NATIONAL PARK SERVICE Fauna of the National Parks of the United States No. 5 The Wolves of Mount McKinley

Description

ALL THE ALASKA SHEEP have been referred to as Ovis dalli dalli, except those found on the Kenai Peninsula, which have been described as Ovis dalli kenaiensis. Only slight skull differences separate this latter form from Ovis dalli dalli.

The Dall sheep is smaller than the Rocky Mountain bighorn (Ovis canadensis canadensis), has more slender and gracefully curved horns, and is white in color. On a dark background Dall sheep appear to be pure white, but in the snow they are seen to be slightly yellowish. Live rams average slightly less than 200 pounds in weight and the ewes are not quite as heavy.

Figure 17: Dall sheep ram. [Toklat River, May 21, 1939.]

Distribution of Dall Sheep in Alaska

Dall sheep, found in Mount McKinley National Park, are widely distributed over the mountainous regions in Alaska. They are found throughout most of the Alaska Range, in the Nutzotin, Wrangell, and Chugach Mountains, on the Kenai Peninsula, in the Endicott Mountains north of the Arctic Circle, and a few still occur in the Tanana Hills. Their range extends into Yukon Territory where they intergrade with Ovis d. stonei, a subspecies which differs from the Dall sheep mainly in color. Locally, the sheep have been reduced in numbers, or eliminated by hunting, but still they are found over most of their ancestral ranges. Sheep are reported to be still plentiful in the Wood River and Mount Hayes district of the Alaska Range, in the Endicott Mountains and on the Kenai Peninsula. In some areas, such as Rainy Pass, they are reported less abundant than formerly. However, on most of their range the actual status of the sheep is not well known.

History of the Sheep in Mount McKinley National Park

The history of the Dall sheep in Mount McKinley National Park, so far as we know, goes back over a stretch of many thousand years. It would be intensely interesting to know the detailed history of the sheep—the many vicissitudes during this long period. But through it all the sheep have survived. There probably have been many periods of sheep abundance, and many periods of sheep scarcity. During a series of easy winters, the population probably built up and overflowed the rough country into the lower gentle hills. Wolf populations may have had their own periods of scarcity and abundance, and, depending somewhat on the status of the caribou, preyed extensively upon sheep at times, or affected them little. When hares disappeared, leaving the lynx population stranded, no doubt the lynx in desperation hunted sheep. The history is an ever changing mosaic. It is to make possible the continuation of this natural course of events that Mount McKinley National Park has been established.

From 1906 to 1923, when I first visited Mount McKinley National Park, the sheep population was consistently high although there was considerable sheep hunting in the area up to 1920. Charles Sheldon (1930) found sheep abundant during his days there from 1906 to 1908. An "old-timer" told me that in 1915 and 1916, when he had hunted sheep in the area, they were very abundant, and that one winter he had sold 42 sheep carcasses. At this time several other market hunters operated in the region. Many sheep were apparently killed for the market and many were fed to the sled dogs used in hauling the meat. At an old crumbling cabin on East Fork River I found many old ram skulls, most of which were heaped in a pile. There were 142 horns, so at least 71 rams had been brought to this camp. The skulls had been split open, probably to make the brains readily available to the dogs.

Figure 18: Heap of 142 ram horns at the ruins of a hunter's cabin on East Fork River. This indicates the extent of hunting in the early days before establishment of Mount McKinley National Park. [July 1939.]

O. J. Murie in 1920 observed a market hunter operating at Savage River. The hunter had shot a number of sheep on these slopes. At that time many sheep were wintering on the ridges south of Savage Camp.

In 1922 and 1923 I observed sheep summering in considerable numbers at the head of Savage River where now scarcely any are to be found. In 1923 they were plentiful but probably not as plentiful as during the next few years because extensive market hunting had stopped only 2 or 3 years before.

From 1923 to 1928 there was a steady increase in the population. Some think the peak was reached in 1928. The numbers were estimated at 10,000 and upward. Two rangers estimated the population as 10,000 and another member of the park force estimated it as 25,000. No organized counts were made so far as I know, so it is hard to evaluate the estimates. Nevertheless the population in 1928 was no doubt large. With so many sheep occupying the ranges there was not sufficient food for all among the cliffs so that many wintered on gentle slopes and down along the river banks 4 or 5 miles from the mountains. The animals were forced into the lower hills and valleys to feed. On some ridges there were patches of dead willow which I presumed had been killed during the high density of the population. I feel confident, from all the information I have been able to gather, that there were at least 5,000 sheep in the park during the peak. How much larger the figure was it is difficult to say, but it is possible that there were as many as 10,000, as some have estimated.

During this period the history of the wolves is briefly as follows: In 1906—8 Sheldon (1930) found wolves among the caribou herds but few among the sheep. He observed two different wolves hunting sheep in the Toklat River region. Possibly they hunted more to the east of Toklat River, but apparently there were only a few wolves among the sheep. John Romanoff, who hunted sheep in the park in 1915 and 1916, said he did not see a wolf track in the area at that time. In 1922 and 1923 my brother and I saw no wolf signs at Savage River, nor had my brother seen any wolf tracks on a trip up Toklat River in December 1920.

Former Superintendent Karstens in his monthly report for December 1925, mentions the presence of wolves in the park and that they seemed to be on the increase. The wolves were not considered abundant in 1926 and 1927, but in 1928 it was felt that there had been a decided increase. From 1928 to the present time wolves have persisted in the sheep hills, apparently in fair numbers. No one knows what variation has taken place in the wolf population during this period. One ranger estimated 50 wolves in 1929—30 when he felt wolves were at their peak. I estimate the 1941 population at between 40 and 60 wolves in the sheep hills. The population may not have varied greatly from that figure during the last 10 or 12 years.

There apparently had been no severe winter during the time the sheep herd built up, although some sheep apparently were adversely affected by winter conditions at times. One of the rangers captured a ewe in the deep snows of Sable Pass in 1927 and the animal died a few days later. A weak ram found at the same time but not brought in was dead when seen again.

The snow conditions in the winter of 1928—29, according to reports, caused quite a large loss among the sheep. The Park Superintendent's report for March 1929, describes conditions as follows: "The winter has been a hard one on sheep with the deep snow and storms. They have been driven down from the ridges and into the deep snow of the flats in their effort to get feed. They were even noticed out on the flats near the north boundary 4 miles from the range."

In the April report the following statement is found: "The month of April proved to be the hardest one of the year for sheep. Very few places were kept blown bare by the wind. What few bare spots there were, were soon grazed off and the sheep ranged into the flats in search of feed. It is believed that many sheep starved to death. In the vicinity of Igloo the rangers picked up three rams and three ewes with lamb, though one of the ewes was too far gone to recover and died after a few days."

The account is continued in the July report for 1929: "Nyberg and Myers returned from a trip into the mountain [McKinley] on the 27th and reported that the wild sheep in the park look to be about as numerous as ever notwithstanding the hard winter and heavy snows, and report that most of the losses occurred in deep passes where they were marooned in the heavy snow and blizzards of March and April."

One ranger wrote me that the wolves had killed many sheep that winter, but that "the big jolt" had come in April, when heavy snows covered the food. For the part of the range he had investigated near Headquarters he estimated that a third of the sheep population had died. He said that those sheep that perished were mostly the old and the yearlings.

In the spring of 1929 there apparently was a fairly good lamb crop. The following two winters were not severe, but heavy wolf predation on the large sheep population was reported.

The most serious reduction among the sheep apparently took place in the winter of 1931—32, which was much more severe for the sheep than that of 1928—29. The Park Superintendent's report for December 1931 states that the rangers were experiencing difficulty in making patrols because of the heavy snows and that "from all indications the sheep are going to have a hard time finding forage this winter." The January 1932 report states that the month was very cold, that the sheep were all in good condition, but that the late snows had driven them well up toward the summits of the mountains. In February 1932 it was reported that all records for snowfall had been broken, that 72 inches had fallen in 6 days, and that the winter of 1931—32 would be remembered as the "year of the big snow." The Superintendent reported: "During the heavy snowfall which came on the 3d of the month, I was alone at headquarters and taxed to the utmost in shoveling snow from the roofs of the buildings. It was thought for a while that several would go down, as the snow was 4 feet deep in places. I called up on the phone to get a man from the station to come up and help me out. He left there at 8 a. m. and arrived here at 3:10 p. m. It took him just 7 hours to go the 2 miles notwithstanding the fact that he had on a good pair of snowshoes. The heavy snows that came during the fore part of the month were followed by 2 days of rain, then below-zero weather, and a heavy crust was formed which has caused untold suffering amongst the wild animals of the interior of the park. This is especially true concerning the moose. Their legs from the knee down are worn to the bone, and each moose trail is covered with blood. It is possible to walk right up on a moose as they have not the courage or strength to run away."

In the April 1932 report, after some investigation of conditions among the sheep had been made, the following statements are found: "We have suffered a severe loss of mountain sheep during the winter as a result of the heavy snows; also the predatory animals have taken their toll. Ranger Rumohr counted 15 dead sheep while on his way in from Toklat. He examined many of the carcasses for evidence of wolf kills, but in most cases it appeared that the deaths were the result of starvation."

Former Ranger Lee Swisher wrote me that he found a great many dead sheep that spring which were not wolf kills. Dixon (1938, p. 231) reports that Ranger Swisher had told him in August 1932 that he did not believe there were more than 1,500 sheep in the park at that time as contrasted with an estimate of 10,000 to 15,000 he had made in 1929. In view of the deep snow and the severe crust, and the huge sheep population, which quickly consumed the food available on the more favorable ridges, it is not surprising that such a catastrophe occurred.

The lamb crop in 1932, probably because of the hard winter, was very poor, so that deducting normal winter losses, the numbers must have been smaller in 1933. Ivar Skarland, anthropologist of the University of Alaska, told me that in a trip into the park in April 1931 he had counted 830 sheep. In 1934 he made the same trip in April and saw scarcely any sheep, which agrees with the other reports on the reduction of the population.

For the period from 1933 to 1939 variations in sheep populations are not well known. One ranger thought the ebb was reached about 1935 or 1936. In 1939 there was an excellent lamb crop and a good survival of the yearlings of the previous year. Another ranger said he thought there were more sheep in 1939 than there had been for 4 or 5 years.

The yearling losses were heavier in the winter of 1939—40, and in the spring of 1940 the lamb crop was far below par. These losses and the small lamb crop were a definite set-back to the population. The lamb crop in 1941 was excellent, but there were very few yearlings because of the few lambs the year before. The present population is estimated to be between 1,000 and 1,500 sheep, perhaps not far different from the population in the spring of 1932, when Lee Swisher estimated the number at not more than 1,500. It appears to me that since 1932 the sheep population has not varied greatly.

That, in brief, is the recent history of the sheep in Mount McKinley National Park, so far as I have been able to outline it from available records.

Distribution of Dall Sheep in the Park

The main ranges of Dall sheep in Mount McKinley National Park lie north of the backbone of the Alaska Range where the snowfall is much less than on the south slope. In an east-and-west direction, the sheep are found from the Nenana River to Mount Eielson, about 68 miles to the west. From Mount Eielson to the western boundary there are no sheep now except for some sporadic records. Scarcity of these animals west of Mount Eielson is apparently due to the absence of foothill ranges, there being too much snow on the short high spurs coming from the main range for much winter use. West of the park, hills occupied by sheep are again found in the Tonzana River region. Along the eastern border of the park, a few sheep occur between McKinley Park Station and Windy, near the Alaska Railroad. Eastward from the park boundary, sheep distribution continues on all suitable locations throughout the Alaska Range.

Because less range is used in winter than in summer, the winter and summer distribution will be discussed separately. (The areas of year-long use and purely summer use are marked on the map on page 7.)

Figure 19: Double Mountain and Teklanika River. This mountain is much used by sheep in summer, but is frequented very little by them in winter. Many caribou use the pass between the two peaks in crossing between Sanctuary and Teklanika Rivers. [May 17, 1939.]

WINTER DISTRIBUTION

In winter the sheep are found from the Nenana River to Mount Eielson, in suitable cliffs and slopes in the foothills and on the north end of a few of the spurs coming off the main range. From the Nenana River to Teklanika River sheep are largely confined to the long single "outside" or foothill ridge. The canyons through this ridge made by the Savage, Sanctuary, and Teklanika Rivers are especially suitable for winter range because of their ruggedness. Farther west, between Teklanika River and Stony River, the mountains available to sheep are a dozen miles in breadth. A depression a mile or more wide separates the "outside" range or foothills proper from the spur ridges coming off the main Alaska Range. At Savage River the tips of the spur ridges were formerly much used by sheep in winter, but this is no longer the case. Possibly the sheep had been too vulnerable to wolf attack there. At Toklat River the spur ridges which are separated from the foothills only by the gravel bars of the streams are still a part of the winter range.

The ridges on the winter range are mostly under 6,000 feet elevation and the sheep are usually found between 3,000 and 5,000 feet. When conditions are suitable they may range from a stream bed at 3,000 feet up the slope to 4,000 or 5,000 feet. At altitudes in excess of 5,000 feet food becomes scarce. As a rule there is relatively little snow on the winter range, even on the flats, although deep snows do occur occasionally. Prevailing south winds, which often become strong, blow many of the ridges and slopes free of snow. Although the prevailing winds are from the south, many north slopes also are blown bare. Much snow is blown into gullies and ravines and thus may at times impede free travel from one slope to another.

SUMMER DISTRIBUTION

When the winter snows melt sufficiently to permit freedom of travel the range of the sheep expands greatly. Many then move nearer the main Alaska Range, some going to the heads of glacial streams. In mid winter the snow is too deep in these regions to permit their use. Not all the sheep make these movements, for during the summer some sheep may be found over much of the winter range. Thus, on the winter range near East Fork, between 100 and 200 rams and some ewes may be found throughout the summer. Many of these rams have moved into the area from other parts of the winter range. There is apparently considerable movement by those sheep which summer on the winter range. Sheep are found all year in some numbers on Sable, Cathedral, and Igloo Mountains, and on winter range along the Toklat River as far north as the last canyon. The exclusive summer ranges are slightly higher in elevation.

Figure 20: Some of a large group of rams that spent the summer on the hills along East Fork River. This is also a winter range. [July 15, 1940.]

Migration

The movements to the summer range take place during much of June and sometimes as late as July. The return to the winter range begins in August, but most of the fall migration takes place in September. Stragglers have been seen migrating in early October.

Although many details concerning the migrations are not known, a number of observations were made of the general movements. Sheep wintering on the outside range near the Nenana River apparently summer in some of the mountains on the west side of Riley Creek. They were seen crossing to the winter range in late August at Mile 5. Some of the sheep at Savage Canyon cross a mile or more of low country and follow the ridges to the head of Savage River. Some of these sheep were seen moving southwest toward Double Mountain and others may go to Sanctuary River before going southward. Most of the sheep wintering in Teklanika and Sanctuary Canyons move south 9 or 10 miles to Double Mountain and points beyond. Some of the sheep wintering on the hills west of Big Creek go eastward to Double Mountain by way of Igloo and Cathedral Mountains. Sheep from the hills west of Sable Mountain and the head of Big Creek and possibly from lower East Fork River move eastward to Cathedral Mountain and cross the Teklanika River to Double Mountain and ridges between the forks of Teklanika River. This is a pronounced movement.

Some sheep summer at the head of the east fork of the East Fork River but I do not know where they come from. They could come from the East Fork or Toklat River winter ranges. Some of those wintering on the Toklat River move to the heads of the two forks of the Toklat 7 or 8 miles beyond the boundaries of the winter range. The few sheep on Stony Creek move to the head of the creek, and those wintering in the cliffs across from Mount Eielson move 8 or 9 miles to the vicinity of Sunshine Glacier.

Figure 21: Looking up to the head of the east branch of the Teklanika River. This is mountain sheep summer range. Vegetation, especially grass, is sparse in these mountains. [May 17, 1939.]

Before venturing a crossing over low country the sheep may spend hours looking over the region from the slopes, apparently to be sure the coast is clear. Sometimes they may spend a day or two in watching before making the attempt. Often ewes and rams move across together in a compact band.

The sheep at Sanctuary and Teklanika Canyons have 3 or 4 miles of low, rolling country to cross to reach the hills adjoining Double Mountain, their immediate destination. Three or four well-defined migration trails lead across the low country.

On June 7, 1940, a band of about 64 sheep, both ewes and rams, crossed from Sanctuary Canyon to the low hills adjoining Double Mountain. They started crossing about 2 p. m., and did not arrive at the hills until 5:30 p. m. Most of the way the sheep traveled in a compact group, stopping frequently to brook ahead. Through tall willows and scattered spruce woods they walked in single file. Just before reaching the first hills they fed for about 15 minutes on the flats at their base. They probably were all hungry and came upon some choice food. When they emerged from the woods to the open hills they were strung out considerably and galloped up the slope in high spirits, seeming relieved to have made the crossing.

On July 27, 1939, eight sheep made a belated migration from Sanctuary Canyon to the hills north of Double Mountain. From these hills I saw a group consisting of two old ewes, two lambs, and two yearlings, one 2 year-old ram, and a young ewe coming across the flats, alternately galloping and trotting, and occasionally stopping briefly to look both ahead and behind. I lost sight of them as they approached the hills to one side of me but presently I saw them cross a creek and climb a long slope to the south. They disappeared over the top but in a few minutes all eight sheep reappeared in precipitous flight and galloped down the slope with miraculous speed and abandon. The lambs kept up with the others. They passed within 30 feet of the place where I crouched in the willows so I could see clearly how they panted hard with open mouths. They climbed a slope north of me and stood above some cliffs surveying the terrain they had crossed. No pursuer was seen but it appeared they had been badly frightened. Later they regained their composure and fed and rested.

When sheep cross a valley they often follow an old rocky stream bed even though the travel would be much easier on the sod beside the stream. They probably have an instinctive feeling of safety when traveling among boulders even on level terrain since such terrain resembles the cliffs where they are safe. This instinct for seeking the boulder-strewn country on the level would serve a good purpose if such areas were extensive enough so that an enemy could not follow on firm smooth sod close by, as they can along the stream beds. On one occasion two ewes which were captured by wolves might have escaped if they had not followed a rocky stream while the wolves ran on the firm sod alongside the rocks.

