www.nps.gov/yell/learn/nature/upload/predation.pdf Both grizzly bears (Ursus arctos) and American black
bears (U. americanus) live in Yellowstone National Park
(YNP), which is located primarily in Wyoming, USA. In
areas where grizzly bears and black bears are sympatric,
temporal isolation and behavioral differences likely reduce
direct competition between the species (Aune 1994).
In the Greater Yellowstone Ecosystem (GYE), grizzly
bears are generally most active during nocturnal and crepuscular
times (Schleyer 1983, Holm et al. 1999), whereas
black bears are mostly diurnal (Barnes and Bray 1967,
Holm et al. 1999). Grizzly bears evolved to exploit nonforested
habitats, whereas black bears are primarily forest
adapted (Herrero 1978). Grizzlies are also generally
larger than black bears and much more aggressive in defending
themselves and their offspring from conspecifics
and other predators (Herrero 1978), whereas black bears
typically escape predators by running into forest cover or
climbing trees (Herrero 1985).
Due to their larger body size, grizzly bears have a competitive
advantage over black bears in large non-forested
areas (Herrero 1977). Although displacement of black
bears by grizzly bears from high quality habitat has been
documented (Shaffer 1971, Kendall 1984, Aune 1994),
interspecific killing of black bears by grizzly bears has
only occasionally been reported (Arnold 1930, Jonkel and
Cowan 1971, Murie 1981, Ross et al. 1988, Mattson et al.
1992).
We documented probable grizzly bear predation on an
adult male black bear in Hayden Valley, in central YNP.
Hayden Valley is a large (>8,500 ha) non-forested valley
surrounded by the forested Central Plateau. Flora in the
valley is dominated by sagebrush (Artemesia spp.) and a
variety of forbs, grasses, and sedges (Meagher 1973).
Numerous graminoid-dominated wetlands are present in
the valley. Lodgepole pine (Pinus contorta) forest types
that occur on infertile rhyolite soils dominate the forested
plateau surrounding Hayden Valley (Despain 1990).
Spruce (Picea engelmannii)–fir (Abies lasiocarpa) stands
are interspersed throughout the lodgepole pine zone in
areas of more favorable moisture regimes such as pond
margins, north slopes, and drainages (Graham 1978).
Grizzly bears are active in both the forested and non-forested
areas of Hayden Valley throughout the non-denning
season (Gunther et al. 1995). Black bears are mostly observed
within and near the edges of the forested portions
of the valley and rarely far from forest cover in the nonforested
areas (Gunther et al. 1995).
On 2 August 1998 we received a report of a dead black
bear on the northeast side of the Yellowstone River in
Hayden Valley, across from the Grizzly Overlook interpretive
sign along the Grand Loop road. We investigated
the report and found a dead adult male black bear in tall
sedges on the bank of the river, 174 meters from the road.
Field inspection of the carcass revealed that the dead black
bear had canine puncture wounds to the head and nose as
well as a crushed skull and left eye orbital. The penis,
baculum bone, and testicles were bitten off and found next
to the carcass. There were canine marks on the scrotum,
and the left hind quarter was partially consumed. The
carcass had not been buried. Two fresh scats containing
vegetation were observed next to the carcass and near (<3
meters) the partially consumed hind quarter. The predator
that scavenged and likely killed the black bear probably
defecated these scats while feeding on the carcass.
We collected the black bear carcass for necropsy to determine
cause of death and to obtain evidence as to the species
of predator that killed it. We collected the scats found
next to the dead black bear for DNA analysis to confirm
the species of the predator.
Laboratory necropsy indicated the dead black bear had
been in fair to good physiologic condition given the time
of year (kidney fat index measured 24%), and no evidence
of disease was observed. The carcass (minus the eaten
tissue) weighed 77.6 kg. Based on body size and condition,
we estimated that the black bear probably weighed
approximately 91 kg prior to being partially consumed.
The hide and musculature of the left side of the head were
torn away, exposing a portion of the skull. The left zygomatic
arch was shattered and bone fragments were missing.
A puncture wound had penetrated the skull ventral
and posterior to the orbital process. Numerous puncture
wounds were observed in the hide surrounding the head,
but no damage was noted to the neck. Although several
ribs on the left side of the thoracic cavity were broken,
the lack of hematoma and tissue damage to that region
indicated that the damage occurred post-mortem. Much
of the tissue surrounding the left hind quarter and a portion
of the large intestine had been eaten. The cause of
death was determined to be trauma to the head.
