005, due to a landslide that rendered any access to
the area impossible. Fieldwork was conducted between
11:00 and 16:00, during sunny days, since amphibians
are more active in such weather conditions (Heyes et
al., 1994). At each site, 15 randomly selected plots
(5×2 m) were monitored in a fixed area of 200 m in
length. Two observers were engaged in the sampling
effort in order to minimize escaping of the animals.
The plots were 3 to 5 m apart, in order to avoid pre-
sampling disturbances of animals and physical obsta-
cles. We located the animals after overturning stones
and logs and checking crevices, exposed roots of trees
and among riparian vegetation. All captured animals
were identified and released after the investigation,
which lasted approximately 10 to 13 min per plot. In
order to detect differences in population sizes between
the sites, we applied independent t-tests and we con-
ducted Kruskal-Wallis tests to assess differences in
the mean density among the months sampled.
Diet analysis
Individuals of B.variegata and R.graeca were colle-
cted by hand or dip netting. They were measured for
snout-vent length (SVL) with a ruler to the nearest 1
mm. The specimens of each species were kept in dif-
ferent baskets so that toxic secretions of B.variegata
could not affect R.graeca individuals. Analysis of the
stomach content of each specimen was performed
within half an hour after capture, by applying stomach
flushing without anaesthesia (Legler & Sullivan,
1979). For this purpose, we used a syringe of 20 ml
and the watered down stomach contents were indi-
vidually stored in vials containing 70% ethanol as
preservative. None of the specimens died during this
application. After this procedure, individuals were
released at the proximity of their capture point.
In the laboratory, all prey items were usually iden-
tified to the order level. We measured length and
width of each prey item to the nearest 0.1 mm under
a stereomicroscope with an ocular micrometer. The
presence of pebbles, plant remains and ingested skin
of amphibians was also recorded. The diet composi-
tion was calculated as both relative abundance and
frequency of occurrence.
Chi-square contingency tables were used to test
the null hypothesis of equal proportions of prey
groups consumed throughout surveys by the two spe-
cies. Wilcoxon signed-rank tests were applied to
search for differences in the size of the prey con-
sumed. All tests were performed using the SPSS soft-
ware (version 13.0). In all tests alpha was set to 0.05.
RESULTS
Population density estimation
At Zesto, monthly density values of B.variegata ran-
ged from 0.13 adults/ 10 m2(September 2004) to 2.47
adults/ 10 m2(August 2005) (Table 1). The density of
B.variegata was higher in late summer and decreased
thereafter until September. For R.graeca, monthly
density values ranged from 0.33 adults/ 10 m2(Octo-
Rika Bisa et al. — Population density and food analysis of Bombina variegata and Rana graeca 131
TABLE 1. Numbers of individuals (N) encountered in the 15 plots used, mean and standard error (S.E.) of density (per 10
m2) for adults of B.variegata and of R.graeca, number (Nplots) and percentage of plots (in parenthesis) in which species were
syntopic at the two study sites at each month sampled (Au: August, Se: September, Oc: October, Ma: May, Jn: June, Jl: July)
Zesto Nazeti
B. variegata R. graeca Syntopic B. variegata R. graeca Syntopic
Month N Mean ± S.E. N Mean ± S.E. Nplots N Mean ± S.E. N Mean ± S.E. Nplots
Au 04 24 1.60 ± 0.32 20 1.33 ± 0.32 9 (0.60) 4 0.18 ± 0.12 27 1.80 ± 0.33 3 (0.20)
Se 04 2 0.13 ± 0.09 10 0.67 ± 0.23 1 (0.07) 0 0.00 5 0.33 ± 0.16 0
Oc 04 0 0.00 5 0.33 ± 0.21 0 0 0.00 3 0.20 ± 0.14 0
Ma 05 1 0.07 ± 0.07 7 0.47 ± 0.17 1 (0.07)
Jn 05 13 0.87 ± 0.41 7 0.47 ± 0.22 1 (0.07) 1 0.07 ± 0.07 8 0.53 ± 0.19 1 (0.07)
Jl 05 25 1.67 ± 0.45 16 1.07 ± 0.27 6 (0.40) 8 0.53 ± 0.22 9 0.60 ± 0.19 4 (0.27)
Au 05 37 2.47 ± 0.50 33 2.20 ± 0.37 11 (0.70) 6 0.40 ± 0.16 28 1.87 ± 0.32 5 (0.33)

ber 2004) to 2.20 adults/ 10 m2(August 2005). The
mean density values among the months were signifi-
cantly different for B.variegata (Kruskal-Wallis test:
¯2=34.2, df=5, p< 0.001) and R.graeca (Kruskal-
Wallis test: ¯2=28.3, df=5, p< 0.001). For both spe-
cies, densities obtained in August 2005 were much
higher than those observed in August 2004. In late
summer, individuals of B.variegata and R.graeca fre-
quently coexisted at the same plot as other ephemeral
water bodies dried up (Table 1).
