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


Food Habits

The timber wolf is a big-game predator. Smaller animals including birds, rodents, and lagomorphs are eaten, but I know of no wolf population which has thrived on small animals alone. Only one record was found which indicated that the majority of wolf scats from an area contained anything other than big game.

Tener (1954) reported that on Ellesmere Island 83 percent of 85 wolf scats contained arctic hare (Lepus arcticus) remains, whereas 17 percent contained muskoxen (Ovibos moschatus).

Murie (1944) found remains of caribou (Rangifer arcticus), Dall sheep (Ovis dalli), or moose in 935 of 1,174 wolf scats collected in Mount McKinley National Park; and big-game remains composed approximately 70 percent of the 1,350 food items. In eight wolf stomachs and eight scats from Michigan, deer and hare remains were represented equally (Stebler, 1944). Cowan (1947) reported that 80 percent of 420 wolf scats from the Rocky Mountain national parks of Canada contained remains of elk (Cervus canadensis), bighorn (Ovis canadensis), mountain goat (Oreamnos americanus), moose, caribou, or mule deer (Odocoileus hemionus). White-tailed deer remains occurred in 97 percent of 435 scats from Wisconsin (Thompson, 1952), and in 80 percent of 51 wolf stomachs collected in winter from Minnesota (Stenlund, 1955). Fuller and Novakowski (1955) found remains of bison (Bison bison) in 32 of 49 wolf stomachs from northern Alberta. Caribou remains composed 58 percent of the items in 62 scats from the Northwest Territories (Banfield, 1954).

On Isle Royale the moose population represents the only potential food supply which could support the present wolf population; beavers (except in winter) and snowshoe hares are available supplements. Of 87 scats collected in May 1952, 56 percent contained moose remains; 24 percent, snowshoe hare; and 20 percent, beaver (Cole, 1952a). In 1954, Cole reported that "sixty-five percent of the scats contained moose hair and 35 percent beaver hair," but the number of scats examined was not given.

A total of 438 wolf scats were analyzed during the present study (table 10). These were collected from 100 miles of foot trails in spring and summer from 1958 to 1960. Since at the time, it was not known whether coyotes were still present, only scats over 1 inch in diameter were considered wolf scats, in accordance with information presented by Thompson (1952). Whenever possible, scats were aged to the nearest month, and those of unknown age were designated "old."


1958 1959 1960 Total
FreshaOldbTotal FreshcOldTotal FreshaOldTotal FreshOldGrand
Number of scats274370 1041102147477 154205230438
Number of occurrences3654 9013112425586 85171253263516
Food items Percent of occurrence
Moose, unidentified3...1 122518999 9.514.812.2
Moose, calf39484443 22360143848.2 15.531.5
Moose, adult

  Moose, total












Snowshoe hare5662335...
Unidentified mammal.........2tr.1...tr.
Red fox............tr.tr.............tr.tr.
Red squirrel...21.....................tr.tr.
Deer mouse............tr.tr.............tr.tr.
aMay through August.
bUnknown age.
cMay through August, plus 3 scats from October.

Although grass or sedge made up about 6 percent of the items, these are not considered food. They often are found in canid scats and even have been reported from mountain lion (Felis concolor) scats (Robinette et al., 1959). Some vegetation may be eaten inadvertently with the prey. Isle Royale wolves eat bloody snow while waiting for a wounded moose to weaken; perhaps in summer they eat blood-spattered grass or even bloody soil. Murie (1944) found that some of the wolf scats containing grass also held several round-worms, and suggested that grass may act as a scour.

calf carcass
Figure 61—All that remains of a 2-1/2-month-old calf, killed by wolves.

Moose remains composed 76 percent of the total (516) items. In scats from May through August, they constituted 74 percent of the occurences, and in "old" scats they formed 78 percent. (Old scats probably were from autumn and winter primarily.) Beaver remains composed approximately 11 percent of the total items, so beavers appear to be the only other important food.


