The only indication we have of the life expectancy of a Death Valley bighorn is the fact that the great majority of the skulls collected are from animals less than 1 year old or over 10, indicating the probability that if a sheep lives past its first year it can expect to reach its 10th. The difficulty of a lamb surviving its first year in Death Valley is indicated by the fact that we saw no yearlings at Nevares Spring in 1957, with no accurate lamb count for the year before. The 1957 lamb count of nine was reduced to eight by Fuzzy's death (see "Cast of Characters" and fig. 44), and we counted a survival of three yearlings for 1958. At Willow Creek in the summer of 1956, we had 12 ewes, 12 lambs, 4 rams, and 3 yearlings.
Of the 30 individuals observed in the Navel Spring area in 195657, only 3 were yearlings. The survival ratio of lambs is therefore indicated as being somewhere between 0 and 3 out of 10, with a suggestion that we may have something in the nature of alternating years in good and bad lamb crops and survival.
Herd composition, with the exception of the off-season segregation of the "bachelor" rams, appears to be more coincidence than anything else. This may be a local situation brought about by the low-density ground cover and population of the Death Valley region.
In the Desert Game Range in 1953, we saw several bands composed solely of ewes with lambs. On one occasion we counted nothing but yearling ewes in a band of seven.
The above is not true in Death Valley. The only age class and sex segregation that seems to take place regularly is that of the mature rams between mating seasons.
The 1960 Quartz Spring band was composed of 12 ewes, 12 lambs, and 2 yearlings, which might be called typical of what we observed at the Desert Game Range. But we were able to stay with them only 2 days, not long enough to determine the duration of specificity in this instance.
Two mature rams observed in the same canyon with them on the second day did not actually join the original band. If they had, it would have made a more typical Death Valley band.
Both of the Death Valley Buttes bands had two rams among them; there was a yearling ram at Badwater, a 3 year old at Furnace Creek, and a 4 year old at Jubilee Pass.
The family influence on herd composition is probably positive but temporary. The Badwater band retained its specificity for 4 weeks.
On February 12, 1955, four of the six returned with Droopy, and the yearling was missing, and finally by March 20 the band of six was reduced to one.
There is some indication that these splits are into immediate family groups, but how general this pattern may be is not known. Family resemblance is often noticeable in small bands, but by no means can it be considered the rule. Juvenile attachments probably carry over and to some extent influence herd composition. Some bands seem to be held together by a leader and break up with her loss, but apparently the strongest factor involved is the general gregariousness of the bighorn, which would probably result in much larger herds and bands if the food and water of the habitat would permit.
Seasonal influence on band composition is pronounced only in the sense that the word "seasonal" is defined as meaning wet or dry cycles in its application to the habitat involved in this study. The wet and dry cycles may be highly localized and variable within a small area. A wet season was created in Furnace Creek wash by high-country rainfall in 1955 and 1956 at the head of the Furnace Creek drainage. This wet season lasted from October 1955 through 1957, producing sufficient forage to permit the bighorn to exercise their natural tendency toward gregariousness.
This had happened, with variations, on the alluvial fans at Badwater in 195455 and set the observational pattern for this study.
The Furnace Creek runoff produced an oasis of Bebbia, Stephanomeria, and Eriogonum, which in turn produced another unique opportunity to observe two bands of relatively sustained specificity and family resemblance that formed the basis for much of the life history of this report.
Finally, the early concentration of flowers on Death Valley Buttes in February 1958 gave us our only sustained observation of a band of bighorn in the Grapevines.
Good ranges may be produced by wet seasons at any time of the calendar year; and when food supplies will allow it, bighorn tend to form larger bands and stay together longer. (See "Food" and "Water.")
Poor ranges result in breaking down of numbers to smaller groups of one to three animals, thus reducing the travel distance necessary to gather a sustaining diet.
Prolonged drought tends to reduce the bands to smaller numbers by decimation as well as by fragmentation. (See "Food" and "Water.")
We saw no specific bands at Nevares during the entire series of sustained observations from 1956 through 1960, but we recorded many instances of bands forming as they neared the watering area and breaking up again as they left. Such temporary bands, observed but once and then for but a few minutes or even hours, has led to many misconceptions regarding the average size of bands to be expected in the field.
The longest record we have of band specificity is that of the seven Badwater sheep who remained under constant observation for 5 weeks from December 18, 1954, to January 24, 1955.
Some concept can be gained of the transitory nature of band composition within the entire herd by data gathered on Paleomesa between November 4 and 25, 1956:
On November 4, Old Eighty, Scarface, Bad Boy, and Whitey came down from the mesa into Big Wash for the first time since Old Mama led the band of eight up Pyramid Peak the previous January. By November 15 this band had increased to nine. On November 18 we were still observing this band of nine in Big Wash when Old Mama with two strangers came down from the mesa on the south side of the wash, and Kinky and five others came down from the north. By 2 p.m. we were photographing our first band of 18 bighorn in Death Valley. They bedded together that night and were still resting at 6 the next morning. But by 7:20 Kinky and eight others, led by a ewe unknown to us, headed north and disappeared toward the Hole-in-the-Wall; we now had two bands again of nine each.
We followed Old Eighty and her band to water on the 24th of November. On the way in they were joined by four strangers, including a mature ram who left at dusk. At dawn of the next day the old gray ewe, Pearl, her lamb, and a yearling preceded the others in to Navel Spring, drank and left by way of nearby Hanging Canyon. At 2 p.m., Scarface, Bad Boy, and a new 2-year-old ewe went into Navel Spring wash; Whitehorns and Old Eighty bedded down with two others in the box canyon; Little Whitey, torn by uncertainty as to which group to follow, finally made an anxious exit alone toward Deadman's Curve. So in just 1 week our 2-day band of 18 had disintegrated into small groups of 1 to 4.
This same pattern was repeated in January 1961, on Death Valley Buttes: On January 1 there were 9; on the 3d, 11; on the 5th, 13 began the day with 2 mature rams heading back to Corkscrew in the afternoon. Nineteen had gathered among the flowers by the 10th, but the next morning eight of these went over the crest and back into the foothills of Corkscrew Peak as two others appeared briefly at the mouth of Bebbia Canyon. By dusk two small bands were left, one of five and one of six. Eight were counted the next day and on the 13th, none.