At Nazeti, R. graeca was the prevailing amphibian
dweller and was recorded in all months sampled.
Monthly densities ranged from 0.20 adults/ 10 m2
(October 2004) to 1.87 adults/ 10 m2(August 2005)
(Table 1). Bombina variegata was found mainly during
the breeding period and its density ranged from total
absence (September 2004) to 0.53 adults/10 m2(July
2005). Throughout the months sampled, the mean
density values significantly differed for both species
(Kruskal-Wallis tests, B.variegata: ¯2=19.3, df= 6,
p<0.005; R.graeca:¯2=45.6, df =6, p< 0.001).
The comparison of the population density of each
species between the two sites revealed that the abun-
dances of adult specimens of R.graeca did not differ
(independent t-test: t193=1.117, p=0.27), but the
abundance of B.variegata at Zesto was significantly
higher than that at Nazeti (independent t-test: t193=
5.700, p<0.001).
Diet analysis
At Zesto, we identified 479 prey items in the stom-
achs of 37 adult individuals of B.variegata, 227 prey
items in 20 adults and 127 prey items in 16 sub-adults
of R.graeca (Table 2). The mean number of prey
items/stomach was higher for B.variegata (13.3) than
for R.graeca (adults: 11.35 and sub-adults: 7.88). The
highest number of prey items/stomach was found in
an adult specimen of B.variegata (snout-vent length
35 mm) which had consumed 109 ants and one spider.
The diet of adults of both species mainly consisted
of five major prey groups, namely spiders, terrestrial
beetles (Carabidae, Clambidae, Cleridae, Chrysomel-
idae, Curculionidae, Elateridae, Lymexylidae, Sta-
phylinidae), ants, flies and aquatic insect larvae (Dip-
tera, Ephemeroptera nymphs, Odonata, Plecoptera
nymphs, Trichoptera). Throughout surveys conducted
at Zesto, the adults of B.variegata and R.graeca did
not differ significantly in the consumption of the five
main prey groups (¯2=7.19, df=4, p= 0.13). Bombi-
na variegata consumed often ants and aquatic insect
larvae and R.graeca consumed mostly ants and spi-
ders (Fig. 2).
The diet of sub-adults of R.graeca mainly consisted
of the same five main groups mentioned for adults.
Throughout surveys, sub-adults of R.graeca did not
differ significantly in the consumption of the main
prey groups from adults (¯2=3.60, df=4, p= 0.42).
They preyed mostly on ants, spiders and flies (Fig. 2).
At Nazeti, we identified 158 prey items in the stom-
ach contents of 29 adults of B.variegata, 110 prey items
in 21 adults and 61 prey items in 15 sub-adults of R.
graeca. The mean number of prey items/stomach was
similar for both species (Table 2). Throughout the
sampling period, B.variegata and R.graeca mostly
preyed on five prey groups: spiders, terrestrial beetles
(Bostrichidae, Byrrhidae, Carabidae, Cleridae,
Chrysomelidae, Curculionidae, Elateridae, Histeridae,
Scolytidae) flies, ants and aquatic insect larvae
(Diptera, Ephemeroptera nymphs, Odonata, Ple-
coptera nymphs, Trichoptera). Over all surveys, we
recorded a significant variation in the consumption of
the main prey groups in the adults of B.variegata and
R.graeca (¯2=22.03, df =4, p< 0.001). Bombina varie-
gata consumed principally flies, ants and aquatic insect
larvae, whereas R.graeca mostly preyed on spiders.
