Bos grunniens Linnaeus, 1766

Leslie, David M. & Schaller, George B., 2009, Bos grunniens and Bos mutus (Artiodactyla: Bovidae), Mammalian Species 836 (3), pp. 1-17 : 2-9

Bos grunniens Linnaeus, 1766

Domestic Yak Bos mutus ( Przewalski, 1883) Wild Yak

[ Bos ] grunniens Linnaeus, 1766:99 . Type locality ‘‘Asia boreali;’’ first use of the current name combination and now considered the binomial for the domestic form (Gentry et al. 2004; International Commission on Zoological Nomenclature 2003).

[ Bos ?] corriculus von Schreber, 1789:? Vide Grubb (2005); see ‘‘Nomenclatural Notes.’’

Bos gruniens Ghainouk Kerr, 1792:338 . Type locality not mentioned and incorrect subsequent spelling of Bos grunniens Linnaeus, 1766 .

Bos gruniens Sarlyk Kerr, 1792:338 . Nomen nudum and incorrect subsequent spelling of Bos grunniens Linnaeus, 1766 .

Bos gruniens ecornis Kerr, 1792:338 . No type locality mentioned and incorrect subsequent spelling of Bos grunniens Linnaeus, 1766 .

Bos Poephagus Pallas, 1811:248 , table xxii. Replacement name for Bos grunniens Linnaeus, 1766 .

B [os ( Bison )]. poephagus: Hamilton-Smith, 1827b:374 . Name combination.

Bison poephagus: Jardine, 1836:259 . Name combination.

[ Bisonus ] Poephagus: Hodgson, 1841:217 . Name combination; said to occur as ‘‘tame and wild samples.’’

Poephagus gruniens: Gray, 1843:153 . Name combination and incorrect subsequent spelling of Bos grunniens Linnaeus, 1766 .

B [ison]. grunniens: Turner, 1850:177 . Name combination.

Poe¨phagus grunniens domesticus Fitzinger, 1860:294. No type locality mentioned; described generally as the domestic form in Tibet and Mongolia .

Poëphagus grunniens , ferus Przewalski, 1879:85. Type locality ‘‘ Altyn-tagh [Mountains],’’ Xinjiang Province, China .

Poe¨phagus mutus Przewalski, 1883:191 , unnumbered plate. Type locality ‘‘Alpine region of the western part of the Nan Shan (approximately lat. 39 u 209N., 95 u E.), between the Anembar-Ula in the west and the Humboldt Range on the east; cf. Harper, 1940, pp. 325–326’’ vide Harper (1945:528).

Bos (Poe¨phagus) grunniens: Huet, 1891:334 . Name combination vide Allen (1940:1259).

Bos [( Bison )] grunniens: Lydekker, 1898:51 . Name combination.

Bos grunniens mutus Lydekker, 1913:33 . Type locality ‘‘eastern part of Ladak [5Ladakh, India], in the neighbourhood of Chang-Chenmo (where they now appear to be exterminated) as far east as Kan-su and northwards to the Kuen-lun, at elevations between 14,000 and 20,000 feet;’’ described as ‘‘the wild race’’ ( Lydekker 1913:32).

Poe¨phagus grunniens mutus: Harper, 1945:528 . Name combination.

Bos (Poe¨phagus) mutus grunniens: Bohlken, 1958:168 . Name combination.

Bos mutus: Bohlken, 1964:325 . First use of the current name combination; current binomial for the wild form of Bos grunniens Linnaeus, 1766 (Gentry et al. 2004; International Commission on Zoological Nomenclature 2003).

Poephagus muths Li, Jiang, and Wang, 1999:49 . Incorrect subsequent spelling of Poephagus mutus Przewalski, 1883 .

B [os]. runniens Wang et al., 2008:76. Incorrect subsequent spelling of Bos grunniens Linnaeus, 1766 .

CONTEXT AND CONTENT. Context as for genus. No subspecies are recognized (Grubb 2005).

NOMENCLATURAL NOTES. We were unable to verify Grubb’s (2005) assertion that corriculus von Schreber, 1789, was a synonym of grunniens Linnaeus, 1766 . All plates (A. L. Gardner, pers. comm.) and text (D. Wingreen-Mason, pers. comm.) associated with J. C. D. von Schreber’s Die Säugthiere in Abbildungen der Natur mit Beschreibungen in the Smithsonian Institution’s Cullman Library were reviewed, and no mention of corriculus was found. Review of all 30 volumes of Die Naturforscher (Halle, Germany) edited by J. E. I. Walch (1774– 1779) and von Schreber (1780–1804) also failed to identify any use of corriculus. No other literature by von Schreber was located that revealed use of corriculus in the nomenclatural history of Bos grunniens . Nevertheless, we retain corriculus von Schreber, 1789, in our synonymy, affiliate it with [ Bos ?], but question its validity.

