83
Does local feeding specialization exist in Eurasian
badgers?
Eloy Revilla and Francisco Palomares
Abstract: Several local populations of the otherwise trophic-generalist Eurasian badger (Meles meles) have been defined as specializing locally on temporally variable food resources such as earthworms (Lumbricus spp.), olive fruits
(Olea europaea), or young rabbits (Oryctolagus cuniculus), owing to a lack of correlation between resource availability
and use. However, theoretical models predict that temporal variation in resources reduces the probability of diet specialization. To understand the relationship between temporal resource variability and local feeding specialization, we
studied temporal variation in diet composition and diversity (using fecal analysis), the availability of a temporally stable key resource, and the relation between consumption and availability of rabbits (key prey) and invertebrates (secondary prey) for a badger population previously described as specialized on young rabbits. We found strong variations in
the use of different resources (including young rabbits) and in diet diversity among seasons and years. The main food
resource was young rabbits during winter and spring, fruits in autumn, and reptiles in summer. Diet diversity was inversely related to consumption of young rabbits and directly related to consumption of secondary prey (invertebrates).
Consumption of rabbits (both young and adults) was correlated with their abundance in the field, with a type 3 functional response in the consumption of young rabbits, which is typical of a generalist to whom alternative prey are
available. There was no relationship between the abundance of invertebrates and their consumption. Our results show
that badgers in the study area were not locally specialized, therefore care should be taken when referring to a population as specialized without an adequate test of the predictions.
Résumé : Plusieurs populations locales du blaireau eurasien (Meles meles), aux habitudes alimentaires ordinairement
généralistes, sont reconnues comme des spécialistes de ressources alimentaires temporaires variables, telles que les vers
de terre (Lumbricus spp.), les olives (Olea europaea) ou les jeunes lapins (Oryctolagus cuniculus), parce qu’il n’y a
pas de corrélation entre la disponibilité de ces ressources et leur utilisation. Cependant, les modèles théoriques prédisent que la variation temporelle des ressources réduit la probabilité de la spécialisation du régime alimentaire. Pour
comprendre la relation entre la variabilité des ressources et la spécialisation alimentaire locale, nous avons étudié la variation temporelle de la composition et de la diversité du régime (par analyse des fèces), la disponibilité d’une ressource importante stable et la relation entre la consommation et la disponibilité de lapins (proies principales) et
d’invertébrés (proies secondaires) chez une population de blaireaux connue comme spécialisée dans la consommation
de lapins. Nous avons trouvé d’importantes variations dans l’utilisation des différentes ressources (y compris les lapereaux) et la diversité des aliments selon la saison et l’année. Les lapereaux sont la ressource principale en hiver et au
printemps, à l’automne, ce sont les fruits et, en été, les reptiles. La diversité dans le régime est inversement proportionnelle à la consommation de lapereaux et directement proportionnelle à la consommation des proies secondaires (invertébrés). La consommation de lapins (jeunes et adultes) est en corrélation avec leur abondance selon une réponse
fonctionnelle de type 3 dans le cas des lapereaux, ce qui est typique d’un organisme généraliste qui a accès à des
proies de rechange. Il n’y a pas de relation entre l’abondance des invertébrés et leur consommation. Les blaireaux de
notre site d’étude ne sont donc pas localement des consommateurs spécialisés, ce qui souligne l’importance de ne pas
considérer une population comme spécialisée sans avoir éprouvé l’exactitude des prédictions.
[Traduit par la Rédaction]
Introduction
A species or population is considered more specialized
than others if its diet breadth is narrower than theirs (Begon
and Mortimer 1986; Futuyma and Moreno 1988). However,
because a quantitative measurement of diet breadth is difficult to obtain (Feinsinger et al. 1981), a second criterion is
frequently used, which is a comparison of resource use and
availability. In this case, a more generalist species varies its
intake of alternative prey in response to fluctuations in their
.
E. Revilla.1 Department of Applied Biology, Estación Biológica de Doñana, Consejo Superior de Investigaciones Científicas,
Avenida Maria Luisa s/n, 41013 Sevilla, Spain, and Department of Ecological Modelling, UFZ (Centre for Environmental Research
Leipzig-Halle), Permoserstraße 15, D-04318 Leipzig, Germany.
F. Palomares. Department of Applied Biology, Estación Biológica de Doñana, Consejo Superior de Investigaciones Científicas,
Avenida Maria Luisa s/n, 41013 Sevilla, Spain.
1
Corresponding author (e-mail: revilla@ebd.csic.es).
abundance. However, even using both criteria, distinguishing between the extremes of the specialist–generalist continuum may be very difficult.
Allopatric populations of generalist species may specialize
by means of behavioral responses (i.e., behavioral specialization) to variation in abundance of prey, hence reducing
their diet breadth (Futuyma and Moreno 1988; Partridge and
Green 1985; West-Eberhard 1989). Behavioral specialization
in a population is characterized by overuse of the key resource independently of its availability, and with few, if any,
possibilities of reversal to the original generalist pattern
(Futuyma and Moreno 1988). These specialists should cope
with resource scarcity by using alternative mechanisms (other
than increasing their food spectrum), such as allocating more
time to searching for the key prey or increasing their energetic
efficiency (e.g., Ward and Krebs 1985; Fietz and Ganzhorn
1999).
The Eurasian badger, Meles meles, is a territorial carnivore with a broad distribution in the Palearctic (Long and
Killingley 1983; Lüps and Wandeler 1993). It is considered
a trophic generalist (Roper 1994; Roper and Mickevicius 1995;
Neal and Cheeseman 1996) because it uses a wide variety
of food resources (e.g., Skoog 1970; Ciampalini and Lovari
1985; Pigozzi 1988; Rodríguez and Delibes 1992; Roper
and Mickevicius 1995; Kauhala et al. 1998). Badgers exploit
areas of their territory with some degree of spatiotemporal
predictability, following a foraging strategy of searching within
a restricted area (clumped resources; Mellgren and Roper
1986). Within this conceptual framework, local feeding specialization has been reported for several badger populations
and for different resources such as earthworms (Lumbricus
spp.) (Kruuk and Parish 1981), olive fruits (Olea europaea)
(Kruuk and de Kock 1981), and young rabbits (Oryctolagus
cuniculus) (Martín et al. 1995). In all these cases, local feeding specialization is defined as the stable use of a main resource and the absence of a correlation between the use of
this resource and its availability (even when there are wide
fluctuations in abundance; Kruuk and Parish 1981; Martín et
al. 1995). However, theory predicts that the use of resources
with a high degree of temporal predictability can lead to
specialization (Wilson and Yoshimura 1994), therefore the
question of how badgers in these specialized populations
cope with temporal variability in the main resource is still
unanswered.