132 Rika Bisa et al. — Population density and food analysis of Bombina variegata and Rana graeca
TABLE 2. Total number of stomach-flushed specimens (NTotal), number and percentage (in parenthesis) of full stomachs
(Nfull), total number of prey items (n), mean number and standard error (S.E.) of prey items/stomach of adults (ad) of B.var-
iegata and adults and sub-adults (subad) of R.graeca at the two study sites
Zesto Nazeti
B. variegata R. graeca B. variegata R. graeca
(ad) (ad) (subad) (ad) (ad) (subad)
NTotal 37 20 17 29 21 16
Nfull 36 (97.3%) 19 (97.3%) 17 (100%) 29 (100%) 20 (97.3%) 16 (100%)
n 479 227 129 158 110 61
Mean ± S.E. 13.3 ± 3.6 11.35 ± 2.91 7.88 ± 1.62 5.40 ± 0.6 5.24 ± 0.82 4.07 ± 0.75

From the comparison of different age classes of R.
graeca, we observed that throughout the sampling
period, the diet of sub-adults was significantly differ-
ent from that of adults (¯2 = 23.55, df = 4, p <
0.001). Specifically, sub-adults of R.graeca mostly
preyed on flies and ants (Fig. 2). Between sites, we
obtained interpopulation differences in the diet of
adults of B.variegata (¯2=14.96, df=4, p< 0.05) and
of adults of R.graeca (¯2=17.07, df= 4, p< 0.005).
Sub-adults of R.graeca did not differ (¯2=3.89,
df=4, p=0.42).
Other prey taxa identified in the stomach contents
of both species were flying insects, bugs and inverte-
brates associated with soil, litter or vegetation (Table
3). Opiliones were identified in the stomach contents
of R.graeca, whereas in the stomach samples of B.
variegata opiliones were never found. Pebbles and
plant material were present in most stomach samples
of both species, but their quantity was insignificant,
and most probably had been accidentally swallowed
while capturing invertebrate prey. Parasitic nema-
todes were found in two adults of B.variegata and in
one adult of R.graeca, and trematodes in one adult of
R.graeca.
Prey size
At both sites, throughout surveys, the mean length
(L) and width (W) of the prey consumed from the
adults of B.variegata were lower than those from the
adults of R.graeca (Table 4).
At Zesto, the mean values of the maximum length
and maximum width of the prey consumed were sig-
nificantly different between the adult specimens of
the two species (Wilcoxon signed-rank test, maxL: Z
=–2.398, maxW: Z=–2.173, p<0.05 in both cases).
Also, at Nazeti, the adults of B.variegata and R.grae-
ca significantly differed in mean length, mean maxi-
mum length, mean width and mean maximum width
of the prey consumed (L: Z=–2.529, W: Z=–2.876
maxL: Z=–2.743, maxW: Z=–2.473, p<0.05 in all
cases).
A similar comparison between different age classes
of R.graeca revealed differences in the prey size con-
Rika Bisa et al. — Population density and food analysis of Bombina variegata and Rana graeca 133
B. variegata
36 ad
R. graeca
17 subad
R. graeca
19 ad
0 20 40 60 80 100
B. variegata
36 ad
R. graeca
17 subad
R. graeca
19 ad
0 20 40 60 80 100
B. variegata
29 ad
R. graeca
16 subad
R. graeca
20 ad
0 20 40 60 80 100
B. variegata
29 ad
R. graeca
16 subad
R. graeca
20 ad
0 20 40 60 80 100
Araneae
Coleoptera
Diptera
Formicidae
Aquatic insect
larvae
Zesto Nazeti
Relative abundance
Frequency of occurrence
FIG. 2. Relative abundance and frequency of occurrence of the five main prey groups identified in the stomach contents of
adults (ad) of B.variegata and of adults and sub-adults (subad) of R. graeca at Zesto and Nazeti.