In migrating along ridges where there are cliffs the sheep move leisurely, frequently stopping in places for a day or longer to feed. Sometimes they may remain on a mountain along the way for a week or more. A band of about 100 ewes and lambs fed on Igloo Mountain for a week before continuing their migration. They moved slowly around the mountain in their feeding, some days going only 300 or 400 yards. Sometimes a band will retrace its steps a half mile or so before going forward again. The movements in both spring and fall are similar. although the fall migration may at times be hurried a little by snow. However, the sheep frequently begin their fall movements before the coming of heavy snows. Single animals, or bands containing up to 100, may be seen in migration.

Figure 22: Looking southward up Igloo Creek to Sable Pass, a much-used caribou migration route. The mountains in the foreground are grazed by sheep the year around. The distant ridges are spurs of the Alaska Range. The highway can be seen along Igloo Creek (lower left). [May 17, 1939.]

The causes of migration are difficult to determine. Vegetation influences migration at least to the extent that it must be satisfactory on the range sought by the sheep. Grazing animals will often follow the snow line in spring as though they were seeking the tender new vegetation. This may be partly a cause-and-effect relationship, but, at least in part, it may be purely an incidental correlation; in following the snow line the animals may simply be driven by an urge to return to a remembered summer habitat, and not necessarily because of the seasonal stage of the plant growth. In Mount McKinley National Park it was not at all evident that the sheep were following very closely the appearance of new vegetation. Near the glaciers, where some of the sheep go, it is true the plant growth is much later than on parts of the winter range, but in many cases the growth stages of vegetation on winter and summer habitats are quite similar. On north slopes, especially, the growth on winter range is late and is green and succulent all summer. Indeed, in some respects the food on winter range seems better than on purely summer range. On winter range there is more grass and a greater variety and abundance of plant species. On the other hand, many of the species that are especially attractive to the white sheep occur on the purely summer ranges in quantity that is ample for their needs. If the migration were a complete one from winter to purely summer range, with no exceptions, one might conclude that the summer range was preferred and was not utilized in winter because of deep snows. This may still be true to some extent. A detailed study of the vegetation over all the ranges might reveal some interesting correlations.

We may assume that migratory habits had a beginning in the racial history of a species. At such an early period the sheep may have remained on the remote summer ranges until driven out by snow. Then, after such a movement had become habitual and more or less punctual, it is reasonable to assume that their fall migration might anticipate the coming of the snow. Or, the inception of the fall movement may be due to the deterioration of vegetation (by ripening and drying) on the purely summer range, for the sheep usually begin to move before the snow drives them out.

Figure 23: Excellent all-year Dall sheep range along East Fork River. [July 15, 1940.]

Insects are often stressed as a factor causing migration of big game and no doubt do at times influence local movements. Occasionally flies annoyed the sheep and caused them to seek the shade of cliffs, but usually the flies were not much in evidence, probably because of the cool breezes on the ridges. Since insects are relatively scarce on purely summer and winter ranges, it seems unlikely that any difference in their incidence would be sufficient to cause migratory movements. As suggested above, the insects might cause local movements such as from one side of a ridge to the other or from the sunshine into the shade.

One factor which may be of importance in explaining the migrations is the natural tendency of animals to wander. Let loose some horses and they will wander widely in their grazing. In any area the sheep wander about considerably over a period of days. Repeated wandering movements to certain localities may in time have established definite migration habits. In other words, the sheep may migrate simply because they like to travel.

The factors originally causing the movements of sheep may have disappeared, so that the animals may now be migratory largely because of habit which has no present-day use. When the population was excessively large, as it was in the late 1920's and often has been in the past, the sheep may have moved from an overgrazed winter range to the fresh pastures unavailable in winter. Now that there is no need for fresh pastures, the sheep may continue their treks because of a habit handed down to them.

The causes motivating sheep migration require more study and careful interpretation before any conclusions can be reached.

Food Habits

The general pattern of the food habits of the sheep was established by observing their feeding, examining the vegetation where they had fed, examining stomach contents. A detailed study of the summer food habits was not made so that no doubt many minor food items are not here recorded. It was of interest to find that, although the main food consisted of grasses and sedges, considerable browse was eaten.

Figure 24: Igloo Creek is in the foreground; the far mountains are on the other side of Teklanika River. Sheep frequently descend to the creek bottom to feed on the willows. The spruce in the middle distance is at timber line on Igloo Creek. [September 1939.]

WINTER FOOD HABITS

Much of the information on the winter food habits was secured from the analysis of the stomach contents of sheep which had died on the range. The stomach contents of 75 sheep were analyzed. Of these, 54 had died in winter when no green grass was available, and 21 had died during the fall and spring, when a small amount of green grass was to be obtained. Since the only essential difference between the two series is the presence of some green grass in one of them, they have been combined in Table 3, page 76. Seventeen of the stomachs were from rams, 26 from ewes, 2 from lambs, 5 from yearlings, and 25 were undetermined.

Since most of the sheep from which these stomachs came were old or diseased it is possible that the contents are not entirely representative of the entire population. Healthy sheep might wander down in the bottoms more to feed on willow or sage, and weak animals probably would not move about a great deal in search of variety. However, the series of stomachs probably gives a fairly good picture of the winter food habits.

TABLE 3—Results of food analyses of 75 Dall sheep stomachs collected in Mount McKinley National Park, from animals that succumbed in autumn, winter, and early spring.

An annotated list of the foods known to be eaten by the sheep in winter follows:

Grasses and Sedges.—The principal winter food of the mountain sheep consists of grasses and sedges. Every stomach contained some grass, and the average amount present was 81.5 percent.

in the middle of September, when the ground was covered with 14 inches of snow, and again in October after a heavy snowfall, I noted, in following trails, that the sheep were feeding extensively on the seed heads of Calamagrostis canadensis, C. langsdorfi, Festuca rubra, Agropyron latiglume, and Trisetum spicatum. Heads of other grasses are perhaps also eaten. Such feed must be very nourishing. Considerable amounts of dry grasses were eaten by early October.

There is a good growth of grass and sedge on the winter range, and since the wind generally blows the snow off the exposed ridges, usually these foods are readily available, On the high ridges species of Poa and Festuca form much of the grass cover. Along with these highly palatable grasses there is a much-utilized short sedge (Carex hepburnii) which grows in solid stands.

Figure 25: The ewe and ram are on one of the exposed grassy ridges which are grazed closely in winter. [Polychrome Pass, May 17, 1939.]

Willow.—Willows (Salix spp.), both the tall and dwarf types, are eaten at all seasons. They were present in 63 stomachs and the average percentage in these was 10.2. The largest percentage found in any stomach was 75. The species of willow are not listed separately except in the case of one of the dwarf willows (Salix reticulata) which was noted in two stomachs. Some willows may be somewhat more palatable than others but several species among both the tall and dwarf varieties are highly relished.

Dwarf willows are widely distributed over the ridges, and the tall willows grow along the streams and in swales between the spur ridges, sometimes far up the slopes where the sheep can feed upon them without too much danger from wolves. Often in October and November sheep descended to creek bottoms to feed on willows. They also fed extensively on those growing in the swales between the ridges. Here in the fall they waded through 14 inches of snow in search of willow leaves that were still green. Dried leaves were also eaten at this time. Along with the leaves, twigs up to a quarter of an inch in diameter were eaten. Generally the leaves are picked off the sides of the twigs, sometimes down the twig a foot or two. Because willow is eaten so much during the early winter, I expected a higher representation in the winter stomachs, but as the snow deepens in the swales, willow probably is not so readily available.

Horsetail.—Although horsetail (Equisetum arvense) was found in only four stomachs, it is nevertheless highly relished. The species is found in swales and wet spots, and sometimes in good stands quite high up the slopes. Because of its occurrence In swales it is probably not readily available in winter because of snow. At East Fork at the base of the ridges there is an extensive stand which had been closely cropped in the spring of 1941. There were many sheep droppings on the area. In the fall after the snow appeared several places were noted where sheep had pawed for it through the snow. Two ewes killed by wolves on October 5, 1939, had been feeding in a swale on this plant when surprised. Their stomachs contained almost 100 percent Equisetum.

Figure 26: Ewes feeding in about 14 inches of snow. They pawed readily down to grass but fed mostly on the leaves and twigs of willows and grass heads protruding above the snow. [Igloo Creek, September 19, 1939.]

Sage.—There are at least three species of sage which are highly palatable. Artemisia arctica and A. frigida occur scatteringly and were nowhere found to be abundant.

These sages are probably eaten in winter although none was recognized in the stomach contents. In late fall they were found much eaten. The tall herbaceous sage, Artemisia hookeriana, is highly palatable at all times. It is most abundant along streams and reaches far up the slopes in the draws between ridges. It often grows in dense clumps. In September and October, when the leaves and stems were dry, the sheep frequently fed on this herbaceous sage in the creek bottoms. Tracks at Big Creek showed that a band had crossed the creek and fed on a sage patch near the mouth of a ravine. In the high draws at East Fork this sage was uniformly closely browsed by spring. It was found in 20 of the stomachs and averaged 5.4 percent of the contents. The greatest amount of it found in a stomach was 35 percent of the contents.

Cranberry.—Cranberry (Vaccinium vitis-idaea minus) is an abundant evergreen widely distributed over the hills. Although eaten in only small quantities, it apparently is frequently sampled. It was found in 30 stomachs. Usually only a trace was noted, but in one stomach it made up 70 percent of the contents. In October, when a variety of food was available, the sheep were several times observed feeding on cranberry.

Blueberry.—Blueberry (Vaccinium uliginosum) was found in eight stomachs, making up 60, 40, and 15 percent of the contents of three of them, and 1 percent of the contents in each of five others. The berries, leaves, and twigs were eaten, but the berries were especially sought. The stomach of a lamb which was killed on October 4, 1939, contained 40 percent blueberries. On that date a band had been feeding extensively on blueberries. On October 7 and 12 it was again noted that sheep had fed heavily on them. Some of the berries had been picked up from the ground. The deterioration of the berries as the season advances, and their reduced availability because of snow, probably is the explanation for the relatively few stomachs containing blueberry in the winter.

Crowberry.—Crowberry (Empetrum nigrum) was present in 14 stomachs, making up 30 percent of the contents of one stomach, but on the average comprising 2.7 percent of the contents. The twigs and leaves of this evergreen were found in the stomachs, but very likely the berries were also eaten, It was found browsed in early October. Perhaps it would be eaten more extensively if it were more abundant on high ridges and grew taller. Crowberry is most plentiful on the lower slopes.

Figure 27: A swale, between lateral ridges of Igloo Mountain, where sheep often fed in the fall on the willow which protruded through the snow. [September 19, 1939.]

Mountain Avens.—Mountain avens (Dryas spp.) was found in 19 stomachs averaging 0.4 percent of the contents. There are about five species of dryas in the park. Some grow mainly on the old river bars where they form extensive mats, while others are found on the ridges. The principal cover of many slopes is dryas, consequently these slopes are poor winter range for sheep. Because it is much eaten in summer it has sometimes been referred to as an important winter food. However, in winter it becomes dry and brittle and loses much of its palatability.

Loco.—Loco (Oxytropis) was found in seven stomachs. Some species were highly relished in the fall while the leaves were still green. The remains found in the stomach contents consisted of the basal part of the plant nipped off at the root. Dry leaves may have been present in other stomachs and escaped detection.

Hedysarum.—Both Hedysarum americanum and H. mackenzi are highly relished in the fall but none was found in the stomachs. The seed pods were often noted eaten in September.

Dwarf Birch.—Dwarf Arctic birch (Betula nana) was found in only two stomachs, The leaves and fine twigs of dwarf birch were noted eaten in late September. It is not readily available on the high ridges.

The following species were found in the stomachs in small quantities: Saxifraga tricuspidata, Ledum groenlandicum, Loiseleuria procumbens, Potentilla fruticosa, Rhododendron lapponicum, Alnus fruticosa, Arctostaphylus uva-ursi, Aster sp., Cladonia sp., leaf lichen sp., and moss.

The following species were found eaten in late fall: Heracleum lanatum. (seed heads and leaves), Delphinium sp. (when dry), Shepherdia canadensis (mainly berries), and Mertensia sp.

SUMMER FOOD HABITS

In summer, as in winter, the principal food of the sheep Consists of grasses, sedges, and willows. However, at this season much Dryas is also eaten, and a variety of other species are eaten in small amounts. Frequently sheep were seen feeding on the leaves of willow. For example, on June 11, 1940, 36 sheep fed steadily on two species of willow for 40 minutes, and on June 20, 1940, a band fed for a half hour on willows. On some of the higher slopes where the shrubby willows are absent there is an abundance of dwarf willows only a few inches high which are grazed extensively. Some species of Oxytropis and Hedysarum are much sought. The leaves and flowers of herbaceous cinquefoil, Aconitum sp., Campanula lasiocarpa, Arnica, and Conioselinum. were eaten.

The contents of one stomach, that of a 2-year-old ewe that died on June 8 as a result of a snag injury, were available for analysis. The following determinations were made: green grass, 85 percent; willow, 14 percent; Oxytropis sp., trace; Pyrola sp., trace; and Salix reticulata, trace.

MINERAL LICKS

Sheep were often seen at licks, and well-defined trails led to some of them. A lick on Ewe Creek was much used, and two in Teklanika Canyon, similar to the preceding, were used extensively. On East Fork the sheep visited a lick on the side of the slope where water seeped out. The mud in this lick was black, while in some of the others the material was gray. On Double Mountain a lick was used by both sheep and caribou. On Toklat River two licks were also found in use.

Above the Ewe Creek lick many sheep droppings were found which were made up entirely of clay, so it is evident that considerable quantities of the earth are eaten at times.

Dixon (1938, p. 223) had samples of the Ewe Creek lick analyzed by Dr. G. L. Foster of the Division of Biochemistry, University of California. He reported that calcium and iron phosphate were the two minerals present in the lick which would be soluble in digestive fluids. Insoluble substances, chiefly magnesia and silicates, were also present.

Carrying Capacity of the Winter Range

After observing so many overgrazed big-game ranges in the States it was a pleasant experience to find the sheep range in Mount McKinley National Park not overutilized. By spring the grasses on the more accessible ledges are closely grazed but they are not harmed even in these places where grazing is concentrated. Willows are not now overutilized, although in years past, when sheep were more abundant, patches on the ridges were killed off by overbrowsing.

There are now from 1,000 to 1,500 sheep using the winter range which comprises an area of between 200 and 250 square miles. But of course not all of this area is a source of food. Many cliffs and rock slides support little growth; the vegetation on many slopes, ravines, and stream bottoms is deeply covered with snow; and extensive areas have a cover which is mainly Dryas, thus furnishing little winter grazing. Hence, without an extensive survey of cover types, it would be difficult to determine the amount of range actually available to the sheep in winter, and this amount would fluctuate from year to year because of varying snow conditions.

Figure 28: Cathedral Mountain in foreground; south extension of Double Mountain on center sky line. Dryas predominates on so many slopes that Cathedral Mountain is a rather poor winter range for Dall sheep. [September 1939.]

The carrying capacity of the winter range is several times the number of sheep now present in the park. As will be discussed later in detail, it is the wolf, rather than the food supply, that appears to be the chief factor limiting the size of the sheep population. The wolf reduces the size of the sheep range by confining sheep to the more rugged country where they are less vulnerable to wolf attack than elsewhere.

In the rugged country, the wolf preys on the sheep sufficiently to keep them below the carrying capacity of the range as determined by food supply. Thus the present sheep numbers apparently are dependent on the extent of cliff protection and the degree of wolf pressure.

Disease and Parasites

ACTINOMYCOSIS

From an examination of the skulls, it is obvious that the sheep are subject to two diseases, common among big game animals, which are known as actinomycosis and necrotic stomatitis. These ailments are described in detail on p. 117. Much of our information on necrotic. stomatitis was secured among the elk in Jackson Hole, Wyo., where it was also found among the deer, bighorns, and moose. In the case of elk, many calves are subject to it. Death among them is often rapid, occurring before the lesions have affected the bone tissue. Many older animals which have recovered are often left with serious dental deformities or loss of teeth. The annual loss among the elk due to necrotic stomatitis apparently becomes more severe during years of food scarcity. The data available indicate that the Dall sheep may be affected by this disease in a manner similar to the elk. The annual loss due to this disease is not extensive.

LUNGWORM

In the lungs of a yearling collected April 24, 1939, and examined by Dr. Frans Goble, numerous eggs, larvae, and adults of a nematode were found which apparently belonged to the genus Protostrongylus or a close ally. (Goble and Murie 1942.) This is the first record of lungworm in the Dall sheep, but the parasite is probably common among them. Lack of previous records is probably due to lack of material for examination. Lungworms might affect the health of some animals, but no data on this were secured. Coughing, so prevalent in the sheep in Yellowstone National Park and Jackson Hole, Wyo., was not observed.

SCABIES

No animals were seen which showed any evidence in the appearance of the coat that scabies existed among them. No mites were found in patches of hide examined. Scabies is quite prevalent among some of the Rocky Mountain bighorns.

A SICK LAMB

About the middle of June 1940, a sick lamb with its mother was seen by a ranger at a tributary of Igloo Creek on the east side of Sable Pass. The lamb was too weak to walk, and saliva drooled from its mouth. It was carried to the creek where it was seen the following day, unable to rise. When I learned about the incident the lamb was gone.

LAMB DIES IN CAPTIVITY

In April 1929, three weak ewes were rescued from starvation. Two lived and gave birth to lambs at Park Headquarters. One of the lambs did very well, but the other died on the second day after it was born. When it was discovered that the ewe had no milk it was too late to save the lamb. Possibly this type of tragedy accounts for the loss of some lambs at birth in the wild, especially after hard winters.