The nomenclatural history of Linnaeus’s grunniens has involved placements under the genera Bos , Poephagus , and Bison (Gray 1846; Groves 1981; Olsen 1990; Pal 1996; Turner 1850). Harper (1945) and Ellerman and Morrison-Scott (1966) incorrectly attributed Poephagus grunniens mutus , the wild yak, to Przewalski (1883), who named the wild yak, Poephagus mutus , in his original Russian publication. Lydekker (1913) appears to be the 1st to use Poephagus grunniens mutus . Nomenclatural distinction between the wild and domestic forms has been attempted frequently in the literature. The recent Opinion 2027 of the International Commission on Zoological Nomenclature (2003) retained Linnaeus’s grunniens and Przewalski’s mutus to distinguish between the domestic and wild forms of the yak, respectively (Gentry et al. 2004).

The etymology of Bos in Latin is ox, grunniens is grunting, and mutus is mute (a poor description because wild yaks are quite noisy). Along with yak (K in Russian), other common names include drong, brong-dong (wild), ya (domestic male), dri (domestic female), pegu (tame), banchour, kuch-gau, boku (old male), and kotass. Various metaphorical expressions for the domestic yak emphasize its importance for transportation of goods and services throughout western Asia: ‘‘ship of the cold region’’ ( Prasad 1997:517), ‘‘biological snow plough’’ ( Wiener et al. 2003:81), and ‘‘boat of the plateau’’ ( Wiener et al. 2003:165).

DIAGNOSIS

The subfamily Bovinae has 9 genera (Grubb 2005) with species of large size, stout bodies, hollow horns, relatively short legs, long tails with at least a terminal tuft of hair ( Bos grunniens and B. mutus fully haired), broad muzzles, and no facial, pedal, or inguinal glands (Blanford 1888; Lydekker 1913). Five of the 9 genera in Bovinae (Grubb 2005) are currently considered in the tribe Bovini : Bison , Bos , Bubalus , Pseudoryx , and Syncerus . Both sexes of extant species of Bovini have typically smooth horns (often relatively large in females), arising far apart and generally outward and then turning inward; upper molars are strongly hypsodont with ‘‘broad prismatic crowns and an accessory column between the two main columns on the inner side’’ ( Lydekker 1913:11).

Bos mutus , B. grunniens , Bison bison (American bison— Meagher 1986), and Bison bonasus (European wisent) have 14 dorsal and 5 lumbar vertebrae, unlike other Bovini that have 13 dorsal and 6 lumbar vertebrae (Groves 1981; Vasey 1857). B. mutus and B. grunniens can be distinguished from B. bison and B. bonasus by long draping hair on the former’s chest, flanks and thighs, described as ‘‘splendid tresses like a ‘skirt,’ which imparts … an entirely distinctive appearance’’ ( Heptner et al. 1989:550; Lydekker 1898, 1913). Olsen (1990:78) noted that an ‘‘extension of the dorsal margin of the maxilla prevent[ed] the nasal from reaching the premaxillae’’ in B. grunniens and B. mutus but not in Bison . Mass varies widely among Bos , but B. mutus is generally considered the largest in the genus and the 3rd largest extant mammal in Asia ( Harris 2008) after the Asian elephant ( Elephas maximus —Shoshani and Eisenberg 1982) and Indian rhinoceros ( Rhinoceros unicornis — Laurie et al. 1983).

GENERAL CHARACTERS

We focused this monograph on the wild yak, unless particular information was considered comparable between the 2 forms (e.g., physiology, anatomy, and morphology). Specific aspects related to domestication of the yak (e.g., reproductive performance, rangeland management, and meat quality) are confined to the ‘‘Husbandry’’ section.

Female domestic yaks are about 35% lighter than males, which is probably similar for wild yaks (Buchholtz and Sambraus 1990; Przewalski 1876; Schaller 1998). Both sexes have nearly smooth, cylindrical, gray-to-black horns, but those of males are larger and longer and sweep outward and forward more than the upright smaller horns of females (Allen 1940; Blanford 1888; Fitzinger 1860; Harper 1945; Heptner et al. 1989; Lydekker 1898; Schaller 1998); the forehead is ‘‘short, wide, and slightly convex’’ ( Lydekker 1913:30).

General descriptions of the wild yak have been consistent through time (Blanford 1888; de Pousargues 1898; Lydekker 1898; Przewalski 1876; Schaller 1998; Wiener et al. 2003): massive body on sturdy short legs but compact ( Fig. 1 ); small ears; no dewlap; large and rounded hooves ( Wiener et al. 2003); conspicuous hump, more pronounced in males, arising abruptly behind the short neck as a result of elongated neural spines of cervical and dorsal vertebrae tapering level at the mid-back ( Lydekker 1913) and ‘‘not falling away above the hips’’ (Blanford 1888:490); black pelage with rust-brown hues and sometimes ‘‘a sprinkling of gray on the head and neck’’ (Blanford 1888; Lydekker 1898:53) of older adults (except for a rare light golden-brown mutation in about 2% of animals around the Aru Basin, Tibet —Deasy 1901; Schaller 1998:128); tip of muzzle grayish; young dark brown; pelage dense with an undercoat of wool and long coarse guard hairs ( Wiener et al. 2003); long draping hair on chest, flanks and thighs, which is longer and almost to the ground in mature males (# 70 cm long—Schaller 1998); tail long and bushy on the lower onehalf, often described as horselike ( Heptner et al. 1989); few functional sweat glands ( Wiener et al. 2003); and no preorbital glands or associated lachrymal fossa. Generally, Bos grunniens shares similar physical characteristics, but it is smaller, and coloration ranges from black to brown, white, and pied (Blanford 1888; Vasey 1857; Wiener et al. 2003).

