On-line Book

Book Cover
Fauna Series No. 7







Study Area

Isle Royale Mammal History

Methods and Extent of Present Research


Wolf-Moose Coaction




Fauna of the National Parks — No. 7
The Wolves of Isle Royale
National Park Service Arrowhead


Social Behavior

Although few opportunities existed for studying at close range the actions of the Isle Royale wolves, certain behavior was noticeable from the aircraft. No attempt was made to use foreign urine, dummies, or howling records for analyzing the actions of the wolves, but these techniques are suggested for future workers. The behavioral information presented is based upon distant observation of undisturbed wolves.


According to Young and Goldman (1944:120):

The pack is generally a pair of wolves and their yearling or two-year-old off spring. At times, however, there will be an intermingling of several wolf families to form a pack; but the duration of such bands is short.

Olson (1938) also asserted that packs are family groups and that larger packs consist of two or more families. Murie (1944), through recognition of individual wolves, as certained that the "family theory" of the pack held for two groups of wolves in Alaska. He observed members of the East Fork family together at various times from May 15, 1940, to March 17, 1941. In 1941 two females from this pack had young, each in a separate den, but by June 30, one group had moved in with the other. Another pack, seen in August and in December, each time contained the same wolves, three adults and six pups.

Figure 56—Part of large pack traveling across ice.

Theories conflict regarding the status of the pack in summer. Schenkel (1948) believes that each mated pair leaves the pack toward winter's end, the unpaired, weaker, and younger animals staying in small groups for some time. Cowan (1947) also thinks that in summer the hunting packs are broken up and that the animals hunt in smaller groups. However, Murie observed 15 wolves (including 2 litters of pups) in one group in July. He also saw five adults at a den in June and July. Whether members of a pack dissociate in summer may depend upon the type of terrain, the prey species, the composition of the pack, and several other variables.

On Isle Royale, most spring, summer, and autumn observations have been of single animals, which may have strayed, at least temporarily, from packs (table 9). Tracks, howling, and sightings of three associated wolves in summer and autumn indicate that probably the pack of three functions as an entity most of the year. A larger group also was heard howling several times. This, plus an observation of a group of six adults, and tracks of five adults show that on Isle Royale some wolves associate with others during the summer, at least at times. It may be that bonds among adult wolves are stronger in winter but that members of packs are together frequently in summer.


NumberaDate LocationObserver

6 July 1958 Huginnin Cove Trail Reported
3 Aug. 1958 Head of Tobin Harbor Reported
1 Aug. 1958 Rainbow Cove Reported
3 Oct. 1958 S. shore of Amygdaloid Channel Reported
1 May 1959 Lake Desor Author
1 May 1959 Hay Bay Reported
1 June 1959 Sugar Mountain Author
1 July 1959 Daisy Farm Reported
1 July 1959 Lake Desor Author
3 Oct. 1959 Mt. Ojibway Author (heard)
Several May 1960 Daisy Farm Author (heard)
Several May 1960 Daisy Farm Author (heard)
3 May 1960 Daisy Farm Reported (heard)
1 May 1960 Mt. Ojibway Reported
Several June 1960 N. of Siskiwit Lake Author (heard)
Several July 1960 Daisy Farm Author (heard)
Several July 1960 Daisy Farm Author (heard)
3 July 1960 Mt. Ojibway Reported
2 or 3 July 1960 Conglomerate Bay Reported (heard)
5 Aug. 1960 Attwood Beach Author (tracks)
3 Aug. 1960 Huginnin Cove Trail Reported
3 or 4 Aug. 1960 Chicken bone Lake Author (from air)
4 Aug. 1960 Chickenbone Lake Author (heard)
Several Aug. 1960 Daisy Farm Author (heard)
5 May 1961 Long Point Author (tracks)
2 or more May 1961 Malone Bay Author (tracks)
3 May 1961 Conglomerate Bay Author (tracks)

a In any of these cases, more animals could have been nearby but unobserved.

Murie (1944:45) explained what might cause a large pack to break up:

The size of the pack may be limited by the law of diminishing returns. Beyond a certain size, advantages may disappear. A pack might be so large that, after the strongest members had finished feeding on a kill, there would be little or nothing left for the rest. In such a situation, hungry ones would go off to hunt again, and the strong ones, already fed, would remain where they were. There thus might result a natural division of a band which was too large to function advantageously for all its members. One would expect that where game is scarce the wolves would operate in smaller units than where food is abundant.

