ANDINTRASPECIFIC
INTERSPECIFIC
SOCIALINTERACTIONS
AMONGBROWN
INAN ENCLOSURE
BEARSANDWOLVES
PAUL KOENE, Ethology Group, Department of Animal Sciences, Wageningen University, PO Box 338, 6700 AH Wageningen, the
Netherlands,email:[email protected]
JENTINAARDESCH,EthologyGroup,Departmentof AnimalSciences, WageningenUniversity,PO Box 338, 6700 AHWageningen,
the Netherlands
ANNETTE LUDRIKS, Ethology Group, Department of Animal Sciences, Wageningen University, PO Box 338, 6700 AH Wageningen,
the Netherlands
EGBERT URFF, Ethology Group, Department of Animal Sciences, Wageningen University, PO Box 338, 6700 AH Wageningen, the
etho.vh.wau.nl
Netherlands,email:egberrt.urff@
LUDGERWENZELIDES,
EthologyGroup,Departmentof AnimalSciences, WageningenUniversity,PO Box 338, 6700 AH
Wageningen,the Netherlands
VERENA WITTENBERG, Ethology Group, Department of Animal Sciences, Wageningen University, PO Box 338, 6700 AH
Wageningen, the Netherlands
Abstract: We investigatedbear-bear,wolf-wolf, and bear-wolf interactionsin a 2-ha enclosure, which was occupied by 13 brown bears (Ursus
arctos) and a wolf (Canis lupus) pack. The hierarchyof bears and wolves was determinedby behavioralobservationand rankorderanalysis. All
behavioralinteractionsbetween bears and wolves were systematicallyvideotapedfor 1 month and analyzed.Bears and wolves had a hierarchical
organizationsimilarto thatfound in nature,althoughwe found a relativelylow position in the rankorderof blind bears.Social interactionsbetween
to be veryseriousat times.Bear-wolfinteractions
weremostlyplayful,butweresometimesagonistic.Youngbearsweremore
wolvesappeared
thanotherbearsandweresometimesseriouslybitten.However,no deathswererecorded,
to descriptions
oftenvictimsof wolfharassment
contrary
from naturalbear-wolf interactions. We concluded thatbears and wolves could be kept safely togetherin a single enclosure with apparentlyfew
consequences.Interactionswe observedwere comparableto interactionsbetween bearsand wolves undernaturalconditionsand concernedmainly
competitionover food or a den (wolf or bear).The size of the food (largepreycarcassesin the wild vs. chickenin the enclosure)probablyplays a role
in theseverityandtheeventualsolutionof thebear-wolfconflict.
Ursus 13:85-93 (2002)
Key words: bear-wolf interactions,brownbear,Canis lupus, enclosure, Ursus arctos, wolf
Bear researchin enclosureshas advantagesand disadvantagesover bearresearchin naturalconditions.Disadvantages include the very small area, compared with
naturalhome ranges,andthe absenceof escape possibilities. These factors can increase social stress even in animals kept in largerenclosures. Advantagesof relatively
largeenclosureslie in thepossibilitiesof intensiveresearch
concerninganimalbehavior,welfare, and health.
Bear behavior varies considerablyamong individuals
(Fagen and Fagen 1996, Koene 1998). This can be demonstratedto the publicin a bearenclosureandcan be used
to illustratethe individualdistinctivenessandpersonality
of bears(FagenandFagen 1996), togetherwith theircognitive abilities. Thus, enclosures are importantin public
education. Furthermore,behavioral and veterinary research in enclosures can add to knowledge acquiredin
nature,as for example, testing radiotelemetryequipment
in relationto activity.Also, the studyof behavioralchanges
relativeto environmentalconditionscan be relatedto the
welfare of the animals;for example,a decreasein stereotypic behavioror an increasein play behaviormay indicate a better relationship between a bear and its
environmentor increased well being (Koene 1998). In
this paper we present an analysis of social interactions
between brownbears and wolves, coexisting in the Bear
Forest enclosure in the OuwehandZoo in Rhenen, the
Netherlands.
We addressedthe following questionsin this paper:(1)
How arebearsandwolves socially organizedin the 'Bear
Forest' enclosure of the OuwehandZoo, in Rhenen, the
Netherlands?(2) Do bearsand wolves interact?(3) How
does the behavior of bears and wolves in the enclosure
compareto free-ranginganimals? and (4) What are the
managementimplicationsfor similarsituationsin captivity?
STUDYAREA
Bear Forest, a refuge for bears, was created in the
OuwehandZoo in Rhenen, the Netherlands,in 1993 by
the InternationalBear Foundation(Koene 1998). The 2ha area is enclosed by a 3.5-m high double fence (3 m
apart)electrified on top with a 1-m electric fence inside
the double fence. The area is bisected into northernand
southernsectionsby a publicwalkwaythroughthe enclosure.Tunnelsunderthe walkwayconnectthe northernand
southernsections. The refuge houses former zoo bears,
circusbears,Turkishdancingbears,andbearsfromBosnia
relocatedbecauseof the war.In 1994, 3 wolves fromBelgium and Germany were moved to the Bear Forest in
Rhenen along with 7 wolves from the OuwehandZoo.
Since then interspecific and intraspecificbehavioralinteractionsamong bearsand wolves have been noted, but
not systematicallyinvestigated.
86
Ursus 13:2002
ThirteenEuropeanbrownbears(6 male:7female)of
differentages(range3-23 years)residedin theenclosure
(Table1). Threeof thebearswereblinddueto previous
mistreatment.
Yearsof birthwereknownforallbearsexthe
blind
anda veterinarian
bears
estimatedtheirapcept
in
We
3
1993.
proximate
age
designatedageclasses:young
(0-4 years),adult(5-10 years)andold(>10yearsof age).
Bearswereindividually
bytheirposture,color,
recognized
andfacialdetails.