These two accounts of several small bands appearing briefly as one in watering or special feeding areas illustrate the rule and not the exception. It has been observed repeatedly at Willow Creek, Virgin Canyon, Indian Pass, and at Nevares Seeps. Its effect on sign reading is discussed under that heading. Unawareness of this pattern can be very misleading in determinations of numbers and distribution, and its discussion will continue there.
Numbers and distribution pose three common inquiries which no one can answer with certainty for the Death Valley area: How many are there? Are they increasing or decreasing in number? Where can they be seen? Answers to these questions can be no more than a recording of the development of our knowledge of the subject at the present time.
Beginning in October 1935, with a preliminary survey by Field Naturalist Joseph Dixon, efforts were made on a gradually expanding scale to count the bighorn of Death Valley in sample areas. From these samplings, estimates of the total population subsequently were derived. During Dixon's 1935 survey, reports by others of 30 and 42 animals were accepted for the Cottonwood Mountains at Quartz and Big Dodds Springs, respectively. From tracks and pellets Dixon estimated 21 additional bighorn at other springs in the Grapevine, Funeral, Black, and Panamint Mountains (Dixon, 1935). From this total of 93 animals seen, or inferred, from sign, covering less than 10 percent of the total bighorn range, Dixon did not attempt to derive a population estimate for the entire monument.
In September 1938, another incomplete survey by Dixon, Sumner, and members of the park staff, in cooperation with members of the California Department of Fish and Game, resulted in a total of 65 bighorn seen, or inferred from sign (Dixon and Sumner, 1939). Rains dispersed the animals before the survey could be completed, and no population estimate was made for the entire monument.
In July 1939, a second cooperative effort by the same group yielded 66 animals seen and 208 inferred from sign. Summer rains again dispersed the bighorn before the survey could be completed (Sumner, 1939). However, from sample areas covered, Sumner derived a total population estimate for the monument of 500.
In July 1955, a third major effort was made by Sumner, the park staff, and the present authors, in cooperation with the California Department of Fish and Game. This was the last and most comprehensive of the cooperative surveys. The number of sheep actually seen by the official census takers was 90. An additional 15 were seen by unofficial but reliable cooperators, making a total of 105 seen. The results of this 1955 census and of subsequent samplings by us are shown in table 12.
TABLE 12.Bighorn census surveys, Death
Valley National Monument, 195559
These surveys suggest that more intensive and continuous observations at high concentration localities would, with luck, yield even better results.
An estimated total population figure is always open to question, and no matter what the figure arrived at or the method by which it is obtained it is likely to be challenged. The 1955 estimate of 261 total for the population within the monument (table 12), the result of a concensus of field estimates made by the individual teams, was based on what they saw in their areas only. This was an ultraconservative figure, derived from the questionable assumption that the census takers had a reasonable expectation of finding nearly half of all the sheep in the areas they covered.
However, our subsequent controlled experiments at Badwater, with many competent observers and a known number of animals in a known area, would lead us to believe that if a quarter of the total population in a given area were sighted we were lucky indeed. There the band of six climbed the rugged western slope of the Black Mountains and browsed and bedded at about 1,000 feet elevation for 3 days. This was the peak period of their publicity, and hundreds of observers, including park rangers and naturalists from other areas, came to satisfy their curiosity. As far as we were able to determine, not once during this period was the band located without our assistance. We ourselves kept track of them only by constant surveillance from the earliest streaks of dawn when they left their bedding ground until sometime after dark when we could tell only by sound where they had bedded down again.
By this line of reasoning, it still seems conservative to estimate the total population at 4 times 105, the number seen, which would be 420.
There is nothing to be gained by being either conservative or liberal in estimating the total bighorn population, but both approaches were used for purposes of comparison. Using the "liberal" approach, a compilation of all the known sight records of bighorn was laid out on the map to give theoretical limits of both vertical and horizontal distribution. These limits were then modified by considerations of known forage and water supply, suitable terrain, seasonal migration (if any), etc. National Park Service Engineer Lin Spaulding then computed this sheep range as comprising 2,100 square miles (70 per cent of the total monument area). The areas marked on the topographic maps as covered by each field team were then computed to be 703 square miles, or 33 percent of the total possible sheep range of 2,100 square miles. By these figures, we arrive at a possible total of 3 x 420, or 1,260.
An apparently obvious fallacy could immediately be pointed out here; the 33 percent of sheep range that we covered included all the areas of high concentration.
It is quite likely that we did cover the majority of the high concentration areas, but we do not know. For example, Indian Pass in the Funeral Mountains, with the highest count of any area censused in 1955, had been left out of all consideration for 20 years preceding this last census. Borell in 1935 wrote: "There are very few, if any, sheep between Daylight Pass and Echo Canyon. This takes in most of the area now known as the Funeral Mountains." Don Curry, Acting Park Naturalist in 1939, wrote that he "had never seen sheep sign at Poison Spring in the central Funeral Mountains, although it is the only source of water for many miles in any direction."
On the other hand, we still do not know how many unmapped sources of free water may be known to the bighorn. These, if known, might add materially to the population figures. Several new springs were found during the 1955 census, notably Bighorn Spring, which was discovered and named by Lowell Sumner and Fred Jones. Our subsequent discovery of large populations watering at two uncensused major water sources, Blackwater and Twin Springs, provides another example that will be more fully described later.
Another difficulty in deciding whether an estimate is "conservative" or "liberal" arises from the fact that up to a certain point the poorer the conditions are for the bighorn, the higher their census reports are likely to be, because the animals that are surviving these adverse conditions of drought or malnutrition are more and more concentrated in permanent spring areas accessible to observation.
Rainfall appears to be the controlling factor in the normal low-density sheep population of the entire Death Valley region. During a year when general rains fall throughout the high-mountain country, vegetation is relatively abundant and nutritious; temporary water supplies in the back country such as tinajas, or "tanks," are plentiful; conditions are ripe for fertility, high lamb survival, and a low disease ratio.
For example, during the winter months from January 1950 until December 1954, we were able to find but one bighorn throughout the entire monument, and none was reported by visitors. High-country rainfall made it unnecessary for the sheep to use areas accessible to humans. In short, when everything points toward an optimum count, very few sheep can be found.
On the other hand, from December 1954 until January 1957, with high-country rainfall scarce and spotty, we were in almost constant touch with at least one band of sheep and occasionally two or three at a time. We saw more sheep and more were reported by visitors during this period than at any other time on record, and it was taken for granted that the population was increasing.