In this paper we aim to answer this question for a population of badgers in southwestern Spain, where they have been
reported as specializing on young rabbits (Martín et al. 1995;
Fedriani et al. 1998). We consider two alternative hypotheses
about how the diet of badgers is affected by temporal variation in the availability of the main food type. If badgers
specialize locally on young rabbits, then we can predict that
they will show stable resource use even when the abundance
of the principal resource fluctuates (between seasons and
years). The use of other secondary prey will also be relatively stable, as will diet diversity.
However, if we consider that badgers in the area are not
locally specialized (as in most of their geographic distribution; Roper 1994), when one of several resources was very
abundant, it would be a key resource for badgers whilst it
was available. When the abundance of the key resource drops,
animals should switch to other available resources (e.g.,
O’Mahony et al. 1999). Thus, with seasonal or interannual
variations in resource availability there should be fluctuations in resource use, while the use of secondary resources
should depend on the availability of the key resources (e.g.,
Hanson and Green 1989). Diet diversity should be inversely
proportional to the use of key resources and directly proportional to the use of secondary resources.
To test these alternative hypotheses we studied the diet of
badgers in the Doñana area of southwestern Spain and
recorded seasonal and interannual variation in diet, the existence of key trophic resources fitting the pattern of variation,
and trophic diversity. We also studied the availability of the
key trophic resource and one secondary resource (rabbits
and invertebrates, respectively; defined after Martín et al.
1995) and the relationship between their availability and
their use by badgers.
Methods
Study sites and badger populations
The study was carried out in Doñana National Park, Spain,
in the southwestern Iberian Peninsula. The climate is subhumid
with mild wet winters and hot dry summers. The main study
area, Coto del Rey, is located in the north of the National
Park and includes a large patch of Mediterranean scrubland
adjacent to marshland and dominated by mastic shrubs (Pistacia
lentiscus) and cork oaks (Quercus suber). The Mediterranean
scrubland is surrounded by plantations of Pinus pinea, pastureland, dehesa, and marshland (for more details see Revilla et al.
2000, 2001a). This area holds the highest density of rabbits
in the Doñana area. We also studied the diet of badgers in the
Reserva Biológica. The Reserva study site is drier (because
of a lower water table), with the vegetation dominated by
xerophytic bushes, mostly Halimium sp., Genista sp., and
Rosmarinus sp., and pine plantations (Revilla et al. 2000).
During the period of study, rabbit density in the Reserva was
far lower than in Coto del Rey (20-fold; E. Revilla, unpublished
data). The Reserva site is located west of that in Martín et al.
(1995), which was characterized by its proximity to marshland, and so by a higher density of rabbits. Badger density in
Coto del Rey was three times higher than in the Reserva
(0.67 and 0.23/km2, respectively; Revilla et al. 1999).
We studied the badgers’ diet by means of fecal analysis.
We collected feces from latrines surrounding the diurnal resting
places (setts) used by badgers (Revilla et al. 2001b). Discovery of setts was facilitated by daily location of radio-collared
badgers in both areas (44–88% of the estimated annual population in Coto del Rey and 38% of that in the Reserva were
marked during the study; Revilla et al. 1999). This also allowed most of the collected fecal samples to be assigned to
badger territories (five territories in Coto del Rey and three
in the Reserva). We took special care not to disturb badgers
during the collection of feces (which was usually done when
the badgers gave up using the sett). We collected 1258 badger
feces in Coto del Rey between January 1995 and November
1997, while in the Reserva the study period was between
March and November 1997, with a total of 114 fecal samples collected.
© 2002 NRC Canada
Table 1. Percentages of prey occurrence, dry mass, and ingested biomass in the diet of
Eurasian badgers (Meles meles) in Coto del Rey (CR) between January 1995 and October
1997 and in the Reserva Biológica (RB) between April and October 1997.
Percent
occurrence
a
Percent dry
mass
Percent biomass
Type of prey
CR
RB
CR
RB
CR
RB
Young rabbits (19.8)
Other rabbits (14.1)
Small mammals (12.1)
Carrion (55.3)
Birds (17.3b or 45c)
Reptiles (19.4)
Amphibians (15.8)
Invertebrate larvae (12.0)
Invertebrate imagoes (2.5)
Fruits (20.4)
Fungi (40.0)
Otherd
Non-nourishing material
35.7
31.2
9.6
3.3
5.9
19.9
27.2
35.2
85.5
31.1
2.4
3.5
70.0
8.8
12.3
10.5
0.9
4.4
36.8
78.1
62.3
95.6
17.5
38.6
0.9
50.0
27.0
15.2
1.2
0.5
1.0
1.6
3.2
7.2
22.9
14.2
0.6
0.6
5.0
5.0
5.1
1.1
0.1
0.3
3.0
14.4
12.2
38.7
7.3
6.9
<0.1
6.8
37.3
15.5
1.03
4.3
1.8
2.8
3.5
6.0
4.6
20.2
1.7
0.8
—
8.2
7.1
1.0
0.5
0.6
8.0
19.1
11.3
8.3
12.4
23.0
0.6
—
Note: The total number of faeces analysed was 1258 and 114 for CR and RB, respectively.
a
Numbers in parentheses show the correction factor.
b
Passerines.
c
Anseriformes.
d
We used the value for the most similar resource (e.g., carrion-correction factor for fish).
Fecal analysis
In the study of the badgers’ diet we followed the analytical
procedures used by previous authors (Kruuk 1989; Rodríguez
and Delibes 1992; Martín et al. 1995; Revilla and Palomares
2001). We present the results as percentage of occurrence,
percentage of dry mass, and estimated biomass ingested. We
calculated our own correction factors (see Table 1) for most
of the trophic resources used by badgers in Doñana National
Park during a feeding trial conducted on a captive badger
from the studied population (carried out at the Centro de
Recuperación de Fauna Silvestre of Doñana National Park,
Spanish Ministry of Environment).
We grouped food types following taxonomic criteria and
the foraging method used by badgers in their food-gathering.