Inadditionto the 13bears,7 wolveslivedin theenclosureatthetimeof observation
(Table1).Eachwolf was
facial
and
All male
bodycharacteristics.
distinguished
by
andallfemaleswere
animalsexceptCazarwerecastrated
sterilized.Thebearsandwolveslivedoutdoorsall year
round;thebearshadunimpededaccessto 8 dens,while
the wolvesdug theirown den.Normallythe bearsand
hours.During
wolveswerefed 3 timesdailyat irregular
thefirstfeeding,onlydogpelletsweregiven.Thesecond
andthirdfeedingconsistedmainlyof apples,bread,cabbage,andcarrotsfor the bearsandrats,chicken,ducks,
fishes,oroffalforthewolves.Therewere2 feedingplaces,
thefoodfromthewalkway.
distributed
andthecaretakers
METHODS
Bear-bearInteractions
Weobservedbearsatfeedingtimesduring13-21May
1997.Observations
began0.5 hrbeforeandended1.5hr
recordedagonisafterfoodwasoffered.We(2 observers)
whichwe definedas anysituationwhere
tic encounters,
>2 bearsinteractedwitheachotherin sucha way as to
disrupttheir ongoing patternsof moving or feeding
Table1. Brownbearsandwolvesresidinginthe BearForest,
OuwehandZoo, Rhenen,the Netherlands,for a behavioral
studyconductedin 1997.
Species
Name
Label
Sex
Age(yr)
M
23
Ba
Battir
22
M
Ge
Geert
M
Ko
>1Ob
Koroglf
7
M
Mk
Mackenzie
Bo
M
Bora
5-10b
4
M
Ax
Axel
F
23
Ne
Nelly
F
Fi
Fiona
5-10
F
8
Ma
Mascha
5
F
Ni
Niki
4
F
To
Tory
4
F
Wo
Wolke
3
F
Bj
Bjorna
7
Ca
M
Wolf
Cazar
4
M
Tr
Traan
4
M
St
Streep
4
M
Po
Poot
4
F
Br
Brenda
4
F
No
Noeska
F
4
Sv
Svlvia
a Wolves were not categorizedby age class.
b Blind bear;age estimatedby a veterinarian.
Bear
Age class
old
old
old
adult
adult
young
old
adult
adult
adult
young
young
young
adult
adult
adult
adult
adult
adult
adult
(StonorovandStokes1972).A bearwas labeleddominant(DOM)whenit approached,
andmaintained
charged,
a frontalorientation
to anotherbear.A bearwaslabeled
subordinate
(SUB)whenit ranor walkedaway,backed
and
had
a
lateralorientation
to anotherbear.Wecalup,
culated2 measuresof rankorder.TheDS (dominancesubordinate)rankorderwas based on reorderingthe
dominance-subordinate
matrix(Martin
andBateson1993)
usingthe MatManprogram(De Vries1994).The DVrankorderwasbasedon theindividualdominancevalue
(DV;Lehner1996)of eachanimal(No. DOMencounters/total[No.DOM+ No. SUB]encounters).
Wolf-wolfInteractions
Weobservedwolvesduring3-28 February1997.We
normallyobservedfrom0800-1700, Monday-Friday,
timewas
usinga Latinsquaredesign.Totalobservation
hrs.
We
data
144
collected by behaviorsampling(Martin
andBateson1993)on a portablerecorderwithTheObserversoftware(Noldus1991).Wedistinguished
22 behavioralelementsthatoccurred
duringsocialinteractions
basedon thewolf sociogramfromZimen(1982).Actor,
ontape.Activesubbehavior,andreceiverwererecorded
missionappearedto be thebestindicationof rankorder
accordingto our analysisandVanHooff andWensing
(1987).Thus,we basedrankorderonly on this behavis a behavioral
ioralelement.Activesubmission
complex
in whichtheactoractivelyseekscontactwitha recipient
it in a crouchedmannerwithcurvedback
by approaching
bent
the
and
legs, tailcurleddown,oftenwagging,with
earsfoldedback.Fromthisposition,the animaltriesto
contactthe recipientby lickingits nose (VanHooffand
elements
Wensing1987,Derix1994).Whenconstituent
occurredseparately,we recordedthemas separatecategories.Wealsocalculatedthe same2 measuresof rank
orderforthewolves.
Bear-wolfInteractions
on 3-21 February
Weobservedbear-wolfinteractions
1997aroundthefeedinghours(1000-1400),whenmost
occurred(Koene1998).We videotapedininteractions
a member
teractionswhen(1) a bearorwolf approached
of the otherspecies,or (2) bearsor wolvesstartedrunobvioustrigger.
ningwithoutanapparent
We recordedthe followingdatafromthe videotaped
sequences:date,time,andcontext(bearden,wolf den,
we recorded
the approach,
food,unknown).Concerning
and
sex
or
theapproacher
ageof first,
(bear wolf);name;
and
and
third
wolf
and
first,second, thirdbear;
second,
distance from which approachstarted;behaviorof
approacher(approach,attack,bite, etc.); whetherthe
approacherstoppedor passed;and distancewhen the
BEARSANDWOLVESIN CAPTIVITY* Koene
approacherstoppedor passed the approached.Concerning the shortdistance interaction,we recordedthe number of bears and wolves participatingin the interaction;
actor (bear or wolf), behavior of the actor (same as
approacherbehavior);recipient(bearor wolf); behavior
of the recipient(bear,wolf, unknown);distancebetween
the participantsof the interaction;andloser (bearor wolf).
We collected some datafrom interactionsin the wild for
the same variablesandcomparedthe enclosuredatawith
data from the wild. Data were analyzed with SAS 6.01
(1990).