From January 1957, when the rains came, until July, the sheep vanished, but a dry summer brought good observations once more, at springs only, until a late rain again dispersed the sheep.
Thus, it has been generally accepted that the waterhole count is the most reliable method of census taking, but now we find that the most favorable conditions for a count are the most unfavorable for the sheep.
The waterhole count is still the best method we have, but it is likely to be more misleading than we had thought. The 1955 census, based on quick visits to springs, shows no sheep counted and no estimate of a population at Nevares Spring. Yet under comparable weather conditions between August 11 and September 10, 1957, when continuous vigils were kept, 11 ewes, 9 lambs, and 27 rams were counted there. It was assumed in 1955 that it was reasonable to expect to count no more than 25 percent of the total population in a 4-day vigil at a waterhole, but we now know that at the climax of a long drought, such as August and September 1957, the entire population may use the waterhole and be counted in a period of 4 days.
The majority of the 27 rams counted at Nevares Spring from August 11 to September 10 were travelers. They were counted only once or twice in 30 days, and could possibly have been counted several times at different springs in any 4 days of the 30-day period. How far these rams travel during the rut is not known, but 2 of the 27 we have seen repeatedly over 20 miles south of Nevares in the Big Wash area, and both of them went on north when they left Nevares after a stopover of but 2 or 3 hours.
We have made no attempt at a census since 1955. We counted 105 bighorn then, and we estimated four times that many. Subsequent observations make that figure seem ultraconservative to the point of inadequacy even when the variables described above have been considered.
For example, in 1955 we counted 4 sheep at Navel Spring and estimated a total population of 12. Between that date and September 1957, we made prolonged observations of over 30 individuals at Navel Spring, and 18 of these animals were in one band. It is true that several things had happened in the meantime: We had improved the water supply at Navel Spring from a bee-seep to 75 gallons. Contiguous areas had practically no rainfall for nearly 3 years, possibly forcing a temporary migration into the washes above Navel Spring where flash-flood runoff had produced a substantial growth of Bebbia and Stephanomeria. But probably most important of all, prolonged observations were carried out in the area for the first time.
Present data still are insufficient for a really good estimate of the number of sheep in the Death Valley region. No clue is offered as to whether the numbers are increasing or decreasing at the present time. However, increase and decrease have relatively little importance in the long range picture, since they undoubtedly vary a great deal with the wet and dry cycles normal to the region. Nevertheless, evidence generally points toward a relatively substantial and healthy herd in the area.
In review, the "liberal" population estimate of 1,260 bighorn computed from areas of known favorable range following the 1955 census seems more plausible now than it did then. A bighorn distribution map is shown on page 154, and results of census surveys are given in table 12. Since 1955 additional evidence on the side of more liberal estimates has accumulated: Sighting or positive signs of current use by bighorn have been recorded throughout the Panamint and Cottonwood ranges on both the eastern and western slopes from Wingate Pass in the south to Sand Springs in the north, indicating a much higher density population for the region censused than was estimated in 1955.
Of still greater significance is the discovery that two previously uncensused water sources in the north Panamints probably are supporting two of the largest bands of bighorn yet recorded for the region. These new sources are Twin Springs and Blackwater Spring. They add approximately 200 square miles of high (1,000 to 6,000 feet) and relatively dense burro-free forage to the known bighorn range, and include practically all of Tucki Mountain.
An estimated population of 50 sheep for the areas surrounding each of these springs seems plausible. The size of the areas served, the food supplies, and the cooler climate are all factors comparing favorably with the Nevares Spring area, where 47 sheep were counted in 1957.
In general, the same can be said of a third area so far uncensused: Corkscrew Peak, with Corkscrew Spring at 4,000 feet elevation as the hub, probably supports another 30 or 40 animals.
On September 1, 1956, we saw six apparently resident bighorn at Monarch Canyon, which is exactly twice the number estimated in 1955. In 1956 and 1957, we saw at least 30 individuals water at Navel Spring and 47 at Nevares, 2-1/2 and 16 times as many as were estimated in 1955.
The highly mineralized waters of the Keane Wonder Springs at the base of Chloride Cliff were blamed for the lack of bighorn sign there in 1955, when so little sign could be found that the area was eliminated from consideration in the final figures of the census. Annual surveys confirmed this view until 1959, when we found the area around Sulphur Spring had been "turned into a barnyard" during the previous summer, with sign indicating a population comparable in size with that of Navel or Nevares Springs.
To confuse the picture somewhat, however, the sign had been substantially reduced by November 1960, indicating much less use and suggesting the temporary presence of a drought-displaced population from another area.
Our rediscovery on March 14, 1960, of Lost Spring (4 miles south of Warm Springs Canyon) disclosed the existence of an uncounted but stable herd in the heart of the burro country between Warm Springs Canyon and Wingate Pass. (See "Wild Burro.")
At Quartz Spring, where no sheep were seen in 1955 but 15 were estimated, we actually observed 28 in one band on the first day of June 1960.
So we find that we have more home areas and more bighorn to the area than we thought we had in 1955. If the picture that has developed at Navel, Quartz, Lost, and Nevares Springs could be projected over the entire habitat, an estimated total herd of 1,000 would be conservative (and has been suggested). Present data, however, do not warrant an optimistic generalization.
If all the home areas of the Death Valley bighorn were substantially comparable with those studied in the Black and Funeral Mountains between and including Virgin Spring in the south and Klare Spring in the north, it would appear reasonable to use this region for projecting a total population estimate. The region in question comprises an area approximately 75 air-line miles long and 8 air-line miles widea one-story ground cover terrain completely free of predation, forage competition (except rodents), and relatively free of human encroachment. We have counted approximately 100 individuals in 3 contiguous areas of the 10 population centers involved: Lemonade, Navel, and Nevares Springs. This is an average of 30 sheep to an area and projects a not unreasonable total of 300 bighorn for the entire 600 square miles under immediate discussion. This represents a 33 percent increase over 200 estimated for the same area in 1955.
With the exception of Tucki Mountain and the Butte Valley-Lost Spring area, the remainder of the Death Valley habitat comprising the northern Grapevines, the Cottonwoods, and the Panamints is dominated by high desert communities of drought-resistant conifers, presenting infinite possibilities of variation in the life history of the local bighorn populations.