DISTRIBUTION

The wild yak occurs on the Tibetan Plateau at elevations of 3,000 –5,500 m, where it ‘‘inhabits the coldest, wildest, and most desolate [treeless] mountains’’ (Blanford 1888:491). It is currently restricted to a small part of Indian Ladak (Fox et al. 1991; Ul-Haq 2002) and Chinese provinces of Tibet, Qinghai, and Xinjiang, with 1 isolated population on the border of Qinghai and Gansu and another near the northern boarder of Tibet and Nepal (Achuff and Petocz 1988; R. B. Harris, pers. comm.; Miller et al. 1994; Schaller 1998; Fig. 2 ). The core range of the wild yak has shrunk northward, and only isolated and fragmented populations occur south and east of that core area in northern Tibet and northwestern Qinghai ( Fig. 2 ). Recent protection from illegal hunting may be permitting wild yaks to recolonize former habitat and increase in numbers ( Harris et al. 2005; Harris and Loggers 2004; Schaller et al. 2005). About 14 million domestic yaks occur from Afghanistan east through China (about 90%) and northward in Mongolia and Russia, with more elsewhere in the world where ambient conditions permit ( Harris 2008; Wiener et al 2003; Zhang et al. 1994). There are probably no more than 15,000 wild yaks in remote high-elevation areas of the Tibetan Plateau ( Harris 2008; Miller et al. 1994; Schaller 1998; Schaller and Liu 1996).

The 300,000-km 2 Chang Tang Reserve (hereafter, Chang Tang), located in north-central Tibet ( Fig. 2 ), was established as a nature reserve in 1993 and upgraded to a national reserve in 1999. Important contiguous reserves to the north in Xinjiang include West Kunlun Reserve (30,000 km 2), Mid-Kunlun Reserve (32,000 km 2), and Arjin Shan Reserve (45,000 km 2). Kekexili Reserve (45,000 km 2) and Sanjiangyuan Reserve (150,000 km 2) are east of Chang Tang in Qinghai. Despite this impressive reserve network, extant populations of Bos mutus and other Tibetan Plateau fauna are still threatened by human activities, including illegal harvest, mining activities and associated roads, and competition with domestic livestock (see ‘‘Conservation’’ section— Harris 2008; Leslie and Schaller 2008; Schaller 1998).

FOSSIL RECORD

The fossil record for bovids from the Tibetan Plateau is fragmentary ( Olsen 1990), but areas to the south in India may have been the ‘‘developmental centre,’’ or close to it, of Bovinae because from the Miocene ‘‘onward the number and variety of Bovine [fossil] genera found in India is out of all proportion to what is the case in other parts of the world’’ ( Pilgrim 1939:27). Bovinae differentiated considerably during the late Miocene (McKenna and Bell 1997:445), giving rise to the early forms such as Proleptobos , Proamphibos , and Parabos ( Pilgrim 1939) .

Pilgrim (1939:253) considered the yak to be a species of Poephagus and, based on the fossil record, placed it in his Taurina group that included Bos , Bibos , and Bison . The Taurina group was thought to have arisen from a common ancestor, Proleptobos , at the beginning of the late Miocene (Groves 1981; Pilgrim 1939). Pilgrim (1939:327) concluded that Poephagus shared characters most associated with Bibos and Bison , but their common ‘‘hypothetical’’ ancestor that lived before the late Pliocene has not been identified. Isotope analyses of fossil and extant herbivores from Kunlun Basin in the northern Tibetan Plateau suggest that the climate was milder and wetter and habitat diversity greater in the Pliocene 2–3 million years ago than they are now ( Wang et al. 2008); such conditions could have led to greater diversification of Bos .

Particular alignment of the yak with fossil species such as Bison sivalensis is debated because of incomplete and lost fossil material (Groves 1981; Olsen 1990; Pilgrim 1939). Nevertheless, Bos mutus likely shares a common ancestry with the North American Bison bison at some point in the past ( Lydekker 1898; Olsen 1990). Most agree that both evolved in central Asia from a common ancestor (Groves 1981). The yak remained in western Asia, but Bison lineages spread north and eventually crossed the Bering Land Bridge into North America sometime in the middle to late Pleistocene ( McDonald 1981; Meagher 1986). Late Pleistocene fossils of extinct yaks have been found in eastern Russia (e.g., Poephagus baikalensis — Verestchagin 1954 not seen, cited in Abramov et al. 1992), Tibet, and Nepal ( Olsen 1990). A skull and mandible from a single wild yak have been described from Quaternary deposits in the Pakistani Himalayas ( Thewissen et al. 1997).