We distinguished between young rabbits (classified as those
with milk teeth in their jaws) and subadult and adult rabbits
(with permanent teeth, and by comparison with a reference
collection). Badgers obtain young rabbits by digging them
out of warrens, while other age classes are only accessible as
carrion or as ill individuals during recurrent outbreaks of
myxomatosis and rabbit hemorrhagic disease (Rogers et al.
1994). We also distinguished between larvae and imagoes
of invertebrates. Larvae that are consumed (mostly of dung
beetles) live underground and so are also dug up by badgers.
Carrion was composed of ungulates and the group referred
to as “other” consisted of fish and human refuse. Other food
types are shown in Table 1. Non-nutritive material was mostly
of vegetal origin and considered to have been accidentally
ingested (it was not considered in the estimations of biomass).
Prey abundance
Rabbit abundance was estimated by line-transect sampling
(Buckland et al. 1993; Palomares et al. 2001; Revilla and
Palomares 2001) carried out in randomly selected areas within
the Mediterranean scrubland habitat. Transects ranged from
1100 to 2040 m in length and were surveyed slowly (ca.
1.5–2.5 km/h) on foot at dusk (from 15–20 min before sunset to 25–30 min after sunset), the best time of the day to
conduct rabbit censuses in Doñana National Park (Villafuerte
et al. 1993). We grouped observation distances of rabbits in
10-m intervals and truncated the distance at 50 m. Because
rabbits could move prior to detection, grouping improved
the robustness of the density estimator. The truncation distance and grouping options were decided after three pilot
samplings. Transects were always the same, did not follow
any track or road in the study area, followed a straight line,
and were always walked by the same observer. Rabbits were
counted every season from winter 1995 to autumn 1997. Rabbit densities were estimated using the program TRANSECT
(Burnham et al. 1980). For further details on methodology
and on rabbit abundance over time and between habitats see
Palomares et al. (2001) and Revilla and Palomares (2001).
During 1996 and 1997 we livetrapped rabbits to determine
the proportion of breeding females (a female was considered
breeding when lactating and (or) pregnant; for details on
rabbit livetrapping see Calzada 2000). In total we captured
129 female rabbits (Table 3; Calzada 2000). The seasonal
density of females (number per hectare, calculated by dividing the rabbit density by 2 (the sex ratio in rabbit populations is 1:1); Calzada 2000) was multiplied by the seasonal
proportion of breeding females. The value obtained was used
as an index of the number of rabbit litters per hectare. During 1996 and 1997 we also counted every young rabbit (after
emergence from the warren, n = 176) and ill rabbit (individuals affected by myxomatosis and rabbit hemorrhagic disease,
n = 71) seen in the field during standard fieldwork (i.e.,
with a constant effort throughout the year, mostly during
© 2002 NRC Canada
radio-tracking). For the sightings of young rabbits we subtracted
19 days (average time spent inside the warren; Calzada 2000)
from the sighting date to ensure that the sighting was allocated to the season of birth. We calculated the proportion of
sightings of young rabbits in each season by dividing the total for the year. Following the same procedure we calculated
an index of the number of ill rabbits in the field using the
proportion of sightings of ill rabbits.
To obtain a relative index of abundance of invertebrates
(mostly beetles) we counted the signs of activity (sand piles
produced by larvae and imagoes of dung beetles), imagoes,
and dung piles (horse dung and cow pats) in transects distributed in every habitat in Coto del Rey. Transect width was
1 m and length was divided into 50-m sections. In total we
sampled 21 different transects (for a total of 91 times) with
1430 sections (i.e., 7.17 km2), counting 9580 signs of activity, 224 imagoes, and 1249 dung piles. Transects were sampled in spring and summer 1996 and in winter (except for
those in flooded marshland), spring, and summer 1997.
Data analysis
Analysis was performed using season as the temporal unit
(winter: January to March; spring: April to June; summer:
July to September; autumn: October to December). Variations in the frequency of occurrence of each type of prey in
Coto del Rey were analyzed with multiway analysis of frequencies using general log-linear models (procedure CATMOD in
SAS; SAS Institute Inc. 1990). Thus, the unit of analysis
was badger scats and the response variable was the presence
or absence of the prey considered. We fitted the complete
models (i.e., considering all the interactions) using season
and year as independent variables. The level of significance of
post-hoc comparisons was assessed with Bonferroni correction.
To define the importance of trophic resources in the characterization of seasonal fluctuations in the badgers’ diet we
grouped seasons using an UPGMA cluster analysis (using
squared Euclidean distances with the CLUSTER procedure
in SAS; Romesburg 1990; SAS Institute Inc. 1990). The sets
of values of the proportions of consumed biomass (untransformed) of each type of prey were used as the characteristics
for defining groups for each season, year, and area of study
(considering both Coto del Rey and the Reserva). Then we
performed a multivariate analysis of variance (MANOVA,
GLM procedure in SAS with type III sum of squares) using
the grouping classification obtained in the cluster analysis as
an independent variable, and beginning with the simplest hierarchic classification (i.e., two groups) until a significant
model was obtained. As dependent variables we used the angular-transformed biomass values for each type of prey. For
defining the most important trophic resources we analysed
the univariate significant contribution of each type of prey to
the final multivariate model and the degree of partial correlation between them.
We used Shannon’s diversity index (H′ ) to estimate variations in diet diversity. A first measure was calculated using
the seasonal proportion of ingested biomass for each of the
territories where we collected feces. A diversity value was
also calculated using the proportion of biomass estimated for
each fecal sample. This second measure can be used as an
indicator of diet diversity in the short term (one or a few foraging sessions). The influence of the different types of prey
on seasonal diversity was analyzed using a multiple regression, where H′ was the dependent variable and the proportions of ingested biomass of each type of prey (after angular
transformation) were the predictors. The model was fitted
with the predictors in decreasing order, according to the
strength of correlation between the dependent variable and
each predictor, and following backward elimination of the
nonsignificant ones.
The pattern of temporal variability in rabbit abundance
has already been studied for the Coto del Rey population
during most of our study period (Palomares et al. 2001). The
three indexes of invertebrate abundance in Coto del Rey
were analyzed with general linear models (GLM procedure
in SAS). As factors we used the type of habitat, season, and
year, and as dependent variable the log-transformed index
value for each 50-m section.