The presence of mountain lions in the forested areas indicates the possibility of significant predation; exotic competition for food exists in all of these regions; mining activity, with a concomitant human usurpation of water and forage withdrawal is prevalent; poaching is believed to be common practice throughout these sections of the bighorn range; and deer-bighorn competition in the Grapevines and Cottonwood ranges is beginning to assume significant proportions. (See "Deer.")
All of these factors complicate census estimates, but since it seems imperative that an estimate of the population be made, this report offers the following figures as of January 1, 1961: The theoretical or "liberal" high of 1,260 for 1955 should be increased at least 15 percent by the new populations discovered since that time, which would bring the figure to 1,450. A wet cycle of 3 years could conceivably result in that many bighorn for a brief time. The actual maximum estimate is indicated in table 12 to be 915. For purposes of discussion, it is suggested that these figures be rounded out to 1,400 as the theoretical high, and 900 as the actual maximum. Deduct 30 percent for errors in censusing due to ram overlap and to provide for annual lamb mortality for a minimum of 662, or 600, to round out the minimum figure.
The above minimum figures take into account the fact that the best waterhole counts occur during July, August, and September, while the lamb crop is usually at a peak and rams may be counted more than once in their overlap of service at the springs. The minimum figure of 600 is believed to represent the average carrying capacity of the total biota when annual lamb mortality has restored the balance around January 1 of each year.
There is no evidence to indicate a perceptible general increase or decrease in numbers of the total herd. By February 1959 the sheep had responded to the autumn rains by dispersing to the outlying ranges not readily available to observation. Reports of sightings dwindled, and once more recurrent rumor had the bighorn dying out. History began to round itself out with autumn rains in November 1960. January flowers and bighorn were in the washes and on the alluvial fans, and bighorn were "on the increase" again in 1961.
Two types of density fluctuation must be expected: The annual appearance and disappearance of the lamb crop, and the intermittent but inevitable drought reduction of the entire range to a critical degree.
If a spring census were possible, the age class and sex ratio would probably be indicated at about 35 percent ewes, 35 percent rams, and 30 percent lambs. A winter count of the same population would indicate about the same number of ewes and rams, but the ratio would now be at 45 percent ewes, 45 percent rams, and 10 percent yearlings and younger.
To what degree prolonged droughts affect the general population is not known. (See "Food" and "Water" and "Chief Causes of Mortality.")
Distribution is affected by the variability and fluctuation of food and water supply, but as indicated in the bighorn distribution map, bighorn tend to find their way to some extent into all areas of the region supporting bighorn forage. There appears to be no other natural factor of limitation. Left to his own resources, the Death Valley bighorn will apparently forage at any elevation in any season, below sea level at Badwater or atop Telescope Peak, summer or winter, as his demand for food meets the law of supply. A possible exception to this is indicated by the fact that we have never found bighorn tracks in the snow on Telescope Peak, which at least suggests a tendency to drop below the snowline when the snow is heavy. The absence of sheep or sheep sign at Doe Spring in the Grapevines is discussed under "Competitors and Enemies."
No food exists in much of the salt-flat area of the valley floor, but the rumor still persists that "bighorn cross it from one mountain range to the other."
In 1952, Ranger-Naturalist Richard Hartesveldt took pictures of bighorn tracks "about a mile out in the salt flats west of National Park Service headquarters." These he believed to be made by sheep crossing the valley floor, although coyote tracks have been misidentified as those of sheep under similar conditions. In October 1958, a ram was observed by the road crew "crossing the road 4 miles north of headquarters and heading out across the valley." Whether the ram continued to "cross the valley" is not known, since the crew moved on with their work. There have been other indications that sheep may for some obscure reason occasionally cross the salt flats to the other side.
That they do not do it regularly in any sense of the word is indicated by the fact that during the entire winter of 195455, when we were making the Badwater observations, we repeatedly checked the edges of the salt marshes from Mormon Point to Salt Creek, but we found no sign, old or new, of sheep crossing the highway and salt flats beyond at any point.
In the autumn of 1955, Geologist Charles Hunt began a 6-year mapping project focused on the salt deposits of the entire region. During the first four winters, he walked across the valley more than a hundred times at short enough intervals of distance to permit accurate mapping of every foot of the terrain, including all roads, and trails of all kinds. He found no sheep trails, no sign of bighorn ever having crossed the valley.
Ancient game trails indicate prehistoric use of such lower elevation springs as Travertine (in Furnace Creek, 1-1/2 miles from the Inn), Texas (at the campground), and Cow Creek (at the Government residence area). During that time sheep undoubtedly ventured regularly into the salt flats as far as the mesquite habitat extended. We did not know until the summer of 1959 that some of these areas are still subject to periodic utilization. On August 28 of that year, I followed a 4-year-old ram through the heat waves as he headed away from Raven Spring, away from Nevares Peak, away from the Funeral Mountains toward the valley floor.
For 2 miles he walked steadily across the desert varnish of the mesas, then dropped down into the wash above Cow Creek Spring. There, to my amazement, I found fresh beds, tracks, and droppings of an entire band in the shade of cut banks, along the trails descending into the wash, and surrounding the Bebbia plants on which the sheep had been browsing. The sign showed the animals had stopped a hundred yards above the present utility area.
Human usurpation of water is a more prevalent factor in locally limiting the number and distribution of bighorn in the Death Valley region than is generally believed. The burro, on the other hand, has had less effect on the bighorn than it has had on the habitat in general. This will be discussed more fully under "Competitors and Enemies."
Field identification becomes an increasingly important factor in determining numbers, distribution, and movement of bighorn in the low density populations of the Death Valley region. The foreseeable future holds no prospects of success for any other method. Belt transects could be laid out by the dozen and yield no sheep at all. Trapping, drugging, marking, and the application of Lincoln's index involve a larger personnel than is likely to be available here.
There is probably no way of proving to the skeptical the practicability of field identification as a substitute for trapping and marking and no effort will be made in this report to do so. Acceptance of the principle would carry with it the responsibility of practice, with a greater expenditure in the field of time, energy, and money than appears to be available in most bighorn study areas.
Figures 25, 26, 27, 28, 29, 30, 31, 32, 33 present a brief introduction to the method of development in field identification that was followed during the investigations upon which this report is based.
No two sets of horns are alike, nor are two single horns identical. Determining the difference between them may be as rewarding to the fieldworker as the precise determination of fingerprints is to the sheriff's office. An understanding of horn development as an indicator of age and sex is necessary for work with live bighorn in the field. Old horns and bones sometimes constitute the only sign of the former presence of bighorn, and are indispensable to the trained observer, who may find nothing to record except what the bighorn has left behind it.