One of the most important figures obtained during this study is the rate of moose kill by the pack of 15 wolves (table 11). (All animals fed upon by wolves are considered "kills," as is discussed on p. 115.) I believe that every kill made by this pack from February 5 to March 4, 1959, and from February 5 to March 20, 1960 was located. In 1961 most of the kills were found, but the wolves' activities were unknown on 11 of the 48 days between February 2 and March 20. During the periods in which the rate of kill was known in 1961, it averaged the same as in 1959 and 1960—one moose per 3 days. However, the wolves once killed two moose in 2 days, and the longest period we found between kills was at least 118 hours, and may have been as much as 137 hours, between March 7 and 12, 1960. The chronological distribution of kills is shown in table 11.


[Underlined dates indicate known dates of kill; others may vary by a day]

1959 1960 1961
DateAge and sex DateAge and sex DateAge and sex

Feb. 5cow Feb. 5calf Feb. 2cow
Feb. 8calf Feb. 7calf Feb. 4bull
Feb. 11calf Feb. 9calf Feb. 7—8calf
Feb. 14adult Feb. 12cow (Feb. 9—16 unknown)
Feb. 17calf Feb. 16cow Feb. 17bull
Feb. 18cow Feb. 20abull Feb. 18calfc
Feb. 21adult Feb. 22abull Feb. 23cow
Feb. 24cow Feb. 27calf Feb. 25calf
Mar. 1—4bull Mar. 1calf Feb. 27adult

Mar. 5calf Mar. 1—5fed on old killsd

Mar. 7calf (Mar. 6—11 unknown)

Mar 12acow Mar. 12adult

Mar. 15bcalf Mar. 14calf

Mar. 17calf Mar. 16adult

Mar. 18—20calfc Mar. 18—20bull
28 days: 9 moose 45 days: 15 moose 37 days: 12 moose

a May have been found dead.
bFound dead, but see p. 137.
c Human activity interfered with previous kill.
dA calf may have been killed during this period.

Apparently, the pack of 15 makes fewer kills than do wolves in other areas. Field men in the Rocky Mountain national parks of Canada determined the rate of kill of two packs of five or six wolves (Cowan, 1947). Each pack killed three elk per 2 weeks, "with indications that two small elk might be taken in a week," Conversion of these figures on a basis of pack size for comparison with those from Isle Royale gives a rate of one kill per 1.6 days. Stenlund (1955) estimated that a pack of 3 wolves in Minnesota would kill about one deer per 4 days, so 15 wolves probably would make one kill per .8 day. Through aerial observations of a pack of 10 Alaskan wolves, Burkholder (1959) found 21 kills (14 caribou and 8 moose calves) made in 35 days, an adjusted rate of one kill per 1.2 days. Some of the differences among these figures can be accounted for on the basis of prey size, as will be apparent in the next section. Undoubtedly, variation in availability of prey and in the methods used to derive these figures also contributes to the differences.

Since Isle Royale's smaller wolf packs never were studied closely for several weeks in a row, little is known about their rate of kill. Probably, both small groups (totaling five) did not together kill over a third the number of moose killed by the large pack. During the 1961 study period (48 days), when I became more proficient in locating kills made by the smaller packs, five of these were found. This is a third of the large pack's expected total (16), although all kills of the small packs may not have been found. These groups probably have more difficulty killing moose than the 15 wolves do, but each carcass should last them longer. Members of the small packs frequently wander far from their kills, so they may hunt new prospects while still able to resort to a previous kill for food. By the time one carcass is eaten, they may have another. In addition, each pack might feed on the other's kill. These speculations are based on limited evidence, but they might indicate direction for future research.

The lone wolf probably kills few moose in winter; it usually feeds on remains left by other wolves. Only once was evidence found that this animal made its own kill. The moose had been wounded and abandoned by the large pack a few days earlier. After finishing it off on March 12, 1961, the lone wolf fed without competition (except from foxes and ravens) at least until March 20.


No kill was weighed, so all consumption figures are based on moose weights given in the literature. According to information from Kellum (1941), Skuncke (vide Peterson, 1955:77), and Simkin (1962), fully adult cows average about 800 pounds, and bulls, 1,000 pounds (see page 93). Possibly these figures are a bit high for animals killed in winter, but since the amount of possible weight decrease occurring over winter is unknown, these figures will be used. Nine-month-old calves apparently weigh about 300 pounds (modified from Peterson). For ease in assessing kill figures from both the present study and from the literature, I have adapted an arbitrary unit to allow for the varying weights of big game. One "prey unit" is considered to be 100 pounds. Following are the assumed prey units for various ages and species of big game:moose calf, 3; cow, 8; bull, 10; deer, 1; elk, 6; caribou, 3. Adult moose of unknown sex are assumed to be cows, since the sex ratio of kills favored cows strongly.