RESULTS
Bear-bearInteractions
As a rule,agonisticinteractionsbetweenbearsoccurred
as soon as the caretakeroffered food to the bears. Some
animals gained access to the center of the feeding spot
with the highestfood concentration.Othershadto wait or
were even forced to stay entirely away from the place
whereall the food was thrown.As bearsgraduallyleft the
feeding spot, the numberof encountersdecreasedrapidly
with time after feeding. Most encounterswere without
physicalcontactandoccurredas follows: a dominantbear
would appearand subordinatebears would back up or
even run or walk away, turningthe head away from the
opponent;ears of both bears were back. If the subordinate beardid not back away, the dominantwould usually
chargeit. In a chargethe dominantbearran,frontallyoriented, directlytowardthe otherwith earsbackandmouth
slightly open. It was possible to distinguishthe dominant
and subordinatebears in most encounters.We recorded
419 agonistic encountersbetween the bears and determined the DS-rankorderof the 13 bears (Table2).
et al.
87
The DS-rank order yielded the following dominance
hierarchy:(1) old males were highest in rank, followed
by an adult male, (2) next an old female followed by an
adultfemale, the oldest blind male, and the last adultfemale, and (3) finally the younger male and females, the
old blind female, and the adultblind male.
Althoughthe hierarchydid not differ between feeding
spots,the amountof agonisticbehaviorof some individual
bearsdifferedsignificantlyfor the differentfeeding spots.
For example, Geert,highest in DS-rank,was involved in
moreaggressiveinteractionswith otherbearsat the place
he occupiedmost thanat otherfeeding places. Battir,second in DS-rank,was less involved in aggressive encounters in the home rangeof Geert.
Not all bears were present at all feeding places. For
example, 1 adult blind male (Koroglfi) did not walk
throughthe tunnels and did not eat when food was offered at the otherside of the tunnel.The old blind female
sometimescontinuedher stereotypicbehavioranddid not
eat.
Based on the dominancevalue, a somewhat different
rankorderwas found(Table2). Fourgroupscould be distinguishedbased on the dominance-value:(1) Geert and
Battir (DV 0.97 and 0.92, respectively), (2) Koroglu
(blind),Mascha,Nelly, andMackenzie(DV = 0.50-0.69),
(3) Niki, Bora,andFiona(bothblind)andAxel (DV=0.140.41) and (4) Wolke,Bjora, andTory(DV = 0.01-0.05).
Wolf-wolfInteractions
The behavioralelement associatedwith most observations of the total (415 of 1,491) was active submission.
Brenda was the only female that copulated with Cazar;
she was the alpha female. Noeska was the most active
wolf in the pack, with 46% of the total submissivebehaviors, of which 57%were directedtowardBrendaand39%
Table 2. Rank ordered dominant (DOM:rows) - subordinate (SUB: columns) matrixof 13 brown bears based
upon agonistic
Interactionswon and lost over food for a 1997 behavioralstudy in a 2-ha enclosure at Ouwehand
Zoo, Rhenen, the Netherlands.
A dash indicates that no encounters were observed between this pair of animals.
Label of
dominant
animal
Ge
Ge
Ba
Mk
Ne
Ma
Koa
Ni
Ba
Mk
Ne
Ma
Koa
Ni
Ax
Fia
Boa
To
Wo
Bj
Total
DOM
3
14
19
4
24
1
9
30
0
2
1
2
1
1
-
22
25
0
2
0
1
4
2
2
8
2
3
3
1
2
5
1
3
3
2
3
11
10
5
17
4
16
7
3
11
11
16
3
5
69
137
37
36
66
11
35
2
0
-
5
9
9
10
14
1
4
old
old
adult
old
adult
old
adult
2
4
1
-
-
-
2
2
2
2
2
8
1
3
62
1
419
2
0
0
0
0
0
2
1
4
0
2
1
1
1
1
Ax
0
0
0
Fia
-
0
-
Boa
To
Wo
0
0
0
0
0
Bj
Total SUB
0
2
0
12
a Blind bear
0
0
0
0
0
0
0 -
0
0
-
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
1
2
-
0
-
-
0
37
0
0
29
1
42
0
5
0
51
0
24
8
0
16
68
1
-
0
54
Age
DSrank
DV
DVrank
1
2
3
4
5
6
7
0.97
0.92
0.50
0.55
0.61
0.69
0.41
1
2
6
5
4
3
7
young
8
0.14
10
old
9
0.20
9
adult
young
young
10
11
12
0.33
0.01
0.05
8
13
11
young
13
0.02
12
88
Ursus 13:2002
towardCazar.But she was also very aggressiveto Sylvia.
The majorityof all dominantbehaviorswere directedtoward Sylvia. The dominance-subordinatematrix based
on active submission(Table3) rankedthe wolves (dominantto subordinate)as Cazar,Brenda,Poot,Traan,Streep,
Noeska, and Sylvia.
We founda DV-rank order(Table3) differentfromthe
DS-matrixforthe malesPoot andTraan.Wedistinguished
3 groups based on the dominance-value:(1) Cazar and
Brenda(DV = 1.00and0.96, respectively),(2) Traan,Poot,
and Streep(DV= 0.19-0.43), (3) Noeska and Sylvia (DV
0.07 and 0.00, respectively).
Bear-wolfInteractions
We recorded 19.5 hours of video including 79 bearwolf interactions.The context of these interactionsin the
enclosure were as follows: interactionsover food 19%,
nearwolf den 8%,nearbearden 5%, and unknown68%.
In 70% of the interactions,only 1 bearwas involved; 8%
involved 2 bears, 3% involved 3 bears, and 1%,4 bears.
In 57% of the interactionsonly 1 wolf was involved; in
14%,2 wolves; in 4%, 3 wolves; in 8%,4 wolves; and in
18%>4 wolves. In 28%, the bearswere the actors (initiating the interaction),and in 72% the wolves were the
actors.Bears retreatedin 23% of the encounters,wolves
in 77% (Table4).
When approachinganotheranimal, bears and wolves
stoppedor passed in the expectedfrequency(%2= 0.01, P
= 0.92). The animalthatapproachedwas usually also the
actor in the interaction(X2= 63.94, P = 0.001). Wolves
initiated more interactionsthan bears (X2 = 15.51, P =
0.001), and wolves retreatedmore often than bears, as
expected (X2= 0.41, P = 0.52).