INSECTS SOMETIMES ANNOY SHEEP

On a few occasions flies were observed annoying the sheep considerably. On July 26, 1940, among the rock chimneys near Sable Pass all the sheep had sought the shade to avoid the large flies which were attacking them. Sheep out in the sun, where they had moved when I disturbed them, were much troubled by these pests. In desperation they would hurry to the shade of a rock. A ewe climbing a long slope opposite me was much bothered by flies, and stopped in the shade of the first cliffs she reached. An old ram lying in the shade was directly in the route I planned to follow to climb out of some rough cliffs into which I had descended. He moved off at my close approach, but when the flies attacked him in the sun he suddenly hurried back to the shade near me. As I approached him he faced me and looked as though he might hold the passage at all costs.

On July 19, 1939, I saw a ewe stamping her foot and watching an insect buzzing around her. She dashed rapidly away, then stood perfectly still, apparently waiting to see if the fly had followed, Three times I saw this performance, in which the sheep behaved much like a caribou when attacked by a botfly.

As a rule the sheep did not seem to be greatly bothered by flies. On the high ridges there is generally enough breeze to greatly minimize this annoyance.

Mosquitoes are said to have attacked some captive sheep at Park Headquarters, but it is not probable that they are of much significance in the sheep hills, where they usually are not plentiful.

Accidents

In 1939 a number of crippled animals were observed. Strangely, no cripples were seen the following two summers except one which had been snagged and was found dead. This difference was due in part to the greater time spent among the sheep in 1939, and possibly, in part, to lighter wolf pressure on the sheep in 1939. The animals injured will be enumerated in order to give an idea of some of the accidents to which sheep are subjected and to show that there is a certain percentage of obviously vulnerable animals, and some accidental casualties.

A lone yearling was found near the road (on April 24, 1939, and it was so weak it could barely rise. When it did get to its feet it stumbled toward me (probably because I was between it and the rocks) and fell at my feet. The hoof of one forefoot was elongated, showing that it had not been used for a long time. One rib was broken and the area around it festered and bloodshot. Other spots showing serious suppuration were the injured foot and the knee, and two spots along the spinal column. There was a sore on each side of the upper palate and sores around the lips. The bruises and broken rib suggested that the yearling had fallen in the rocks.

Figure 29: A yearling sheep, whose right foreleg had been injured, apparently by a fall. The foot had been so little used that the hoof had grown long. [Polychrome Pass, April 24, 1939.]

On May 13, 1939, a ram about 5 years old had a decided limp on a foreleg.

On June 18, 1939, a young ram on the east side of Cathedral Mountain was lame in a foreleg. When the animal fed, it either held this lame leg off the ground or rested on the knee.

A ram about 6 years old, observed at East Fork on June 27, 1939, had a horn which, instead of curving outward, curved toward the neck so that in time, as it grew longer, it might gouge into the top of the neck.

On July 26, 1939, a lamb was noted at the head of East Fork River which had a seriously lame foreleg.

On Big Creek on August 16, 1939, Emmett Edwards saw a lamb with a severe limp in a foreleg. On August 18 he saw a ewe with a large raw spot on the right ham which appeared to be the result of a severe bruise.

On September 11, 1939, a lamb was seen on Polychrome Pass with a long deep gash across the rear of one ham. The hair around the wound was red from recent bleeding. The cut appeared to have been made by a rock. Possibly the lamb had fallen.

On September 12, 1939 on Igloo Creek, a lamb carried a crippled foreleg, not using it at all.

On September 20, 1939 I noted an occasional drop of blood in the tracks of some sheep which had been feeding in the snow. Nothing serious, but it indicates that the feet may be come quite sore in crusted snow.

On September 29, 1939, a yearling was seen with a decided limp in a foreleg. A ewe in the same band was soiled from the tail to the ankle. In the distance it appeared to be dried blood, but of this I could not be certain. It may have been due to diarrhea.

On June 8, 1940, a 2-year-old ewe was found dead at Igloo Creek on a low slope. A snag had penetrated the abdominal cavity and the small intestine.

Some years a go a ranger found a sheep at Igloo Creek which had died from a fall incurred by slipping on a glacier.

At least two instances of sheep frozen in overflow on the rivers have come to my attention.

In September 1928, five young rams were found dead in Savage Canyon. They had died from eating dynamite.

In the Park Superintendent's report for 1927 it is stated that a ranger found the carcass of a sheep at Toklat River. When found, the carcass was fresh and there were no marks on it except where birds had picked a hole in the entrails. The tracks in the snow showed that the animal had walked off some glare ice a distance of 25 yards and had dropped dead. The suggestion is made that is may have slipped on the ice and hurt itself internally.

THE MENACE OF DEEP SNOW

The deep snows of the winters of 1928—29 and 1931—32 took a heavy toll of the sheep. Apparently in some cases entire bands, in crossing from one mountain to another, became exhausted in the valleys and died en masse. There is no detailed report on the situation but it is stated that bunches of sheep were found dead with no indication that they had been killed by wolves. Some of the sheep were no doubt later fed upon by wolves and foxes, making it impossible so know whether the animals had starved or been killed. Many of the sheep killed by wolves during these two hard winters were no doubt first weakened by snow conditions. Perhaps most of these sheep were doomed, but it is possible that some of those killed by wolves might have survived the severe weather.

In April 1927. one of the rangers found a ram and ewe in a weakened condition in the deep snow as Sable Pass. The ewe was brought to Headquarters but died a few days later. The ram was found dead when the spot was again visited.

In February 1929, when the snow was unusually deep, two rams were seen to jump on a drift of loose snow and sink completely out of sight. In such a circumstance they would be quite helpless if a predator were near.

In April 1929, three ewes and three rams were rescued at Igloo Creek in a weak condition due to deep snows. One of the ewes was too far gone and died in a few days. After the sheep had been kept a while at Igloo and regained some strength they were taken to Headquarters. They were later moved to the University of Alaska where they were kept for several years.

Ranger John Rumohr stated that in the spring of 1932 when the snows were deep he had seen 30 old rams on the low banks bordering East Fork River south of the road. Later he found five of them dead, untouched by any animal.

The deep snows have played an important role in the control of the sheep population. Probably the larger the population the more drastically are the sheep affected. A large population quickly devours the food available on ridge tops and they must all go down to the deeper snow in search of food. A smaller number might find sufficient food on the ridge tops so pull them through a severe winter. This is an example of one of the regulatory devices of Nature.

The Rut

Observations on the rutting of the mountain sheep were limited because during much of the rutting period I was checking on wolf movements along the north boundary, and snow made is difficult to cover much territory. The time of the beginning and ending of the rut was not learned. The rutting was probably early because the following spring the lambing period was early.

On November 15, 1940, I noted that the rut was in progress. On Polychrome Pass one old ram and 2 younger rams were with 5 ewes, and another old ram was very active rounding up a flock of 5 ewes. The following day considerable rutting activity was noted on East Fork River. A large ram herding 4 ewes walked a short distance to a large ram in possession of 4 other ewes. After standing around about 10 minutes he returned to his own harem just over the rise. One old ram had 7 ewes, and another was lying near a single ewe. Eleven large rams, 3 young rams, and 12 ewes were scattered over a small area observed. Some of the rams were not in possession of any ewes at the time.

Sheldon (1930) in 1907 found active rutting from about the middle of November to the middle of December. He noted several encounters between rams, and makes the following comments on the rut (p. 240); "Friendly rams remained among the ewes, serving them indifferently. The fights always occurred when stranger rams attempted to enter the bands. As times new rams intruded among the ewes and shared their privileges without apparent objection on the part of rams already with the band."

The rutting activities of the white sheep are apparently very similar to those of the Rocky Mountain bighorn. One ram may serve a number of females and two or more rams may serve the same one. The ewes come in heat over a period of about a month, but judging from the lambing time it appears that most ewes come in heat during a two-week period.

Gestation Period

In 1907 Sheldon (1930) observed the sheep actively rutting in the latter half of November and early December. How late in December active rutting took place is not known because Sheldon's observation of the sheep was interrupted by trips after moose. The following spring he observed the first lamb on May 25. In 1940 I observed sheep rutting in mid-November but do not have data on the duration of the rut. The first lamb the following spring was seen on May 8. The gestation period seems to be between 5-1/2 and 6 months.

Figure 30: Two ewes with their lambs which are only a few days old. As young lambs remain close to their mothers, eagles have little opportunity to attack them successfully. [Polychrome Pass, June 3, 1939.]

Lambs

PRECOCITY

Lambs a day or two old, so small that they can walk erect under their mothers, clamber up cliffs so precipitous that even the mothers can scarcely find footing. Not only can the lambs climb but they possess unexpected endurance. Frequently very young lambs were seen hurrying after their mothers from near the base of a mountain to its very top without resting.

The travels of a captive lamb illustrate the endurance they possess. On May 13, 1928, a ranger brought to his cabin a lamb only a few hours old. It soon became a pet and amused itself by leaping about on the chairs and beds. It followed him and his cabin mate around all day and showed no desiretso join the other sheep that were sometimes in view. It was raised with a bottle, on powdered milk. When only 2-1/2 weeks old it followed the men for 30 miles over rough ground and through glacial streams. When a month old is became wet in a glacial stream when overheated and died of "pneumonia."

It surprised me when I first observed the speed of the lambs on relatively smooth slopes. On May 28, 1939, when most lambs were a little less than 2 weeks old, I came suddenly upon six or seven ewes and their lambs. My nearness gave them quite a fright, causing them to flee full speed down a gentle slope to the bottom of a draw and up the other side. Although the ewes galloped at full speed, the lambs kept up easily and at times two or three of them even forged out ahead. Upon reaching the bottom of the ravine and starting up the other side, the lambs seemed to have an advantage and "flowed" smoothly up the slope ahead of the ewes. When the group finally stopped to look around the lambs were as fresh as ever, and seemed anxious to make another run. This precocity of the lambs is, of course, of great value in avoiding wolves and other predators.

Figure 31: Ewes with lambs frequently congregate on rugged cliffs when the lambs are young. When they go off to feed, some ewes leave their lambs in the care of other females. [Polychrome Pass, May 1939.]

EWE-LAMB RELATIONSHIPS

During the first week or two, ewes that have lambs are especially wary. When approached, they hurry off with their lambs, often crossing over to another ridge so as to leave a draw between them and the intruder, or else climb high into rough cliffs. Before the lambs are born these same ewes might be rather tame, and as the lambs become older they again become tame.

The lamb is not left lying alone as are the young of deer, elk, and antelope. It remains near its mother, pressing so close to her at times when traveling that it may be almost under her. When resting, the young lamb usually lies against the mother or only a few feet away. This habit of remaining close to the mother is of high protective value should eagles attack.

When the lambs are young, several mothers often congregate in rough cliffs. In such places a few of the ewes may remain with the lambs and the others go out 300 or 400 yards to feed on the gentler slopes. On May 26, 1939, 13 lambs were seen frisking about on some cliffs on a rocky dike on East Fork. With the lambs were 6 ewes—3 lying down just above the group, and 3 standing in their midst. While I watched, one of the ewes moved off 150 yards and commenced to feed. A lamb followed her 10 yards, then dashed back to the others. One lamb ventured out alone 20 yards and a ewe at once became alert and moved four or five steps in its direction, whereupon the lamb hastened back to the group. Four ewes were feeding on a slope about 250 yards away from the cliffs. Occasionally one of them would look toward the lamb assemblage. From a point about a quarter of a mile away three ewes stopped feeding and came galloping across the face of a shale slope, calling loudly as they came. When they were still about 100 yards away two lambs galloped forth, each joining its mother and nursing at once. The third ewe did not have such an alert or hungry lamb, for she smelled of five lambs in the top group, then walked briskly down to the rest of the lambs 15 yards below where she found her lamb, who belatedly came forth to nurse.

Figure 32: A ewe and her lamb in the fall of the year. [Polychrome Pass, September 7, 1939.]

On June 5, 1939, there were 22 lambs on this outcrop of cliffs and an equal number of ewes were seen, but some of the ewes were feeding 200 or 300 yards away. One lamb on this occasion ran out about 60 yards to meet its mother and to nurse.

On July 12, 1939, a band of 52 ewes, lambs, and yearlings were feeding on some rather gentle slopes of Cathedral Mountain. The mothers frequently moved off 100 or 200 yards from their lambs which remained in a scattered group resting on the slide rock. Five or six ewes at different times stopped feeding and called loudly. Each time, a lamb would recognize the call of its mother and gallop down to nurse, usually for a minute or less. The sheep were quite noisy at times, the ewes "ba-a-ing" loudly, and the lambs answering more softly.

The tendency for ewes with lambs to bunch up is probably a natural outcome of their all having the same inclination to remain in the rougher terrain. Later there is more intermingling of the ewes with lambs and those without lambs.

Figure 33: A band of ewes, lambs, and yearlings on Cathedral Mountain. Vegetation on the loose slide rock is dominantly dryas. A wide variety of herbs and some grass are also present. Note how some of the lambs remain close to their mothers. Others rested alone while their mothers fed lower on the slope. [July 12, 1939.]

LOST LAMBS

When a ewe leaves her lamb in a group and goes off to feed she may lose her lamb temporarily. On June 7, 1940, I climbed to a small group of ewes and lambs which ran out of sight when I neared them. I was standing on the spot where they had been, when a ewe, which I had passed on the way up, approached me calling. There was not much room on the cliff where I stood so, when the ewe was seven or eight paces from me, I stepped to one side to let her pass. She was evidently perturbed at finding me where her lamb had been, for she called loudly and came toward me with lowered head which she jerked threateningly upward as though to hook me with her sharp horns. After I had warded her off with my tripod she stood for a few moments calling, then hurried in the general direction the others had taken and no doubt soon found her lamb, for the band had not gone far.

On June 13, 1941, I spent some time on Igloo Mountain observing about 50 ewes and 30 lambs. At first most of them were in a broad grassy basin. One ewe nursed her lamb, walked off 200 yards and fed with another ewe for an hour, never once looking toward the lamb. As I climbed, a small group of ewes and lambs, including the lamb of one of the two feeding ewes, moved upward, feeding higher and higher until they were out of my view. When the two ewes rejoined the main band which in the meantime had moved 200 or 300 yards to some cliffs, one of them found her lamb at once but the other was not so fortunate. Calling continually, she searched through the entire band, duplicating her visits to some parts of it, but still with no success. She finally recrossed a gulch and climbed in the direction taken by the little group with which her lamb had gone. She showed good sense, and no doubt soon found her lamb. After the sheep had rested in the cliffs for some time another ewe suddenly commenced to call at short intervals from a prominent point. The ewe had called a long time when two lambs came forth from a large crevice in a rock a short distance away where they had been lying. The ewe saw them emerge and hurried to join them. One of them was her lamb and it immediately nursed.

Figure 34: Ewe, lamb, and yearling plow through a September snow. [Igloo Creek, September 19, 1939.]

PLAY

The lambs play a great deal, romping with speed and agility. Sometimes they butt each other, coming together after a 2- or 3-foot charge. Occasionally a ewe may play with the lambs. Once one played with three lambs, dashing after one, then another. A lamb would leave its mother, get chased, and rush back again to its parent. Apparently is was great fun.

WEANING

Lambs begin to nibble grass when they are a week or 10 days old, and as the summer progresses they feed more and more on green vegetation. Some of the lambs continue nursing well into the winter and I have seen them when they were yearlings trying to nurse. Sheldon (1930, p. 278) reports lambs nursing in February. Weaning is not an abrupt process. I expect that long before nursing is discontinued the lambs could get along very well without this nourishment. The lambs remain with the mother a year and sometimes in the case of female yearlings, another 6 months. It is not uncommon to see a ewe, yearling, and lamb together during the summer and fall. Often yearlings, 2-year-olds, and other youngish animals, along with barren ewes, may be found traveling together. In the spring male yearlings sometimes join the bands of rams.

Relation with Animals of Minor Importance

SNOW BUNTINGS COMPETE FOR FOOD

On October 11, 1939, a flock of about 100 snow buntings (Plectrophenax n. nivalis) in a high pass were feeding on the heads of Calamagrostis. A large patch had been quite thoroughly threshed by the snowbirds. Here was competition for food between animals which we would not expect to have much relation with one another. I had noticed that the sheep had been feeding extensively on these same grass heads which stuck up through the snow. Calamagrostis is a very common grass, so there is much of is available. But it so happened that the particular patch eaten by the snow buntings was up high where it would be available to the sheep when deep snows covered that growing lower down. However, this competition is no doubt extremely slight and is mentioned here only to show an unexpected relationship.

WOLVERINE AND SHEEP

The wolverine (Gulo hylaeus) is widely distributed in the mountains in the park. In 1940 and 1941, after a period of low numbers, is was reported to be on the increase. It was reported as abundant in 1927 and 1928, but I do not know when the numbers became reduced. Wolverine abundance may have been correlated with the high sheep population when much carrion was undoubtedly available. Several times I found the scattered remnants of a sheep carcass which had been investigated by a wolverine, but I have no authentic data on the relationship of the wolverine to the sheep. One circumstance cited as evidence that wolverines prey on sheep is that the former are commonly found on mountains near timber line where mountain sheep range. However, it should be added that wolverines occupy a similar habitat in areas where there are no sheep. The number killed by wolverines under any circumstance is undoubtedly small.

LYNX AND SHEEP

During the period of this study the lynx (Lynx canadensis canadensis) was very scarce; not a single track was noted.

In the winter of 1907—8, Sheldon (1930) found two sheep which had been killed by lynxes. One of the victims was a 2-year-old ewe and the other a ram, about 2 years old. In each case, the predator had lain in ambush, leaped from above to the sheep's back, and had bitten the sheep around the eyes.

Ordinarily the lynx feeds on snowshoe hares, but when the hares become scarce during the ebb of their cycle, the lynxes are left without this usual food supply. Under these conditions, they are forced to seek other food. The fact that hares were scarce in 1907—8 probably accounts for the lynxes hunting sheep at that time. After the hares disappear the lynxes also go into a decline, so their predation on sheep is only for a short time and probably is never a serious factor.

In 1927 there was a big die-off in the hare population. It is reported that hares were unusually abundant in the fore part of the summer but that during August they became extremely scarce. It is interesting to note that in the Superintendent's reports for 1927 and 1928 the lynx was reported increasing. Possibly due to the hare die-off, lynxes began to wander about in search of food, especially in the sheep hills, and so were noticeable. After 1928 there is no mention of lynx abundance, and they have been scarce since that time.