The reason wolves form larger packs in winter seems to be unknown. Possibly, much more food is required during this season, and a larger pack might hunt more efficiently. The latter assumption might not be valid on Isle Royale, for while the large pack chases a moose, usually only five or six animals stay close to it; the others fall far behind. Because of this, it seems that the most efficient pack would contain five or six animals. Two packs of this size, operating independently, could travel twice as far as one, and therefore could locate, on the average, twice the number of vulnerable moose. It may be significant that in 1961, when fewer calves were present, the large group was split into two packs about half the time. Since calves provide much of the winter wolf food (about half during the 1960 study period), a shortage of this age class could cause more difficult hunting, which might force the wolves to operate in smaller groups. Indeed, in the 1961 study period when the large pack split up several times, the 15 wolves consumed more food than in either of the previous two periods.

At times, a larger pack might be more advantageous. Many a moose stands its ground when cornered by wolves. Such an animal usually is safe, for if the wolves cannot force it to run, they soon leave. Possibly a moose is more inclined to flee when confronted with several wolves than with few. If that is true, a larger pack would be more advantageous, for when a moose runs, it is much more vulnerable.

In the Rocky Mountain national parks of Canada, "the usual winter hunting pack consists of from four to seven individuals, with five or six the most frequent numbers. Packs of 10 or 12 have been reported once or twice in the Jasper area. The largest group recorded was believed to contain 14 individuals" (Cowan, 1947:157). Stenlund (1955) reported that a pack of 15 occurred in Minnesota, and Olson (1938) gave records of packs containing 20 and 30 wolves, also in Minnesota. Murie mentioned a sighting of 22 wolves, tracks of 24, and a report of 50, in Alaska.

According to Schenkel (1948:83) packs are formed in early winter. He provides the following description:

Chorus, howling, joint wanderings and hunting, and fairly early rivalries concerning leadership and sexual partnership denote this period. During this time the pack becomes a close (exclusive) society. Its core comprises the bitch wolf, presumably the only mature one of the pack, and the male "lead wolf." Whether the isolation of the mature female wolves from one another is the result of rivalries, what course these rivalries take in any event, and what effect they have on the formation of a pack is not known. The lead wolf and bitch more and more plainly be come a pair—first in the pack group—then at winter's end they separate from the pack and occupy a family area for the summer.

The possible beginning of a break up in the large pack was noticed in mid-March 1960. The 16 wolves had remained together from February 4 to March 16, but on the 17th and 18th only 13 were seen. On these dates the animals were inland and could have been miscounted, but on March 20 (the last day of the study period) they crossed Siskiwit Bay; only 13 were present. Perhaps this disbanding was only temporary, but it might have been the beginning of a seasonal breakup.


In the zoological gardens where Schenkel studied wolf behavior, more than one mature male and female were present in a pack, but one of each sex was dominant. These two highest ranking individuals he called "alpha animals." Murie (1944) believed that an unmated male was leader of one of the Mount McKinley packs, for other wolves approached this individual cowering. Apparently, this animal was dominant even to the mated pair within the pack.

Isle Royale's large pack contains at least three mature females, but only one pair (a male and female closely associated for 2 or 3 weeks) was observed each winter. On February 19 and 22, 1960, the female led the pack, while the male remained beside her, half a body length behind. Copulation was attempted several times. The female appeared inexperienced at leading, for she backtracked twice and was often shortcut by other wolves. Each time after shortcutting the lead pair, the rest of the pack waited "respectfully," and as the leaders passed, each individual assumed the submissive position described by Schenkel (1948:fig. 2). While passing these animals, the leaders held their tails high, in the dominant position (Schenkel, 1948:fig. 30a).

On February 6, 1961, the small female also led the pack, followed closely by a male. Both held their tails in the dominant position when approaching an old kill, whereas the rest of the animals held theirs normally. This pair probably was the "alpha" pair. Apparently, the male usually was leader, but while his mate was in estrus, he took the advantageous position behind her. Fuller and Novakowski (1955) reported that during a wolf-poisoning campaign in Wood Buffalo National Park, Canada, males dominated in taking the bait in two instances in autumn, whereas the only instance during the mating season showed that a female was dominant.