Bear and wolf behaviorduringinteractionswas rather
comparable(Table5). Bears and wolves did not differ in
behavior elements as approacher(X2= 4.83, 9 df, P =
0.78) or as an actor(X2= 9.82, 9 df, P = 0.28). As recipient, however, wolves avoided bears more often than expected (X2= 21.52, 9 df, P = 0.011).
Mostbearsthatinteractedwithwolves wereyoungbears
(Table6); Bjorna,a 3-yearold female, accountedfor 47%
of all bear-wolf interactions.There was a significantrelationshipbetween DV-rankorderand numberof interactions(r = 0.60, P = 0.039, n = 13);the relationshipwas
strongerin sightedbears(rs= 0.81, P = 0.016, n = 10). In
most cases the wolf or wolves involved in the interactions could not be identified(87%).Traan(9%) andPoot
(3%),the subordinatemales, were involved in most identified (recognizedindividuals)interactions,and Brenda,
the alphafemale, only once (1%).
Table3. Rankordereddominant(DOM:
rows)- subordinate(SUB:columns)matrixof 7 adultwolvesbasedon the behavioral
elementactivesubmissionfora 1997behavioralstudy in a 2-haenclosureat OuwehandZoo,Rhenen,the Netherlands.
Label of dominantanimal
Ca
Br
Po
Tr
St
No
Sy
TotalSUB
Ca
0
0
0
0
0
0
0
Br
Po
Tr
St
No
4
17
2
5
2
1
25
4
1
0
100
133
5
2
3
2
2
1
0
0
9
0
0
0
0
19
0
0
0
8
0
0
30
0
243
Table4. Pathwayrepresentationof bear-wolfinteractions'
fora 1997behavioralstudyof 13 brownbearsand7 wolves
in a 2-ha enclosure at Ouwehand Zoo, Rhenen, the
Netherlands.
Approacher Pass or stop
Bear
pass
Bear
stop
pass
Bear
Bear
stop
Wolf
pass
Wolf
stop
pass
Wolf
Wolf
stop
Total
Actor
Bear retreat
Wolf retreat
bear
bear
wolf
wolf
bear
bear
wolf
wolf
2
2
0
0
0
0
9
5
18
5
10
0
0
1
2
11
32
61
a Protocolwas startedwhen a subjectapproacheda subjectof the other
species afterwhich the actionof the approachercould be eitherstops or
passes. The actorin the interactioncould be eitherthe approacheror the
approachedanimal. Frequency of withdrawingby brown bears and
wolves is presenteddependenton the type of approacher,the action of
the approachernearthe approached(pass or stop), and the type of actor
in the interaction.
Sy
9
73
0
2
3
19
106
Total DOM
160
214
9
6
7
19
0
415
Age
adult
adult
adult
adult
adult
adult
adult
DV-rank
DV
DS-rank
1
2
4
3
5
6
7
1.00
0.96
0.32
0.43
0.19
0.07
0.00
1
2
3
4
5
6
7
Table5. Behavioralelements shown by brownbears and
wolves as approacher(app),actor(act),and recipient(rec),
expressed as percentageof total numberof interactions.
Datafroma 1997behavioralstudyof 13 brownbearsand 7
wolves in a 2-haenclosureat OuwehandZoo, Rhenen,the
Netherlands.
Approacher
Behavior
Approach
Attack
Bite
Chase
Defend
Look at
Avoid
Flee
Rob food
Sniff
Threat
Threatbite
Total number
of interactions
bear wolf
70
63
7
16
7
5
2
0
0
0
5
11
0
0
0
0
2
0
0
2
5
0
2
5
60
19
Actor
bear
41
14
5
14
0
9
0
0
0
0
5
14
22
wolf
39
16
11
2
0
14
0
0
4
2
11
4
57
Recipient
bear
0
4
0
0
33
23
2
21
0
2
9
7
57
wolf
5
5
5
0
23
9
27
18
0
0
0
9
22
BEARS ANDWOLVESIN CAPTIVITY* Koene
DISCUSSION
Bear-bearInteractions
All bear species are primarilysolitaryin the wild, and
individuals avoid each other if possible (Brown 1993).
Bearsthatdo associatearetypicallya motherandhercubs,
siblings that recently left their mother, or male and female duringthe breedingseason (Brown 1993). The best
known example of a social groupingis when bears congregate temporarilyon places where good food is abundant,such as garbagedumpsand salmon (Oncorhynchus
spp.) streams (Stonorov and Stokes 1972, Egbert and
Stokes 1976). In this situation,bearsdevelop a stable social structureto use resources as efficiently as possible
withoutspendingtoo muchtime fighting.Theirrepertoire
of threateningbehaviors enables them to keep distance
withoutshowing overt aggression.The Bear Forest,with
its highfood availability,compareswell to thebrownbears
in Alaska studiedby Stonorovand Stokes (1972). Weber
(1988) reportedthat252 of 478 (52%)bear-bearencounters at naturalfeeding places in Romania were aggressive. Unfortunately,no rankorderwas presented.Weber
(1988) found the same repertoirein his bearsas Stonorov
and Stokes (1972) found, except for the behavior "jawing", which was not recordedin the Romanianbears.
We provokedagonistic interactionsamongthe bearsin
the Bear Forestby offering all the food for a meal in one
place insteadof scatteredaround.The dominancehierarchy we found was based on the dominance-subordinate
matrix and on the dominance value of each individual.
Blind bearswere rankedlower thanexpectedon the basis
of theirage andsex. However,basedon dominance-value,
the blindfemale Fionawas the only one with a lower rank
than expected based on her age and sex. Despite their
handicap,the blind bears still dominatedyounger bears
in the Bear Forest.
et al.
89
Literatureon social structurein brown bears revealed
the existence of temporaryhierarchiesin groupsof bears
on rich feeding places (Stonorovand Stokes 1972, Weber
1988). Largeadultmales were highest in rank,followed
by females with cubs, adultmales, andotherfemales. Finally young bears deferredto all the otherbears (Brown
1993). The hierarchyfound in the Bear Forest was very
similar to that in the wild under favorable food conditions. Furthermore,as in nature, agonistic interactions
withoutphysical contactwere used to maintainthis hierarchy,with strongbear-beartolerance(Colmenaresand
Rivero 1983).