COYOTE AND SHEEP

The coyote, Canis latrans incolatus Hall, was so scarce in the sheep hills that there was no opportunity to study its possible effect on the mountain sheep population. I saw a coyote on three occasions at Sable Pass, and once near Teklanika Canyon. These were the only coyotes seen during the three summers and one winter which I spent in the field. In recent years they have been reported somewhat common near the northeast corner of the park where snowshoe hares have been present in moderate numbers; and in the low country north of the park a few are reported by trappers.

The coyote frequently has been listed as a destroyer of the Rocky Mountain bighorn in the States, but the few studies which have been made indicate that the coyote is not a serious enemy of these sheep. In Yellowstone National Park I found no evidence of coyote predation on the bighorn, not even in summer when lambs in some situations seemed to be vulnerable to coyote attack, and coyotes on these summer ranges were common (Murie, 1941). Other investigators in Yellowstone National Park have not found evidence that coyotes were a serious menace to the mountain sheep. In Jackson Hole, Wyo., coyotes were suspected of destroying many lambs, but investigations indicated that the lambs were dying of disease rather than coyote predation (Honess and Frost, 1942). In a study of a prospering bighorn herd in Colorado, where coyotes are common in the sheep area, known losses from coyotes were small (Spencer, 1943).

The scarcity of coyotes in the sheep hills in Mount McKinley National Park may indicate that they cannot find sufficient food there. Of course in the absence of the wolf, there would be more weak sheep and carrion available to them, and more coyotes might then occupy the sheep hills. For this food supply they probably cannot compete with the wolf, since they are a weaker predator in respect to the sheep. It is possible that the wolf tends to drive out the coyote by attacking it, although both species once occupied the same general territory in the States. I am inclined to believe that the main reason that the coyotes are not in the sheep hills is that the staple rodent supply is more abundant in the lower country. Possibly coyote distribution is much influenced by the distribution of snowshoe hares also.

In 1932, when many sheep were succumbing to the unusually severe winter conditions and others were weakened by them, it is reported that some sheep were killed by coyotes. Such predation on weakened animals is not at all unlikely, but the information at hand does not indicate any consistent or serious predation by coyotes on the mountain sheep herds of Mount McKinley National Park.

GOLDEN EAGLE AND SHEEP

Lambs Well Protected When Young.—Charles Sheldon (1930, p. 33) writes as follows concerning his observations on relationships between the golden eagle (Aquila chrysaetos canadensis) and white sheep: "While hunting sheep in the Yukon Territory during the two previous years, and also while in the Alaska Range that summer, I observed the relations between golden eagles and sheep, not once noticing any antagonism between them—but only complete indifference." On May 25, 1908, Sheldon saw an eagle swooping at some ewes and lambs and states that it was the first time that he had seen an eagle swoop at sheep or even notice them. The ewes stood over their lambs to protect them from the eagle. On June 7, 1908, Sheldon (1930, p. 382) reports seeing eagles swooping at lambs but states that no capture was made, and that the ewes were very watchful when the eagle was near. An eagle's nest was nearby and near it Sheldon found lamb remains. He shot the eagle on the nest and found the stomach filled with ground squirrel remains. He concluded that it was probably true that the eagles preyed heavily on young lambs but that they were seldom molested by them after they were a month old. The evidence to support his conclusions that "it is undoubtedly true that golden eagles take a heavy toll of the newly born lambs" was his observation of some eagles swooping at ewes and lambs, and the lamb remains found at the one nest. His other experiences had caused him to conclude that eagles did not molest the sheep.

Joseph S. Dixon (1938) spent the summers of 1926 and 1932 making wildlife studies in Mount McKinley National Park. In 1932 he made observations on four eagle nests but found no fresh lamb remains in or below them. However, it should be taken into account that the lamb crop that year was very poor, so fewer lambs than usual were available to the eagles. Dixon (1938, p. 46) draws the following conclusion concerning eagle-sheep relationships; "Our experience in the region both in 1926 and in 1932 indicated that during these two seasons lambs were rarely taken by eagles, which were found to live chiefly upon ground squirrels and marmots."

Former Ranger Lee Swisher stated that he had visited several eagle nests but never found any lamb remains at them.

I have observed eagles dive at ewes and lambs, but such maneuvers do not necessarily gage the degree of eagle predation on lambs. Eagles have also been observed swooping low over grizzlies and wolves at times when there was no intent of predation. Once an eagle dove at an adult wolf which was standing near its den. About a dozen times the eagle swooped, barely avoiding the wolf which each time jumped into the air and snapped at it. The eagle turned upward at the right moment to avoid the leap, and apparently was enjoying the game.

During the first few weeks the lamb remains close to its mother. It usually lies down beside her when she is resting, follows her closely, or lies down only a few feet away while she feeds. In traveling, the young lamb often presses close to its mother's side, sometimes appearing to be partially under her. The lamb is near its mother except when left with other lambs, usually on a cliff, while she goes off a short distance to feed. But at such times the group of lambs is watched over by some of the ewes. So, at the time when the lamb would be most vulnerable to eagle attack, it is generally well protected, giving to the eagle little opportunity to prey on it.

Figure 35: "Come on!"

After a few weeks the lambs move about with more freedom and gambol over the meadows in little groups. Judging from their behavior, they are not greatly worried about eagles after 3 or 4 weeks. In late June and early July I have seen eagles fly low over lambs, separated from their mothers, without attempting to strike and without alarming them. Throughout the summer the eagle may occasionally dip downward at the sheep just as is does at other animals. On September 10 an eagle swooped low over two ewes and a lamb, giving them quite a start. It sailed close over them several times, calling as is passed. But after the first start the sheep seemed unafraid.

Few Lamb Remains Found at Nests.—Upon examining the vicinity of seven unoccupied eagle nests only one lamb bone was found. It was a skull which lay in slide rock some distance below the nest.

As the 13 occupied nests examined during the three summers, one old leg bone of a lamb was found below one nest and at another were the remains of two recently eaten lambs. The presence of two lambs at the one nest may have indicated that this particular pair of eagles was more inclined to attack lambs than were other eagles, or else were more fortunate in finding unprotected lambs, or possibly dead ones. One eagle was seen feeding upon a lamb which appeared to be about a week mild. As 11 of the 13 nests occupied no lamb remains, old or recent were found.

It should be pointed out that in some cases the eagle eggs do not hatch until the lambs are 3 or 4 weeks old and probably past the danger period. In these instances possibly lambs would not be taken to the nest. However, they might be brought to favorite perches near the nest where some of them would have been found if any significant number had been consumed.

During the period that the eagles are at the nest the parent birds bring to it twigs of birch, heather, or willow, probably to cover debris accumulating in it. In one nest an eagle had brought several mats of shed sheep hair, the presence of which could easily have been misconstrued as evidence of predation.

Eagle Pellets Contain Few Lamb Remains.—During each of the three summers spent on the study, substantial numbers of eagle pellets were collected from perches and nests in the lambing area. The pellets give a fairly good index of the food habits of the eagle during these 3 years. Of 632 pellets analyzed, only 6 (2 each year) contained remains of lambs. This low incidence of lamb in the eagle pellets strongly corroborates the evidence from nest examinations and general observations. The considerable information available supports the conclusion that only occasionally does an eagle feed upon a lamb. Such lambs may have been carrion or may have been killed by the eagles. If killed, they may have been healthy, weak, or deserted. In any event, whatever eagle predation exists, it is apparent that it would have no appreciable effect on the mountain sheep population.

In two recent publications dealing with the Rocky Mountain bighorn both of which have come to my attention since the above was written, no evidence was found that eagles preyed on bighorn. In a study of sheep in Wyoming (Honess and Frost, 1942, p. 56) the following statement on bighorn-golden eagle relationships is given: "To date no case of predation by eagles has been seen by a Survey member nor has one been reported for the Crystal Creek area." These eagles were living mainly on jack rabbits and ground squirrels. In a Colorado bighorn study (Spencer, 1943, p. 9) the following statement is made; "The hunting and food habits of the golden eagles in the Tarryall Mountains were observed and studied with a great deal of interest because many people consider these birds responsible for considerable predation on the bighorn sheep, especially the lambs. During the entire period of this study, not a single eagle was observed to attack or molest the bighorn sheep in any manner. The sheep were not disturbed when eagles came close so them, although the ewes were alert even when a raven came close to the lambing grounds." A study of the food habits of eagles nesting close to the lambing grounds showed that they were feeding mainly on prairie dogs. As many as seven prairie dogs were observed in the nest at one time.

Wolf and Sheep

Some of the information on the relationships between sheep and wolves that we should like to have would take several years to gather. Ideally, one should study the sheep in the absence of wolves for a few years, and in the presence of them for a similar length of time, and in several localities. What is not evident in one area is often readily revealed in another. But some data have been obtained which reveal their method of hunting mountain sheep, the sections of the sheep population suffering the heaviest mortality from them, and the lamb crop and its survival in the presence of wolves.

Significant observations on the actions of these animals in Mount McKinley National Park, and the wolf's hunting methods (with a number of hunting incidents) are enumerated below. This is followed by data regarding the portion of the white sheep population most subject to predation, based on examination of 829 sheep remains. Finally, consideration is given to the lamb crop and the question of its survival.

REACTION OF SHEEP TO WOLVES

A few incidents are here related which do not involve hunting but show how sheep and wolves sometimes behave when they meet.

On May 7, 1940, a scattered band of sheep moved slowly from one ridge to an adjoining one. The movement was so definite and consistent that I suspected the sheep were moving away from danger. A little later I peered into the draw below the ridge first occupied by the sheep and saw a black wolf investigating some cleanly picked sheep bones. The sheep had simply preferred to have a ridge between themselves and the wolf.

While eating breakfast at the Igloo cabin on May 3, 1940, a wolf was heard howling nearby. Stepping outside, we noticed three alert rams, each on a pinnacle, peering intently below them. They continued to watch for some time, evidently keeping an eye on the wolf we had heard. They apparently felt safe where they were even though the wolf was directly below them.

On August 3, 1940, Dr. Ira N. Gabrielson and I watched a black wolf trot leisurely down a short draw on the ridge opposite us and descend the narrow stream bed bordered by steep slopes. Two rams on the slope below us watched the wolf, and when it trotted out of their sight they moved to a point where they could again see it. Seven other rams grazing a short distance from the two paused but briefly to look. The wolf stopped a few times to look up at the rams, but continued on its way until finally hidden by a ridge. The rams and the wolf had shown a definite interest in one another, but that was all. The wolf probably examines all sheep in the hope of discovering an opportunity for a successful hunt, and the sheep keep alert to the movements of the wolf so as not to be taken by surprise or at a disadvantage.

On June 29, 1941, about 60 ewes and lambs on the south side of Sable Mountain all moved up the slope 100 yards or more and stood with their attention centered on the terrain between us. A search with the field glasses revealed a gray wolf loping westward between me and the sheep, about a half mile from them. Some of the sheep soon began to feed, others watched until the wolf had passed. The day was dark and rainy, so that the wolf, whose legs and lower sides had become wet from the brush, was unusually hard to see, yet the sheep had quickly discovered is.

Former Ranger Lee Swisher told me that he had seen six wolves suddenly come close to seven rams feeding out on some flats at Stony Creek. The rams bunched up and the wolves stopped 100 yards away. They made no move toward the rams, which, still bunched up, walked slowly and stiffly toward the cliffs. The slow gait was maintained until the cliffs were almost reached, then the rams broke into a gallop and quickly ascended the rocky slope. The incident seems to indicate that a wolf may to some extent recognize the ability of the rams to defend themselves. The wolves on this occasion may not have been hungry; possibly under other circumstances they at least would have made some attempt to single out one of the rams.

On one occasion Mr. Swisher said he let a sled dog chase some rams. They turned and faced the dog with lowered horns. After thus threatening the dog the rams started up the slope, and when the dog followed, they again turned on him. This incident is indicative of what the seven rams in the above incident would have done had they been attacked.

A dog belonging to Joe Quigley, a miner, is said to have escaped one night from his camp in the sheep hills. The dog returned 2 or 3 days later all battered. Sometime following this event when the team was driven up to some sheep carcasses this dog was not at all anxious to approach them. The inference drawn was that the bruises the dog had suffered during his absence had been administered by a sheep.

On June 19, 1939, 22 ewes and lambs were seen feeding among the cliffs a short distance above four resting wolves, one of which was lying only about 200 yards away. The sheep had already become accustomed to the presence of the wolves when I saw them for they grazed unconcernedly. Their confidence was probably due to the proximity of exceptionally rugged cliffs to which they could quickly retreat should the wolves attack.

On August 3, 1939, a band of 20 sheep ran up the slope of a ridge bordering East Fork River. A little later a wolf climbed the slope, making slow progress. Twelve sheep watched from a point up the ridge, three from some rocks not far from where the wolf went over the ridge top. Two eagles swooped at the wolf a number of times, continuing to do so after the wolf was out of my view so that its progress could be followed by watching the eagles. The sheep quickly resumed grazing. They had not moved far from the wolf but had watched to see what it was up to.

HUNTING INCIDENTS

Five Wolves on a Hunt.—On the morning of September 15, 1939, five wolves (the East Fork band) trotted along the road to Igloo Mountain then climbed half way up the slope which was covered with several inches of snow, and followed a contour level. I saw them a mile beyond Igloo cabin; three were traveling loosely together, a little ahead of the other two, Sometimes they were strung out, 50 yards or more apart. Generally they trotted, but occasionally broke into a spirited gallop as though overflowing with excess energy. Opposite me they descended to a low point, and the two gray wolves which had brought up the rear dropped to the creek bottom. A black one rounded a point and came upon three rams which it chased. The animals then went out of sight, but in a few minutes the wolf returned and I saw the rams descend another ridge and cross Igloo Creek. As they climbed a low ridge on Cathedral Mountain they kept looking back. When crossing the creek bottom they had not been far from the two gray ones which traveled a mile up the creek and then returned to join the others.

After chasing the three rams, the black wolf joined another black and the dark gray female and all moved up the slope. One of them stopped to howl, possibly a call to the two grays on the creek bottom. These two turned back about that time and later joined the others.

On a pinnacle of a high ridge stood a ewe peering down at the approaching wolves. She watched a long time but moved away while the hunters were still far off. The wolves went out of view on the other side of this same ridge. Later I saw some ewes farther along the ridge looking steadfastly to the west in the direction the wolves had last taken, and beyond these ewes there were three more gazing intently in the same direction. The sheep had fled to the highest points and were definitely cautious and concerned because of the presence of the wolves, which seemed to he coursing over the hills hoping to surprise a sheep at a disadvantage. The day before, I had seen a lamb in this vicinity with a front leg injured so severely it was not used. Such a cripple would not last long if found by these wolves. The habit of cruising far in his hunting gives the wolf opportunity to find weak sheep over a large range and to come upon undisturbed sheep, some of which he may find in a vulnerable location.

Sheldon observes Wolves Hunting.—Charles Sheldon (1930, p. 315) tells of following trails of two separate wolves in March and finding that on nine occasions they had chased sheep unsuccessfully. On eight of the chases they had descended on the sheep from above. He said that the sheep in the region had become badly frightened and that "most of them kept very high." So many unsuccessful hunts suggests that these two wolves were testing out each band, hoping eventually to find a weakened animal or gain some advantage. It appears that wolves chase many sheep unsuccessfully and that their persistence weeds out the weaker ones.

Sheep Escape a Wolf.—On October 7, 1939, I saw the track of a single wolf that had crossed Igloo Creek and then had moved up the slope of Igloo Mountain. After following the creek about a mile, I saw two ewes and two lambs on a spur of Igloo Mountain watching a black wolf which was curled up on a prominent knoll on the next ridge about 200 yards to the west. The two ears of the wolf were turned in my direction, so in order not to alarm him I walked along as though I had not seen him until I was out of his view. Then I doubled back close to the bank and ascended the draw toward him. But when I came near the knoll he was gone and the sheep were lying down.

After back-tracking the wolf, I deduced that its actions were as follows: After crossing Igloo Creek, it climbed part way up the mountain. It followed a trail along a contour at the edge of some spruces near the point where a wolf (perhaps the same one) had surprised a lamb among the spruces a few days before. The wolf crossed some draws and small spur ridges and arrived at a ridge on the other side of which the four sheep were feeding in a broad swale. The wolf climbed up the ridge, out of sight of the sheep, for 50 yards, so that he was slightly above them. He then advanced slowly until within 150 yards of the two ewes and two lambs and then galloped down the slope toward the sheep. The latter had escaped to the next ridge from which they were watching the wolf when I first saw them. The wolf chased up the slope after them but a short distance and then continued westward to the next ridge where he curled up in the snow. This time the sheep had the advantage and escaped. It is significant that they did not run far beyond the wolf; apparently they were confident that they were safe since above them was much rugged country. The behavior of the sheep is definitely conditioned by the terrain they are in and their position in relation to the enemy. Approach a sheep from above and he feels insecure and hurries away. A sheep on a flat is much more wary and timid than one in rugged country.

A Lamb Surprised in Woods.—On October 4, 1939, by back-tracking a wolf I found a freshly killed male lamb. The tracks in the snow plainly told this story: The wolf was following a trail along the side of Igloo Mountain within the edge of the uppermost timber. Suddenly he came upon five or six sheep feeding among the trees. Those above the trail ran up the slope to safety, but a lamb which had been feeding farthest down found itself cut off from a possible escape to the top of the mountain and was forced to run on a contour in the direction from which the wolf had just come. In chasing the lamb, the wolf was able to gallop back over the easy trail it had been traveling and thus keep above the lamb which was endeavoring to swing upward in front of it. The lamb traveled parallel to the trail over terrain broken up by the heads of numerous small draws. In one place he veered slightly upward until he came to the trail, but he must have been hard pressed for he again turned downward and now the wolf followed him directly. The chase led gradually down the slope, the lamb apparently keeping as much altitude as possible in the hope of gaining the rocks above. But he was too hard pressed to cut upward ahead of the wolf. Finally he descended a steep gravel bank to the creek bottom, crossed and recrossed the creek, started up the steep gravel bank at an angle, and returned to the creek where he was killed. He ran slightly more than a half mile before being overtaken. This time the wolf had the advantage not only of coming suddenly upon the lamb from above, but also in having a trail to follow while the victim was galloping over rough brushy country. If the sheep had been feeding in the open as they almost always do, the wolf probably would have been discovered before he was so close to them. This is an example of a situation where the predator gains an unexpected advantage.