The alpha pair does not always head the string of wolves, but position in line does not necessarily reflect social order. In Alaska, Murie (1944) noticed that the first animal in line was not always the leader. In the present study there were three main activities during which one wolf appeared outstanding: journeying overland, hunting, and arousing the pack from its rest. (I do not know if this wolf is the alpha male as identified above, or even whether it is the same individual each time.) When the pack journeyed through deep snow, the first wolf in line frequently was 25 to 50 yards ahead of the rest, even though it broke trail. When this animal rested, the others did also, and when it began to travel, the others followed.

Figure 57—Social behavior among members of the large pack. Note oral "besnuffing" in uppermost pair, and varying tail positions. The smaller wolves (arrows) maybe females.

During hunts in 1959 and 1961, one wolf often seemed more aggressive. Sometimes one animal threatened a moose while the others paid no attention. During several chases, one wolf caught up to the moose before the others did, and in some instances an individual continued chasing for 100 yards farther than the others. In a few cases, although several wolves threatened a moose, only one actually attacked the animal. Once when most of the wolves were lying around waiting for a wounded moose to weaken, one individual, its front legs covered with the blood of the prey, continued to harass the belligerent moose.

A hunt on February 12, 1960, produced some unusual behavior of a different type. After a three-quarter-mile chase during which most of the wolves ran alongside a moose for about 300 yards without attacking it, the lead wolf suddenly stopped, turned around and lunged at those behind, as if to stop them from continuing the chase. It succeeded, for the other wolves turned and ran up their backtrail.

One wolf usually arouses the rest from their slumber. After stretching, this animal goes from wolf to wolf, touching noses and awakening each individual. As each wolf arises, it duplicates the procedure until the entire pack is active. Perhaps it is not necessarily the leader which initiates the arousing; it could be the first animal which awakens. (Murie describes similar arousing behavior begun by an individual other than the leader.) Nevertheless, I frequently saw the leader begin such activity.

Without identifiable individuals in the pack, it was impossible to distinguish the dominance position of the middle-ranking animals. However, status-demonstration was observed often among the wolves in the large pack. Frequent urination, oral, anal, and genital "besnuffling" (figure 57), presentation and withdrawal of anal parts, tail wagging, and mock attack upon weaker members of the pack (an energy displacement) were evident almost every day the pack was observed. Schenkel (1948) describes the above behaviorisms and explains their significance. The large pack performed the most noticeable social behavior in the following situations: (1) upon awakening, (2) when stopping to rest, and (3) while reassembling after a hunt or after splitting up.

Schenkel (1948:87) explains why this behavior occurs so frequently:

Every mature wolf has an ever ready "expansion power," a tendency to widen, not necessarily his personal territory, but rather, his own social behaviour freedom, and to repress his "Kumpans" of the same sex. Consequently, he maintains a constant watchful interest in all socially important happenings within the pack. In particular, status quarrels are never private affairs between two individuals; the whole society takes a more or less active part in them.

Murie (1944), Young and Goldman (1944), and Crisler (1958) also have observed one or more of the social behaviorisms.

The most repressed individual in the large group of wolves was the most conspicuous. This was the lone wolf which followed the other 15. We first saw the animal on February 23, 1959, about 100 yards behind the pack. It held its head low, ears back, and tail between its legs in the submissive position, and appeared to fear the other wolves. On February 24, while the pack rested on a lake, this animal was able to join the last two wolves in line. These slept for long periods and were not interested in the mating activity in which the rest engaged. The lone wolf wandered around near these two, and suddenly two others bolted toward it. The low-ranking individual ran off, directly past the two friendlier animals, but these remained unconcerned.

After a 100-yard chase, the pursuing wolves cornered the lone wolf near a snowbank and attacked it. They fought the animal for a few seconds and then, rejoined the excited pack. The lone wolf followed slowly, and again the two wolves attacked it momentarily. This happened a third time, after which the lone wolf did not attempt to follow the attackers. The details of the fights were not observable, but the lone wolf fought hard and did not appear to be injured. Lorenz (1952) asserts, on the basis of observations in the zoological park, that a wolf submits to its aggressor by presenting its throat, a maneuver which tends to inhibit the aggressive tendency in the attacker. Perhaps this is what caused the hostile animals to end their attacks so suddenly.

The lone wolf continued to follow the pack for the remainder of the 1959 study period and throughout the 1960 winter study period. We did not notice any more attacks on the animal, but it stayed away from most of the wolves and remained submissive. Schenkel reports that "energy displacements" directed at subordinates often occur in packs, and may take the form of ambushes, sneak attacks, and fights. He often saw cases in which several wolves directed their attack against one animal over a long period. This individual "... steadily lost the significance of environmental social partnership, was robbed of all social initiative and, in certain circumstances, with repeated attacks, became mortally wounded" (1948:88).