FORM AND FUNCTION

Form.— Most of the published research that relates to form and function has been conducted on the domestic yak, but results likely parallel, or even understate, characteristics and adaptations of the wild yak under wild conditions ( Jianlin et al. 2002; Wiener et al. 2003). Both forms are highly adapted for existence under extreme conditions of low temperature, high elevation and associated low oxygen availability, extreme solar radiation at southern latitudes, and relatively arid conditions (e.g., Jianlin et al. 2002; Wiener et al. 2003). Even under husbandry, domestic yaks do not do well when ambient conditions depart from their ancestral condition ( Wiener et al. 2003).

The pelage consists of 3 types of hairs: long, coarse guard hairs 52 Mm in diameter, intermediate down fibers 25– 52 Mm in diameter, and dense, fine down fibers,25 Mm in diameter ( Wiener et al. 2003). Down fibers grow dense in winter, particularly on the neck, shoulders, and back increasing to 17–30% of the pelage by weight in winter ( Xi et al. 1983). Density of down fiber can be as high as about 3,000/cm 2 ( Wiener et al. 2003). Pelage of domestic yak calves,6 months of age is almost entirely down fiber with few guard hairs; the proportion by weight declines to 62% of the pelage at 1 year of age, 52% at 2 years of age, 44% at 3 years of age, and 43% at 4–5 years of age ( Wiener et al. 2003; Zhang et al. 1982).

Relative to mass, a small female wild yak may be only one-third the size of a large male; in contrast, female domestic yaks are 25–50% smaller ( Harris 2008; Miller et al. 1994; Wiener et al. 2003). Body mass (kg) of adult male wild yaks has been estimated at. 800 kg (Engelmann 1938) and as high as 1,000 kg (Scha¨ fer 1937; Wiener et al. 2003:43) and 1,200 kg ( Lu 2000; Lu and Li 1994); females are about 350 kg (Schaller 1998). Wild yak calves at 3 months of age (62.5 kg, n 5 5) are nearly twice as large as domestic yak calves (33.6 kg, n 5 19), but in captivity, calves grow slower relative to their weight such that at 16 months old, the wild yak is 63% heavier than the domestic yak ( Wiener et al. 2003). Shoulder heights (cm) of wild yak are 175–203 for adult males and 137–156 for adult females (Schaller 1998); 1 newborn was 67 cm at the shoulder ( Zhang et al. 1994). Although not completely disjunct geographically, 2 ‘‘ecological types’’ of wild yaks have been described based on body characteristics, temperament, and geographical location: the smaller, more docile Qilian Mountain type and the massive, aggressive Kunlun Mountain type ( Lu 2000; Lu and Li 1994; Lu et al. 1993).

Horns of male and female wild yaks vary in size and shape and are far more massive in males ( Fig. 3 ). Generally, they have a ‘‘wide lateral sweep, turning then forward and finally upward and slightly bent inward,’’ are smooth except for a ‘‘few low transverse ridges at the base,’’ and vary among individuals (Allen 1940:1260). Early descriptions provide fragmentary summaries of various horn measurements (Allen 1940; Blanford 1888; Lydekker 1898, 1913; Przewalski 1876, 1883). A recent sample of 53 adult male and 12 adult female wild yaks from the Chang Tang provides a contemporary reference (cm): length of outside curve, male 47.5–99.0, female 37.0–64.5; basal circumference, male 26.0–42.0, female 17.5–23.0; and tip-to-tip, male 26–83, female 18–67 (Schaller 1998). In Yeniugou (‘‘Wild Yak Valley’’), Qinghai, a particularly large male had a basal circumference of 45 cm (Miller et al. 1994). Such wild yak horns are used as milk pails by nomadic peoples (Ekvall 1968).

Although there are few published cranial measurements ( Olsen 1990), Allen (1940) provided a general description ( Fig. 4 ): heavy; broad nasals with tapering ends; narrow lachrymal; upper edge of maxillary in contact with middle of nasals; and outer sides of premaxillaries nearly parallel, not tapering. Cranial measurements (mm) from 2 large male wild yaks from eastern Tibet were: tip of premaxillaries to vertex of skull, 576–610; basal length, 506–528; condylobasal length, 540–555; nasal length, 230–255; combined nasal width, 81–97.5; and width of occipital shield, 250–252 (Allen 1940). Numbers of vertebrae are 7 C, 14 T, 5 L, 5 S, 14 Ca, total 45 ( Vasey 1857).