Two Old Ewes Killed.—On October 5, 1939, there were fresh tracks of foxes, wolves, and a grizzly in the snow, all leading up a small cliff-bordered stream flowing into Igloo Creek. It was evident that there was some special attraction near at hand. My companion and I proceeded cautiously and soon saw a wolf run off. Farther on we saw a lone lamb in the rocks. An eagle flew away, and then we noticed a grizzly chewing on the skull of a ewe. I hoped the bear would leave the skull so I could get the age of the animal and examine the teeth, but when he finally moved off and climbed a low promontory the skull was in his jaws. He soon finished crunching the bones, then climbed to a rock a little higher where he lay down and after a few minutes went to sleep. The skull was a little close to the bear to retrieve, but that difficulty was solved by two magpies which, in fighting over it, knocked it off the cliff to a spot where we could safely get it. After considerable searching we found the skull of a second ewe on a grassy knoll a few feet above the gravel bed.

The two ewes had been killed 50 yards apart. Nothing now remained except a few large pieces of hide, some legs, entrails, stomach contents, and the two broken-up skulls. The assemblage of animals gathered at the kill may have consumed all the meat, or perhaps the wolves and foxes had cached what was not eaten.

The snow on the ground made it easy to back-track the chase. The story was simple. The two ewes and a lamb had been feeding on Equisetum on a broad moist swale not far above Igloo Creek. (The stomach contents of the two victims were made up mainly of Equisetum.) The wolves had been following along the mountain slope at a level slightly higher than that where the sheep fed. Coming over a rise they spied the sheep feeding in the swale 150 yards below them. The tracks showed that the wolves did not start running until they were within 75 yards of the sheep. The latter galloped out of the swale and ran downward at an angle toward Igloo Creek which they crossed after descending a steep dirt bank about 100 feet high. The wolves followed directly after the sheep, but, instead of running among the large rocks in the canyon stream bed as the sheep had done, they ran alongside the rocks on the smoother ground covered with a sod of Dryas. A distance of 250 yards up this creek the two ewes were captured just before reaching some cliffs. The lamb escaped, and, as stated, we saw it alone a short distance above the carcasses. The ewes had run a half mile before the capture.

These two ewes had lived almost their full span of life, for one was 10 years old and the other was 11. The teeth had been used up, and they were no longer pushing out to compensate for wear on the surfaces. In one ewe the tooth surface was worn below the gum. In the other a molar had worn a "crease" in the palate. They were weak animals, less able to escape the wolves than a lamb, which itself is apparently in a vulnerable age class.

Old sheep killed by wolves.—Late in the winter of 1940, in the rugged draws on the north side of the Outside Range between Savage and Sanctuary Rivers, Harold Herning and Frank Glaser found much sheep hair, bloody pieces of hide, and wolf tracks, all indicating that the wolves had been killing sheep. No skull remains were found. The bones were no doubt hidden in the deep snow which filled the bottom of the draws. After the snow disappeared I hunted in these draws for skull remains. The skulls of four recently killed sheep and the hair remains of a fifth sheep were found. Two of the skulls were those of 11-year-old rams, one of a 12-year-old ram, and the other of a 12-year-old ewe. Here the wolves had apparently eliminated some animals doomed soon to die of old age.

Old Ram Chased Down Bluff.—On January 7, 1941, the tracks of five or six wolves were seen along the road toward Savage Canyon. A short distance above the canyon the wolves abruptly left the road and climbed a gentle slope to the top of an isolated rocky promontory. They chased a lone ram down its steep side to the creek bottom where I found remnants of hide, as well as the stomach contents and the skull. Apparently the wolves had seen the ram on this isolated bluff and had turned aside to circle behind him and cut off his retreat to high ground. He was 12 years old, past his prime, a weak animal. The method employed in capturing him—that of coming down from above and driving him down the slope—seems to be a typical hunting technique.

Figure 36: The lamb made it!

Ewe Captured.—On March 12, 1941, Ranger Raymond McIntyre found the carcass of a recently killed 6-year-old ewe at Toklat River. Tracks showed that two wolves had chased her down a steep slope and captured her at its base. The usual manner of hunting, coming down on the sheep from above, had been practiced. There was no indication of necrosis on the skull. Of course the sheep may have had some other weakness not apparent in the skull.

Lamb Captured in Migration.—On September 28, 1939, Mrs. John Howard of Lignite, Alaska, saw three sheep, at least one of which was a lamb, crossing a valley a mile or more wide, between the "outside" mountain range and the mountains up Savage River. A little later she saw two of the sheep galloping up the long gentle slope leading to the "outside" range. The lamb ran down the slope and disappeared in a draw. The two sheep continued to the ridge, but she did not see the lamb again. Later in the morning she saw two black wolves near the spot where the lamb had disappeared. One of these crossed the flat to the south and returned, followed by two pups. Mr. John Howard investigated and found the lamb partly eaten and saw the wolves nearby. Sheep are especially vulnerable when crossing valleys, for wolves can outrun them on the flats.

Sheep easily avoid a Lone Wolf.—Some observations made in the sheep hills bordering East Fork River on the morning of May 26, 1939, show that, on a relatively steep smooth slope, sheep are easily able to avoid a single wolf. With a companion I had climbed to the top of a ridge from which I had a view of some snow-free ridges on the other side of a small creek below me. I noticed a band of sheep resting on a smooth slope, slanting at about a 40° angle or less. While I watched, the sheep bunched up and ran off to one side about 30 yards. Through the glasses I saw a gray wolf a short distance above. He loped toward them and the band split in two, some going upward around the wolf, the others circling below it. When the wolf stopped, so did the sheep, only 30 or 40 yards from him. He galloped after the lower band, which ran downward and then circled, easily eluding him. Compared to the sheep, the wolf appeared awkward. After a few more sallies the wolf lay down, with feet stretched out in front. One band lay down about 70 yards above him, the other about 50 yards below him. Only one sheep in the lower band faced him; the others as usual faced in various directions. One sheep fed a little before lying down. The lower group consisted of five ewes, one yearling, and three rams. In the upper group were four ewes, four yearlings, one 2-year-old, and two rams. All rested for 1 hour. Then the wolf again chased the lower band, which evaded him as before by running in a small circle around him. A flurry of snow then obscured my view. When it cleared a few minutes later, the wolf was disappearing in a draw and the sheep were grouped on the ridge above him.

In a short time he reappeared and slowly worked his way down the ridge to the creek bottom. Nine ewes, each with a lamb, appeared on the ridge near the draw which the wolf had just left. The lambs were at the time only a week or so old but still they apparently had been able to avoid the wolf. The utter lack of fear exhibited by these sheep is quite significant, indicating that a single wolf can easily be avoided on a slope.

Sheep avoids sled dog.—The behavior of a ram chased by Charles Sheldon's dog is somewhat similar to the behavior of the sheep attacked by the lone wolf. Sheldon (1930, pp. 368, 369) gives an excellent description of the event, as follows:

At that moment, a short distance ahead, I saw a three-year-old ram crossing the divide toward Intermediate Mountain. Here was a rare opportunity to observe the actions of a sheep when chased by a wolf. Quickly taking the pack off Silas, I led him ahead to within a hundred yards of the ram, which had not yet seen us. Silas dashed at him full speed. The ram rushed toward the slope a hundred yards ahead. For a hundred feet the dog did not gain, but during the next hundred he gained at least twenty-five feet, and during the next hundred he was gaining rapidly. Although the dog was clearly the speedier of the two, I thought that the ram deliberately slackened his speed as he neared the slope, which was sharply inclined. The dog was not forty feet behind when the ram reached it. Up he went bounding for forty feet; then turned and coolly stood a moment to watch the dog, which was running up at almost equal speed. Then the ram turned and rather leisurely ran upward a hundred feet, gaining somewhat on the dog, who by that time was going much more slowly. This time the ram stood and watched until the dog was within twenty feet, then easily ran up another hundred feet and again stood and looked at the dog. Silas, however, was now only trotting and his panting showed that he could not run upward any more. Yet he followed the ram, which kept repeating the same tactics, never losing sight of the progress of the dog, until within a hundred feet of the crest, where a sharp projecting rock rose almost perpendicularly from the slope. The ram quickly climbed to the top and looked down at the dog, which now was only walking. Nor did he move when Silas reached a point fifty feet below him. Then the two stood looking at each other. Finally the dog turned and trotted back to us.

Not once, after the first burst of speed on the level, did the ram show any fright. When he knew he could reach the slope he was deliberate in every movement, and after reaching it he coolly played with the eager dog. After each advance, however, he was careful to turn and watch his pursuer. He seemed to know that the dog would soon give up the chase, yet I believe he did not credit Silas with the persistence he had displayed. The actions of the ram led me to suspect that a wolf would not have followed more than a few feet up such a slope, its experience, which Silas lacked, having taught it that a sheep could easily escape when once headed upward on a steep slope."

Highway an Advantage to Wolf.—At several places, notably Igloo Creek, Polychrome Pass, Toklat, and a stretch opposite Mount Eielson, the automobile highway passes through winter sheep range. In some places it cuts into the heart of the more rugged cliffs utilized by sheep. The highway favors the wolves in three ways. First, it gives them an easy trail along the entire winter range so that they can move more readily from one part of it to another. Second, it gives the wolves easy access into the cliffs themselves; they need not make a laborious climb to get among the sheep but can follow a smooth easy grade. Finally, each blind corner in the road—and there are many of them—is a hazard, for the sudden appearance of a wolf may give the sheep no time to escape. This is especially true when the sheep are bedded down on the road.

A ewe, lamb, and yearling were killed on the road at Mile 67 on September 20, 1939. The victims had been bedded down near a sharp corner. Four or five wolves had come around the corner, made a dash at the sheep and captured them before they had run more than a few yards.

Several sheep killed by wolves were found on and beside the road at Polychrome Pass and at Igloo Creek.

While the road affects seriously only a small part of the winter range, it is a good illustration of disruption of natural wildlife relationships by an artificial intrusion.

Diseased Yearlings Captured.—In a high draw three yearlings were found which had been killed in early spring. Two of them showed necrotic lesions on the mandibles so probably were not in good health when captured by the wolves.

Lone Diseased Ewe Captured in Cliffs.—On the snow-covered mountain slope above our cabin at Igloo Creek on the morning of October 14, 1939, I saw a fox feeding, and on the same slope a raven and some magpies. Obviously there was a fresh kill. We climbed to the spot and found the remains of a 9-year-old ewe affected with severe necrosis of the jaws. The horns were short and stubby, perhaps indicative of a prolonged chronic ailment. The teeth were irregularly worn. Some were long and sharp, others were worn down to the gum. One tooth was shoved entirely out of line. A premolar was broken off. An upper premolar was bent outward. Cavities along the teeth were packed with vegetation. The animal had been alone on the cliff (sick animals are often solitary). Five wolves had killed the sheep on a steep slope below some cliffs. They had crushed the skull and eaten the brains. After feeding, the wolves had curled up in the snow on a rocky spur to one side of the remains, about a quarter of a mile from the cabin. There were seven beds, but only five of them were coated with ice. In the other two beds the wolves had not lain long. Although the ewe had been found in rough country under conditions favorable for escape, she was so weak that the wolves were able to make the capture.

Figure 37: The sheep feeding on this gentle terrain have access to nearby slide rock and cliffs for safety. [Igloo Creek, June 13, 1939.]

DISCUSSION OF WOLF HUNTING HABITS

A few generalizations can be made concerning the methods of the wolves in hunting Dall sheep. It is my impression that the wolves course over the hills in search of vulnerable animals. Many bands seem to be chased, given a trial, and if no advantage is gained or weak animals discovered, the wolves travel on to chase other bands until an advantage can be seized. The sheep may be vulnerable because of their poor physical condition, due to old age, disease, or winter hardships. Sheep in their first year also seem to be specially susceptible to the rigors of winter. The animals may be vulnerable because of the situation in which they are surprised. If discovered out on the fiats the sheep may be overtaken before gaining safety in the cliffs. If weak animals were in the band, their speed and endurance would be less than that of the strong and they would naturally be the first victims.

A wolf hunting alone apparently can be avoided easily by healthy sheep on a slope. The lone wolf must find his animals at a decided disadvantage to be successful. Two or more wolves can hunt with much more efficiency. The method is to get above a sheep and force it to run down, for a sheep running upward can quickly outdistance the wolves and escape. Sheep on somewhat isolated bluffs where space for maneuvering is limited are in danger of having their upward retreat cut off and of being forced to run down to the bottom. A number of carcasses were found in situations suggesting that the sheep had been chased down bluffs of limited extent. My general observations indicate that weak animals are the ones most likely to be found in such vulnerable situations. They often lack the energy to climb to more safe retreats to rest. Where the wolf population is relatively large its pressure on the sheep is probably proportionately great, eliminating a high percentage of weak animals and capturing more strong animals surprised at a disadvantage.

Do the Wolves Prey Mainly Upon the Weak Sheep?

THE PROBLEM

In any predation problem involving big game it becomes important to learn if the predation is indiscriminate or if it affects primarily certain classes of vulnerable animals such as those past their prime, those weakened by disease, or the young. The possibility is generally recognized that through predation the weak and diseased are eliminated, so that in the long run what seems so harmful may be beneficial to the species. Perhaps the evolution of the mountain sheep has progressed to a point whore it is in equilibrium with its environment but still requires environmental stresses such as the wolf to maintain this equilibrium.

To learn what kind of predation is taking place and what effect it may be having on the population we must have information on the type of animals killed. There are difficulties in collecting quantitative data of this kind. First, it is hard to find a sufficient number of remains to give significant results. Furthermore, in the majority of cases, it cannot be determined whether or not the animal was killed by the predator. There are some criteria that can be used, but usually the evidence is missing. Bloody pieces of hide are a strong indication of wolf predation, but this evidence in time disappears. The location of remains may be significant. If they are found beneath a bluff the chances are that wolves maneuvered the animal down to the bottom, but one cannot be certain of the cause of death in these cases. After knowing the immediate cause of death one must ascertain the condition of the animal when it died.

The number of deaths which may be assigned definitely to the predator are usually a small proportion of the total. Such being the case, one must often broaden the scope of the study from determining the type of animal being killed by predators to the type of animal dying from all causes. If the results are highly uniform, then significant conclusions on predation can still be secured. For instance, if it is found that most of the animals were weak at the time of death, then it may be concluded that whatever predation there was affected the weak part of the population.

Fortunately not all evidence is destroyed when the animal is eaten. The age of the animal can be closely determined from the horns and, in the absence of horns, the age group can be approximated from the teeth. The condition of the animal in a great many cases can also be determined from the condition of the teeth and skull bones.

SEARCH FOR SHEEP SKULLS

Although I began to find an occasional sheep skull and other remains after arriving in the field, it. did not seem possible that enough skulls could be gathered for any quantitative treatment of them. But as I continued my excursions among the sheep hills I began to learn where carcasses were most likely to be found. Small tributary streams having narrow bars bounded by steep cliffs were the most likely localities. Skulls were found in all sorts of situations, but the best places were along these narrow streams, especially beneath the steep rocky slopes. About this time I acquired the services of two boys from a Civilian Conservation Corps camp. With their help I made a methodical search along all the likely streams and draws. Of course those streams on the best winter ranges yielded the greatest returns. In going up the narrow streams and draws one of us would walk in the center, and the other two would follow the sides of the stream bed. In places where numerous small draws cut up a ridge we sometimes would each follow a draw. There was competition to find the most skulls and the "best" ones, which were those showing the most severe diseased conditions. In time we exhausted all the more likely places and discontinued the organized search. But whenever it did not interfere with my main objective I usually followed a course over the hills where chances were best for finding skulls. Thus I continued to pick up a few of them in all sorts of locations.

When a wolf or bear (or both, as often happens) has finished with a sheep carcass there usually is not much left. Sometimes there may be only a few patches of hair or some broken pieces of long bones. But usually the entire skull or parts of it are present in the vicinity. The skull may be considerably damaged, so that the vital parts, the tooth rows and horns, are broken apart and scattered. Wolves often eat away the entire face and most of the brain case of even an old ram, and, if the wolves fail to chew up the skull, a grizzly may come along and do so. When the skull, or parts of it, were not found at the location of the largest bunch of hair or hide, or at the stomach contents, hunting proceeded farther and farther away, and we were often repaid for our perseverance. Knolls above the kill, where it was likely a wolf or fox would go to chew on a bone, were examined, often with success. I was always anxious to find at least one horn for age determination and, if possible, all the tooth rows, because sometimes two or three tooth rows might look normal and the fourth show a severe condition of necrosis. Hence it was desirable to have as many of the tooth rows as possible. In many cases only a single mandible could be found. Although the bones last a long time, the horns soften up when they remain on the ground where it is damp and disintegrate relatively soon. Many large ram horns found lying in the grass were so soft where they were in contact with the ground that their surfaces could be scraped away with one's finger.

The mandibles and upper tooth rows of the ram skulls, and sometimes the entire skull and horns, wore collected. Generally the skulls of the ewes were saved, since their horns and horn cores are not large and can be more easily carried.

Figure 38: These two horns show the variation in size in Dall sheep rams. The small horn is from a 7-year-old animal; the large horn is from a 4-year-old animal.

A total of 829 sheep skulls were gathered. All but about 100 of them were picked up the first year in the field.

Some of the skulls were recent in origin; others were of animals which had died several years ago. Those designated as recent were, for the most part, those of animals that had died since 1938, though some probably dated back to 1937. The accumulation of old skulls would be considerably greater if there were not so much movement of gravel in the stream beds during the spring freshets and summer rainstorms. Skulls lying in the loose gravel bars are usually carried away or completely covered with gravel. Fortunately, those on the grassy benches or the sides of the stream beds remain findable for a much longer period of time.