The large number of puncture wounds inflicted to the
bear’s head made it difficult to locate a matching pair of
canine puncture marks. However, one set of marks believed
to be caused by the lower canines was observed
near the right mandible. The center-to-center distance of
these canine puncture marks was 59 mm, typical of average
size, adult male grizzly bears in the GYE. In grizzly
bears, lower canine widths range from 35 to 66 mm (x
_
=
53 mm, SD = 6 mm, n = 35). Based on measurements
taken from reference skulls, a lower canine width of 59
mm is too large to have been inflicted by even a large
black bear, wolf (Canis lupus), or mountain lion (Felis
concolor) from the GYE. The distance between lower
canines range from 37 to 55 mm (x
_
= 45 mm, SD = 4 mm,
n = 31) in black bears, 35 to 48 mm ( x
_
= 40 mm, SD = 3
mm, n = 33) in wolves, and 29 to 41 mm ( x
_
= 35 mm, SD
= 4 mm, n = 56) in mountain lions. Predation by wolves
can be further ruled out because in 1998, wolves had only
recently been reintroduced to YNP and no wolves had yet
established territories in Hayden Valley (Smith et al. 1999).
Predation by a mountain lion is not likely either. In YNP,
mountain lions typically inhabit Douglas-fir (Pseudotsuga
menziesii) and spruce–fir forest types containing numerous
rocky canyons and outcrops (T. Ruth, Hornocker
Wildlife Institute, Gardiner, Montana, USA, personal communication,
2001). The kill site was located in a large,
non-forested valley bottom without these features, not
typical of mountain lion habitat in YNP. Based on canine
widths alone, we cannot completely rule out predation by
a black bear because it is possible that an exceptionally
large, old adult male black bear in YNP might have a 59-
mm lower canine width. However, the identification of
the predatory bear as a grizzly based on canine widths
was also supported through laboratory analysis of DNA
extracted from the bear scats collected at the kill site. DNA
extraction and species identification using mitochondrial
DNA fragment analyses (Murphy et al. 2000) unambiguously
identified the scats collected next to the dead black
bear as being from a grizzly bear.
The availability of trees as a means of escape through
climbing or hiding was an important selective force in the
evolution of black bears (Herrero 1977, 1978, 1985). In
this incident, no tracks were left in the lush grass to enable
us to determine the events that led to the black bear
being killed. Therefore, we do not know if the black bear
was stalked, ambushed, or chased to the kill site. There
were no ungulate carcasses nearby that would have attracted
multiple large carnivores to the area. To reach the
potential security of a climbable tree to escape from the
kill site, the black bear had 3 options. The nearest climbable
tree was a dead snag 72 meters west of the kill site,
on a small island (0.21 ha) in the Yellowstone River. The
black bear would have had to swim and run through the
river for 68 meters and run 4 meters on land to reach the
snag. The nearest climbable, live trees were in a small
(1.2 ha) stand, 129 meters to the southwest, and on the
shore opposite from where the black bear was killed. To
reach these trees the black bear would have had to swim
and run through the river for approximately 118 meters
and run 11 meters on land. The nearest climbable trees
that could be reached without swimming were approximately
900 meters northeast and uphill from the kill site.
These trees were also the nearest contiguous forest large
enough to potentially have provided escape or hiding
cover.
In areas where grizzly bears and black bears are sympatric,
differences in morphology, behavior, food preferences,
and habitat selection (Herrero 1978) generally allow
each species to exploit different subniches and coexist
within common geographic areas (Aune 1994). Holm et
al. (1999) reported overlap in home ranges of black and
grizzly bears in the GYE. In that study, black bears included
more forested habitats within their home ranges,
whereas grizzly bears selected more nonforested habitats
(Holm et al. 1999). Gunther et al. (1995) reported overlap
in areas of activity of grizzly and black bears throughout
most of YNP, but reported that black bears were seldom
observed far from forest cover in large non-forested areas
such as Pelican (Gunther 1991) and Hayden Valleys.
Black bears may underutilize large non-forested areas due
to habitat and food preferences (Aune 1994) or to avoid
potentially aggressive interactions with grizzly bears,
wolves, and coyotes (C. latrans; Herrero 1985). Our observation
of interspecific killing gives insight into (1) potential
selective pressures that may influence the
distribution of black bears, and (2) subniche separation
between black bears and grizzly bears in YNP.