Dental formula of adult yaks is: i 0/4, c 0/0, p 3/3, m 3/3, total 32. No information exists on replacement and wear of teeth in wild yaks, but they have been evaluated in domestic yaks ( Pal et al. 2002). Unlike domestic cattle, domestic yak neonates are not born with their deciduous incisors; the 1st pair erupts after about 1 week, with successive pairs erupting weekly thereafter ending at 4 weeks of age; fully erupted deciduous incisors are 1.2–1.6 cm in length and 0.6–1.1 cm in width ( Pal et al. 2002). At about 2 years of age, the 1st pair of deciduous incisors is replaced by permanent incisors, and that process continues until about 5 years of age; fully erupted permanent incisors are 0.8–2.0 cm in length and 0.8– 1.4 cm in width ( Pal et al. 2002). Wear of permanent incisors is purported to be useful in aging after 5 years, albeit specific standards were not provided by Pal et al. (2002).

Morphology of the penis of the yak is characterized by ‘‘a urethral canal [that] is produced into a short tube free from the terminal cushion-like thickening of the glans’’ ( Pocock 1918:454–455). These characteristics parallel the penal morphology in the genus Bibos and in Bos frontalis and B. javanicus but are disparate from B. taurus (Allen 1940) . The scrotum of the yak is relatively small and hairy, an adaptation to the cold ( Wiener et al. 2003). Semen of the wild yak has 2.13 3 10 10 spermatozoa/ml with motility of 63% (about twice that in domestic yaks), defective rate of 6.3%, pH of 6.6, specific gravity of 1.055, and osmotic pressure of 0.65 ( Lu 2000).

Female reproductive organs of domestic yaks differ by breed and from those of domestic cattle; cervix averages 5.0 cm long and 3.2 cm in diameter with 3 or 4 transverse circles each with small tight folds; corpus uteri are short, averaging 2.1 cm; and ovarian weight is only 2 g (Cui and Yu 1999a; Li 1980 not seen, cited in Wiener et al. 2003). Four mammae are present, and the udder is small and haired. Morphology and anatomy of the ovary (Cui and Yu 1999b), tongue ( Sarma et al. 2005), nasal cavity (Kalita and Kalita 2005), bronchioles (Kalita and Bordolop 2005), spinal nerves ( Kulbhushan et al. 1999), sternum ( Sarma et al. 1997), and thyroid gland (Baishya et al. 1998) also have been described for B. grunniens .

The alimentary organs of the yak have evolved to deal with the limited forage availability and quality in its native range ( Wiener et al. 2003). The mouth is broad, muzzle small, and lips flexible. Incisors have flat grinding surfaces, and the tongue is broad and blunt with highly cutinized and developed papillae. Such adaptations allow yaks to forage like cattle on long grasses or like sheep on grasses as short as 2–3 cm. In winter when sedges such as Kobresia are short and brittle, yaks simply ‘‘lick’’ them up with their rough tongue. Relative percentages of the rumen and omasum of the domestic yak are about 50% larger and 200% smaller, respectively, than in some domestic cattle breeds—the former maximizes intake and microbial fermentation of low-quality forages ( Wiener et al. 2003).

Function.— Unlike some other species of Bos , yaks possess physiological adaptations to the extreme conditions of high elevation, high solar radiation, low temperature, and aridity under which they live (Christopherson et al. 1978; Prasad 1997; Wiener et al. 2003). Adaptations to maximize oxygen exchange at high elevations include an expanded thoracic capacity with 14 widely spaced and relatively thin ribs and large ‘‘trachea supported by annular cartilages at considerable distances’’ ( Prasad 1997:518); attenuation of the hypoxic pulmonary vasoconstrictor response (Anand et al. 1986; Heath et al. 1984), nitric oxide–regulated pulmonary circulation ( Ishizaki et al. 2005), and associated genetic adaptations to hypoxia ( Wang et al. 2006); small pulmonary arteries of 75–250 Mm of smooth muscle with long, wide, and rounded endothelial cells (Durmowicz et al. 1993; Heath et al. 1984); transitional pulmonary arteries of 228–760 Mm in diameter (Heath et al 1984); hemoglobin with a high affinity for oxygen ( Lalthantluanga et al. 1985; Prasad 1997); and persistent fetal hemoglobin with its high affinity for oxygen through life, unlike most other mammals ( Sarkar et al. 1999b). The ratio of right-to-left ventricular weight of the heart is 0.37, lower than would be expected if a species experienced chronic hypertension due to high elevation, as is seen in domestic cattle ( Heath et al. 1984).

Consistently low temperatures and low primary productivity in the range of the wild yak, and most domestic yaks, result in a strategy of heat conservation rather heat production ( Sarkar et al. 1999a; Wiener et al. 2003), although digestive efficiency of low-quality forage may be enhanced ( Richmond et al. 1977; Schaefer et al. 1978). In Tibet, average annual temperatures are only 24 u C and winter temperatures as low as 240 u C are common; most areas have no frost-free days. Adaptations for heat conservation include a compact body, despite a large mass, with relatively short legs, neck, and ears and a low surfaceto-volume ratio; thick pelage particularly on neck, back, and rump; pelage and skin pigmentation always dark in the wild yak to minimize effects of intense solar radiation but maximize heat absorption; thick unwrinkled skin with nonfunctional apocrine sweat glands, except on the muzzle, but with highly developed piloerection muscles; and a thick, but seasonal, subcutaneous fat layer (Wiener et al 2003). Adaptations to the cold are so developed that even the domestic yak shows signs of heat exhaustion when ambient temperatures exceed 13 u C; heart rate and respiration increase and most activity ceases when ambient temperatures approach 20 u C ( Li et al. 1981 not seen, cited in Wiener et al. 2003). Early accounts note the propensity of wild yaks to maximize heat dissipation and minimize heat production by seeking the coldest spots and shade, bedding in snow, and standing in icy water even during inclement weather ( Przewalski 1876).