Results
Diet and its variations
The resources used by badgers in Coto del Rey showed a
high degree of taxonomic diversity, which is in accordance
with the results of previous studies (Martín et al. 1995; Fedriani
et al. 1998). Animal prey ranging from scavenged horse and
cattle carcasses to dwarf shrews (Suncus etruscus), amphibians, reptiles, birds, fish, Mollusca, earthworms, scorpions,
Scolopendridae, Crustacea, Diplopoda, Isopoda, and up to
10 different orders of insects were found. Fruits, bulbs, seeds
of Graminae, and fungi were also included in the badgers’
diet, as well as small amounts of human refuse. Young rabbits were the most important prey, accounting for more than
37% of ingested biomass, followed by fruits with 20% of ingested biomass (mostly dwarf palm (Chamaerops humilis)
nuts and wild pears (Pyrus bourgeana)) and older rabbits
with 15% of biomass. Invertebrates (larvae and imagoes) accounted for about 11% of ingested biomass (Table 1).
The diet of badgers in Coto del Rey was characterized by
marked seasonality and interannual variation affecting all prey
types (Fig. 1). Consumption of young rabbits varied significantly among years and seasons (Fig. 1, Table 2). Consumption
of older rabbits, fruit, and invertebrate larvae and imagoes
also varied among years, seasons, and their interaction
(Fig. 1, Table 2).
In the Reserva badgers also had a broad trophic spectrum.
Fungi, amphibians, fruits, and invertebrates in that order were
the most important prey types (Table 1). As we only had
information for three seasons we did not analyze variability
(but see Fig. 1).
Key trophic resources
The simplest classification obtained in the cluster analysis
(two groups) was not significant (MANOVA Pillai’s trace =
0.989, P > 0.05). In the case of the next most simple classification, i.e., three groups, the multivariate model was significant (Pillai’s trace = 1.982, P < 0.01). These groups were
defined by winter and spring in Coto del Rey, winter and
spring in the Reserva, and summer and autumn in both areas.
Only 4 of the 11 types of prey considered in the analysis
were significant; in order of importance these were young
rabbits, fruits, fungi, and reptiles (F[2,11] = 53.55–4.45, P <
0.04 in all cases). The matrix of partial correlation between
© 2002 NRC Canada
Fig. 1. Proportions of biomass of the main prey types ingested by Eurasian badgers (Meles meles) in Coto del Rey, Doñana National
Park, Spain (a), and the Reserva Biológica (b) according to season (winter (WI), spring (SP), summer (SU), and autumn (AU)) between
1995 and 1997. The numbers above the graph show the number of feces analyzed for each season.
(a)
Proportion of biomass
1.0
156
151
182
115
137
73
WI
SP
SU
AU
WI
SP
110
62
117
55
89
WI
SP
SU
0.8
0.6
0.4
0.2
0.0
1995
SU
1996
AU
1997
(b)
Proportion of biomass
48
21
1.0
27
Young rabbits
Other rabbits
Fruits
Invertebrate imagos
Invertebrate larvae
Other
0.8
0.6
0.4
0.2
0.0
WI
SP
SU
1997
Table 2. Results of log-linear models for the frequencies of occurrence of prey types in the diet of badgers in Coto del Rey, Doñana
area, Spain, for the effect of year, season, and their interaction.
Years
χ[2 ]
2
Seasons
Interaction
P
χ [23]
P
χ [26]
P
Maximum-likelihood contrasts
1995 < > 1996, significant differences between all
seasons
1995 < > 1996, summer different from the rest,
winter < > spring
summer < > autumn
1996 < > 1995 = 1997, winter < > summer = autumn
1995 < > 1996 < > 1997, summer different from the rest
summer = autumn < > winter = spring
1995 < > 1996 = 1997, summer different from the rest
Young rabbits
10.81
<0.005
148.01
<0.0001
31.71
<0.0001
Other rabbits
38.41
<0.0001
131.85
<0.0001
34.53
<0.0001
Fruits
Amphibians
Invertebrate larvae
Invertebrate imagoes
Reptiles
Birds
Fungi
6.49
69.28
92.75
0.98
20.46
1.82
6.79
0.0389
<0.0001
<0.0001
0.611
<0.0001
0.4033
0.0335
118.94
44.55
36.55
33.89
58.94
1.62
7.31
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
0.656
0.0627
56.07
24.28
20.69
18.69
26.85
19.80
6.36
<0.0001
<0.0001
0.0021
0.0047
0.0002
0.003
0.3842
© 2002 NRC Canada
Fig. 2. (a) Variations in diet diversity (calculated with Shannon’s diversity index, H′ ) in Coto del Rey (•, average (±1 SE) seasonal
proportion of ingested biomass for the territories where we collected feces; 0, average (±1 SE) proportion of biomass estimated for
each faecal sample (for the sample size see Methods and Fig. 1). (b) Relationship between H′ (estimated seasonally for the different
badger territories) and percent biomass of invertebrate imagoes (•) and young rabbits (0), with their respective trend lines.
(a)
2.0
Diet diversity
1.5
1.0
0.5
0.0
WI
SP
SU
AU
WI
1995
SP
SU
AU
WI
1996
SP
SU
AU
1997
Season and year
(b)
2.0
Diet diversity
1.5
1.0
0.5
0.0
0
20
40
60
80
Percent biomass
dependent variables a showed significant correlation between
young and older rabbits (R = –0.72, P < 0.01) and between
fruits and invertebrate imagoes (R = –0.73, P < 0.01). These
results indicate that in Coto del Rey, rabbits characterize the
diet of badgers during winter and spring, and in the Reserva,
fungi characterize the diet of badgers during the same seasons and fruits, reptiles, and invertebrate imagoes during
summer and autumn.
Trophic diversity
The general linear model for the trophic-diversity value
calculated for badger territories in Coto del Rey, with season
and year as independent variables, was significant (F[11,22] =
5.33, P < 0.001; Fig. 2). Main effects of year and season
were significant (F[2] = 9.36, P = 0.001, and F[3] = 8.19, P <
0.001, respectively), but not the interaction. Diversity was
lower in 1996 than in 1995 (least-squares means 0.90 and
1.38, respectively; Fisher’s protected least significant difference (LSD), P < 0.001). H′ during summer was significantly
larger than during winter and spring (least-squares means
1.46, 0.98, and 0.95, respectively; LSD, P < 0.001).