Sign reading as a science suffers from the fact that the average fieldworker has never proved to himself how wrong he can be. This conclusion is inescapable to anyone who has spent considerable time actually studying the sign left by known numbers of bighorn, of known age classes and sex, of known time and activity, and of known climatic conditions.
The importance of sign reading can scarcely be overemphasized in a barren region like Death Valley, where the entire population of bighorn may vanish into the back-country for months or even years at a time and very few if any of them can be found for actual observation. Concepts of numbers and distribution, age and sex ratios, lamb crops and survival, forage conditions, and adequacies of water supplies are formed from the accounts of happenings recorded only by old bones, footprints, bedding basins, fecal pellets, and partially eaten plants.
The importance of accuracy in conclusions drawn from sign reading is obvious, and yet the need for training in method, or for testing the extent of knowledge of the subject is generally ignored. The margin of error widens as one false premise is drawn from another and the stockpile of misinformation reaches such proportions that facts become hopelessly buried beneath it.
A population estimate based on conclusions drawn by an incompetent observer from sign alone can be less than worthlessit can be extremely costly to a management program. The cost of man-hours wasted in gathering misinformation and the launching and carrying out of faulty projects generated by this blundering activity could be staggering.
Probably the most outstanding example of this is the case of the feral burro which will be referred to under "Competitors and Enemies" but must be mentioned here because of the probability that the incredible mass of misinformation about the wild burro that has been accumulated and disseminated through highly respected channels was initially the result of the misreading of sign at waterholes. The obvious abundance of burro sign and the apparent lack of bighorn sign in the vicinity led to a quick and specious conclusion that there was no bighorn sign present, that therefore bighorn did not utilize the water source, and that this failure was due to the fact that burros had some how rendered it unusable for bighorn, thereby forcing them to abandon the entire area.
There are many springs in the Death Valley region today which apparently tell the same storyunless the observer has learned to look beneath the surface of the dust or gravel surrounding the water where the bighorn pellets have been inadvertently buried by the shuffling feet of burros milling about the water source during the night. The drinking, mating, fighting activity of a band of half a dozen burros can obliterate a 6-month collection of sheep sign in one visit to a spring. And when it is known that burros not only leave much more sign while watering but water three times as often as bighorn, it becomes increasingly clear why the sign is so often misread and misinterpreted.
The case against the burro was further heightened by a lack of knowledge concerning the discoloration of standing water by the chemical action of the soil supporting it.
During our reconnaissance for the 1955 Death Valley bighorn census, we made our first survey of "burro-fouled springs," and our failure to find any led to some surprising conclusions. We visited 45 springs which were in constant utilization by burros and found only one contaminated by defecation. With two observers on each side of Cottonwood Creek, we had walked its entire length and found exactly one incidence of defecation in the water.
It was there we learned where to look for bighorn sign at burro watering places. We found sheep tracks coming down to within 25 feet of the water, then vanishing in the loose earth which had been churned up by much shuffling of burro hooves. It occurred to us for the first time that the bighorn sign was underneath the burro sign and immediate investigation proved it to be the case; thus, the entire picture was changed.
At Rest Spring in the Cottonwoods, we found no droppings within 30 feet of the pool, but the water had been trampled in by burros and was a brownish urine color and had an odor that could be detected while standing above it. We assumed that here was an instance of water pollution and photographed it as such. We saw several other springs in the burro country bearing the same evidence and listed them accordingly. It was not until we began working in the Funeral Range, where there were no burros, and found springs with no animal tracks of any kind in or near them and yet found the water dark brown and with a strong odor that we realized that the "fouling" was coming from the ground itself.
Poison Spring in Indian Pass is a typical case in point. The water is brown and bitter the year around, but it is perfectly acceptable to bighorn. Caught there in July without water, we dug a small shaft above the established point of discharge, which filled with clear, uncloudy but exceedingly bitter water, and we drank it with no ill effects whateverbut we thought we knew why it had been named "Poison."
By the time the 1955 census was over, our lessons in burro-bighorn sign reading had progressed to the point that we wrote, "as far as his exact influence on the bighorn is concerned, we do not know enough to say." The first fact regarding our knowledge of burro-bighorn interrelationships had been recorded.
Some of the problems in the reading and interpretation of browsing sign has been discussed under "Food" and will be developed further under "Competitors and Enemies." The subject of sign is also discussed under "Water," "Trails," and "General Habits."
We must emphasize the necessity for study of sign reading by field-workers. Bighorn sign cannot be studied apart from the bighorn themselves. It is not feasible to expect an observer to be able to interpret the meaning of sign when he is not familiar with the type of activity producing it.
The "dance" described under "Play" could scarcely be envisioned from the sign unless a similar incident had been observed previously. We had often wondered how certain bushes and mesquite trees acquired such a devastated appearance until we became aware of the "signpost" aspect of their destruction. (See fig. 40.) Likewise, the churned surfaces left by the preliminary hassling of a ram bout had repeatedly left us puzzled until we saw it happening.
Usefulness of Beds
By the same token, the usefulness of beds as a definitive factor in sign reading appears to be more limited than was to be expected. The presence, number, size, placement, time and season observed, and weathering of beds contribute to estimates of population, herd composition, and activity but provide the experienced observer with only one certainty: Bighorn have been there.
The number of beds is not an accurate indication of the number of sheep involved. The Badwater band of 6 usually left but 2 or 3 beds behind them, but on one occasion the Furnace Creek band of 7 left 22 fresh beds on one site. Certain bighorn seldom prepare basins of any dimension, while others prepare and occupy several in one night. The suitability of some areas for bedding may cause repeated use of the same site, each use complicating the story to be read by the observer. We have never seen a band bed in the same place on consecutive nights.
The quality and quantity of soil available for bedding must be considered before conclusions are drawn. The crest of Paleomesa, a vantage point above Navel Spring, is covered with hundreds of beds of all conceivable sizes and ages along the ancient trails that contour the mesa ridges and the edge of cliffs towering 200 feet above the spring. The soil of the mesa is apparently ideal for both the preparation and the preservation of sheep beds and could contribute to an overestimation of the population.