DETERMINATION OF AGE

The Dall sheep is an exceptionally suitable species for a study of predation because the age of the animals can be established rather accurately from an examination of the horns, and, in the absence of the horns, from the teeth, although the teeth, except for the first 2 years, are less satisfactory as an index of age.

Each year the horn makes its main growth during the summer at a time when much nutritive food is available. During the winter there is scarcely any growth in the horn, probably due to the poorer quality of feed available. This season leaves its mark upon the horn which in the early years may be a slight swelling, and in later years a groove, more accentuated than other wrinkles on the horn. Thus the annual growths in the horn made in summer are divided by annual growth rings formed in winter.

The horns of a captive ram which I examined did not follow the development sequence found in the wild, probably because there was not such a marked seasonal change in its food. The ram was 5-1/2 years old, yet had eight definite growth rings.

Figure 39: Two ewes, one showing diverging type of horns, the other, the more usual type, in which the horns diverge very little. [Cathedral Mountain, July 12, 1939.]

The base of the horn increases in circumference with age and the length of the annual growth increases the first 2 or 3 years, after which it tends to become shorter each year. The annual horn growths of a 12-year-old ram, measured on the outside of the curve, were as follows: First year, 1/2 inch; second year, 8-7/8 inches; third year, 6-3/4 inches; fourth year, 4-1/4; inches; fifth year, 3 inches; sixth year, 3-1/8 inches; seventh year, 2-1/2; inches; eighth year, 2-7/8 inches; ninth year, 2-1/8 inches; tenth year, 1-3/8 inches; eleventh year, 1 inch; twelfth year, 7/8 inch. The circumference of this horn was 9-3/4 inches at end of fourth year; 11 inches at end of fifth year; 11-1/2; inches at end of seventh year; and 12-5/8 inches at end of twelfth year.

The growth rings in ewes fall very close together in an old animal so that in the case of some horns it is difficult to get an accurate age determination. However, it is felt that the age can be rather accurately established in most cases.

Figure 40: Horns of a 6-year-old ewe. The upper one retains on the tip the entire first year's growth. Only an indication of the growth ring in the lower horn can be seen near the tip. The horns of many ewes and some rams have tips as in the lower specimen; this fact must be considered in determining age.

In determining age one must carefully examine the tip of a horn to learn how much of the first year's growth remains. The first year's growth is often worn almost completely away so that the annual growth ring at the end of the first year merges with the tip of the horn. Figure 40 shows the horns of a 6-year-old ewe in which the first year's growth is present in the upper horn and worn down to the annual ring in the lower horn. When the first year's growth of horn is worn away there is generally some indication of the annual growth ring at the tip. A clue to the amount of horn missing can be gained by examining the pattern of the annual growths present below the tip. Usually there is not much difficulty, however, in interpreting the tips of the horns.

When the horns are not found the age can be approximated by an examination of the teeth. The yearlings and 2-year-olds and probably the 3-year-olds can be definitely identified from the tooth formula.

A male lamb which died September 28 had horns an inch long. The premolars were fully erupted, and the first molars were just erupting. The lower molars were a trifle more advanced than the upper molars. There was scarcely any noticeable wear on the incisors.

In a yearling the first molars were fully functioning and showed some wear. There was no sign of the second and third molars.

In a 2-year-old female which died on June 8 the first and second molars were fully erupted, but the third molar was not. All the deciduous premolars except the third upper premolar on one side were shed. The first and second permanent incisors were functioning. The third and fourth deciduous incisors were still present.

Figure 41: The lower mandible is that of a 4-year-old ram and the upper one that of a 10-year-old ram. The bone has been chipped away to expose the molars. The last two molars of the 4-year-old animal retain practically their full length, while those of the 10-year-old animal are almost worn away. As the surface wears down the tooth pushes upward until, after 10 or 11 years, little of it remains.

In animals older than 2 or 3 years the approximate age can be determined by breaking away part of the jawbone and measuring the length of the molar teeth. In a young animal the roots of the molars reach to the base of the jawbone (Figure) 41. As the surface of the tooth wears away the tooth pushes upward, becoming shorter with age. The third molar in a 5-year-old measured 53 mm. in length; a 10-year-old, 17 mm.; and in a 13-year-old ram, 10 mm. There is considerable individual variation in the length of the molars so that one cannot allocate a skull to annual age classes without the possibility of some error. For instance, an 8-year-old sheep may have as long molars as a 7-year-old animal, or there may be an overlapping of 2 years. However, for deciding whether the animal should be placed in the old-age group or in the next younger group this method of age determination is very useful.

DETERMINATION OF SEX

The sex can be easily determined from the horns. The rams have large curved horns and heavy horn cores; the ewes have small, slightly curved horns and small horn cores. In the absence of horns, the sexes of sheep older than about 2 years can be determined by examining the mandibles. In the ewes, these bones, just anterior to the first premolars, are noticeably more slender than those of the rams.

SEGREGATION OF SHEEP REMAINS INTO CLASSES

Old Age Class.—An animal definitely past its prime becomes a member of the weaker part of the herd. Therefore, an old-age group can be segregated and placed in the category of weak animals which in this case are doomed to a relatively early death even in the absence of predation.

The oldest ram found was 14 years of age, and there were three of this age. Only one ram was found in the 13-year class, Many rams belonged to the 10-, 11-, and 12-year-old classes. It appears that few rams live beyond 12 years of age. When a ram is 11 or 12 years old, little remains of his teeth. Tooth wear indicates that an 11- or 12-year-old ram has about completed his life span, and animals a year or two younger are probably beginning to fail.

Four females were 12 years old; these were the oldest recorded. The growth rings on a female horn are close together when the animal is old so that a person could make a mistake of a year or so. Consequently what is recorded as a 12-year-old might possibly be a 13-year-old, but, even so, there were still only four in this class. While there are many rams in the 12-year age class, there are very few ewes reaching that age and there also are few ewes in the 11-year age class. These data on longevity strongly suggest that the ewes are shorter lived than the rams. Since few ewes pass the 10-year mark, it appears justifiable to place those 9 years or older in the old-age class. Possibly the old age group for rams should begin at 10 years, but in the interest of uniformity and simplicity they have been classified the same as the ewes. So far as these calculations are concerned, if old age for rams should be construed as beginning at 10 years, the change would not greatly alter the final figures.

Two- to 8-year Class—Animals in Their Prime.—Animals from 2 to 8 years old are placed in one age group. It is thought that such animals are in their prime and are least vulnerable to predation and other factors of the environment.

Lambs and Yearlings.—The young are arbitrarily called lambs until about October, and after that until they are about 15 months old, they are called yearlings. Most yearling skulls are from animals less than a year old, from animals which have died during the winter.

It is difficult to evaluate the vulnerability of the young sheep. Everything indicates that it is probably correct to call them weaker than healthy animals in the 2- to 8-year class. But lambs a few weeks old are not easily captured by wolves, so one would expect that after a few months they would be able to avoid wolves almost as well as the sheep in their prime. Perhaps the young sheep, up until early winter, are not especially vulnerable, although they would be handicapped by inexperience. But as the winter advances and the quality of the food decreases they perhaps become progressively weaker just as the adult sheep become progressively thinner. Because they are still growing they may be more affected by hardship than older animals. Many of course succumb to winter hardship, and apparently the strength of the healthier ones is somewhat reduced so they become less able to avoid the wolf. It is thought that on the average the term weak may be applicable to the yearlings so they have all been so classified in discussing the skull data.

Diseased and Injured Animals.—Actinomycosis: The ungulates in general seem to be subject to certain diseases which cause lesions in the mouth. These lesions may be caused by at least three diseases—actinomycosis, necrotic stomatitis and actinobacillosis. The mortality in necrotic stomatitis is high, while in the other two diseases the infection is often chronic. There is some confusion as to the causative organisms, but actinomycosis and necrotic stomatitis appear to be caused by two species of a ray fungus, respectively known as Actinomyces bovis and Actinomyces necrophorus. Actinobacillosis is caused by another organism of similar appearance called Actinobacillus lignierisi. Tissues in the mouth cavity are invaded by these organisms through abrasions such as are sometimes caused by coarse vegetation or grass awns. The lesions may occur on the tongue, hard palate, cheeks, and near the teeth, especially around the molariform teeth. Actinomycosis is thought to generally affect the bone, and the other two diseases the soft tissues. But there is a possibility that necrotic stomatitis, when chronic, also affects the bones. Much more bacteriological research is needed, especially among game animals, to solve all aspects of the diseases. An affected bone may be eaten away or enlarged and made spongy. If the bone is necrosed near the teeth, which is frequently the case, the latter may drop out or grow at abnormal angles. When a tooth drops out, the one that opposed it usually becomes elongated and sometimes spikelike because it meets no surface or strikes only the side of another tooth. Mandibles affected are sometimes bent or so necrosed that they are almost severed. The mechanical difficulties in the teeth resulting from the disease are frequently considerable, and since mastication is so important to an ungulate, any injury to the teeth becomes a serious handicap.

In Jackson Hole, Wyo., where much work has been done on the diseases of elk, a large percentage of the elk deaths during the winter and spring months are due to necrotic stomatitis (O. J. Murie, 1930). The calves in their first winter are the principal victims, but older animals also succumb. In the acute form of the disease the animals do not last long and die while still in good flesh. In older animals the disease at times seems to take on a chronic form, causing severe necrosis of the bone. Whether more than one disease is present has not been completely worked out, but further bacteriological studies are in progress.

Evidence of what appears to be necrotic stomatitis has been found among wildlife in many localities, and in deer, antelope, moose, caribou, and Rocky Mountain bighorn, as well as in elk. W. Reid Blair (1907) reported finding actinomycosis in a severe stage in three of six skulls of Ovis stonei from the Stikine River country in northern British Columbia.

Figure 42: The upper specimen shows one side of the palate of an 8-year-old ewe partially destroyed by necrosis. The lower specimen shows necrosis and exostosis near the angle of the mandible of an old ewe.

In the Jackson Hole elk herds the disease seems to take a regular toll. Some years it is a little more serious than others and it especially affects the calves. The young are especially susceptible, possibly because their mouths are tender and easily pierced by vegetation and the wounds in the mouth attending changes in the dentition afford a portal of entry. Although usually the incidence of the disease is not high and (according to paleontological evidence) has afflicted the animals for thousands of years, still it is important in the consideration of game losses and should be recognized in order that losses can be better analyzed.

The extensive necrosis and exostosis found in the skulls indicates that many sheep in Mount McKinley National Park are affected by either necrotic stomatitis or actinomycosis. Because of the bone conditions the disease has been referred to as actinomycosis although bacteriological studies would be necessary to determine the organism. Evidence of the disease was found in animals of all ages. Many had undoubtedly recovered but had been left handicapped with misshapen jaws and loose and missing teeth. Some yearling skulls were found which showed strong evidence of the disease. But since it appears that yearlings often die before the bone is affected, especially from necrotic stomatitis, many yearlings afflicted with it would not show evidence of its presence.

Figure 43: Mandibles of a 9-year-old ram. The one on the left, greatly enlarged and porous, shows an extreme diseased condition.

In the present study, animals whose skulls showed severe conditions of necrosis or exostosis were considered weakened by disease. Those with but slight necrosis were considered healthy. Figures 42 to 45 show malformations resulting from necrosis and exostosis.

Some animals affected by these bone changes would be unable to masticate the food well enough or fast enough. Others would have chronic infections in the mouth when the food kept the sores open and irritated. Still others would be ailing with the disease itself. All but six of the diseased sheep listed in Tables 4 to 7 inclusive (pp. 122 and 123) are those showing severe cases of what has been referred to as actinomycosis.

Some notes on domestic sheep are suggestive of the effect of old age and missing teeth, or what is termed a "broken mouth" by sheepmen. O. J. Murie recently discussed the matter with some experienced, successful sheepmen. One of them said that after ewes are 6 years old they do not survive well on the desert in winter even though they have all their teeth. Teeth generally "spread" and fall out after the sixth year and the losses then become heavier. After the ewes are 6 years old they are disposed of or put on a ranch. It was also stated that the sheep probably never exceed 14 years of age. Another sheepman thought that ewes wintered fairly well at 7 years and stated that sometimes he had wintered them on the desert even as old as 9 years. But all of the sheepmen agreed that as soon as a ewe had missing teeth her chances for wintering successfully on the desert dropped considerably.

Figure 44: Lower jaw bones of Dall sheep. The mandible at top shows necrosis on the side of the molars. The mandible below is bent due to necrosis, and a tooth is missing. The side of the molar behind the cavity has become worn, thus making a sharp-pointed tooth.

C. L. Forsling (1924, pp. 20—21), in his studies, has come to similar conclusions concerning the effect of age and tooth deterioration on domestic sheep. He writes: "Ewes should not be kept in the range bands after they are 5 or 6 years of age. Animals that have reached that age are less able to rustle forage than are younger sheep. Furthermore, if such animals are held over, they may not be in condition to mother lambs or survive a critical period. The condition of the teeth usually determines whether or not ewes will do well on range feed. Gummers, or animals with few or irregular teeth, are of doubtful value on the range and should be disposed of." These conclusions correlate tooth deterioration with loss in stamina, suggesting that a similar correlation may exist in mountain sheep.

Severe necrosis may only be present in one mandible or along one tooth row, and the remainder of the tooth rows may appear quite normal. On a few occasions two or three of the tooth rows were picked up, none showing at the most more than slight necrosis, and the presence of severe necrosis was not known until the last tooth row was found. Since a number of skulls are represented by only one or two tooth rows, some cases of severe necrosis were no doubt missed so that the incidence of the disease was probably higher than recorded in Tables 4 to 7 (pp. 122 and 123). Diseases other than actinomycosis cannot be determined from the remains, so that some sheep designated as non-diseased might possibly have been diseased.

Figure 45: Teeth of Dall sheep. The specimen at lower left shows crooked teeth resulting from necrosis and missing molars. The other three tooth rows show an abnormal condition in a 6-year-old ewe. The upper and lower molars should have met on the flat surface. Instead they slid past each other, wearing the sides so that sharp chisel-edges resulted. The white areas on the sides of the teeth indicate wear.

Malformed Teeth: Four animals were found in which there was severe malocclusion of the molariform teeth which seemed not to have been caused by disease. The teeth did not occlude normally, on a flat surface, but slid past each other wearing off the sides of the teeth and forming sharp-edged chisel-shaped tooth rows. As an example, in one case the lower molars were worn on the outer side, and the upper molars on the inner side, resulting in beveled teeth. On one specimen the lower molars slid past the uppers and gouged into the upper gum. Since these animals were undoubtedly greatly handicapped they are listed as diseased.

Lungworm: The lungs of one yearling were examined and were found to contain many adults, larvae, and eggs of a nematode which probably belonged to Protostrongylus or a related form (Goble and Murie, 1942). Further information is not available as to the incidence of lungworm or its effect on the sheep. No animals are listed as being weakened by heavy lungworm infestation.

Injuries: A yearling was found weakened by internal infections apparently resulting from a fall. A 2-year-old died as the result of a snag puncturing the abdominal cavity. These are listed as diseased.

DISCUSSION OF SKULL DATA

The wolf preys mainly on the weak sheep.—For an analysis of the data, the skulls have been placed in two groups according to the years in which the sheep died. One group is composed of skulls from animals which died between 1937 and 1941. The skulls in this group, at the time I found them, showed that the sheep had died in the last year or two. They had died during winters which were not severe, perhaps a little more favorable to them than the average winter. A crust on the snow in 1940 compensated somewhat for the otherwise favorable nature of these winters. This recent material portrays the current picture and is thought to present more accurate information of sheep mortality in the presence of wolves than does the old material.

The second group of skulls includes those from sheep which died previous to about 1937. These skulls are segregated because the winter conditions when the animals died are not so well known. A few of these skulls may date back to the days before the wolf became abundant, and an unknown number of them, no doubt, date back to the severe winters of 1929 and 1932 when apparently in many cases the strong and the weak alike succumbed in the deep snows. Skulls of sheep which died in 1929 and 1932 would probably show a lower proportion of old age and diseased animals and a greater number of healthy animals in their prime years. More of these old skulls are incomplete, several being represented by single mandibles. The incomplete skulls would be less likely to show presence of disease.

The data for each group are presented in Tables 4 to 7 (pp. 122 and 123). In Tables 4 and 6 the number of sheep in yearly age classes is given, and in Tables 5 and 7 the yearly age classes are combined into the 2- to 8-year and old-age classes. The lamb and yearling data are the same in both sets of tables. Besides being segregated according to ages. the skulls are segregated according to sex and the presence or absence of severely diseased skull bones.

TABLE 4—Skulls of 221 sheep which died between about 1937 and 1941, showing number of diseased and nondiseased animals in annual age classes. Sexes of lambs and yearlings are combined.

¹Exact age not known because of absence of horns, but teeth showed that the animals were 9 years of age or older.
²Lambs and yearlings.
³Adults.

TABLE 5.—Skulls of 221 sheep which died between about 1937 and 1941, segregated into four age groups. Skulls showing serious diseased conditions are segregated from the normal ones.

¹Lambs and yearlings.
²Adults.

TABLE 6.—Skulls of 608 sheep which died before about 1937, showing number of diseased and nondiseased animals in annual age classes. Sexes of lambs and yearlings are combined since usually they are not known.

¹Exact age not known because of absence of horns, but teeth showed that the animals were 9 years of age or older.
²Lambs and yearlings.
³Adults.

TABLE 7.—The skulls of 608 sheep which died previous to about 1937 are here segregated into four age groups. Skulls showing serious diseased conditions are segregated from the normal ones.

¹Lambs and yearlings.
²Adults.

In the recent material there are 221 skulls. Of these, 153, or 69 percent, are in the old-age class; 21, or 9 percent, are in the 2- to 8-year class but are severely diseased ; 10, or 4 percent, are in the 2-to 8-year class and show no sign of disease; 7, or 3 percent, are from yearlings showing signs of disease; 22, or 10 percent, are yearlings with no evidence of disease; and 8, or 3 percent, are lambs.