In the analysis of the influence of prey type on seasonal
diversity we included invertebrate imagoes, young rabbits,
invertebrate larvae, reptiles, and older rabbits (in that order)
© 2002 NRC Canada
Revilla and Palomares
Table 3. Rabbit densities (number/ha), percentages of breeding females (pregnant, lactating, or both), and percentages of young and ill (mostly myxomatous) rabbits over the year
in the Mediterranean scrubland in Coto del Rey, and the total number of rabbits used in
calculating the percentages.
Season and
year
Rabbit
densitya
Percentage
of breeding
femalesb
Percentage of
young rabbits
seen per year
Percentage of
ill rabbits seen
per year
Winter 1995
Spring 1995
Summer 1995
Autumn 1995
Winter 1996
Spring 1996
Summer 1996
Autumn 1996
Winter 1997
Spring 1997
Summer 1997
Autumn 1997
Total
29.30
39.35
21.20
7.00
5.20
12.45
3.30
2.00
8.00
21.26
3.46
4.01
—
—
—
—
—
85.7
70.0
0.0
68.8
69.2
42.4
20.0
27.8
129
—
—
—
—
20.7
66.6
4.5
8.1
43.1
55.4
0.0
1.5
176
—
—
—
—
0.0
0.0
100
0.0
8.7
87.0
4.4
0.0
71
a
Data for 1995 and 1996 are extracted from Palomares et al. (2001).
For more information see Calzada (2000).
b
in the multiple-regression model (Spearman’s rank correlation with H′ was positive for all, between R = 0.70 and R =
0.32, except for young rabbits, which was negative: R =
–0.64). From this initial model, only invertebrate imagoes
and young rabbits remained in the final model (Fig. 2),
which took the form H′ = 0.026 arcsine(invertebrate imagoes) –
0.007 arcsine(young rabbits) + 1.047(adjusted R2 of the final
model = 0.65, F[2,31] = 31.90, P < 0.001).
The general linear model for the trophic-diversity value
calculated with the percentages of biomass of each prey type
in faeces was significant (F[11,1242] = 10.07, P < 0.0001).
Season and year had significant effects (F[2] = 21.70, P <
0.0001, and F[2] = 9.77, P < 0.0001, respectively), but not
their interaction. Intra-faeces diversity was higher in 1995
than in 1996 and 1997 (0.50 versus 0.39 and 0.43, respectively; P < 0.023). Values in winter and spring did not differ
significantly from each other, while both differed from summer and autumn, and finally, summer and autumn differed
from each other (H′ = 0.38, 0.34, 0.56, and 0.48, respectively, for winter, spring, summer, and autumn; LSD, P <
0.045 in all cases).
To compare trophic diversity in Coto del Rey and the
Reserva we repeated the same analysis for winter, spring,
and summer 1997, but included territories in the model to
control local variability within the two subpopulations (F[14,298] =
7.81, P < 0.0001). Territory and its interaction with season
had significant effects (F[4] = 13.01, P < 0.0001, and F[8] =
5.04, P < 0.0001, respectively), but not the main effect of
season (F[2] = 1.51, P = 0.2225). Territories in Coto del Rey
were separated from those in the Reserva in all cases (range
0.37–0.52 in Coto del Rey and 0.77–0.88 in the Reserva;
LSD, P < 0.015 in all comparisons). The interaction was
due to different patterns of seasonal diversity for territories
in both areas: in Coto del Rey trophic diversity increased
significantly from winter to summer, while in the Reserva,
trophic diversity was significantly higher during winter than
during spring and summer.
Key prey: rabbit availability and use
Rabbit abundance varied seasonally, with maximum densities in spring and minimum densities in summer–autumn
(Palomares et al. 2001; Table 3). There was a reduction in
rabbit numbers after the winter of 1996. Reproduction occurred mostly in winter and spring and disease outbreaks in
spring 1997 and summer 1996 (Table 3).
Grouping data by season, and using the same bibliographic
data on rabbit reproductive activity as Martín et al. (1995)
for measuring young rabbit availability, we were unable to
find any relationship with total ingested biomass of rabbits
(i.e., grouping together young and older rabbits) or with ingested biomass of young rabbits only (R = 0.20, P = 0.56,
and R = 0.02, P = 0.95, respectively). Nor was there a relationship between rabbit density and total biomass of rabbits,
biomass of young rabbits, and biomass of non-young rabbits
in Coto del Rey (Figs. 3a–3c).
We also compared rabbit consumption with more direct
measures of rabbit availability: the proportion of young and
ill (myxomatous) rabbits sighted per season and an index of
the density of rabbit litters (number per hectare). In these
cases there were clear relationships between the estimated
resource availability and the amount of resource used, which
explained 65–71% of the variance in the case of young rabbits and 87% in older rabbits (for linear regressions see
Figs. 3d–3f ). For young rabbits, the data showed a good fit
with a sigmoidal function (adjusted R 2 = 0.98, F[3,6] > 45.25,
P < 0.005, for both measures of availability), showing that
badgers switch to young rabbits when densities exceed 1.5
rabbit litters per hectare (Figs. 3e and 3f ).
Secondary prey: invertebrate availability and use
We compared the abundance of signs of activity and dung
piles (both the observed and the least-squares means predicted by GLMs) with seasonal consumption of invertebrate
imagoes, invertebrate larvae, and the two summed, and compared the abundance of beetles with consumption of inverte© 2002 NRC Canada
Fig. 3. Relationship between rabbit biomass ingested by badgers and different measures of rabbit availability in Coto del Rey, on a
seasonal basis. (a) Total ingested rabbit biomass fitted to rabbit density during 1995, 1996, and winter, spring, and summer 1997.
(b) As a, but considering only young-rabbit ingested biomass. (c) As a, but considering only non-young-rabbit ingested biomass.
(d) Non-young-rabbit ingested biomass in relation to the proportion of sightings of ill rabbits during 1996 and winter, spring, and summer 1997. (e) Proportion of young-rabbit ingested biomass in relation to the density of rabbit litters. (f) As e, but considering the proportion of sightings of young rabbits. Proportions were angular-transformed.