The opposite is true of the rocky terrain above Nevares Spring, where a population at least as large and of as long standing as that at Navel has left practically no beds at all. The solid rock of the mountainside successfully resists all bighorn techniques of basin building and offers so few natural bedding sites that bighorn seldom spend the night within sight of the springs and, when they do, leave little but droppings for the record. The big ram Broken Nose on September 2, 1957, lay in one of the most uncomfortable beds on record for over 4 hours, and he left no tracks or basin and but one small group of small "lamb-size" pellets.
The term "vantage point" draws attention to the significance of placementwhere the bed is located with relation to the surrounding terrain. The tendency of bighorn is to survey the watering area and surrounding terrain both before and after wateringbefore, to assure themselves that it is safe to water, and after, apparently to give their bodies a better opportunity to absorb the water ingested. And, of course, during the rut the promontories become important points of contact for both sexes.
In the summer, a fresh bed on the southern exposure of a promontory at an elevation of 1,000 feet is very likely a night bed; on a northern exposure it may be either a day bed or a night bed, as it would be on a southern exposure with sufficient elevation to reduce the temperature to a point of toleration for daytime bedding.
A typical observation was recorded at Deadman's Curve on March 22, 1956: "9:45. Temperature (shade), 78°. Ground temperature 94°. All sheep (band of seven) are bedded in the open wash in full sun. Apparently they like the 94°. They seem to have no feeling of security there one-quarter of a mile from even a low ridge, let alone a cliff for escape." We have no record of bighorn bedding in low open washes either at night or in the summer.
During the precensus reconnaissance work in July 1955, we camped overnight to check a report of abundant fresh sign on the rim of Indian Pass, 1,500 feet above and a mile northeast of Poison Spring.
The approach to the rim looked hopeless: flat rolling hills almost barren of vegetation of any kind, no cliffs, no place for sheep. Sign of mining operations of the 1930's was everywhere.
But there was a change at the rim. In the vast canyon area there were many miles of dusty sheep trails bending down from the rim edge toward the springs below. The dominant vegetation was Encelia sp., much browsed. A quarter-mile strip on the rim disclosed three current bedding areas, yellow-dusty and rimmed with clustered pellet groups and "apple-seed" lamb droppings. There were three groups of 17 beds, 30 beds, and 24 beds, less than 100 yards apart. Other smaller groups were more widely scattered.
Let us consider this observation step by step: (a) The presence of the beds indicated the past presence of sheep. (b) The freshness of the beds (yellow-dusty) indicated use of the beds within the last few days. (c) More than 70 fresh beds suggested a relatively high number of sheep, certainly more than 2 or 3, probably at least 12 adults. (d) The relative size of some of the beds (as well as of the pellets) indicated several lambs, a probable minimum of six. (e) The placement on a high rim in full sun close to a water source suggested a lambing ground, which checked with (f) the season during which the beds had been and were being used. And finally, (g) the weathering of all the beds, both old and new, indicated annual use of the area, or a resident herd.
The usefulness of weathering in determining the age of a bed is modified by the nature of the terrain where the bed is made. Exposure to wind, water, and sun cause weathering, and the more protection a bed has from them the less the bed will change with time. Therefore it becomes obvious that a bed lying well back in a north-facing cave, protected from all three modifying factors, will look fresh longer than the one on the south-facing promontory fully exposed to them.
Examination of a fresh Death Valley bighorn bed will disclose dust as the denominator of freshness in most cases. If the exposed subsurface dust of a bed in the open has not been blown away or disturbed by the wind, washed back into the soil or gravel, packed down by rain, or crusted, bleached, or burned back to the color of the surrounding earth by the sun, it probably has not been there very long.
Scraping out a similar basin by hand and comparing the quantity, color, and looseness of the dust in both may be helpful, but it must be remembered that a sheep may have lain in the bed all night and packed it down. However, the manner of their rising usually disturbs the surface enough to make comparisons profitable.
All data point to many variables in "bed reading." During a dry winter with no rain, a low sun, and a hazy sky, a bighorn bed could remain "fresh" as long as no wind blew to remove the dust and fill in the basin with gravel and bits of vegetation. On the other hand, the "reading" of a bed can be made impossible in a matter of minutes by either rainfall or heavy wind.
Usefulness of Pellets
Pellets are probably the least understood and most misleading single factor in sign reading. There are so many factors, many of them still unknown to us, governing the distinguishing characteristics of pellet deposits, that definitive classification by age class and sex will not be attempted here.
Earlier in this study we shared with many other observers the comfortable conviction that the size of ram, ewe, and lamb droppings would naturally reflect a relative difference in the size of rams, ewes, and lambs, and therefore render field determination of age class and sex of unseen animals a reasonably accurate procedure. The largest pellets, especially those deposited in cowlike clusters, were by this concept ram pellets, the intermediate size, ewes, and the smallest, lambs.
As soon as we began in 1954 to follow the Badwater band and to collect and tabulate the pellets of sheep of known size, sex, and age, the smooth flow of misinformation from the old formula was interrupted. A long and enlightening series of lessons in sign reading had begun.
The first of our 16 interim reports contains these notes:
By the time the 1955 census data had been digested and our fourth report written, these conclusions had been reached:
August 24, 1955, found us camping above Willow Creek in an effort to keep the spring under full-time observation. Seven weeks after the census, we found "sign plentiful, with a full complement of lamb sign now, where in July, sign showed 400 adult pellet groups to 8 lambs."
Our confidence in the old relative size formula for determining lambs was still unshaken. Still firm, too, was our conviction that relative brightness of pellet varnish was a dependable measure of the time lapse since deposition. This time-honored practice received its first setback on December 22, 1955:
We spent the day with a ewe and a 4- or 5-month-old lamb. The pictures we took identified her as one of the band later to be observed.
There was a very strong wind blowing all day, and from this we were able to make a valuable observation.
During the time the sheep were there, we stayed out of the wash to keep from "spooking" them. At the same time, we were anxious to show some friends who were helping us what sheep sign looked like and how much would be left by a ewe and lamb being very active in one area all day. The sheep left about 3 o'clock, and we went into the wash to study the sign in two or three marked areas. To our surprise, none could be found that looked less than a week or two old. The wind had filled in the tracks with gravel, dried and scattered the pellet groups, and disturbed browsing sign to such an extent that had we not just seen the ewe and lamb there all day we would have guessed that none had been there for several weeks. We found several adult pellet groups but no lamb pellets.
This was our first realization of how quickly and how much the desert winds can modify a story written on the face of the land.