In the old material there are 608 skulls. Of these, 348, or 57 percent, are in the old-age group; 64, or 10 percent, are in the 2- to 8-year class but are severely diseased; 75, or 12 percent, are in the 2- to 8-year class, without evidence of disease; 3, or .5 percent, are from yearlings and show signs of disease; 85, or 14 percent, are from yearlings with no indication of disease; 33, or 5 percent, are lambs.

These figures are remarkable because they show that most of the sheep dying belonged to the weak classes. In the absence of predation we would expect the mortality to be distributed among the weak, namely the old, diseased, and the young, but in the presence of a strong predator like the wolf, known to be preying extensively on the sheep, it is interesting that so few animals in their prime are represented. In the recent material 211 skulls, or 95 percent, were from the weak classes in the population and only 10 skulls, or 5 percent, were from sheep in their prime which were healthy. In the old material 533, or 88 percent, of the skulls were from weak animals and only 12 percent from sheep in their prime which were healthy so far as known. The figures for both groups are roughly similar, considering the fact that in many cases disease or weakness would not be shown in bony remains.

It is significant that the incidence of disease is much higher in the skulls of the 2- to 8-year class than in the old-age class. In the recent material, 68 percent of the skulls in the 2- to 8-year class show evidence of serious disease, as compared with only 20 percent in the old-age class. In the old material 46 percent of the skulls in the 2- to 8-year class show a serious diseased condition, while diseased animals comprised only 25 percent in the old-age group. From these figures it becomes apparent that disease is an important factor in the predation among sheep in the 2- to 8-year group and not so important in the old-age group. In the old-age group the effects of age in weakening the animal are, as we would expect, more important than disease.

It is of interest to note that in the old-age group the youngest animals (those 9 years old) are the part of the group showing the highest incidence of disease. This is logical, for the older the animals become the more their weakness would be due to old age rather than disease. In the old material the incidence of disease is significantly higher among the ewes than among the rams. Possibly more of the ram skulls date back to the hard winters of 1929 and 1932 when conditions were so severe that disease was not an important factor. This is suggested as an explanation, since ram skulls would be expected to last longer than ewe skulls. In the recent material there is no appreciable sexual difference in disease incidence.

It is not known how many of the dead sheep the wolves killed but it is certain that they killed many of them. The figures on mortality are so uniform in showing that the weak are the ones dying that for our purpose we do not need to know how many of the sheep the wolf killed. Because of the high percentage of vulnerable sheep (95 percent in recent material) among the dead, whatever predation was done by the wolf would necessarily mainly affect the weak sheep. Perhaps we should expect the wolf predation to operate in this way, for prey-predator relationships between large animals, where numbers are limited, must be rather finely adjusted from the standpoint of species preservation. The wolf and the mountain sheep have existed together under conditions largely as at present for a long time, so that an adjustment between them, whereby both can survive, should be expected. If the wolf were powerful enough to capture the sheep indiscriminately it would long ago have exterminated them, for the law of diminishing returns in the case of mountain sheep hunting would not begin to function very early in the process of decrease. The adjustment seems to be such that, in mountain habitat having a considerable degree of ruggedness and extent, the wolf catches a few of the strong animals but preys mainly on the weak members of the population. This conclusion is supported by the skull data, by field observations of animals killed, and by the hunting methods whereby successive bands of sheep are chased by the wolf as he courses over the hills.

Some related big game-predator findings.—Many persons have taken for granted that the large predators prey upon the weak. It seemed to them that such a relationship was necessary and logical. Others have been skeptical of any such relationship, feeling that the predators preyed indiscriminately on all classes. Actual data have been scarce because quantitative information bearing on this type of problem is difficult to accumulate. I have not reviewed the literature thoroughly, but two reports of studies bearing on the subject have come to my attention. Some observations which I made on the coyote and mule deer in Yellowstone National Park are pertinent. Because these three studies are closely related to the wolf-sheep problem, the conclusions will be given briefly.

Some such relationship as exists among the wolves and Dall sheep also appeared to exist between the mule deer and the coyote in Yellowstone National Park (Murie, 1941). On the better ranges, where the fawns were strong, there was little predation, but on a poor range, where the fawns weakened as the winter progressed, they were preyed upon by coyotes. The prey-predator adjustment seemed to be one in which, as a rule, the prey was not taken until somewhat weakened. Here the predation affected mainly the weak and tended to quicken the adjustment of the deer population to the condition of the range.

Sigurd F. Olson (1938, p. 335) in a paper on the wolves of northern Minnesota, stated: "Long investigation indicates that the great majority of the killings are of old, diseased, or crippled animals. Such purely salvage killings are assuredly not detrimental to either deer or moose, for without the constant elimination of the unfit the breeding stock would suffer."

In an interesting study of the cougar, Hibben (1937), reported on 11 deer which had been killed by cougars. He wrote as follows concerning these deer: "The fact that, of the 11 deer, all showed abnormal or subnormal characteristics is almost too complete to be mere coincidence, and yet coincidence it may be. It is certainly logical that not every deer which a lion kills is subnormal. It is claimed by many hunters that the lion kills at will and takes the best as he chooses. The evidence does not seem to support this theory. Often in the course of this survey lions have been followed which were hunting. These were obviously trying to catch any deer they could. One lion made at least three attempts before he secured a meal. When he did get one, it was * * * the one with the abscessed neck. * * * All the evidence seems to show that the lion catches what he can and that may very well be the less ablebodied prey." Here again is a large predator whose ability to catch its prey seems to be closely adjusted to the ability of the prey to escape.

EFFECT OF WOLF PREDATION ON THE SHEEP AS A SPECIES

It is with a definite hesitancy that I venture to discuss the effects of the wolf on the sheep as a species because of the limitation of knowledge in this general field. The ways in which Nature operates to bring about or retain transformations are still rather mysterious. There is much that we do not know about fundamental premises, such as the causes of variations, which may alter our evaluation of other phases of the problem such as the elimination of the weak. The mathematics involve figures which we do not have, and a time element which we cannot evaluate. A slight trend which may to us seem insignificant could perhaps be of major importance if given a long enough period in which to operate. But because the elimination of the weak at once suggests a racial benefit, it seems apropos to consider briefly predation on the different categories of weak sheep as it may have a bearing on natural selection.

The predation on the old-age group would seem to have little bearing on evolution since the old animals are the successful ones, and their characteristics have already been lodged in many offspring. But some sheep may become infirm at an earlier age than others and be the first of their age group to be eliminated by the wolf. The stronger would continue breeding another year or two and impart their characteristics to a few additional offspring. So even in this group predation may result in a greater number of offspring from the stronger sheep. Characters such as tooth durability related to longevity would be favorably affected. The influence of the wolf might be less than appears, however, for some of the early infirm sheep taken by the wolves probably would have been ineffective breeders. But at least a tendency for selection of favorable characters may be conceded.

Predation on sheep weak from disease may operate in more than one way. Disease itself may be a selective agency, eliminating the weak; or it may affect the strong and weak alike. Actinomycosis is perhaps generally acquired by both the weak and the strong, depending somewhat upon the character of the food and accidental presence of abrasions in the mouth. If the wolf should prey on a sheep which has recovered from this disease but has become weakened because of resulting malformed dentition, the wolf may be eliminating an animal with inherent resistance and thus neutralize the selective influence of the disease. If a sheep acquired the disease because of faulty tooth succession and recovered, then the wolf, in destroying it, would be a selective agent. Thus the wolf's predation on the diseased sheep might operate to neutralize selection due to disease or at times might be eradicating some unfit animals. On the whole, it seems that predation on the diseased animals is not an important factor in natural selection or survival.

Because losses are so heavy among sheep in their first year, it is among these animals that selection would have to operate to be most effective. But unfortunately it is in this group that the information regarding natural selection is least complete, because the condition of the yearlings at the time of death is usually not known. The most important information not obtainable from an examination of the skulls is the general physical vigor and strength of the yearlings that are taken by wolves. No doubt the crippled yearlings and those diseased are eliminated early. In the fall most of the lambs are probably strong enough to avoid the wolf under favorable circumstances.

As winter progresses all the sheep become thinner and some of the growing yearlings succumb, regardless of predation. Other yearlings weaken but would probably recover in the absence of the wolf. It is in the predation on these yearlings that the wolf would have an important bearing on selection.

The percentages involved in these theoretical calculations are not known. Some day we may find a situation which will give us this information on yearlings. The escape ability of the yearlings is so high that we would expect eventually to find the wolves largely eliminating the less able individuals in this group.

It is difficult to arrive at a definite conclusion. Many of the deaths among the sheep have no bearing on selection so far as wolf activities are concerned; others seem to be significant in this respect. What effect predation on the weak has on a species is hard for any of us to say in the present state of our knowledge. It may be much greater than many believe; it may be important along with other forces in Nature. If the sheep has reached a point, at least for the time being, in equilibrium with its environment, then perhaps the wolf is important in maintaining the type.

We have been discussing the elimination of the weak and the part this activity may play in the maintenance or improvement of the species. There is another angle that could be considered. As an evolutionary force the wolf may function most effectively by causing the sheep to dwell in a rocky habitat.

I have tried in this brief review to point out that much of the wolf predation, although affecting the weak sheep, may have a limited selective value, but that there is a tendency toward some selection which may be of great importance to the sheep as a species.

Classified Counts of the Mountain Sheep

Basic for our understanding of mountain sheep ecology is a knowledge of lamb and yearling numbers. Knowing the size of the lamb crop, we can then determine by later counts what time of the year losses among them occurred. The counts in the spring give the size of the new lamb crop, and also the number of yearlings which have survived the winter, a critical period in the sheep's life. The counts in late summer and fall show the lamb survival during the summer.

The lamb and yearling ratios are obtained by comparing the lamb and yearling counts with the ewe counts, rather than with the combined counts of ewes and rams. The basic relationship desired is the ratio of young to the ewes. Knowing the number of rams, the ratio can then be given in terms of the entire population if desired. The figures would be more precise if the 2-year-old ewes could be consistently segregated. Small errors in yearling ratios in the fall counts may be present because at this time some yearlings are with the rams and the proportion of rams counted may vary. During the winter a certain number of ewes would die which brings in a variable that tends to raise the yearling-ewe ratio. But the errors introduced by the roughness of the data in these particulars are apparently slight and unimportant. In comparing spring and fall counts, specific localities are not compared, for unless total counts are available for the locality one would expect enough variation in the sampling to make comparisons unprofitable.

Figure 46: A ewe with two lambs which appeared to be her twins. No other twins were identified as such. [Polychrome Pass, September 25, 1939.]

In counting the mountain sheep, they were generally classified as ewe, lamb, yearling, and ram. When the 2-year-old animals were tabulated the sexes were lumped. Young rams—those up to 3 or 4 years old—were tabulated separately on a few occasions. These partial tabulations of 2-year-olds and young rams are retained in Table 8 and the 2-year-olds also in Tables 9 and 10 (pp. 130 and 131), but it should be understood that this differentiation is far from complete.

When only a part of a band was classified the record was not placed in the tabulations. Since ewes with lambs tend to segregate when the lambs are young, special effort was frequently made to classify the whole ewe population in an area in order to include the bands with many lambs and those with few lambs and thus attain a representative figure. Classifications were begun in early spring before the lambs were born and were continued through the summer and early fall, except in 1941 when the work was terminated in early August.

Sheep from practically all parts of the range were classified during each of the three summers that counts were made. The opportune time to obtain classifications by locality is early June before the sheep have begun their summer migrations and late enough to include most of the lambs. In the fall the sheep return to the spring haunts, so the populations can again be classified by locality and the summer losses can be determined.

No dependence should be placed on the sex ratios shown in the classified counts, because rams and ewes segregate for most of the year, and unless one is certain that a complete count in an area has been made one does not know what proportion of the ewes or the rams in the area has been included.

The greater number of ewes in most of the tables is largely due to the special efforts made to classify the ewe bands. The data from the skull collections indicate that the sex ratio among the sheep is about 50—50. In the recent skull material there were 91 females and 93 males above the yearling age. In the old material there were 213 females and 274 males, but here the number of male skulls may be greater only because they are found more readily. Ewe skulls are more easily covered by gravel and hidden by vegetation. In the recent material, however, the skulls of ewes are found almost as easily as those of rams because hair remains are often present to direct one to the skull, and few skulls have as yet been covered by debris.

The lambing period may vary considerably in different years. In 1908 Sheldon (1930, p. 366) saw the first lamb of the season on May 25. Dixon (1938, p. 216) states that in 1926 he encountered the first lamb on May 5, and in 1932 on May 31. In 1939 I saw two lambs on May 14, and one was reported on May 11. Soon after May 14 many lambs appeared, and by June 1 most of them had been born. In 1940 the lambing was late. None was seen until June 4. By the middle of June the lambing was largely finished. In 1941 two lambs were seen in Nenana Canyon on May 8. Most of the 1941 lambs were born about the middle of May.

The majority of the lambs seem to be born within the first 2 weeks of the lambing period. During the third and fourth weeks there are additional arrivals, and sometimes births occur quite late in the season. On July 19, 1939, a lamb was seen which was about half the size of the others. It appeared to be about 2 weeks old which would place its birth early in July. Twins are a rarity. Only one pair was noted. Often a ewe is followed by two lambs (or more), but if a person watches long enough he generally will see a second ewe appear and claim one of the lambs.

Sheep classified totaled 4,985 in 1939, 1,157 in 1940, and 2,732 in 1941. There were many duplications since the total number of sheep in the park is not more than 1,500.

TABLE 8—Classified Spring Counts—1939¹

1 The best counts are included, following as nearly as possible the close of the lambing period. There is very little duplication, and figures give lamb crop and yearling survival for majority of mountain sheep in Mount McKinley National Park.

2 Arranged from east to west.

TABLE 9 —Classified Fall Counts—1939¹

¹Includes the most complete fail counts available for different localities. Lamb survival through the summer and proportion of yearlings are shown.
²Arranged from east to west.

TABLE 10.—Classified Spring Counts—1940¹

¹The sample is small but representative. The lamb crop is extremely small and the yearling survival low.
²Arranged from east to west.

TABLE 11.—Classified Fall Counts—1940¹

¹The small lamb crop revealed in spring count is again evident here.
²Arranged from east to west.

TABLE 12.—Classified Spring Counts 1941¹

¹Shows 1941 lamb crop and survival of 1940 lambs to yearling age. There is a minimum of duplication.
²Arranged from east to west.

TABLE 13—Classified Summer Counts—1941¹

¹Field work was terminated in early August, so these counts were made earlier than the late-season counts for 1930 and 1940.
²Arranged from east to west.

TABLE 14.—Comparison of lamb-ewe and yearling-ewe ratios for the years 1939, 1940, 1941

¹The "spring count" and "fall count" consist of the most complete counts in each locality over the range rather than a total of all counts during the spring or fall season. Thus duplication was avoided.
²"Total count" consists of all the classified counts made during time year.

CLASSIFICATIONS FOR 1939

Total count.—Between the latter part of May, when practically all lambs had been born, and late October 4,985 sheep were classified. There were many duplications, but the count probably included 90 percent or more of the sheep. Some sheep were undoubtedly counted oftener than others, so that in the total count there is some danger of the averages being unduly influenced by some of the duplications. However, the results are about the same as the selected counts where there was very little, if any, duplication. The figures for the total classification are as follows: Ewes, 2,055; lambs, 1,064; yearlings, 549; 2-year-olds, 163; rams 1,154. The lamb-ewe ratio is 51 per cent, that is 51 percent of the ewes had lambs. The yearling-ewe ratio is 26 percent. If the lamb crop in 1938 was about the same as in 1939, then one-half of the lambs had succumbed during the first year. This I would presume to be a normal, good survival. The 2-year-olds were not counted consistently, so the count of them is smaller than it should be. In most other counts 2-year-olds were not segregated, so it may be best here to place half of them with the ewes and half with the rams so as to make all the figures comparable. This would reduce the figure for the lamb-ewe ratio from 51 percent to 50 percent and the yearling-ewe ratio from 26 percent to 25 percent.

Spring Count.—In the total count there were more duplications in some parts of the range than in others and if they happened to fall in localities where the lamb and yearling numbers were especially high or low, the final figures for the lamb and yearling ratios would be too high or too low. One representative count has therefore been selected from the field record in each locality. These were made as soon after lambing as a good count could be obtained, in order to have a classification of most of the population with a minimum of duplications. Although the counts were not made on the same date (there is more than a month separating some of them) I am confident there is little duplication. Where there could have been duplication, judging from the localities and dates appearing in Table 8, the movements of the bands were well enough known during the critical period to eliminate largely any probability of it. A range-wide count closely following the completion of the lambing gives a lamb-ewe ratio before potential summer losses have occurred.

The selected early 1939 counts, totaling 1,141 animals, are given in Table 8. The lamb-ewe ratio is 53 percent and the yearling-ewe ratio is 30 percent. If one-half of the 2-year-olds are placed with the ewes, the lamb-ewe ratio becomes 49 percent, and the yearling-ewe ratio 28 percent. These ratios do not differ greatly from those derived from the total count.

In this particular early count a fairly accurate count of 2-year-olds was obtained, although it is perhaps a little lower than it should be, since it was not always possible to count this age group. However, the figure gives some indication of the survival of sheep to that age. Sixty-seven 2-year-olds were counted, giving a ratio of 2-year-olds to ewes of 15 percent. The size of the lamb crop from which these 2-year-olds came and the number of these lambs surviving to the yearling stage is not known, but if similar to the 1939 lamb and yearling ratios, one might guess that a little more than one-half the yearlings had survived to the 2-year-old class. However, there is too much conjecture involved to accept this figure for the yearling loss with any assurance.

Fall count.—In order to check the survival of lambs during the summer, I have listed the most comprehensive fall counts for 1939 in such a way as to avoid duplication. Counts are available from fewer localities than in the spring, but the figures seem large enough to be significant. This late 1939 count is given in Table 9. One-half (5) of the 2-year-olds listed are added to the ewe figures. The lamb-ewe ratio is 49 percent, the same as in the spring, and the yearling-ewe ratio is 32 percent, four points higher than in the spring. The figures indicate that there were few losses of lambs and yearlings during the summer months. This was also the impression I gained from general field observations, for no indications of disease were noted, and wolves, after the appearance of the caribou, did not molest the sheep.