(a)
(b)
1.4
Arcsine (% of young
rabbit biomass)
Arcsine (% of total
rabbit biomass)
1.4
1.2
1.0
0.8
0.6
0.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
0.2
0
10
20
30
0
40
10
No. of rabbits/ha
Arcsine (% of non-young
rabbit biomass)
Arcsine (% of non-young
rabbit biomass)
(c)
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
10
20
30
40
(d)
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0.0
50
Arcsine (% of young
rabbit biomass)
Arcsine (% of young
rabbit biomass)
(e)
1
2
3
4
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
Adj. R2 = 0.774, F[1,5]=21.550 P= 0.006
Y= 0.136 +0.301 · X
Adj. R2 = 0.162, F[1,9]=2.938 P= 0.121
0
0.2
Arcsine (% of mixomatose sightings)
No. of rabbits/ha
1.2
1.1
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
40
No. of rabbits/ha
0.8
0
30
2
Adj. R = -0.028, F[1,9]=0.728 P= 0.416
2
Adj. R = 0.290, F[1,9]=5.032 P= 0.051
0.9
20
5
No. of rabbit litters/ ha
2
Adj. R = 0.721, F[1,5]=16.472 P= 0.010
Y= 0.202 + 0.215 · X
brate imagoes. In none of these comparisons was there a
significant correlation (R < 0.69, P > 0.196 in all cases).
However, the three indexes of invertebrate abundance showed
variation between habitats, seasons, and years (F > 4.40, P <
(f)
1.2
1.1
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0.0
0.2
0.4
0.6
0.8
1.0
Arcsine (% of young rabbit sightings)
Adj. R 2 = 0.724, F[1,5]=16.710 P= 0.009
Y= 0.0913 +1.136 · X
0.0001, for the three models). Signs of activity were more
abundant in Mediterranean scrubland, pine plantations, and
pastureland, while values were lowest in marshland (F[5,63] =
19.83, P < 0.0001). Seasonal abundance reached a maximum
© 2002 NRC Canada
in winter and a minimum in summer (F[2] = 3.51, P = 0.036).
There was no difference between years, and interactions
were not significant (F < 2.05, P > 0.157, in all cases). Beetles
were more abundant in Mediterranean scrubland and pine
plantations (F[5,82] = 7.42, P < 0.0001), especially during
spring and summer (F[5] = 13.02, P < 0.0001), with lowest
values during winter. Year had no significant effect in the
model (F[1] = 1.17, P = 0.282). Abundance of dung piles
varied between seasons, years, habitat types, and the season
by habitat interaction (F > 2.70, P < 0.04, in all cases).
Dehesa, pastureland, and pine plantations were the habitats
with the highest density of dung piles. The seasonal maximum occurred in winter and the minimum during summer,
except in marshland, where the maximum also occurred in
summer. The density of dung piles was significantly lower
in 1996 than in 1997.
Discussion
According to the hypothesis of local specialization, consumption of the main prey should be relatively stable, without significant variation in the use of resources over time
(Kruuk and de Kock 1981; Kruuk and Parish 1981; Martín et
al. 1995). In the case of badgers in Coto del Rey there is
strong seasonal and interannual variation in the consumption
of not only young and older rabbits but also of other resources. The role of young rabbits as the prey that characterizes the diet of badgers in Doñana National Park was
corroborated for winter and spring in Coto del Rey, while
fungi were the most characteristic food in the same seasons
in the Reserva, fruits during autumn in Coto del Rey, and
reptiles during summer in both areas. Therefore, the apparent role of young rabbits as the only key resource in Coto
del Rey over the whole study period (as can be interpreted
by collapsing Table 1) is the result of grouping together seasons and years when the main resources differed.
The variations in the use of resources were followed by
similar variations in seasonal diet diversity (Fig. 2). Consistent with the hypothesis of a lack of local specialization, diet
diversity was inversely related to the use of the main resource (young rabbits) and directly related to the use of a
secondary prey (invertebrate imagoes). This pattern of variation is also apparent at a smaller temporal scale (intra-faeces
diversity), which is probably a reflection of individual behavior during one or a few foraging bouts. Furthermore, significant differences in diversity between Coto del Rey and
the Reserva during the same seasons show that even at small
spatial scales (the two areas are separated by only 8 km), the
functional responses of badgers can vary according to local
conditions.
Consumption of both age classes of rabbits in Coto del
Rey can be predicted from their availability in field. In the
case of young rabbits, the relationship between consumption
and abundance resembles a prey switch, with a type 3 functional response (Figs. 3e and 3f ), which is characteristic of
generalist predators for which alternative prey are available
(Murdoch and Oaten 1975). This prey switch would explain
the lack of a relationship between the use of invertebrates
and their abundance in the field, and the inverse relationship
between consumption of young rabbits and diet diversity.
Therefore, we may conclude that badgers in Doñana National Park responded to temporal variations in the availability of the main trophic resource as we may expect for a
generalist species. Nevertheless, badgers in this area have
been previously regarded as specializing locally on young
rabbits, owing to the lack of correlation between rabbit use
(badgers consume rabbits in the summer also) and an index
of rabbit availability taken from the literature (Martín et al.
1995). Given that rabbits can breed year-round whenever
protein-rich grass is available (Rogers et al. 1994), and therefore in wet summers, we may expect that badgers ate rabbits
during summer in years when the rabbit breeding season was
prolonged from spring to summer and breeding began earlier
in autumn. Consequently, the results obtained by Martín et
al. (1995) could also be interpreted as a temporal functional
response (i.e., consumption of rabbits in summer) because
the availability of rabbits was not measured during their
study. Additionally, when we used the same bibliographic
data on rabbit abundance (which represents population density
including adult rabbits) as Martín et al. (1995), we failed to
detect any correlation. We detected the relationship between
resource use and availability only after using a clear definition of both age classes of rabbits and their actual availability in the field. Badgers can use only two fractions of the
rabbit population, one defined as litters (young rabbits before emergence) and the other as ill subadult and adult rabbits, and only when they are available in the field. Therefore,
defining prey types correctly from the point of view of the
predator is very important.