Wolf-wolfInteractions
Whereas in bears social grouping is an exception, in
wolves it is expected.Therefore,dominancerelationships
are well established. It is nearly impossible to observe
social relationshipsof a wolf pack in the wilderness,because wolves have home rangesbetween 50-1,000 km2.
As a result, nearly all studies of social relationshipsof
wolves have been conducted on captive packs in large
enclosures.The enclosurein Zimen'sstudy(1982) of wolf
pack sociogramwas 6 ha. Mech (1970) used Isle Royale
as a naturalenclosurewith an areaof 54,400 ha. VanHooff
and Wensing(1987) describedpack structureundercaptive conditionsin a smallenclosure(0.3 ha) andconcluded
that: (1) there are differentrank orders for females and
males, especially for the higherrankingmembers,(2) the
sexual rankorderis mainly structuredaccordingto age,
with older animalsgenerallydominantover youngeranimals;(3) the rankdifferencesarelargerin the higherranks
andlowerbetweenthe lowerrankingwolves; and(4) there
is no cross-sex dominance relationship,as long as the
wolves have the same ranklevel in their sexual rankorder; if the ranklevels are differentand there are significant age differences, there are cross-sex dominance
Table 6. Brown bear-wolf interactions in relation to bear dominance in descending order, for a 1997 behavioral study of 13
brown bears and 7 wolves in a 2-ha enclosure at Ouwehand Zoo, Rhenen, the Netherlands.
Bear
Sex
Bj
Wo
To
Mk
Ax
Ge
Ni
Ma
Ne
Ba
female
female
female
male
male
male
female
female
female
male
male
male
female
Koa
Boa
Fia
a Blind bear
Age
DS-rank
young
young
young
adult
young
old
adult
adult
old
old
old
adult
old
13
12
11
3
8
1
7
5
4
2
6
10
9
DV-rank
12
11
13
6
10
1
7
4
5
2
3
8
9
IA (No.)
39
4
4
12
9
8
1
1
1
0
0
0
0
79
IA (%)
49.4
5.0
5.0
15.2
11.4
10.1
1.3
1.3
1.3
0
0
0
0
100%
90
Ursus 13:2002
relationships.The wolf dominancehierarchyis thus best
describedas a pyramidwith the alphamale andfemale at
the highestposition,followed by the betamale and sometimes the betafemale, thenadultsubordinatefemales and
males.
In the Rhenenwolf pack,therewere 4 especially active
wolves:Cazar,Brenda,Noeska,andSylvia.Cazarshowed
no submissivebehavior,muchdominantbehavior,andwas
apparentlythe alpha male of the pack. He was also the
only intactmale in the pack (with testicles and thus hormonal activity). Cazarwas the only wolf observedcopulating with a female (Brenda, the alpha female). There
was a clearlinearhierarchywithinthe femalesof thepack:
Brenda,Noeska, andSylvia. The relationshipsamongthe
wolves were sometimes very aggressive and led to injuries, especially for Sylvia, the lowest in rank.
Only 6% of all interactionsoccurredin the non-alpha
malesTraan,Streep,andPoot.It was difficultto rankthese
wolves basedon a small numberof interactions.Poot and
Streep showed more submissive behaviors than Traan;
Traancould be the beta male andthe othersthe adultsubdominants.This agreeswith the dominancerankfromthe
dominance-value(DV). On the otherhand,Landau'sindex of linearity(Lehner1996) for the pack was 0.77, indicatingno clearlinearhierarchy(a linearityindex of 0.9
is consideredas indicativeof a linearhierarchy).
Comparisonof our resultswith the hierarchyof a freerangingwolf pack is difficult. In the Rhenen wolf pack,
the social interactionsseemed to be more aggressive and
moreconcentratedon only some individualsthanin a freeranging wolf pack. Mech (1999) emphasized that freeliving wolf packs are family groupsin which dominance
displays are uncommon except during competition for
food. He suggested that active submissionis primarilya
food-begging gestureand not primarilyan indicatorof a
dominancerelationship.At the beginningof each breeding season, Brenda, the alpha female, severely injured
Sylvia by bitingher in the backnearthe tail, causingdeep
open wounds. In the wild, biting will not occur because
low rankedfemales leave the groupbefore the breeding
season (Mech 1999). In captivitythereis no possibilityto
leave the area and to avoid the alpha female. We speculate thatin the wild, Sylvia probablywould have left the
packto avoidbeingattackedby BrendaandNoeska.Traan,
Streep,and Poot would have been more socially active if
they had retainedtheir naturalhormonallevel. No sexually related behavior was observed in these 3 animals.
Althoughthe hierarchyappearedto be somewhatnormal
for wolves, it seemed to be restrictedby the non-intact
males and the restrictionsof the enclosure.
Bear-wolfinteractions
Magoun (1976) observedbearsand wolves neara car-
cass simultaneouslyin the wild, and during 173 5-min
periods she recorded39 aggressive acts by the bears towardwolves andonly 1 aggressiveact by a wolf towarda
bear.Unfortunately,the behavioris not describedin detail; most aggressive acts of bearsinvolved a shortlunge
or swipe, althoughsometimesa bearchaseda wolf. Murie
(1981) also describedwolves as the losers in bear-wolf
interactions,but othersobservedthe opposite (see Mech
1970, Brown 1993). In NelchinaBasin, brownbearscontested wolves in 13.1%of 130 observationsof wolves on
kills (Ballard1982; Table7).
A more recent descriptionof interactionsshowed that
wolves attack bears fiercely, especially when bears are
nearthe wolves' den (Kehoe 1995). Wolves mainly initiatedthe bear-wolf interactionsin the BearForest.Wolves
also retreatedthe most from interactionsand were therefore the probablelosers.The interactionswere sometimes
related to fights concerning the bear den, wolf den, or
food. However, most interactions(68%) were still of an
unknownorigin. Bears and wolves showed approachbehavioras actorsand defensive or flee behavioras recipients.