134 Rika Bisa et al. — Population density and food analysis of Bombina variegata and Rana graeca
TABLE 3. Relative abundance (A%) and frequency of occurrence (F%) of other prey taxa encountered in stomach contents
of adults (ad) of B.variegata and adults and sub-adults (subad) of R.graeca at the two study sites
Zesto Nazeti
B. variegata R. graeca B. variegata R. graeca
36 (ad) 20 (ad) 16 (subad) 29 (ad) 21 (ad) 15 (subad)
Prey taxa A% F% A% F% A% F% A% F% A% F% A% F%
Acari 0.3 5.6 0.4 5.0 1.7 12.2 1.5 6.9 0.3 4.7 1.7 6.6
CollembÔla 2.0 7.3 – – 5.1 18.7 – – – – 1.7 6.6
Chilopoda 0.6 2.8 0.5 5.0 1.0 6.2 2.2 6.9 – – 0.6 6.6
Dermaptera 0.2 2.8 2.3 10.0 – – – – – – 1.7 6.6
Dictyoptera – – 0.4 5.0 – – 3.6 13.8 2.8 14.3 1.3 6.6
Diplopoda – – 0.6 5.0 – – 1.2 3.4 2.8 9.5 1.3 6.6
Gastropoda 1.6 5.6 1.5 5.0 1.0 6.2 – – 0.3 4.7 – –
Hemiptera 5.7 25.0 1.3 15.0 1.9 12.5 8.8 27.6 4.0 23.8 10.6 26.6
Hydrophilidae 2.9 11.1 0.4 5.0 – – 2.6 13.8 – – – –
Hymenoptera 2.7 22.2 1.5 10.0 3.7 18.7 6.3 24.1 4.4 14.3 3.8 20.0
Isopoda – – – – – – – – 1.9 14.3 8.9 13.3
Lepidoptera 1.4 2.8 – – – – – – – – 2.2 6.6
Orthoptera 0.6 5.6 1.3 15.0 1.2 6.2 1.2 6.9 2.1 9.5 1.7 6.6
Plecoptera 0.8 5.6 – – – – 0.4 3.4 – – – –
Pseudoscorpiones – – – – – – – – – – 1.1 6.6
Trichoptera – – 0.6 5.0 – – – – – – –
Coleoptera larvae 3.7 16.6 0.4 5.0 2.3 12.5 2.9 10.3 2.4 9.5 3.6 13.3
Hymenoptera larvae – – 0.4 5.0 1.2 6.2 – – 2.1 9.5 0.6 6.6
Lepidoptera larvae – – 4.2 15.0 3.1 12.5 0.4 3.4 2.6 14.3 – –
Neuroptera larvae 0.7 2.8 – – – – 0.5 3.4 – – – –
Opiliones – – 5.3 20.0 3.0 18.7 – – 5.6 23.8 3.3 13.3
Undetermined 3.9 13.8 3.5 10.0 – – 1.6 6.9 5.1 19.0 0.6 6.6
TABLE 4. Values (mean, standard error and range) of the variables of prey size for adults (ad) of B. variegata and adults and
sub-adults (subad) of R. graeca at the two sites. SVL: snout-vent length of anurans studied, L: length, maxL: maximum length,
W: width, maxW: maximum width, S.E.: standard error (all values are expressed in mm)
Zesto
B. variegata (N = 36) R. graeca (N = 20) R. graeca (N = 16)
(ad) (ad) (subad)
Variables Range Mean ± S.E. Range Mean ± S.E. Range Mean ± S.E.
SVL 32.0-49.0 42.09±0.81 35.0-62.0 48.07±1.99 16.0-32.5 24.85±1.42
L 2.3-12.2 4.88±0.37 3.2-14.6 6.75±0.71 1.6-7.8 4.48±0.41
maxL 2.4-16.8 7.34±0.57 2.2-26.1 12.15±1.32 2.0-22.9 8.51±1.18
W 0.5-2.1 1.23±0.07 0.9-4.2 1.74±0.19 0.6-2.1 1.16±0.09
maxW 0.5-3.8 1.95±0.13 1.1-6.0 2.58±0.24 0.8-2.9 1.90±0.13
Nazeti
B. variegata (N = 29) R. graeca (N = 21) R. graeca (N = 15)
(ad) (ad) (subad)
Variables Range Mean ± S.E. Range Mean ± S.E. Range Mean ± S.E.