Independently of the degree of specialization, animals have
a rank of preferences among their prey. Whenever possible,
they maximize the use of preferred prey, mostly by means of
behavioral responses (such as using search images, switching
habitat use, or changing foraging tactics; e.g., O’Donoghue
et al. 1997, 1998a, 1998b). Thus, by interpreting behavioral
flexibility and preference ranking as effects of local specialization we confound optimization models on foraging with
theoretical evolutionary models on feeding specialization
(e.g., Fedriani et al. 1998). The first deal with the choices
made by individuals among a spectrum of alternative prey
(Stephens and Krebs 1986), while the second analyze why
that spectrum includes those particular prey instead of being
wider or narrower (e.g., Wilson and Yoshimura 1994). Optimalforaging models can only be taken to predict the evolution of
an optimal diet (i.e., behavioral specialization) by supposing
that the optimal decision becomes genetically fixed (Futuyma
and Moreno 1988). In contrast, models that include genotypespecific resource preferences and fitness may differ from
models of optimal-foraging theory, such as those that predict
specialization on a sufficiently abundant but less profitable
resource (Futuyma and Moreno 1988). As we stated above,
behavioral flexibility is more marked in trophic generalists,
while a ranking of preferences among prey types is common
to most predators (Begon and Mortimer 1986; Begon et al.
1990).
In summary, populations (e.g., Thompson 1998), and even
individuals (e.g., Beaudoin et al. 1999), may specialize locally on one or a few preferred prey. However, to constitute
a specialization the pattern has to be consistent throughout
time and (or) the life of individuals, despite variability in
prey abundance. This local or individual behavioral special© 2002 NRC Canada
ization might also occur in carnivores (Kruuk 1986; Vargas
and Anderson 1998, 1999). However, misinterpretation of
the behavioral flexibility of predators and the use of shortterm studies without considering temporal variability can
lead to a false impression of local specialization. Animals
may have only one prey type available, making the distinction even more difficult. This is especially true for mammals
because their high behavioral flexibility in terms of alternative prey (and their spatiotemporal fluctuations) makes it
difficult to test many of the predictions of theoretical models
(e.g., O’Donoghue et al. 1997, 1998a, 1998b; O’Mahony et
al. 1999), and means that the existence of local specialization is very unlikely. The distinction between specialist and
generalist predators is important for interpreting their role in
many ecological processes (such as predator–prey relationships or population and community dynamics). In the case
of the badger, its complex variation in social organization is
generally explained by the key importance of trophic resources and their variations in spatiotemporal availability
(reviewed in Woodroffe and Macdonald 1993). Therefore,
care should be taken when referring to a population as locally specialized without making an adequate contrast between
predictions.
Acknowledgements
This research was founded by the Direción General de
Investigación Científica y Técnica and the Direcíon General
de Enseñanza Superior (projects PB94-0480 and PB97-1163)
and sponsored by Rover España. E.R. was supported by a
predoctoral grant from the Spanish Ministry of Education
and Culture. Fieldwork was conducted with the permission of
Doñana National Park (Spanish Ministry of Environment) and
Agencia de Medio Ambiente (Junta de Andalucía). The Centro
de Recuperación de Fauna Silvestre, Doñana National Park,
allowed and helped with the feeding trial. C. Arcocha, J. Ayala,
A. Devenoges, J. Calzada, G. Chapron, K. Elsner, N. Fernández,
A. Flores, G. Lariccia, M.A. López, A. Meijers, J.M. Pérez,
J.C. Rivilla, C. Rodríguez, A. de Roos, and V. Salvatori helped
with field and (or) laboratory work. Lively discussions with
and (or) comments from B. Blake, J. Calzada, M. Delibes,
J. Goszczyñski, R. Delahay, J.M. Fedriani, N. Fernández,
P. Ferreras, K. Moloney, and an anonymous reviewer improved
the content of previous drafts.
References
Beaudoin, C.P., Tonn, W.M., Prepas, E.E., and Wassenaar, L.I.
1999. Individual specialization and trophic adaptability of northern
pike (Esox lucius): an isotope and dietary analysis. Oecologia,
120: 386–396.
Begon, M., and Mortimer, M. 1986. Population ecology: a unified
study of animals and plants. 2nd ed. Blackwell Scientific Publications, Oxford.
Begon, M., Harper, J.L., and Townsend, C.R. 1990. Ecology: individuals, populations and communities. 2nd ed. Blackwell Scientific Publications, Oxford.
Buckland, S.T., Anderson, D.R., Burnham, K.P., and Laake, J.L.
1993. Distance sampling. Chapman and Hall, London.
Burnham, K.P., Anderson, D.R., and Laake, J.L. 1980. Estimation
of density from line transect sampling of biological populations.
Wildl. Monogr. No. 72.
Calzada, J. 2000. Selección de presa e impacto de depredación del
lince ibérico y el zorro sobre el conejo en el Parque Nacional de
Doñana. Ph.D. thesis, University of León, León, Spain.
Ciampalini, B., and Lovari, S. 1985. Food habits and trophic niche
overlap of the badger (Meles meles L.) and the red fox (Vulpes
vulpes L.) in a Mediterranean coastal area. Z. Saeugetierkd. 50:
226–234.
Fedriani, J.M., Ferreras, P., and Delibes, M. 1998. Dietary response
of the Eurasian badger, Meles meles, to a decline of its main
prey in the Doñana National Park. J. Zool. (Lond.), 245: 214–
218.
Feinsinger, P., Spears, E.E., and Poole, R.W. 1981. A simple measure of niche breadth. Ecology, 62: 27–32.
Fietz, J., and Ganzhorn, J.U. 1999. Feeding ecology of the hibernating primate Cheirogaleus medius: how does it get so fat?
Oecologia, 121: 157–164.
Futuyma, D.J., and Moreno, G. 1988. The evolution of ecological
specialization. Annu. Rev. Ecol. Syst. 19: 207–233.
Hanson, J., and Green, L. 1989. Foraging decisions: prey choice by
pigeons. Anim. Behav. 37: 429–443.
Kauhala, K., Laukkanen, P., and von Rége, I. 1998. Summer food
composition and food niche overlap of the raccoon dog, red fox
and badger in Finland. Ecography, 21: 457–463.
Kruuk, H. 1986. Interaction between Felidae and their prey species: a review. In Cats of the world: biology, conservation and
management. Edited by S.D. Miller and D.D. Everett. National
Wildlife Federation, Washington, D.C. pp. 353–374.
Kruuk, H. 1989. The social badger. Oxford University Press, Oxford.
Kruuk, H., and de Kock, L. 1981. Food and habitat of badgers
(Meles meles L.) on Monte Baldo, northern Italy. Z. Saeugetierkd.
46: 295–301.
Kruuk, H., and Parish, T. 1981. Feeding specialization of the European
badger Meles meles in Scotland. J. Anim. Ecol. 50: 773–788.