The wolves did not often harassthe blind bears. Only
one video showed interactionsbetween Bora and wolves
andKorogluandwolves. The firstconfrontationbetween
the blind bearsand the wolves was describedearlier,"all
blind bears were at least once attackedby the group of
wolves, butthe bearsdid not reactandthe attackstopped"
(Koene 1998: 582). A possible explanationis thatthe interactionswere mostly play motivatedand that the blind
bears reactedinadequatelyto the wolf behavioralinitiative.
Comparisonbetween the bear-wolf interactionsin the
enclosure and interactionsin the wild is rathercomplicated.Bearsdo not normallyconflictwithwolves. Wolves
cannot win an aggressive encounterwith an adult bear.
Still wolves often kill bears, but only when the bear is
ratheryoung, ill, or trappedin the den. In NorthAmerica,
cases of bears as victims of wolves have been reported,
butalso thereverseis found.We madean attemptto evaluate bear-wolf interactionsas reportedin the availableliterature(Table7).
Based on the data collected in the Bear Forest and the
dataextractedfromthe literature(Table8), we madesome
generalcomparisons.Contraryto expectation,interactions
in the Bear Forest were more of unknownorigin than in
the wild (x2=37.14, 4 df, P=0.001). Wolves andbearsin
the BearForestbeganinteractionsin the same proportion
as in the wild (X2= 1.31, P=0.25). Losers of the interactions in the wild werewolves in fewercases thanexpected
(x2=47.24, 3 df, P =0.001), but wolves were more often
the losers in the Bear Forest.
Conditionsof the Bear Forest may explain these dis-
* Koene et al.
BEARSANDWOLVES
INCAPTIVITY
91
reviewof bear-wolfinteractionsrecordedin the wild.
Table7. Literature
Actor
Behavior
Loser
wolf
bear
bear
wolf
wolf
bear
bear
wolf
wolf
bear
bear
wolf
wolf
bear
bear
bear
bear
bear
bear
bear
wolf
bear
bear
bear
wolf
bear
wolf
bear
wolf
wolf
wolf
wolf
wolf
wolf
wolf
wolf
bear
wolf
wolf
wolf
bear
bear
wolf
wolf
wolf
wolf
wolf
wolf
wolf
bear
wolf
wolf
chase
kill
attack
attack
attack
approach
eat
eat
eat
kill
attack
kill
kill
kill
attack
attack
attack
attack
attack
attack
attack
chase
chase
chase
eat
chase
chase
displace?
chase?
displace
kill
displace
displace
displace
kill
eat
eat
kill
kill
kill
approach
kill
attack
attack
attack
approach
chase
attack
stand
chase
chase
attack
bear
wolf
unknown
unknown
bear
unknown
unknown
unknown
bear
wolf
unknown
bear
bear
wolf
wolf
wolf
wolf
none
none
wolf
bear
wolf
wolf
wolf
bear
wolf
bear
wolf
bear
unknown
bear
unknown
unknown
unknown
bear
bear
wolf
bear
bear
bear
unknown
wolf
unknown
unknown
unknown
bear+cubs
bear
wolf
unknown
unknown
unknown
bear
Bears (No.)
1
1
4
4
1
3
1
1
1
~~1
1
1
1
?
1
4
4
1
1
3
1
1
>3
1
>3
4
4
1
2
1
1
1
1
1
?
3
1
1
1
1
3
1
1
1
2
3
1
4
1
1
1
1
Wolves (No.)
5
5
5
5
5
2
1
1
5
4
5
5
5
1
5
5
5
1
1
1
12
1
1
1
9
2
2
1
3
4
2
1
3
4
1
2
4
10?
2
5
1
3
3
1
5
1
2
1
1
1
1
1
Bear species
brown
brown
brown
brown
brown
brown
brown
brown
brown
black
black
black
brown
brown
brown
brown
brown
brown
brown
brown
brown
black
black
black
black
brown
brown
brown
brown
brown
brown
brown
brown
brown
black
polar
brown
black
black
black
brown
brown
black
black
brown
brown
brown
brown
brown
brown
brown
brown
Source
Mills 1919
Mills 1919
Murie 1944
Murie 1944
Murie 1944
Murie 1944
Lent 1964
Lent 1964
Pulliainen1965
Joslin 1966
RutterandPimlot 1968
Rutterand Pimlot 1968
trapper(in Mech) 1970
Ballard1980
Herning(in Murie) 1981
Murie 1981
Murie 1981
Murie 1981
Murie 1981
Murie 1981
Murie 1981
RogersandMech 1981
Rogersand Mech 1981
RogersandMech 1981
Rogers andMech 1981
Ballard1982
Ballard1982
Ballard1982
Ballard1982
Ballard1982
Ballard1982
Ballard1982
Ballard1982
Ballard1982
Horejsiet al. 1984
Ramsayand Stirling1984
Hornbeckand Horejsi1986
PaquetandCarbyn1986
Paquetand Carbyn1986
PaquetandCarbyn1986
Hayes andMossop 1987
Hayes andBaer 1992
Gehring1993
Veitchet al. 1993
Kehoe 1995
Koenea1995
Follmannb1997
Jamesb1997
Magob1997
Reynoldsb1997
Reynoldsb1997
Stephensonb1997
a Koene,
personalobservation.During the excursion of the TenthIBA (InternationalBear Association) conference in Fairbanks,Alaska, 1995, a
wolf coming from a long distanceencountereda grizzly bearmotherwith 2 cubs. The cubs were directedhigherup hill, while the motherstood on
her hind legs. The wolf stoppedand made a half circle aroundthe bearand her cubs and continuedon its way. Immediatelythereafter,the cubs ran
towardthe mother(as if they had been called) and the motherbear suckled them for a short while (as if to calm them). They then continuedtheir
foragingbehavior.
b Data from northwesternBrooks Range near the Kokolik River, Alaska (H. Reynolds, Alaska Departmentof Fish and Game, Fairbanks,Alaska,
USA, personalcommunication,1998).
crepancies. First, the food was provided and, although
there was competition, food was abundant.The size of
the food (chicken carcasses in contrastwith large prey
carcassesin the wild) probablyplayed a role in the sever-
ity andsolutionof the bear-wolf conflict. Second,wolves
normallyoutnumberbears in interactionsin the wild. In
the Bear Forest bear density was exceptionallyhigh, resultingin a higherpotentialdangerfor the wolves. Third,
92
Ursus 13:2002
Table8. Comparisonof bear-wolfinteractionsin the wild
(Table7) andin BearForestfora 1997behavioralstudyof 13
brownbearsand7 wolves in a 2-haenclosureat Ouwehand
Zoo, Rhenen,the Netherlands.