SVL 30.0-53.0 41.65±1.27 36.0-78.0 47.14±2.26 15.0-30.0 23.57±1.32
L 2.9-13.4 5.20±0.41 3.5-15.7 7.38±0.57 2.6-9.2 4.91±0.54
maxL 3.1-22.3 7.89±0.67 0.6-4.0 2.02±0.16 2.6-11.6 7.39±0.74
W 0.5-2.9 1.45±0.11 3.9-24.1 10.87±0.97 0.8-3.5 1.46±0.19
maxW 0.7-4.7 2.21±0.18 0.6-5.5 2.87±0.23 1.2-3.5 1.99±0.18

sumption of the sub-adults and adults. At Zesto, sub-
adults and adults differed in the mean width of the
prey (W: Z=–1.992, p<0.05), whereas at Nazeti dif-
ferent age classes of R.graeca differed in the mean
length and width of the prey ingested (L: Z=–2.188,
W: Z=–2.160, p<0.05).
DISCUSSION
During our study it was revealed that the population
densities of B.variegata and R.graeca follow similar
seasonal patterns. However, R.graeca remains active
for a longer period of time. The monthly variation in
the observed densities of the adults of both species
may be attributed to two possible factors: 1) fluctua-
tion of the density due to breeding activity, 2) expan-
sion or shrinkage of the area used, due to changes in
weather conditions. As adults of B.variegata assem-
ble and breed from June to August (personal obser-
vations), it is likely that during summer, the popula-
tion density increased due to mating activity. This
does not seem to be the case for R.graeca, which in
altitudes similar to those of the study sites, breeds in
mid-spring (Asimakopoulos, 1992). Besides, with
regard to the second factor, during May and June
2005, rainfalls were frequent and intense at the study
sites. Therefore, the densities recorded during this
period of the year were low and it is very likely that
individuals of both species spread in the surrounding
area. In late summer, water resources decreased and
ephemeral water bodies dried up. As the life style of
both species is closely related to water, the area with
favorable conditions correspondingly shrunk. This is
probably the reason why during that period, the high-
est population densities were obtained and the high-
est proportion of coexistence in number of individuals
of both species at each sampling plot was recorded.
At Zesto, population densities for both species
were by more than one third higher in August 2005
compared with those obtained in August 2004. The
section of the river sampled in August 2004 (narrow
riverbed, with deep points and large stones and
rocks) differed morphologically from the section of
the river surveyed the following August (wide river-
bed with small stones) enabling a better estimation of
the population sizes. In any case, it is possible that the
population densities recorded for both species may
constitute an underestimation of the true values due
to difficulties in tracing all possible anuran shelters.
Our data on diet composition of B.variegata and
R.graeca clearly underpin their status as carnivorous
generalists. The stomach contents of both species
consisted only of invertebrate prey. At Zesto, the diet
composition was similar for both species with ants
and spiders being generally consumed in higher pro-
portions. As mentioned above, high population den-
sities for both species were also recorded at this site.
Nevertheless, opportunistic predation benefits the
two syntopic species and the local anuran community
can coexist in high densities, sharing the same space
and food resources. However, at Nazeti interspecific
differences in the consumption of the five main prey
groups were observed between the adults of B.varie-
gata and of R.graeca. At this site, B.variegata mainly
preyed on flies and ants, whereas R.graeca mostly
consumed spiders, ants and aquatic insect larvae.
Most likely, based on the availability of food resour-
ces in their habitats, the studied species adjust their
diets accordingly.