When the Isle Royale rates of kill are examined in terms of prey units, they do not appear so uniform. The pack of 15 consumed 2.11 p.u. per day in 1959, 1.64 per day in 1960, and 2.21 per day in 1961. Comparable figures calculated from the literature are: Cowan (1947), 3.85 p.u. per day; Stenlund (1955), 1.25; and Burkholder (1959), 2.80. Since Stenlund's figure is an estimate, probably it is not as valid as those from Cowan and Burkholder, which are known rates of kill, or at least minimum rates closely approaching actuality.

Figures of average consumption per wolf-day should include consideration of the weight of uneaten remains. Since remains were not weighed, a standard estimate must suffice. I believe that unconsumed bones, skin, and hair averaged about 50 pounds per adult moose, and 15 pounds per calf. On this basis, the pack of 15 devoured approximately 5,555 pounds of moose in 28 days during 1959, or 13.2 pounds per wolf-day. In 1960, the 16 wolves consumed about 7,000 pounds in 45 days, or 9.7 pounds per wolf per day. The 1961 consumption by 15 wolves was approximately 7,740 pounds in 37 days, or 13.9 pounds per wolf-day.

No average-daily-consumption figures for wolves were found in the literature. However, Wright (1960) reported that African lions (Panthera leo) consumed an estimated .11 to .13 pounds per pound of lion per day, and wild dogs (Lycaon pictus lupinus) ate .15 pounds per pound of dog per day. On the assumption that the Isle Royale wolf pack contains 5 females at 61 pounds, and 10 males at 78 pounds (weights from Stenlund, 1955), the average wolf weighs 72 pounds. Therefore, the average daily consumption rates per pound of wolf per day would be: (1959) .18; (1960) .13; and (1961) .19. These figures compare favorably with Wright's.

Although average rates are useful figures, they also are misleading, for a wolf's feeding schedule is quite erratic. When food is available, wolves gorge; then they may go several days without eating. Young and Goldman (1944:120) explain it as follows:

Of all the members of the canine family, the wolf, when in its prime, can be most irregular in its feeding habits. Equipped with abundant power to kill, the preference is for large prey in order to sustain its large body. When large prey is not available, long intervals between meals may be endured rather than spend time and energy in quest of small animals. Then when opportunity again occurs the animals fill themselves to repletion.

As would be expected, the capacity of a wolf's stomach is extremely large. Several times, the pack of 15 devoured a calf within 24 hours, a rate of about 20 pounds per individual per day. On one occasion these animals consumed approximately half a cow in less than 2 hours. They killed the moose at 2:40 p.m. on February 12, 1960, and immediately began to feed. By 4:10 p.m. only three wolves were feeding, and the carcass appeared at least half eaten. The cow was mature, but possibly it was smaller than average. Even if it weighed only 600 pounds, the 15 wolves ate about 20 pounds apiece in 1-1/2 hours.

Other authors have recorded similar feats. Young and Goldman (1944) reported a wolf stomach weighing 18 pounds and another weighing 19 pounds, 3 ounces. According to Cowan (1947), four wolves devoured most of a doe mule deer in 4 hours; and in 5 days, three wolves consumed two mule deer and a calf elk. Cole (1957) found two instances in which a pack of seven or eight Isle Royale wolves ate about three-quarters of an adult moose in 2 days, a consumption rate of about 35 pounds per wolf-day.

Besides being able to consume great amounts of food in short periods, a wolf also can fast several days with no evident hardship. In 1960, the pack of 15 went at least 95 hours (March 8, 3 p.m. to March 12, 2:30 p.m.) apparently without eating anything except possibly hair and bones which they might have gleaned from old kills. In 1961, half of the large pack spent from March 1 to March 5 without any food except a beaver and scraps from old kills. (Possibly the wolves fed on material cached at old kills, but I think this is unlikely. Before leaving a kill, they clean it so completely that there probably is no extra food to cache.) Young and Goldman (1944) reported on a captive wild wolf which fasted a week and then gorged on the eighth day. E. H. McCleery of Kane, Pa., who has raised wolves—as many as 34 at a time—for many years, wrote me that in winter and spring he feeds his animals every 5 days, and in summer, every 5 to 10 days.