Sense of smell is keen; eye sight and hearing less so (Blanford 1888; Bower 1894). Przewalski (1876) and others described the ease with which wild yaks could be stalked, particularly upwind, yet other early accounts and present day researchers often remark on the species’ wariness; when startled, they often flee many kilometers ( Rockhill 1895; Schaller 1998).

ONTOGENY AND REPRODUCTION

Estrus has been described in detail in the domestic yak ( Wiener et al. 2003); we presume it to be comparable in the wild yak. Both wild and domestic yaks are seasonal breeders ( Zi 2003). Generally, 1–4 estrous cycles of about 20 days each occur during summer, and up to 75% of female domestic yaks conceive during their 1st estrus of the year. Estrus generally lasts,1 day (Sarkar and Prakash 2005). Physical changes of female domestic yaks in estrus include swollen vulva, vaginal redness, mucus discharge, raised tail, and frequent urination (Sarkar and Prakash 2005; Wiener et al. 2003).

The majority of females breed for the 1st time at 3–4 years of age, but this, and annual timing of estrus, varies depending on climate, latitude, elevation, and availability of nutritious forage ( Wiener et al. 2003; Yu and Li 2001; Zi 2003). Gestation is 258–270 days, and premature termination of pregnancies from unknown causes can be 5–10% in domestic yaks ( Wiener et al. 2003). Postpartum anestrus is about 125 days. Peak productivity of female domestic yaks occurs at 5–6 years old and declines after 9 years of age. Domestic yaks generally produce a calf every other year, or longer (Buchholtz and Sambraus 1990; Wiener et al. 2003), which parallels observation of wild yaks (Miller et al. 1994; Schaller 1998). Most calves in Chang Tang are born from mid-May through June (Schaller 1998).

Parturition of the domestic yak occurs during the day, rarely at night, in a sheltered location away from the herd ( Wiener et al. 2003). Birth is often from a standing position although the female may spend considerable time lying on her side. Females may be very aggressive during parturition. Twinning is rare,,0.5% of births of domestic yaks. Offspring are precocial and attempt to stand within about 10 min postpartum; 1st nursing occurs 11–30 min postpartum and may last 3–5 min ( Wiener et al. 2003). Females and their offspring rejoin the herd shortly thereafter ( Fig. 5 ). Similar to muskoxen ( Ovibos moschatus — Lent 1988), groups of wild yaks will protect offspring from threats by forming a ‘‘phalanx, calves in the centre [and] some of the full-grown males advancing to reconnoiter’’ ( Przewalski 1876:190; Rawling 1905; Schaller 1998).

Ratios of young of the year to adult + juvenile (2–3 year olds) females in Yeniugou, Qinghai, were 20 calves: 100 females and ranged from 9.7 to 49.0 in various herds (Miller et al. 1994). To the west in Chang Tang, percentages of calves to females were considerably lower and ranged from 1.0% to 12.7% in the early 1990s, with 2 years of apparent reproductive failure or loss of all offspring to predators (Schaller 1998). Given the vulnerable status of the wild yak, recent interspecific cloning experiments ( Li et al. 2007) with other bovine species may be applied in the future.

ECOLOGY

Population characteristics.— Accurate densities of wild yaks are difficult to estimate because of the large size of the Tibetan Plateau, seasonal movements, and greatly reduced numbers from past and present illegal hunting (Clark 1954; Harris 2008; Harris et al. 2005; Schaller 1998). These factors and demarcation and size of survey areas result in widely disparate density estimates that may do little more than reflect a highly clumped and seasonally dynamic distribution of extant wild yaks. For example, in Yeniugou, Qinghai, Harris (2008) estimated that 1,200 –1,700 wild yaks occupied 1,100 km 2 from the early 1990s through 2002, or 1.1–1.5 individuals/km 2. In sharp contrast, only 9 male wild yaks were counted along a transect that covered 20,000 km 2 in western Qinghai just south of Yeniugou (Schaller 1998; Schaller et al. 1991). Regardless, overall densities of wild yaks are clearly much lower now than they were historically (Bower 1894; Harris 2008; Przewalski 1876; Schaller 1998).