On the morning of December 29 came the first of many opportunities to study the story told by pellet groups at bedding sites. A band of 5 had bedded down the day before at dusk, and by 7:30 that morning the 5 sheep had left 11 beds and 14 pellet groups, with nothing to indicate that one of them was a lamb.
The following day the band increased to five ewes and two 5-month-old lambs. A week later we were making a special effort with this band to "check variation in pellet colors, find lamb pellets if possible. Have not yet found any, although we have come across dozens of adult groups, some brown, some black. This variation is probably caused by their spending half time on green and half on dry food."
Although we intensified our search for lamb pellets, taking turns watching the sheep with glasses and trying to trace their course after they had passed by in the wash, it was not until January 15, over 2 weeks after they came into the area, that we "found lamb pellets for the first time. The absence of lamb pellets from these two continues to amaze us."
On February 20, after 3 weeks of constant daily observation of Old Mama and her new lamb, we noted that the "ewe's pellet groups are numerous, pellets large, squarish, and both black and brown by groups. This size and shape have been considered ram sign." We had been collecting droppings from this ewe for over 2 months, and comparison showed a considerable increase in size during that time.
Later the same day, "Buddy and I went up and down the wash, combing it for lamb droppings. Found some looking fresh, but big enough for a 6-month-old animal. Is it possible that this feed, making ewes' droppings so much larger than they were a month ago, is also making this lamb's droppings much larger than would be expected ?"
The old formula was up for examination. Here were ewe droppings which we would have unhesitatingly labeled ram droppings had we not seen them fall. Questions began to multiply. What caused this change? Was it the type of forage available now? Was it because the ewe was old? Did parturition have something to do with it? Why could we find no "apple-seed" lamb pellets? Was it possible that a 4-week-old lamb utilized its food to such a degree that fecal deposits were so small that they could not be detected?
On February 25, we combed the wash all day for sign. Still no lamb sign of any kind, but we found fresh, wet, soft ewe pellets that I had seen fall 45 minutes beforeOld Mama's, still extra large (ram size), and jetblack. These I placed in various places, in shade and out, exposed to the wind and sheltered from it to check aging. So, belatedly, controls in the study of sheep sign had been inaugurated.
The exhausting task of keeping the lamb under constant fieldglass surveillance was finally rewarded on the 29th: "Lamb is a ewe. We finally caught a urination in good view. Nursed 11 seconds at 10:20. Several 'apple seeds' hung on lamb's whitening rump patch, indicating that (larger) pellets found in wash so far were of the older lambs of the previous band. At 10:25, we finally saw pellets in process of deposition. We now know why they have been so hard to find. Total of 5 in group. One-eighth inch in diameter, one-fourth inch long. Genuine 'apple seeds'."
The small number of pellets in the lamb deposit served as a reminder that the ewes' pellet groups seemed to be more numerous and to contain less pellets per group than previously. Why was this? The type of food? The type of activity engaged in at the time of deposition? Did age class and sex enter into it? Would the number deposited while moving along a trail be consistently different from the number deposited while browsing? Would the group left at a daybed differ from that at a nightbed? Was a physiological malfunction in evidence?
On March 2,1956, by 2:10 p.m., we had counted 6 (browsing) pellet groups. They contained 50, 54, 64, 74, and 80 pellets respectively, all adult, but we found no more from the lamb for several days.
Old Mama and her lamb were joined on March 13 by a 2- or 3-year-old ram, three ewes, and a lamb about the same age as her own6 to 7 weeks. On March 17, we followed them up on the mesa. Along the crest ledge [a bedding area] we found droppings of all colors from white (bleached by at least a year's weathering in the open) to jetblack. Five groups were still wet at 12:30 (5-1/2 hours after band left nightbeds, suggesting both an abundance of green in forage and cool and relatively damp atmosphere). We counted pellets in 46 groups: 41 adult, 3 young, 2 lamb (67 weeks, those with this band). The count per pellet group ranged for adults from 36 to 406, for young from 43 to 74, and for lambs from 70 to 174.
There were 7 groups with over 300 pellets on the edge of the beds, suggesting that after the sheep had an all-night rest the number of pellets greatly increased, for the average on the trail was something under 100.
It is a point of interest that of 46 groups only 2 of the young lambs' (apple-seed size and shape but brown) were found. It was becoming quite apparent that the ratio of lambs in a group (band) is not indicated by the ratio of the pellet groups.
On March 20, 1956, we counted pellets in 40 groups. The groups dropped while feeding averaged 128; daybeds, 175; nightbeds, 310.
On March 22, 1956, at 10:45, we noted that the seven observed sheep had taken an hour's siesta in the sun. We went to study sign and found five beds, three adult pellet groups, all nearly as large individual pellets now as Old Mama's have been and all very wet and jetblack. There was no sign of pellet groups of the two lambs, although they both bedded there. Individual count of each daybed adult pellet group: 105, 124, and 128.
By June 27, 1956, observations on pellet deposits in another area were cautious: "Nevares. Much use. Wet pellets indicate bighorn still in area. Examined all seeps, all used. Long Spring used the most. Large and small, black and brown pellets, big enough for rams, and small enough for lambs, some apple seeds."
Throughout the winter of 195657, we counted, collected, and compared droppings from known animals and of known times of deposition, and we were gradually becoming aware of a staggering number of modifying agents which could contribute to the condition of droppings before we found them in the field.
By now we had a constantly increasing number of "control stations" under observation. These "stations" were simply marked deposits of all types of pellets from known animals of all age classes and sexes, left in the field in as nearly all types of terrain as could be kept under observation. We noted the effects of protection from the sun and the lack of it; we checked the bleaching effect of rain in both shaded and unshaded terrain. We began to see a correlation between the rate of decomposition and the elevation of the area where the deposit was located. It became apparent that the type of food ingested affected the size and color of the pellet, as before noted; additionally, the amount and quality of "varnish" on each pellet appeared to be at least partially determined by the chemistry and growth stage of the plant eaten: thus, we began to see that pellets with the thick, black coating, generally associated with abundant green forage, retained their sheen and original color much longer than those with the thin, light-brown "varnish" often found in association with dry forage.
The word "association" began to assume a new importance in dropping analysis, and eventually our entire concept of sign reading began to re-form into what we now refer to as the associative principle of sign reading. This means simply that any one factor in a sign picture acquires authority only from its association with other factors inherently present in every picture.