Interaction Item
Context
bearden
wolfden
food
cubs
unknown
Actor
bear
wolf
bear
Loser
wolf
none
unknown
BearForest
4
6
15
0
54
22
57
18
61
0
0
Wild
3
9
20
3
10
17
28
16
10
3
16
neitherbearsnor wolves could leave the area.Fourth,no
reproducingbearsor wolves were kept so no interactions
concerningcubscouldbe recorded.Fifth,bearsandwolves
in the enclosurewere familiarwith each other,a situation
thathardlywill occur underwild conditions.This familiarityimplies thatsocial contactsbetweenindividualsinclude morelearned(predictableandcontrollable)aspects,
and hence implies less explorationand emotional reactions. Bears and wolves seemed to tolerateeach otherin
the BearForest,as sometimesalso occurs in naturalsituations (Lent 1964).
IMPLICATIONS
MANAGEMENT
Duringtheresearchno severelyinjuriousbear-bearand
bear-wolf interactionswere recorded.However, wolves
sometimes bit and injuredother wolves in attacks that
could be fatal. Recently, Mech (1999) pointed to differences in social organizationof captive wolf packs and
naturalwolf packs.The typicalnaturalwolf packis a family, a breedingpairwith 1-3 generationsof offspring.This
may imply that a stable organizedwolf pack should be
setup from the startfrom a pairthatdevelops stablerelations by breeding.
As long as thereare no reproducinganimalsin the enclosure, no severe problems are expected from having
bearsand wolves together.However,we had 2 concerns.
The distancesbetween bear and wolf dens were unnaturally small, and because young bears, wolves, and cubs
seem to induce bear-wolf interactionsin the wild, wolf
and bear cubs in enclosures may cause severe managementproblems.Also, the feeding scheduleshouldbe varied to keep the animals more active, so that they do not
expect food every day at the same time.
To keep a wolf pack as naturalas possible, the males
should not be castrated- sterilized if necessary - to
maintaina naturalhormonallevel within the pack. Thus,
the pack will be more stable and activities will be more
equally distributedamongthe wolves. Whetherthe same
is truefor bearsis not clear.If a groupof intactbearsand
intactwolves arekept together,many interactionsshould
be expected. It is probablydifficult - maybe even impossible - to keep intactanimalsin such an enclosure.
CONCLUSIONS
Brownbearsin the BearForestdevelopeda social hierarchybasedupon sex andage. The blindbearsoccupieda
lower rankthanexpected based on their age and sex. As
in a free-rangingpack, the rankorderof the wolves was
sexually related.The linearhierarchyin the females was:
(1) alphafemale Brenda,(2) beta female Noeska and (3)
lowest ranking female Sylvia. The interactionsamong
these females were very aggressive.
Bear-wolf interactionsoccurredmoreoften in the Bear
Forest than at first thought.Interactionsof an unknown
originwere morecommonthanin the wild, andmay have
been play interactionsbetweenbearsandwolves familiar
with each other.However,moreresearchconcerningspecific behaviorsequencesis neededto elucidatethis question.
Comparingour results of interactionsin the enclosure
to interactionsin the wild, we found that (1) the dominance hierarchyin bears is similar to that found in the
wild duringsalmonfishing, that(2) the wolf hierarchyis
most probablysimilarto thatfound in the wild, and that
(3) bear-wolf interactionsareless severe in the enclosure
thanin the wild. In the enclosurewolves killed no bears
and bears killed no wolves. There seemed to be a high
tolerancelevel, as was sometimes describedfor natural
interactionsbetweenbearsand wolves.
ACKNOWLEDGMENTS
Special thanksaredue to OuwehandZoo, Rhenen,and
the Netherlands.Fundingfor the projectandtravelto the
EleventhInternationalConferenceon BearResearchand
Managementin Grazin 1997 came fromthe International
Bear Foundation,The Netherlands.Also thanksare due
to 2 refereesand H. Reynolds who providedinformation
about bear-wolf interactionsin the wild and stimulated
us to explore the comparisonof interactionsin the 2-ha
enclosurewith interactionsbetween bears and wolves in
the wild.
CITED
LITERATURE
W.B. 1980. Brown bear kills gray wolf. Canadian
BALLARD,
Field-Naturalist94:91.
. 1982. Gray wolf-brown bear relationships in the
Nelchinabasinof south-centralAlaska.Pages 71-80 in E.H.
Harringtonand P.C. Paquet, editors. Wolves of the world.
BEARSANDWOLVES
INCAPTIVITY
* Koene et al.
Noyes Publications,ParkRidge, New Jersey,USA.
G. 1993. The great bear almanac.Lyons and Burford,
BROWN,
New York,New York,USA.
COLMENARES,
E, ANDH. RIVERO.1983. Male-male tolerance,
matesharingandsocial bondsamongadultmalebrownbears
living under group conditions in captivity.Acta Zoologica
Fennica 174:149-151.
R. 1994. The social organizationof wolves and African
DERIX,
wild dogs. Dissertation,University of Utrecht,Utrecht,the
Netherlands.