Maximum life span of the yak in captivity is generally about 20 years ( Wiener et al. 2003). One wild yak lived 22 years and 9 months in the Beijing Zoo, China ( Weigl 2005). Longevity probably is comparable in the wild (Schaller 1998). Miller et al. (1994) found that the oldest of 6 reliably aged wild yaks died at 16 years, based on cementum annuli. Sex ratios are difficult to estimate because many males occur singly or in small groups and are widely spaced. Of 507 wild male yaks observed in the Aru Basin, 36% were alone, 43% in groups of 2–5, 13% in groups of 6–10, and the rest in groups up to 19. Observations from Chang Tang suggest a sex ratio of 67– 75 males: 100 females (Schaller 1998).

Space use.— The wild yak is now restricted to very highelevation and remote uplands, usually free of human harassment. It is not daunted by mountainous terrain (Schaller 1998) because of its ‘‘strong limbs and small hooves of compact texture, with a narrow and sharp hoof tip, hard hoof edges and a close hoof fork’’ ( Wiener et al. 2003:81). The Tibetan Plateau contains as many as 17 vegetation types, but alpine meadows (45%), alpine steppe (29%), and desert-type grasslands and steppe (14%) comprise 88% of the land cover ( Sheehy et al. 2006). The wild yak occurs in greatest abundance on alpine meadows, less so in alpine steppe, and is scarce in desert steppe (Schaller and Liu 1996). Preferred habitats in Chang Tang include partially glaciated mountains with slopes of alpine meadows, seasonally lush alpine steppe that may green up 2–3 weeks before the plains, and edges of streams (Schaller 1998). Male wild yaks occur often on gentle slopes, and female herds occur more often on high hills and upper slopes ( Harris 1993; Miller et al. 1994; Schaller 1998).

The wild yak is capable of long-distance and unpredictable movements ( Harris 1993, 2008), some of which may be associated with avoidance of human activities. The wild yak is not migratory, typically moves up and down slopes seasonally to take advantage of the best forage availability, and may shift ranges seasonally or if harassed (Schaller 1998). Most of the Tibetan Plateau has sparse vegetative cover (e.g., only about 10–15% in alpine steppe of the Chang Tang—Schaller and Ren 1988) with low primary productivity (80–160 kg /ha dry matter—Schaller 1998; Schaller et al. 2005), but alpine meadows, preferred by wild yaks, can be up to 9 times as productive as alpine steppe and alpine desert-type habitats ( Long 2003a; Sheehy et al. 2006). Such meadows are frequently covered with a heavily grazed turf of the sedge Kobresia about 5 cm above the ground ( Koizumi et al. 1993; Rockhill 1895).

Diet.— The yak is a herbivorous ruminant. Foraging preferences of the wild yak are understood mainly from limited microhistological analyses of feces (Harris and Miller 1995; Miller et al. 1994; Schaller and Liu 1996). The yak is a grazer ( Poephagus 5 grass eater), seasonally eating grasses, sedges, and forbs. Hedin (1934:29) noted that wild yaks ‘‘find nourishment in the mosses and lichens on mountain slopes and among old and new moraines.’’

For all ungulates of the Tibetan Plateau ( Harris 2008; Schaller 1998), dietary diversity is constrained substantially by seasonally limited forage availability and diversity, but sedges and grasses, followed by forbs, dominate diets during the short summer growing season (Harris and Miller 1995; Miller et al. 1994). In Chang Tang, analyses of wild yak feces show a preference for grasses and sedges ( Stipa , 52%; Kobresia , 4%; Carex moorcroftii , 14%; and other grasses, 4%), followed by herbaceous plants (12%) and the dwarf shrub Ceratoides compacta (10%—Schaller and Liu 1996). In Yeniugou, Qinghai, wild yak feces in summer contain 85.5% sedges and grass (sedges: 67.1% Kobresia and 5.3% Carex ; and grasses: 13.1%) and almost 4% mosses (Harris and Miller 1995). In autumn, grasses dominate (68.8%) the diet of wild yaks in Yeniugou, and sedges become less important (25.3%—Miller et al. 1994).

Generally, ungulates of the Tibetan Plateau must contend with nutritionally deficient diets from winter and early spring ( Long 2003a; Schaller 1998; Wiener et al. 2003). Diets of wild and domestic yaks are low in protein (about 6%) from October to May ( Long 2003b; Ping et al. 2002; Schaller 1998). Deficiencies of sodium ( Ping et al. 2002), copper (Clauss and Dierenfeld 1999; Shen et al. 2006), and molybdenum ( Long 2003b) and plant-induced pyrrolizidine alkaloid poisoning in India ( Mondal et al. 1999) and Bhutan ( Winter et al. 1993) have been noted in domestic yaks. Little is known about the specific water requirements of wild yaks, but early chroniclers noted frequent visits to mineral-rich warm springs ( Przewalski 1876) and rivers ( Rockhill 1894) and consumption of snow. Herders drive domestic yaks to water sources as often as twice a day, particularly under twice-a-day milking regimes ( Wiener et al. 2003).