Wolves feed for the first few hours after making a kill. Apparently, a cow moose is not quite large enough for the entire 15 animals to feed on at once, for at the one cow we saw killed, two wolves had to wait off at one side; the others, packed solidly around the carcass, were tugging voraciously at it. After gorging on a fresh kill, the wolves usually curl up nearby and sleep. Each probably feeds at least twice during the first half day, for seldom during this period are there no individuals feeding. About mid-morning the whole pack often heads for an open ridge or stretch of ice, sometimes over a mile away, where each animal sprawls in the sun for several hours (figure 62). A few wander back to the kill now and then if it is not far. Around midafternoon the pack returns to feed. On the second day, if the carcass is large, the pack frequently travels leisurely for a few miles to a resting spot. By then little is left of the kill except the intact skeleton. When the wolves return again, they dismember the skeleton and spend hours gnawing bones. Usually 2, and some times 3, days are spent at the carcass of an adult, and one or 2 at a calf carcass. Although there is much variability in the above-described routine, it seems to be the basic pattern.

Figure 62—Resting attitudes of engorged wolves.

Two calves were examined on the ground soon after being killed, and information was obtained on the parts eaten first. On February 5, 1960, one of a set of twins was killed at 4:40 p.m. By 5:10 p.m. the neck and left side of the chest had been skinned; the heart, part of the lungs, the rump, and the nose were eaten, and there was a hole in the side of the abdomen (figure 63). The second kill, made at 11:45 a.m., March 17, 1960, was investigated within 45 minutes. Most of the meat was missing from one side of the head and throat, and from the upper hind leg and pelvic region; part of one shoulder was eaten. One side of the abdomen was wide open, with intestines pulled out and partly eaten, and the liver was gone. Whether these parts are preferred or whether they merely represent the points of attack is unknown.

Figure 63—Remains of 9-month-old calf 45 minutes after being killed by the large pack.

There is some indication that the pelvic and abdominal regions and the nose are preferred. An adult killed about 7:30 a.m. on August 26, 1960, was examined a few hours later. The wolves had been frightened from the carcass about 8 a.m., and the only meat missing was about 15 pounds from around the pelvis. Undoubtedly, most of the wolves were reluctant to return, for 2 days later just the nose and left side of the abdomen had been eaten. A bull wounded by the large pack and killed by a lone wolf about 6:30 p.m., March 12, 1961, was checked the next day at 11 a.m. The meat and a few pieces of intestine in the pelvic region were all that had been eaten.

Before the wolves abandon a carcass, all the viscera and flesh and about half the skin and hair are consumed. Sometimes the skin is left on the lower legs, but if the carcass is revisited, this is eaten also. Calves killed in winter usually are dismembered completely; all that remain are a chunk of hide, the disarticulated long bones, the mandible and upper tooth rows, and a great patch of hair (figure 64). In summer, the skin and most of the bones of calves apparently are devoured. Cow remains include the skull and anterior half of the backbone in one piece, and the pelvis and posterior part of the backbone in another. The legs are detached from the skeleton, but most of the bones of each remain together. The ends of the ribs and long bones. and the edges of the scapulae and mandibles are ragged from being chewed (figure 65). Usually, bull skeletons are less pulled apart but are thoroughly cleaned of meat. The completeness with which carcasses are consumed may attest that wolves have difficulty obtaining prey.

Figure 64—Remains of the calf in Fig. 63, 24 hours later.

cow moose
Figure 65—Cow moose remains gathered just after the large pack abandoned them.

Burkholder (1959:9) found the following usage pattern of caribou and calf moose carcasses:

The first parts of the animal eaten are the viscera, except for the stomach contents. The soft parts of the neck and ribs appear to be preferred over the more massive tissue structure. In many cases the entire animal is consumed, including hoofs, long bones, and skull, with only hair and stomach contents remaining.

Continued >>>

top of page Top

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

National Park Service's ParkNet Home