Despite the hostilities shown it by certain members, the lone wolf seemed to be accepted by part of the pack. The two friendly wolves mentioned above provide one example. On another occasion, when most of the animals were resting, two wolves backtracked around a point about 25 yards to meet the lone wolf. They sniffed the cowering individual a few seconds and accepted it. The three then moved back around the point a few yards to the rest of the pack. When the lone wolf saw the pack, it ran about 25 yards, lay down, and remained there. A third case of differential behavior toward this individual occurred on March 4, 1959. Ten of the wolves started traveling, while the other five (including the three lighter-colored, lankier animals) rested. The lone wolf joined these five. When they left, it accompanied them and was completely indistinguishable from them; no trouble ensued. A similar situation occurred on March 8. Six wolves, including the lanky individuals, were several miles behind the rest of the pack. The lone wolf joined these and accompanied them, without any apparent fear, to the rest of the animals. When they approached the main pack, the usual sniffing, tail wagging, and other greeting behavior took place; but the lone wolf quickly ran off and remained away from the others.

The reasons for the differential reactions to the lowest-ranking individual are unknown. Possibly, each of the members which accepted it also held low social status. Indeed, there were reasons to conjecture that the three lanky individuals were pups.

Although the lone wolf was suppressed by the large pack, it joined the group in pursuit of a strange wolf. All 16 animals chased the alien for half a mile. According to Schenkel (1948), this type of behavior is not unusual. He observed that despite the sometimes-violent relationships within a pack, the members present a united front toward aliens and become a unit during friendly activity such as chorus bowling.


The full significance of wolf howling is unknown. Murie (1944) described situations in which several animals howled before leaving the den area to hunt. From his descriptions, it appears that the howling was merely a manifestation of the wolves' restlessness. This seemed to be the case also during an observation of Isle Royale wolves. On February 9, 1961, four wolves lay on the ice at the head of Washington Harbor from 8:45 to 9:15 a.m., after which they headed into the woods. At 10 a.m., a wolf howled twice from the woods near the shore, and a few minutes later an animal appeared on the ice and howled three more times. Each time, the wolf's muzzle pointed skyward; the howls were low-pitched and drawn out. Five minutes later, four more wolves appeared, one at a time. They walked about 150 yards onto the ice and lay down.

About 2 p.m. one animal arose, stretched, lay back down, and howled a few times, arousing the nearest wolf. Then it approached this individual, with tail straight up and tip cocked forward, and sniffed its nose. The second animal rolled over and extended its paws toward the first. Meanwhile, the other three wolves arose, and all five walked about 200 yards westward and disappeared up a creek bed. A few minutes later, a single wolf emerged from the woods, sniffed the tracks of the others, cowered, and howled for a few minutes. Then it wandered eastward along the shore for about 50 yards and entered the woods.

Crisler (1958:151) believes that howling is an emotional outlet for wolves. She writes:

Like a community sing, a howl is not mere noise, it is a happy social occasion. Wolves love a howl. When it is started, they instantly seek contact with one an other, troop together, fur to fur. Some wolves . . . love a sing more than others do and will run from any distance, panting and bright-eyed, to join in, uttering, as they near, fervent little wows, jaws wide, hardly able to wait to sing.

Seton (1937) and Young and Goldman (1944) believe that wolves vocalize when chasing prey. This supposition seems logical, for vocal expression might help keep members of the pack together as they chase their quarry. However, the only evidence I have found in the literature to support this contention was an observation in 1875 reported to Seton (p. 281) by a logger. During the present study, all hunting was observed from an aircraft, so only indirect information was obtained on the subject.

Murray and Mech
Figure 58—Pilot Don Murray and author examining a fresh kill.

Two observations indicated that wolves did not vocalize while chasing moose. In one instance, the large pack chased and wounded a moose, while a second animal lay about 100 yards away. The latter moose eventually wandered away but did not seem cognizant of the 16 wolves nearby, as it probably would have if the wolves had been "tonguing." In the second situation, the large pack chased a moose, and a few of the animals caught up with it and held it at bay. Meanwhile, the others were wandering around searching for the moose. They finally found it by following the trail left by those which had cornered it. I believe that if any vocal communication had occurred, these animals would have run directly to the cornered moose.