Other studies on the diet of B.variegata per-
formed in forest ecosystems of Romania (Ghiurca˘&
Zaharia, 2005; Sas et al., 2005) report the prevalence
of terrestrial prey categories in the diet of this spe-
cies. In particular, Ghiurca˘& Zaharia (2005) found
that bugs, flies, spiders and ants were the most impor-
tant prey categories in the diet of B.variegata. These
prey categories are important in the diet of B.varie-
gata with the difference that ants were eaten in higher
proportions. Sas et al. (2005) reported that apart from
bugs, ants, and flies, snails constitute a significant
trophic resource. Although we identified snails in the
stomach contents of B.variegata, this prey was rarely
consumed.
At both sites, the comparison of prey size con-
sumed by adults of B.variegata and R.graeca showed
that the former preyed on thinner and smaller prey
than the latter. Most likely, morphological differences
between the species, such as mouth opening and pre-
dator size, contribute principally to the trophic re-
source partitioning. Kuzmin (1990) and Coga˘lniceanu
et al. (2001) also report the significance of the prey
size in relation to the predator size in the trophic re-
source partitioning of syntopic species.
Regarding the diet of the sub-adults of R.graeca,
it was found that their trophic spectrum consisted
principally of the same main groups specified for
adults. At Zesto, both age classes of R.graeca preyed
upon the same groups, and ants and spiders were
mostly consumed. At Nazeti, sub-adults and adults
differed in the consumption of the main groups and
immature specimens preyed mostly on ants and flies,
whereas mature specimens mainly consumed spiders.
Rika Bisa et al. — Population density and food analysis of Bombina variegata and Rana graeca 135

The analysis of prey size revealed intraspecific differ-
ences between different age classes of R.graeca, with
sub-adults ingesting smaller prey than adults. In the
present study, as expected, we evinced that prey size
is an important factor in the partitioning of food re-
sources between sub-adults and adults, but further
studies focusing on the diet of different age classes of
R.graeca will provide more detailed results.
ACKNOWLEDGEMENTS
We thank the officers of the Forest Management of
Grevena and Metsovo and especially Yiannis Tzatza-
nis for his precious collaboration regarding visits in
the park. We are grateful to Robert Schabetsberger
and George Mitsainas for their valuable comments
on the manuscript. We also thank Kostas Sotiropou-
los for technical support and Dimitra Bisa for helping
at the field-work. This research was partially financed
by the WWF-Hellas.
REFERENCES
Arnold N, Ovenden D, 2004. Field guide of reptiles and am-
phibians. Harper Collins Publishers, London.
Asimakopoulos V, 1992. Geographical distribution, biolo-
gy and ecology of the Balkan stream frog Rana graeca
Boulenger in Greece. Ph. D. Thesis, University of
Thessaloniki.
Barandun J, 1990. Reproduction of yellow bellied-toads
Bombina variegata in a man made habitat. Amphibia-
reptilia, 11: 277-284.
Barandun J, Reyer HU, Bradley A, 1997. Reproductive
ecology of Bombina variegata: aspects of life history.
Amphibia-reptilia, 18: 347-355.
Bisa R, 2006. Study of the diversity of amphibians and of
the trophic ecology of characteristic species in riveri-
ne ecosystems of Northern Pindos. M. Sc. Thesis, U-
niversity of Ioannina.
Buschmann H, 2002. Fecundity of yellow bellied toads
Bombina variegata under free-range conditions: an
indication of risk-spreading strategy. Amphibia-rep-
tilia, 23: 362-369.
Coga˘lniceanu D, Palmer MW, Ciubuc C, 2001. Feeding in
anuran communities on islands in the Danube flood-
plain. Amphibia-reptilia, 22: 1-19.
Dafis S, Papastergiadou E, Georghiou K, Babalonas D,
Georgiadis T, Papageorgiou M, Lazaridou T, Tsia-
oussi V, 1996. Directive 92/43/EEC. The Greak “Ha-
bitat” Project NATURA 2000: An overview. Life
Contract B4-3200/94/756, Commission of the Euro-
pean Communities DG XI, The Goulandris Natural
History Museum – Greek Biotope/Wetland Center.
Denoël M, 2004. Distribution and characteristics of aquatic
habitats of newts and yellow-bellied toads in the dis-
trict of Ioannina (Epirus, Greece). Herpetozoa, 17:
49-64.