Diseases and parasites.— Rockhill (1894:118) mentioned a type of ‘‘cattle plague’’ in eastern Tibet that killed pastoralists’ livestock and was particularly hard on wild yaks in the late 1800s. Przewalski (1876) described ‘‘mange’’ (‘‘homun’’ in Mongolian) on wild yaks and considerable loss of hair on some individuals that he shot. Currently, no known pathogen or disease singularly affects extant populations of wild yaks, but they are at serious risk of disease transmission from association with domestic yaks, which frequently associate with domestic cattle, particularly on winter range (Dorji et al. 2003).

Many of the serious disease- and mortality-causing pathogens of domestic cattle can be transmitted to, and many of them have been found in, domestic yaks (Dorji et al. 2003; Pal and Kar 1999), including bacterial (anthrax, brucellosis, bovine pleuropneumonia, Chlamydia , and Salmonella [ Sharma et al. 1996]) and viral (foot-and-mouth disease [Barman et al. 1999] and infectious bovine rhinotracheitis) diseases. Although impractical for wild yaks, domestic yaks can be effectively vaccinated against many of these. Various ecto- and endoparasites, such as warble fly larvae ( Li et al. 2004), ticks ( Haemaphysalis — Yin et al. 2002), and the bladder larval tapeworm Coenurus cerebralis (Sharma and Chauhan 2006) , among others (Dorji et al. 2003; Heath et al. 1984), infect domestic yaks and probably wild yaks.

Interspecific interactions.— The Tibetan Plateau has a rich wild ungulate fauna, although it has been diminished greatly by human activities ( Harris 2008; Schaller 1998). Wild yaks can be sympatric with chiru or Tibetan antelope ( Pantholops hodgsonii —Leslie and Schaller 2008), Tibetan gazelle ( Procapra picticaudata —Schaller 1998), kiang or Tibetan wild ass ( Equus kiang —St-Louis and Côté 2009), bharal or blue sheep ( Pseudois nayaur —Wang and Hoffmann 1987), Tibetan argali ( Ovis ammon hodgsoni —Fedosenko and Blank 2005), and occasionally others, such as white-lipped deer ( Przewalskium albirostris —Harris and Miller 1995; Schaller 1998). As in mixed ungulate assemblages elsewhere, Tibetan species likely partition food and space, relative to size and digestive capabilities, to minimize competition (Harris and Miller 1995; Schaller 1998; Schaller et al. 1991). For example, wild yaks and argalis tend to use hilly to mountainous areas, chirus share flatlands with Tibetan gazelles, and the kiang uses both ( Schaller et al. 1991). Nikol’skii and Ulak (2006) concluded that habitats of Himalayan marmots ( Marmota himalayana ) benefitted from heavy use in the past by wild yaks and currently by domestic yaks.

Scant information exists on the predator–prey dynamics on the Tibetan Plateau, and current dynamics are a product of greatly reduced populations of both due to various human activities. The degree to which ungulates are preyed on or scavenged is largely unknown, and separating wild and domestic yaks in predators’ feces, for example, is difficult in places where they both occur. In Kekexili Nature Reserve, Qinghai, contents of feces from Tibetan brown bear ( Ursus arctos pruinosus ; predator and scavenger) suggested a summer diet of 31% wild yak (dry weight in feces— Xu et al. 2006), but Schaller (1998) noted only 0.4% in bear feces from Chang Tang. Feces of wolves ( Canis lupus ) contain 0– 10.4% yak in various parts of Tibet, Qinghai, and Xinjiang (Schaller 1998), but depredation of domestic yaks can represent 60% of the total livestock losses in India ( Namgail et al. 2007). The snow leopard ( Uncia uncia ) preys on domestic yaks in limited areas, notably Mongolia and Nepal ( Ikeda 2004; Namgail et al. 2007; Oli 1994; Oli et al. 1993; Schaller 1998). The lynx ( Lynx lynx ) is an uncommon predator of yaks (Namigail et al. 2007).

Figures

Fig. 1.—Mature male wild yak (Bos mutus) in Yeniugou, central Qinghai, China. Photograph courtesy of Milo Burcham (www. milophotos.com).

Fig. 2.—Distribution of the wild yak (Bos mutus) is restricted to the Tibetan Plateau of western China and includes at least 20 fragmented populations; map adapted from Schaller (1998:131) with updates from R. B. Harris (pers. comm.).

Fig. 3.—Skulls of male (left) and female (right) wild yaks (Bos mutus) from Yeniugou, central Qinghai, China, highlighting relative mass and sexual dimorphism in skull size and horn shape. Photograph courtesy of Daniel J. Miller.

Fig. 4.—Dorsal, ventral, and lateral views of skull and lateral view of mandible of an adult male domestic yak (Bos grunniens); zoo specimen of unknown origin (National Museum of Natural History, specimen 174734). Greatest length of skull 525 mm.

Fig. 5.—Wild yak (Bos mutus) adult females and calves in Yeniugou, central Qinghai, China. Photograph courtesy of Milo Burcham (www.milophotos.com).