One function of howling may be to aid in assembling. After long chases, the 15 wolves sometimes were scattered over a large area. On one such occasion, we noticed that a wolf ascended the nearest ridge and appeared to howl. Several others approached the first, and about 150 yards away another animal appeared to howl. Eventually, most of the pack assembled on the ridge. Murie (1944:102) described a similar episode.

On February 4, 1960, when we landed the aircraft within 100 yards of the 16 wolves on Intermediate Lake, the animals eventually scattered into the woods. A few minutes later we heard some howling, which soon increased in volume until the entire pack seemed to be involved. The whining, yelping, and howling (much of which was high-pitched) continued for about 30 seconds and then gradually diminished; a few single howls were emitted after the chorus had subsided. Tracks later showed that the wolves had assembled on a small knoll, where most of the howling probably originated.

Another time the pilot and I frightened the large pack from a freshly killed moose. As the wolves retreated, several barked hoarsely. We remained at the carcass for about 2-1/2 hours, and heard distant howling and barking intermittently throughout the period.

In the last instances some or all of the howling could have resulted from frustration or emotion. This undoubtedly was the case on an occasion in August 1960. From 9 to 11 p.m., I sat 20 feet up in a tree above a freshly killed moose. At 9:30 p.m. at least four wolves began howling about 200 yards away. Howling continued off and on for the next half an hour, but it gradually became more distant. A check the next morning showed that the wolves had not returned to the carcass.

Howling was heard several times near the Daisy Farm campsite (across Rock Harbor from my cabin) in the summer of 1960. The earliest time of day that I heard it was 5:40 p.m., and the latest, 12:30 a.m. It consisted of the usual medley of yips, barks, deep "mournful" howls, and extended calls of ever-changing pitch. Sometimes it occurred for only a few seconds, but once it lasted about 2 minutes. On one occasion when 22 campers were present at the camp site, several wolves howled directly behind the area. The animals sounded to me to be about 100 yards behind the lean-tos, although the campers thought they were closer. The reasons for all the howling in this area are unknown, but perhaps the sound of humans stimulated the wolves. Young and Goldman (1944) wrote that whistles and other human disturbances often stimulate wolves to howl.

Pimlott (1960) found that human "wolf" howling and recordings of wolf howling would cause wild wolves to perform. After extensive testing of this method he concluded (p. 7):

It appears that the stimulus of wolves to howl is, at least in part, directly proportional to the length of time since they last howled. It is frequently difficult to evoke a response within 15 to 20 minutes, or even longer, after they last howled.

Phonograph records of wolf howling were tried during the present study, and replies were obtained four times. The records were also played after "natural" howls, to determine whether the wolves would vocalize again within a few minutes of their first howl. Although only a few trials were made, results supported Pimlott's conclusion.


Activity resembling play was noticed on March 6, 1960. The 16 wolves had just left a kill and were traveling along the shore toward Cumberland Point. Several animals chased one another back and forth and in circles, but sometimes a group would chase one individual and then suddenly turn on another. It appeared that the pursued animal carried something, possibly a bone, and that as it dropped the object, another would pick it up and attempt to out run the rest. The pack cut across Cumberland Point, and in the woods the activity reached its maximum. The entire pack became involved, some in ambushing, others in chasing, until eventually the animals tired. Several other times, I have noticed tracks indicating that the wolves had engaged in similar "sport."

Over-cautious behavior on the part of a single wolf was observed on February 28, 1961. At 6:05 p.m., seven wolves started northward across Hay Bay from Hay Point. Most of the ice was bare, and the wolves were reluctant to walk on it, probably because snow-free ice usually is new and thin. They tried to keep on the chunks of old, snow-covered ice, which were frozen together by new, bare ice. When there were no more snow-covered chunks, the wolves walked on the opaque cracks across the bare ice. However, one wolf would not follow the pack onto the snow-free ice, although it was tempted. Instead, it headed westward into Hay Bay on snow-covered chunks, being careful not to walk on bare ice. When the animal reached the end of these chunks and faced bare ice, it returned the 250 yards to where the pack had crossed.

Again the wolf started following the tracks, but once they left snow-covered ice, it would not continue. This time the animal ran about 150 yards southwestward, back into the bay where the snow-covered ice was continuous. It crossed this without hesitation. Meanwhile, the pack had reached shore and was about a mile away. The cautious wolf hurried to them, catching up at 6:35 p.m.

Continued >>>

top of page Top

Last Modified: Thurs, Jul 4 2002 10:00:00 pm PDT

National Park Service's ParkNet Home