DE VRIES,H. 1994. MatMan, a program for the analysis of
sociometricmatrices,Version2.51-2.53. Ethology& Social
Ecology group of the University of Utrecht, Utrecht, the
Netherlands.
EGBERT,
A.L., ANDA.W. STOKES.1976. The social behavior of
brown bears on an Alaskan salmon stream.Proceedingsof
the International Conference of Bear Research and
Management3:41-56.
FAGEN,R.M., ANDJ.M. FAGEN.1996. Individual distinctiveness
in brownbears, Ursus arctos L. Ethology 102:212-226.
T.M. 1993. Adult black bear, Ursus americanus,
GEHRING,
displacedfrom a kill by a wolf, Canis lupus,pack.Canadian
Field-Naturalist107:373-374.
R.D., ANDA. BAER.1992. Brown bear, Ursus arctos,
HAYES,
preying upon gray wolf, Canis lupus, pups at a wolf den.
CanadianField-Naturalist106:381-382
, ANDD.H. MossoP. 1987. Interactionsof wolves, Canis
lupus, and brown bears, Ursus arctos, at a wolf den in the
northern Yukon (Canada). Canadian Field-Naturalist
101:603-604.
HOREJSI,
B.L., G.E. HORNBECK,AND M.R. RAINE.1984. Wolves,
Canis lupus, kill female black bear, Ursus americanus, in
Alberta.CanadianField-Naturalist98:368-369.
1986. Grizzly bear, Ursus
HORNBECK,
G.E., ANDB.L. HOREJSI.
arctos, usurps wolf, Canis lupus, kill. Canadian Field-
Naturalist100:259-260.
P.W.B.1966.Summeractivitiesof two timberwolf (Canis
JOSLIN,
lupus) packs in Algonquin Park. Thesis, University of
Toronto,Toronto,Ontario,Canada.
N.M. 1995. Grizzly bear, Ursus arctos-wolf, Canis
KEHOE,
lupus, interaction in Glacier National Park, Montana.
CanadianField-Naturalist109:117-118.
P. 1998. Adaptation of blind brown bears to a new
KOENE,
environmentandits residents:stereotypyandplay as welfare
indicators.Ursus 10:579-587.
P.N. 1996. Handbookof ethological methods. Second
LEHNER,
edition. CambridgeUniversityPress, Cambridge,UK.
P.C.1964.Tolerancebetweengrizzliesandwolves. Journal
LENT,
of Mammalogy45:304-305.
A.J. 1976. Summerscavengingactivityin northeastern
MAGOUN,
Alaska. Thesis, University of Alaska, Fairbanks,Alaska,
USA.
MARTIN,P., ANDP.P.G. BATESON.1993. Measuring behaviour.
CambridgeUniversityPress, Cambridge,UK.
MECH,L.D. 1970. The wolf. University of Minnesota Press,
93
Minneapolis,Minnesota,USA.
. 1999. Alpha status, dominance,and division of labor
in wolf packs. CanadianJournalof Zoology 77:1196-1203.
E.A. 1919. The grizzly.Comstock,Sausalito,California,
MILLS,
USA.
A. 1944. The wolves of MountMcKinley.NationalPark
MURIE,
Service, FaunaSeries No. 5., USA.
.1981. The grizzlies of Mount McKinley. Scientific
MonographSeries No. 14. U.S. Departmentof the Interior,
NationalParkService. 1985.Universityof WashingtonPress,
Seattle,Washington,USA.
L.P.J. 1991. The observer: a software system for
NOLDUS,
collection and analysis of observational data. Behavior
Research,Methods,InstrumentsandComputers23:415-429.
PAQUET, P.C., AND L.N. CARBYN. 1986. Wolves, Canis lupus,
killing denningblackbears,Ursusamericanus,in the Riding
MountainNationalParkArea(Manitoba,Canada).Canadian
Field-Naturalist100:371-372.
PULLIAINEN,E. 1965. Studies of the wolf (Canis lupus L.) in
Finland.Annales Zoologici Fennici 2:215-259.
1984. Interactions of wolves
RAMSAY,M.A., ANDI. STIRLING.
and polar bears in Northern Manitoba. Journal of
Mammalogy65:693-694.
ROGERS,L.L., ANDL.D. MECH. 1981. Interactions of wolves and
black bears in northeastern Minnesota. Journal of
Mammalogy62:434-436.
RUTTER, R.J., AND D.H. PIMLOT. 1968. The world of the wolf.
J.B. Lippincott,Philadelphia,Pennsylvania,USA.
SAS INSTITUTE,
INC. 1990. SAS/ETS user's guide, version 6
edition. SAS Institute,Cary,North Carolina,USA.
STONOROV,D., AND A.W. STOKES. 1972. Social behaviorof
the
Alaska brown bear. Pages 232-254 in S. Herrero,editor.
Bears-their biology andmanagement.InternationalUnion
for Conservationof NatureandNaturalResources,Morges,
Switzerland.
VANHOOFF,
1987. Dominance
J.A.R.A.M., ANDJ.A.B. WENSING.
and its behaviouralmeasuresin a captive wolf pack. Pages
219-252 in H. Frank, editor. Man and wolf. Dr W. Junk
Publishers,Dordrecht,the Netherlands.
VEITCH, A.M., W.E. CLARK, AND F.H. HARRINGTON. 1993.
Observations of an interaction between a barren-ground
black bear,Ursus americanus,and a wolf, Canis lupus, at a
wolf den in NorthernLabrador.CanadianField-Naturalist
107:95-97.
WEBER,P. 1988.Beobeachtungenzu gegenseitigenBegegnungen
von BareninnerhalbeinerindividuellbekanntenPopulation.
Folia Zoologica 37:231-239. (In German.)
ZIMEN,E. 1982. A wolf pack sociogram.Pages 282-322 in F.H.
Harrington,P.C.Paquet,editors.Wolves of the world.Noyes
Publications,Mill Road, ParkRidge, New Jersey,USA.
Received: 4 September 1997.
Accepted: 7 June 2002.
Associate Editor: Kate Kendall.