CLASSIFICATIONS FOR 1940

Total Count.—In 1940 the total count—1,157—was small, but the various counts were well distributed according to locality. The results of the total classification are as follows: Ewes, 479; lambs, 93; yearlings, 78; 2-year-olds, 20; rams, 487. The lamb-ewe ratio is 19 percent, and the yearling-ewe ratio is 16 percent.

Spring Count.—The best early classified counts are given in Table 10. They do not include two important localities—Savage River Canyon and Sanctuary and Teklanika Canyons—which were classified a little too early to get the lamb counts. The count is as follows: Total, 291; ewes, 97; lambs, 11; yearlings, 19; 2-year-olds, 6; rams, 158. The lamb-ewe ratio is 11 percent; the yearling-ewe ratio is 19 percent.

Fall Count.—The fall count given in Table 11 is a good sample. It is as follows: Total, 421; ewes, 208; lambs, 35; yearlings, 31; rams, 147. The lamb-ewe ratio is 16 percent, and the yearling-ewe ratio is 15 percent.

It will be noted that the lamb-ewe ratio in the spring was 11 percent, while in the fall, when it would be expected to be the same or smaller, it was 16 percent. If three bands, which were classified in the fall, had been classified in the spring, the lamb-ewe ratio for the spring would be raised to about 17 percent. It just happened that the lamb-ewe ratio was higher than average in these particular bands. There were apparently few lamb losses during the summer.

The losses in the 1939 lamb crop were too large to maintain the herd. The 49 percent lamb-ewe ratio of the fall of 1939 had been reduced by the following spring to a yearling-ewe ratio of 19 percent or less.

CLASSIFICATIONS FOR 1941

Total Count.—In 1941 the total count was as follows: Total, 2,732; ewes, 1,331; lambs, 722; yearlings, 38; rams, 641. The lamb-ewe ratio is 54 percent; the yearling-ewe ratio is about 3 percent.

Spring Count.—In the spring of 1941 large counts from many localities are available as shown in Table 12. Some of the counts were made as early as May 26, but by this date most lambs had been born. The count is as follows: Total, 844; ewes, 386; lambs, 209; yearlings, 16; rams, 233. The lamb-ewe ratio is 54 percent; the yearling-ewe ratio is 4 percent.

Summer Count.—In 1941 I left the park in early August so could not make counts in the fall. Most of the counts included in Table 13 were made in July and early August. Counts made about the middle of June at Savage and Sanctuary River Canyons are used in order that these localities can be included to make this count more comparable with the spring count. The count is as follows: Total, 452; ewes, 195; lambs, 111; yearlings, 1; rams, 145. The lamb-ewe ratio is about 57 percent, slightly higher than the 54 percent obtained In the "total" and "spring" tabulations (Table 14). Indications are that few lambs were lost during the summer.

The yearling-ewe ratio is about 0.5 percent, only one yearling being recorded in this late count. If the yearling-ewe ratio in the fall had been the same as in the spring there would have been eight yearlings in the fall count. The number of yearlings involved is so small that one hesitates to conclude that there was a yearling loss during the summer. A band or two harboring a few odd yearlings would bring the fall yearling ratio up to the spring ratio.

All the counts in 1941 show a high lamb crop (more than 50 percent) and an extremely low number of yearlings. Very few of the 1940 lambs survived the winter.

DISCUSSION

Variations in lamb crop. —The lamb crop in 1939 was about 49 percent; in 1940 about 16 percent; and in 1941 about 54 percent. The 1939 and 1941 lamb crops were probably slightly better than average, with relatively small losses occurring during lambing time. The analyses of wolf droppings indicated some predation on the 1939 lambs before the caribou moved into the region with their calves. Possibly this wolf predation was just sufficient in 1939 to lower the 1939 lamb crop below that of 1941. However, we are dealing with rather small differences, and the slight wolf predation on lambs probably had little effect on the ratio.

The 1940 lamb crop was abnormal in two respects; the season was very late and the number of lambs unusually small. In 1940 the first lamb was found on June 4. On this date the lambing had been practically completed in the 1939 and 1941 seasons. The lamb crop in 1940 was one-third or less than that of 1939 or 1941.

It is not easy to account for the small lamb crop. According to some of the trappers there is occasionally an exceptionally poor lambing year. It is possible that the small 1940 lamb crop is correlated with crusted snow conditions reported for January of that year. The ewes may have been affected in some manner, causing a lowering in the lamb numbers. Since the lambing was so late, possibly the early bred ewes were affected most by the snow conditions. It may be significant that in 1932, after the severe winter, the lambing was also late and the lamb crop very small. On the other hand, the winters preceding the high 1939 and 1941 lamb crops were favorable for the sheep. The crop, in part, may be dependent on winter conditions.

Ewe-lamb ratio normally about 50 percent.—A good average lamb-ewe ratio appears to be about 50 percent, although this is based on the classification of only three lamb crops, one of which was almost a complete failure. But a 50 percent lamb crop is about what one would expect in a good year judging from the reproduction rates of big game in general. The 2-year-old ewes are included in the ewe counts even though ewes probably do not produce lambs until they are 3 years old. However, their inclusion does not change the figures more than a few points, still leaving the ratio near 50 percent.

In 1941 L. J. Palmer (1941) classified a number of sheep in the Mount Hayes region of the Alaska Range east of Mount McKinley National Park. He did not segregate the yearlings so the figures are not strictly comparable. He counted 501 ewes and 214 lambs, and estimated that one-half of the 2-year-olds and yearlings survived. If one-half the yearlings survived, we would have about 100 to subtract from the ewe count, to make his ewe count comparable with mine. This would give us about 400 ewes. After making this adjustment, which still includes the 2-year-olds as in my figures, we get a lamb-ewe ratio of 53 percent. This percentage is surprisingly close to my "total" and "spring" counts the same year (Table 14). This close agreement between the figures for the Mount Hayes area, where wolf predation is probably less than at Mount McKinley National Park, and those in the park, strengthens the conclusion that a lamb-ewe ratio of about 50 percent is a good normal lamb crop.

Lamb and yearling ratios in different parts of range.—The lamb-ewe ratios tended to be uniform over the range during the 3 years that these ratios were determined.

In considering yearling survival, however, there does seem to be some real difference between localities. The yearling survival in the general East Fork area was below average, while at Savage and Sanctuary River Canyons and at Toklat River the yearling survival was consistently above average. In the areas where there is a slightly higher than average yearling ratio it appears that the country is rougher and of such a nature as to give greater protection to the yearlings from wolves. Sanctuary and Teklanika River Canyons with a mountain between them seem to be especially favorable localities for yearling survival. Although there apparently are local differences, the general average in different localities each year does not vary greatly.

Summer losses.—In comparing the spring and fall counts in each of the 3 years it is obvious that the losses of both lambs and yearlings during the summer were slight. There was relatively little wolf predation on lambs in 1939, and practically none in 1940 and 1941 so far as I could ascertain. The only sick animal noted was a young lamb observed in June by rangers.

Lamb survival first winter.—We do not know what the lamb-ewe ratio was in 1938, but judging from the number of yearlings present in 1939 it apparently was high. If it were somewhere near 50 percent, which seems likely, then more than half of the lambs survived the winter. Such a survival is probably sufficient to cause a definite upward trend in the population. The number of 2-year-olds in 1939 was also quite high, so that the sheep seemed to have done well for two successive years.

The excellent 1939 lamb crop suffered heavy casualties during the winter and a number of these casualties may have resulted from the crusted snow conditions prevailing for part of the winter or heavy wolf pressure due to scarcity of caribou in the sheep hills. The yearling-ewe ratio in 1940 varies from 15 to 19 percent (the lamb crop in 1939 was about 50 percent). About two-thirds of the lambs failed to come through the winter. Such a heavy loss no doubt means no increase, or a possible decrease in the population, for the yearling survivals would hardly compensate for the winter losses among the older animals.

As has been pointed out, the lamb crop in 1940 was only about a third the size of the crop in 1939 and 1941. Of this small crop, 20 percent or less survived the winter as yearlings, so that there were scarcely any yearlings in the population in the summer of 1941.

Thus for two successive years the survival of yearlings was so small that it is probable that the sheep population suffered a decline in numbers each year. So far as known, there was no disease among the yearlings except actinomycosis which generally affects only a nominal number of the animals. The cause of heavy losses appeared to be due to wolf predation.

Some Wolf-Sheep Relationships

EFFECT OF ARTIFICIAL INTRUSIONS ON WOLF PREDATION

In a national park the objective is to preserve a piece of primitive nature where natural interrelation ships may prevail. The more complete the biotic unit and the larger the area the greater is the opportunity to achieve that ideal. But unfortunately few national parks are large enough to be uninfluenced by artificial activities taking place both within and outside their boundaries.

Figure 47: At Polychrome Pass the high way traverses the heart of excellent winter range for mountain sheep. This affords wolves ready access to the range, and an opportunity to surprise sheep at sharp curves on the road. [October 22, 1939.]

In Mount McKinley National Park the automobile highway passes along the sheep hills for some 70 miles. This road gives the wolves a special advantage in that they have an easy trail to different parts of the sheep range. In several places the road winds among the cliffs used by sheep in winter, saving wolves a laborious climb. Because of the many sharp turns in the road the wolves have an opportunity to surprise the sheep at close range. These advantages to the wolves are not especially great because the road affects vitally only a small part of the sheep range. This illustrates the need for carefully examining all proposals for park developments.

Another activity which intrudes on the natural conditions within the park is the trapping of wolves on lands outside the park. Trapping adjacent to the park boundaries, and a certain amount of control within the park, annually has eliminated some of the park wolves. The effect on the wolf-sheep relationships is not well known, but probably such trapping does not take the annual increase during most years or greatly alter the wolf population. The status of the wolf within the park probably is, at the present time, similar to the status of the wolf in Alaska as a whole.

Still another artificial intrusion is the shooting of caribou adjacent to the park. Although this hunting does not now appear to be important it apparently has been extensive in the past. The reduction or curbing of the caribou herds by shooting might increase the wolf predation on the sheep by leaving a smaller caribou population.

The effect of the artificial intrusions will have to be carefully evaluated in considering the conservation of the wildlife in the park.

DO CARIBOU DIVERT PREDATION FROM SHEEP?

Frequently there is serious direct competition for food between ungulates. Although the food habits of sheep and caribou overlap broadly, their ranges, especially in winter overlap very little. Generally the caribou feed on the lower hills and rolling terrain where lichens are more abundant, and the sheep are up on the wind-swept ridges.

The caribou is an exceptionally good example of a buffer species. This is especially noticeable in the spring. When the caribou calves make their appearance they furnish to the wolves a large and easily obtainable food supply. At this time wolf predation on sheep practically ceases.

The caribou also serve as a buffer in winter. There are some suggestive correlations between wolf predation on sheep and the presence or absence of caribou in the vicinity of the sheep hills in winter. In a year that caribou wintered among the sheep hills the predation on sheep appeared to be less than when caribou were absent. In the winter of 1938—39 between 1,500 and 2,000 caribou wintered between Teklanika and Savage Rivers and others wintered along the north border of the park. That year there was a high survival of yearlings among the sheep. In the winter of 1939—40 there were practically no caribou in the sheep areas. The survival of sheep yearlings that winter was about half of what it had been the preceding winter (assuming that lamb crops were about equal). In the absence of the caribou the wolves turned to the sheep. However, crusted snow conditions also favored the wolf in his sheep hunting in the winter of 1939—40, so the two winters are not strictly comparable. The 1940 lamb crop was so small that one can not draw many conclusions from the figures. There was a heavy winter mortality among the few lambs present even though caribou were present to act as a buffer. Wolf pressure on individual lambs may have been higher than usual because there were so few lambs.

Although the caribou are very definitely a buffer species at times, there is still a question whether or not they increase or decrease the wolf predation on the sheep over a period of years. It is likely that because of the presence of the caribou, the sheep range supports a larger wolf population. True, when caribou are present they probably bear the brunt of the predation. But when caribou winter elsewhere, the wolf population, whose size is probably adjusted to their presence, is, so to speak, left stranded, and dependent on sheep for sustenance. The majority of the wolves in the park do not move off with the caribou as some persons have assumed. Wolves are accustomed to obtaining their food in the one territory. Thus it appears that the caribou herds, because of their periodic absences from the sheep ranges, may influence predation in such a way as, in the long run, to lower the size of the sheep population. But because of the small amount of information on this point our conclusions must be tentative. There is no doubt, however, that the caribou at times is a highly important buffer species for the sheep.

HISTORICAL EVIDENCE THAT WOLVES ARE CURBING SHEEP NUMBERS

During the period of wolf scarcity beginning sometime between 1908 and 1916 the sheep prospered, in fact built up to what apparently was an overpopulation about 1928. Unusually severe snow conditions in 1932 reduced the population which already had been depleted by the hard winter of 1929. According to one observer, not more than 1,500 sheep remained. This population was not far from that present on the range in 1941, 9 years later. We do not know definitely the number of sheep in 1932 after the die-off but it probably was not less than at present. During this 9-year period, so far as we know, a rather uniform wolf population existed and it is known that wolves preyed extensively on sheep. The fact that sheep increased rapidly in the absence of wolves and have not increased during their presence strongly indicates that the wolves have been the factor preventing the sheep from increasing. I am fully aware how frequently "obvious conclusions" are wrong, especially in prey-predator relationships. However, I found no other factor which seemed sufficiently operative to hold the sheep numbers in check. It seems therefore, that the wolves are the controlling influence. Granting that the wolves have held the sheep in check for several years, that does not necessarily mean that the sheep will not increase or decrease in the future in the presence of the same wolf population. The presence and numbers of the caribou is an uncertain factor which influences wolf predation on sheep so the yearly movements of the caribou and their status must be considered. Furthermore, the abundance or scarcity of snowshoe hares may have an important bearing on relationships. Other factors, of course, may change to alter the relationship, so one cannot predict with certainty the future relationships and population trends.

The historical data corroborate the evidence obtained in the field study that wolves are the chief factor limiting the sheep population at the present time.

EFFECT OF THE WOLF ON SHEEP NUMBERS

At the time that the sheep population was at its maximum many of the sheep lived away from the traditional rocky habitat and occupied low hills and even high banks along the rivers. The use of this gentle terrain was probably necessary because of food shortages in the rugged hills. In the absence of the wolf they were able to survive in the areas lacking protective cliffs. But when the wolf appeared and became plentiful the sheep in these places became so vulnerable that no doubt the strong as well as the weak could be captured. In the presence of the wolf, therefore, the range of the sheep was gradually constricted, and they were confined to more rugged habitats where they could cope with the wolf on better terms. in confining the sheep to the cliffs, the wolf is an important factor in molding their habits, and through the past ages probably has done much to develop and preserve the sheep as we know them today.

The relationship between wolves and sheep probably differs according to the types of terrain. Where the cliffs are exceptionally rugged and extensive a sheep population might be largely immune from wolf attack until it becomes so large that it over flows into less rugged surrounding country. Where cliffs are only moderately rugged or not very extensive, the sheep might remain rather low in numbers, perhaps at a point where the law of diminishing returns makes it unprofitable for the wolf to hunt them; in some unfavorable cliffs perhaps all the sheep would be destroyed. Therefore we would expect to find considerable variation in the abundance of sheep, in the presence of wolves, in different sheep areas, due primarily to the character of the country.

Since the wolves appear to hold the sheep in check it is of interest to discuss how predation, which apparently is mainly upon the weak, operates to control the population.

The predation on the old-age animals and those weakened by disease curbs the numbers to some extent. A few of the old ewes, although weaker than the average and doomed to die in a year or two from old age, might still bring forth a healthy lamb if not eliminated by the wolf. The same is probably true for some of the sheep suffering from malformed dentition as a result of disease. Predation on the healthy sheep, captured by chance or otherwise eliminates a few members of the population and their potential offspring. But this predation does not appear to be as important as that on the yearlings.

From the skull study it could not be determined how much the wolves increased the death rate among the yearlings. Comparative studies in periods of wolf scarcity would be enlightening in this regard. But it was evident that the losses among the yearlings in the winters of 1939-40 and 1940-41 were so great that, combined with all other losses, they resulted in a reduction in the total sheep population. It appears to me that the large losses were due mainly to heavy wolf predation. The lamb and yearling losses in mountain sheep herds in Wyoming, in the absence of any serious predator, are sometimes fully as great as in Mount McKinley National Park. But in Wyoming it was evident that the young were diseased for they coughed considerably in late summer, and many died during the summer and fall. This disease is probably correlated with poor range conditions. In Mount McKinley National Park the young sheep appeared healthy at the beginning of winter. No coughing or other evidence of disease was noted excepting actinomycosis which would probably affect only a nominal number of animals. On the other hand, the wolf was known to prey upon the sheep extensively at times. From the total observations made, the conclusion seemed evident that large losses among the yearlings are caused by wolf predation at times, and that these losses are probably most important in holding the sheep numbers in check.

THE SHEEP AND THE WOLVES MAY NOW BE IN EQUILIBRIUM

It is unfortunate that we do not have accurate information on sheep numbers since the die-off of 1932 so that we would know more about the trend of the population. One ranger thought that the low was reached about 1936, and that in 1939 there was a definite increase over the past few years. These thoughts were based on general observations. In 1940 and again in 1941 there were decreases. My impression is that the sheep numbers have not varied greatly since 1932. If this is actually the case then it is probable that the sheep and the wolves have reached an equilibrium, with, of course, some ups and downs. If we were assured that there would not now be any striking reduction in sheep numbers, the wolf-sheep relationship might be considered a satisfactory one, especially to anyone who has seen the ungulate over-populations and the over-grazed big game ranges in the States. Artificial intrusions may give the wolf an undue advantage over the sheep, hence it is desirable to continue observations in order that the status of the sheep may at all times be known and proper steps taken when needed to assure their perpetuation under as natural conditions as possible.