January rains released the bighorn from our "control" areas around permanent water supplies, which gave us 6 months in the field with not only the opportunity but the necessity of depending on sign only for the accumulation of bighorn data for that period. Our observations of droppings were beginning to reduce them to the auxiliary status which they occupy in the sign story today. Some of our notes of this period show the beginning development of the associative principle:
An entry made later the same day in the Nevares area indicates a further maturing of the associative principle in sign reading:
By now we were considering the possibility of the small size of the ram pellets being associated with the rutting season. We had seen some of these rams and checked their ram-sized pellets during the previous winter. We wondered if the smaller pellets could indicate a more complete utilization of food materials by the fully mature rams during the rut, which might at least partially explain why they remain so physically fit and yet seem to eat so little during the peak of the mating season. The younger rams, whose pellets remain relatively normal during the same period, seem to eat more normally, but some also evince definite signs of strain and fatigue by the end of the season.
October rains ended 1957 sheep observations abruptly but also presented us with one of our most valuable lessons in sign reading: With the exception of browsing evidence, all sign at Nevares in the immediate spring area began to disappear within a week or two after the rains; and by January 1958 practically none could be found except high up on the mountain where the angle of repose and lack of soil precluded water retention.
Tracks and droppings had completely disappeared, disintegrated by the action of ground salts being carried to the surface and impregnating the tracks and droppings through capillary action. The ground turned white and fluffy, and the outlines of tracks and trails first softened, then blurred, and finally disappeared. Droppings became encased in salt crystals forming rough ovals from a half to a full inch in length, containing the pellets which were eventually reduced to particles by the expansion of the salts and finally returned to soil as the salt crystals eroded away.
We now understood why Nevares Spring had been left out of the 1955 census effort, but the full extent of how far we might have been misled by this phenomenon is indicated by a notation made on December 1, 1955: "Survey of Nevares Spring area. Found complete reinvestment of bighorn of that area. From 20 to 30 bighorn using grasses and shrubs of area as well as water. No sign of past habitation, indicating this as first year of reinvestment since vacating of premises by Adolph Nevares and installation of gate to access." We estimated no population at Nevares in 1955, yet counted 47 individuals there in 1957.
The pellet story for this report was given a final fillip by the band of 14 bighorn luxuriating among the flowerbeds of Death Valley Buttes on February 11, 1958:
This much time and space has been devoted to the development of the specific study of bighorn droppings in the hope that through the illumination of some of the false concepts common to the study, some light might be shed on a true one.
Usefulness of Tracks and Soil
The associative principle must be applied to tracks as well as to the other elements of the sign picture.
Figure 42 illustrates some of the problems of determining how long ago a track was made and by how many animals. The fact that each animal leaves two sizes of tracks often is not taken into consideration by the observer, and he tends to double the number that should be estimated. In figure 42, the nature and placement of the matrix in which the print was left preserved the fresh look of these tracks for several years before they were finally covered by another runoff. This is true in varying degrees in many spring areas.
The opposite can be true as well. At Nevares Spring, for example, the self-rising nature of the soil often obliterates tracks as well as droppings in a few weeks.
The soil around Navel Spring, owing to the natural cement in the soil, will resist the imprint of a track completely if the ground has had time to set up after a rain. On the other hand, if the track is made during a rain or while the ground is still wet after a rain it will look fresh until it is finally broken up by other watering animals or covered up by another runoff.
In soft gravel it is sometimes impossible to tell what made a track unless some pellets can be found to further a diagnosis.
Sometimes, too, bighorn will step in the tracks of the animal preceding them, thus confusing the sign as to the number of animals passing.
When the band of 19 sheep on Death Valley Buttes was breaking up in January 1961, I watched 8 of them racing across the north wash toward Corkscrew Peak, but when I reached their crossing no amount of searching revealed more than the tracks of 4 animals.
Definite conclusions from signs as to how many animals did what and when are seldom made by workers who have taken the trouble to find out how misleading signs can be, and therefore how tentative must be the conclusions drawn from them.
Chief Causes of Mortality
The chief causes of mortality have not been conclusively determined. Evidence indicates the greatest age class incidence of mortality to be during the first and tenth years. (See "Longevity".) We have seen no sick adults, with the possible exception of The Old Lady, where old age made other diagnosis difficult.
Old Mama had afterbirth complications, a cough, a "runny nose," and old age to contend with, and yet she reflected no sign of ill health in her daily activity and was the leader (to our knowledge) of her fourth band when last seen on December 20, 1956. All evidence indicates old age as the chief cause of death in mature ewes and rams. Contrary to considerable data from other areas, loss of teeth in old age plays no significant part in mortality here. Collected skulls of old animals usually display remarkably sound dentition.
Accidents claim some lives. We find carcasses, particularly lambs, at the foot of cliffs, suggesting death by fall. However, it is also known that sick animals tend to bed at the foot of cliffs. We have never found a carcass soon enough after the animal's death to determine the cause of its demise.
In the autumn of 1954, a ewe with a broken leg was found by a visitor at Klare Spring and was reported to the National Park Service before we returned to duty for the winter. Park Service personnel found her dead, and they disposed of her without acquiring further data. In 1940 Sumner (1940) found at Bighorn Gorge the skeleton of an adult ewe that had received a fracture of the right hind leg (metatarsal) a considerable time, probably several years, before its death. The fractured bone had fully healed, although in a bent and shortened position.
By far the majority of bighorn deaths occur during the first year, but the actual cause of death is not known. Poor condition appears to be accompanied by a cough, rough pelage, and increasing lethargy. We have not observed the last stages of this deterioration, but its ultimate fatality is attested to by the continued scarcity of yearlings.
The lambs' need of water and "emergency rations" in spring areas in hot and dry weather appears to be more urgent than in adults, suggesting at least a contribution to juvenile mortality by the harsh environment.
The probable prevalence of lungworm, with pneumonic complications encouraged by dietary deficiencies, should be included in mortality postulations.
Symptoms of trace element deficiency in domestic sheep of Australia, described by Anderson and Underwood (1959), are similar enough to those of the 1956 Nevares herd to justify an attempt to determine whether such a deficiency, especially of cobalt, is a mortality factor here. Through the efforts of Charles Hunt of the Geological Survey, U.S. Department of the Interior, and Lowell Sumner, the assistance of several specialists in the field was solicited for the project, but no conclusive data were forthcoming by the time of this writing.