Denoël M, Schabetsberger R, 2003. Resource partitioning
in two heterochronic populations of Greek Alpine
newts, Triturus alpestris veluchiensis. Acta oecologica,
24: 55-64.
Ghiurca˘D, Zaharia L, 2005. Data regarding the trophic
spectrum of some population of Bombina variegata
from Baca˘u county. North-western journal of zoology,
1: 15-24.
Heyes RW, Donnelly MA, McDiarmid RW, Hayek LA,
Foster MS, 1994. Measuring and monitoring biological
diversity:standard methods for amphibians. Smith-
sonian Institution Press, Washington DC.
IUCN (World Conservation Union), Conservation Inter-
national, Nature Serve (2004). Global amphibian as-
sessment. IUCN, Gland, Switzerland and Cambrid-
ge, United Kingdom. Available from http://www.glo
balamphibians.org
Kiesecker JM, Blaustein AR, Belden LK, 2001. Complex
causes of amphibian population declines. Nature,
410: 681-684.
Kuzmin SL, 1990. Trophic niche overlap in syntopic post-
metamorphic amphibians of the Carpathian moun-
tain (Ukraine: Soviet Union). Herpetozoa, 3: 13-24.
Legler JM, Sullivan LJ, 1979. The application of stomach-
flushing to lizards and anurans. Herpetologica, 35:
107-110.
Lips KR, Mendelson JR, Munoz-Alonso A, Canseco-Ma´r-
quez L, Mulcahy DG, 2004. Amphibian population
declines in montane southern Mexico: resurveys of
historical localities. Biological conservation, 119: 555-
564.
Malakou M, Ravasini M, Tsunis G, 1986. The amphibians
and reptiles of the north-eastern Pindos Mountain
Reserve in Central Greece. Herptile, 11: 56-61.
Nürnberger B, Hofman S, Förg-Brey B, Praetzel G, Ma-
clean A, Szymura JM, Abbott M, Barton NH, 2003.
A linkage map for the hybridising toads Bombina
bombina and B. variegata (Anura: Discoglossidae).
Heredity, 91: 136-142.
Pounds AJ, Bustamante MR, Coloma LA, Consuegra JA,
Fogden MPL, Foster PN, La Marca E, Masters KL,
Merino-Viteri A, Puschendorf R, Ron SR, Sa´nchez-
Azofeifa GA, Still CJ, Young BE, 2006. Widespread
amphibian extinctions from epidemic disease driven
by global warming. Nature, 439: 161-167.
Rafinska A, 1991. Reproductive biology of the fire-bellied
toads, Bombina bombina and Bombina variegata (A-
nura: Discoglossidae): egg size, clutch size and larval
period length differences. Biological journal of the lin-
nean society, 43: 197-210.
Sas I, Covaciu-Marcov SD, Cups¸a D, Cicort-Lucaciu AS,
Popa I, 2005. Food analysis in adults (males/females)
136 Rika Bisa et al. — Population density and food analysis of Bombina variegata and Rana graeca

and juveniles of B.variegata. Analele scientifice ale
universitatii “Al.I. Cuza” biologie animala˘, Tom LI:
169-177.
Schneider H, Sofianidou TS, Kyriakopoulou-Sklavounou P,
1984. Bioacoustic and morphometric studies in water
frogs (genus Rana) of lake Ioannina in Greece, and
description of a new species (Anura, Amphibia). Zeit-
schrift für zoologische systematik und evolutionsfor-
schung, 22: 349-366.
Sotiropoulos K, 2004. Genetic and morphological variation
of Triturus species (Amphibia, Urodela) in the Hel-
lenic region: historical and ecological inferences. Ph.
D. Thesis, University of Athens.
Stuart SN, Chanson JS, Cox NA, Young BE, Rodrigues
AS, Fischman DL, Waller RW, 2004. Status and
trends of amphibian declines and extinctions world-
wide. Science, 306: 1783-1786.
Szymura JK, 1998. Origin of the yellow-bellied toad popu-
lation, B.variegata, from Goritzhain in Saxony. Her-
petological journal, 8: 201-205.
Rika Bisa et al. — Population density and food analysis of Bombina variegata and Rana graeca 137