Long, C.A., and Killingley, C.A. 1983. The badgers of the world.
Charles C. Thomas, Springfield, Ill.
Lüps, P., and Wandeler, A.I. 1993. Meles Brisson, 1762. In Handbuch
der Säugetiere Europas. Edited by J. Niethammer and F. Krapp.
Aula-Verlag, Wiesbaden, Germany. pp. 855–906.
Martín, R., Rodríguez, A., and Delibes, M. 1995. Local feeding
specialization by badgers (Meles meles) in a Mediterranean environment. Oecologia, 101: 45–50.
Mellgren, R.L., and Roper, T.J. 1986. Spatial learning and discrimination of food patches in the European badger (Meles meles L.).
Anim. Behav. 34: 1129–1134.
Murdoch, W.W., and Oaten, A. 1975. Predation and population stability. Adv. Ecol. Res. 9: 1–131.
Neal, E., and Cheeseman, C. 1996. Badgers. T. & A.D. Poyser
Natural History, London.
O’Donoghue, M., Boutin, S., Krebs, C.J., and Hofer, E.J. 1997.
Numerical responses of coyotes and lynx to the snowshoe hare
cycle. Oikos, 80: 150–162.
O’Donoghue, M., Boutin, S., Krebs, C.J., Murray, D.L., and Hofer,
E.J. 1998a. Behavioral responses of coyotes and lynx to the
snowshoe hare cycle. Oikos, 82: 169–183.
O’Donoghue, M., Boutin, S., Krebs, C.J., Zuleta, G., Murray, D.L.,
and Hofer, E.J. 1998b. Functional responses of coyotes and lynx
to the snowshoe hare cycle. Ecology, 79: 1193–1208.
O’Mahony, D., Lambin, X., MacKinnon, J.L., and Coles, C.F. 1999.
Fox predation on cyclic field vole populations in Britain.
Ecography, 22: 575–581.
Palomares, F., Delibes, M., Revilla, E., Calzada, J., and Fedriani,
J.M. 2001. Spatial ecology of the Iberian lynx and abundance of
in European rabbits in southwestern Spain. Wildl. Monogr.
No. 148. pp. 1–36.
© 2002 NRC Canada
Partridge, L., and Green, P. 1985. Intraspecific feeding specializations and population dynamics. In Behavioural ecology: ecological consequences of adaptive behaviour. Edited by R.M. Sibly
and R.H. Smith. Blackwell Scientific Publications, Oxford.
pp. 207–226.
Pigozzi, G. 1988. Diet of the European badger (Meles meles L.) in
the Maremma Natural Park, central Italy. Mammal Rev. 18: 73–
75.
Revilla, E., and Palomares, F. 2001. Differences in key habitat use
between dominant and subordinate animals: intraterritorial dominance payoffs in Eurasian badgers? Can. J. Zool. 79: 165–170.
Revilla, E., Delibes, M., Travaini, A., and Palomares, F. 1999.
Physical and population parameters of Eurasian badgers (Meles
meles L.) from Mediterranean Spain. Z. Saeugetierkd. 64: 269–
276.
Revilla, E., Palomares, F., and Delibes, M. 2000. Defining key habitats for low density populations of Eurasian badgers in Mediterranean environments. Biol. Conserv. 95: 269–277.
Revilla, E., Palomares, F., and Delibes, M. 2001a. Edge–core effects and the effectiveness of traditional reserves in conservation: Eurasian badgers in Doñana National Park. Conserv. Biol.
15: 148–158.
Revilla, E., Palomares, F., and Fernández, N. 2001b. Den characteristics and selection by Eurasian badgers in a low density area.
J. Zool. (Lond.), 255: 291–299.
Rodríguez, A., and Delibes, M. 1992. Food habits of badgers (Meles
meles) in an arid habitat. J. Zool. (Lond.), 227: 347–350.
Rogers, P.M., Arthur, C.P., and Soriguer, R.C. 1994. The rabbit in
continental Europe. In The European rabbit: the history of a
successful colonizer. Edited by H.V. Thompson and C.M. King.
Oxford University Press, Oxford. pp. 22–63.
Romesburg, H.C. 1990. Cluster analysis for researchers. Robert E.
Krieger Publishing Co., Malabar, Fla.
Roper, T.J. 1994. The European badger Meles meles: food specialist or generalist? J. Zool. (Lond.), 234: 437–452.
Roper, T.J., and Mickevicius, E. 1995. Badger Meles meles diet: a
review of literature from the former Soviet Union. Mammal Rev.
25: 117–129.
SAS Institute Inc. 1990. SAS/STAT user’s guide. Vols. 1 and 2,
release 6.03. SAS Institute Inc., Cary, N.C.
Skoog, P. 1970. The food of the Swedish badger Meles meles L.
Viltrevy, 7: 1–120.
Stephens, D.W., and Krebs, J.R. 1986. Foraging theory. Princeton
University Press, Princeton, N.J.
Thompson, J.N. 1998. The evolution of diet breadth: monophagy
and polyphagy in swallowtail butterflies. J. Evol. Biol. 11: 563–
578.
Vargas, A., and Anderson, S. 1998. Ontogeny of black-footed ferret predatory behavior towards prairie dogs. Can. J. Zool. 76:
1696–1704.
Vargas, A., and Anderson, S.H. 1999. Effects of experience and
cage enrichment on predatory skills of black-footed ferrets (Mustela
nigripes). J. Mammal. 80: 263–269.
Villafuerte, R., Kufner, M.B., Delibes, M., and Moreno, S. 1993.
Environmental factors influencing the seasonal daily activity of
the European rabbit (Oryctolagus cuniculus) in a Mediterranean
area. Mammalia, 57: 341–347.
Ward, R.M.P., and Krebs, C.J. 1985. Behavioural responses of lynx
to declining snowshoe hare abundance. Can. J. Zool. 63: 2817–
2824.
West-Eberhard, M.J. 1989. Phenotypic plasticity and the origins of
diversity. Annu. Rev. Ecol. Syst. 20: 249–278.
Wilson, D.S., and Yoshimura, J. 1994. On the coexistence of specialist and generalist. Am. Nat. 144: 692–707.
Woodroffe, R., and Macdonald, D.W. 1993. Badger sociality—
models of spatial grouping. Symp. Zool. Soc. Lond. No. 65.
pp. 145–169.
© 2002 NRC Canada